Oral zinc for treating diarrhoea in children

Abstract

Background

In developing countries, diarrhoea causes around 500,000 child deaths annually. Zinc supplementation during acute diarrhoea is currently recommended by the World Health Organization (WHO) and the United Nations Children's Fund (UNICEF).

Objectives

To evaluate oral zinc supplementation for treating children with acute or persistent diarrhoea.

Search methods

We searched the Cochrane Infectious Diseases Group Specialized Register, CENTRAL (the Cochrane Library 2016, Issue 5), MEDLINE, Embase, LILACS, CINAHL, mRCT, and reference lists up to 30 September 2016. We also contacted researchers.

Selection criteria

Randomized controlled trials (RCTs) that compared oral zinc supplementation with placebo in children aged one month to five years with acute or persistent diarrhoea, including dysentery.

Data collection and analysis

Both review authors assessed trial eligibility and risk of bias, extracted and analysed data, and drafted the review. The primary outcomes were diarrhoea duration and severity. We summarized dichotomous outcomes using risk ratios (RR) and continuous outcomes using mean differences (MD) with 95% confidence intervals (CI). Where appropriate, we combined data in meta‐analyses (using either a fixed‐effect or random‐effects model) and assessed heterogeneity.

We assessed the certainty of the evidence using the GRADE approach.

Main results

Thirty‐three trials that included 10,841 children met our inclusion criteria. Most included trials were conducted in Asian countries that were at high risk of zinc deficiency.

Acute diarrhoea

There is currently not enough evidence from well‐conducted RCTs to be able to say whether zinc supplementation during acute diarrhoea reduces death or number of children hospitalized (very low certainty evidence).

In children older than six months of age, zinc supplementation may shorten the average duration of diarrhoea by around half a day (MD −11.46 hours, 95% CI −19.72 to −3.19; 2581 children, 9 trials, low certainty evidence), and probably reduces the number of children whose diarrhoea persists until day seven (RR 0.73, 95% CI 0.61 to 0.88; 3865 children, 6 trials, moderate certainty evidence). In children with signs of malnutrition the effect appears greater, reducing the duration of diarrhoea by around a day (MD −26.39 hours, 95% CI −36.54 to −16.23; 419 children, 5 trials, high certainty evidence).

Conversely, in children younger than six months of age, the available evidence suggests zinc supplementation may have no effect on the mean duration of diarrhoea (MD 5.23 hours, 95% CI −4.00 to 14.45; 1334 children, 2 trials, low certainty evidence), or the number of children who still have diarrhoea on day seven (RR 1.24, 95% CI 0.99 to 1.54; 1074 children, 1 trial, low certainty evidence).

None of the included trials reported serious adverse events. However, zinc supplementation increased the risk of vomiting in both age groups (children greater than six months of age: RR 1.57, 95% CI 1.32 to 1.86; 2605 children, 6 trials, moderate certainty evidence; children less than six months of age: RR 1.54, 95% CI 1.05 to 2.24; 1334 children, 2 trials, moderate certainty evidence).

Persistent diarrhoea

In children with persistent diarrhoea, zinc supplementation probably shortens the average duration of diarrhoea by around 16 hours (MD −15.84 hours, 95% CI −25.43 to −6.24; 529 children, 5 trials, moderate certainty evidence).

Authors' conclusions

In areas where the prevalence of zinc deficiency or the prevalence of malnutrition is high, zinc may be of benefit in children aged six months or more. The current evidence does not support the use of zinc supplementation in children less six months of age, in well‐nourished children, and in settings where children are at low risk of zinc deficiency.

12 April 2019

Up to date

All studies incorporated from most recent search

All eligible published studies found in the last search (30 Sep, 2016) were included and two ongoing studies have been identified (see 'Characteristics of ongoing studies' section)

Plain language summary

Oral zinc supplementation for treating diarrhoea in children

In low‐ and middle‐income countries, millions of children suffer from severe diarrhoea every year and many die from dehydration. Giving fluids by mouth (using an oral rehydration solution (ORS)) has been shown to save children's lives, but it has no effect on the length of time the children suffer with diarrhoea. Zinc supplementation could help reduce the duration and the severity of diarrhoea, and therefore have an additional benefit over ORS in reducing children mortality.

What is oral zinc and how may it shorten the duration and severity of diarrhoea

Zinc is usually given as zinc sulphate, zinc acetate, or zinc gluconate, which are all water‐soluble compounds. The World Health Organization (WHO) and the United Nations Children's Fund (UNICEF) recommend 10 mg to 20 mg of zinc per day for children with diarrhoea. There are several mechanism of action of zinc on acute diarrhoea, some of which are specific to the gastrointestinal system: zinc restores mucosal barrier integrity and enterocyte brush‐border enzyme activity, it promotes the production of antibodies and circulating lymphocytes against intestinal pathogens, and has a direct effect on ion channels, acting as a potassium channel blocker of adenosine 3‐5‐cyclic monophosphate‐mediated chlorine secretion. Cochrane researchers examined the evidence available up to 30 September 2016.

What the evidence in the review suggests

Thirty‐three trials that included 10,841 children met the inclusion criteria of this review.

Among children with acute diarrhoea, we don't know if treating children with zinc has an effect on death or number of children hospitalized (very low certainty evidence). In children older than six months, zinc supplementation may shorten the average duration of diarrhoea by around half a day (low certainty evidence), and probably reduces the number of children whose diarrhoea persists until day seven (moderate certainty evidence). In children with signs of malnutrition the effect appears greater, reducing the duration of diarrhoea by around a day (high certainty evidence). Conversely, in children younger than six months, the available evidence suggests zinc supplementation may have no effect on the mean duration of diarrhoea (low certainty evidence), or the number of children who still have diarrhoea on day seven (low certainty evidence). Zinc supplementation increased the risk of vomiting in both age groups (moderate certainty evidence). No other adverse effects were reported.

Among children with persistent diarrhoea, zinc supplementation probably shortens the average duration of diarrhoea by around 16 hours (moderate certainty) but it probably increases the risk of vomiting (moderate certainty evidence).

In areas where the prevalence of zinc deficiency or the prevalence of malnutrition is high, zinc may be of benefit in children aged six months or more. The current evidence does not support the use of zinc supplementation in children less six months of age, in well‐nourished children, and in settings where children are at low risk of zinc deficiency.

Summary of findings

Summary of findings for the main comparison. 'Summary of findings' table 1.

Zinc compared to placebo for children more than 6 months of age with acute diarrhoea
Patient or population: children with acute diarrhoea Settings: all countries Intervention: zinc Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	Number of participants (trials)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Zinc
Duration of diarrhoea	All trials		MD −11.46 (−19.72 to −3.19)	2581 (9 trials)	⊕⊕⊝⊝ low1,2	No comment
	The mean duration of diarrhoea among placebo ranged from 31.2 to 169.5 hours	The mean duration of diarrhoea among zinc ranged from 28.8 to 147.6 hours
	Trials limited to children with signs of malnutrition		MD −26.39 (−36.54 to −16.23)	419 (5 trials)	⊕⊕⊕⊕ high	No comment
	The mean duration of diarrhoea among placebo ranged from 103.4 to 146.4 hours	The mean duration of diarrhoea among zinc ranged from 70.4 to 120.0 hours
Diarrhoea on day 7	128 per 1000	93 per 1000 (78 to 113)	RR 0.73 (0.61 to 0.88)	3865 (6 trials)	⊕⊕⊕⊝ moderate3	No comment
Number of children hospitalized (community trials only)	—	—	—	276 (1 trial)	⊕⊝⊝⊝ very low4,5	No events
Death	5 per 1000	1 per 1000 (0 to 11)	RR 0.29 (0.04 to 2.20)	1134 (4 trials)	⊕⊝⊝⊝ very low6,7	Few events
Adverse events (vomiting)	119 per 1000	188 per 1000 (173 to 242)	RR 1.57 (1.32 to 1.86)	2605 (6 trials)	⊕⊕⊕⊝ moderate8	No comment
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). Abbreviations: CI: confidence interval; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

¹Downgraded by 1 for indirectness: all trials were conducted in Asia.
 ²Downgraded by 1 for serious imprecision: wide CI.
 ³Downgraded by 1 for serious indirectness: these trials were all conducted in Asia in countries at high risk of zinc deficiency.
 ⁴Downgraded by 1 for serious indirectness: only one small community trial reported on number of children hospitalized.
 ⁵Downgraded by 2 for very serious imprecision: no hospitalizations occurred in this trial.
 ⁶Downgraded by 1 for serious indirectness: the included trials were mostly conducted in hospitals and are therefore likely to underestimate death at the community level.
 ⁷Downgraded by 2 for very serious imprecision: only three deaths occurred in these two trials, consequently the trials are significantly underpowered to detect or exclude an effect.
 ⁸Downgraded by 1 for serious risk of bias: two trials reported no details on sequence generation, allocation concealment, blinding, and incomplete outcome data, while one did not give any details regarding allocation concealment.

Summary of findings 2. 'Summary of findings' table 2.

Zinc compared to placebo for children aged less than 6 months with acute diarrhoea
Patient or population: children aged less than 6 months with acute diarrhoea Settings: all countries Intervention: zinc Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	Number of participants (trials)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Zinc
Duration of diarrhoea	The mean duration of diarrhoea among placebo ranged from 97.9 to 133.2 hours	The mean duration of diarrhoea among zinc ranged from 105.6 to 133.2 hours	MD 5.23 (−4.00 to 14.45)	1334 (2 trials)	⊕⊕⊝⊝ low1,2	No comment
Diarrhoea on day 7	203 per 1000	252 per 1000 (201 to 313)	RR 1.24 (0.99 to 1.54)	1074 (1 trial)	⊕⊕⊝⊝ low3,4	No comment
Number of children hospitalized (community trials only)	—	—	—	1074 (1 trial)	⊕⊝⊝⊝ very low5,6	No events
Death	2 per 1000	2 per 1000 (0 to 32)	RR 1.00 (0.06 to 15.89)	1334 (2 trials)	⊕⊝⊝⊝ very low7,8	Only 1 event in each treatment group
Adverse events (vomiting)	64 per 1000	104 per 1000 (67 to 143)	RR 1.54 (1.05 to 2.24)	1334 (2 trials)	⊕⊕⊕⊝ moderate9	No comment
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). Abbreviations: CI: confidence interval; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

¹Downgraded by 1 for inconsistency: only two trials were done and both had inconsistent results.
 ²Downgraded by 1 for imprecision: large CI.
 ³Downgraded by 1 for inconsistency: different results in the subgroups.
 ⁴Downgraded by 1 for indirectness: only one trial (although multi‐country) as it is not possible to generalize these results.
 ⁵Downgraded by 2 for imprecision: only one hospitalization was recorded in 1074 participants. Much larger trials would be necessary to prove or exclude an effect.
 ⁶Downgraded by 1 for imprecision: the result is not statistically significant.
 ⁷Downgraded by 1 for indirectness: most of this data is from Asia and may not be applicable elsewhere.
 ⁸Downgraded by 2 for imprecision: in these two trials deaths were very rare, and consequently these trials are significantly underpowered to detect or exclude an effect on mortality.
 ⁹Downgraded by 1 under consistency because only two trials were done.

Summary of findings 3. 'Summary of findings' table 3.

Zinc compared to placebo for children with persistent diarrhoea
Patient or population: children with persistent diarrhoea Settings: all countries Intervention: zinc Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	Number of participants (trials)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Zinc
Duration of diarrhoea	The mean duration of diarrhoea among placebo ranged from 84 to 132	The mean duration of diarrhoea among zinc ranged from 69.4 to 122.4 hours	MD −15.84 (−25.43 to −6.24)	529 (5 trials)	⊕⊕⊕⊝ moderate1	No comment
Diarrhoea on day 7	191 per 1000	99 per 1000 (52 to 195)	RR 0.52 (0.27 to 1.02)	221 (2 trials)	⊕⊕⊝⊝ low1,2	No comment
Hospitalization	—	—	—	275 (1 trial)	⊕⊝⊝⊝ very low1,3	No events
Death	—	—	—	402 (3 trials)	⊕⊝⊝⊝ very low1,4	No events
Adverse events (vomiting)	8 per 1000	16 per 1000 (3 to 85)	RR 1.97 (0.37 to 10.59)	505 (4 trials)	⊕⊝⊝⊝ very low1,5	No comment
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). Abbreviations: CI: confidence interval; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

¹Downgraded by 1 for indirectness: most of this data is from Asia and may not be applicable elsewhere.
 ²Downgraded by 1 for imprecision: the result does not reach statistically significance, and the number of recorded events is low.
 ³Downgraded by 2 for imprecision: no hospitalizations were recorded. Much larger trials would be necessary to prove or exclude an effect.
 ⁴Downgraded by 2 for imprecision: in these three trials deaths were very rare, and consequently these trials are significantly underpowered to detect or exclude an effect on mortality.
 ⁵Downgraded by 2 for imprecision: vomiting was very uncommon in these trials.

Background

Description of the condition

Despite improving trends in mortality rates, diarrhoea still causes nearly 10% of all deaths in children under five years of age and accounts for about 500,000 child deaths in developing countries every year (Liu 2015a; Liu 2015b). The incidence of diarrhoea decreased from 3.4 episodes per child‐year in 1990 to 2.9 episodes per child‐year in 2010. However, it still remains one of the most common reasons of hospital admission, with an estimated 1731 million episodes of childhood diarrhoea reported in 2011 of which 2% progressed to severe disease (Das 2014). Diarrhoea is also an important cause of malnutrition, particularly when it is prolonged (Brown 2003).

Zinc deficiency is mainly due to inadequate dietary intake and is estimated to be common in many countries, especially in children (IZiNCG 2004; Wagstaff 2004; Hess 2009). According to recent estimates, 17.3% of the world’s population is currently at risk of inadequate zinc intake (Wessells 2012). The regional estimated prevalence of inadequate zinc intake ranges from 7.5% in high‐income regions to 30% in South Asia (Wessells 2012). Foods more rich in zinc are 'expensive foods', such as meat and fish (IZiNCG 2004). Zinc is also present in nuts, seeds, legumes, and wholegrain cereal, but the high phytate content of these foods interferes with its absorption (IZiNCG 2004). Zinc cannot be stored in the body, and nearly 50% of zinc excretion takes place through the gastrointestinal tract and is increased during episodes of diarrhoea (IZiNCG 2004). Zinc requirement varies with age and is highest in children due to their rapid rates of growth. As a consequence, young children who are regularly exposed to gastrointestinal pathogens and have diets low in animal products and high in phytate‐rich foods are most at risk of zinc deficiency (IZiNCG 2004).

Description of the intervention

Treatment of diarrhoea with oral rehydration solution (ORS) reduces mortality due to dehydration (Liu 2015a; Liu 2015b). In addition to ORS, the World Health Organization (WHO) and United Nations Children's Fund (UNICEF) recommend for children under five years of age with diarrhoea a supplementation with 10 to 20 mg of zinc per day, at least twice the recommended daily allowance (WHO/UNICEF 2004). Zinc is usually given as zinc sulphate, zinc acetate, or zinc gluconate, which are all water‐soluble compounds (IZiNCG 2004).

How the intervention might work

There are several different mechanism of action of zinc on acute diarrhoea (Berni Canani 2010; Krebs 2014). Zinc influences the activity of over 300 enzymes, some of which are responsible for DNA replication and transcription (IZiNCG 2004). Zinc promotes immunity, skin and mucosal resistance to infection, growth, and development of the nervous system (MacDonald 2000; Prasad 2008; Hess 2009). It is also an important antioxidant and preserves cellular membrane integrity (O'Dell 2000; Powell 2000; Prasad 2014). At the level of gastrointestinal system, zinc restores mucosal barrier integrity and enterocyte brush‐border enzyme activity (Roy 1992; Shankar 1998), it promotes the production of antibodies and circulating lymphocytes against intestinal pathogens (Sazawal 1997b; Albert 2003; Raqib 2004), and has a direct effect on ion channels, acting as a potassium channel blocker of adenosine 3‐5‐cyclic monophosphate‐mediated chlorine secretion (Hoque 2005; Hoque 2009).

