Unmet need for family planning is an appraisal concept that is widely used for advocacy, the development of family planning policies and the implementation and monitoring of family planning programme worldwide. Although the concept of unmet need for family planning is the discrepancy between fertility preferences and contraceptive use, the definition of unmet need for family planning has been under continuous revision and development since the 1960s. Unmet need was initiated in the 1960s through the first generation fertility and family planning surveys, it had rightly pointed out the proper direction for family planning programme showing the KAP-gap, which means in most of the societies a discrepancy exists between woman’s reproductive preference and their contraceptive behavior (Mauldin, 1965). At that time, the demand for family planning services was the important focus especially for investment and resource allocation purposes. As a result Freedman et al.. (1972) redefined this as the desire to terminate childbearing but not using any contraceptive methods. However, before the introduction of the exact wording of the term ” unmet need for family planning” by Charles F. Westoff in 1978, many other estimates of unmet need for family planning had been done using the World Fertility Survey (WFS) and later Contraceptive Prevalence Survey (CPS) data. In his first definition of unmet need for family planning, Westoff (1978) proposed eight measures to estimate proportion of women who do not want any more children and are not practising contraception. His analysis considered different groups of women as a denominator ranging from all currently married women of reproductive age to totally excluding the pregnant and amenorrheic women from the analysis. In 1981, Westoff and Pebley, using WFS data from 18 countries, showed that a different definition of unmet need produced widely differing estimates. They proposed 12 alternative definitions. Their estimates using different definitions showed that the proportion of unmet need vary substantially from one definition to another. They also recommended enlarging the concept of unmet need by covering both desire to space and limit childbearing. Nortman (1982) advocated a further broadening of the definition based on data from CPS which included data on women’s preference for timing births as well as for limiting birth. Nortman’s (1982) argument was that pregnant, breastfeeding and amenorrheic women should be included in the analysis of unmet need for family planning because, ” many women who give birth each year did not seek to become pregnant but rather conceived because of failure to practice contraception or to practice it effectively”. This argument was supported by Westoff and Ochoa (1991) when they argued that many pregnant and postpartum amenorrheic women might be in that state at the time of the survey because they were not using contraception, but did not want to become pregnant; that is, they had a prior need for family planning that was not met. They suggested that women who were pregnant or postpartum amenorrheic be assigned an unmet need status based on the retrospective wantedness of their current pregnancy or last birth. Despite critiques of using retrospective fertility intentions as a measure of unmet need status for pregnant and postpartum amenorrheic women as expressed by Ross and Winfrey (2001). This approach has been used by DHS since approximately the 1990s. After the introduction of Demographic and Health Survey (DHS) in late 80s, the estimates of unmet need had been more refined because DHS asked women how soon they wanted to have their next birth. DHS also asked pregnant and amenorrheic women whether their current or most recent pregnancy was mistimed or unwanted. DHS could also identify the pregnancy due to method failure because women were asked whether they were using contraception at the time of conception (Casterline & Sindings, 2000). Based on all these information, Westoff (1988) proposed a framework for estimating unmet need for family planning, which is a widely used conventional definition till today. According to him, unmet need consists of four discrete subsets of currently married women who compose of two types of unmet need. Spacing unmet need composes of two subsets, which includes pregnant and amenorrhoeic women whose current or most recent pregnancy was mistimed and non-pregnant and non-amenorrhoeic fecund women who wish to postpone their next birth. On the other hand, limiting unmet need composes of another two subsets which include pregnant and amenorrhoeic women whose current or most recent pregnancy were unwanted and non-pregnant and non-amenorrhoeic fecund women who wish to terminate childbearing altogether. However, the validity of this concept had been questioned in different points of view including reproductive health needs, use of ineffective methods, unmet need of unmarried women and men, and even unmet need for couple. According to Dixon-Muller and Germain (1992), the conventional definition neglects reproductive health needs of women and it does not cover both unmarried women and men although they are also at risk of unintended pregnancy. Their view in fact supports Hartmann’s (1987) earlier argument that unmet need for family planning is not a reproductive health measure because it can poorly serve a woman’s health need. In 1991 Casterline and Palmore et al.. proposed separate health-risk based definitions for unmet need for family planning. According to Casterline (1991), women who are fecund but not practising contraceptive and have one of more health-risks such as high-order births (above 4), closely spaced births (birth occurred within 15 months of previous birth), and even early (birth by women aged under 18 or 20) or late (birth by women aged over 35) childbearing are considered as having unmet need for family planning. Govindasamy and Rutstein (1993) and Degraff and De Silva (1996) also proposed almost similar health measures like Casterline and Sindings (2000). Another critical point of conventional unmet need is that it does not include women who are pregnant due to method failure and women who are using an inappropriate method (Dixon-Muller and Germain 1992; Foreit & Mostajo 1993). The study seeks to identify proximate determinants of unmet need for family planning and not estimates of unmet need for family planning. Therefore, the conventional definition for unmet need for family planning as adopted by DHS would be used.
Determinants of unmet need for family planning
The focus on examining covariations in unmet need for family planning is to statistically identify variables that significantly relate to others, to aid in developing appropriate programme strategies to fulfill the demand services. This is a widely used approach in the identification of important predictors for specific outcomes. Since the study is using three sets of variables, the following sections would present a review of previous studies for understanding the specific effect of independent variables.
Demographic characteristics and unmet need for family planning
Several studies tried to explore differentials of unmet need in regard to four demographic variables. Among them age and number of living children were commonly used by most of the studies. However, other studies considered age at first marriage as a demographic determinant of unmet need for family planning. There is a distinct variation of unmet need in regard to age of women in 80s and 90s. This variation may be due to a significant reduction of unmet need. Most of the descriptive studies revealed that in 80s, the proportion of unmet need was increasing with an increase in age (Nortman & Lewis, 1984; Kabir et al., 1987; Sabir and Ali, 1993). In 90s, the aggregate level variation was in reverse order (Mitra et al., 1994; Mitra et al., 1997; Mitra et al., 2001). Singh et al., (1997) also found maternal age as one of the best predictors of unmet need. However, in comparison to middle age (20-34 years) group teenagers have very high unmet need and older age group have a relatively low unmet need (Islam et al., 2000). Most of the descriptive studies observed that spacing unmet need is higher among younger and limiting unmet need is higher among older women. Perez and Palmore (1997) did this analysis in the Philippines, which could confirm the descriptive results. Moreover, they observed that the health-risks related to high-parity are significantly higher among older (30-39) women and short spacing births are significantly higher among young (15-29) women. The contraception need prevailing among both the younger and the old remain poor and for different purposes. For example, the need for spacing is among younger women and the need for limiting for older women (Perez & Palmore, 1997). Age at first marriage as a predictor of unmet need for family planning was observed as an important predictor for health-risk and poor contraception unmet needs along with spacing need in the Philippines (Perez & Palmore, 1997). Since early marriage and early childbearing are common behavioural patterns of Ghanaian women, age at first marriage may be a useful predictor in explaining the variations of spacing and health-risk unmet needs. As women have more children, their unmet need for spacing birth tend to decrease, while unmet need for limiting increases. (Bradley et al.. 2012). According to Khuda et al.. (1999), unmet need increases with an increase in the number of living children. They observed number of living children as the best predictor of unmet need among several demographic, socioeconomic and geographic variables. The effect of number of living on both unmet need for limiting and spacing is almost similar to the effect of age on unmet need. According to Bradley et al.. (2012), in almost all surveys, unmet need for limiting increases with parity, but trends are less consistent for unmet need for spacing.
