Putting Unmet Need to the Test: Community-Based Distribution of Family Planning in Pakistan
Context: Unmet need for family planning in the developing world, as measured through surveys, is high. But it is important to determine whether there is a significant level of dormant demand for actual contraceptive services waiting to be satisfied, especially in a country such as Pakistan, where efforts to promote family planning have been disappointing.
Methods: Records from six household contraceptive distribution projects in Pakistan are used to determine contraceptive prevalence over 13-22-month periods. An independent professional team conducted an external evaluation, interviewing project supervisors, fieldworkers and clients.
Results: Contraceptive use increased dramatically in all six projects, from an average of 12% to 39% in less than two years. The external evaluation team found the contraceptive prevalence measurements to be generally accurate, but identified additional improvements in access and quality that might further increase contraceptive use.
Conclusions: Increased use of contraceptives that result from improvements in service delivery confirm that a substantial unmet need exists. The evidence suggests that even greater improvements in access to and quality of services will further increase contraceptive use.
International Family Planning Perspectives, 1999, 25(4):191-195
The existence of a sizable "unmet need" or "unmet demand" for family planning services has been a fundamental tenet of three decades of global efforts to provide voluntary family planning services.1 But what does "unmet need" really mean? As it is generally used, the term is a construct derived from large-scale surveys, such as the Demographic and Health Surveys (DHS). In these surveys, a substantial number of women who are at risk of pregnancy report either that they want no more children or that they wish to postpone childbearing, and are not practicing contraception. But does this survey-based construct reflect a latent demand for family planning that can be addressed by improving services and making them more available? It is important to understand what unmet need means operationally to help guide program efforts.
A major review of unmet need proposed maximizing access to high-quality services as the first approach to take to address the issue.2 Conversely, a recent analysis of survey data from the Philippines concluded that lack of access to family planning services does not contribute to the country's high level of unmet need.3 Similarly, an analysis of survey and qualitative data in Punjab, Pakistan, determined that the social costs of contraception (e.g., fear of side effects and spousal, cultural and social acceptance) were the decisive obstacles to its use, rather than the monetary and related direct costs of obtaining supplies.4
But survey methodologies often used to measure unmet need may have intrinsic limitations for assessing what is, after all, a complicated behavioral issue. We can take a more direct approach to determine whether unmet need can be satisfied by making services more available simply by providing services where they previously were unavailable. Such a supply-side approach (i.e., the provision of a number of direct services, including information, counseling and contraceptive methods) can be distinguished from demand-side activities (such as efforts focused on changing reproductive intentions and otherwise altering people's motivation).
The most widely known example of the supply-side approach is the Matlab project in Bangladesh, where programmatic efforts have had a clear and dramatic effect on contraceptive use and fertility (far beyond the expectations of many contemporary skeptics) in a difficult socio-economic and cultural context.5 In our view, the Matlab experience demonstrates the existence of a clear, unmet, latent demand for family planning and supports the merits of a supply-side approach that emphasizes access to and quality of family planning services. Yet this effort also has been dismissed as artificial, "massive and expensive,"6 despite its success in increasing contraceptive prevalence and decreasing fertility.
In the formative 1970s, support for the supply-side approach came from a whole family of operations research and other demonstration projects, especially community-based distribution efforts. These clearly demonstrated that a variety of trained nonprofessionals could effectively provide contraceptives and significantly increase use in a short period of time.7 In the ensuing years, similar community-based distribution projects have been implemented in numerous developing countries.
Fieldworkers involved in community-based distribution often visit each household in the community and conduct an ongoing census of contraceptive use, and thus engage in an ongoing assessment of contraceptive prevalence. Major increases in contraceptive use often occur as services become better and more accessible. In fact, this phenomenon has been so commonplace that it has not been considered news, and therefore has gone largely unreported in the literature.
