Population Attributable Fractions of Caffeine and Water Pipe on low birth weight: The Bandar-Abbas Population-based Cohort in suburban areas

Background: Low Birth Weight (LBW) poses a major health challenge in low-resource suburban communities. Despite relatively commonality, there is little evidence on the effects of water pipe and dietary caffeine on reproductive outcomes in the Middle-East region. The Bandar Abbas Pregnancy Cohort (BAPC), as a population-based prospective study, has been investigating the effects of lifestyle and environmental factors on maternal wellbeing and child growth and development in suburban communities in the south of Iran. This study used the BAPC data to estimate the attribution of LBW incidence to fully or partially removal of dietary caffeine intake and water pipe smoking during pregnancy. Methods: The present study used data on 861 live singleton pregnancies in the rst two BAPC visits (response rate= 95.42%). Water pipe smoking (yes/no) was measured during pregnancy using a structured checklist recommended by the World Health Organization (WHO). Dietary caffeine intake was measured during pregnancy using a structured checklist and categorized into low for 0-99 mg/day, and high for>100 mg/day. LBW (yes/no) was determined using recorded birth weight in infant`s vaccination card. Exposure relative risks were calculated using Modied Poisson regression models. Population Attributable Fractions (PAFs) and Generalized Impact Fractions (GIFs) were calculated on relative risk scale. The prevalence of counter fact was set as 3% for water pipe smokers and 14.9% for intake of >100 mg/day caffeine. Results: The cumulative incidence of LBW was 16.1%. An estimated 19% (95%CI: 6, 30%) and 11% (95%CI: 8, 14%) of LBW were attributed to high intake of dietary caffeine and water pipe smoking, respectively. Reducing the intake of caffeine to less than 100 mg/day or prevalence of water pipe to 3% would prevent 10.7% (95%CI: 6.6, 25.3%) and 5.7% (95%CI: 5.0, 6.8%) of LBW incidence, respectively. Conclusions: Our ndings have provided evidence on quantitative contributions of caffeine and water pipe on LBW using real-world data.

informal settlements have shaped in many Iranian cities with an estimated population of approximately 10.3 million (8). Suburban dwellers are among the most vulnerable sub-populations with inappropriate environmental conditions, speci cally in lifestyle (9). They usually suffer from unhealthy lifestyle with no food diversity, malnutrition, and abundant smoking behaviour (10), which prompt context-speci c epidemiologic evidence. Population Attributable Fraction (PAF), de ned as the fraction of all cases of a particular health condition that is attributable to a speci c exposure, is an epidemiologic measure widely in use to assess public health impact of various exposures. However, interpretation of PAF relies heavily on interventions that can perfectly remove exposure of interest. In other words, PAF interprets poorly for interventions for which success rate is below 100% (11). To overcome this conceptual pitfall, Generalized Impact Fraction (GIF), which calculates the change in disease burden when a risk factor is altered, is proposed. While PAF addresses complete elimination of a risk factor, GIF estimates the proportional reduction in disease incidence given a graded reduction in the prevalence of a risk factor (12).
Little is known about the likely impact of preventive interventions to control LBW by addressing water pipe or caffeine, especially in suburban communities. The present study aimed to predict the incidence of LBW following interventions to completely eliminate or partially decease the level of caffeine intake and water pipe during pregnancy using data from a population-based prospective cohort study.

