Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

Samson Gebremedhin; Tilahun Bekele

doi:10.1371/journal.pone.0252247

. 2021 May 26;16(5):e0252247. doi: 10.1371/journal.pone.0252247

Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

Samson Gebremedhin ^1,^*, Tilahun Bekele ²

Editor: José Antonio Ortega³

PMCID: PMC8153429 PMID: 34038488

Abstract

Background

Inadequate or excess gestational weight gain (GWG) leads to multiple undesirable birth outcomes. Yet, in sub-Saharan Africa (SSA) little is known about the weight gain pattern in pregnancy. The purpose of the study is to estimate the average gestational weight gain (GWG) in sub-Saharan Africa (SSA) and to examined whether there had been recent improvements or not.

Methods

Based on cross-sectional anthropometric data extracted from multiple Demographic and Health Surveys conducted in SSA, we estimated the average GWG in the region. Pseudo-cohort design was used to reconstruct GWG trajectories based on aggregated data of 110,482 women extracted from 30 recent surveys. Trend in GWG between 2000 and 2015 was determined using the data of 11 SSA countries. Pre-pregnancy weight was estimated based on the weight of non-pregnant women at risk of conception.

Results

On average, women in SSA gain inadequate weight (6.6 kg, 95% confidence interval, 6.0–7.2) over pregnancy. No meaningful gain was observed in the first trimester; whereas, women in the second and third trimesters put on 2.2 and 3.2 kg, respectively. The highest weight gain (10.5, 8.2–12.9 kg) was observed in Southern African sub-region and the lowest in Western Africa (5.8, 5.0–6.6 kg). The GWG among women who had secondary or above education (9.5, 8.2–10.9 kg) was higher than women with lower education (5.0, 4.3–5.8 kg). Likewise, GWG in women from richest households (9.0, 7.2–10.7 kg) was superior to those from poorest households (6.1, 5.3–7.0 kg). The estimated recent (2015–20) mean GWG (6.6, 5.8–7.4 kg) was not significantly different from what had been at beginning of the new millennium (6.7, 5.9–7.5 kg).

Conclusion

In SSA GWG is extremely low and is not showing improvements.

Background

Gestational weight gain (GWG)–the weight increase between conception and just before the birth of the baby–is an important predictor of maternofoetal nutrition [1]. Data on GWG is imperative because insufficient gain is a proxy indicator for maternal and foetal undernutrition and shows strong correlation with intrauterine growth retardation and low birthweight [1, 2]. Low GWG may also increase risk of maternal and perinatal death [3]. On the other hand, excessive GWG leads to increased risks of macrosomia, caesarean delivery, preeclampsia, childhood obesity and maternal postpartum weight retention [4–7].

In 2009 the Institute of Medicine (IOM) of the National Academies put forth a new guideline on rate of GWG [8]. Though the guideline had been developed for the US population, the available few studies suggested its potential applicability in other settings as well [9–11]. According to the guideline, women with normal pre-pregnancy body mass index (BMI) and having singleton pregnancy should put on 11.5–16.0 kg of weight, assuming 0.5–2 kg gain in the first and 0.42 kg/week in the last two trimesters. The gain should also be as high as 18 kg in women with low pre-pregnancy BMI [8]. GWG is affected by multiple factors including socioeconomic status, dietary intake, co-morbidities, multifetal pregnancy and genetic factors [12].

In many low-income settings including the sub-Saharan African (SSA) limited information is available on the rate of GWG and there is no GWG guideline specific to the region [9, 13]. This is probably because estimation of GWG requires women to be followed from pre-pregnancy to near to childbirth, which is frequently infeasible in low-income settings where preconception care is rudimentary and births are largely happening at home [14, 15]. However, the available limited studies reported high rates of inadequate GWG in Africa and Asia [9, 13]. A systematic review concluded that 31% of women in Asia, as compared to 18–21% in Europe and US, gain inadequate gestational weight [10]. A study that reviewed the available few studies in SSA reported that the prevalence of optimal GWG ranged from 3% to 62% [13].

A recent study by Wang and colleagues (2020) based on modelling of Demographic and Health Surveys (DHS) reported that in 2015, the mean GWGs in Latin America and Caribbean region (11.8 kg) and Central Europe and Eastern Europe (11.2 kg) was considerable higher than the corresponding estimates for Sub-Saharan Africa (6.6 kg) and North Africa and Middle East (6.8 kg) [16]. However, this study did not provide information on the trends in GWG in the region, sub-national estimates and differences in GWG trajectories across basis maternal socio-demographic characteristics.

The purpose of the current study is to: (i) estimate mean GWG in SSA based on aggregated data from multiple nationally representative cross-sectional surveys and; (ii) to compare changes in mean GWG between 2000 and 2015 in the region. Information on the extent of GWG important because weight gain rate is a proxy indicator for maternal nutrition and helps to monitor the indicator for informing maternal nutrition related services and programs.

Methods

Study design

Pseudo-cohort design was used to estimate GWG based on aggregated cross-sectional anthropometric data of pregnant and non-pregnant women enrolled in recent and nationally representative DHS. Weight gain trajectories during pregnancy were reconstructed using the data of pregnant women at different gestational durations (1–9 months) whereas pre-pregnancy weight was estimated using the data of non-pregnant women at risk of conception.

Data source and inclusion criteria

The analysis was made based on the secondary data of Standard DHS. Other types of DHS including Continuous DHS, Malaria Indicator Surveys and AIDS Indicator Surveys that normally do not collected anthropometric data were not considered.

In order to estimate mean GWG, we analysed the data of 110,482 women in reproductive age (19,850 pregnant and 90,642 non-pregnant women) enrolled in DHS implemented in 30 SSA countries since 2010. With the intension of limiting the study to the recent period, surveys conducted before 2010 were excluded. At times when two or more eligible surveys were available for a country, the most recent was considered.

In order evaluate trends in GWG between 2000 and 2015, we compared the data of 11 SSA countries that implemented DHS around the beginning of the new millennium (1997–2003) and on or after 2015 (2015–2018). We were not limited to studies conducted in 2000 and 2015, rather we analysed studies conducted around these two periods because very few countries implemented surveys exactly on 2000 and 2015. In this specific analysis the data of 54,603 subjects (9,508 pregnant and 41,095 non-pregnant women) from surveys implemented between 1997 and 2003 and the data 49,127 subjects (9,055 pregnant and 40,072 non-pregnant women) from surveys taken place after 2014 were compared (Table 1).

Table 1. Demographic and health surveys included in the analysis.

Country	Year of survey
	Surveys used for estimating mean GWG		Control surveys for assessing trends in GWG
	Year	Sample size^‡	Year	Sample size^‡
Benin	2017/18	4,394	2001	3,655
Burkina Faso	2010	4,306	-	-
Burundi	2016/17	4,695	-	-
Cameroon	2018	3,879	1998	512
Chad	2014/15	6,451	1996/97	1,926
Comoros	2012	3,444	-	-
Congo	2011/12	1,898	-	-
Congo Democratic Republic	2013/14	4,636	-	-
Cote d’Ivoire	2011–12	2,625	-	-
Ethiopia	2016	7,275	2000	10,302
Gabon	2012	2,725	-	-
Gambia	2013	2,738	-	-
Ghana	2014	2,622	-	-
Guinea	2018	2,891	1999	1,358
Kenya	2014	6,134	-	-
Lesotho	2014	2,292	-	-
Liberia	2013	1,235	-	-
Malawi	2015/16	2,705	2000	7,048
Mali	2018	3,107	2001	7,504
Namibia	2013	1,504	-	-
Niger	2012	2,752	-	-
Nigeria	2018	8,177	2003	4,792
Rwanda	2014/15	3,459	2000	6,788
Senegal	2010/11	3,500	-	-
Sierra Leone	2013	3,845	-	-
South Africa	2016	1,324	-	-
Tanzania	2015/16	6,458	-	-
Togo	2013/14	2,622	-	-
Uganda	2016	2,829	2000/01	3,624
Zimbabwe	2015	3,960	1999	3,094
Total		110,482		50,603

Open in a new tab

^‡ sample size for both pregnant and non-pregnant women; GWG–Gestational weight gain.

S1 File outlines the process of refining the dataset for the final analysis. Regarding pregnant women, data lines having complete information on gestational age and body weight were eligible for analysis. For non-pregnant women those at risk of conception were eligible. Women at risk of conception was operationally defined as menstruating women in reproductive age who were not abstaining from sex and not using any modern or natural contraceptive methods at the time of the survey. Furthermore, women who recently gave birth (in the last one year) were excluded from the analysis (S1 File).

Sampling design and approach of data collection in DHS

Demographic and Health Surveys are standardized national surveys being regularly implemented by the DHS Program and national agencies in many low- and middle-income countries for monitoring vital statistics and population health indicators [17]. The DHS identify eligible subjects including women 15–49 years of age, using two-stage multiple cluster sampling approach that is intended to provide representative data at national and subnational (regions or states) levels in both urban and rural settings [18].

In DHS socio-demographic data are collected using standardized questionnaires and anthropometric (weight and height) measurements are taken following standard procedures. Pregnancy status is determined based on self-report without any further validation. Gestational duration is determined in months (1–9 months) based on self-report and the date of last normal menstrual period (LNMP) and is approximated to the nearest month.

Estimation of pre-pregnancy weight

Pre-pregnancy weight was estimated based on the mean body weight of women at risk of conception. Sexually active and menstruating women, who were not using any contraceptive at the time of the survey were considered to be at risk of conception. In this regard, the data of women who gave birth in the last one year was not used considering that postpartum weight retention can make us to overestimate pre-pregnancy weight. The weight of women at risk of conception, rather than that of non-pregnant women was used because the latter are likely to be systematically different from pregnant women in basic socio-demographic characteristics.