Zinc supplementation may have different effects according to the level of zinc deficiency in the country and in the individual. It is important to verify whether zinc supplementation is effective in countries with high, or even medium or low risk of zinc deficiency (IZiNCG 2004). Despite an accurate estimation of the prevalence of zinc deficiency in populations is hampered by the lack of reliable indicators or biomarkers (Wieringa 2015), indirect indicators such as the prevalence of stunting or anaemia, and the absorbable zinc content of the national food supply are currently used at to estimate the prevalence of zinc deficiency in populations (IZiNCG 2004). Zinc requirements are higher in malnourished children because nutritional zinc deficiency is considered more severe in these children (IZiNCG 2004). However, infants have lower requirements (IZiNCG 2004), as healthy normal birthweight infants have adequate zinc levels at birth from maternal sources even if maternal stores are suboptimal (Iqbal 2001). Infants may also be able to mobilize hepatic stores accumulated during gestation (Zlotkin 1988), and are less likely to have had a zinc‐depleting illness. Breastfeeding will provide zinc supplementation and protective immune factors against infections (Krebs 1999).

Zinc can cause vomiting because of its metallic taste (Fontaine 2001). In high doses, zinc can also cause epigastric pain, lethargy, and fatigue (IZiNCG 2004). One small study suggested a possible increase in mortality in malnourished children supplemented with 6 mg/kg/day of zinc compared to those supplemented with 1.5 mg/kg/day (Doherty 1998). Copper deficiency with zinc supplementation can occur although usually only when zinc is consumed in very high doses (100 to 300 mg/day for adults) over a long period of time (IZiNCG 2004), and malnourished children are at particularly high risk of this due to lower basal copper levels.

Iron and zinc deficiencies often co‐exist. These two compounds may compete for the same absorptive pathways, and iron may interfere with zinc utilization (Gunshin 1997; Kordas 2004). A review of combined supplementation showed that giving zinc with iron resulted in a lower increase in iron levels compared to giving iron alone; iron supplementation alone had no effect on zinc status (Fischer Walker 2005). A trial that assessed combined supplementation on diarrhoea and malaria morbidity showed that zinc combined with iron reduced zinc's protective effect against diarrhoea (Richard 2006). Several trials have also reported a negative interaction of the combined supplementation on physical growth and development (Rosado 1997; Dijkhuizen 2001; Zlotkin 2003; Lind 2004; Bhandari 2007). Some protocols suggest supplementing malnourished children also with copper because these children are also prone to copper deficiency (Beshgetoor 1998).

Why it is important to do this review

Previous meta‐analysis and systematic reviews have indicated that zinc supplementation in diarrhoea is effective (Bhutta 2000b; Lukacik 2008; Patro 2008; Haider 2009; Liberato 2015; Zou 2015; Lazzerini 2016). This Cochrane Review will have an up‐to‐date extensive search for trials, will explore more outcome measures of interest, and will report on more possible sources of heterogeneity. This Cochrane Review updates the last published version of this review (Lazzerini 2013).

Objectives

To evaluate oral zinc supplementation for treating children with acute or persistent diarrhoea.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (RCTs).

Types of participants

Children between one month and five years of age with acute or persistent diarrhoea, including dysentery.

We excluded trials of infants below one month of age and studies that exclusively enrolled children with particular conditions, such preterm or low birthweight infants and children with HIV.

Acute diarrhoea is usually defined as three or more loose stools in a 24‐hour period. Persistent diarrhoea is defined as diarrhoea lasting more than 14 days. Dysentery is a diarrhoeal illness in which blood is observed in the stool. The final day of diarrhoea is usually defined as the last day meeting the above definition followed by 48 hours without diarrhoea.

Types of interventions

Intervention

Oral zinc supplementation of any zinc salt at doses of 5 mg/day or more for any duration.

Control

Placebo.

Concurrent supplementation of other minerals and vitamins are eligible only if administered to both the intervention and control groups.

We excluded ORS plus zinc and food fortification interventions (such as milk fortification) as the amount of ORS/food consumed, and hence the zinc intake, would be less certain.

Types of outcome measures

Primary outcomes

Measures of diarrhoea duration

• Diarrhoea duration.

• Diarrhoea at 3, 5, and 7 days after starting the intervention.

Measures of diarrhoea severity

• Stool frequency.

• Stool output.

Secondary outcomes

• Hospitalization (number of children hospitalized).

• Death (from any cause and diarrhoea specific).

Adverse events

• Serious adverse events (life‐threatening or requiring hospitalization).

• Any adverse event that results in the discontinuation of treatment.

• Other adverse events, such as vomiting and reduced copper levels.

Search methods for identification of studies

We attempted to identify all relevant trials regardless of language or publication status (published, unpublished, in press, and in progress).

Electronic searches

We searched the following databases using the search terms and strategy described in Table 4: the Cochrane Infectious Diseases Group Specialized Register (30 September 2016); Cochrane Central Register of Controlled Trials (CENTRAL), published in the Cochrane Library (2016, Issue 9); MEDLINE (1966 to 30 September 2016); Embase (1974 to 30 September 2016); LILACS (1982 to 30 September 2016); CINAHL (1982 to 30 September 2016), the metaRegister of Current Controlled Trials (mRCT; 30 September 2016), ClinicalTrials.gov (30 September 2016), and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (30 September 2016).

1. Detailed search strategies.

Search set	CIDG SR1	CENTRAL	MEDLINE2	EMBASE2	LILACS2	CINAHL	CCT
1	zinc	zinc	zinc	zinc	zinc	zinc	zinc
2	diarrhoea	diarrhoea	ZINC	ZINC	diarrhoea	diarrhoea	diarrhoea
3	vomiting	morbidity	1 or 2	1 or 2	morbidity	morbidity	vomiting
4	adverse effects	2 or 3	diarrhoea	diarrhoea	2 or 3	2 or 3	adverse effects
5	—	1 and 4	diarrhoea	morbidity	1 and 4	1 and 4	—
6	—	vomiting	morbidity	4 or 5	vomiting	vomiting	—
7	—	adverse effects	MORBIDITY	3 and 6	adverse effects	adverse effects	—
8	—	6 or 7	4 or 5 or 6 or 7	Limit 7 to human	6 or 7	6 or 7	—
9	—	1 and 2 and 8	3 and 8	vomiting	1 and 2 and 8	1 and 8	—
10	—	—	Limit 9 to human	adverse effects	—	—	—
11	—	—	vomiting	9 or 10	—	—	—
12	—	—	adverse effects	3 and 4 and 11	—	—	—
13	—	—	11 or 12	—	—	—	—
14	—	—	3 and (4 or 5) and 13	—	—	—	—

¹Cochrane Infectious Diseases Group Specialized Register.
 ²Search terms used in combination with the search strategy for retrieving trials developed by Cochrane (Higgins 2006); upper case: MeSH or EMTREE heading; lower case: free text term.

Searching other resources

Researchers and organizations

For unpublished and ongoing trials, we contacted individual researchers working in the field, including researchers at the WHO.

Reference lists

We checked the reference lists of all studies identified by the above methods.

Data collection and analysis

Selection of studies

Both review authors screened all trials identified by the search strategy by title/abstract, and we retrieved the full‐text articles of all potentially relevant trials. Both review authors independently applied the inclusion criteria to the full‐text reports using a pilot‐tested data extraction form, and scrutinized publications to ensure we included each trial only once. We contacted the trial authors for clarification if necessary, and resolved any disagreements through discussion and consensus after referring to the protocol; we recorded and reported their solutions. We listed studies excluded after full‐text assessment and their reasons for exclusion in a 'Characteristics of excluded studies' table. We constructed a PRISMA flow diagram to illustrate the study selection process.

Data extraction and management

Both review authors independently extracted data using a pilot‐tested data extraction form and entered the data into Review Manager 5 (RevMan 5) (Review Manager 5). When data were missing or unclear, we contacted the trial authors for clarification. For dichotomous outcomes, we recorded the number of participants that experienced the event and the number of participants assessed in each group. For continuous outcomes, we extracted the arithmetic mean values, standard deviation (SD) values, and the number of participants assessed in each group. If a trial reported continuous data using geometric means, we extracted the SD values on the log scale; we extracted median and range values and reported these in a table.

Assessment of risk of bias in included studies

Both review authors independently assessed the risk of bias for each trial using the Cochrane 'Risk of bias' assessment tool (Higgins 2011). We categorized our judgments as either at low, high, or unclear risk of bias, and we used this information to guide our interpretation of the results. Where the judgement for efficacy trials was 'unclear' we attempted to contact the trial authors for clarification and we resolved any differences of opinion through discussion. If data were missing or unclear, we contacted the trial authors for further information.

Measures of treatment effect

For dichotomous data, we reported outcome measures using the risk ratio (RR). For continuous data summarized by arithmetic means and SDs, we used the mean difference (MD) values to combine the results in a meta‐analysis. We presented continuous data summarized using other summary statistics that we could not combine in a meta‐analysis in a table. We calculated the geometric mean ratios and transformed them in the log scale for analysis, and presented them on the natural scale

Unit of analysis issues

There were no cluster RCTs.

Dealing with missing data

We planned to conduct an analysis so that we included all individuals with a recorded outcome in the analysis. If there was missing information or we needed more details on reported measures, we sought further clarification from the trial authors. To take into account the participants for whom there was no outcome data available, we also conducted an intention‐to‐treat (ITT) analysis for worst‐case/best‐case scenarios.

Assessment of heterogeneity

We assessed heterogeneity between trials by visually inspecting the forest plot, using the Chi² test for heterogeneity with a 5% level of statistical significance, and the I² statistic with a value of 50% representing a moderate level of heterogeneity. If we detected significant heterogeneity but considered it was appropriate to pool data, we used the random‐effects model.

Assessment of reporting biases

We constructed funnel plots to look for evidence of publication bias for the outcome duration of diarrhoea and diarrhoea at day 7.

Data synthesis

We analysed the data using RevMan 5 (Review Manager 5). We presented all results with 95% confidence intervals (CIs).

Quality of the evidence

We assessed the certainty of the evidence using the GRADE approach (GRADEpro GDT 2014). We used GRADEpro Guideline Development Tool (GDT) software to construct the 'Summary of findings' tables (GRADEpro GDT 2014). The GRADE system considers ‘certainty’ to be a judgment of the extent to which we can be confident that the estimates of effect are correct. The level of ‘certainty’ is judged on a four‐point scale. Evidence from RCTs is initially graded as high and downgraded by either 1, 2, or 3 levels after full consideration of: any limitations in the design of the studies, the directness (or applicability) of the evidence, the consistency and precision of the results, and the possibility of publication bias.

We have displayed the estimates of effect, and the GRADE assessments of our confidence in these estimates in 'Summary of findings tables' for the main comparisons. Where we have downgraded the evidence our reasons for doing so are displayed in footnotes.

When making conclusions about the relative effects of the interventions we used language that reflected the GRADE assessments and our confidence in the estimates, that is if the evidence was of high certainty we said "zinc reduces"; if it was of moderate certainty we stated "zinc probably reduces"; it was of low certainty we used "zinc may reduce"; and where the evidence was of very low certainty we did not draw conclusions.

Subgroup analysis and investigation of heterogeneity

We stratified the analyses for acute diarrhoea or persistent diarrhoea as these are different conditions. We also stratified the results by age (children aged less than and greater than six months) because we observed a clear difference in zinc effect according to the age of children enrolled and significant heterogeneity if we pooled all the trials together. We explored the following potential sources of heterogeneity using subgroup analyses: nutritional status (malnourished children versus well‐nourished plus moderate malnourished); geographical region (by continent and by high versus medium estimated risk of zinc deficiency as defined by the International Zinc Nutrition Consultative Group (IZiNCG) (IZiNCG 2004)); zinc dose (less than versus greater than 20 mg/day); zinc salt (zinc sulphate versus zinc acetate versus zinc gluconate versus other type); concomitant copper or iron supplementation; and trial setting (hospital versus community trials). We also explored the effect of sex, although we did not specify this in the original Cochrane Protocol (Lazzerini 2005).

Sensitivity analysis

We conducted a sensitivity analysis in which we limited the analyses to those trials with adequate allocation concealment, blinding (excluded those trials classified as unclear), and those that included an adequate number of randomized participants in the analysis (excluded those trials classified as inadequate or unclear). To take into account the participants for whom no outcome data were available, we also conducted an ITT analysis for worst‐case/best‐case scenarios.

Results

Description of studies

Results of the search

In the previous version of this review, Lazzerini 2013, 24 trials met the inclusion criteria. We updated the literature search to 30 September 2016 and identified 68 records after removal of duplicates. After screening the articles by title and abstract, we excluded 48 articles and assessed 20 articles for eligibility. Nine new trials met the inclusion criteria of the review and thus we included 33 trials (10,841 children) in this review update. We have presented a PRISMA study flow diagram in Figure 1 and we have reported the trial selection process in Table 5.

1.

Study flow diagram.

2. Results of the study selection.

Total number of studies identified through the search (up to 30 September 2016)	306 trials
Total number of studies excluded as clearly did not concern the topic of interest	126 trials
Studies further evaluated and excluded1 Not RCTs: 25 trials Not placebo‐controlled RCTs: 8 trials RCTs on prevention of diarrhoea, not on treatment: 51 trials Not concerning the population of interest (for example, studies on low birthweight, HIV): 13 trials Not concerning the interventions of interest (for example, studies on zinc in oral rehydration solution (ORS), multiple micronutrients, probiotics, food fortification): 19 trials Concerning different outcomes (for example, studies on serology, appetite, mental or motor development, malnutrition): 16 trials Could not be compared with other studies because of methodological problems (enrolling the same children more than once) and types of outcomes (episodes of diarrhoea and not children with diarrhoea): 1 trial Secondary analysis of other studies: 8	141 trials
Duplicates of included studies Folwaczny 1996; Darmon 1997 are review articles of the same trial (Sazawal 1995) Roy 1991 is a duplication of Roy 1997 and Roy 1998 Roy 1998 is an abstract of Khatun 2001 Cuevas 2000 is an abstract of Al‐Sonboli 2003 Patel 2013 is a cost effectiveness analysis based on data reported in Patel 2009	6 trials
Independent trials included in the review	33 trials (10,841 participants)

¹See the 'Characteristics of excluded studies' section.
 Abbreviations: HIV: human immunodeficiency virus; RCT: randomized controlled trial.

Included studies

Thirty‐three trials in total met the inclusion criteria of this review. We have listed the details of the included trials in the 'Characteristics of included studies' table. Three included trials presented results divided in two or more subgroups, and specifically: one trial presented two intervention groups of zinc 20 mg and zinc 5 mg, and one control group (Brooks 2005a); one trial presented data for three different study sites (Fischer Walker 2006); one trial presented the results as children with low and normal zinc serum levels (Polat 2003). For these three trials there was no way to combine mean values and standard deviation (SD) values, and thus we entered the data separately as Brooks 2005a (20 mg), Brooks 2005a (5 mg), Fischer Walker 2006 ETH, Fischer Walker 2006 IND, Fischer Walker 2006 PAK, Polat 2003 (low Zn), and Polat 2003 (normal Zn).

Type of diarrhoea

Most trials included children with acute diarrhoea only. Of these, 13 used the definition for acute diarrhoea that we used in this Cochrane Review (Faruque 1999; Dutta 2000; Strand 2002; Al‐Sonboli 2003; Polat 2003; Bhatnagar 2004a; Brooks 2005a; Fischer Walker 2006; Boran 2006; Dutta 2011; Crisinel 2015; Passariello 2015; Tran 2015), two trials defined diarrhoea as the presence of either four (Sazawal 1995), or five (Bahl 2002), unformed stools in 24 hours, one trial defined diarrhoea as acute onset of change in stool frequency and consistency (Karamyyar 2013), one shigellosis trial included participants with bloody mucoid diarrhoea (dysentery) or febrile diarrhoea less than five days' duration (Roy 2008a).

Three trials enrolled only children with rotavirus infection (Dalgic 2011; Jin 2013; Jiang 2016).

