3
Infection-Free Sex and Reproduction

Sexual relations and reproductive events should be free of infection. Reproductive tract infection (RTI) is a generic term we use to cover three types of infection: sexually transmitted diseases (and infections) (STDs), endogenous vaginal infections, and infections related to reproductive tract procedures.

RTIs are caused by a variety of bacterial, viral, parasitic, and fungal microorganisms, and they have major consequences that include infertility, ectopic pregnancy, chronic pelvic pain, genital neoplasia, and enhanced transmission of the human immunodeficiency virus (HIV). The sexually transmitted infections are associated with a spectrum of acute, chronic, and pregnancy-related conditions (Brunham and Ronald, 1991); see Table 3-1. Endogenous vaginal infections include bacterial vaginosis and candidiasis, both the result of overgrowth of organisms normally present in the vagina. Endogenous infections have also been associated with prematurity and low birth weight (Gravett et al., 1986). Procedure-related infections can involve the lower or upper reproductive tract and can result in both acute sepsis and such long-term complications as infertility.

RTIs are a persistent global health problem: as syphilis preoccupied clinicians at the beginning of the twentieth century, RTIs are a major international public health problem as it ends (Holmes et al., 1990; Wasserheit, 1994; Piot and Islam, 1994). Trends in STDs have become increasingly important indicators of unsafe sexual behavior in both developed and developing countries, and changes in trends monitor the effectiveness



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions 3 Infection-Free Sex and Reproduction Sexual relations and reproductive events should be free of infection. Reproductive tract infection (RTI) is a generic term we use to cover three types of infection: sexually transmitted diseases (and infections) (STDs), endogenous vaginal infections, and infections related to reproductive tract procedures. RTIs are caused by a variety of bacterial, viral, parasitic, and fungal microorganisms, and they have major consequences that include infertility, ectopic pregnancy, chronic pelvic pain, genital neoplasia, and enhanced transmission of the human immunodeficiency virus (HIV). The sexually transmitted infections are associated with a spectrum of acute, chronic, and pregnancy-related conditions (Brunham and Ronald, 1991); see Table 3-1. Endogenous vaginal infections include bacterial vaginosis and candidiasis, both the result of overgrowth of organisms normally present in the vagina. Endogenous infections have also been associated with prematurity and low birth weight (Gravett et al., 1986). Procedure-related infections can involve the lower or upper reproductive tract and can result in both acute sepsis and such long-term complications as infertility. RTIs are a persistent global health problem: as syphilis preoccupied clinicians at the beginning of the twentieth century, RTIs are a major international public health problem as it ends (Holmes et al., 1990; Wasserheit, 1994; Piot and Islam, 1994). Trends in STDs have become increasingly important indicators of unsafe sexual behavior in both developed and developing countries, and changes in trends monitor the effectiveness

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions TABLE 3-1 Major Sexually Transmitted Infection Microbial Agents and Their Effects Sexually Transmitted Infection Acute Disease Chronic Disease Pregnancy-Associated Disease Bacterial       Neisseria gonorrhoeae Urethritis; Cervicitis; Salpingitis Infertility; Ectopic pregnancy; Chronic pelvic pain Prematurity; Septic abortion; Ophthalmia neonatorum; Postpartum endometritis Chlamydia trachomatis Urethritis; Cervicitis; Salpingitis Infertility; Ectopic pregnancy; Chronic pelvic pain Ophthalmia neonatorum; Pneumonia Postpartum endometritis; Prematurity (?) Treponema pallidum Primary and secondary syphilis Neurosyphilis; Cardiovascular syphilis; Gumma Spontaneous abortion; Stillbirth; Congenital syphilis Haemophilus ducreyi Genital ulcer None known None known Viral       Human immunode-ficiency virus (HIV) Mononucleosis-like syndrome AIDS Prematurity; Stillbirth; Perinatal HIV infection Human papilloma virus (HPV) Genital warts Genital cancer Laryngeal papillomatosis Herpes simplex virus type 2 (HSV-2) Genital ulcer Recurrent; genital herpes Congenital and neonatal; HSV; Prematurity Hepatitis B virus (HBV) Acute hepatitis Chronic hepatitis; Cirrhosis; Hepatoma; Vasculitis Perinatal HBV Parasitic       Trichomonas vaginalis Vaginitis; Urethritis   Prematurity; Low birthweight (?), Evidence is weaker than for other effects.   SOURCE: Adapted from Brunham and Ronald (1991:62).