Reconstructing weight gain patterns during pregnancy

Gestational weight gain during pregnancy was reconstructed by taking the mean weight of pregnant women at different gestational months (1–9 months) and assuming that pregnant women in the region have passed through the month-specific weight gain trajectories.

Estimation of weight at the time of delivery

As described earlier, in DHS gestational duration is measured to the nearest month. Accordingly, maternal weight at a given month represents the average gestational weight in the +/- 2 gestational weeks range. Accordingly, the average gestational weight at the 9 month underestimates maternal weight at the time of birth. According maternal weight at the time of delivery was estimated by extrapolating the weight gain rate in the third trimester (6–9 months) to the estimated date of birth (40^th week or 9 months and 10 days).

Data management and analysis

We used SPSS v24 for data management and analysis. The datasets for all the eligible surveys were downloaded from the DHS Program database [19] and merged into a spreadsheet.

Weighted data analysis was employed based on the sampling weights readily available in the data and post-stratification weight developed using the 2020 population size of the countries [20]. DHS calculates sampling weights based on sample selection probabilities at household and individual levels [18]. Post-stratification weight was computed by dividing the 2020 total population size of that country to the total population size of the countries represented in the dataset for the same year. The ultimate data weight was calculated by multiplying the sample weight readily available in the dataset by the post-stratification weight with linearization to balance the weighted and unweighted sample size. The analysed data is publicly available from https://dhsprogram.com/data/.

When available, gestational age was determined based on LNMP otherwise it was estimated based on self-report of the women. Very small proportion (about 0.3%) of pregnant women reported gestational age of 10 months and during analysis it was recoded to 9 months.

Average gestational weight gain was computed by subtracting estimated pre-pregnancy weight with estimated weight at the time of delivery. Weight gain in the first trimester was estimated by subtracting pre-pregnancy weight from estimated weight at the third gestational month. Second trimester weight gain was computed by subtracting weight at third gestational month from weight at the sixth month. Similarly, third trimester weight gain was computed by subtracting weight at sixth gestational month from estimated weight at the time of delivery.

Total GWG and trimester-specific weight gain rates were estimated at regional and sub-regional levels However, country-level estimates have not been provided as many national surveys enrolled inadequate number of pregnant women at different gestational durations. Sub-regional estimates are provided by classifying the region into four geographic areas (Eastern, Southern, Western and Central) according to the African Union classification system [21]. Furthermore, the countries were sub-divided as low-income ($1,035 or less) or lower-middle or upper-income ($1,036 to $12,535) economies based the classification approach proposed by the World Bank [22].

GWG rates were compared across levels of maternal age, place of residence (urban vs rural), maternal educational status, household wealth index and maternal height. Wealth index was computed as an index of household economic standing using Principal Component Analysis based on ownership of selected household assets and materials used for house construction [18]. For comparing weight gain trajectories across levels of the aforementioned variables, 95% confidence intervals (CIs) for mean GWG were used.

Results

Basic characteristics

The data of 88,602 non-pregnant and 21,822 pregnant women from 30 countries including 13 Western, 6 Eastern, 6 Central and 5 Southern African countries, was used for estimating mean GWG in SSA. The mean (± SD) age of the women was 26.9 (± 10.1) years and nearly one-third (30.2%) were under the age of 20 years. Two thirds were rural residents and one third had no formal education. The number of pregnant women across the gestational months ranged from 934 in the first to 3,204 in the eighth month. Women in the first trimester were slightly underrepresented (25.6%) while women in the second trimester were overrepresented (40.5%). About a quarter (26.8%) the women were stunted (height < 155 cm) and among non-pregnant women 19 years or above 12.1% were underweight and 24.4% were overweight/obese (Table 2).

Table 2. Basic characteristics of the study subjects, sub-Saharan Africa, 2010–2018.

Characteristics (n = 110,428)	Frequency	Percentage
Pregnancy status
Pregnant	21,822	19.8
Non-pregnant	88,606	80.2
Gestational trimester (n = 21,822)
First (1–3)	5,682	25.6
Second (4–6)	8,897	40.5
Third (7–9)	7,399	33.7
Gestational month (n = 21,822)
1	934	4.3
2	2,070	9.5
3	2,638	12.1
4	2,640	12.1
5	3,182	14.6
6	3,011	13.8
7	3,104	14.2
8	3,204	14.7
9	1,038	4.8
Type of place of residence
Urban	38,085	34.5
Rural	72,343	65.5
Maternal age
15–19	33,309	30.2
20–34	49,402	44.7
35–49	27,717	25.1
Level of education (n = 110,417)
No education	37,487	33.9
Primary	35,962	32.6
Secondary	31,562	28.6
Higher	5,406	4.9
Wealth index
Poorest	18,201	16.5
Poorer	19,075	17.3
Middle	20,739	18.8
Richer	22,836	20.7
Richest	29,577	26.8
Maternal height (n = 110,243)
< 145 cm	2,030	1.8
145–154 cm	27,612	25.0
155 or above	80,601	73.0
Body mass index in non-pregnant women 19 years or above (n = 62,011)
Underweight	7484	12.1
Normal	39380	63.5
Overweight or obese	15,148	24.4
Mean (± standard deviation) weight (kg)
Non-pregnant	57.1 (± 12.7)
First trimester (1–3)	57.0 (± 11.0)
Second trimester (4–6)	59.4 (± 11.0)
Third trimester (7–9)	62.1 (± 11.2)

Open in a new tab

Estimated gestational weight gain in sub-Saharan Africa

Fig 1 depicts the GWG trajectory in SSA. GWG increased by 6.6 kg (95% CI: 6.0–7.2) from the estimated pre-pregnancy weight of 57.1 kg (95% CI: 57.0–57.2) to 63.7 kg (95% CI: 63.0–64.4) at the end of pregnancy. No meaningful weight gain was observed in the first trimester; whereas, average gains were 2.2 kg in second and 3.2 kg in the third trimesters, respectively (Fig 1).

Table 3 compares the estimated pre-pregnancy weight and GWG in different geographic regions and economic categories of 30 SSA countries. The Southern Africa (59.8 kg: 95% CI: 59.5–60.1) and Eastern Africa (55.2 kg: 95% CI: 55.0–55.3) sub-regions had the highest and lowest pre-pregnancy weight, respectively. On the other hand, the highest GWG (10.5 kg: 95% CI: 8.2–12.9) was in Southern African sub-region and the lowest was in Western Africa (5.8 kg: 95% CI: 5.0–6.6). In Central and Eastern regions, the estimated GWG rates were 6.8 kg (95% CI: 5.4–8.1) and 6.6 kg (95% CI: 5.5–7.6), respectively.

Table 3. Gestational weight gain by household and maternal characteristics, sub-Saharan Africa, 2010–2018.

Characteristics	Estimated mean weight (kg) (95% CI)				Estimated gestational weight gain (kg) (95% CI)
Characteristics	Pre-pregnancy	Third month	Sixth month	End of pregnancy	Estimated gestational weight gain (kg) (95% CI)
Geographic classification of the country
Southern	59.8 (59.5–60.1)	59.6 (58.2–61.1)	64.4 (62.7–66.1)	70.3 (67.7–73.0)	10.5 (8.2–12.9)
Central	56.1 (55.9–56.3)	56.1 (55.2–56.9)	59.3 (58.4–60.1)	62.9 (61.3–64.5)	6.8 (5.4–8.1)
Eastern	55.2 (55.0–55.3)	55.8 (55.0–56.6)	59.8 (59.1–60.6)	61.7 (60.6–62.9)	6.6 (5.5–7.6)
Western	57.8 (57.7–58.0)	57.6 (56.9–58.2)	60.7 (60.1–61.4)	63.6 (62.7–64.6)	5.8 (5.0–6.6)
Economic classification of the country
Low-income	55.4 (55.3–55.5)	55.5 (55.0–56.0)	58.6 (58.1–59.1)	62.2 (61.3–63.0)	6.4 (6.1–7.7)
Lower-middle-income	58.9 (58.8–59.1)	58.6 (57.9–59.3)	62.7 (62.0–63.4)	65.7 (64.6–66.9)	6.8 (5.8–7.8)
Maternal age (years)
15–19	52.5 (52.4–52.6)	53.7 (53.0–54.4)	56.8 (56.1–57.6)	59.4 (58.0–60.7)	6.9 (5.7–8.1)
20–34	57.8 (57.7–57.9)	57.1 (56.6–57.6)	60.9 (60.3–61.4)	64.1 (63.2–64.9)	6.2 (5.5–6.9)
35 or above	58.7 (58.5–58.9)	59.2 (57.9–60.4)	62.4 (61.2–63.5)	65.2 (63.4–67.0)	6.5 (4.8–8.1)
Place of residence
Urban	60.4 (60.2–60.5)	60.2 (59.3–61.1)	65.0 (64.2–65.9)	67.9 (66.6–69.3)	7.6 (6.4–8.7)
Rural	55.5 (55.4–55.6)	55.6 (55.2–56.0)	58.4 (58.0–58.9)	61.7 (60.9–62.4)	6.2 (5.5–6.8)
Educational status
No formal education	55.8 (55.7–55.9)	55.1 (54.6–55.7)	58.2 (57.6–58.7)	60.8 (59.9–61.7)	5.0 (4.3–5.8)
Primary education	56.3 (56.2–56.4)	56.2 (55.5–56.9)	60.0 (59.3–60.7)	63.8 (62.6–65.0)	7.6 (6.5–8.6)
Secondary education	58.8 (58.6–58.9)	60.6 (59.7–61.5)	64.6 (63.7–65.6)	68.3 (66.8–69.8)	9.5 (8.2–10.9)
Household wealth index
Poor	53.5 (53.3–53.6)	54.6 (53.9–55.3)	56.7 (56.0–57.4)	59.6 (58.6–60.7)	6.1 (5.3–7.0)
Poorer	54.9 (54.7–55.1)	55.7 (55.0–56.4)	59.4 (58.6–60.1)	61.8 (60.6–63.0)	6.9 (5.9–8.0)
Middle	56.4 (56.3–56.6)	56.2 (55.3–57.0)	59.0 (58.2–59.8)	63.2 (61.7–64.7)	6.8 (5.5–8.1)
Richer	58.2 (58.1–58.4)	57.7 (56.6–58.7)	62.9 (61.9–63.9)	66.3 (64.4–68.1)	8.0 (6.4–9.7)
Richest	60.7 (60.5–60.9)	62.3 (61.0–63.6)	66.7 (65.4–67.9)	69.7 (67.8–71.6)	9.0 (7.2–10.7)
Maternal height
< 155 cm	51.1 (50.9–51.2)	51.8 (51.1–52.5)	54.2 (53.6–54.9)	57.4 (56.2–58.6)	6.3 (5.3–7.3)
> = 155 cm	58.9 (58.8–59.0)	58.6 (58.2–59.1)	62.5 (62.0–63.0)	65.2 (64.4–66.0)	6.3 (5.6–7.0)