Eight trials did not report the definition of acute diarrhoea (Sachdev 1988; Roy 1997; Larson 2005;Fajolu 2008; Shimelis 2008; Patel 2009; Patro 2010; Jiang 2016).

Five trials were on children with persistent diarrhoea (Sachdev 1990; Roy 1998; Bhutta 1999; Penny 1999; Khatun 2001).

Age

Two trials only enrolled children under six months of age (Brooks 2005a; Fischer Walker 2006). Seventeen trials only enrolled children over six months of age (Sachdev 1988; Sachdev 1990; Sazawal 1995; Bhutta 1999; Faruque 1999; Penny 1999; Khatun 2001; Bahl 2002; Strand 2002; Boran 2006; Roy 2008a; Fajolu 2008; Patel 2009; Dutta 2011; Karamyyar 2013; Passariello 2015; Tran 2015). Fourteen trials included children of different ages greater than one month (Roy 1997; Roy 1998; Dutta 2000; Al‐Sonboli 2003; Polat 2003; Bhatnagar 2004a; Larson 2005; Shimelis 2008; Patro 2010; Dalgic 2011; Jin 2013; Crisinel 2015; Patel 2015; Jiang 2016).

Nutritional status

Eight trials only enrolled malnourished children (Roy 1997; Roy 1998; Bhutta 1999; Dutta 2000; Khatun 2001; Polat 2003; Roy 2008a; Passariello 2015). Two trials included well‐nourished children (Boran 2006; Patro 2010), and one trial enrolled children regardless of their nutritional status (Larson 2005), while the remaining 20 trials enrolled children who were well nourished or with mild or moderate malnutrition. No trials included only severe malnourished children. Two trials did not mention the nutritional status of children (Jin 2013; Jiang 2016). There was some variability between trials regarding the definition of malnutrition (most used 'weight/age'; only some used 'weight/height'); therefore we were unable to follow the definition of malnutrition proposed in our protocol (Lazzerini 2005).

Sex

Four trials only included males (Dutta 2000; Bhatnagar 2004a; Brooks 2005a; Dutta 2011), while the remaining 29 trials enrolled children of both sexes.

Geographical region

Most included trials were conducted in Asia. Only three trials were conducted in Europe (Patro 2010; Crisinel 2015; Passariello 2015), two in South America (Al‐Sonboli 2003; Penny 1999), two in Africa (Fajolu 2008; Shimelis 2008), one multicentre trial in Asia and Africa (Fischer Walker 2006), and one trial in Australia (Tran 2015). Thus, participants were from Bangladesh (Roy 1997; Roy 1998; Faruque 1999; Khatun 2001; Brooks 2005a; Larson 2005; Roy 2008a), India (Sachdev 1988; Sachdev 1990; Sazawal 1995; Dutta 2000; Bahl 2002; Bhatnagar 2004a; Fischer Walker 2006 IND; Patel 2009; Patel 2015), Pakistan (Bhutta 1999; Fischer Walker 2006 PAK), Nepal (Strand 2002), China (Jin 2013; Jiang 2016), Turkey (Polat 2003; Boran 2006; Dalgic 2011), Brazil (Al‐Sonboli 2003), Peru (Penny 1999), Ethiopia (Fischer Walker 2006 ETH; Shimelis 2008), Nigeria (Fajolu 2008), Poland (Patro 2010), Italy (Passariello 2015), Switzerland (Crisinel 2015), and Australia (Tran 2015).

Risk of zinc deficiency

Most trials were conducted in countries ranked as at high risk of zinc deficiency (IZiNCG 2004). Nine trials were conducted in countries at medium risk: Nepal (Strand 2002); Turkey (Polat 2003; Boran 2006, Dalgic 2011); Brazil (Al‐Sonboli 2003), China (Jin 2013), Iran (Karamyyar 2013), Nigeria (Fajolu 2008), and Ethiopia (Shimelis 2008). Four trials were conducted in countries where zinc deficiency is considered rare: Poland (Patro 2010), Italy (Passariello 2015), Switzerland (Crisinel 2015), and Australia (Tran 2015).

Zinc dose

The most frequent zinc dose was 20 mg/day. Only three trials administered higher zinc doses: 40 mg/day (Dutta 2000); 22 or 45 mg/day (Al‐Sonboli 2003); 20 and 40 mg respectively in children under and above six months of age (Passariello 2015). Two trials, of which one was of children aged less than six months only, gave 10 mg/day zinc (Fischer Walker 2006; Roy 2008a).

Seven trials used different dosages based on the age of children (under and above six months of age): 5 mg and 10 mg (Boran 2006), 10 mg and 20 mg (Fajolu 2008; Patro 2010; Crisinel 2015; Jiang 2016 ), 20 and 40 mg (Passariello 2015;); one trial used zinc at two different dosages (5 mg and 20 mg) in children aged less than six months (Brooks 2005a),

Seven trials used different doses depending on age (zinc < 20 mg in infants and ≥ 20 mg in older children), but they did not report results separately for each treatment group (Faruque 1999; Bahl 2002; Strand 2002; Bhatnagar 2004a; Boran 2006; Crisinel 2015; Passariello 2015). We classified these trials as 'not assignable' and could not include them in the sensitivity analysis for zinc dose.

Three trials reported a per kilo dose: 1 mg/kg/day (Karamyyar 2013); 2 mg/kg/day (Bhutta 1999); 3 mg/kg/day (Patel 2009). We were unable to include these trials in the subgroup analyses.

Type of zinc salt

Eight trials used zinc acetate (Roy 1997; Roy 1998; Faruque 1999; Khatun 2001; Strand 2002; Brooks 2005a; Roy 2008a; Dalgic 2011), five used zinc gluconate (Sazawal 1995; Penny 1999; Bahl 2002; Jin 2013; Jiang 2016), and three did not specify (Shimelis 2008; Dutta 2011; Passariello 2015), while all the remaining trials used zinc sulphate.

Concomitant copper or iron supplementation

One trial compared zinc alone versus zinc and copper versus placebo (Patel 2009).

Study setting

Most trials were conducted in hospitals, with the exception of six community‐based trials (Penny 1999; Bahl 2002; Strand 2002; Fischer Walker 2006; Boran 2006; Passariello 2015), and one trial was held in both hospital and community settings (Larson 2005).

Treatment regimen

Treatment duration

About half of trials administered zinc for two weeks. Of the remaining trials, one gave zinc for a total of four days (Tran 2015), three gave zinc for seven days after recovery (Bahl 2002; Strand 2002; Polat 2003), four gave zinc until recovery (Al‐Sonboli 2003; Brooks 2005a; Karamyyar 2013; Passariello 2015), one trial gave zinc for seven days (Khatun 2001), and three trials gave zinc for 10 days (Patro 2010; Crisinel 2015; Jiang 2016). Five trials were unclear in respect of duration of zinc supplementation (Sachdev 1988; Sachdev 1990; Sazawal 1995; Dutta 2000; Dalgic 2011). One trial on adverse events administered only one dose of zinc (Larson 2005).

Formulation

Most included trials administered zinc as syrup. Seven used dispersible tablets (Al‐Sonboli 2003; Larson 2005; Fischer Walker 2006; Shimelis 2008; Jin 2013; Crisinel 2015: Jiang 2016), four used powder (Sachdev 1988; Sachdev 1990; Penny 1999; Dalgic 2011), two mixed it with ORS (Passariello 2015; Tran 2015), and one did not specify (Fajolu 2008; Patel 2015).

Dose frequency

Zinc was administered once a day in most of the included trials. It was administered twice a day in five trials (Sachdev 1988; Sachdev 1990; Khatun 2001; Roy 2008a; Patro 2010), three times a day in six trials (Roy 1997; Roy 1998; Dutta 2000; Polat 2003; Bhatnagar 2004a), and together with ORS depending on stool frequency in two trials (Passariello 2015; Tran 2015). One trial administered zinc twice a day to infants, and a single dose to children over six months (Dalgic 2011). Three trials did not specify the dose frequency (Fajolu 2008; Patel 2009, Patel 2015).

Additional treatment

Most trials administered zinc alone. Seven trials used zinc and multivitamin, which did not contain iron (Sazawal 1995; Roy 1997; Roy 1998; Bhutta 1999; Khatun 2001; Bhatnagar 2004a; Roy 2008a). One trial used zinc and vitamin A (Faruque 1999). One trial used concomitant copper (Patel 2009).

Excluded studies

For this review update we excluded 11 studies after full‐text assessment. We have provided the total results of the study selection (that is, for the previous versions of this review plus this update) in Table 5, and listed the reasons for exclusion of studies after full‐text assessment in the 'Characteristics of excluded studies' table.

Risk of bias in included studies

See Figure 2 and Figure 3 for the risk of bias in the included trials.

2.

'Risk of bias' graph: review authors' judgements about each 'Risk of bias' item presented as percentages across all included trials.

3.

'Risk of bias' summary: review authors' judgements about each 'Risk of bias' item for each included trial.

Allocation

Twenty‐six trials used adequate methods to generate the allocation sequence. The methods used in the other trials was either at unclear (Sachdev 1988; Sachdev 1990; Khatun 2001; Fajolu 2008; Jin 2013; Jiang 2016), or at high risk of bias (Shimelis 2008).

Nineteen trials reported methods that assured adequate allocation concealment. Of the remaining trials, thirteen were at unclear risk of bias (Sachdev 1988; Sachdev 1990; Roy 1998; Khatun 2001; Al‐Sonboli 2003; Brooks 2005a; Boran 2006; Fajolu 2008; Dalgic 2011; Jin 2013; Patel 2015; Tran 2015; Jiang 2016), and one was at high risk of bias (Shimelis 2008).

Blinding

Twenty‐two trials were double blinded. Eight trials were at unclear risk of bias regarding the use of blinding (Sachdev 1988; Sachdev 1990; Khatun 2001; Fajolu 2008; Dalgic 2011; Jin 2013; Passariello 2015; Jiang 2016), and three trials were at high risk of bias (Boran 2006; Shimelis 2008; Patel 2015).

Incomplete outcome data

Twenty‐one trials included more than 90% of the randomized participants in the analysis. Seven trials included less than 90% of the randomized participants, which we assessed as at high risk of bias (Roy 1997; Bhutta 1999; Roy 2008a; Patro 2010; Karamyyar 2013; Crisinel 2015; Tran 2015), and the five remaining trials were at unclear risk of bias (Sachdev 1988; Sachdev 1990; Roy 1998; Dutta 2000; Fajolu 2008).

Selective reporting

Only three included trials were at low risk of bias regarding selective reporting (Roy 2008a; Patel 2009; Karamyyar 2013). The risk of bias was unclear for all other included trials, and the most frequent reason for this was the fact that the trial was not registered.

Other potential sources of bias

No information was available to evaluate other sources of bias. Therefore we judged each of the included trials as at unclear risk of bias regarding other potential sources of bias.

Effects of interventions

See: Table 1; Table 2; Table 3

Comparison 1: Zinc versus placebo for children with acute diarrhoea

Diarrhoea duration

On average, the mean duration of diarrhoea in children given zinc was around 13 hours shorter than those given placebo (mean difference (MD) −13.42 hours, 95% confidence interval (CI) −20.52 to −6.31; 5096 children, 20 trials, 24 comparisons; Analysis 1.1; Figure 4), but there was substantial statistical heterogeneity between trials (I² statistic = 84%).

1.1. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 1 Diarrhoea duration (hours).

4.

Zinc versus placebo for acute diarrhoea: diarrhoea duration (h)

In the primary analysis stratified by age, the benefit was only apparent in trials that recruited children over six months of age (MD −11.46 hours, 95% CI −19.72 to −3.19 hours; 2581 children, 9 trials; Table 1; Analysis 1.1; Figure 4), and trials that recruited all age groups (MD −22.18 hours, 95% CI −32.57 to −11.78 hours; 1181 children, 9 trials, 10 comparisons). In trials that only recruited children less than six months of age, no effect was demonstrated (1334 children, 2 trials, 5 comparisons; Table 2).

This age stratification did not adequately explain the statistical heterogeneity so we conducted a series of further subgroup analyses excluding the trials which only recruited children less than six months of age (Analysis 1.2 to Analysis 1.9). None of these subgroupings adequately explained the heterogeneity, but several observations are worth noting.

1.2. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 2 Diarrhoea duration (hours): subgrouped by nutritional status.

1.9. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 9 Diarrhoea duration (hours): subgrouped by concomitant treatment.

• When subgrouped by the nutritional status of participants, the smallest average effect was seen in trials that only recruited well‐nourished children, and the largest average effect in trials that only recruited children with signs of malnutrition (Analysis 1.2).

• There was only one included trial from the African continent, and this trial failed to show a benefit (Analysis 1.4).

• When subgrouped by the national risk of zinc deficiency, the smallest average effect was in countries at low risk of zinc deficiency (Analysis 1.5).

1.4. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 4 Diarrhoea duration (hours); subgrouped by continent.

1.5. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 5 Diarrhoea duration (hours): subgrouped by national risk of zinc deficiency.

Diarrhoea on days 3, 5, and 7

On average, treatment with zinc resulted in fewer children continuing to have diarrhoea at day three (risk ratio (RR) 0.77, 95% CI 0.69 to 0.86; 2063 children, 8 trials, 9 comparisons, Analysis 1.10), at day five (RR 0.76, 95% CI 0.64 to 0.91; 2307 children, eight trials, Analysis 1.11), and at day seven (RR 0.82, 95% CI 0.72 to 0.94; 5528 children, 10 trials, 13 comparisons; Analysis 1.12; Figure 5; Table 1; Table 2). For all the three outcomes there was significant statistical heterogeneity between trials.

1.10. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 10 Diarrhoea on day 3.

1.11. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 11 Diarrhoea on day 5.

1.12. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 12 Diarrhoea on day 7.

5.

Zinc versus placebo for acute diarrhoea: diarrhoea on day 7

For diarrhoea at day seven we conducted a series of subgroup analyses to explore the heterogeneity (Analysis 1.13 to Analysis 1.20), which found similar patterns as seen with duration of diarrhoea.

1.13. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 13 Diarrhoea on day 7: subgrouped by nutritional status.

1.20. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 20 Diarrhoea on day 7: subgrouped by concomitant treatment.

• No subgrouping completely explained the statistical heterogeneity.

• There was no evidence of benefit in the single trial that recruited only children less than six months of age (1074 children, 1 trial, 3 comparisons; Analysis 1.12). This was also the only trial conducted in the African continent.

• The average effect was largest in trials that only recruited children with signs of undernutrition (Analysis 1.13).

Stool frequency

There was no significant benefit of zinc on reducing stool frequency (RR −0.10, 95% CI −0.25 to 0.04; 2643 children, 7 trials, 10 comparisons; Analysis 1.21). Heterogeneity was markedly reduced if results were stratified by age, and while no benefit of zinc was detected in children under six months of age (1334 children, 2 trials, 5 comparisons), zinc had a significant benefit in children older than six months (RR −0.32, 95% CI −0.58 to −0.06; 1235 children, 4 trials) and in the trial that recruited both age groups (RR −5.90, 95% CI −9.44 to −2.36; 74 children, 1 trial; Analysis 1.21).

1.21. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 21 Stool frequency (stools /day).

Stool output

The included trials measured stool output using different units at different time points, thus we were unable to pool results (Table 6). We expressed results are expressed as arithmetic mean difference (AMD) or geometric mean ratio (GMR) values.

