The effect of family planning exposure on fertility choices and reproductive health care in rural Pakistan

Abstract

This paper investigates whether family planning exposure mainly through the Lady Health Worker Program has had any effect on women’s fertility choices and use of reproductive health care service in rural Pakistan, using the 2013 Pakistan Demographic and Health Survey (PDHS). Exploiting variations in the program intensity across regions, we conduct instrumental variables (IV) estimation on the impact of family planning exposure on women’s fertility preferences and reproductive behaviors. Our analysis shows that even in the presence of strong son preferences in Pakistan, increasing family planning exposure reduces women’s incentive to have additional children. Moreover, family planning exposure is associated with more antenatal care visits, more deliveries performed by skilled professionals, and a greater chance of delivery at the health center. Finally, we examine the heterogeneity in the effect on fertility preference by age group and education level, and find that the effect is mainly significant for the educated and young women.

Introduction

Over the past few decades, a number of family planning programs have been implemented in developing countries, using various modes of service delivery. These programs have played a vital role in increasing women’s knowledge about family planning in countries such as Bangladesh (Cleland, Phillips, Amin, & Kamal, 1994), Nigeria (Bankole, Rodriguez, & Westoff, 1996), Nepal (Storey, Boulay, Karki, Heckert, & Karmacha, 1999), Tanzania (Jato et al., 1999), Malawi (Kane, Gueye, Speizer, Pacque-Margolis, & Baron, 1998), and Indonesia (Jayachandran, 2014). Family planning programs have been shown to play major roles in increasing contraceptive use, decreasing the number of unwanted pregnancies, and improving maternal and child health in many developing countries (Bongaarts, 1994; Casterline & Sinding, 2000; Jayachandran, 2014).

Although these programs have achieved considerable progress, large variations in fertility patterns persist across countries (Pritchett, 1994). There are two popular views about the determinants of fertility in public policy discussions (Becker, 1996; Pritchett, 1994). Some proposes that differences in fertility result primarily from different ideas as to the desired target number of children. For example, Pritchett (1994) argues that 90 % of variation in fertility could be explained by the desire for fertility, whereas only a very small effect could be attributed to family planning services, in the cross-country regressions. However, some recent studies evaluating the effect of family planning services show that those variations in fertility can be driven by inadequate use of contraception methods, which in turn results from lack of access to contraceptives, lack of knowledge, and costs (Cleland et al., 2006). Cleland et al. (2006) further show that family-planning programs have played a major part in raising the prevalence of contraceptive practice and reducing fertility in developing countries in the past four decades.

In the country study, for example, Caldwell, Caldwell, and McDonald (2002) show that the family planning program in Bangladesh decrease the birth rate per woman of childbearing age from 6.3 to 3.3 children over 20 years. This initiative in the Matlab district of Bangladesh contributed to a significant increase in contraceptive use by mitigating the fear of side effects through information dissemination (Cleland et al., 1994).

This paper is in line with the literature to investigate the effect of a community-based family planning programs, namely Lady Health Worker Program (LHWP) in Pakistan. The Program equipped local female health workers with the skills to provide essential maternal and child health care including family planning, management of minor and common ailments and imparting health education, targeted mainly at rural communities typically with high illiteracy and strong traditional cultural norms. LHWs may help rural women receive and comprehend the information, and motivate them to translate this knowledge into actual health outcomes. As a result, we expect that women who have exposure to LHWs would demonstrate actionable knowledge and better health outcomes measured as antenatal health care and fertility choices in this paper.

A few studies have investigated the program’s impact on fertility and use of maternal health care service (Cernada, Rob, Ameen, & Ahmad, 1993; Douthwaite & Ward, 2005; Khan, 1999, 2011; Sultan, Cleland, & Ali, 2002). For instance, Sultan et al. (2002) find that women living within the five-kilometer radius of community-based workers are more likely to use modern contraceptives compared to women who had no access. Douthwaite and Ward (2005) also find that the presence of lady health workers increased the use of modern contraceptives. Family planning workers can go beyond simply offering access to contraceptives to reshape traditional norms about desired family size by establishing themselves as trusted friends and resources in local communities (Phillips, Simmons, Koenig, & Chakraborty, 1988; Simmons, 1996).

These studies have some limitations, however. Sultan et al. (2002) do not investigate the LHWP’s effect on women living outside the stipulated five-kilometer radius, whereas Douthwaite and Ward (2005) use only the presence of lady health workers, not women’s actual exposure to a lady health worker, as their independent variable. Furthermore, these studies do not discuss actual fertility outcomes.

Our study seeks to address these limitations. First, we use women’s actual exposure to lady health workers, rather than the availability of these workers, as the main explanatory variable. Second, we take into account the intensity of the family planning program. As existing cultural norms are stronger in rural areas, it takes time for lady health workers to perform their duties and begin to advocate effectively about family planning. Furthermore, since the literacy rate is very low in rural areas of Pakistan (about 28 % in our sample), it is difficult for women to understand new knowledge and translate it into actual health outcomes without frequent exposure to family planning instructions. We believe that failing to take these variations into account could bias estimates of the program’s effect.

The rest of the paper is organized as follows. Section 2 introduces the Lady Health Worker Program (LHWP) in Pakistan. Section 3 describes the data and variables. Section 4 discusses the estimation strategy. Section 5 presents the main results. Section 6 concludes.