Open in a new tab

The estimated GWG in low-income countries was 6.4 kg (95% CI: 6.1–7.7) while the corresponding level for lower-middle or upper-income countries was 6.8 kg (95% CI: 5.8–7.8). The overlap in the confidence intervals suggested absence of statistically significant difference between the two groups (Table 3).

The GWG pattern was also compared across selected socio-demographic characteristics. The weight gain among women who had secondary or above level of education (9.5 kg: 95% CI: 8.2–10.9) was significantly higher than women who had no formal education (5.0 kg: 95% CI: 4.3–5.8). Similarly, the GWG among women from richest households (9.0 kg: 95% CI: 7.2–10.7) was also superior to those from poorest households (6.1 kg: 95% CI: 5.3–7.0). No significant differences were observed across levels of maternal age, place of residence and maternal height (Table 3).

Trends in gestational weight gain

Fig 2 compares the GWG trajectory in 11 SSA countries between the beginning of the new millennium (1997–2003) and after 2014. Over the period, the pre-pregnancy weight significantly increased from 54.4 kg (95% CI: 54.3–54.5) to 56.9 kg (95% CI: 56.8–57.1). Nevertheless, the GWG remains more or less the same: 6.7 kg (95% CI: 5.9–7.5) and 6.6 kg (95% CI: 5.8–7.4), respectively for the two periods (Fig 2).

Discussion

In SSA information on GWG remains scarce. This study, based on aggregated cross-sectional data from multiple national surveys, reconstructed weight gain trajectories during pregnancy and estimated that mean GWG is very low in the region and did not show meaningful changes over the last 15 years. Furthermore, GWG showed significant heterogeneity across the sub-regions of SSA and levels of socio-economic status including maternal educational status and household wealth index.

According to IOM, pregnant women with normal pre-pregnancy BMI should gain 11.5–16.0 kg of weight and women with low pre-pregnancy BMI are expected to gain as high as 18 kg of weight [8]. Though the utility the IOM recommendation in settings outside the United States can still be contended [23], the mean GWG that we estimated (6.6 kg) is only about half of the minimum recommended gain for women with normal baseline BMI. A study that estimated GWG in SSA and India based on surveys conducted between 2000 and 2010 also reported GWG in both regions is grossly inadequate at 7 kg [24]. A systematic review also found that the prevalence of inadequate weight gain exceeded 50% in most of the studies from SSA [13]. The low GWG trajectory observed in the region is reflective of several underlying determinants including suboptimal maternal nutrition, poor access to social services and low socio-economic status [12]. Genetic factors can also significantly affect GWG [12, 25].

Our analysis suggested that GWG in SSA is only increased by 6.6 kg from the estimated pre-pregnancy weight of 57.1 kg to 63.7 kg at the end of pregnancy. This is in conformation with the findings of a recent modelling analysis of DHS by Wang and colleagues [15]. The study reported that in 2015, the estimated GWG in SSA was 6.6 kg, which was much lower than that of Latin America and Caribbean (11.8 kg), Central and Eastern Europe (11.2 kg) and Central Asia (11.2 kg). On the other hand, the GWG in North Africa and Middle East found to be grossly inadequate (6.8 kg). In general, our study and that of Wang et al [15]. followed similar pseudo cohort design to estimate the mean total GWG. Yet, the statistical analysis approaches employed were different.

Sub-regional analysis also suggested that the mean GWG in southern African region (10.5 kg) is considerably higher than the eastern (6.6 kg) and western (5.8kg) sub-regions. This is likely the indirect reflection of the better socioeconomical status of southern African countries like Namibia and South Africa included in the analysis.

We also observed that over the last 15 years of so, pre-pregnancy weight has significantly increased on average by 2.4 kg in SSA yet GWG showed no significant improvement. The finding looks paradoxical because the better nutrition that resulted in improvements in pre-pregnancy weight should also uplift GWG. A study that compared the GWG trajectories among Indians and Africans also reported that both regions have similar GWG despite the pre-pregnancy weight was substantially lower by 8 kg among Indian women [24]. Interventional and prospective studies are required to evaluate why GWG is not improving in SSA despite nutritional improvements observed over the last two decades.

In the current study better maternal education and household wealth standing were significantly associated with higher rates of GWG. The same had been witnessed by other studies conducted in low- or middle-income countries. In Brazil, pregnant women with seven or less years of schooling were two times more likely to encounter insufficient weight gains as compared to those with better education; however, significant difference have not observed between categories of economic class [26]. In Ilam province of Iran, prevalence of inadequate weight gain was higher among women with lower educational and income status [27]. Better socioeconomic status is likely to increase GWG though advancing access to better nutrition and social services.

The findings of the study should be interpreted in consideration of the following strength and limitations. The major strength is that GWG was estimated based on data coming from several nationally surveys and this assures the representativeness of the findings at regional and sub-regional levels. Conversely, the following limitations have to be noted. (1) We estimated GWG using aggregated cross-sectional, rather than the individual-based longitudinal data. Consequently, prevalence of inadequate or excess weight gain could not be determined and the heterogeneity observed across levels of socio-economic variables could not be statistically adjusted for potential confounders. In addition, GWG estimates cannot be provided for different levels of pre-pregnancy BMI. This was because BMI values were not calculated for pregnant women and reconstructing GWG based on BMI levels was not possible. (2) We determined mean GWG assuming normal gestation length but in actual sense GWG is directly affected by gestational duration [12] and significant proportion (about 15% in total) of pregnancies end up in either pre- or post-term births [28, 29]. This may have resulted in over- or under-estimation of GWG. (3) DHS determine pregnancy status and gestational duration based on self-reports without any further validation. Future more, gestational duration is measured in months rather than weeks. These may have made us to over- or under-estimate the extent of GWG in SSA.

Furthermore, we assessed adequacy of GWG considering the US IOM guideline as a standard. Though the applicability of the IOM standard outside US especially in Asia had been reported [9–11], its usability in African setup had not been investigated. We used the IOM guidelines as a standard because we have not come across with SSA-specific GWG guideline.

While we estimate the trends in GWG between 2000 and 2015 we encountered with the problem that nearly all countries do not have data in these specific years. Accordingly, we have included surveys conducted around 2000 (1997–2003) and 2015 (2015–2018). This decision may have made us to marginally over or underestimate the rate of GWG between 2000 and 2015.

In the current study the presence or absence of statistically significant differences between two mean GWG trajectories was assessed using the overlap between the two associated confidence intervals. We were not able to calculate p-values because estimation of GWG was not made based on individual-level data. While the method of examining differences based on overlap is simple and convenient it may lead to type II error especially when the associated p-values are marginally significant [30, 31].

Though 30 of the 46 SSA are represented in the analysis, the generalizability of the findings to the entire region can still be doubted. This is because there is no guarantee that countries that recently implemented DHS and included in the analysis are comparable with other countries in terms of pertinent variables that many affect GWG. Furthermore, we could not be able to provide country-level estimates because reconstructing GWG trajectories using aggregated data requires large sample size and many of the national surveys enrolled inadequate number of pregnant women.

Conclusion

The mean GWG in SSA is very low and did not show meaningful changes over the last 15 years. Women from southern African region tend to have better weight gain trajectories than the other sub-regions of the continent. Similarly, women with lower socio-economic status tend to put on smaller weights during pregnancy.

Supporting information

S1 File. Flow chart of the study.

(DOCX)

Click here for additional data file.^{(29.4KB, docx)}

Acknowledgments

The authors acknowledge the DHS Program for making the DHS data accessible for public use.

Data Availability

The analysed data is publicly available from https://dhsprogram.com/data/.

Funding Statement

The authors received no specific funding for this work.