3. Stool output: acute diarrhoea.

Trial ID	Outcome	Units	Zinc		Placebo		Mean difference	Statistical test
			N	Values	N	Values
Age < 6 months
Brooks 2005a (5 mg)	Total (mL)	Mean (95% CI)	85	229 (180 to 256)	45	202 (180 to 246)	27 (−23.3 to 77.3)1	Not significant
Brooks 2005a (20 mg)	Total (mL)	Mean (95% CI)	86	240 (200 to 266)	44	202 (180 to 246)	38 (−8.6 to 84.6)1	Not significant
Age > 6 months
Patel 2009a (zinc)	Total (g)	Mean (95% CI)	248	972 (858 to 1087)	247	877 (728 to 1026)	−95 (−283 to 92)	Not significant
Dutta 2011	Total (L)	Mean (95% CI)	41	1.2 (0.6 to 1.8)	43	2.0 (1.3 to 2.7)	‐ 0.8 (‐1.1 to 1.5)	P < 0.0001
Dutta 2011	Per day (ml/kg/day)	Mean (95% CI)	41	51.22 (27.39 to 79.05)	43	66.83 (42.41 to 71.25)	‐ 15.61 (‐22.9 to ‐8.2)	P = 0.0001
Ages < and > 6 months
Bhatnagar 2004a	Total (g/kg)	Geometric mean (95% CI)	132	111 (86 to 147)	134	148 (116 to 190)	0.69 (0.48 to 0.99)2	P < 0.05
	Per day (g/kg/day)	Geometric mean (95% CI)	132	62 (51 to 78)	134	78 (68 to 91)	0.76 (0.59 to 0.98)2	P < 0.05
Dutta 2000	Total (kg)	Mean (95% CI)	44	1.5 (1.3 to 1.7)	36	2.4 (2.2 to 2.6)	−0.9 (−1.2 to −0.6)1	P = 0.0001
Roy 1997	Per day (g/kg/day)	Median (range)	37	238 (35 to 2416)	37	329 (32 to 1464)	−91 (NA)	P = 0.06

¹Arithmetic mean difference (95% CI) for means.
 ²Geometric mean ratio (95% CI) for geometric means, adjusted for confounders. (Stool output using zinc is 0.69 and 0.76 times that of participants using placebo, which means a 31% and 24% less stool output under zinc treatment.)

Abbreviations: CI: confidence interval; NA: not applicable.

One trial reported on children less than six months of age with no evidence of a difference (Brooks 2005a). Two trials reported on children more than six months of age with inconsistent results (Patel 2009; Dutta 2011). Three trials reported on children aged less than and greater than six months: two of these trials showed a reduction in stool output with zinc (Dutta 2000; Bhatnagar 2004a), while one trial showed no evidence of an effect (Roy 1997).

Hospitalization

Two community trials reported no hospitalizations in the zinc group and only one in the placebo group (Fischer Walker 2006, 1074 participants under six months of age; Penny 1999, 276 children over six months of age).

Death

The trials reported a low number of deaths without significant difference between the zinc group and placebo group (Analysis 1.22).

1.22. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 22 Death.

Adverse events

Vomiting was more common in those given zinc across all age groups (RR 1.54, 95% CI 1.28 to 1.85; 5942 children, 15 comparisons, 13 trials; Figure 6). There was moderate heterogeneity among trials (P = 0.005; I² statistic = 56%), and differences in control event rates (from 0.4% to 13.5%). In one large trial with adequate allocation concealment that was designed to look at safety reports, vomiting was limited to one episode in most children and mainly occurred within 10 minutes of administration (Larson 2005). Two trials found no difference in time to resolution of vomiting between zinc and placebo, although we could not pool the results (mean duration 13.63 ± 10.33 hours versus 16.35 ± 11.34 hours, P = 0.1; Dalgic 2011; median duration 2 days (interquartile range (IQR) 1 to 3) versus 2.5 days (IQR 1 to 5), P > 0.5; Crisinel 2015).

6.

Zinc versus placebo for acute diarrhoea: adverse events (vomiting)

One small trial reported a non‐statistically significant difference between the two treatment groups for difficulties in treatment administration (19/45 (45%) in the zinc group versus 20/44 (44%) in the placebo group (Crisinel 2015).

Three trials reported on copper levels, with no significant differences between the zinc and placebo groups. Two trials reported the mean change in serum copper on the last day of supplementation (seven and 14 days after recovery): −1.1 ± 5.5 µmol/dL in the zinc group versus −1.5 ± 4.2 µmol/dL in the placebo group in Strand 2002, and −41.2 ± 418.8 µg/dL in the zinc group versus −79.4 ± 429.2 µg/dL in the placebo group in Patel 2009. Mean serum copper after 14 days was 121 mg/L in zinc group versus 127 mg/L in the control in Bhatnagar 2004a.

No other side effects were reported.

Publication bias

We constructed funnel plots for trials that reported diarrhoea duration (Figure 7) and diarrhoea at day 7 (Figure 8). The funnel plots are both asymmetric due to the absence of smaller trials at the base and, for diarrhoea at day 7, also at the right of the pooled estimate. Asymmetry in the funnel plot could result from possible selection bias where smaller studies reporting greater treatment benefit for the experimental group were published (publication bias). The gap in the bottom corner of the graph suggests that smaller studies without statistically significant effects remain unpublished. However, asymmetry in the funnel plot may also be due to other reasons, such as poor methodological quality in smaller studies leading to spuriously inflated effects in the treatment effect, true heterogeneity, sampling variation or chance (Higgins 2011).

7.

Funnel plot of comparison: 1 Zinc versus placebo for children with acute diarrhoea, outcome: 1.1 Diarrhoea duration (hours).

8.

Funnel plot of comparison: 1 Zinc versus placebo for children with acute diarrhoea, outcome: 1.12 Diarrhoea on day 7.

Sensitivity analysis

The sensitivity analysis against markers of methodological quality did not affect the direction of results. There was some loss of significance with diarrhoea duration, but overall the analysis did not change the point estimate of effects. The ITT analysis for worst‐case/best‐case scenarios did not alter the statistical significance of the results.

Comparison 2: Zinc versus placebo for children with persistent diarrhoea

All trials of persistent diarrhoea enrolled children aged over six months.

Diarrhoea duration

On average, zinc supplementation reduced the duration of persistent diarrhoea by around 16 hours (MD −15.84 hours, 95% CI −25.43 to −6.24 hours; 529 children, 5 trials; Analysis 2.1), with no evidence of heterogeneity.

2.1. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 1 Diarrhoea duration (hours).

Diarrhoea on days 3, 5, and 7

There was no evidence of a benefit with zinc in the one trial that reported on diarrhoea at days three (Analysis 2.2) and five (Analysis 2.3) (Penny 1999), and two trials that reported on diarrhoea at day seven (Analysis 2.4; Penny 1999; Khatun 2001).

2.2. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 2 Diarrhoea on day 3.

2.3. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 3 Diarrhoea on day 5.

2.4. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 4 Diarrhoea on day 7.

Stool frequency

One small trial reported on stool frequency, Sachdev 1990, but the result did not reach statistical significance (40 participants, Analysis 2.5).

2.5. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 5 Stool frequency (stools/day).

Stool output

Stool output was measured using different units at different time points, thus we could not pool results (Table 7). We expressed the results as the AMD or GMR. Two trials, Bhutta 1999 and Khatun 2001, reported on children greater than six months of age, with five comparisons (Additional tables). Of these, one trial reported a significant reduction in cumulative stool output at day seven in the zinc group (AMD −338 mg/kg bodyweight, 95% CI −413.6 to −262.4 mg/kg bodyweight; P ≤ 0.001) (Khatun 2001).

4. Stool output: persistent diarrhoea.

Trial ID	Outcome	Zinc		Placebo		Mean difference 1	Statistical test
		N	Mean (95% CI)	N	Mean (95% CI)
Age > 6 months
Bhutta 1999	Per day of diarrhoea, day 1 (g/kg/day)	43	116.8 (85.8 to 147.8)	44	141.9 (91.2 to 192.6)	−25.1 (−84.5 to 34.3)	Not significant
	Per day of diarrhoea, day 7 (g/kg/day)	43	66.7 (40.9 to 92.4)	44	43.9 (32.1 to 55.7)	22.8 (−5.5 to 51.1)	Not significant
	Per day of diarrhoea, day 14 (g/kg/day)	43	24.9 (20.1 to 29.7)	44	27.8 (18.5 to 37.1)	‐2.9 (−13.4 to 7.6)	Not significant
Khatun 2001	Cumulative day 1 (mg/kg)	24	127 (113 to 141)	24	137 (121 to 153)	−10 (−31.6 to 11.6)	Not significant
	Cumulative day 7 (mg/kg)	24	528 (472 to 584)	24	866 (815 to 917)	−338 (−413.6 to −262.4)	P ≤ 0.001

¹Arithmetic mean difference (95% CI) for means.

Abbreviations: CI: confidence interval; ID: identification.

Hospitalization

The only community trial that reported on hospitalization did not observe any hospitalizations in the zinc or placebo group (Penny 1999; 275 participants).

Death

One trial reported one death in the zinc group compared to five deaths in the placebo group, out of 95 participants in each group (Roy 1998). Two trials did not observe deaths in any participants, irrespective of their allocated group (Penny 1999; Khatun 2001).

Adverse events

Four trials that reported on vomiting (505 children) showed no difference between the zinc and placebo groups (Analysis 2.6): three of the trials reported no incidences of vomiting in either group (Khatun 2001; Roy 1998; Sachdev 1990); one trial that used 3 mg/kg/day zinc for 14 days in moderately malnourished and severely malnourished children reported a significantly lower plasma copper levels in the zinc‐treated group by the end of the second week of therapy (56.2 ± 17.8 µg/dL versus 72.7 ± 18.3 µg/dL, P = 0.02; Bhutta 1999, 87 children).

2.6. Analysis.

Comparison 2 Zinc versus placebo for children with persistent diarrhoea, Outcome 6 Adverse events (vomiting).

Statistical heterogeneity

There was heterogeneity between two trials for diarrhoea at day seven. This may be explained by differences in the geographical regions (India and Peru) or to other factors not explored in this Cochrane Review. Reporting of vomiting was heterogeneous between trials, and this may be due to difference in the population or in the definition of event, or to reporting bias.

Sensitivity analysis

The sensitivity analyses did not affect the direction of results. There was some loss of significance with diarrhoea duration, but no changes in the point estimate of effects. An ITT analysis for worst‐case/best‐case scenarios did not alter the point estimate or the significance of results.

Discussion

Summary of main results

Thirty‐three trials, enrolling 10,841 children, met our inclusion criteria. Most included trials were conducted in Asian countries where the risk of zinc deficiency is high.

Acute diarrhoea

There is currently not enough evidence from well conducted trials to be able to say whether zinc supplementation during acute diarrhoea reduces the number of deaths or the number of children hospitalized (very low certainty evidence).

In children aged greater than six months, zinc supplementation may shorten the duration of diarrhoea by around half a day (low certainty evidence), and probably reduces the number of children whose diarrhoea persists until day seven (moderate certainty evidence). In children with signs of moderate malnutrition the effect appears greater, reducing the duration of diarrhoea by around a day (high certainty evidence).

Conversely, in children less than six months of age, the available evidence suggests zinc supplementation may have no impact on the duration of diarrhoea (low certainty evidence), and may increase the proportion of children whose diarrhoea persists until day seven (low certainty evidence).

No trials reported serious adverse events, but zinc supplementation during acute diarrhoea causes vomiting in both age groups (moderate certainty evidence).

Persistent diarrhoea

In children with persistent diarrhoea, zinc supplementation probably shortens the duration of diarrhoea by around 16 hours (moderate certainty evidence).

Overall completeness and applicability of evidence

This Cochrane Review showed that zinc overall reduced the duration of acute diarrhoea. However, most trials were conducted in populations with moderate to high risk of zinc deficiency (Asia, Africa, children over six months of age and with some degree of malnutrition). Transferability of these results to other countries is therefore likely to depend on local risk of zinc deficiency and other population characteristics such as the degree of malnutrition and breastfeeding habits. The few trials conducted in populations at low risk of zinc deficiency, namely well‐nourished children in countries and continents where zinc deficiency is uncommon (Europe, Australia), overall showed no benefit of zinc.

Most trials were conducted in hospital where participants are more likely to adhere to the intervention, but some community trials also showed a benefit with zinc, which suggests that zinc could be used both at hospital and at community level.

The observed increase in vomiting was consistent across trials in all age groups with one large trial reporting that vomiting was limited to one episode in most children and mainly occurring within 10 minutes of administration (Larson 2005). Zinc has a metallic after‐taste, and development of a more palatable formulation may minimize this adverse effect.

Quality of the evidence

We assessed the certainty of the evidence using the GRADE methodology and displayed it in three 'Summary of findings' tables (Table 1; Table 2; Table 3).

In general, the methodological quality of the trials included in this review was good.

The evidence for benefits on diarrhoea duration in children aged greater than six months of age is of low to moderate certainty. This implies that we can have some confidence in the results but further research may alter the estimates of benefit and harm. The main reasons to downgrade were 'quality of trials' and 'inconsistency' in the results. Heterogeneity between trials was often high. This is perhaps not surprising given the variations in populations, settings, and interventions. We were unable to completely explain this heterogeneity through subgroup analysis, and so our confidence that zinc supplementation can be broadly applied was decreased.

Most trials were conducted in hospitals where death rates were low, and were consequently not powered to detect an effect on mortality. Large community trials are needed to explore whether zinc treatment for diarrhoea reduces hospitalization and death.

Potential biases in the review process

We attempted to limit bias by following the rigorous methods provided by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We conducted an extensive search for studies, including ongoing studies. We only included peer‐reviewed trials in this review. Two review authors independently scrutinized the studies, assessed them for eligibility, extracted data, inserted data into RevMan 5 (Review Manager 5), and double checked the final version of the review. The findings of the funnel plots may suggest publication bias. However, asymmetry in the funnel plots may also be due to other reasons, such as poor methodological quality in smaller studies leading to spuriously inflated effects, true heterogeneity in the treatment effect, sampling variation, or chance (Higgins 2011).

Agreements and disagreements with other studies or reviews

Our results agree with those of other systematic reviews of zinc for treating children that have diarrhoea (Bhutta 2000b; Lukacik 2008; Patro 2008; Haider 2009; Liberato 2015; Zou 2015), except for the finding of no effect of zinc in children aged less than six months, and in populations at low risk of zinc deficiency. Compared to the other recent reviews (Liberato 2015; Zou 2015; Lazzerini 2016), this Cochrane Review includes several new trials, includes a more extensive subgroup analysis, and reports on diarrhoea at different time points, diarrhoea severity, death, and adverse events.

The results of this Cochrane Review in children over six months of age support the current WHO/UNICEF policy to give zinc to children with diarrhoea (WHO/UNICEF 2004), while currently there are no evidence from randomized controlled trials to provide zinc in children younger than six months of age.

Authors' conclusions

Implications for practice.

In areas where diarrhoea is an important cause of child mortality, and the prevalence of zinc deficiency or mild/moderate malnutrition is high, zinc may be of benefit in children with diarrhoea aged six months or more.

Implications for research.

Causes of heterogeneity in the effect of zinc in children over six months should be further explored, and further research is necessary to justify continued supplementation in children less than six months of age and in children with low risk of zinc deficiency.

What's new

Date	Event	Description
25 April 2017	Amended	We have amended the GRADE assessment for two outcomes, duration of diarrhea and diarrhea on day 7 for children aged less than 6 months with acute diarrhoea and treated with zinc, which we had scored incorrectly in the review update. We had stated that these outcomes had a GRADE score of moderate. However, we had downgraded the certainty of the evidence by 2, and thus the certainty of the evidence was low. We have amended the certainty of the evidence for these two outcomes to low in 'Summary of findings' table 2 and the review text.

History

Protocol first published: Issue 3, 2005
 Review first published: Issue 3, 2008

Date	Event	Description
19 December 2016	New citation required but conclusions have not changed	We included 33 trials in total in this review update, of which nine were new trials. We updated the 'Summary of findings' tables according to the GRADE approach, and included a PRISMA study flow diagram and funnel plots.
19 December 2016	New search has been performed	We amended the author team. Luca Ronfani stepped down as an author, and Humphrey Wanzira joined as an author. We updated the literature search to 30 September 2016, and nine new trials met the inclusion criteria of the review update.
6 December 2012	New citation required but conclusions have not changed	We corrected the Abstract.
6 December 2012	Amended	An error was spotted in the abstract (number of participants and number of studies was incorrect). We have corrected this and republished the review to ensure the correct details are documented.
22 March 2012	New search has been performed	We updated the search on 20 February 2012, and included two new trials. We updated the Background and undertook a more detailed assessment of the risk of bias in all included trials. We updated the 'Summary of findings' tables according to the GRADE methodology.
22 March 2012	New citation required but conclusions have not changed	Update.
11 February 2011	New search has been performed	We updated the search on 1 December 2010, and included four new trials. We updated the Background and performed a more detailed assessment of the risk of bias in all included trials. We added 'Summary of findings' tables according to the GRADE methodology.

