Clean delivery kit use in low-and middle-income countries: a systematic review and meta-analysis

Abstract

Background

The use of Clean Delivery Kits (CDKs) promotes healthy birth outcome. Nevertheless, representative data are lacking for low- and middle-income countries. Thus, study aimed at determining the overall estimate of clean delivery kit use among low and middle income countries.

Methods

A comprehensive review of the Science Direct, Scopus, Google Scholar, EMBASE, and PubMed databases was part of the current investigation. Microsoft Excel (version 14) was used to extract the data, and STATA statistical software was used for analysis. The presence of publication bias was assessed using DOI Plots in the study. By calculating I² and doing an overall estimation analysis, heterogeneity was evaluated. Furthermore, a subgroup analysis was carried out according to the sample size, publication year, and study design.

Results

Out of 654 articles, 12 met the inclusion criteria and were included in the analysis, covering 19,889 study participants. In low- and middle-income countries, the pooled proportion of clean delivery kits use was 54% (95%CI: 31. 60, 76.82, I² = 70%). Notably, the cross-sectional study revealed the lowest incidence of delivery kit use, with a proportion of 54.29%, while the cohort study revealed the highest proportion, at 72.37%. In studies with sample sizes < 500, clean delivery kit use was higher (60.28%) compared to studies with sample sizes > 500 (45.75%). Lastly, the proportion of clean delivery kits was found to be 62.53% for studies published after 2010, compared to 48.25% for studies published before 2010.

Conclusion

In summary, the proportion of clean delivery kits use was 54%. Healthcare systems should focus on different strategies to improve clean delivery kit use.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-024-06913-1.

Introduction

An estimated 303 000 mothers died annually worldwide in 2015 [1], while roughly 2. 62 million infants pass away within 28 days of birth [2]. Infections are a leading cause of maternal and neonatal morbidity and mortality [3–7] and are responsible for an estimated 10.7% of all maternal deaths [8] and 44% of all neonatal deaths [9] worldwide each year. In the poorest quintile of home births in SSA, 41% were attended by a traditional birth attendant (TBA), compared to 56% that were unattended [10]. Neonates born in these nations have three to twenty time’s higher rates of neonatal infections than newborns born in high-income nations [4]. Due to insufficient intra-partum and postpartum infection control procedures, the risk of infection is still high in both facilities and the community [4].

Puerperal sepsis is the major cause of maternal morbidity and mortality worldwide while about 94% of maternal mortality occurs in low and middle- income countries [11]. Neonatal sepsis is a major cause of mortality and morbidity in developing countries [12]. Up to 40% of infections leading to neonatal sepsis are transmitted at the time of birth, and are classified in early-onset (EOS) sepsis and late onset sepsis (LOS) [13]. Obtaining cause-specific estimates of maternal and neonatal morbidity and mortality for low- and middle-income countries is a notoriously challenging task [14, 15].

The World Health Organization (WHO) reports that 30% of newborn deaths and 10 points 7% of maternal deaths are caused by infections acquired during childbirth [8]. The WHO recommends following the “six cleans” during delivery in order to prevent sepsis. These include keeping the hands, perineum, delivery surface, cord and tying instruments, and cutting surfaces clean [16]. Clean delivery kits (also known as CDKs) are potentially low-cost interventions to support clean delivery practices. They contain items like soap, gloves, cord ties, and other sterile equipment to enable the “six cleans” [17]. Three studies examined the impact of complex intervention packages, which include clean delivery kits, on maternal outcomes. The impact of these studies on the health of mothers and newborns was assessed in a recent systematic review [18–21]. These studies’ findings imply that clean delivery procedures, especially the use of clean kits, can enhance the prognosis of mothers who have puerperal sepsis [19–21].

Despite the existence of global policies and promotional efforts, the implementation of clean birth practices is poorly implemented in both institutional [22–24] and community settings in low- and middle-income countries (LMIC) [25, 26]. Previous research on clean birth practices in LMIC has mainly focused on umbilical cord care [27] and hygiene practices among traditional birth attendants [28]. These practices are crucial for promoting hygienic birth and postnatal care and are essential for the effective implementation of infection prevention measures and the provision of quality care in general [29].

By 2030, the UN’s Sustainable Development Goals (SDGs) seek to bring the rate of neonatal mortality down to at least 12 per 1,000 live births and the rate of maternal mortality down to less than 70 per 100,000 births [30]. According to estimates, maintaining cleanliness during birth procedures can lower the risk of tetanus-related neonatal mortality by 30% in home births and 38% in facility births [31]. Furthermore, up to 23% of maternal deaths in low-income nations may be avoided by implementing clean birth practices in conjunction with active management, third-stage surveillance, and monitoring [32].