References

1.Kleinman KP, Oken E, Radesky JS, Rich-Edwards JW, Peterson KE, Gillman MW. How should gestational weight gain be assessed? A comparison of existing methods and a novel method, area under the weight gain curve. International Journal of Epidemiology 2007; 36(6): 1275–1282. 10.1093/ije/dym156 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Strauss RS, Dietz WH. Low maternal weight gain in the second or third trimester increases the risk for intrauterine growth retardation. The Journal of Nutrition 1999; 129(5): 988–993. 10.1093/jn/129.5.988 [DOI] [PubMed] [Google Scholar]
3.Ukah UV, Bayrampour H, Sabr Y, Razaz N, Chan W, Lim KI, et al. Association between gestational weight gain and severe adverse birth outcomes in Washington State, US: A population-based retrospective cohort study, 2004–2013. PLoS Med 2019; 16(12): e1003009. 10.1371/journal.pmed.1003009 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Chung JGY, Taylor RS, Thompson JMD, Anderson NH, Dekker GA, Kenny LC, et al. Gestational weight gain and adverse pregnancy outcomes in a nulliparous cohort. European Journal of Obstetrics & Gynecology and Reproductive Biology 2012; 167(2): 149–53. 10.1016/j.ejogrb.2012.11.020 [DOI] [PubMed] [Google Scholar]
5.Bhavadharini B, Anjana RM, Deepa M, Jayashree G, Nrutya S, Shobana M, et al. Gestational weight gain and pregnancy outcomes in relation to body mass index in Asian Indian women. Indian Journal of Endocrinolgy and Metabolism 2017; 21(4): 588–593. 10.4103/ijem.IJEM_557_16 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Ludwig DS, Currie J. The relationship between pregnancy weight gain and birth weight: a within family comparison. Lancet; 2010: 376(9745), 984–990. 10.1016/S0140-6736(10)60751-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Truong YN, Yee LM, Caughey AB, Cheng YW. Weight gain in pregnancy: does the Institute of Medicine have it right? American Journal of Obstetrics and Gynecology 2015; 212: 362.e1–8. 10.1016/j.ajog.2015.01.027 [DOI] [PubMed] [Google Scholar]
8.Institute of Medicine. Report Brief: Weight gain during pregnancy: Re-examining the guidelines. Washington (DC): National Academies Press; 2009. [Google Scholar]
9.Goldstein RF, Abell SK, Ranasinha S, Misso ML, Boyle JA, Harrison CL, et al. Gestational weight gain across continents and ethnicity: systematic review and meta-analysis of maternal and infant outcomes in more than one million women. BMC Medicine 2018; 153: 16. 10.1186/s12916-018-1128-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Nomura K., Nagashima K., Suzuki S. Itoh H. Application of Japanese guidelines for gestational weight gain to multiple pregnancy outcomes and its optimal range in 101,336 Japanese women. Sci Rep 2019; 9: 17310. 10.1038/s41598-019-53809-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Jh Wie, Park IY, Namkung J, Seo HW, Jeong MJ, Kwon JY. Is it appropriate for Korean women to adopt the 2009 Institute of Medicine recommendations for gestational weight gain? PLoS ONE 2017; 12(7): e0181164. 10.1371/journal.pone.0181164 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Rasmussen KM, Yaktine AL (eds). Weight gain during pregnancy: Re-examining the Guidelines. Institute of Medicine (US) and National Research Council (US) Committee to Re-examine IOM Pregnancy Weight Guidelines. Washington (DC): National Academies Press; 2009. [PubMed] [Google Scholar]
13.Asefa F, Cummins A, Dessie Y, Hayen A, Foureur M. Gestational weight gain and its effect on birth outcomes in sub-Saharan Africa: Systematic review and meta-analysis. PLoS ONE; 2020: 15(4), e0231889. 10.1371/journal.pone.0231889 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Doctor HV, Nkhana-Salimu S, Abdulsalam-Anibilowo M. Health facility delivery in sub-Saharan Africa: successes, challenges, and implications for the 2030 development agenda. BMC Public Health 2018); 18(1): 765. 10.1186/s12889-018-5695-z [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Wang D, Wang M, Darling AM, Perumal N, Liu E, Danae G et al. Gestational weight gain in low-income and middle-income countries: a modelling analysis using nationally representative data. BMJ Global Health. 2020;5:e003423. 10.1136/bmjgh-2020-003423 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Dean SV, Lassi ZS, Imam AM, Bhutta ZA. Preconception care: Closing the gap in the continuum of care to accelerate improvements in maternal, newborn and child health. Reproductive Health; 2014: 11(Suppl 3), S1. 10.1186/1742-4755-11-S3-S1 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Corsi DJ, Neuman M, Finlay JE, Subramanian SV. Demographic and health surveys: a profile. International Journal of Epidemiology 2012; 41(6): 1602–1613. 10.1093/ije/dys184 [DOI] [PubMed] [Google Scholar]
18.MEASURE DHS/ICF International. Demographic and Health Surveys methodology: Sampling and household listing manual. Calverton, MD: ICF International; 2012. [Google Scholar]
19.The DHS Program. Retrieved from https://dhsprogram.com/Data/ on Nov 02, 2020.
20.United Nations Population Division. World population prospects. Retrieved from https://population.un.org/wpp/ on Nov 02, 2020.
21.African Union. Member states. Retrieved from https://au.int/en/member_states/countryprofiles2 on Nov 02, 2020.
22.Bank World. Country and lending groups: Country classification. Retrieved from https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups. on November 02, 2020. [Google Scholar]
23.Arora P Aeri BT. Gestational weight gain among healthy pregnant women from Asia in comparison with Institute of Medicine (IOM) guidelines-2009: A systematic review. Journal of Pregnancy. 2019; 2019: 3849596. 10.1155/2019/3849596 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Coffey D. Pre-pregnancy body mass and weight gain during pregnancy in India and sub-Saharan Africa. PNAS 2015; 112(11): 3302–3307. 10.1073/pnas.1416964112 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Meng Y, Groth SW, Li D. The association between obesity-risk genes and gestational weight gain is modified by dietary intake in African American women. Journal of Nutrition and Metabolism 2018: 5080492. 10.1155/2018/5080492 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Fraga AC, Filha MM. Factors associated with gestational weight gain in pregnant women in Rio de Janeiro, Brazil, 2008. Cadernos de Saúde Pública 2014; 30(3): 633–644. 10.1590/0102-311x00085313 [DOI] [PubMed] [Google Scholar]
27.Dolatian M, Sharifi N, Mahmoodi Z, Fathnezhad‐kazemi A, Bahrami‐vazir E, Rashidian T. Weight gain during pregnancy and its associated factors: A Path analysis. NursingOpen 2020; 7(5): 1568–1577. 10.1002/nop2.539 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Galal M, Symonds I, Murray H, Petraglia F, Smith R. Postterm pregnancy. Facts, Views & Vision in ObGyn 2012; 4(3): 175–187. [PMC free article] [PubMed] [Google Scholar]
29.Tucker J, McGuire W. Epidemiology of preterm birth. BMJ 2004; 329(7467): 675–678. 10.1136/bmj.329.7467.675 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Austin PC, Hux JE. A brief note on overlapping confidence intervals. Journal of Vascular Surgery 2002; 36(1):194–5. 10.1067/mva.2002.125015 [DOI] [PubMed] [Google Scholar]
31.Schenker N, Gentleman JF. On judging the significance of differences by examining the overlap between confidence intervals. The American Statistician 2001; 55(3): 182–186. 10.1198/000313001317097960 [DOI] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0252247.r001

Decision Letter 0

José Antonio Ortega

1 Mar 2021

PONE-D-21-02601

Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

PLOS ONE

Dear Dr. Gebremedhin,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The paper looks promising but it currently is lacking in several respects.

With respect to the introduction, the paper should provide background that puts the manuscript into context and allows readers outside the field to understand the purpose and significance of the study and include a brief review of the key literature. In the present case, as reviewer 1 points out, the article is missing a key reference that looks at the same topic with the same data and similar methods but for more countries. It is basic in this case to note the differences in approach or in scope. Note that otherwise PLOS ONE policy states that if a submitted study replicates or is very similar to previous work, authors must provide a sound scientific rationale for the submitted work and clearly reference and discuss the existing literature. Submissions that replicate or are derivative of existing work will likely be rejected if authors do not provide adequate justification.

Regarding methods, there are some limitations that should be addressed.

A prior consideration, is that the methods should be adequately described to allow for replication. As noted by reviewer 1, this is not the case, and the missing information might have confused, for instance, reviewer 3 in understanding the method.

First, note the concerns regarding the use of relevant guidelines as suggested by reviewers 1, 2 and 3.

Second, the problem of the relevant comparison group raised by reviewers 2 and 3. Reviewer 2 suggestion of reweighting non-pregnant women to match pregnant women according to weight is interesting and doable. If you carry this out show, at least for some instances, both sets of estimates to see if the strange patterns for first trimester disappear. Alternatively, a more sophisticated approach would be the use of propensity score matching to ensure that women in the synthetic control group (nonpregnant) are comparable to those pregnant.

Comment by reviewer 3 regarding grouping by weight I don’t believe it is feasible since you do not observe prepregnancy weight. However, you should bear it in mind when analyzing the patterns of estimated gestational weight gain. Since you have computed the proportions underweight and overweight you could see if the pattern of estimated weight gain at the survey level correlate with the proportions underweight and overweight. That will be very helpful for the discussion and also to tease out possible factors behind the much smaller GWG in Africa compared to other developing regions according to the missing source noted by reviewer 1.

Third, as noted by reviewer 2, not using data from women in the first two months of pregnancy is probably a good idea. For instance, it is known that DHS patterns of reported pregnancy termination are extremely low in most sub-Saharan African countries (https://doi.org/10.1371/journal.pone.0221178) suggesting that many early terminations are missing because the woman was unaware of the pregnancy. Since this proportion would vary from survey to survey and according to socioeconomic characteristics, it would induce a bias in estimate weight gain.

In your revision note that PLOS ONE endorses the use of the STROBE checklist (http://www.strobe-statement.org) for observational studies such as this one. Make sure that all the questions in the checklist are addressed.