Data and analyses

Comparison 1. Zinc versus placebo for children with acute diarrhoea.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Diarrhoea duration (hours)	24	5096	Mean Difference (IV, Random, 95% CI)	‐13.42 [‐20.52, ‐6.31]
1.1 Age < 6 months	5	1334	Mean Difference (IV, Random, 95% CI)	5.23 [‐2.00, 14.45]
1.2 Age > 6 months	9	2581	Mean Difference (IV, Random, 95% CI)	‐11.46 [‐19.72, ‐3.19]
1.3 Ages both < and > 6 months	10	1181	Mean Difference (IV, Random, 95% CI)	‐22.18 [‐32.57, ‐11.78]
2 Diarrhoea duration (hours): subgrouped by nutritional status	17	3518	Mean Difference (IV, Random, 95% CI)	‐17.54 [‐25.49, ‐9.58]
2.1 Nutritional status: only well‐nourished	2	406	Mean Difference (IV, Random, 95% CI)	‐6.79 [‐23.84, 10.26]
2.2 Nutritional status: well‐nourished plus moderately malnourished	10	2693	Mean Difference (IV, Random, 95% CI)	‐15.46 [‐25.55, ‐5.36]
2.3 Nutritional status: malnourished	5	419	Mean Difference (IV, Random, 95% CI)	‐26.39 [‐36.54, ‐16.23]
3 Diarrhoea duration (hours): subgrouped by sex	18	3621	Mean Difference (IV, Random, 95% CI)	‐17.33 [‐25.03, ‐9.62]
3.1 Sex: male	3	430	Mean Difference (IV, Random, 95% CI)	‐22.35 [‐36.40, ‐8.31]
3.2 Sex: male and female	15	3191	Mean Difference (IV, Random, 95% CI)	‐16.13 [‐24.71, ‐7.55]
4 Diarrhoea duration (hours); subgrouped by continent	18	3621	Mean Difference (IV, Random, 95% CI)	‐17.33 [‐25.03, ‐9.62]
4.1 Continent: Africa	1	60	Mean Difference (IV, Random, 95% CI)	‐2.40 [‐33.25, 28.45]
4.2 Continent: Asia	13	3205	Mean Difference (IV, Random, 95% CI)	‐19.01 [‐28.19, ‐9.82]
4.3 Continent: South America	1	74	Mean Difference (IV, Random, 95% CI)	‐31.20 [‐46.43, ‐15.97]
4.4 Continent: Europe	2	224	Mean Difference (IV, Random, 95% CI)	‐10.19 [‐34.29, 13.91]
4.5 Continent: Australia	1	58	Mean Difference (IV, Random, 95% CI)	‐2.40 [‐20.93, 16.13]
5 Diarrhoea duration (hours): subgrouped by national risk of zinc deficiency	16	3253	Mean Difference (IV, Random, 95% CI)	‐16.99 [‐25.49, ‐8.50]
5.1 Region: countries ranked as high risk of zinc deficiency	8	2535	Mean Difference (IV, Random, 95% CI)	‐14.97 [‐26.21, ‐3.72]
5.2 Region: countries ranked as medium risk of zinc deficiency	5	436	Mean Difference (IV, Random, 95% CI)	‐25.92 [‐44.80, ‐7.04]
5.3 Region: countries ranked as low risk of zinc deficiency	3	282	Mean Difference (IV, Random, 95% CI)	‐7.63 [‐22.74, 7.48]
6 Diarrhoea duration (hours): subgrouped by zinc dose	13	2018	Mean Difference (IV, Random, 95% CI)	‐20.24 [‐28.84, ‐11.63]
6.1 Zinc dose:≤ 20 mg	9	976	Mean Difference (IV, Random, 95% CI)	‐18.45 [‐30.19, ‐6.71]
6.2 Zinc dose: > 20 mg	4	1042	Mean Difference (IV, Random, 95% CI)	‐23.33 [‐38.30, ‐8.35]
7 Diarrhoea duration (hours): subgrouped by zinc type	16	3454	Mean Difference (IV, Random, 95% CI)	‐16.50 [‐25.11, ‐7.89]
7.1 Zinc type: zinc acetate	3	875	Mean Difference (IV, Random, 95% CI)	‐30.55 [‐49.29, ‐11.82]
7.2 Zinc type: gluconate	2	908	Mean Difference (IV, Random, 95% CI)	‐14.51 [‐30.84, 1.81]
7.3 Zinc type: zinc sulphate	11	1671	Mean Difference (IV, Random, 95% CI)	‐13.21 [‐24.16, ‐2.27]
8 Diarrhoea duration (hours): subgrouped by study setting	18	3621	Mean Difference (IV, Random, 95% CI)	‐17.33 [‐25.03, ‐9.62]
8.1 Study setting: hospital	15	2468	Mean Difference (IV, Random, 95% CI)	‐17.86 [‐27.01, ‐8.70]
8.2 Study setting:community	3	1153	Mean Difference (IV, Random, 95% CI)	‐12.65 [‐21.76, ‐3.54]
9 Diarrhoea duration (hours): subgrouped by concomitant treatment	18	3777	Mean Difference (IV, Random, 95% CI)	‐15.68 [‐23.53, ‐7.82]
9.1 Concomitant treatment: zinc alone	17	3394	Mean Difference (IV, Random, 95% CI)	‐16.95 [‐24.85, ‐9.05]
9.2 Concomitant treatment: zinc plus copper	1	383	Mean Difference (IV, Random, 95% CI)	2.20 [‐5.08, 9.48]
10 Diarrhoea on day 3	9	2063	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.69, 0.86]
10.1 Age > 6 months	4	1599	Risk Ratio (M‐H, Fixed, 95% CI)	0.83 [0.72, 0.94]
10.2 Ages both < and > 6 months	5	464	Risk Ratio (M‐H, Fixed, 95% CI)	0.66 [0.56, 0.79]
11 Diarrhoea on day 5	8	2307	Risk Ratio (M‐H, Fixed, 95% CI)	0.76 [0.64, 0.91]
11.1 Age > 6 months	3	1384	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.52, 1.01]
11.2 Ages both < and > 6 months	5	923	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.64, 0.96]
12 Diarrhoea on day 7	13	5528	Risk Ratio (M‐H, Fixed, 95% CI)	0.82 [0.72, 0.94]
12.1 Age < 6 months	3	1074	Risk Ratio (M‐H, Fixed, 95% CI)	1.24 [0.99, 1.54]
12.2 Age > 6 months	6	3865	Risk Ratio (M‐H, Fixed, 95% CI)	0.73 [0.61, 0.88]
12.3 Ages both < and > 6 months	4	589	Risk Ratio (M‐H, Fixed, 95% CI)	0.31 [0.18, 0.52]
13 Diarrhoea on day 7: subgrouped by nutritional status	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
13.1 Nutritional status: only well‐nourished	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.04, 3.26]
13.2 Nutritional status: well‐nourished plus moderately malnourished	6	4075	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.60, 0.86]
13.3 Nutritional status: malnourished	3	238	Risk Ratio (M‐H, Fixed, 95% CI)	0.37 [0.22, 0.61]
14 Diarrhoea on day 7: subgrouped by sex	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
14.1 Sex: male	1	266	Risk Ratio (M‐H, Fixed, 95% CI)	0.11 [0.01, 0.88]
14.2 Sex: male and female	9	4188	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.58, 0.81]
15 Diarrhoea on day 7: subgrouped by continent	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
15.1 Region: Asia	9	4313	Risk Ratio (M‐H, Fixed, 95% CI)	0.67 [0.56, 0.79]
15.2 Region: Europe	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.04, 3.26]
16 Diarrhoea on day 7: subgrouped by national risk of zinc deficiency	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
16.1 Region: countries ranked as high risk of zinc deficiency	6	3240	Risk Ratio (M‐H, Fixed, 95% CI)	0.75 [0.62, 0.92]
16.2 Region: countries ranked as medium risk of zinc deficiency	3	1073	Risk Ratio (M‐H, Fixed, 95% CI)	0.49 [0.35, 0.68]
16.3 Region: countries ranked as low risk of zinc deficiency	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.04, 3.26]
17 Diarrhoea on day 7: subgrouped by zinc dose	9		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
17.1 Zinc dose: 20 mg	8	3154	Risk Ratio (M‐H, Fixed, 95% CI)	0.62 [0.51, 0.74]
17.2 Zinc dose: >20mg	1	805	Risk Ratio (M‐H, Fixed, 95% CI)	0.67 [0.38, 1.19]
18 Diarrhoea on day 7: subgrouped by zinc type	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
18.1 Zinc type: zinc acetate	3	1628	Risk Ratio (M‐H, Fixed, 95% CI)	0.60 [0.45, 0.79]
18.2 Zinc type:gluconate	1	805	Risk Ratio (M‐H, Fixed, 95% CI)	0.67 [0.38, 1.19]
18.3 Zinc type: zinc sulphate	6	2021	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.57, 0.90]
19 Diarrhoea on day 7: subgrouped by study setting	10		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
19.1 Study setting: hospital	8	2758	Risk Ratio (M‐H, Fixed, 95% CI)	0.69 [0.56, 0.84]
19.2 Study setting: community	2	1696	Risk Ratio (M‐H, Fixed, 95% CI)	0.61 [0.44, 0.85]
20 Diarrhoea on day 7: subgrouped by concomitant treatment	11		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
20.1 Concomitant treatment: zinc alone	10	4330	Risk Ratio (M‐H, Fixed, 95% CI)	0.65 [0.55, 0.78]
20.2 Concomitant treatment: zinc plus copper	1	383	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.43, 2.45]
21 Stool frequency (stools /day)	10	2643	Mean Difference (IV, Fixed, 95% CI)	‐0.10 [‐0.25, 0.04]
21.1 Age < 6 months	5	1334	Mean Difference (IV, Fixed, 95% CI)	0.0 [‐0.17, 0.17]
21.2 Age > 6 months	4	1235	Mean Difference (IV, Fixed, 95% CI)	‐0.32 [‐0.58, ‐0.06]
21.3 Ages both < and > 6 months	1	74	Mean Difference (IV, Fixed, 95% CI)	‐5.9 [‐9.44, ‐2.36]
22 Death	8	2609	Risk Ratio (M‐H, Fixed, 95% CI)	0.31 [0.09, 1.07]
22.1 Age < 6 months	2	1334	Risk Ratio (M‐H, Fixed, 95% CI)	1.00 [0.06, 15.89]
22.2 Age > 6 months	5	1134	Risk Ratio (M‐H, Fixed, 95% CI)	0.29 [0.04, 2.20]
22.3 Ages both < and > 6 months	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.19 [0.02, 1.55]
23 Adverse events (vomiting)	15	5942	Risk Ratio (M‐H, Random, 95% CI)	1.54 [1.28, 1.85]
23.1 Age < 6 months	3	1334	Risk Ratio (M‐H, Random, 95% CI)	1.54 [1.05, 2.24]
23.2 Age > 6 months	6	2605	Risk Ratio (M‐H, Random, 95% CI)	1.57 [1.32, 1.86]
23.3 Ages both < and > 6 months	6	2003	Risk Ratio (M‐H, Random, 95% CI)	1.63 [1.14, 2.34]
24 Difficulties in treatment administration	1	87	Odds Ratio (M‐H, Fixed, 95% CI)	1.03 [0.44, 2.41]

1.3. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 3 Diarrhoea duration (hours): subgrouped by sex.

1.6. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 6 Diarrhoea duration (hours): subgrouped by zinc dose.

1.7. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 7 Diarrhoea duration (hours): subgrouped by zinc type.

1.8. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 8 Diarrhoea duration (hours): subgrouped by study setting.

1.14. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 14 Diarrhoea on day 7: subgrouped by sex.

1.15. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 15 Diarrhoea on day 7: subgrouped by continent.

1.16. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 16 Diarrhoea on day 7: subgrouped by national risk of zinc deficiency.

1.17. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 17 Diarrhoea on day 7: subgrouped by zinc dose.

1.18. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 18 Diarrhoea on day 7: subgrouped by zinc type.

1.19. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 19 Diarrhoea on day 7: subgrouped by study setting.

1.23. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 23 Adverse events (vomiting).

1.24. Analysis.

Comparison 1 Zinc versus placebo for children with acute diarrhoea, Outcome 24 Difficulties in treatment administration.

Comparison 2. Zinc versus placebo for children with persistent diarrhoea.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Diarrhoea duration (hours)	5	529	Mean Difference (IV, Fixed, 95% CI)	‐15.84 [‐25.43, ‐6.24]
1.1 Age > 6 months	4	388	Mean Difference (IV, Fixed, 95% CI)	‐16.01 [‐26.16, ‐5.86]
1.2 Ages both < and > 6 months	1	141	Mean Difference (IV, Fixed, 95% CI)	‐14.40 [‐43.77, 14.97]
2 Diarrhoea on day 3	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
2.1 Age > 6 months	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
3 Diarrhoea on day 5	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
3.1 Age > 6 months	1		Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4 Diarrhoea on day 7	2	221	Risk Ratio (M‐H, Fixed, 95% CI)	0.52 [0.27, 1.02]
4.1 Age > 6 months	2	221	Risk Ratio (M‐H, Fixed, 95% CI)	0.52 [0.27, 1.02]
5 Stool frequency (stools/day)	1		Mean Difference (IV, Fixed, 95% CI)	Totals not selected
5.1 Age > 6 months	1		Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
6 Adverse events (vomiting)	4	505	Risk Ratio (M‐H, Fixed, 95% CI)	1.97 [0.37, 10.59]
6.1 Age > 6 months	3	364	Risk Ratio (M‐H, Fixed, 95% CI)	1.97 [0.37, 10.59]
6.2 Ages both < and > 6 months	1	141	Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Al‐Sonboli 2003.

Methods	Randomized controlled trial (RCT)
Participants	Number of participants (N): 81 participants Inclusion criteria: age 3 to 60 months; diarrhoea < 7 days, or 1 or more loose stool containing blood in the previous 24 hours and at least mild dehydration Exclusion criteria: suspected or confirmed severe systemic infections; antimicrobial or antidiarrhoeal treatment within 72 hours before admission; severe malnutrition (< 60% median for weight for age of the National Center for Health Statistic (NCHC) standards)
Interventions	Zinc sulphate: 22.5 mg (3 to 6 months) or 45 mg (7 to 60 months). Placebo.
Outcomes	Average duration of diarrhoea. Stool frequency.
Notes	Location: Brazil Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used a random numbers table to randomize participants.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not provide any details regarding allocation concealment.
Blinding (performance bias and detection bias) All outcomes	Low risk	This trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	8.6% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available.

Bahl 2002.

Methods	RCT
Participants	N: 1219 participants Inclusion criteria: age 6 to 35 months; acute diarrhoea (less than 4 days duration) Exclusion criteria: visible blood in stools; likely to emigrate in the next 4 weeks; required hospitalization; previously enrolled; sibling concurrently enrolled; refusal of consent
Interventions	Zinc gluconate 30 mg (≥ 12 months) or 15 mg (< 12 months). Placebo.
Outcomes	Average duration of diarrhoea. Diarrhoea at day 3. Diarrhoea at day 5. Diarrhoea at day 7. Stool frequency. Adverse events (vomiting).
Notes	Location: India Setting: community
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	This trial used computer‐generated randomization lists.
Allocation concealment (selection bias)	Low risk	An independent individual who was not involved in participant enrolment labelled the glass bottles that contained the products with the participant’s number that corresponded to the randomization list. Randomization codes were secured until the completion of data collection and initial analysis. There was no difference between zinc and the placebo in appearance; a minor metallic aftertaste of zinc was hardly detectable.
Blinding (performance bias and detection bias) All outcomes	Low risk	Four‐blinded (participant, intervention provider, data collector, data analyst).
Incomplete outcome data (attrition bias) All outcomes	Low risk	2% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no available protocol.
Other bias	Unclear risk	There was no further information available.