Methods:
The present study used data from the rst two visits of a population-based prospective cohort study in Bandar Abbas city entitled Bandar Abbas Pregnancy Cohort (BAPC). Since 2014, BAPC is investigating the effects of lifestyle and environmental factors on maternal wellbeing and child growth and development in Bandar Abbas city, the capital of Hormozgan Province in the south of Iran. Through doorby-door inquiry in the suburban neighborhoods, BAPC has recruited over 1,500 pregnant women. Any pregnant woman aged above 16 years who is residing in the suburban areas was eligible to participate in the study. Those who were unable to communicate in Farsi or unwilling to participate were excluded.
Following providing the informed consent, data from subjects and their babies are gathering through four visits in pregnancy, 1, 6, and 12 months after birth. The details of the cohort methodology have been published elsewhere (13).
The present paper used the data of 897 subjects who completed the second visit of BAPC (participation rate: 92%, response rate: 95.42%) during September 2016 to May 2019. For this study, the following subjects were further excluded: miscarriage (n = 24, 2.55%), stillbirth or multiple pregnancies (n = 8, 0.85%), and self-report cigarette smoking during pregnancy (n = 4, 0.44%). Therefore, data of 861 live singleton pregnancies (mean ± SD gestational age at recruitment: 22.62 ± 9.66 weeks) was included in the analysis (Fig. 1). Data of the rst visit of BAPC is collected during pregnancy and the second visit is performed during post-partum period (0-42 days post-delivery) by in-person or telephone visit. The main study`s outcome was set as LBW and was de ned as birth weight below 2,500 grams (yes/no) (1). The main exposures were water pipe and caffeine intake during pregnancy, which both were measured during the rst visit of BAPC. Water pipe during pregnancy (regular/only for leisure/never) was measured by a checklist recommended by the World Health Organization (WHO) (14), and was merged into yes/no categories. The checklist was validated by a group of healthcare professionals and epidemiologists while the reliability was checked on a subset of BAPC subjects (n = 25, Cronbach`s alpha = 0.78). The checklist also contained additional information on age at water pipe initiation, duration of water pipe smoking, number of water pipe sessions per day, and Environmental Tobacco Smoking (ETS). Caffeine was measured by a structured checklist using Bunker categorization as a guideline (15). The checklist was validated by a team of nutritionist, epidemiologist and gynecologist to measure dietary intake of caffeine during pregnancy. The checklist contained questions on daily intake of any type of caffeinated beverages available in the local market (including black coffee, instant coffee, black tea, green tea, hot chocolate, soft drinks) and caffeinated medications (e.g., painkillers). The reliability was further checked on a subset of BAPC subjects (n = 15, Cronbach`s alpha = 0.64). The cumulative daily dose of caffeine was then dichotomized to normal (0-99 mg/day) and high (≥ 100 mg/day) (16). Number of prenatal care visits to healthcare centers and/or gynecologist o ce was dichotomized into regular (i.e., at least nine visits recommended by the national guideline,) and irregular (i.e., less than nine visits). Monthly expenditure as a proxy of socio-economic status was de ned as monthly average of usual household expenses during the last six months on housing, food, clothing, and healthcare.
Confounders were selected using the Change-In-Estimate (CIE) strategy. The CIE selects covariates on the basis of how much their control changes exposure effect estimates, i.e. amount of confounding by the covariate. Suppose RR a and RR u denote the estimated risk ratio with and without adjustment for the covariate; then RR a /RR u is traditionally used to judge change importance. By this strategy, an "important covariate" was determined as whether the change in the exposure effect estimate from adjusting for the covariate falls outside an interval of practical equivalence; e.g., 0.91 < RR a /RR u < 1.1 (which is the 10%change rule for the risk ratio modi ed to be proportionally symmetric) (17). Based on the CIE strategy, duration of water pipe smoking, maternal education, intake of iron supplement during pregnancy, infant sex, preterm birth, history of LBW infant, and monthly expenditure were included in the nal regression model. Adjusted Relative Risks (ARRs) for the effects of the main exposures on LBW were calculated using Modi ed Poisson regression models (18). The Miettinen formula was applied to calculate PAFs for caffeine intake and water pipe (11). Accordingly, we estimated the PAFs from the estimated ARRs for the exposures of interest (water pipe and caffeine intake, both as dichotomous variables). The prevalence of exposure among cases (p c ) was estimated as 22.3% for water pipe smoking and 67.63% for high caffeine intake. The PAF nally estimated as: PAF = p c (1 − 1/RR) (12).
GIFs were calculated using the following formula: Where, P i denoted the proportion of the population in exposure category i (fact) (8.59% for water pipe smoking, 56.98% for high caffeine intake). P i` denoted the proportion of the population in exposure category after an intervention or other change is implemented (counter fact) (3% as hypothesized prevalence of water pipe). Using the WHO recommended safety threshold of 100 mg/day caffeine in pregnancy, we hypothesized that an effective intervention would successfully decrease the proportion of women with high intake of caffeine to 14.9%. Adjusted relative risks were derived from the modi ed Poisson regression model (12). GIFs were calculated based on a series of proposed action plans to: A) Decrease caffeine intake to less than 100 mg/day only among subjects with high intake of caffeine; and B) Decrease the prevalence of water pipe to national report of the prevalence of water pipe among Iranian women in reproductive age (i.e. 3%) (5). All the analyses were performed using Stata version 13 (Stata Corp., College Station, TX, USA). P-values less than 0.05 were considered statistically signi cant for the nal model.