Please submit your revised manuscript by Apr 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

José Antonio Ortega, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this study by Gebremedhin and Bekele, the authors used data from the Demographic and Health Surveys (DHS) program to estimate the mean gestational weight gain (GWG) in sub-Saharan African countries. The authors reported that the level of GWG in sub-Saharan Africa was low and did not improve with time.

The use of nationally representative data and the extensive subgroup analyses are the primary strengths of this work. The limitations of this study were the use of cross-sectional data and the inability to estimate the prevalence of women with inadequate or excessive GWG. The manuscript is well-written in general, but I find that the statistical methodology of the analysis is inadequately described. Please find my specific questions, comments, and suggestions below.

1) This work is similar to a previous study in terms of the research question. In Wang et al. (Gestational weight gain in low-income and middle-income countries: a modelling analysis using nationally representative data. BMJ global health. 2020), a multilevel modeling framework was applied to the DHS data to estimate the mean GWG in the year 2015 in all low- and middle-income countries and regions of the world, including in sub-Saharan Africa. Surprisingly, the Wang et al. paper was not referenced or mentioned in this work. Discussing the similarities and differences in the results and methodology between the two works will be necessary.

2) On page 4: “assuming 0.5-2 kg gain in the first and 0.45 kg/week in the last two trimesters." However, the weekly weight gain in the second and third trimesters recommended in the IOM guidelines is 0.42 kg (0.35-0.50). It would be good to provide a reference for the value of 0,45 kg/week or revise it to be consistent with the IOM guidelines.

3) I find the statistical methods of the study poorly described in general. There is virtually no explanation on how the pseudo-cohort design was reconstructed using the cross-sectional data in the DHS Program, how the total GWG and trimester-specific GWG were computed, or how the pre-pregnancy weight was estimated using the weights of women at risk of conception. I suggest that the statistical analysis and technical details of the methods be much more adequately described.

4) It is not straightforward to combine the survey data across 30 countries into one unified dataset while still preserving the sampling designs (weight, stratification, clustering) appropriate for each dataset. Therefore, I suggest clarifying how the datasets were combined and how stratification and clustering of the original surveys were accounted for. The explanation on the derivation of the new weights using the 2020 population size in the analysis also appears brief to me and will benefit from more details.

5) In Table 2, when classifying women into underweight, normal weight, and overweight or obese, were the WHO cutoffs (i.e., < 18.5, 18.5-<25, >=25) used? For adolescents aged 15 to 19 years old, were the WHO growth references (based on BMI-for-age Z-scores) for children/adolescents used instead of the cutoffs for adults?

6) Page 13: “The overlap in the confidence intervals suggested absence of statistically significant difference between the two groups.” This statement is incorrect in that two CIs may well overlap while still having a statistically significant difference. I suggest the statistical significance in the subgroup analyses be rigorously and quantitatively determined. Otherwise, a qualitative description of the difference may be sufficient without resorting to a discourse on statistical significance.

7) It appears that 11 countries were included in the analysis for 1997-2003, while 30 countries were included in the analysis for 2015-2020. As a result, quite different numbers of countries were used to compare the two time periods (1997-2003 and post-2015). This discrepancy greatly complicates the comparison, and the interpretation of the comparison becomes tricky and unclear.

Reviewer #2: This paper uses DHS data to estimate gestation weight gain in SSA as a whole. It also presents estimates for regions within SSA. I think it is an important project and is well-written and well-presented.

I'd suggest the following major revisions:

1. When you compute the average (body) weight of women who are at risk of conception, I suggest creating and using (statistical) weights that make the non-pregnant group match the age profile of currently pregnant women. The reason to do this is that older women in their 40s with low fecundity may report menstruating and not using contraception, but are unlikely to become pregnant. Yet, (I believe) the current computation considers them equally “at risk of conception” as women in their 20s who are not using contraception. Therefore, their body weights count equally in the average weight of potentially pregnant women. If age is correlated with weight (which it is in many societies), you would be overestimating the average weight of women at risk for pregnancy, which would lead you to underestimate GWG. Both the tables and Figure 1 suggest this is happening because the (body) weights of pre-pregnant women are higher, on average, than the body weights of women in their first trimester. (Although a very small number of women do lose weight in the first trimester, it is not the norm.)

If adjusting for the age profile of potentially pregnant women doesn't bring the pre-pregnancy weight down below first trimester weights, you could try creating statistical weights based on other factors as well (such as number of children or education) but I think age will cover it.

2. I would suggest not including results for women in the first and second months of pregnancy. The women who report very early pregnancy are likely systematically different from the pregnant population, and from those who report 2nd and 3rd trimester pregnancies. I think they will be more educated and perhaps come from less gender-conservative backgrounds. Indeed the fact that such a small fraction of pregnant women report being 1 or 2 months pregnant reinforces the idea that there is under-reporting of pregnancy in these months.

I'd suggest the following minor revisions:

1. The second paragraph of the introduction reads: “In 2019 the Institute of Medicine (IOM) of the National Academies put forth a new guideline on rate of GWG.” It would be good to mention that this guideline is for the US population. If the authors know of any guidelines specific to any SSA country/countries it would be helpful to mention them.

2. Please mention how you code gestational age in the main text. The DHS asks both month pregnant and last period; in some cases the information is missing for one variable and not for the other. Which do you preference? Also, how often is gestational month 10 recorded? You might consider clubbing it with month 9 since it basically descries a full-term pregnancy, but the way that different societies talk about full term pregnancies differs slightly.

3. Please include a more in-depth discussion of the statistical weights in the text.

Reviewer #3: Thank you for the opportunity to review this interesting manuscript. It is an important and interesting topic that can have the potential to improve outcomes for women in sub-Saharan Africa. The paper is well written using clear language. There are, however, some comments I make for your consideration before I believe the paper is ready for publication.

Abstract

1. The authors can add comma after the word ‘On average’ in the result section of the abstract.

2. Please be consistent while reporting mean gestational weight gain with its confidence interval. Sometime you put the mean outside of the bracket and confidence interval inside the bracket while you put all in the bracket in the other time, for example, 6.6 kg (95% confidence interval, 6.0-7.2) vs (10.5, 8.2-12.9 kg)

3. Please be consistent while using ‘kg or kgs’

Background

4. The authors defined Gestational weight gain (GWG) as “the weight increase between conception and just before the birth of the infant.” Could the term ‘baby’ is more appropriate than ‘infant’ here?

5. At the beginning of the second paragraph, the authors stated “In 2019 the Institute of Medicine (IOM) of the National Academies put forth a new guideline on rate of GWG”. I am not sure if the National Academies released a GWG guideline in 2019. Please correct me if I am wrong.

Methods

6. Not clears why and how non-pregnant women were included in the analysis of GWG.

7. I am a bit confused about data inclusion criteria. In the abstract section, the authors stated “Trend in GWG between 2000 and 2015 was determined using the data of 11 SSA countries”. In the method section, the authors stated contradictory statement “In order to estimate mean GWG, we analysed the data … enrolled in DHS implemented in 30 SSA countries since 2010.” At the same time, the authors included data collected since 1997. Although the authors reported that they used DHS data since 2010 to estimate mean GWG, the finding on the figure two showed that the authors estimated mean GWG since 1997. The authors may need to clarify more about this.

Result and discussion

8. Can the authors estimate GWG for underweight, normal weight, over weight and obese women separately? For example, if we see their GWG estimate for Southern African sub-region, 10.5 kgs (95% CI: 8.2-12.9), this value is excessive for obese women; adequate for overweight women; and inadequate for normal weight and underweight women. Therefore, the GWG estimates in this paper are difficult to interpret unless stratified based on pre-pregnancy body mass index of the women. These findings tell us nothing about the adequacy of GWG in the way that the authors reported. In the discussion section, the authors interpreted GWG as “we estimated (6.6 kg) is only about half of the minimum recommended gain for women with normal baseline BMI”. However, authors didn’t tell us whether 6.6 kg was estimated for normal weight women or obese women. The meanings and implications of this finding, GWG of 6.6kg, is totally different for normal weight or obese women.

Conclusion

The authors concluded that “The mean GWG in SSA is very low”. However, the nature of their result do not allow them to conclude so.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 May 26;16(5):e0252247. doi: 10.1371/journal.pone.0252247.r002

Author response to Decision Letter 0

20 Mar 2021

Comments from the Academic Editor

Comment 1: With respect to the introduction, the paper should provide background that puts the manuscript into context and allows readers outside the field to understand the purpose and significance of the study and include a brief review of the key literature. In the present case, as reviewer 1 points out, the article is missing a key reference that looks at the same topic with the same data and similar methods but for more countries. It is basic in this case to note the differences in approach or in scope. Note that otherwise PLOS ONE policy states that if a submitted study replicates or is very similar to previous work, authors must provide a sound scientific rationale for the submitted work and clearly reference and discuss the existing literature. Submissions that replicate or are derivative of existing work will likely be rejected if authors do not provide adequate justification.

Response: The concern is right. The key literature (Wang et al., 2020) identified by the first reviewer was not included because it was published after the submission of this manuscript. Now we have cited this article and described its findings both in the introduction and discussion sections. Further we have presented how our study is different from this article, including our peculiar contribution to the existing body of literature on the topic. It is important to note that the study by Wang et al., has not provided information the following information that we did: (1) trends in GWG in SSA, (2) sub-national estimates with SSA, and (3) differences in GWG trajectories across basis maternal socio-demographic characteristics. The same is now stated in the fourth paragraph of the Introduction section (Page 5). Furthermore, to allow readers outside the field to understand the purpose and significance of the study, we have now strengthened the last paragraph of the introduction section.