The poor performance of Pakistan's family planning efforts over the last 30 years has been discussed intensively. In the late 1960s, Pakistan was routinely cited as a leader in family planning efforts.8 In the years since the country was divided, however, contraceptive use has increased little in what was the economically more advanced former West Pakistan, despite numerous programmatic efforts. In sharp contrast, contraceptive use has increased dramatically and fertility has fallen in the less advanced former East Pakistan, now Bangladesh.9
Ironically, the level of unmet need in Pakistan is among the highest in the world.10 Some informed observers maintain that the vast majority of people do not have access to services of even minimal quality.11 Others attribute poor performance to an intrinsic resistance to family planning related to cultural conservatism, religious influences and the low status of women.12
For these reasons, Pakistan presents a challenging opportunity to probe wheth-er there is an unmet demand for family planning that can be met operationally through the provision of accessible, high-quality family planning services. To do that, we report on a set of community-based distribution programs carried out in the country in the early 1990s
The six community-based distribution programs in Pakistan described here were developed with support from the U.S. Agency for International Development (USAID) as a single project of the Nongovernmental Organizations Coordinating Council (NGOCC), with technical assistance from the Asia Foundation.13 The individual projects were initiated in 1991 or early 1992 and were carried out by a variety of local nongovernmental organizations (NGOs) in all four provinces of Pakistan and in both urban and rural areas (Table 1).
|Table 1. Family planning community-based distribution projects, Pakistan|
|Area||Province||Start date||Implementing NGO|
|Lyari||Sind||July 1991||Lyari Community Development|
|Orangi||Sind||Oct. 1991||Orangi Pilot Project|
|Korangi||Sind||Oct. 1991||Pakistan Voluntary Health and Nutrition Association|
|Swabi||Northwest Frontier Province||Dec. 1991||Swabi Women Welfare Society|
The projects' service delivery was designed to follow a standard pattern. In each area, about 20-27 newly recruited female fieldworkers received 10 days of training in family planning. In general, they were residents of the project area and had roughly a 10th grade education. Each fieldworker was assigned a catchment population with a minimum of 650 married women of reproductive age; the number of married women of reproductive age in each of the six projects combined was about 15,000.
The fieldworker was expected to visit each woman every two months. (There was little formal outreach directed to men.) Besides offering information and counseling, fieldworkers provided oral contraceptives (including the initial supply), condoms and contraceptive foaming tablets (at nominal cost), and referred clients to clinical facilities for other methods (injectables, IUDs and sterilization). While fieldworkers mainly provided family planning services, they also referred women for some other health services. At every visit, the woman's contraceptive status was recorded. Fieldworkers were supervised, and supervisors also made household spot checks to validate record-keeping.
In each of the six projects, contraceptive prevalence increased dramatically and substantially over a short period of time (Figure 1), from an average of 12% at baseline to 33% after one year. Notably, prevalence increased from 7% to 31% in the conservative rural area of Swabi in the Northwest Frontier Province. Additionally, it increased from about 16% to more than 40% in Quetta, the capital of Balochistan (a province considered to be one of the most conservative culturally). By June 1993 (about 20 months after program initiation), the average prevalence for all six projects was 39% (not shown).
|Figure 1. Contraceptive prevalence rate, by number of months following project implementation, according to project site, Pakistan, 1991-1993|
|Note: The rate is calculated based on use of all modern methods.|
Not surprisingly for a community-based distribution program, supply methods were well represented. A large proportion of women chose supply methods: Forty-one percent used condoms and 10% chose oral contraceptives. Another 28% of clients were sterilized, 11% used IUDs and 9% chose the injectable. Sterilization clients were somewhat older (37) than were users of oral contraceptives (28) or users of the condom or IUD (30-32).
In an external evaluation of the projects conducted in June 1993, an independent professional team used rapid-assessment and survey techniques to interview project supervisors, fieldworkers and 289 clients.14 Project documents, computer records and reports also were reviewed. The evaluation team was very positive in its assessment of the reliability of the program's monitoring and record system, and found the contraceptive prevalence measurements to be generally accurate. The few inaccuracies reported were judged to be minimal by the evaluation team.
On the other hand, the evaluation found that fieldworker performance could be substantially improved in some areas (such as in counseling about side effects), and that referral and transport to clinical services were weak. They also identified other substantial barriers to access, even in the context of this community-based approach. For example, 18% of fieldworkers thought that a couple should have at least two sons before practicing contraception. Further, 45% believed that no one younger than 21-25 should receive oral contraceptives.