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions of STD prevention programs, including those aimed at controlling the spread of the HIV. In this chapter we examine RTIs in both the developed and developing world: their magnitude and dimensions, their determinants and consequences, the available means to prevent and treat them, and the strengths and limitations of possible intervention strategies. Unfortunately, the goal of infection-free reproductive health appears as far away today as it did 100 years ago, and we conclude that a multifaceted public health approach is necessary to get us closer to the goal of infection-free sex. REPRODUCTIVE TRACT INFECTIONS Measures Estimates of the incidence and prevalence of RTIs vary according to the source of data and the methods used to detect them (Rothenberg, 1990). In most developed countries, sources generally include reportable infections (e.g., gonorrhea and syphilis), visits to office-based practices, national surveys of representative populations, and data on patients at specialized health facilities (e.g., STD clinics, family planning clinics). Unfortunately, each of these sources has limitations. Data on reported infections are affected by differences in the completeness of reporting among different health care sources. Infections diagnosed in public facilities are reported more frequently, so these data are susceptible to biases related to the characteristics of individuals who tend to use public clinics. Data from private clinicians' practices are often affected by the absence of diagnostic validation. National surveys are limited by their sporadicity and the superficial nature of the analytic variables. Data from specific health facilities suffer from the problem of patient selection bias, as well as geographic variation. In developing countries, data sources are even less representative (Meheus, Schulz, and Cates, 1990; Wasserheit, 1989; Over and Piot, 1993; World Health Organization, 1995b). Few nations have even rudimentary surveillance systems, so RTI incidence is usually derived from patient visits to health care facilities. RTI prevalence is typically extrapolated from studies of selected high-risk populations. Although these data provide useful estimates, they must be viewed with even more caution than reports from the developed world. Studies have shown wide discrepancy between women's self-reported symptoms and medical diagnoses of prevalent conditions (Younis et al., 1993; Zurayk et al., 1995). This discrepancy could be due to two factors: RTIs are sometimes asymptomatic, and even when symptomatic, women's perceptions of the symptoms may

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions not prompt help-seeking behavior. For example, women may consider vaginal discharge a ''normal" occurrence, even when accompanied by color or odor, because it is so widespread in the community (Zurayk et al., 1995). Alternatively, women may recognize the symptom but may not want to report it because of stigmatization. In developing countries, research is needed on women's perceptions of reproductive health conditions in order to design better instruments to measure RTIs at the community level. Finally, data on specific RTIs also vary by the type of infection (Centers for Disease Control and Prevention, 1994; Laga, 1994; World Health Organization, 1995b), depending on whether current or cumulative infection is being measured. In both developed and developing countries, symptomatic viral infections (measured by physician visits) occur less frequently than serologic or cytologic indicators of the cumulative number of infected persons. Thus, care must be used in making comparisons among different measures of RTIs. Prevalence in Developed Countries In North America and Europe, the incidence of genital chlamydial infections and viral STDs steadily increased during the 1970s and 1980s, while the incidence of gonorrhea generally declined (Gershman and Rolfs, 1991); levels of syphilis varied among different population subgroups (Wasserheit, 1994; Over and Piot, 1993). Endogenous infections such as bacterial vaginosis and candidiasis remained high, accounting for up to 5 percent of all primary care visits (Berg, 1990). Overall, syphilis incidence rose during World War II, but fell thereafter, coinciding with the introduction of penicillin. The lowest levels were observed at the end of the 1950s, increasing from the 1960s on. A rapidly rising male-to-female ratio coincided with the spread of syphilis among men having sex with men throughout the 1970s. In the 1980s, however, probably as a result of the safer sexual behaviors stimulated by HIV prevention messages, syphilis rates in gay males declined dramatically (Rolfs and Nakashima, 1990; Webster and Rolfs, 1993). At the same time, in the United States and other developed countries with heterogeneous populations, syphilis rates climbed during the late 1980s among heterosexuals of minority races. By the mid-1990s, syphilis levels were again falling in most developed countries. The trends for gonorrhea have been more consistent. Gonorrhea incidence generally increased in the 1960s, and then, depending on the country, declined at different points in the 1970s. In the United States, most of the decline occurred among older, white populations, with gonorrhea rates remaining high among minority races and adolescents (Webster,