Bhatnagar 2004a.

Methods	RCT
Participants	N: 287 participants Inclusion criteria: male; 3 to 36 months; acute diarrhoea (< 72 hours) with mild dehydration Exclusion criteria: severe malnutrition (weight/height < 65% of NCHS median); visible blood in stool; severe systemic illness
Interventions	Zinc sulphate: 15 mg (< 12 months) or 30 mg (> 12 months) syrup. Placebo. Both groups: multivitamin
Outcomes	Average duration of diarrhoea. Diarrhoea at day 5. Diarrhoea at day 7. Stool output. Adverse events (vomiting). Adverse events (copper levels).
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial authors used a table of random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Central randomization was performed at a site remote from trial location (World Health Organization (WHO), Geneva).
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	7% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no available protocol.
Other bias	Unclear risk	There was no further information available on other risks of bias.

Bhutta 1999.

Methods	RCT
Participants	N: 87 participants Inclusion criteria: 6 to 36 months; persistent diarrhoea (> 4 unformed stools/day for at least 14 days); malnutrition (weight‐for‐age z score < −2.0) Exclusion criteria: kwashiorkor; clinical signs of vitamin A or zinc deficiency; needing intravenous fluids or unable to tolerate oral feeds after a 24‐hour period of stabilization
Interventions	Zinc sulphate: 3 mg/kg/day. Placebo. Both groups: multivitamins
Outcomes	Average duration of diarrhoea. Stool output. Adverse events (copper levels).
Notes	Location: Pakistan Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random numbers table.
Allocation concealment (selection bias)	Low risk	An independent pharmacy performed central randomization; the pharmacy maintained the table block randomization.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	11% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available.

Boran 2006.

Methods	RCT
Participants	N: 280 participants Inclusion criteria: acute diarrhoea of <14 days presenting at the paediatric emergency and outpatient clinic Exclusion criteria: refusal of consent, malnutrition, medical condition requiring hospitalization, received anti‐diarrhoea medication or antibiotics
Interventions	3 RDA zinc sulphate in a syrup once daily (15 mg zinc for 6 to 12 months children and 30 mg for 12 to 60 months) for 14 days + ORS. ORS.
Outcomes	Duration of diarrhoea. Adverse events (vomiting).
Notes	We requested additional information from the trial author, but did not receive any reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used block randomization with 8 numbers in each block.
Allocation concealment (selection bias)	Unclear risk	The trial did not mention these details.
Blinding (performance bias and detection bias) All outcomes	High risk	There was no blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Fifteen participants (5.36%) were lost to follow‐up
Selective reporting (reporting bias)	Unclear risk	The trial authors did not include the RCT protocol registration number.
Other bias	Unclear risk	There was no further information available on other sources of bias.

Brooks 2005a.

Methods	RCT
Participants	N: 275 participants Inclusion criteria: male, 1 to 6 months; onset < 72 hours; some dehydration or > 100 mL of watery stool within a 4‐hour observation period Exclusion criteria: clinical signs of zinc deficiency; kwashiorkor, weight/age < 60% NCHS; grossly bloody stool comorbidity; cholera
Interventions	Zinc acetate: 20 mg. Zinc acetate: 5 mg. Placebo.
Outcomes	Death. Average duration of diarrhoea. Stool output. Stool frequency. Adverse events (vomiting).
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used a random numbers table to randomize participants to treatment.
Allocation concealment (selection bias)	Unclear risk	The trial used bottles labelled with randomization numbers; but did not provide any other details.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	5% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no information available on other sources of bias.

Brooks 2005a (20 mg).

Methods	See Brooks 2005a
Participants	N: 91 participants (5% lost to follow‐up)
Interventions	Zinc acetate: 20 mg. Placebo.
Outcomes	See Brooks 2005a
Notes	See Brooks 2005a
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Brooks 2005a for all descriptions.
Allocation concealment (selection bias)	Unclear risk	See Brooks 2005a for all descriptions.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Brooks 2005a for all descriptions.
Incomplete outcome data (attrition bias) All outcomes	Low risk	5% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	See Brooks 2005a for all descriptions.
Other bias	Unclear risk	See Brooks 2005a for all descriptions.

Brooks 2005a (5 mg).

Methods	See Brooks 2005a
Participants	N: 91 participants (7% lost to follow‐up)
Interventions	Zinc acetate: 5 mg. Placebo.
Outcomes	See Brooks 2005a
Notes	See Brooks 2005a
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Brooks 2005a for all descriptions.
Allocation concealment (selection bias)	Unclear risk	See Brooks 2005a for all descriptions.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Brooks 2005a for all descriptions.
Incomplete outcome data (attrition bias) All outcomes	Low risk	7% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	See Brooks 2005a for all descriptions.
Other bias	Unclear risk	See Brooks 2005a for all descriptions.

Crisinel 2015.

Methods	RCT
Participants	N = 148 participants Inclusion criteria: children 2 months to 5 years of age, acute diarrhoea (3 or more stools a day for < 72 hours) at emergency department Exclusion criteria: severe malnutrition (−3 standard deviations (SDs)), ongoing zinc treatment, overwhelming chronic medical condition, non‐French speaking parents, hypersensitivity to component of zinc or placebo, phenylketonuria
Interventions	Zinc tablets of 10 mg (children < 6 months) or 20 mg (children ≥ 6 months) once a day for 10 days plus ORS. Placebo plus ORS.
Outcomes	Diarrhoea at day 3 and day 5. Adverse events (vomiting, difficulties in treatment administration).
Notes	Results recorded as medians with IQR because data was not normally distributed.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	WHO performed block randomization.
Allocation concealment (selection bias)	Low risk	An institutional pharmacy assigned a study number to each package of zinc or placebo.
Blinding (performance bias and detection bias) All outcomes	Low risk	The same packaging and dosage were used for the intervention and control groups.
Incomplete outcome data (attrition bias) All outcomes	High risk	Over 40% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	This RCT was not registered.
Other bias	Unclear risk	No information.

Dalgic 2011.

Methods	RCT
Participants	N: 120 participants Inclusion criteria: 1 to 28 months and, on admission, stool positive for rotavirus antigen.   Exclusion criteria: severe malnutrition (weight for height < −3SD as for WHO standards): duration of diarrhoea > 96 hours; severe dehydration; exclusively breast‐feeding; toxic clinical appearance; immunosuppression; any known allergies to any drugs or foods.
Interventions	Zinc 20 mg/day. Placebo.
Outcomes	Average duration of diarrhoea. Hospitalization.
Notes	Location: Turkey Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment was not specified.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors stated the trial was "single blind", but did not provide further details.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All children completed the study.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available.

Dutta 2000.

Methods	RCT
Participants	N: 80 participants Inclusion criteria: male, 3 to 24 months; malnourished (< 80% Harvard Standard weight for age); clinical signs of dehydration Exclusion criteria: antibiotics; systemic infections; chronic diseases; need for intensive care; exclusively breastfed
Interventions	Zinc sulphate: 40 mg/day. Placebo.
Outcomes	Average duration of diarrhoea. Diarrhoea at day 5. Stool output.
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The trial kept code numbers in a sealed envelope; zinc and placebo bottles were identical.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The trial did not specify the number of participants lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available on other sources of bias.

Dutta 2011.

Methods	RCT
Participants	N: 84 participants Inclusion criteria: age 6 to 24 months, history of acute watery diarrhoea, moderate dehydration Exclusion criteria: severe malnutrition (weight on height < −3SD WHO reference); systemic illness; chronic underlying disease (for example, tuberculosis, liver diseases) or needing intensive care; exclusively breastfed; antibiotics before enrolment or vitamin A within the previous 6 months
Interventions	Zinc 20 mg/day. Placebo.
Outcomes	Average duration of diarrhoea. Diarrhoea at day 5. Stool output.
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The trial kept code numbers in a sealed envelope; zinc and placebo bottles were identical.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	The number of participants lost to follow‐up was < 10%.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available on sources of bias.

Fajolu 2008.

Methods	RCT
Participants	N: 60 participants Inclusion criteria: age 6 to 24 months; acute diarrhoea (less than 14 days duration) Exclusion criteria: refusal of consent; protein energy malnutrition; use of stool hardeners, antimotility drugs ant antibiotics; other medical condition requiring hospitalization
Interventions	Zinc sulphate 20 mg (>12 months) or 10 mg (< 12 months). Placebo.
Outcomes	Average duration of diarrhoea. Stool frequency.
Notes	Location: Nigeria Setting: hospital (follow‐up in the community)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial authors did not provide these details.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not provide these details.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors did not provide these details.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The trial authors did not specify the number of participants lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available on other sources of bias.

Faruque 1999.

Methods	RCT
Participants	N: 684 participants Inclusion criteria: children 6 to 24 months with acute diarrhoea, some dehydration and no severe dehydration; underweight or stunted children were not excluded Exclusion criteria: marasmus; kwashiorkor; systemic illnesses
Interventions	Zinc acetate: 14.2 mg (first 417 children) or 40 mg (other 273 children randomized). Placebo. Both groups: vitamin A
Outcomes	Average duration of diarrhoea. Diarrhoea at day 7.
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The trial used bottles serially numbered according to the randomization schedule to correspond to the serial number of the participant; a pharmaceutical company prepared the supplements and provided them in dark‐coloured bottles.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	4% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no information available on other sources of bias.

Fischer Walker 2006.

Methods	RCT
Participants	N: 1110 participants Inclusion criteria: infants 1 to 5 months of age with acute diarrhoea (< 72 hours) Exclusion criteria: severe malnutrition (< −3 z score weight for age); signs of pneumonia if < 2 months (cough and difficult or fast breathing with a respiratory rate of > 60 breaths/minute); signs severe pneumonia if 2 to 5 months of age (cough or difficult fast breathing and chest indrawing, nasal flaring, or grunting); required hospitalization (overnight stay at a healthcare facility) for any reason; known major congenital malformation; any other serious pre‐existing medical condition; lived out of or planned to move out of study area within following 3 months; previously enrolled in the study
Interventions	Zinc sulphate: 10 mg. Placebo.
Outcomes	Death. Average duration of diarrhoea. Diarrhoea at day 7. Stool frequency. Hospitalization. Adverse events (vomiting).
Notes	Location: Ethiopia, India, and Pakistan Setting: community
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The trial assigned the randomization scheme in Geneva and kept it secure until completion of data collection and initial analysis; upon enrolment, infants were assigned chronological study identifiers corresponding to a prelabelled blister pack of either zinc or placebo tablets.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	3% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available on other sources of bias.

Fischer Walker 2006 ETH.

Methods	See Fischer Walker 2006
Participants	N: 177 participants (8% lost at follow‐up)
Interventions	See Fischer Walker 2006
Outcomes	See Fischer Walker 2006
Notes	Location: Ethiopia
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Fischer Walker 2006.
Allocation concealment (selection bias)	Low risk	See Fischer Walker 2006.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Fischer Walker 2006.
Incomplete outcome data (attrition bias) All outcomes	Low risk	8% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	See Fischer Walker 2006.
Other bias	Unclear risk	See Fischer Walker 2006.

Fischer Walker 2006 IND.

Methods	See Fischer Walker 2006
Participants	N: 373 participants (1% lost to follow‐up)
Interventions	See Fischer Walker 2006
Outcomes	See Fischer Walker 2006
Notes	Location: India
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Fischer Walker 2006.
Allocation concealment (selection bias)	Low risk	See Fischer Walker 2006.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Fischer Walker 2006.
Incomplete outcome data (attrition bias) All outcomes	Low risk	1% participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	See Fischer Walker 2006.
Other bias	Unclear risk	See Fischer Walker 2006.

Fischer Walker 2006 PAK.

Methods	See Fischer Walker 2006
Participants	N: 560 participants (3% lost to follow‐up)
Interventions	See Fischer Walker 2006
Outcomes	See Fischer Walker 2006
Notes	Location: Pakistan
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Fischer Walker 2006.
Allocation concealment (selection bias)	Low risk	See Fischer Walker 2006.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Fischer Walker 2006.
Incomplete outcome data (attrition bias) All outcomes	Low risk	3% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	See Fischer Walker 2006.
Other bias	Unclear risk	See Fischer Walker 2006.

Jiang 2016.

Methods	RCT
Participants	N: 103 participants Inclusion criteria: children diagnosed with acute diarrhoea,duration within 48 hours, age 3 months to 3 years, rotavirus enteritis (colloidal gold method should be used to detect RV antigen expression) and informed consent Exclusion criteria: mucous and bloody stool, stool routine shows white blood cells > 5 /high power field or red blood cells > 5 /high power field, total white blood cells > 12 × 109/L, C reactive protein > 10 mg/L, children with some underlying diseases such as congenital heart disease, hepatopathy and epilepsy
Interventions	Zinc gluconate granules (10mg) in children 3 to 6 months, and 20 mg in children over 6 months. Microecologic products with some extra treatments for myocardial nutrients, protect liver, relieve cough, reduce phlegm, and improve microcirculation (vitamin C) for children with abnormal laboratory indexes.
Outcomes	Duration of diarrhoea. Diarrhoea at day 3.
Notes	None
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial authors did not mention how allocation sequence was generated.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not mention how allocation concealment was conducted.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors did not mention how blinding was achieved for participants, intervention providers, and assessors.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There was no loss to follow‐up of participants.
Selective reporting (reporting bias)	Unclear risk	There was no number of protocol registration.
Other bias	Unclear risk	The trial was not registered.

Jin 2013.

Methods	RCT
Participants	N: 103 participants Inclusion criteria: 4 to 40 months of age; infants with acute rotavirus diarrhoea; parental consent Exclusion criteria: not reported
Interventions	Zinc gluconate (20 mg of zinc/day). Montmorillonite.
Outcomes	Diarrhoea at day 3. Average number of hospitalization (days).
Notes	We requested additional information from the trial author, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	There was no mention of how random sequence was generated.
Allocation concealment (selection bias)	Unclear risk	There was no mention of how allocation concealment was performed.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	It was unclear whether or not blinding was done.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There was no loss of participants to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no RCT protocol registration number.
Other bias	Unclear risk	There was no information available on other sources of bias.

Karamyyar 2013.

Methods	RCT
Participants	N: 379 participants Inclusion criteria: children aged 9 months to 5 years, admission to hospital with acute watery diarrhoea and moderate dehydration Exclusion criteria: chronic diseases (cystic fibrosis, inflammatory bowel disease, malabsorption), severe malnutrition (weight curve under 3% for age), dysentery and bloody diarrhoea with red blood cells (RBCs) or white blood cells (WBCs) in stool, recent consumption of antibiotics, severe dehydration, persistent vomiting, consumption of zinc supplements (in the last month), drug intolerance, refusal to consent
Interventions	Zinc supplementation (syrup 1ml/kg/day with 1 mg zinc sulphate divided into two doses) + ORS. ORS.
Outcomes	Stool frequency.
Notes	We requested additional information from the trial author, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used computer‐generated allocation sequence.
Allocation concealment (selection bias)	Low risk	A randomization list (simple randomly allocation of two group) was given to the pharmacist prior to enrolment. Randomization codes were secured until the completion of data collection and neither the physician, participants (or their parents), nor nurse were unaware of the drug or placebo.
Blinding (performance bias and detection bias) All outcomes	Low risk	The glass bottles that contained the products (zinc Sulfate or placebo) were labelled with participants’ code (with keeping the names of participants) by pharmacists who was not involved in the treatments. A placebo with similar taste, colour, and smell and with a similar option (1 mL/kg/day) was given to the control group.
Incomplete outcome data (attrition bias) All outcomes	High risk	15.5% of participants were lost to follow‐up (> 10%).
Selective reporting (reporting bias)	Low risk	The outcomes were reported according to the protocol. The RCT was registered (IRCT201201241580N2).
Other bias	Unclear risk	There was no information available on other sources of bias.

Khatun 2001.