The external evaluation found that client shyness, especially at baseline registration, might have resulted in an underestimation of contraceptive prevalence. Nevertheless, the project's baseline contraceptive prevalence rate was roughly the same as the 1990-1991 Pakistan DHS rate (12%). Moreover, even if one uses the higher level at first follow-up as a baseline, the increase in contraceptive prevalence was still substantial for a short period of time.
The evaluators also assessed the cost-effectiveness of the six projects for the first year (excluding commodity and technical assistance) at US$3.80 per couple-year of protection. Bearing in mind the difficulties of generalizing such cost calculations, this level is relatively inexpensive, compared with other programs.15
Impact of a Supply-Side Approach
Results from these six projects show that unmet need actually does exist in a programmatic sense. Increased access to basic family planning services can meet much of the substantial unmet demand for family planning. Although this programmatic phenomenon is far from unique, the Pakistan experience is unusually well-documented and validated. Notably, latent demand can be satisfied in a variety of settings within Pakistan—a country where programmatic efforts have been largely unsuccessful and which some believe has, for cultural reasons, a low demand for family planning. Our findings corroborate those of an early demonstration project in Sialkot, Pakistan, which even in the early 1970s showed a marked increase in contraceptive prevalence (from 3% to 20%) after the introduction of truly accessible household family planning services.16
COMPARISON WITH SURVEY FINDINGS
Unmet need, as measured in the 1990-1991 Pakistan DHS, was 32%.17 Superficially, this level of unmet need corresponds well with the 28% increase in contraceptive use that occurred after the availability of services was enhanced through the six community-based distribution projects. But the relationship between the survey construct of unmet need and the operational unmet need we documented must be more complex. Some individuals who would be classified as having unmet need in such a survey will not make use of family planning services, even if they are highly available. The reverse is also true, however. Indeed, a sizable number of women do not fall into the traditional unmet need classification but nevertheless say they intend to practice contraception.18 Clearly, the phenomena we are dealing with—human ideation, motivation and behavior—are complicated.
CONTINUUM OF DEMAND
It is helpful to view demand for family planning on a continuum. Some people are highly motivated and even desperate to avoid births; they will go to great lengths to prevent them. For others, demand is much less concrete, and perhaps not even consciously formulated. Many people in the developing world face serious, daily, immediate demands on their time and energy. Any action is strongly influenced by competing needs and desires, and highly dependent on the practical opportunities available. Making family planning services available can make using contraceptives a practical reality and can crystallize latent demand.19 Further, the literature on behavior change clearly advances a variety of ideational and contextual factors that can influence human behavior, above and beyond well-defined purposeful intent, including the concept that behavior change may proceed through stages of decision formation.20
Although our findings demonstrate a clear, rapid and substantial increase in contraceptive use in response to contraceptive availability, several issues should be considered. First, these community-based distribution projects were not randomized studies, and therefore have no formal comparison areas. Might factors other than increased contraceptive availability have led to the rise in prevalence in all the project areas? We seriously doubt it. The projects represented six replications of a one-group design, and the fact that the results were consistent, immediate and strong across all six areas is reassuring. Moreover, baseline prevalence in these areas was consistent with levels seen in the 1990-1991 DHS. Although there were some new nationwide initiatives in Pakistan in the early 1990s, such as social
marketing, the 1994-1995 Pakistan Contraceptive Prevalence Survey reported a national contraceptive prevalence rate of only 18%, and prevalence of condom use at 4% and of pill use less than 1%.21 Although local areas can vary widely, in effect the remainder of the country (without improved family planning services) serves as a comparison. In our view, there is only one credible explanation for such a rapid, major and consistent increase in contraceptive prevalence in the six project areas—improved availability.
Another concern is that the community-based distribution projects were not purely a supply phenomenon because the fieldworkers provided some motivation to potential clients. We argue that, especially in the short term, such activities serve principally to spur into action those who already have some latent demand, rather than to change fertility preferences. Even in Matlab, after many years, the effect of sustained outreach on preferences for additional children was small.22
But how are these findings on household distribution relevant to family planning service delivery in general? Household distribution offers a high degree of access, and these programs typically can only be part of an overall national program effort. In our view, the key point is that demand can be satisfied by improving access in many ways. While it is generally impractical to cover an entire country with household distribution, Bangladesh is a notable example that this can be a realistic and significant part of a country-wide program.