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions Berman, and Greenspan, 1993). Overall, gonorrhea is associated with a younger mean age than syphilis among all gender and race categories. Chlamydia infections became the most prevalent bacterial STD in the developed world during the 1970s, when gonorrhea levels began declining. Chlamydia is not a universally reported infection; therefore, syndromes have been used as proxies to monitor trends. In England and the United States, nongonococcal urethritis diagnoses exceeded those of gonorrhea in the early 1970s, with the gap widening in recent years (Centers for Disease Control and Prevention, 1994). In one rural Canadian province, active surveillance showed rates of chlamydia nearly three times higher than those of gonorrhea (Alary, Joly, and Poulin, 1989). In all developed countries, chlamydial infections in women exceed those in men, and chlamydial prevalence is strongly correlated with younger age and heterosexual behaviors. Sexually transmitted viral infections are widespread. In the United Kingdom and the United States, the numbers of symptomatic genital herpes and genital warts cases increased 5- to 15-fold during the 1970s and 1980s (Centers for Disease Control and Prevention, 1994). In the developed world, symptomatic genital herpes causes over 10 times more genital ulcer cases than does syphilis. Moreover, recent investigations have shown that symptomatic infections with herpes simplex viruses (HSV) are only a small fraction of the total prevalence (Johnson et al., 1994): for example, HSV-2 has occurred among an estimated 30 million Americans, even though less than one-quarter perceive themselves ever to have had genital herpes. Similarly, diagnoses of both symptomatic genital warts caused by the human papilloma virus (HPV) and of asymptomatic infection have increased enormously during the last two decades (Centers for Disease Control and Prevention, 1994). HPV infections of the cervix and vagina have emerged as the most common STD among sexually active adolescent populations. Since no serologic test is available to determine previous infections and HPV cannot be recovered through tissue culture, determining the full extent of these infections is extremely difficult. The epidemiologic pattern of HIV infection in the developed world is different from that in the developing world (Over and Piot, 1993). Beginning in the mid-1970s, HIV was transmitted among homosexual and bisexual men and resulted in acquired immune deficiency syndrome (AIDS) by the early to mid-1980s. The virus entered the injection drug-using populations in the early 1980s and rapidly spread in Western Europe and North America during the decade. Limited heterosexual transmission occurred in these regions until the late 1980s; since 1989, however, the greatest proportionate increase of reported AIDS cases has been among heterosexuals and this trend is expected to continue (Centers for Disease Control and Prevention, 1995). By the end of 1996, an estimated 1.3 million

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions persons in North America, Western Europe, Australia, and New Zealand were living with HIV/AIDS (UNAIDS and World Health Organization, 1996) Among all AIDS cases reported to the World Health Organization (WHO) by 1993, more than one-half came from the United States and Europe (40% and 13%, respectively) (Way and Stanecki, 1994). However, because reports from the developing world are less complete than those from the developed world, WHO estimates that cases reported by the United States actually represent less than one-sixth of the world total of AIDS cases (World Health Organization, 1995a). Overall, in most developed countries, the incidence of classical RTIs such as gonorrhea and syphilis declined rapidly during the 1980s among middle and upper socioeconomic strata; in North America, however, the incidence of these same RTIs remained stable or actually increased within young, low-income, minority populations. HIV infection has also become entrenched in these same disenfranchised groups, and the proportion that is spread through heterosexual behaviors is increasing. Prevalence in Developing Countries The epidemiology of RTIs in developing countries differs greatly from that in developed countries (Wasserheit, 1989; Brunham and Embree, 1992; Over and Piot, 1993; Piot and Islam, 1994; World Health Organization, 1995b). Overall, RTIs are a more frequent health problem in developing countries. WHO estimates at least 333 million new cases of curable STDs occurred globally in 1995 (World Health Organization, 1995b), mostly in developing countries. RTIs are among the top five causes of consultation at health services in Cameroon, representative of many African countries; and among adults, RTI is the leading diagnosis (Meheus, Schulz, and Cates, 1990). In Zimbabwe, up to 10 percent of the population had a documentable RTI (Laga, 1994). Intensive studies of women in India, Bangladesh, and Egypt have found RTI rates ranging from 52 percent to 92 percent, less than one-half of which were recognized by the women as abnormal (Bang et al., 1989; Wasserheit et al., 1989; Younis et al., 1993; Singh et al., 1995). Among RTI syndromes, the etiology of genital ulcer infection differs significantly from that in the developed world: syphilis and chancroid are the major causes of genital ulcers in tropical countries, with genital herpes accounting for a smaller proportion (Brunham and Ronald, 1991). Table 3-2 shows the prevalence of RTIs among pregnant women in some developing countries for which there are data. Syphilis in developing countries remains at levels that were seen in developed countries a century ago. One must be cautious in looking at the data, however, because a seropositive serological test for syphilis