Comment 2: Regarding methods, there are some limitations that should be addressed. A prior consideration, is that the methods should be adequately described to allow for replication. As noted by reviewer 1, this is not the case, and the missing information might have confused, for instance, reviewer 3 in understanding the method.

Response: Additional methodological descriptions are now provided at different parts of the section.

Comment 3: First, note the concerns regarding the use of relevant guidelines as suggested by reviewers 1, 2 and 3.

Response: To the best of our knowledge, there is no GWG guideline specific to the SSA region or a country in this region. It is important to note that, even though the IOM guideline was developed for US population, its applicability to other settings had been somehow validated (Goldstein et al. 2018; Nomura et al. 2019; and Wie et al. 2017). The same is now stated in the second and third paragraphs of the introduction section. The absence of SSA-specific guideline is now discussed in the discussion section.

Comment 4: The problem of the relevant comparison group raised by reviewers 2 and 3. Reviewer 2 suggestion of reweighting non-pregnant women to match pregnant women according to weight is interesting and doable. If you carry this out show, at least for some instances, both sets of estimates to see if the strange patterns for first trimester disappear. Alternatively, a more sophisticated approach would be the use of propensity score matching to ensure that women in the synthetic control group (nonpregnant) are comparable to those pregnant.

Response: Reviewer 2 recommended for adjusting pregnant and non-pregnant women by age. As we tried to respond to the reviewer below, theoretically non-pregnant women are likely to be older and tend to have higher weight due to postpartum weight retention. Yet, in our dataset there was no meaningful difference in the age profile of non-pregnant women (26.9 ± 10.1 years) and women in the first (26.6 ± 6.9 years), second (26.8 ± 6.7 years), and third (26.9 ± 6.5 years) trimesters. Accordingly, we felt that additional age adjustment does not add value to the validity of the estimation.

Comment 5: Comment by reviewer 3 regarding grouping by weight I don’t believe it is feasible since you do not observe prepregnancy weight. However, you should bear it in mind when analyzing the patterns of estimated gestational weight gain. Since you have computed the proportions underweight and overweight you could see if the pattern of estimated weight gain at the survey level correlate with the proportions underweight and overweight. That will be very helpful for the discussion and also to tease out possible factors behind the much smaller GWG in Africa compared to other developing regions according to the missing source noted by reviewer 1.

Response: Inability to stratify GWG based on prepregnancy BMI or weight is now discussed as a limitation of the study and the possibility of over or underestimation of GWG is also stated (Discussion section, sixth paragraph, page 19). It is important to note that most of the limitations of this study can lead to both under or overestimation of GWG.

Comment 6: Third, as noted by reviewer 2, not using data from women in the first two months of pregnancy is probably a good idea. For instance, it is known that DHS patterns of reported pregnancy termination are extremely low in most sub-Saharan African countries (https://doi.org/10.1371/journal.pone.0221178) suggesting that many early terminations are missing because the woman was unaware of the pregnancy. Since this proportion would vary from survey to survey and according to socioeconomic characteristics, it would induce a bias in estimate weight gain.

Response: Please see the response we have provided to the reviewer below. Thank you.

Comment 7: In your revision note that PLOS ONE endorses the use of the STROBE checklist (http://www.strobe-statement.org) for observational studies such as this one. Make sure that all the questions in the checklist are addressed.

Response: We have completed the STROBE checklist and provided as a supporting file.

Reviewer #I

Comment 1: This work is similar to a previous study in terms of the research question. In Wang et al. (Gestational weight gain in low-income and middle-income countries: a modelling analysis using nationally representative data. BMJ global health. 2020), a multilevel modelling framework was applied to the DHS data to estimate the mean GWG in the year 2015 in all low- and middle-income countries and regions of the world, including in sub-Saharan Africa. Surprisingly, the Wang et al. paper was not referenced or mentioned in this work. Discussing the similarities and differences in the results and methodology between the two works will be necessary.

Response: Thank you for proposing this important paper. Earlier we did not cite this paper because it was not yet published at the time of this manuscript was submitted to this journal. Now the findings of the study are discussed and compared (Discussion Section, Paragraph 3, page 17) and what the current study would add in light with the new literature are described (Introduction Section, Paragraph 4).

Comment 2: On page 4: “assuming 0.5-2 kg gain in the first and 0.45 kg/week in the last two trimesters." However, the weekly weight gain in the second and third trimesters recommended in the IOM guidelines is 0.42 kg (0.35-0.50). It would be good to provide a reference for the value of 0,45 kg/week or revise it to be consistent with the IOM guidelines.

Response: Thank you. We have now corrected it as 0.42 kg/week.

Comment 3: I find the statistical methods of the study poorly described in general. There is virtually no explanation on how the pseudo-cohort design was reconstructed using the cross-sectional data in the DHS Program, how the total GWG and trimester-specific GWG were computed, or how the pre-pregnancy weight was estimated using the weights of women at risk of conception. I suggest that the statistical analysis and technical details of the methods be much more adequately described.

Response: In the data management and analysis section (Page 10), we have now described how GWG and trimester specific GWG were computed. In page 9, we have now stated how GWG trajectories were reconstructed based on cross-sectional data.

Comment 4: It is not straightforward to combine the survey data across 30 countries into one unified dataset while still preserving the sampling designs (weight, stratification, clustering) appropriate for each dataset. Therefore, I suggest clarifying how the datasets were combined and how stratification and clustering of the original surveys were accounted for. The explanation on the derivation of the new weights using the 2020 population size in the analysis also appears brief to me and will benefit from more details.

Response: Merging the datasets was not a major problem in this study because all the DHS surveys used similar sampling scheme, sampling weight calculation approach and variable names. What we did was adjusting the existing sample weights for the population size of the countries (i.e. poststratification weighting). Regarding poststratification weighting, we have now provided an additional explanation on how we calculated it based on the population size of the countries in the year 2020.

Comment 5: In Table 2, when classifying women into underweight, normal weight, and overweight or obese, were the WHO cutoffs (i.e., < 18.5, 18.5-<25, >=25) used? For adolescents aged 15 to 19 years old, were the WHO growth references (based on BMI-for-age Z-scores) for children/adolescents used instead of the cutoffs for adults?

Response: Thank you for this interesting comment. We have now provided BMI values only for adults 19 years or above based on the usual BMI cut values.

Comment 6: Page 13: “The overlap in the confidence intervals suggested absence of statistically significant difference between the two groups.” This statement is incorrect in that two CIs may well overlap while still having a statistically significant difference. I suggest the statistical significance in the subgroup analyses be rigorously and quantitatively determined. Otherwise, a qualitative description of the difference may be sufficient without resorting to a discourse on statistical significance.

Response: We disagree with comments of the reviewers. In general, examining statistical difference based on overlap of confidence intervals is a simple and convenient approach specially when graphs of confidence intervals have been presented (Schenker & Gentleman, 2001: On judging the significance of differences by examining the overlap between confidence intervals). Specially in this manuscript, relying on overlap of confidence intervals is the only option because we did not have individual level data to estimate p-values for comparing the distribution of two or more means. However, it is important to note that the method of examining overlap can lead to more type II errors than the standard p-values and it mistakenly accepts the null hypothesis when p-values show marginally significant differences (Schenker & Gentleman, 2001; Austin & Hux , 2002). This limitation is now discussed in the ninth paragraph of the discussion section (Page 20).

Comment 7: It appears that 11 countries were included in the analysis for 1997-2003, while 30 countries were included in the analysis for 2015-2020. As a result, quite different numbers of countries were used to compare the two time periods (1997-2003 and post-2015). This discrepancy greatly complicates the comparison, and the interpretation of the comparison becomes tricky and unclear.

Response: Seems there is misunderstanding here. As we tried to describe at the end of the introduction section (Page 5), this study had two objectives (i) estimate mean GWG in SSA based on aggregated data from multiple nationally representative cross-sectional surveys and; (ii) to compare changes in mean GWG between 2000 and 2015 in the region. For the first objective, the data of 30 countries that implemented DHS between 2015-2020 were included. For the second objective the data of 11 countries that conducted DHS surveys between 1997-2003 and 2015-2020 analysed. So, the trend analysis was conducted based on the data of 11similar set of countries. This is stated in the second and third paragraphs of the “Data source and inclusion criteria section” (Page 6)

Reviewer #2

Comment 1: When you compute the average (body) weight of women who are at risk of conception, I suggest creating and using (statistical) weights that make the non-pregnant group match the age profile of currently pregnant women. The reason to do this is that older women in their 40s with low fecundity may report menstruating and not using contraception, but are unlikely to become pregnant. Yet, (I believe) the current computation considers them equally “at risk of conception” as women in their 20s who are not using contraception. Therefore, their body weights count equally in the average weight of potentially pregnant women. If age is correlated with weight (which it is in many societies), you would be overestimating the average weight of women at risk for pregnancy, which would lead you to underestimate GWG. Both the tables and Figure 1 suggest this is happening because the (body) weights of pre-pregnant women are higher, on average, than the body weights of women in their first trimester. (Although a very small number of women do lose weight in the first trimester, it is not the norm.) If adjusting for the age profile of potentially pregnant women doesn't bring the pre-pregnancy weight down below first trimester weights, you could try creating statistical weights based on other factors as well (such as number of children or education) but I think age will cover it.

Response: Thank you for this interesting comment. Theoretically the suggestion of the reviewer is right. yet, in our data there was no meaningful difference in the age profile of non-pregnant women (26.9 ± 10.1 years) and women in the first (26.6 ± 6.9 years), second (26.8 ± 6.7 years), and third (26.9 ± 6.5 years) trimesters. Accordingly, we felt that additional age adjustment does not add value to improve the comparability of non-pregnant and pregnant women.