Results:
Out of 861 newborns, one hundred and thirty-nine (16.14%) were LBW, while the incidence of LBW was higher in illiterate/ elementary education mothers (21.1%), mothers with no gestational intake of iron supplementation (19.3%), and mothers with a history of LBW newborn (24.5%). Overall, 482 (56%) pregnant women reported high intake of caffeine, among whom the three most frequently reported sources of caffeine were black tea (43.3%), soft drinks (33.8%), and instant coffee (19%). Compared to normal caffeine intake, the proportion of LBW neonates was signi cantly higher in mothers with high caffeine intake (13.2% vs. 18.4%, p = 0.038). Sixty-nine (8.59%) pregnant women reported water pipe smoking, with an average of 1.87 sessions per day. Compared to the non-smokers, the proportion of LBW neonates was signi cantly higher in the water pipe smokers (14.27% vs. 37.68%, p ≤ 0.001). The mean (SD) of duration of water pipe smoking was 6.7 (30.8) months. The duration of water pipe smoking was signi cantly higher in LBW neonates (11.8 (43) months vs. 5.8 (27) months in normal-weight neonates, P = 0.01) ( Table-1). After adjusting on duration of water pipe smoking, maternal education, prenatal iron supplementation, infant sex, preterm birth, history of LBW infant, and monthly expenditure ; intake of more than 100 mg/day caffeine during pregnancy increased the risk of LBW by 36% (ARR = 1.36, 95%CI:1.00,1.88). The risk of LBW was more than two times higher in women who smoked water pipe during pregnancy (ARR = 2.12, 95%CI: 1.32, 3.40) ( By implementing community-based action plans to reduce the proportion of high caffeine intake, 10.7% of LBW neonates would not have occurred (GIF = 10.7, 95%CI: 6.6, 25.3%). Similarly, implementation of action plans to reduce the prevalence of water pipe to the average national level (i.e. 3%) would decrease the number of LBW neonates by 5.7% (GIF = 5.7, 95%CI: 5.0, 6.8%). (Table 3)   Discussion: The present study used data of 861 live singleton births from the BAPC study. Sixteen percent of neonates were born with LBW while one-third of them were preterm as well. The incidence of LBW in our sample of suburban residents was higher than both the United Nations Children's Fund (UNICEF) report for Iran (8%) and previous studies (19,20). Totally, 73% of the pregnant women reported consumption of any dose of dietary caffeine during their current pregnancy while the average dose of caffeine was 104.7 mg/day. By de ning the threshold for the safety dose of caffeine as 100 mg/day (16), we found that 56% of our study subjects reported high caffeine intake. We also observed that women belonging to lower educational and socio-economic groups and those who used water pipe, reported higher doses of caffeine intake. The estimated dose of caffeine intake in our study was similar to those reported among pregnant women in the US and Poland (21,22). The primary sources of dietary caffeine in our study; however, disagreed with the previous reports. While black tea (60%) and soft drinks (44%) were the predominant sources of caffeine intake in our study, black coffee was reported as the dominant source of caffeine intake in the developed world. This discrepancy can be attributed to social and cultural context. Differences in weather circumstances and availability and affordability of black tea and soft drinks advertise their consumption over other sources of caffeine (e.g. coffee or chocolate). Nevertheless, due to lack of national estimates on pattern of caffeine intake in Iran, we cannot provide a comprehensive discussion on caffeine consumption in our sample compared to the general population. Our study showed that consumption of more than 100 mg/day caffeine signi cantly increased the risk of LBW, which was consistent with the results of previous studies (7,23,24). We used one-week dietary recall to measure daily intake of caffeine during pregnancy. Due to many exogenous and endogenous factors, it is likely that the measured dose of caffeine did not necessarily re ect the actual concentration of caffeine metabolites in body (24). Moreover, we are aware that the amount of caffeine in beverages is heavily determined by methods of beverage preparation; therefore, non-differential misclassi cation bias for the measured dose of caffeine is likely (25).
The self-reported prevalence of water pipe during pregnancy was 8% in our study, which was higher than previous estimates in Iran (5,26); yet similar to the estimates from countries of Arabian Peninsula (27). Consistent with previous reports, we found that water pipe during pregnancy signi cantly increased the risk of LBW (2, 28).
