The main objective of this feasibility or pilot study was to explore whether a birth cohort with biological samples could be established in Tarlai Kalan, Islamabad Capital Territory, Pakistan. This involved testing the full strategy to identify, recruit, and follow up 150 pregnant women in their third trimester; administer varied instruments capturing multiple constructs, including those with sensitive questions; and obtain biological samples. We could do all of the set our objectives of this feasibility study with good recruitment and responses rates both for biological samples and electronically collected interview-based data. Thus, illustrating that Pakistan can feasibly be involved as one of the sites for the global birth cohort called Evidence for Better Lives Study .
Between 26-30% of our participants did not provide biological samples. The most frequent reason being the worry about getting late. This concern from women, especially from lower socioeconomic strata, with large families to care for, and cultural norms holding them accountable for their time and activities, is not surprising in the local milieu. Although less frequent, and perhaps less visible, these norms permeate the better-off classes too. Adeel et. al. note in their analysis of Pakistan Time-Use Survey that 80% of all trips outside the house are by males in Pakistan ). A qualitative study in a hospital in Rawalpindi by Armaan Rowthor et. al. found gender norms to be a major cause of constrained agency, decision-making, and prenatal anxiety amongst pregnant women ).
The other objective was to gain initial estimates of key variables. We learned that although there is high community and family support, neighbourhood disorder, widespread p-IPV, perceived stress, poor wellbeing, and alarming frequency of prenatal depression, are major challenges in improving maternal & child health. IPV during pregnancy and otherwise as well as prenatal depressive symptom have been consistently found to be higher in low- and lower – middle-income countries (LMICs) and especially the south Asian region. Our findings are also comparable to earlier studies in Pakistan, reporting similar prevalence rates of depressive symptoms [10-12]. Postnatal depression was markedly lower than prenatal depression, in line with local literature, which estimates the prevalence £ 25% [13-15]. This may point towards strong social and biological determinants of prenatal depression in the Pakistani context.
The overall characteristics of participating mothers, and gender disparities in education and employment, closely resembled those of the general population of Pakistan as reported in PDHS 2017-18 . Respondents belonged to diverse socioeconomic backgrounds. Most respondents had regular access to healthcare and ultrasound facilities. Despite access, high unmet need for family planning among our currently pregnant sample was clearly noted, with over half of the mothers reporting unplanned/unwanted, pregnancies. This finding to is in-keeping with the recent demographic and health survey of Pakistan . The reasons, as described in surveys and studies, are varied across socioeconomic class and ethnicity. Power dynamics also play a role, with only about 7% of current female contraceptive users in Pakistan having made the choice alone to do so along with lack of informed choice .
A significant proportion of the sample did not believe physical punishment for children did not harm them. This is another contextually embedded finding previously reported in literature that represents the general norms of punishment across our setting . While women’s exposure to four or more adversities in their childhood was also a frequent finding in our sample with respondents themselves having experienced physical punishment. Most frequently reported adversity was having witnessed a household member being treated violently. As per literature concerning childhood adversity carries has an intergenerational link .
Lifetime substance use was found to be very low among our participants despite higher stress levels and depressive symptoms. This has been a consistent finding in other studies and reports from Pakistan [19, 20]. It is primarily a reflection of the cultural norms surrounding women’s acceptable behaviour in semi-urban middle-class Pakistan, as well as other conservative societies. The picture changes when high and low-income groups are studied in isolation . Another factor responsible for lower substance use amongst women worldwide could be preferred processing mechanisms in men and women. Where men are more likely to externalize extreme stress through aggression and substance use, there is a propensity for women to internalize, leading to anxiety and mood disorders [22, 23]
The prenatal attachment levels seemed to have been low at first glance, across all categories, with mean values being below the mid-point. Perhaps this can be attributed to the fact that over half of our respondents did not plan their current pregnancy. However, another important consideration is that of the cultural context. For example is considered shameful for pregnant women to talk about their pregnancy and childbirth with family and or express happiness about it since pregnancy is associated with a sexual act . Beliefs about “evil eye” also deter mothers from openly expressing joy and letting people feel their baby’s movements . Furthermore, in our context caring for many children and the extended families living together with feeling fatigued and stressed most of the time does not let women think and enjoy the pregnancy thus expressing as poor prenatal attachment .
We collected data on emotions at the time of childbirth, at the postpartum follow-up, which was reported as largely negative. About one-third of the participants had childbirth through Caesarean Section. The trends worldwide show a major shift towards Caesarean Sections between 1990 and 2014, more marked in the high income countries and more in the populations with a higher wealth quintile [27, 28]. Pakistan is no exception, despite being a LMIC, the rising trends can be attributed to wealthier women having greater access to Caesarean Section option.
About 29% mothers held their babies within the first half hour after birth. Guidelines on the subject ask for immediate skin-to-skin contact between mother and child, foregoing the routine practices of weighing, bathing the baby etc. [29-31]. Almost no participant had child’s metrics from the time of birth in her records except birth-weight. Another indicator of healthcare services for mother and child in Pakistan lagging behind international guidelines.
Only 58% of mothers were exclusively breastfeeding at the postpartum follow up. 32.2% reported using milk substitutes along with breastfeeding in contrast to the 98.6% who showed intent at the prenatal visit. Beliefs about mother producing “insufficient” milk, baby not being satiated, formula milk or cow milk being more nutritious are widespread in Pakistan, which may have caused this. Other factors include maternal depressive symptoms, stress, resumption of household chores in the postpartum period, poor technique, lack of knowledge about the importance of breastfeeding, family physician’s advice, peers’ advice, and lack of facilities and services to initiate proper breastfeeding after birth[32, 33].
Based on our findings we propose that determinants of maternal health and child development are in poor state in Pakistani low and middle-income communities. Further research into the subject is essential to understand and mitigate these problems.
The main strengths of our feasibility study were its approach to identify eldigible sample using both facility based and a community based approach, high response rate at recruitment, postpartum follow-up and acquiring varied biological samples, using constructs informed by expert opinion through a delphi technique , use of validated instruments used in all other EBLS sites making interesting cross cultural comparisons.
Specific to the Pakistan site, this study is novel in the local context in that it takes prenatal life as a period of exposure to adversity and aims to collate these findings with follow-up data over the childhood years.
Our recruitment strategies eventually worked. We had initially planned involving LHWs bringing mothers along from their communities to the health facility for recruitment into our study. This had some initial issues where LHWs could not accompany the pregnant women on the said day for interviews at the facility. Additionally, LHWs reported many women found it inconvenient to come to the facility on the given day to be interviewed by the research team. However, our later approach to have participants come to the community based health houses of the LHWs (which are nearby to the households of the women) was far better for both recruitment and follow ups.
However, this strategy is not without its limitations. All of the population on Tarlai Kalan is not covered by LHWs thus leading to a potential source of selection bias for the eventual birth cohort. Other challenges during the study included scheduling visits with the Lady Health Workers, who are extremely busy and found it difficult to arrange follow up visits for our sample leading to some potential delays. For example, polio campaigns, were a recurring activity during our study period in which LHWs participate. We mitigated this issue by coordinating with the LHWs which were not involved in the polio campaigns. Similarly, during the postpartum follow up phase there was an outbreak of dengue fever in Tarlai Kalan leading to stopping of field activities for nearly two weeks. Thus for the next main round of study, the timelines of the main study should have these delays incorporated from the outset.