In fact, in addition to community-based distribution activities carried out by NGOs n Pakistan, the government of Pakistan has more recently initiated a major effort with two new cadres of village-based family planning and health workers.23 The latest fertility and family planning survey (conducted in 1996-1997) suggests that this approach may be working: Nationwide, prevalence increased to 24%, and by the time of the survey, more than 16% of married women said they had received a home visit by a village-based health worker.24
GREATER PROGRAM IMPACT POSSIBLE
The substantial success of this basic supply-side approach to community-based distribution only foreshadows the potential programmatic impact. First, given the degree of cultural resistance to family planning in Pakistan, program success may come more easily in many other countries. Furthermore, we see immense opportunities to increase contraceptive use by, for example, improving access to clinical methods, enhancing client-provider interaction (including counseling about side effects) and removing major barriers to access (such as the substantial age and parity restrictions described earlier). Such changes are feasible programmatically and are likely to markedly improve both client satisfaction and program performance.25 Matlab demonstrated a rapid leap in contraceptive prevalence in the late 1970s, when program access and quality were improved, and prevalence there has continued to grow to well beyond 60% as access and quality have been improved.26
Moreover, use of mass media and other interventions offer opportunities to increase demand. Family planning programs can also be linked with a variety of other health (especially reproductive health) interventions. But doing this effectively is not necessarily easy. Integration with other health services can have a positive, negative or neutral effect on family planning use.27 Linkage and integration need to be approached in a situation-specific, pragmatic way. The key is to organize services so they enhance the impact of both family planning and other interventions to improve the clients' quality of life.
The community-based distribution projects in Pakistan clearly demonstrate that a substantial number of couples have an unmet need for family planning and confirm the construct of unmet need derived from surveys. Yet the projects also provide evidence that unmet need may be even greater in a programmatic sense. They reveal that simple, direct provision of family planning services can result in a dramatic increase in contraceptive use. In large measure, the adoption of contraception reflects the balance between the existing motivation to use it and the difficulties that hamper use. A sudden shift in that balance with improved access revealed a substantial need.
In our view, an important research and programmatic agenda for the coming decade is how best to provide accessible, client-oriented, high-quality services to meet the large unmet need. Even implementing basic family planning on a broad scale requires a huge amount of painstaking work. Yet we must learn how to organize services so as to maximize access, quality and demand for family planning while at the same time enhancing the linkages with other health interventions. Unmet need is real. Family planning service provision can and does work. The real question is how to make the most of it.
1. Robey RB, Ross J and Bhushan I, Meeting unmet need: new strategies, Population Reports, Series J, No. 43, 1996; and Freedman R, Do family planning programs affect fertility preferences? a literature review, Studies in Family Planning, 1997, 28(1):1-13.
2. Robey RB, Ross J and Bhushan I, 1996, op. cit. (see reference 1).
3. Casterline JB, Perez AE and Biddlecom AE, Factors underlying unmet need for family plannning in the Philippines, Studies in Family Planning, 1997, 28(3):
4. Population Council, The Gap Between Reproductive Intentions and Behavior: A Study of Punjabi Men and Women, Islamabad, Pakistan: Population Council, 1997.
5. Phillips JF et al., Determinants of reproductive change in a traditional society: evidence from Matlab, Ban-
gladesh, Studies in Family Planning, 1988, 19(6):313-334.
6. Prichett LH, Desired fertility and the impact of population policies, Population and Development Review, 1994, 20(1):1-55.
7. Gallen ME and Rinehart W, Operations research: lessons for policy and programs, Population Reports, Series J, No. 31, 1986.
8. Robinson WC, Shah MA and Shah NM, The family planning program in Pakistan: what went wrong? International Family Planning Perspectives, 1981, 7(3):85-92.
9. Shah IH and Cleland JG, High fertility in Bangladesh, Nepal, and Pakistan: motives vs. means, in: Leete R and Alam I, eds., The Revolution In Asian Fertility, Oxford, England: Clarendon Press, 1993.
10. Westoff CF, Unmet need, Demographic and Health Surveys Comparative Studies, No.16, Columbia, MD, USA: Institute for Resource Development/Macro Systems, 1995; and Robey RB, Ross J and Bhushan I, 1996, op. cit. (see reference 1).