Methods	RCT
Participants	N: 100 participants Inclusion criteria: 6 to 36 months; moderately malnourished (61% to 75% of the median NCHS median weight for age); persistent diarrhoea Exclusion criteria: systemic infection; clinical signs of vitamin A deficiency; received vitamin A supplementation within 3 months; received prior antibiotics therapy; bloody mucoid diarrhoea; kwashiorkor; no longer received breast milk
Interventions	Zinc acetate: 20 mg. Placebo. Both groups: multivitamins
Outcomes	Death. Average duration of diarrhoea. Diarrhoea at day 7. Stool output. Adverse events (vomiting).
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial authors did not provide these details.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not provide these details.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors did not provide these details.
Incomplete outcome data (attrition bias) All outcomes	Low risk	4% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no further information available on other sources of bias.

Larson 2005.

Methods	RCT
Participants	N: 1067 participants Inclusion criteria: children aged 3 to 59 months; acute diarrhoea; having taken ORS as instructed; no vomiting in the past 2 hours for the short‐stay ward or 30 minutes in the outpatient clinic, and no longer dehydrated Exclusion criteria: returning to the hospital with diarrhoea; receiving zinc
Interventions	Zinc sulphate: 20 mg. Placebo.
Outcomes	Adverse events (vomiting).
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The trial used opaque envelopes, numbered, in which the assigned zinc tablet, placebo tablet, or a similar‐sized button was placed; and kept the randomization schedule in a locked cabinet.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	There was no information available on other sources of bias.

Passariello 2015.

Methods	RCT
Participants	N: 83 participants Inclusion criteria: children aged between 5 to 36 months, diarrhoea lasting less than 24 hours with mild‐moderate dehydration Exclusion criteria: malnutrition ( weight/height ratio) < 5th percentile), severe dehydration, concomitant severe or chronic systemic illness, immunodeficiency, cystic fibrosis, food allergy, chronic gastrointestinal disease, endocrine disease, use of pre/pro/symbiotic antibiotics, any anti‐diarrhoea medication in the previous 3 weeks
Interventions	Hypotonic super ORS containing zinc in a gel formulation. Standard hypotonic ORS.
Outcomes	Duration of diarrhoea. Diarrhoea on day 3. Adverse events (vomiting).
Notes	We requested additional information from the trial authors, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used computer‐generated randomization to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	The hospital pharmacy produced identical white aluminium fold sachets contained in a blank blinded code‐labelled paper box for intervention and controls.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	We requested additional information from the trial authors, but received no reply.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There was no loss of participants to follow‐up.
Selective reporting (reporting bias)	Unclear risk	The RCT was registered retrospectively (ACTRN12614000028606).
Other bias	Unclear risk	There was no further information available on other sources of bias.

Patel 2009.

Methods	RCT
Participants	N: 808 participants Inclusion criteria: age 6 to 59 months; acute diarrhoea (duration up to 72 hours); ability to accept oral fluids or feeds   Exclusion criteria: severe dehydration and unable to drink, chronic or severe complicating illness, known positive HIV status, kwashiorkor, residing outside a radius of 30 km around the hospital, participating in another study or already enrolled in this study
Interventions	Zinc sulcates 2 mg/kg/day. Zinc sulphate 2 mg/kg/day + copper 0.2 mg/kg/day. Placebo.
Outcomes	Death. Average duration of diarrhoea. Diarrhoea at day 3. Diarrhoea at day 5. Diarrhoea at day 7.
Notes	Location: India Setting: hospital (follow‐up in the community)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Single‐site, blocked randomization procedure with blocks of sizes 3, 6, and 9 in equal proportions.
Allocation concealment (selection bias)	Low risk	Randomization list generated off site by an investigator not directly involved in the data collection. The code list of the placebo and the treatment groups was secured and held only by the pharmacist at the Universal Medicaments Pvt. Ltd, Nagpur, until initial data analysis was completed.
Blinding (performance bias and detection bias) All outcomes	Low risk	Double blind: bottle packs sequentially labelled according to the treatment allocation list and assigned to participants by the research physician.
Incomplete outcome data (attrition bias) All outcomes	Low risk	7% lost at follow‐up.
Selective reporting (reporting bias)	Low risk	The protocol was available. The trial was registered in the metaRegister of Controlled Trials (ISRCTN85071383).
Other bias	Unclear risk	There was no information available on other sources of bias.

Patel 2009a (zinc).

Methods	RCT
Participants	N: 808 participants Inclusion criteria: age 6 to 59 months; acute diarrhoea (duration up to 72 hours); ability to accept oral fluids or feeds   Exclusion criteria: severe dehydration and unable to drink,chronic or severe complicating illness, known positive HIV status, kwashiorkor, residing outside a radius of 30 km around the hospital, participating in another study or already enrolled in this study
Interventions	Zinc sulphate 2 mg/kg/die. Zinc sulphate 2 mg/kg/die + copper 0.2 mg/kg/die. Placebo.
Outcomes	Death. Average duration of diarrhoea. Diarrhoea at day 3. Diarrhoea at day 5. Diarrhoea at day 7.
Notes	Location: India Setting: hospital (follow‐up in the community)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Patel 2009.
Allocation concealment (selection bias)	Low risk	See Patel 2009.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Patel 2009.
Incomplete outcome data (attrition bias) All outcomes	Low risk	See Patel 2009.
Selective reporting (reporting bias)	Low risk	See Patel 2009.
Other bias	Unclear risk	See Patel 2009.

Patel 2009b (zinc + copper).

Methods	See Patel 2009a (zinc)
Participants	See Patel 2009a (zinc)
Interventions	See Patel 2009a (zinc)
Outcomes	See Patel 2009a (zinc)
Notes	See Patel 2009a (zinc)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Patel 2009.
Allocation concealment (selection bias)	Low risk	See Patel 2009.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Patel 2009.
Incomplete outcome data (attrition bias) All outcomes	Low risk	See Patel 2009.
Selective reporting (reporting bias)	Low risk	See Patel 2009.
Other bias	Unclear risk	See Patel 2009.

Patel 2015.

Methods	RCT
Participants	N: 100 participants Inclusion criteria: children < 12 years (but all enrolled had ≤ 5 years), presentation to hospital with diarrhoea Exclusion criteria: serious illness, intensive care admission, use of ventilators, impossibility of communication
Interventions	Oral zinc sulphate (10 mg for < 6 months a day or 20 mg for ≥ 6 months) per 14 days + standard of care (ORS, intravenous fluid, antibiotics). Standard of care (ORS, intravenous fluid, antibiotics).
Outcomes	Diarrhoea at day 3 and 5.
Notes	We requested additional information from the trial authors, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence generation was computer generated.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not clearly describe allocation concealment.
Blinding (performance bias and detection bias) All outcomes	High risk	This was an open label trial.
Incomplete outcome data (attrition bias) All outcomes	Low risk	4% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no RCT registration number.
Other bias	Unclear risk	There was no information on other sources of bias.

Patro 2010.

Methods	RCT
Participants	N: 160 participants Inclusion criteria: age 3 to 48 months diagnosed with acute diarrhoea lasting less than 5 days, with at least some degree of dehydration Exclusion criteria: diarrhoea lasting <1 day or >5 days, recent history of diarrhoea (last 2 weeks before enrolment day), chronic gastrointestinal disease with diarrhoea manifestation, (for example, food allergy, coeliac disease), weight‐to‐height ratio < 5th percentile, severe dehydration, coexistence of serious systemic disease(s), coadministration of antibiotics, exclusive or > 50% breastfeeding, immunodeficiency, immunosuppressive therapy.
Interventions	Zinc sulphate (20 mg in children > 6 months or 10 mg in children < 6 months). Placebo.
Outcomes	Average duration of diarrhoea. Diarrhoea at day 7.
Notes	Location: Poland Setting: hospital (90% of children) and outpatient (10%)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	An investigator at the Medical University of Warsaw computer‐generated 2 different randomization lists for each centre.
Allocation concealment (selection bias)	Low risk	The glass bottles containing the products were labelled with the participant’s number corresponding to the randomization list by an independent individual who was not involved in participant enrolment. Randomization codes were secured until the completion of data collection and initial analysis. The placebo was identically supplied and formulated. There was no difference between zinc and the placebo in appearance; a minor metallic aftertaste of zinc was hardly detectable.
Blinding (performance bias and detection bias) All outcomes	Low risk	Investigators, participants, outcome assessors, and data analysts were blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	11.8% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no protocol available.
Other bias	Unclear risk	Note: the source of funding was Nutricia.

Penny 1999.

Methods	RCT
Participants	Number: 413 Inclusion criteria: 6 to 36 months, persistent diarrhoea Exclusion criteria: vitamins or minerals within 6 weeks; major congenital malformation affecting growth; severe dehydration; requiring hospitalization
Interventions	Zinc gluconate: 20 mg. Placebo.
Outcomes	Death. Hospitalization. Diarrhoea at day 3. Diarrhoea at day 5. Diarrhoea at day 7. Adverse events (vomiting).
Notes	Location: Peru Setting: community
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used computer‐generated random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Randomization numbers were linked to letter codes, each indicating 1 treatment group; codes were kept secret; independent laboratories provided supplements.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available on other sources of bias.

Polat 2003.

Methods	RCT
Participants	N: 200 participants Inclusion criteria: 2 to 29 months; malnourished children (weight for age scale, score < 76% according to NCHS standards); acute non‐bacterial diarrhoea Exclusion criteria: concomitant illness or oedema
Interventions	Zinc sulphate: 20 mg. Placebo.
Outcomes	Average duration of diarrhoea. Diarrhoea at day 3. Diarrhoea at day 7. Adverse events (vomiting).
Notes	Location: Turkey Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Bottles were labelled with randomization numbers.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	9% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available on other sources of bias.

Polat 2003 (low Zn).

Methods	See Polat 2003
Participants	N: 76 participants Children with low zinc serum levels
Interventions	See Polat 2003
Outcomes	See Polat 2003
Notes	See Polat 2003
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Polat 2003.
Allocation concealment (selection bias)	Low risk	See Polat 2003.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Polat 2003.
Incomplete outcome data (attrition bias) All outcomes	Low risk	See Polat 2003.
Selective reporting (reporting bias)	Unclear risk	See Polat 2003.
Other bias	Unclear risk	See Polat 2003.

Polat 2003 (normal Zn).

Methods	See Polat 2003
Participants	N: 106 participants Children with normal zinc serum levels
Interventions	See Polat 2003
Outcomes	See Polat 2003
Notes	See Polat 2003
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	See Polat 2003.
Allocation concealment (selection bias)	Low risk	See Polat 2003.
Blinding (performance bias and detection bias) All outcomes	Low risk	See Polat 2003.
Incomplete outcome data (attrition bias) All outcomes	Low risk	See Polat 2003.
Selective reporting (reporting bias)	Unclear risk	See Polat 2003.
Other bias	Unclear risk	See Polat 2003.

Roy 1997.

Methods	RCT
Participants	N: 111 participants Inclusion criteria: 2 to 24 months; weight below the 76th centile of weight‐for‐age according to the NCHS standard 18 (by Gomez classification, protein energy malnutrition grades II and III included) Exclusion criteria: systemic infection or oedema
Interventions	Zinc acetate: 20 mg. Placebo. Both groups: multivitamin
Outcomes	Average duration of diarrhoea. Stool output.
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used a table of random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Bottles were labelled with randomization numbers.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	32.4% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No protocol was available
Other bias	Unclear risk	No information was available on other sources of bias.

Roy 1998.

Methods	RCT
Participants	N: 190 participants Inclusion criteria: 3 to 24 months; persistent diarrhoea; underweight (low weight‐for‐age) using a cut‐off of 70% weight/age of the 50th centile of the NCHS standard; wasted (low weight/height) using a cut‐off of 80%; short (low height/age) using a cut‐off of less than 95% of the height/age standard Exclusion criteria: none stated
Interventions	Zinc acetate: 20 mg. Placebo. Both groups: multivitamin
Outcomes	Death. Average duration of diarrhoea. Adverse events.
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Unclear risk	The trial did not provide any details on allocation concealment.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	It was unclear whether or not any participants were lost to follow‐up; 11% discontinued the intervention.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available on other sources of bias.

Roy 2008a.

Methods	RCT
Participants	N: 56 participants Inclusion criteria: aged 12 to 59 months; moderately malnourished (weight/age 61% to 75% of NCHS median); history suggestive of dysentery (for example, bloody‐mucoid diarrhoea or febrile diarrhoea less than 5 days' duration); with culture‐proven shigellosis Exclusion criteria: severe malnutrition; receiving zinc supplementation; measles in the last 6 months; living beyond 2 hours of travel time; complications such as haemolytic uraemic syndrome or other systemic illness, including pneumonia, meningitis, and septicaemia
Interventions	Zinc acetate: 10 mg. Placebo. Both groups: multivitamins
Outcomes	Death. Average duration of diarrhoea. Diarrhoea at day 7.
Notes	Location: Bangladesh Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used a table of random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Bottles were identical and labelled with sequential numbers that had earlier been allocated to either intervention or control according to the randomization.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	High risk	11% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Low risk	The trial was registered at ClinicalTrials.gov (NCT00321126).
Other bias	Unclear risk	No information was available on other sources of bias.

Sachdev 1988.

Methods	RCT
Participants	N: 50 participants Inclusion criteria: children 6 to 18 months; dehydration secondary to acute diarrhoea of < 4 days' duration Exclusion criteria: antibiotics; severe malnutrition (grades III and IV); concomitant features (meningitis, pneumonia, liver disease, otitis media, fever > 39°C)
Interventions	Zinc sulphate: 20 mg. Placebo.
Outcomes	Average duration of diarrhoea. Stool frequency. Adverse events (vomiting).
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial authors did not provide any details.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not provide any details.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors did not provide any details.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The trial authors did not provide any details.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available.

Sachdev 1990.

Methods	RCT
Participants	N: 40 participants Inclusion criteria: 6 to 18 months; persistent diarrhoea Exclusion criteria: another diarrhoeal episode 1 month prior; critically ill; obvious parenteral infections; severe malnutrition (grade III and IV)
Interventions	Zinc sulphate: 20 mg. Placebo.
Outcomes	Average duration of diarrhoea. Stool frequency. Adverse events (vomiting).
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The trial authors did not provide any details.
Allocation concealment (selection bias)	Unclear risk	The trial authors did not provide any details.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	The trial authors did not provide any details.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The trial authors did not provide any details.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available on other sources of bias.

Sazawal 1995.

Methods	RCT
Participants	N: 947 participants Inclusion criteria: 6 to 35 months; diarrhoea for 7 days; permanent resident in study area; stunted defined (length for age less than −2 SD) Exclusion criteria: second visit; malnutrition requiring hospitalization; not provide consent
Interventions	Zinc gluconate: 20 mg. Placebo. Both groups: multivitamin
Outcomes	Diarrhoea at day 7. Stool frequency. Adverse events (vomiting).
Notes	Location: India Setting: hospital
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Children were allocated to sequential numbers indicating zinc or placebo; the WHO kept the code, which was not available to the trial investigators.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	2% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available.

Shimelis 2008.

Methods	RCT
Participants	N: 414 participants Inclusion criteria: children 2 to 59 months, presented at the hospital with acute watery diarrhoea for less than 7 days Exclusion criteria:children living far or unsafe areas for follow‐up, children requiring antimicrobial for other conditions, immunocompromised (severely malnourished or with known primary immune deficiency) excluding cases of measles or those with HIV positive status, special fluid requirements (that is, renal disease, health hepatic failure), chronic or persistent diarrhoea and dysentery requiring hospitalization or admitted for in‐patient care, on zinc supplementation, no consent
Interventions	Zinc (2 tablets each containing 10 mg zinc) and ORS. ORS.
Outcomes	Diarrhoea at day 5. Adverse events: vomiting.
Notes	We requested additional information from the trial author, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	The trial used randomly selected days to assign participants to treatment.
Allocation concealment (selection bias)	High risk	There was no randomization concealment since participants were randomized depending on the day they reported to the health facility.
Blinding (performance bias and detection bias) All outcomes	High risk	This was an open label trial.
Incomplete outcome data (attrition bias) All outcomes	Low risk	3% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There was no RCT registration number. We requested additional information from the trial authors but received no reply.
Other bias	Unclear risk	There was no information available on other sources of bias.