Comment 2: I would suggest not including results for women in the first and second months of pregnancy. The women who report very early pregnancy are likely systematically different from the pregnant population, and from those who report 2nd and 3rd trimester pregnancies. I think they will be more educated and perhaps come from less gender-conservative backgrounds. Indeed, the fact that such a small fraction of pregnant women report being 1 or 2 months pregnant reinforces the idea that there is under-reporting of pregnancy in these months.

Response: It is true that the small number of women in the first and second months indicates the presence of under reporting of pregnancy in early pregnancy. It is also theoretically true that women who are aware of pregnancy as early as first or second gestational month are likely to be more educated or come from better off households. However, we don’t think this has affected our primary objectives (estimating average and trimester-specific GWG and determining recent trends in GWG) for the following reasons. (1) GWG is determined as a difference between pre-pregnancy weight and weight at the time of delivery. As a result, imprecision or errors in weight in the first one or two months cannot affect this estimate. (2) Trimester-specific GWG estimates are made based on the maternal weight at the third, sixth and end of pregnancy weights and it is not dependent on the weight on the first one or two months. (3) Practically we did not observe meaningful difference between women in the first two gestational months and women in their third or higher gestational duration in terms of multiple socio-demographic factors including being in lowest wealth quantile (22.0% vs 20.6%), having only primary level of education (34.8% vs 32.7%), parity of 2 or above (21.5% vs 24.9%), age less than 20 years (14.3% vs 13.9%); and, (3) while presenting weight gain trajectories graphically, we did not observe a different or unexpected pattern in those two months.

Comment 3: The second paragraph of the introduction reads: “In 2019 the Institute of Medicine (IOM) of the National Academies put forth a new guideline on rate of GWG.” It would be good to mention that this guideline is for the US population. If the authors know of any guidelines specific to any SSA country/countries it would be helpful to mention them.

Response: The concern raised is right. We have now stated in the second paragraph of the Introduction section that the IOM is US-specific guideline however studies reported that it is applicable in other settings (especially in the Asian continent). The applicability of the IOM guideline had not been validated in Africa setting and to the best of our knowledge there is no SSA-specific gestational weight gain guideline. This is now discussed as a limitation of the study in the seventh paragraph of the discussion section (Page 18).

Comment 4: Please mention how you code gestational age in the main text. The DHS asks both month pregnant and last period; in some cases, the information is missing for one variable and not for the other. Which do you preference? Also, how often is gestational month 10 recorded? You might consider clubbing it with month 9 since it basically descries a full-term pregnancy, but the way that different societies talk about full term pregnancies differs slightly.

Response: The following information is now given under the “Data management and analysis” sub-section: “When available, gestational age was determined based on LNMP otherwise it was estimated based on self-report of the women. Very small proportion (about 0.3%) of pregnant women reported gestational age of 10 months and during analysis it was recoded to 9 months.”

Comment 5: Please include a more in-depth discussion of the statistical weights in the text.

Response: Additional information is now provided in the second paragraph of the data analysis section (page 9-10) on how the sample weighting was made.

Reviewer #3

Comment 1: Abstract. The authors can add comma after the word ‘On average’ in the result section of the abstract. Please also be consistent while reporting mean gestational weight gain with its confidence interval. Sometimes you put the mean outside of the bracket and confidence interval inside the bracket while you put all in the bracket in the other time, for example, 6.6 kg (95% confidence interval, 6.0-7.2) vs (10.5, 8.2-12.9 kg). Please be consistent while using ‘kg or kgs’

Response: corrected.

Comment 2: Background: The authors defined Gestational weight gain (GWG) as “the weight increase between conception and just before the birth of the infant.” Could the term ‘baby’ is more appropriate than ‘infant’ here.

Response: corrected.

Comment 3: At the beginning of the second paragraph, the authors stated “In 2019 the Institute of Medicine (IOM) of the National Academies put forth a new guideline on rate of GWG”. I am not sure if the National Academies released a GWG guideline in 2019. Please correct me if I am wrong.

Response: Apologies for this silly error. We have now changed 2019 to 2009.

Comment 4: Not clears why and how non-pregnant women were included in the analysis of GWG.

Response: As we tried to describe in the study design of the methods section (Page 6, first paragraph) pre-pregnancy weight was estimated using the data of non-pregnant women at risk of conception. Pre-pregnancy weight can not be determined only based on the data of pregnant women, because this may overestimate the baseline weight and underestimate the GWG.

Comment 5: I am a bit confused about data inclusion criteria. In the abstract section, the authors stated “Trend in GWG between 2000 and 2015 was determined using the data of 11 SSA countries”. In the method section, the authors stated contradictory statement “In order to estimate mean GWG, we analysed the data … enrolled in DHS implemented in 30 SSA countries since 2010.” At the same time, the authors included data collected since 1997. Although the authors reported that they used DHS data since 2010 to estimate mean GWG, the finding on the figure two showed that the authors estimated mean GWG since 1997. The authors may need to clarify more about this.

Response: In general, our intention was to assess the trends in GWG between 2000 and 2015. However, nearly all countries do not have data in these specific years. However, they have conducted surveys around 2000 and 2015. Accordingly, we have included surveys conducted

around the beginning of the new millennium (1997-2003) and on or after 2015 (2015-2018). This may marginally over or underestimate the rate of GWG. We have now stated this as a limitation of the study in the eighth paragraph of the discussion section (Page 19).

Comment 6: Result and discussion: Can the authors estimate GWG for underweight, normal weight, overweight and obese women separately? For example, if we see their GWG estimate for Southern African sub-region, 10.5 kgs (95% CI: 8.2-12.9), this value is excessive for obese women; adequate for overweight women; and inadequate for normal weight and underweight women. Therefore, the GWG estimates in this paper are difficult to interpret unless stratified based on pre-pregnancy body mass index of the women. These findings tell us nothing about the adequacy of GWG in the way that the authors reported. In the discussion section, the authors interpreted GWG as “we estimated (6.6 kg) is only about half of the minimum recommended gain for women with normal baseline BMI”. However, authors didn’t tell us whether 6.6 kg was estimated for normal weight women or obese women. The meanings and implications of this finding, GWG of 6.6kg, is totally different for normal weight or obese women.

Response: The concern of the reviewer is right however it was not possible to do stratification by BMI because BMI was only available for pregnant women. BMI was only available for non-pregnant women and hence reconstructing GWG for different BMI categories was not possible. This is discussed as a limitation in the seventh paragraph of the discussion section (Page 19).

Comment 7: Conclusion: The authors concluded that “The mean GWG in SSA is very low”. However, the nature of their result do not allow them to conclude so.

Response: We disagree with this comment of the reviewer. In the results section we have demonstrated the mean GWG in SSA is only 6.6kg and this is largely sub-optimal as compared to the IOM guideline. According to IOM, pregnant women with normal pre-pregnancy BMI should gain 11.5-16.0 kg of weight and women with low pre-pregnancy BMI are expected to gain as high as 18 kg of weight. So, we think that it is reasonable to conclude that the mean GWG in SSA is very low.

PLoS One. doi: 10.1371/journal.pone.0252247.r003

Decision Letter 1

José Antonio Ortega

12 May 2021

Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

PONE-D-21-02601R1

Dear Dr. Gebremedhin,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

José Antonio Ortega, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

[staff editor's edits]

The suggestion [from reviewer 2] to reweight the observations is possible to carry out. In statistical terms doing so would reduce bias at the cost of increasing variance. In the first revision I suggested the authors to consider the idea, which I believe is good at least to try. They preferred not do so and argued regarding balance on age. The reviewer argues that there could be second-order effects connected to variance and there could be lack of balance on other dimensions. There are others instances of published research where this has not been done and I believe the authors should have some leeway, particularly when considering that it cannot be said that doing so will always improve the estimation. If we take a mean squared error criterion it would depend on how large is the bias to see if it compensates for the increased variance.

Based on these considerations, I feel that the article fulfills PLOS ONE publication criteria and that this particular choice is optional for the authors who ultimately are taking responsibility for their published work.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: I suggested that the authors make adjustments to the weights of non-pregnant women to obtain a better estimate of pre-pregnancy weight. The fact that non-pregnant women weight more than first trimester pregnant women is a sign that there is selection into pregnancy that is correlated with weight. I suggested age would be a good place to start looking for factors on which to adjust. The authors did not show additional analyses to respond to this comment. The fact the average age is similar between pregnant and non-pregnant women is similar does not tell us that no adjustment is needed. The distribution of age may be different. There may also be other characteristics (such as number of children, education, asset wealth) that could be used to provide a better estimate of pre-pregnancy weight, a quantity which is at the heart of the authors' analysis.

Further, the authors say that they don't include results for women who are 1 and 2 months pregnant, but Fig 1 does include these women.

I would continue to suggest that the authors give more thought into how selection into pregnancy and pregnancy reporting affects their analysis, and write up their results in such a way as to reflect this. As is, the paper continues to assume that non-pregnant and pregnant women of each gestational age are similar, without providing evidence for these assumptions in the paper.

Reviewer #3: I would like to thank the authors for addressing almost all of my comments expect the one they disagree with. Although the authors disagreed with my last comment, their argument is still not convincing. As the authors rightly stated, the mean gestational weight gain, 6.6kg, was low for normal weight women according to IOM recommendations. However, the mean weight gain of 6.6kg is adequate for obese women given they are recommended to gain 5 to 9kg. Moreover, the authors clearly stated that they do not have data regarding BMI for pregnant women. This means that they are unable to tell whether the mean weight gain of 6.6kg is either for normal weight women or obese women. The mean weight gain of 6.6kg is low for normal weight women but it is adequate for obese women. Anyway, it is up to you and the editor to decide on if convinced on your argument.