The nal hypothesis of our study was the extent to which partial or complete removal of either exposures of interest would reduce the risk of LBW. To answer that, we calculated PAF, which is de ned as the fraction of all cases of an adverse condition in a population that is attributable to a speci c exposure (29). We estimated that 11% and 19% of LBW cases would be prevented if interventions to eliminate water pipe or dietary caffeine intake were to be implemented, respectively. In order to interpret PAF using our observational data, we must assume that removal of the study exposures did not affect or alter the distribution of other risk factors for LBW (such as cigarette smoking replaces water pipe). We also must assume that there was no bias in the study design and data analysis, which means that the estimated effects were adjusted for all measured confounders (11). The estimate of PAF only applies to interventions that can completely remove the exposure of interest, while such interventions, especially in lifestyle factors, may not be pragmatic. To calculate the impact of partially successful interventions, we developed real-time scenarios and calculated GIFs, which measure the reduced fraction of cases that would result from changing the current level of the exposure to some modi ed level (12). The rst intervention targeted decreased proportion of pregnant women with high caffeine intake. By implementing such program, 19% of LBW neonates would be prevented. The second scenario targeted decreases marginal prevalence of water pipe to the estimated national level of 3% among women.
Implementation of such program would result in 11% reduction in LBW cases.
This study used data from a population-based sample to nd the extent to which various preventive interventions on caffeine intake and water pipe during pregnancy would in uence the risk of LBW.
Despite the high generalizability of a population-based sample, our results should be translated carefully. We restricted our study subjects to women who had conception without medical assistance and had no history of infertility, hence our sample might be a healthier sample compared to the general population of women of reproductive age (30). Caution must be taken if the estimated effects of water pipe and high caffeine intake are to be generalized to women who had prior history of infertility or medically-assisted conception. Moreover, generalization of the estimated PAF to other setting requires similar distribution of LBW and valid estimates for the exposure effects in that setting (11). The possible mediation effect of preterm birth in causal pathways between the exposures of interest and LBW is another controversial point. Whether preterm birth lies in the causal pathways from water pipe or caffeine to LBW is not clear yet. Nonetheless, we acknowledged the likelihood of mediating association and we recommend that use of the results of our study to preterm infants should be done with caution.
The present study was one of the rst prospective epidemiologic studies to investigate public health impacts of caffeine and water pipe on reproductive outcomes in Iran. It used data from a prospective cohort study, while suitability of prospective epidemiological studies to investigate pregnancy outcomes has been well-documented (30,31).
Out study had some limitations. We calculated the risk of LBW based on a risk set of total live births, whereas, the true theoretical risk set is de ned as the total of successful conceptions. Therefore, the estimated risk of LBW might be in ated due to exclusion of unsuccessful pregnancies from the denominator (30). We used a structured checklist to measure the prevalence of water pipe based on the WHO recommendations. Nevertheless, exact estimation of the amount of absorbed substances from water pipe smoking is impractical due to lack of a ubiquitous standard tool. Therefore, the estimated magnitude and effects of water pipe on pregnancy outcomes might be different from previous results partly because of using different data collection tool. Additionally, we did not measure the cumulative effects of water pipe smoking as an important predictor of fetal growth. This was because of the inconclusive evidence regarding such association at the time of the BAPC development. Therefore, our results only apply to the effects of water pipe during pregnancy rather than cumulative effects of water pipe during life span.

Conclusions:
Our ndings have provided evidence on quantitative contributions of caffeine and water pipe on LBW using real-world data. Integration of this information into practical action plans to prevent LBW is the next step in our cohort project. Furthermore, to get an overarching picture and in-depth understanding of reproductive effects of caffeine and water pipe, exercising this study`s analytic approach in other contexts is encouraged.

Declarations
Ethics approval and consent to participation All the study`s methods were performed in accordance with the declaration of Helsinki.

Consent for Publication
All the data were analyzed de-coded and in aggregated form. The participants agreed to use the data for publication without reporting any personal identi ers.
Availability of data and materials The dataset generated and analysed during the current study is not publicly available due to the restrictions applied by the National Institute for Medical Research Development as the sponsored agency. The de-identi ed data; however, is available from the authors upon reasonable request and with permission of the sponsor.

Competing interests
The authors declare no competing interests.
Authors' contributions SN contributed in study design, data analysis and manuscript preparation, MM and KH supervised the study conduction, data analysis and manuscript nalization, AR contributed in data analysis and manuscript preparation, AM and HS supervised the eld-data collection and manuscript writing. All the authors have read and approved the manuscript. Figure 1