Strand 2002.

Methods	RCT
Participants	N: 899 participants Inclusion criteria: 6 to 35 months; diarrhoea < 96 hours Exclusion criteria: massive dose of vitamin A; requiring hospitalization; family intended to leave Bhaktapur within 2 months
Interventions	Zinc gluconate: 15 mg for infants; 30 mg for older children. Placebo.
Outcomes	Diarrhoea at day 3. Diarrhoea at day 7. Adverse events (vomiting). Adverse events (copper levels).
Notes	Location: Nepal Setting: community
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The trial used random numbers to assign participants to treatment.
Allocation concealment (selection bias)	Low risk	Packing with serial number; the list was kept in Copenhagen; capsules were identical in appearance; the syrup was identical in appearance and taste.
Blinding (performance bias and detection bias) All outcomes	Low risk	The trial was double blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	1% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	No protocol was available.
Other bias	Unclear risk	No information was available.

Tran 2015.

Methods	RCT
Participants	N: 76 participants Inclusion criteria: children 6 months to 12 years,clinically diagnosed with diarrhoea and tolerate oral feed Exclusion criteria: other gastrointestinal symptoms, history of gastrointestinal surgery with organic disease (excluding previous gastrostomy, pyloric stenosis), phenylketonuric or diabetic, taking gastric acid‐neutralizing antacids, drugs to suppress gastric acid secretion or anti‐diarrhoeal drugs, probiotics or zinc supplement, immunocompromised, proven sucrose intolerance, or previously participated in the study
Interventions	ORS (Gastrolyte‐R sachets) with zinc sulphate fortification (3 mg elemental zinc in total) to be mixed with 200 mL water for 4 days up to a maximum of 4 sachets in 24 hours). ORS (Gastrolyte‐R sachets).
Outcomes	Duration of diarrhoea.
Notes	We requested additional information from the trial author, but received no reply
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated random sequence.
Allocation concealment (selection bias)	Unclear risk	Not clearly described.
Blinding (performance bias and detection bias) All outcomes	Low risk	All bottles packaged by the manufacturing pharmacy.
Incomplete outcome data (attrition bias) All outcomes	High risk	Lost to follow‐up in both study arms = 23.7% (> 10%).
Selective reporting (reporting bias)	Unclear risk	No RCT registration number. We did not receive a response from the trial authors.
Other bias	Unclear risk	No information available.

Abbreviations: N: number of participants; NCHS: National Center for Health Statistics; ORS: oral rehydration solution; RCT: randomized controlled trial; WHO: World Health Organization.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Abraham 2016	This study did not concern the intervention of interest to this review
Adu‐Afarwuah 2007	This study did not concern the intervention of interest (3 types of micronutrients for food fortification)
Adu‐Afarwuah 2008	This study did not concern the intervention of interest (zinc fortification)
Aggarwal 2007	Randomized controlled trial (RCT) on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Agustina 2007	This study did not concern the intervention of interest (probiotic, prebiotic, fibre, and micronutrient mixture)
Alam 2010	Prevention study
Alarcon 2004	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Awasthi 2006	This study did not concern the intervention of interest (zinc in oral rehydration solution (ORS))
Baqui 2002	A community RCT without a placebo group
Baqui 2003	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Baqui 2006	This study did not concern any outcome of interest (serum zinc) to this review
Baum 2010	This study did not concern the population of interest (adults, HIV‐positive)
Becquey 2016	This study did not concern the intervention of interest to this review
Behrens 1990	This study did not concern any outcome of interest (nutritional status)
Bhandari 2002	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Bhandari 2005	This was not a RCT
Bhandari 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Bhandari 2008	A RCT without a placebo group
Bhatnagar 2004b	This was not a RCT
Bhutta 2000a	This study did not concern any outcome of interest (appetite)
Bilenko 2010	This study did not concern the intervention of interest (multiple micronutrients in sprinkles)
Black 2001	Not a RCT
Bobat 2005	This study did not concern the population of interest (only children with HIV enrolled)
Borges 2007	This was not a RCT
Brooks 2005b	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Brown 2007	This study did not concern the Intervention of interest (food fortification)
Bruzzese 2016	This study did not concern the intervention of interest to this review
Chandyo 2010	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Chang 2010	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Chen 2010	This study did not concern the intervention of interest (food fortification with multiple micronutrients)
Chhagan 2009	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Chhagan 2010	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Christian 2009	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment, on a different population (pregnant women)
CIGNIS 2010	This study did not concern the intervention of interest (food fortification with multiple micronutrients)
Colgate 2016	This study did not concern the intervention of interest to this review
Coronel Carbajal 2000	This was not a placebo‐controlled RCT
Cross 2009	Not a RCT
Cárcamo 2006	This study did not concern the population of interest (adults with HIV)
Dhingra 2009	Not a RCT
Doherty 1998	This was not a placebo‐controlled RCT, and the criterion for inclusion of children was malnutrition, not diarrhoea
Ebrahimi 2006	Thist study did not concern any outcome of interest (growth) to this review
Ellis 2007	Not a RCT
Ferraz 2007	Not a RCT
Ferrufino 2007	Not a RCT
Fischer Walker 2008	Secondary analysis of a previously excluded study (Baqui 2002)
Gardner 2005	Not a RCT
Garenne 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Gebremedhin 2016	This study did not concern the outcomes of interest to this review
Gregorio 2007	This study did not concern the intervention of interest (zinc‐fortified ORS)
Gupta 2003	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Gupta 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Habib 2010	A longitudinal cohort study
Habib 2013	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Heinig 2006	This study did not concern any outcome of interest (growth, morbidity, and motor development)
Hess 2015	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Hettiarachchi 2008	This study did not concern the population of interest (children 12 to 16 years), nor the outcomes
Hidayat 1998	A community RCT, but we could not compare the results with other studies because of methodological problems (enrolling the same children more than once) and types of outcomes (episodes of diarrhoea and not children with diarrhoea)
Hoque 2006	Not a RCT (review)
Hyder 2007	This study did not concern the population of interest (adolescent girl), the intervention (multiple micronutrients), nor the outcomes
Iannotti 2010	This study did not concern the population of interest (pregnant women)
Islam 2010	This study did not concern the population of interest (preterm infants), nor any outcome of interest (growth)
Jimenez 2000	This study did not concern any outcome of interest (growth)
Kelly 1999	The intervention and the population (micronutrient supplementation in AIDS diarrhoea‐wasting syndrome) considered in this RCT were not relevant to this review
Kelly 2010	This study did not concern any outcome of interest (intestinal function) to this review
Kianmehr 2016	This study did not concern any interventions of interest to this review
Larson 2010	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Lin 2008	This study was not placebo controlled, and did not report outcomes of interest (weight)
Lind 2004	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Lind 2008	A secondary analysis of a previously excluded study (Lind 2004)
Lira 1998	This study did not concern the population of interest (low birthweight infants)
Long 2006	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Long 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment, with different outcomes (specific intestinal infections)
Lopez 2005	This study did not concern the intervention of interest (multiple micronutrient), nor the outcomes (anaemia, micronutrient status, growth, and morbidity) of interest to this review
Luabeya 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Makonnen 2003a	This study did not concern any outcome of interest to this review
Makonnen 2003b	This study did not concern any outcome of interest to this review
Manger 2008	No placebo control, different intervention (multiple micronutrients), prevention study
Maragkoudaki 2016	This study did not concern any intervention of interest to this review
Martinez‐Estevez 2016	This was a prevention study
Mazariegos 2010	This study did not concern any outcome of interest (linear growth) to this review
Mazumder 2010	This was a secondary analysis of a previously excluded study (Bhandari 2008)
Mda 2010	This study did not concern a population of interest (only children with HIV), and used a different intervention (multiple micronutrient)
Meeks Gardner 1998	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Müller 2001	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Naheed 2009	A secondary analysis of a previously excluded study (Baqui 2002)
Nasrin 2005	Not a RCT
Negi 2014	The study participants were above 5 years (age range 5 to 12 years)
Nga 2009	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Osendarp 2002	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Ouedraogo 2008	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Passariello 2010	This study did not concern the intervention of interest (zinc in ORS) to this review
Patel 2005	This study did not concern the intervention of interest (zinc and copper in ORS)
Patel 2010a	Secondary analysis of an included study (Patel 2009), with no outcome of interest (by isolated microorganism)
Patel 2010b	Not a RCT (review)
Patel 2012	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Penny 2004a	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Penny 2004b	Not a RCT
Polat 2006	Not a placebo‐controlled RCT
Prado 2016	This study used a different intervention that was not of interest to this review
Rahman 2001	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Rahman 2005	This study did not concern any outcome of interest to this review
Raqib 2004	This study did not concern any outcome of interest (immune and inflammatory responses) to this review
Richard 2006	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Rollins 2007	This study did not concern the population of interest (only HIV‐infected children), and different outcomes (growth, immunity)
Rosado 1997	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Rosado 1998	Not a RCT
Rosado 2009	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment, with different outcomes (specific intestinal infections)
Roy 1992	This study did not concern any outcome of interest (intestinal permeability) to this review
Roy 1999	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Roy 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Roy 2008b	This study did not concern the population of interest (children aged between 3 and 14 years)
Ruel 1997	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sabatier 1997	Not a placebo‐controlled RCT
Samuel 1995	Not a RCT
Sazawal 1996	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sazawal 1997a	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sazawal 2004	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sazawal 2007a	This study did not concern the intervention of interest to this review(milk fortification)
Sazawal 2007b	This study did not concern any outcome of interest to this review (plasma retinol)
Sazawal 2007c	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Shamir 2005	Different intervention (zinc and probiotics)
Shankar 1998	Not a RCT (review)
Sharieff 2006	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sheikh 2010	Not a RCT
Sur 2003	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Sáenz De Pipaón 2007	Not a RCT (review)
Taneja 2009	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment, in a different population (low birthweight infants)
Taneja 2010	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment, with different outcomes (growth)
Tielsch 2006	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Tielsch 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Umeta 2000	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Untoro 2005	This study did not concern the intervention of interest (multiple micronutrient), nor any outcome of interest to this review (anaemia, micronutrient status, growth, and morbidity)
Valery 2005	This study did not concern the population of interest (all children aged under 11 years)
Veenemans 2011	This was a prevention study
Wadhwa 2011	This was a study on zinc‐enriched ORS
Walden 2004	Not a RCT
Walker 2007	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment
Wieringa 2010	This study did not concern the population of interest to this review (pregnant women)
Winch 2006	Not a RCT
Winch 2008	Not a RCT
Wuehler 2008	A RCT on zinc supplementation for prevention of diarrhoea episodes, not for diarrhoea treatment

Abbreviations: AIDS: acquired immune deficiency syndrome; HIV: human immunodeficiency virus; ORS: oral rehydration solution; RCT: randomized controlled trial.

Characteristics of ongoing studies [ordered by study ID]

NCT01140074.

Trial name or title	Efficacy of zinc sulfate with probiotics for the treatment of acute diarrhoea in children
Methods	RCT
Participants	Inclusion criteria: age 1 to 36 months; a cute diarrhoea defined as 3 or more watery stools per day; informed consent (parents) Exclusion criteria: severe dehydration (> 10%); coexisting severe infection (for example, sepsis, pneumonia, meningitis); immune deficiency; chronic digestive tract disease (for example, coeliac disease, food allergy); on antibiotic therapy
Interventions	Zinc sulphate 10 to 20 mg per day orally plus probiotics. Zinc sulphate 10 to 20 mg per day orally. Placebo.
Outcomes	Period of diarrhoea in hours (time frame: 15 days) (designated as safety issue: no). Number of stools in consequent days (time frame: 15 days). Hospitalization. Tolerability. Adherence to the therapy.
Starting date	July 2010 (not yet recruiting in December 2010)
Contact information	Contact: Leszek Szenborn, Prof szenborn@zak.am.wroc.pl (principal investigator) Contact: Ernest P. Kuchar, MD kuchar@zak.am.wroc.pl
Notes	Location: Poland Registration number: NCT01140074 Source of funding: unclear Sponsor: University Hospital No 1 Wroclaw

NCT01198587.

Trial name or title	A double blind randomized placebo controlled trial of oral zinc for children with acute diarrhoea in a developed nation
Methods	RCT
Participants	Inclusion criteria Healthy children with non‐bloody diarrhoea illness defined as loose or watery stools. Symptoms must be present for greater than 24 hours but less than 72 hours. Comorbid conditions including: asthma, gastroesophageal reflux (unless followed by a gastroenterologist), mild speech, language, motor delays, benign heart murmurs, isolated atrial septal defect (ASD) or ventricular septal defect VSD, epilepsy (unless developmentally delayed), children born prematurely between 33 to 37 weeks without long term sequelae, repaired tetralogy of Fallot (no cardiac issues for > 6 months), diabetes may be enrolled in the study. Exclusion criteria Children with symptoms less than 24 hours. Children with symptoms greater than 24 hours. Failure to thrive. G or J tube. Major surgery within last 3 months. Minor surgery (e.g. tonsillectomy, ear tubes, skin lesion removal) within last 1 month. Followed by gastrointestinal service for any reason (Crohn, ulcerative colitis, constipation. Developmental delay, patient > 1 year behind milestones. Current brain tumour. Currently being treated for cancer or in remission < 6 months. Intussuception. Antibiotics in the last 14 days or currently taking antibiotics for any reason. Autism. Children born premature < 33 weeks. Cystic fibrosis. Major congenital heart disease (any disease where child's baseline oxygen saturations < 93%). Short gut. Liver disease. History of bowel resection. Age minimum: 6 months Age maximum: 6 years Gender: both
Interventions	Zinc sulfate: for children aged 6 months to 1 year, 12.5 mg orally daily for 14 days mixed in 60 mL of fluid; for children aged 1 year and above 25mg orally daily for 14 days mixed in 60 mL of fluid. Placebo.
Outcomes	Duration of diarrhoea in acute diarrhoeal illnesses in a developed nation while taking zinc or placebo (time frame: 14 days)
Starting date	September 2010
Contact information	Michelle L Niescierenko, MD michelle.niescierenko@childrens.harvard.edu Children's Hospital Boston
Notes	Location: USA Registration number: NCT01198587 Source of funding: unclear Sponsor: Children's Hospital Boston

Abbreviations: MD: medical doctor; RCT: randomized controlled trial.

Differences between protocol and review

2016

We included a PRISMA study flow diagram and funnel plots.

2011

We used GRADE profiler, version 3.2.2 to create 'Summary of findings’ tables for the primary outcomes in the review.

2007, Issue 4 (first review version)

We made the following modifications to the review.

• We changed the inclusion criteria for participant age to "children over one month old" (rather than "two months") to avoid arbitrarily losing trials.

• We amended death to a secondary outcome measure following feedback from referees.

• We stratified the results by age categories since we observed significant heterogeneity when trials were pooled, and a clear difference in zinc effect was evident according to age.

• For subgroup analysis by nutritional status, it was not possible to refer to the definition of malnutrition given in the protocol (weight/height) as most included trials used another definition (weight/age), which is easier to measure. The difference between the two definition is that the first identifies children with acute weight loss or 'wasted', while the second includes both children with acute and chronic malnutrition ('wasted' and 'stunted').

• Two categories of 'zinc dose' were used (20 mg and > 20 mg) as most trials used zinc 20 mg/day, and only two trials used more than 20 mg/day.

• We added sex (male, female) as a subgroup as it was identified as a possible effect modifier (Garenne 2005).

Contributions of authors

Both ML and HW contributed equally to the preparation of this Cochrane Review update.

Sources of support

Internal sources

• Liverpool School of Tropical Medicine, UK.

External sources

• Department for International Development (DFID), UK.

  Grant: 5242

Declarations of interest

Marzia Lazzerini has no known conflicts of interest.
 Humprey Wanzira has no known conflicts of interest.

Unchanged