11. Rukanuddin AR, Rassaque A and Hardee-Cleaveland K, Can family planning succeed in Pakistan? International Family Planning Perspectives, 1992, 18(3):109-115.
12. Sathar Z et al., Women's status and fertility change in Pakistan, Population and Development Review, 1988, 14(3):415-432.
13. Asia Foundation, Close-out report: Non-Governmental Organizations Coordinating Council for Population Welfare (NGOOC): strengthening the provision of family planning services through non-governmental organizations, Islamabad, Pakistan: Asia Foundation, 1994.
14. Enterprise and Development Consulting (EDC), Evaluation of the "A" type community-based distribution of family planning projects implemented by the Non-Governmental Organizations Coordinating Council, Islamabad, Pakistan: EDC, 1993.
15. Bulatao R, Expenditures on Population Programs in Developing Regions: Current Levels and Future Requirements, World Bank Staff Working Paper No. 679, Washington, DC: World Bank, 1985.
16. Osborn RW, The Sialkot experience, Studies in Family Planning, 1974, 5(4):123-129.
17. National Institute of Population Studies (NIPS) and Institute for Resource Development/Macro International, Pakistan Demographic and Health Survey 1990/1991, Islamabad, Pakistan: NIPS, 1992.
18. Ross J and Heaton L, Intended contraceptive use among women without unmet need, International Family Planning Perspectives, 1997, 23(4):148-154.
19. Freedman R, 1997, op. cit. (see reference 1).
20. Rogers EM, Diffusion of Innovation, 4th ed., New York: Free Press, 1995; and Aggleton P et al., Risking everything? risk behavior, behavior change, and AIDS, Science, 1994, 265(5170):341-345.
21. Population Council, 1997, op. cit. (see reference 4).
22. Phillips JF, Hossain MB and Arends-Kuenning M, The long-term demographic role of community-based family planning in rural Bangladesh, Studies in Family Planning, 1996, 27(4):204-219.
23. Rosen JE and Conly S, Pakistan's Population Program: The Challenge Ahead, Country Series #3, Washington, DC: Population Action International, 1996.
24. A. Hakim, Cleland JG and Bhatti MH, Pakistan Fertility and Family Planning Survey 1996-1997, Preliminary Report, Islamabad, Pakistan: NIPS; and London: Centre for Population Studies, London School of Hygiene and Tropical Medicine, 1998.
25. Shelton JD and Davis S, Some priorities in maximizing access to and quality of contraceptive services, Advances in Contraception, 1996, 12(3):233-237.
26. van Ginnekan J et al., Health and Demographic Surveillance System in Matlab: Past, Present and Future, Dhaka, Bangladesh: International Centre for Diarheal Disease Research, Bangladesh, March 1998.
27. Phillips JF et al., Integrating health services into an MCH-FP program: lessons from Matlab, Bangladesh, Studies in Family Planning, 1984, 15(4):153-161.
Contexte: Le niveau de besoin non satisfait de planning familial dans le monde en voie de développement est, selon la mesure des enquêtes, élevé. Il importe de déterminer s'il existe un niveau significatif de demande dormante de prestations contraceptives effectives en attente de satisfaction, surtout dans un pays tel que le Pakistan, où les efforts de promotion du planning familial se sont avérés décevants.
Méthodes: Les dossiers de six projets de distribution de contraceptifs aux ménages du Pakistan servent à déterminer la prévalence contraceptive sur une période de 12 mois. Une équipe professionnelle indépendante a mené une évaluation extérieure, avec interview des surveillants de projet, intervenants sur le terrain et clientes.
Résultats: La pratique contraceptive s'est fortement accrue dans le cadre des six projets, d'une moyenne de 12% à 39% en l'espace de deux années. L'équipe d'évaluation externe a trouvé les mesures de prévalence contracep-tive généralement exactes, non sans toutefois identifier d'autres améliorations d'accès et de qualité susceptibles d'accroître davantage la pratique de la contraception.
Conclusions: La pratique accrue de la contraception observée sous l'effet d'améliorations au niveau de l'offre des prestations confirme l'existence d'un besoin non satisfait substantiel. Les données semblent indiquer qu'une plus grande amélioration de l'accessibilité et de la qualité des prestations permettrait d'accroître davantage encore la pratique contraceptive.