One promising and reasonably priced intervention to support clean delivery practices has been identified: Clean Delivery Kits (also known as CDKs) [33]. The impact of CDK use has been the subject of numerous studies, all of which have consistently found that it increases the use of clean delivery practices [34–36]. CDKs have been demonstrated to be effective in significantly lowering perinatal and neonatal mortality in South Asia when used as part of a package of interventions [19, 37]. Furthermore, it has been demonstrated that CDKs offer defense against tetanus and umbilical cord infections, suggesting their potential to lower neonatal morbidity and mortality [28]. All things considered, these findings imply that CDKs are an effective intervention to enhance the health of expectant mothers and their babies.

There is still a need for up-to-date research to elucidate the use of clean delivery kits (CDKs) in low- and middle-income nations, despite earlier reviews on the subject that mainly emphasized knowledge gaps about the efficacy of CDKs rather than their application [31]. Furthermore, estimates of the worldwide proportion of clean delivery kit usage ranged from 8% [38] to 99.3% [39], with primary studies demonstrating substantial variation and inconsistency in this regard. In order to address the observed variations in the utilization of delivery sets, it becomes imperative to consolidate and utilize this information within the lower and middle income strata. Moreover, the absence of comprehensive data representative of the low and middle income levels necessitates the undertaking of this meta-analysis study. Its objective is to ascertain the overall proportion of clean delivery kit usage in countries classified as low and middle-income.

Methods

Data synthesis and reporting

The current investigation involved examining data obtained from a single measurement outcome, particularly in relation to the use of delivery kits. The results were presented through the implementation of tables, text descriptions and a forest diagram. In accordance with the established PRISMA checklist guideline, a systematic review and meta-analysis was conducted to assess the overall proportion of clean delivery kit use low and middle income countries [40] (Additional file 1). The review protocol has been registered with PROSPERO, number CRD42023464752.

Search strategy

In order to explicitly present our review question, we considered utilizing PECO questions. The study population consisted of women who gave birth(P), with women who utilized a clean delivery kit being the exposure of the study (E), while the prevalence of clean delivery kit usage served as the study’s outcome (CO). These inquiries were constructed by combining relevant words and phrases with the Medical Subject Headings (MeSH) and the Boolean operators “OR” and “AND”. To find as many pertinent primary studies as feasible, we created the following review question using the modified PECO structure described above:

Review question

“What is the magnitude of clean delivery kit use in low and middle income countries?

A comprehensive search was carried out for publications regarding the use of clean delivery kits in low- and middle-income nations for this study. Google Scholar, Pub Med, Science Direct, Scopus, EMBASE, and other international online databases were used in the search. There was no restriction on publication period. The articles published from January 2000 until August 2023 was retrieved. These included “use,” “utilization,” “clean delivery kit,” “birth kit,” “mama kit,” and list of countries of low and middle income countries. The combined and separate search phrases, Boolean operators like “OR” and “AND” were used for searching related articles (Additional file2).

Study outcome

A CBK was defined as a disposable package containing at least the minimum commodities required to facilitate clean cutting and tying of the umbilical cord e.g. a clean blade, clean cord tie or clamp [31].

Low and middle-income countries: It has been defined as based on World Bank list of economic status of countries [41].

Inclusion and exclusion criteria

The studies published in English between, conducted in low- and middle-income countries, and with fully searchable texts were all included in the current meta-analysis. The analysis also included studies that included data on the use of delivery kits. Excluded from this systematic review and meta-analysis were qualitative studies, duplicate sources, randomized controlled trial studies, and incomplete articles.

Quality assessment

The research paper’s quality was evaluated in the current study using a standardized quality assessment checklist created by the Joanna Briggs Institute (JBI) [42] (Additional file3). NAG and KDT, two authors, independently completed the evaluation. The third author, YA, encouraged discussion to settle any disputes that surfaced during the quality assessment. Eight parameters with “Yes,” “No,” “Uncertain,” and “Not Relevant” boxes were included in the critical analysis checklist. Studies that scored 50% or more on the quality assessment indicators were classified as low risk. Other criteria included whether interfering elements were found, whether confounding factor measures were mentioned, whether the results were actually and accurately measured, and whether the statistical analysis was appropriate. The inclusion criteria for the sample were (1) clearly stated, (2) detailed, and the study participants and environment were described in detail, (3) the exposure measurement results were valid and reliable, and (4) the event met the main purpose and accepted standards.

Data extraction

The Joanna Briggs Institute created a standardized data extraction format that served as the foundation for the current investigation. Two authors, NAG and KDT, independently completed the extraction process and extracted all pertinent data. In the event that disagreements arose during the extraction process, KAG, the third author, moderated a discussion to get to an agreement. An automated tool was not used in this study; instead, manual data extraction was done. Paper forms were not used. First author’s name, publication year, study setting, country, design, sample size, proportion of clean delivery kit use, and paper quality were among the extracted data.

Data analysis

The present study comprised the process of extracting pertinent data from a Microsoft Excel spreadsheet and exporting it to STATA software version 14 for analysis. A forest patch that was used to estimate the pooled delivery kit estimate was examined visually to determine whether heterogeneity was present. Based on the publication year and study location, a subgroup analysis was carried out. The impact of individual studies on the total estimate of clean delivery kit use was evaluated using a sensitivity analysis. Hence, this study was meta-analysis of proportion of studies; we performed DOI Plots to assess the presence of publication [43]. To assess the amount of total/residual heterogeneity, test for heterogeneity, and assess the variability brought on by heterogeneity, the Cochran’s Q X² test and I² statistics were used [44].

Results

During the initial phase of our investigation, an extensive search strategy was implemented utilizing a variety of electronic databases on a global scale, leading to the identification of a total of 654 research studies. Following the elimination of 147 duplicate papers, the remaining 507 studies underwent a meticulous evaluation of their titles and abstracts, resulting in the recognition of 168 papers that were considered pertinent to the research inquiry. Subsequently, a detailed examination of the complete articles was carried out, leading to the exclusion of 156 articles for various reasons. Ultimately, a sum of 12 studies [20, 21, 35, 36, 38, 39, 45–50] involving 19,889 study participants met the criteria for inclusion in this systematic review and meta-analysis (Fig. 1).

Fig. 1.

PRISMA flow chart displays the article selection process for clean delivery kit use in low and middle-income countries

A total of 12 studies were examined in the present investigation. These studies consisted of eight cross-sectional studies [20, 21, 36, 38, 39, 46–48], three cohort studies [45, 49, 50], and one case-control study [35]. All of the studies included in the analysis were community-based. Geographically, the studies were distributed as follows: three were conducted in Nepal [38, 47, 48], two in Nigeria [49, 50], two in Tanzania [20, 46], two in Egypt [21, 36], and one each in Pakistan [35], Papua New Guinea [45], and Uganda [39]. The sample sizes of the studies ranged from 93 to 4893. The prevalence of clean delivery kit use varied between 8% and 99.3%. It is worth noting that all of the studies included in the analysis were considered to have a low risk of bias (Table 1).

Table 1.

Characteristics of studies included in the systematic review and meta-analysis of clean delivery kit use in low- and middle-income countries

Author/year	Country	Economic Status	Setting	Design	Sample size	Prevalence	Quality
LM.Valley et.al/2016 [45]	Papua New Guinea	Low-income	Community	Cohort	200	93.4%	Low-risk
Ediau et.al/2013 [39]	Uganda	Low-income	Community	Cross-sectional	202	99.3%	Low-risk
Darmstadt et.al/2009 [21]	Egypt	Middle-income	Community	Cross-sectional	334	71.6%	Low-risk
Osrin et.al/2002 [38]	Nepal	Middle-income	Community	Cross-sectional	4893	8%	Low-risk
Balsara et.al/2009 [36]	Egypt	Middle-income	Community	Cross-sectional	349	74%	Low-risk
Raza & Aven /2013 [35]	Pakistan	Middle-income	Community	Case-control	420	17.1%	Low-risk
Winani et.al/2007 [20]	Tanzania	Low-income	Community	Cross-sectional	3262	59%	Low-risk
Winani et.al/2005 [46]	Tanzania	Low-income	Community	Cross-sectional	3262	59%	Low-risk
Leight et.al/2018 [50]	Nigeria	Middle-income	Community	Cohort	3317	51.7%	Low-risk
Falle et.al/2009 [47]	Nepal	Middle-income	Community	Cross-sectional	93	47%	Low-risk
Sreeramreddy et.al/2006 [48]	Nepal	Middle-income	Community	Cross-sectional	240	19.2%	Low-risk
Abdulkadir & Rainis/2020 [49]	Nigeria	Middle-income	Community	Cohort	3317	51.4	Low-risk

Meta-analysis

Utilization of clean delivery kit use in low-and middle income countries

The current investigation utilized a random-effects model to calculate the combined estimate of clean delivery kit utilization. Results indicated that the utilization of clean delivery kits in low and middle income nations was approximately 54.21% (95%CI: 31.60, 76.82), accompanied by a heterogeneity score of (I²) = 70.9% (Fig. 2).

Fig. 2.

Forestplot for pooled prevalence of clean delivery kit use in low and middle income countries

Sub-group analysis

A subgroup analysis was carried out based on the characteristics of the study sample, study design, and publication year. The results indicated that the use of clean delivery kits was more common in cohort studies (72.37%; 95% CI: 31.21, 93.53) compared to cross-sectional studies (54.29; 95%CI: 26.13, 82.45) (Fig. 3). In studies with sample sizes < 500, clean delivery kit use was higher (60.28%; 95%CI: 32.23, 88.32) compared to studies with sample sizes > 500 (45.75%; 95%CI: 19.95, 71.56) (Fig. 4). Additionally, the use of delivery kits was higher in studies published after 2010 (62.53%; 95% CI: 34.54, 90.53) compared to studies published before 2010 (48.25%; 95%CI: 22.93, 73.56) (Fig. 5).

Fig. 3.

Forest plot displaying sub-group analysis of clean delivery kit use based on study design

Fig. 4.

Forest plot displaying sub-group analysis of clean delivery kit use based on sample size

Fig. 5.

Forest plot displaying sub-group analysis of clean delivery kit use based on publication year

Leave–one–out sensitivity analysis

A sensitivity analysis was performed using the leave-one-out method to evaluate the impact of individual studies on the overall proportion of clean delivery kit utilization. This approach systematically excluded one study at a time. The results, as depicted in Table 2, demonstrate a substantial influence of the individual studies on the overall proportion of delivery kit use. Notably, the exclusion of Ediau et al. and Osrin et al. led to a significant alteration in the overall proportion of clean delivery kit utilization.

Table 2.

The pooled prevalence of clean delivery kit use when one study omitted from the analysis a step at a time

Study omitted	Estimate	95% Conf. Interval
LM.Valley et.al/2016	50.64	27.14–74.15
Ediau et.al/2013	50.10	31.93–68.26
Darmstadt et.al/2009	52.63	28.90-76.37
Osrin et.al/2002	58.46	43.15–73.76
Balsara et.al/2009	52.41	28.68–76.14
Raza & Aven /2013	57.59	33.74–81.43
Winani et.al/2007	53.78	28.71–78.83
Winani et.al/2005	53.78	28.71–78.83
Leight et.al/2018	54.46	29.27–79.66
Falle et.al/2009	54.86	31.21–78.50
Sreeramreddy et.al/2006	57.39	33.64–81.13
Abdulkadir & Rainis/2020	54.46	29.27–79.66
Combined	54.21	31.60-76.82

Heterogeneity and publication bias

The current investigation arrived at a finding concerning subgroup analysis by evaluating the capacity of the study regions to explain the alleged heterogeneity (I² = 70.9%). The presence of publication bias was assessed using DOI Plots in the study (Fig. 6). The DOI plot depicted in Fig. 6 demonstrated a notable asymmetrical distribution of research outcomes (LFK index = 2.22), hinting at the likelihood of publication bias.

Fig. 6.

DOI plot analysis and LFK index of publication bias

Discussion

In recent times, there has been a concentrated endeavor to decrease the rates of maternal and newborn mortality in low-income nations. The primary focus of these endeavors has been the implementation of training programs for obstetricians [19, 51]. The effectiveness of these programs has been consistently demonstrated, emphasizing the crucial role played by skilled caregivers in ensuring safe deliveries in settings with limited resources. However, it is imperative to differentiate between deliveries conducted by qualified attendants and those necessitating skilled care, which encompasses providing women with appropriate hygienic care prior to, during, and after delivery in a clean and supportive environment [52]. Ideally, the delivery environment should be equipped with convenient access to clean delivery supplies, emergency medications, and obstetric care, preferably within a healthcare facility. The achievement of facility-based clean deliveries in low-income countries still remains elusive. Despite efforts to promote such deliveries, progress has been sluggish due to inadequate financial resources and a lack of political determination to integrate this practice into healthcare systems. This obstacle is anticipated to persist in the foreseeable future [31].

The utilization of clean delivery kits is a crucial strategy aimed at improving the health outcomes of both mothers and newborns. Research has demonstrated that the introduction of clean delivery kits has a significant positive impact on neonatal health, leading to a reduction in perinatal mortality [18], neonatal mortality [19, 37], sepsis [37], and umbilical cord infection [20, 21, 52]. Furthermore, several studies have reported notable reductions in adverse maternal outcomes associated with the use of clean delivery kits, including postpartum infections [20, 21] and postpartum hemorrhage [45]. The objective of this systematic review and meta-analysis is to determine the overall prevalence of clean delivery kit utilization among women residing in low and middle-income countries. The findings of this study reveal that the estimated utilization rate of clean delivery kits in these countries is 54.21% (95% CI: 31.60, 76.82), with a corresponding heterogeneity score of (I²) = 70.9%.

The current study reveals that there is a higher prevalence of Clean Delivery Kit (CDK) usage compared to a previous study conducted in South Asia, which reported a range of 3–35% [37]. This difference in results may be due to variations in sample size, study design, and the time interval between the studies. On the other hand, our findings suggest a lower prevalence of CDK usage compared to a systematic review study that reported a range of 8–99% [53]. This inconsistency could be attributed to challenges in delivering the service and a lack of awareness about the use of CDKs [6, 19]. Implementing the distribution of CDKs is a simple and cost-effective measure that can have a significant impact on improving the quality of care during pregnancy, delivery, and the neonatal period, especially at the community level in low-income countries. In settings where facilities are not well-equipped and staffs do not receive continuous training, the use of CDKs can be an effective way to reduce neonatal mortality [54]. These challenges arise from the limited scientific literature on CDK uptake and the focus on distributing kits to health facilities, shopping stores, or skilled birth attendants, rather than documenting the usage of the intervention program. Additionally, the recipients of CDKs can vary [55]. The availability of CDKs can promote adherence to clean delivery and cord care practices for pregnant women and obstetricians, thus maximizing the potential of clean practices to achieve Millennium Development Goal 4 [56].

Funnel plots are not suitable for assessing publication bias in meta-analyses of proportions; thus, Doi plots and the LFK index are valuable alternatives [43]. In terms of sensitivity and specificity, the DOI plot and its correspondence Luis Furuya-Kanamori (LFK) index outperformed the funnel plot and Egger’s test in identification of publication bias [43]. It is visually examined by inspecting the distribution of dots representing individual studies. In Fig. 6 there is a noticeable visual asymmetry, with the studies dispersed towards the right limb. We used the LFK index to give a quantitative interpretation without subjectivity. Therefore, an LFK index value of 2.22 indicates significant asymmetry in cases where there is an expectation that studies with higher prevalence are more likely to be published.

The present investigation allowed researchers to assess the regulations of low- and middle-income countries in relation to the compliance of their health facilities with national health guidelines. More specifically, the study evaluated the accessibility of clean delivery kits to all pregnant women, regardless of their socioeconomic status, as outlined in the policy. The outcomes suggest a strong commitment from the government to improve maternal health and decrease maternal mortality rates.

This evaluation possesses several noteworthy strengths. Firstly, it employed comprehensive electronic search engines with a global reach. Secondly, it presents, for the first time, an overall estimation of the utilization of clean delivery kits, as no prior research has been conducted in this area to the best of our knowledge. Lastly, a standardized quality assessment methodology facilitated the identification of both strengths and limitations within the primary studies. However, this evaluation is not without its limitations. The measurement of outcomes varied in terms of methods and units, which presents challenges in drawing reliable conclusions. The heterogeneity in the design of the supply kits and their components, as well as the definition of compliance with the “use of kit,” may hinder comparability. Furthermore, the majority of studies included in this evaluation were conducted in Africa and Asia, limiting the generalizability of their findings to other regions, such as Latin America. Another limitation of this study is the inconsistent presentation of associated factors, which prevents the identification of elements linked to the utilization of delivery kits. The I² statistic is frequently employed to demonstrate a notable degree of heterogeneity. Nevertheless, it is imperative to acknowledge that the I-square statistic might not consistently function as a clear-cut indicator of heterogeneity, given that the origin of heterogeneity could be linked to the particular command employed (specifically, the “Metan” command). A majority of the studies included were cross-sectional in nature, potentially introducing confounding variables that could impact the outcome variable.

Conclusion

In conclusion, the pooled proportion of clean delivery kits use LMIC was 54%. The finding differed by study design and publication year. Hence, it has bettered from 48.2 to 62.5 before and after 2010 studies respectively. These findings have important implications for healthcare providers, policy-making, and further research.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

NAG conceptualized the study: NAG and KAG contributed during data extraction and analysis: NAG, KDT and GA wrote the result interpretation: NAG, GB and YA Prepared the first draft: NAG, KAG and YA contributed during the conceptualization and interpretation of results and substantial revision: NAG, YA, KDT, GB and KAG Revised and finalized the final draft manuscript. All the authors read and approved the final version of the manuscript.

Funding

The authors received no specific funding for this work.

Data availability

All relevant data are within the Manuscript and its Supporting Information files.

Declarations

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participants

Not applicable because no primary data were collected.