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions TABLE 3-2 Prevalence of Selected Reproductive Tract Infections Among Pregnant Women, Selected Developing Countries, 1980s: in percent Country Syphilisa Gonorrheab Chlamydiab Cameroon — 14.5 — Central Africa Republic 9.5 9.5 — Ethiopia 16.9 — — Gabon — 5.5 9.9 Gambia 11.0 6.7 6.9 Ghana — 3.1 — Kenya — 6.6 29.0 Malawi 13.7 — — Malaysia 2.0 0.5 — Mozambique 6.3 — — Nigeria 2.1 5.2 6.5 Saudi Arabia 0.9 — — Somalia 3.0 — — South Africa 20.8 11.7 12.5 Swaziland 33.3 3.9 — Tanzania 16.4 6.0 — Uganda 40.0 — — Zaire 2.0 — — Zambia 12.5 11.3 — Zimbabwe — 7.0 9.9 a Diagnosis is based on Treponema Pallidum Haemagglutination test (TPHA) and Fluorescent Treponemal Antibody test (FTA-Abs). b Diagnosis is based on culture of vaginal secretion. SOURCE: Data from World Health Organization (1986), Wasserheit (1989), and Over and Piot (1993). could be due to sexually transmitted infection or to previous infection with nonvenereal treponematoses (Larsen, Hunter, and Creighton, 1990). WHO estimates that in 1995, approximately 12 million new cases of adult syphilis will occur worldwide, with the greatest number in South Asia and sub-Saharan Africa (World Health Organization, 1995b). With this limitation in mind, past syphilis infections among pregnant women have ranged from less than 1 percent in Saudi Arabia to more than 33 percent in Swaziland (see Table 3-2). In one population of rural Somalia, nearly one-quarter of men and women in the general population had past evidence of syphilis (Over and Piot, 1993). Overall, syphilis is highly prevalent in developing countries, and considerable risk for congenital syphilis exists in many areas (Brunham and Embree, 1992). Gonorrhea, like syphilis, is more prevalent in developing countries

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions than in developed countries. Estimates for large cities in Africa suggest an annual gonorrhea rate of between 3,000 and 10,000 cases per 100,000 inhabitants (Laga, 1994). These frequencies have been extrapolated mainly from attendance at general health care centers or from special surveys of prevalence in population groups that may not be representative. Surveys of gonorrhea in pregnant women have ranged from one-half of 1 percent in Malaysia to 40 percent in Uganda (see Table 3-2). Among women visiting family planning clinics, rates have ranged from 2 percent in Swaziland to 17 percent in Kenya (Wasserheit, 1989). WHO estimates that approximately 62 million new cases of gonorrhea will have occurred in 1995 among adults worldwide; the largest number will be in South Asia and sub-Saharan Africa (World Health Organization, 1995b). Genital chlamydial infections in the developing world have a prevalence similar to those in the developed world, both occurring at high levels. Among pregnant women, chlamydial infections are more frequent than gonococcal, with rates ranging from 6 percent in Nigeria to 29 percent in Kenya (see Table 3-2). Among men with symptoms of urethritis, rates of chlamydial infection (as measured by nongonococcal urethritis) appear to be lower than in the developed world. However, because chlamydia causes less symptomatic infections, patients may not be motivated to seek treatment in resource-poor areas where health care is difficult to obtain (Laga, 1994). WHO estimates that approximately 89 million new cases of chlamydia will have occurred in 1995 among adults worldwide; again, as with syphilis and gonorrhea, the greatest number will be in South Asia and sub-Saharan Africa (World Health Organization, 1995b). Chancroid is highly endemic in many tropical countries, in particular Southeast Asia and eastern and southern Africa (Piot and Islam, 1994). The global incidence of chancroid is probably equivalent to that of syphilis. (There is a resurgence of interest in this infection due to the availability of new methods for detecting the causative organism, Haemophilus ducreyi.) In both developed and developing countries, commercial sex workers and their clients play a crucial role in the spread of chancroid. On the basis of extrapolations from selected local studies, WHO estimates trichomoniasis is the most common curable STD (World Health Organization, 1995b). Trichomonal infection is frequently asymptomatic in men. Prevalence rates among women attending antenatal clinics range from 12 percent in Kenya to 47 percent in Botswana (World Health Organization, 1995b). Cross-sectional screening has found this infection in 11 percent of Nigerian adolescent women (Brabin et al., 1995). WHO estimates that 170 million new cases of trichomoniasis will have occurred in 1995 among adults worldwide, especially in developing countries (World Health Organization, 1995b).

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions The level of sexually transmitted viral infections appears to be quite high. Serologic studies have found that asymptomatic herpes simplex type 2 infections are frequently more common than evidence of past syphilis (Corey, 1994). Likewise, HPV has been the most prevalent RTI found in selected studies, even in comparison with vaginal bacterial infections (Singh et al., 1995). In Asia and elsewhere, hepatitis B virus (HBV) is widespread (Lemon and Newbold, 1990); this virus is transmitted not only among sexual partners, but also from mothers to their newborns. HIV infection in the developing world has been predominantly transmitted through heterosexual behaviors (Way and Stanecki, 1994). By the end of 1996, more than 22 million persons were infected worldwide, of whom 14 million were in sub-Saharan Africa (UNAIDS and World Health Organization, 1996). The HIV epidemic emerged later in Asia; however, rapid increases have occurred in both South and Southeast Asia. A striking increase in the percentage of HIV-infected commercial sex workers in Thailand and India, for example, provides a harbinger of future levels of HIV infection among the general population in these countries. The level of endogenous RTIs among women in developing countries is typically even higher than that of the traditional STDs. In rural India, with careful physical examination and laboratory investigation, 92 percent of women were found to have genital infections (Bang et al., 1989). Less than one-half of these women had reported any RTI symptoms when interviewed prior to being examined. Similar situations were found both in Egypt (Younis et al., 1993) and in another region in India (Singh et al., 1995). The type of dominant endogenous infection varied among the populations, although bacterial vaginosis and candidiasis were both common. Until recently, the inability to mount a coordinated prevention response, even in the face of the pervasive HIV epidemic, has led to continued high levels of all RTIs in developing countries. DETERMINANTS OF RTIs Many factors affect the current high level of RTIs in both developed and developing countries (Wasserheit, 1994; Holmes, 1994; Cates and Holmes, 1992), and the factors differ not only from nation to nation, but also from community to community. Three interrelated environmental levels affect RTI patterns: physiological microenvironment, personal behavioral environment, and sociocultural macroenvironment. These levels can be further stratified according to specific effects related to the organism, the host, or the situation under which transmission occurs. At the microenvironmental level, microbiologic, hormonal, and immunologic variables most directly influence individual susceptibility, infectiousness, and the development of RTI sequelae. These microenvironments

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions are shaped, in part, by the personal environments created by an individual's sexual and health-related behaviors (Wasserheit, 1994). Substance use behaviors also affect reproductive health. Most direct and dramatic is the role of intravenous drug use in HIV transmission, which has been prominent in the developed countries and in parts of Asia and Latin America. Use of alcohol and other drugs is associated with casual sex (Ferry, 1995), sexual violence and coercion, resulting in unwanted exposure to infection. Besides affecting acquisition of infection and the development of sequelae, personal behaviors mediate risk of exposure to infection. They are, therefore, the determinants that most directly affect changing infection patterns characterized by the emergence, maintenance, or reemergence of RTIs at a community level. Individual behaviors and risk are, in turn, molded by powerful macroenvironmental forces that include socioeconomic, demographic, geographic, political, epidemiologic, and technological factors. Over the past 20 years, the profound changes that have occurred in the macroenvironment of both the developed and developing worlds have largely shaped patterns of RTI prevalence. Physiological Microenvironments Trend data are not available for most of the biological indices of the microenvironments that affect RTI patterns. For example, although it is known that the vaginal flora and acidity (pH) of the microbiological microenvironment influence susceptibility to RTIs (Hillier et al., 1992), it is not known how these factors have changed over time in either developed or developing countries. Similarly, manifestations of the hormonal microenvironment—such as the size of the zone of cervical ectopy, the penetrability of the cervical mucus, the patency of the cervical canal, the phase of the menstrual cycle, and possibly even the composition of seminal and prostatic fluids—may contribute to susceptibility to RTIs or their sequelae (Ehrhardt and Wasserheit, 1991). Trend data are obtainable, however, for a few of the microenvironmental parameters. By altering cervicovaginal ecology, modulating vaginal pH, or other mechanisms, one RTI may increase susceptibility to other RTIs and their complications (Wasserheit, 1992). Thus, recent increases in RTIs may have fueled some of the changing disease patterns biologically, as well as epidemiologically. Moreover, high levels of RTIs have been implicated in the rapid spread of HIV—particularly among homosexual men in the developed countries and heterosexual men and women in Africa and Asia (Over and Piot, 1993). Decreases in the age of menarche, a manifestation of the nutritional and hormonal microenvironment, have been documented in both developed and developing countries (Rees,