Wish you all the best

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

PLoS One. doi: 10.1371/journal.pone.0252247.r004

Acceptance letter

José Antonio Ortega

17 May 2021

PONE-D-21-02601R1

Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

Dear Dr. Gebremedhin:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. José Antonio Ortega

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File. Flow chart of the study.

(DOCX)

Click here for additional data file.^{(29.4KB, docx)}

Data Availability Statement

The analysed data is publicly available from https://dhsprogram.com/data/.

[pone.0252247.ref001] 1.Kleinman KP, Oken E, Radesky JS, Rich-Edwards JW, Peterson KE, Gillman MW. How should gestational weight gain be assessed? A comparison of existing methods and a novel method, area under the weight gain curve. International Journal of Epidemiology 2007; 36(6): 1275–1282. 10.1093/ije/dym156 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref002] 2.Strauss RS, Dietz WH. Low maternal weight gain in the second or third trimester increases the risk for intrauterine growth retardation. The Journal of Nutrition 1999; 129(5): 988–993. 10.1093/jn/129.5.988 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref003] 3.Ukah UV, Bayrampour H, Sabr Y, Razaz N, Chan W, Lim KI, et al. Association between gestational weight gain and severe adverse birth outcomes in Washington State, US: A population-based retrospective cohort study, 2004–2013. PLoS Med 2019; 16(12): e1003009. 10.1371/journal.pmed.1003009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref004] 4.Chung JGY, Taylor RS, Thompson JMD, Anderson NH, Dekker GA, Kenny LC, et al. Gestational weight gain and adverse pregnancy outcomes in a nulliparous cohort. European Journal of Obstetrics & Gynecology and Reproductive Biology 2012; 167(2): 149–53. 10.1016/j.ejogrb.2012.11.020 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref005] 5.Bhavadharini B, Anjana RM, Deepa M, Jayashree G, Nrutya S, Shobana M, et al. Gestational weight gain and pregnancy outcomes in relation to body mass index in Asian Indian women. Indian Journal of Endocrinolgy and Metabolism 2017; 21(4): 588–593. 10.4103/ijem.IJEM_557_16 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref006] 6.Ludwig DS, Currie J. The relationship between pregnancy weight gain and birth weight: a within family comparison. Lancet; 2010: 376(9745), 984–990. 10.1016/S0140-6736(10)60751-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref007] 7.Truong YN, Yee LM, Caughey AB, Cheng YW. Weight gain in pregnancy: does the Institute of Medicine have it right? American Journal of Obstetrics and Gynecology 2015; 212: 362.e1–8. 10.1016/j.ajog.2015.01.027 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref008] 8.Institute of Medicine. Report Brief: Weight gain during pregnancy: Re-examining the guidelines. Washington (DC): National Academies Press; 2009. [Google Scholar]

[pone.0252247.ref009] 9.Goldstein RF, Abell SK, Ranasinha S, Misso ML, Boyle JA, Harrison CL, et al. Gestational weight gain across continents and ethnicity: systematic review and meta-analysis of maternal and infant outcomes in more than one million women. BMC Medicine 2018; 153: 16. 10.1186/s12916-018-1128-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref010] 10.Nomura K., Nagashima K., Suzuki S. Itoh H. Application of Japanese guidelines for gestational weight gain to multiple pregnancy outcomes and its optimal range in 101,336 Japanese women. Sci Rep 2019; 9: 17310. 10.1038/s41598-019-53809-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref011] 11.Jh Wie, Park IY, Namkung J, Seo HW, Jeong MJ, Kwon JY. Is it appropriate for Korean women to adopt the 2009 Institute of Medicine recommendations for gestational weight gain? PLoS ONE 2017; 12(7): e0181164. 10.1371/journal.pone.0181164 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref012] 12.Rasmussen KM, Yaktine AL (eds). Weight gain during pregnancy: Re-examining the Guidelines. Institute of Medicine (US) and National Research Council (US) Committee to Re-examine IOM Pregnancy Weight Guidelines. Washington (DC): National Academies Press; 2009. [PubMed] [Google Scholar]

[pone.0252247.ref013] 13.Asefa F, Cummins A, Dessie Y, Hayen A, Foureur M. Gestational weight gain and its effect on birth outcomes in sub-Saharan Africa: Systematic review and meta-analysis. PLoS ONE; 2020: 15(4), e0231889. 10.1371/journal.pone.0231889 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref014] 14.Doctor HV, Nkhana-Salimu S, Abdulsalam-Anibilowo M. Health facility delivery in sub-Saharan Africa: successes, challenges, and implications for the 2030 development agenda. BMC Public Health 2018); 18(1): 765. 10.1186/s12889-018-5695-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref015] 15.Wang D, Wang M, Darling AM, Perumal N, Liu E, Danae G et al. Gestational weight gain in low-income and middle-income countries: a modelling analysis using nationally representative data. BMJ Global Health. 2020;5:e003423. 10.1136/bmjgh-2020-003423 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref016] 16.Dean SV, Lassi ZS, Imam AM, Bhutta ZA. Preconception care: Closing the gap in the continuum of care to accelerate improvements in maternal, newborn and child health. Reproductive Health; 2014: 11(Suppl 3), S1. 10.1186/1742-4755-11-S3-S1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref017] 17.Corsi DJ, Neuman M, Finlay JE, Subramanian SV. Demographic and health surveys: a profile. International Journal of Epidemiology 2012; 41(6): 1602–1613. 10.1093/ije/dys184 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref018] 18.MEASURE DHS/ICF International. Demographic and Health Surveys methodology: Sampling and household listing manual. Calverton, MD: ICF International; 2012. [Google Scholar]

[pone.0252247.ref019] 19.The DHS Program. Retrieved from https://dhsprogram.com/Data/ on Nov 02, 2020.

[pone.0252247.ref020] 20.United Nations Population Division. World population prospects. Retrieved from https://population.un.org/wpp/ on Nov 02, 2020.

[pone.0252247.ref021] 21.African Union. Member states. Retrieved from https://au.int/en/member_states/countryprofiles2 on Nov 02, 2020.

[pone.0252247.ref022] 22.Bank World. Country and lending groups: Country classification. Retrieved from https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups. on November 02, 2020. [Google Scholar]

[pone.0252247.ref023] 23.Arora P Aeri BT. Gestational weight gain among healthy pregnant women from Asia in comparison with Institute of Medicine (IOM) guidelines-2009: A systematic review. Journal of Pregnancy. 2019; 2019: 3849596. 10.1155/2019/3849596 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref024] 24.Coffey D. Pre-pregnancy body mass and weight gain during pregnancy in India and sub-Saharan Africa. PNAS 2015; 112(11): 3302–3307. 10.1073/pnas.1416964112 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref025] 25.Meng Y, Groth SW, Li D. The association between obesity-risk genes and gestational weight gain is modified by dietary intake in African American women. Journal of Nutrition and Metabolism 2018: 5080492. 10.1155/2018/5080492 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref026] 26.Fraga AC, Filha MM. Factors associated with gestational weight gain in pregnant women in Rio de Janeiro, Brazil, 2008. Cadernos de Saúde Pública 2014; 30(3): 633–644. 10.1590/0102-311x00085313 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref027] 27.Dolatian M, Sharifi N, Mahmoodi Z, Fathnezhad‐kazemi A, Bahrami‐vazir E, Rashidian T. Weight gain during pregnancy and its associated factors: A Path analysis. NursingOpen 2020; 7(5): 1568–1577. 10.1002/nop2.539 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref028] 28.Galal M, Symonds I, Murray H, Petraglia F, Smith R. Postterm pregnancy. Facts, Views & Vision in ObGyn 2012; 4(3): 175–187. [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref029] 29.Tucker J, McGuire W. Epidemiology of preterm birth. BMJ 2004; 329(7467): 675–678. 10.1136/bmj.329.7467.675 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0252247.ref030] 30.Austin PC, Hux JE. A brief note on overlapping confidence intervals. Journal of Vascular Surgery 2002; 36(1):194–5. 10.1067/mva.2002.125015 [DOI] [PubMed] [Google Scholar]

[pone.0252247.ref031] 31.Schenker N, Gentleman JF. On judging the significance of differences by examining the overlap between confidence intervals. The American Statistician 2001; 55(3): 182–186. 10.1198/000313001317097960 [DOI] [Google Scholar]

PERMALINK

Gestational weight gain in sub-Saharan Africa: Estimation based on pseudo-cohort design

Samson Gebremedhin

Tilahun Bekele

Roles

Abstract

Background

Methods

Results

Conclusion

Background

Methods

Study design

Data source and inclusion criteria

Table 1. Demographic and health surveys included in the analysis.

Sampling design and approach of data collection in DHS

Estimation of pre-pregnancy weight

Reconstructing weight gain patterns during pregnancy

Estimation of weight at the time of delivery

Data management and analysis

Results

Basic characteristics

Table 2. Basic characteristics of the study subjects, sub-Saharan Africa, 2010–2018.

Estimated gestational weight gain in sub-Saharan Africa

Fig 1. Gestational weight gain trajectories in sub-Saharan Africa.

Table 3. Gestational weight gain by household and maternal characteristics, sub-Saharan Africa, 2010–2018.

Trends in gestational weight gain

Fig 2. Comparison of gestational weight gain in sub-Saharan Africa between 1997–2003 and after 2014.

Discussion

Conclusion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

José Antonio Ortega

Roles

Author response to Decision Letter 0

Decision Letter 1

José Antonio Ortega

Roles

Acceptance letter

José Antonio Ortega

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases