The question of whether to integrate vertical primary health care programs and services (maternal health, child survival, family planning, etc.) is long-standing, and the arguments on either side have been clearly articulated. Integrated services are thought to expand access to and coverage of critical services and to improve their efficiency and cost-effectiveness by reducing duplication of service delivery functions and delivering more services per client contact.¹ Many countries have already integrated family planning services with maternal and child health services.

The 1994 International Conference on Population and Development (ICPD) and the continuing escalation of the HIV/AIDS epidemic have increased interest in integrating family planning services with those for HIV and other sexually transmitted infections (STIs). However, despite the considerable literature on the subject, there is no consensus on what it means to have integrated services, which services should be integrated or whether integration precludes continuation of stand-alone services.² Discussions often refer to "one-stop shopping" supported by multifunction workers. One review suggests that service integration "might involve the linkage of several provider functions at the service delivery point and would require modifications of worker roles, allocation of time and referral requirements [emphasis added]."³

In this commentary, we broadly define services to include the full range of provider-client interactions, from counseling and behavior-change communication to the performance of clinical procedures and delivery of medications or commodities. Any two services can be considered to be integrated when they are offered at the same facility during the same operating hours, and the provider of one service actively encourages clients to consider using the other service during that visit. According to this definition, integrated services may or may not be offered in the same physical location within the facility, and may or may not be offered by the same service provider.

"Programs" and "service delivery" are not synonymous. Programs set out broad, population-based objectives, whereas individual service delivery points may contribute to some but not all of these objectives. Family planning programs seek to stimulate demand for family planning, help individuals meet their reproductive intentions, encourage men to be more involved in reproductive health, give adequate information to allow informed choice, and provide ready access to a variety of contraceptive supplies and services. STI control programs seek to prevent new infections in the general population, treat persons with symptoms of infection, improve health-seeking behavior among those who self-diagnose infection, strengthen detection and treatment of those with an asymptomatic infection and improve the effectiveness of STI case management and treatment.

The integration of family planning into maternal and child health differs in several ways from the integration of STI and HIV services into family planning. First, unlike family planning services of 25 years ago, many public health systems have long-established STI services, sometimes under such unrelated departments as dermatology. Second, STI clients often self-diagnose and self-medicate, and third, private providers—especially pharmacies—rather than public facilities are often the client's first service delivery contact point.⁴

We contend that not all services should be integrated in all situations, and that even some potentially integrable services should sometimes be offered separately. At least three compelling factors argue against totally integrating STI and HIV services into family planning.

The first of these is that family planning clients may not be at disproportionately high risk for HIV and other STIs, and that groups that are at high risk (men and young people) may not be disposed to seek services at family planning sites. Few attempts have been made to compare STI and HIV prevalence among family planning clients with that in the general population. To the extent that family planning services attend women in union, it is unlikely that they attract large numbers of adolescents and other vulnerable populations such as "core" or "high" STI and HIV transmitters. Misdiagnosis of STIs is common when prevalence is less than 20%.⁵ The one published review of STI prevalence among antenatal and family planning clients found generally low prevalence: In 17 studies from 11 countries published between 1985 and 1998, the prevalence of gonorrheal and chlamydial infection was less than 10% in 11 studies, 10-20% in five studies, and greater than 20% in only one study.⁶ Now that HIV prevalence among antenatal clients surpasses 20% in many African settings, it may be opportune to collect data on other STIs.

Men are a key target for STI services, and their reproductive health needs differ from women's.⁷ There is growing consensus that men are not attracted to family planning settings that serve women. ProPater, a successful vasectomy clinic in São Paulo, Brazil, has never offered services for women, and Profamilia in Colombia tripled its vasectomy caseload when it opened a separate door for men.⁸ It has been suggested that men "may feel intimidated in the presence of a large number of women attending regular family planning clinics," and may be more concerned with privacy and confidentiality than women.⁹ A 1999 literature review found that increased STI coverage was achieved through diversification and special programs for men, rather than through integration of STI services into existing family planning outlets.¹⁰

Young people, especially young women, account for disproportionately high numbers of STIs and new HIV infections. Despite their need for many of the same services as older adults, sexually active, unmarried youth tend not to use existing reproductive health services. Youth advocates now call for separate "youth-friendly" services that would integrate a wide range of reproductive health services, including family planning as well as STI and HIV services.¹¹