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions 1993). Competing factors that influence the immunological microenvironment, such as nutrition, pregnancy, HIV infection, and prior exposure to STDs, can also be traced over time in some populations. Personal Behavioral Environments Both sexual and health-related behaviors affect the prevalence of RTIs; see Figure 3-1. Sexual Behaviors Risky sexual behaviors have been one of the primary determinants of changing patterns of RTIs, including HIV infection. Early sexual debut appears to be associated with subsequent patterns of multiple sexual partners and sex with risky partners in both developed and developing countries (Kost and Forrest, 1992; Caraël et al., 1995). Furthermore, the hormonal microenvironment is age dependent: for young adolescents, behavioral risk factors such as multiple, risky partners combine with biological risk factors such as large zones of cervical ectopy to cause high RTI rates. Both commercial sex (exchange of sex for money or drugs) and specific sexual practices (such as anal intercourse, intercourse during menses, or "dry sex") have also been linked to increased risk of RTIs or their sequelae (Caraël et al., 1995). Occupations that require long intervals away from home (e.g., truck drivers, migrant workers, military personnel) also place persons in higher risk personal environments. In developing countries, marked variability exists among different nations in the reported level of risky sexual behaviors (Cleland and Ferry, 1995). For FIGURE 3-1 Behavioral Personal Environments that Affect RTI Patterns Sexual Behavior Health Behavior Age at coital debut Barrier contraceptive use Number of sex partners Hormonal contraceptive use Commercial sex Use of intrauterine device (IUD) Sexual practices Vaginal douching   Circumcision   Early health care utilization   Compliance with therapy   Provider screening   SOURCE: Wasserheit (1994)

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions Unfortunately, the performance of these algorithms in practice has been variable: the algorithms work well for genital ulcer disease and symptomatic urethral discharge in men; however, in the case of those syndromes most common among women—vaginal discharge and lower abdominal pain—their performance is less than optimal (Vuylsteke et al., 1993). A number of studies have shown a poor correlation between women's symptoms, clinical observations made during vaginal and pelvic examination, and the presence of infection as detected by laboratory testing (Bang et al., 1989; Wasserheit et al., 1989; Grosskurth et al., 1994; Younis et al., 1993; Zurayk et al., 1995). For example, in one study conducted in Zaire to assess the diagnostic validity of the WHO flowcharts, the hierarchical algorithms for vaginal discharge and lower abdominal pain had only 48 percent sensitivity and 75.2 percent specificity when applied to pregnant women and 54.9 percent sensitivity and 52.2 percent specificity when applied to female sex workers (Vuylsteke et al., 1993). A poor correlation with the actual presence of infection means that a large proportion of women are unnecessarily treated, while others who have infections remain undetected. This situation is further compounded by the complexities of health-seeking behavior among men and women who experience symptoms of infection. Many people with RTI symptoms follow a sequence of therapeutic practices, beginning with self-treatment in the home and progressing through remedies provided by family members, pharmacists, and a variety of traditional healers before seeking health services from allopathic providers (Olukoya and Elias, 1994). Whether the development of health education efforts that encourage more prompt health-seeking behavior among symptomatic individuals will improve the utility of clinical algorithms remains to be tested. More timely health-seeking behavior will change the characteristics of the population presenting for syndromic management, which may influence the performance characteristics of the therapeutic algorithms. Consequently, it is important to validate standardized clinical management tools among the populations in which they will actually be used. Although the current WHO algorithms are not ideal, standardized case management must be considered a mandatory intervention for the responsible delivery of contraceptive and other reproductive health services. Many men and women currently come to services complaining of symptoms related to RTIs. Unfortunately, providers are often poorly informed about women's perceptions of reproductive tract symptoms and the cultural idioms of symptom presentation (Zurayk et al., 1995). Failure to address a client's concern undermines the credibility of the service delivery system. Too often individuals in need of care are sent away without any attention given to their presenting complaint, or they

