Methods for studying private sector supply of public health products in developing countries: a conceptual framework and review

Abstract

The private sector is an important supplier of public health products (PHPs) in developing countries. Although there are concerns about the quality and affordability of these products, private providers also offer possibilities for expanding access to key commodities. This paper proposes a conceptual framework for understanding the public health implications of private sales of PHPs. It reviews methods for studying these sales, together with their advantages and shortcomings. Ten methods are identified which can be used for studying the behaviour of providers and consumers. The effects of seasonal variation are discussed, together with the challenges of creating a sampling frame and studying illicit behaviour. We conclude that relatively little is known about the sales of PHPs, that more is known about contraceptives and drugs than about the newer products, and that the demand side of the market has been studied in greater depth than the behaviour of suppliers. The existing toolbox is biased towards formal providers, and thus, probably towards understanding the provision of PHPs to those who are better off. Methods for studying the supply of PHPs in outlets used by poor people is a priority area for further methodological development.

Introduction

Recent years have seen increased recognition of the importance of the private sector in the health systems of low- and middle-income countries (Bennett, McPake, & Mills, 1997; Hanson & Berman, 1998; IHSD & LSHTM, 2000; Smith, Brugha, & Zwi, 2001). Private providers are an important source of care for poor people, who in many countries use the private sector nearly as much as better-off groups (Castro-Leal, Dayton, Demery, & Mehra, 2000).

A deepening understanding of the complexity of the private sector has contributed some description of the range of different providers involved (formal and informal, for-profit and non-profit, ranging from simple shops and solo practitioners through to complex organisations providing tertiary-level inpatient care). There is also growing evidence of the extent to which use of the private sector is segmented by socioeconomic status, suggesting that poor people are more likely to use the more informal parts of the private sector, such as shopkeepers, drug sellers, and solo practitioners who may not be formally qualified (Smith et al., 2001).

The range of types of health care which is purchased through the private sector is highly heterogeneous. The spectrum runs from complex services which include diagnostics and inpatient care, through to the supply of individual public health products (PHPs), such as condoms, mosquito nets, and certain over-the-counter drugs. While there is a growing evidence base concerning health services that are provided by the private sector, the information concerning the supply of commodities such as drugs and contraceptives is scanty, tends to fall outside of the general health services literature, and is often not considered in policy debates about the role of the private sector. We refer to such commodities as PHPs and define them in the next section.

Three arguments provide support for placing a higher priority on understanding the role of the private sector in distributing PHPs. First, private providers are an important and rapidly growing source of a variety of PHPs. Distribution data from the Philippines, for example, show that over the period 1994–2001, the private sector (including social marketing) share of the supply of oral contraceptives (OCs) increased from 26% to 43% of all cycles distributed (PSI, unpublished data). The volume of OCs provided by the private sector in Pakistan doubled over the period 1994–2000 (PSI, unpublished data). In Kenya, Demographic and Health Surveys (DHS) in 1989, 1993 and 1998 show a steady decline in the public share of condom supply (from 55% to 21%) and an increase in the share supplied by ‘other private’ sources (distinguished from ‘private medical’) from 25% to 52% (Measure DHS, 2002). In Indonesia, the government has completely phased out its supply of condoms, and the private sector share of distribution has risen from 50% in 1995 to 100% in 2001 (Purdy, 2002). The number of mosquito nets produced by manufacturers in Tanzania is estimated to have increased from 400,000 in 1994 to 1.2 million in 2000, with approximately half of these distributed through the private retail sector (Hanson & Jones, 2000).

A second justification for studying private supply of PHPs is that these products are often directed at problems of public health importance (e.g., sexually transmitted infections (STIs), malaria). And third, the provision of such essential health products through the private sector may create important new opportunities for expanding their use among target populations, most particularly among poor people. At the same time, there is a range of potential problems with private sector supply of PHPs that needs to be addressed, such as issues of affordability, accessibility and quality—issues around which there is currently relatively little information. Better information is needed about both the opportunities and potential problems of private sector supply of PHPs, in order to balance their effects, and to determine whether the appropriate policy response is to encourage private sector provision, tolerate it, or discourage it.

Typically, routine health information systems have very poor coverage of the private sector. Such information can often prove difficult to collect, as providers may be unwilling to provide details about their activities, or may lack paper records. In the face of this need for better information, the aim of this paper is to review the alternative methods for studying private sector supply of PHPs. While there have been a number of reviews of methods for studying specific products such as drugs (Ross, 1997; Van der Geest, 1982) or OCs (Green, 1988; Janowitz, Suazo, Fried, Bratt, & Bailey, 1992), this paper differs from existing reviews in two respects. First, it takes a broader perspective and focuses on the whole category of PHPs; and second, it focuses specifically on methods appropriate for studying the supply of these products in the private sector. We begin by proposing working definitions for the terms PHP and private providers. We go on to present a conceptual framework for understanding why markets for PHPs emerge, the public health implications of market provision of PHPs, and identify the variables which are of interest when studying the sales of PHPs through the private sector. We then review the main methods for studying PHP sales through the private sector, and some of the cross-cutting methodological issues such as constructing a sampling frame, the effects of seasonality and studying illicit behaviour. Finally, the discussion and conclusions are presented.