A second important consideration in decisions about whether to integrate services is that the delivery and management requirements of family planning services may not always be operationally compatible with those of STI and HIV services. To assess the operational potential for integrating family planning with STI and HIV services, we must specify which services and which service delivery points are being considered. Family planning services that may be considered for integration include counseling and promotion, and provision of supply and clinical methods.* Among the potentially integrable services associated with STI and HIV prevention are promotion (behavior-change communication, dual protection and condom distribution) and voluntary HIV counseling and testing. For STI treatment, such services include the syndromic management of STIs for urethral discharge in men and for genital ulcers in both sexes, and the clinical diagnosis and treatment of STIs for women and men.

Each of these services has minimum service delivery requirements that dictate the settings in which it may be safely and effectively offered. Thus, while a clinical facility may be able to offer the full range of contraceptive counseling and methods, a community distributor may be able to provide only general counseling, barrier methods and oral contraceptives, and referrals for other methods. Similarly, a clinical facility could offer both STI testing and appropriate treatment for women with vaginal discharge and syndromic management for individuals with genital ulcers and men with urethral discharge, while a pharmacy may be able to provide only syndromic management for genital ulcers and male urethral discharge and referrals for women with vaginal discharge. Table 1 describes minimum service delivery requirements for the services listed above.

The third compelling factor arguing against total integration of services is the lack of simple and effective technologies for diagnosing and treating STIs in asymptomatic women or women with vaginal discharge. Syndromic management is not able to pick up asymptomatic infections and is not specific enough to distinguish between vaginal discharge caused by an STI and that resulting from other etiology. The cost of adding routine clinical screening for STIs among family planning clientele may not be justified by the number of infections detected and treated.

Shelton argues that integrating family planning with HIV and STIs should be a high priority for the prevention of disease, but is not a viable curative model. "Unfortunately, at this moment, the primary curative approach for women in the general family planning and maternal and child health context (syndromic management) is difficult to implement, will not have a true impact on the HIV and STI epidemic and is largely ineffective."¹²

CONCLUSIONS

Promotional activities, whether for family planning or for STI and HIV prevention, are clearly integrable in both service settings.¹³ Moving beyond promotion, however, we argue that it makes sense to consider integrating new services into an existing setting only to the extent that the existing clientele is in need of the new services and that the delivery requirements of the new services are compatible with those of the existing services.

Except for infections causing urethral discharge in men and genital ulcers in both sexes, STIs (including HIV) cannot be easily and inexpensively diagnosed and treated. This challenge is the major barrier to integrating family planning with STI and HIV services. With this issue in mind, we make the following programmatic recommendations:

• Family planning services whose clients are at high risk for STI and HIV infection should consider integrating only STI and HIV services whose delivery requirements can easily be met by existing capacity at the delivery site. In clinical settings such as health centers and polyclinics, integration may be achieved by immediate referral to another provider in another location within the same facility, as long as the client is not required to return on another day.

• Family planning services whose clients are not at high risk or that cannot accommodate STI and HIV delivery requirements within existing capacity should build relationships with appropriate services to which clients can be referred.

• Focusing exclusively on integrating STI and HIV services into family planning services is short-sighted and will not by itself resolve the problem of expanding access and coverage for either service—particularly for men. Therefore, consideration should also be given to integrating family planning into STI and HIV service sites, especially STI clinics, voluntary counseling and testing centers and pharmacies.

• Because sexually active, unmarried young people are at disproportionately high risk both of unplanned pregnancy and of STI and HIV infection, youth-friendly services in settings with a high prevalence of STIs or HIV should consider offering the full range of reproductive health services, with the clear understanding this is likely to increase operating costs.

A final note—current work on the integration of family planning with STI and HIV services is taking place in the shadow of health sector reform.¹⁴ Initiated prior to ICPD, health sector reform considers reproductive health as one of many vertical programs to be integrated into the overall health care system, often through essential-services packages. Around the world, countries such as Bangladesh, India, Brazil, Mexico, Senegal, South Africa, Uganda and Zambia are designing and implementing such health care packages. Few publications in the integration literature place reproductive health in the context of health sector reform. Decisions on essential-services packages are being made now. Consensus is urgently needed on what makes sense for integration of family planning services with STI and HIV services, and under what circumstances, to inform integration efforts under health sector reform.