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions are given a vague referral to an STD treatment facility to which they cannot or will not go. The challenge is to refine current guidelines for clinical management of symptomatic RTIs within existing reproductive health facilities. Ideally, this would be done in the light of data concerning the local epidemiology of infection and antibiotic resistance patterns. One benefit of such an approach would be to minimize the costly and unnecessary overtreatment of clients and potentially improve clinical outcomes through standardizing therapy, thereby allowing program managers to select the most cost-effective therapeutic regimens (again, ideally based on local data) and streamline drug procurement processes. In refining the algorithms, one important issue that urgently requires clarification is the utility of linking syndromic management to behavioral risk screening. WHO flowcharts have included a number of screening questions regarding sexual behavior (World Health Organization, 1991). Risk assessment is positive only if patient answers yes to: Does your sexual partner have a discharge from his penis or open sores anywhere in his genital area? Or if she answers yes to two or more of the following: Are you younger than 21 years? Are you unmarried or not in union? Have you been with your husband or sexual partner for less than 3 months? Have you had more than one sexual partner in the last 4 weeks? The purpose of these questions is to help distinguish between cervical infections (which have more serious complications and require more intensive treatment) and vaginal infections (which are more common). Given the generally poor predictive performance of the vaginal discharge algorithm, treating all women with a discharge syndrome for cervical infection has the potential to result in high levels of overtreatment with broad spectrum antibiotics, an outcome that has significant costs both in terms of resources and the possible emergence of antibiotic resistance. Unfortunately, there are few data to determine whether these questions significantly improve the diagnostic performance of the algorithms in the majority of settings in developing countries where they will be applied. As discussed further below, risk screening approaches have sometimes had paradoxical results (e.g., when applied to interventions to reduce maternal morbidity and mortality; Rooks and Winikoff, 1990). Therefore, caution must be used in recommending the combination of an essential, potentially highly cost-effective intervention strategy (standardized case management) with a behavioral risk screening approach that has had limited testing. The combined approach may be helped by local definition of the risk screening criteria. One recent study conducted in a Jamaican STD clinic, for example, evaluated the sensitivity of diagnosing cervical infection with an algorithm designed to combine a woman's symptomatology with

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions an assessment of her risk status. In that population, the WHO flowchart, including risk assessment, was 84 percent sensitive and 40 percent specific for cervicitis diagnosis (Behets et al., 1995). Adaptation of the algorithm on the basis of local data was found to be slightly more specific, though somewhat less sensitive, and a modification of the WHO algorithm was defined for local use. Behets et al. (1995:15) concluded that ''the generation of regional data enhanced local acceptability of algorithm approaches." Ultimately, the predictive performance of standardized algorithms may be further improved by the development of newer and simpler diagnostic tests for RTIs. Recently, a number of initiatives to promote the development of STD diagnostic technology applicable in resource-poor settings have been launched (Berkely, 1994), and there has been some early success. For example, the use of a simple dipstick test for the presence of leukocyte esterase may greatly improve screening efforts among men who either have asymptomatic urethritis or experience mild symptoms without gross discharge (Shafer et al., 1989; Mayaud et al., 1992). Rapid advances in gene amplification and the use of urine as a specimen source for antigen detection (Quinn, 1994) are also promising approaches, but they are still experimental and prohibitively expensive. Developing an acceptable flowchart for standardizing the management of symptomatic individuals is only the first step in treatment. Experience in primary care settings routinely using flowcharts to manage other clinical conditions (e.g., acute respiratory infections in children) suggests that more operational research is needed to successfully introduce such strategies to a variety of providers in a range of clinical practice settings (Pan American Health Organization, 1983). At a minimum, the widespread use of standardized case management strategies will require a sizable investment in provider training and retraining. Efforts to upgrade diagnostic facilities and ensure adequate antibiotic supplies are also needed. One randomized community trial in Mwanza, Tanzania, has recently shown that improved STD case management had a significant impact on HIV incidence (Grosskurth et al., 1995). Primary health care clinics in six rural intervention communities received STD health education, staff training on the use of syndromic treatment algorithms, a regular supply of antibiotics, supervisory visits, and access to an STD reference clinic. The HIV incidence was reduced by more than 40 percent in this population after the integration of this package of STD services within the primary health care system. (In comparison with randomly matched communities, the prevalence of other STDs was also consistently lower in the intervention communities, but these results were not statistically significant.) In addition to using standardized case management to choose therapeutic

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions interventions, efforts are also needed to improve treatment adherence (compliance) and encourage sexual abstinence for the duration of therapy among those diagnosed with STDs. Clinics also have to allow sufficient time for providers to conduct appropriate counseling regarding the primary prevention of RTIs, condom use, and the importance of partner notification and treatment. Some public health researchers have suggested that the opportunity to provide such directed prevention advice may be as important as the treatment itself in reducing the spread of RTIs (O'Reilly and Islam, 1995). Screening for Asymptomatic Infections Interventions aimed at screening for asymptomatic infections (case finding) have historically been an important strategy for STD control. For example, serologic screening for syphilis infection is a standard component of routine antenatal care in many settings (Centers for Disease Control and Prevention, 1988b, 1993a). Providing prompt treatment for mothers has proven to be an extremely cost-effective strategy for preventing congenital syphilis, even in extremely resource poor settings (Hira, Bhat, and Chikamata, 1990; Stray-Pederson, 1983). Syphilis screening requires minimal diagnostic facility, is relatively inexpensive, requires therapy with antibiotics that could be readily available and affordable in most settings, and requires a modest level of client follow-up and treatment adherence. Therefore, the widespread implementation of screening for asymptomatic syphilis infections in antenatal clinics can be seen as a "sentinel" intervention. Yet although the necessary steps to ensure antenatal syphilis screening, maternal follow-up, and appropriate treatment are all relatively simple, they are rarely successfully coordinated in most developing country settings (Temmerman, Mohamedali, and Fransen, 1993). Successful establishment of antenatal syphilis screening could, therefore, serve as a benchmark of the programmatic capacity to design and effectively achieve more complicated interventions, such as case finding for other infections (Schulz, Schulte, and Berman, 1992). In considering screening for other asymptomatic infections, cervical infections in women (primarily caused by gonorrhea and chlamydia) are an important priority, given the high costs associated with PID and its complications. Diagnostic screening for these infections with currently available technology is neither simple nor inexpensive, however, so such screening must be rationed. A number of studies are available to suggest that screening for asymptomatic cervical gonococcal and chlamydial infections is cost-effective in family planning settings in industrialized countries (Trachtenberg, Washington, and Halldorson, 1988; Handsfield et al., 1986; Begley, McGill, and Smith, 1989). Selective screening is most cost-effective

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions when prevalence is generally low (i.e., 2-3%); in high prevalence settings (> 5%), universal screening is the more cost-effective approach, if affordable (Marrazzo et al., 1997). Hence, in selecting those for screening, sub-populations known to have a higher prevalence of infection are sought (Over and Piot, 1993; Aral and Peterman, 1993). Again, the relative cost-effectiveness of this screening may improve as simpler and less expensive diagnostic tests become available. Another potentially important method for identifying people with asymptomatic sexually transmitted infections is through partner notification and referral efforts (formerly known as contact tracing). Identification and treatment of infected partners could also help lower the risk of reinfection for women who have been treated for STDs. Partner notification may be either passive—infected people are expected to personally notify their partners concerning the possibility of infection and the need for treatment ("patient referral")—or active—clinic staff solicit names of sexual contacts and attempt to contact them ("provider referral"). The second approach is obviously more costly, but it is also more effective (Judson and Wolf, 1978). However, studies of the effectiveness of partner notification efforts in both developed and developing country settings have had mixed results (O'Reilly and Islam, 1995; Andrus et al., 1990; Winifield and Latif, 1985; Asuzu, Rotowa, and Ajayi, 1990). More research concerning the optimal design of partner notification efforts is needed to refine this set of interventions. Mass Treatment Approaches Treatment of an entire group of individuals at risk of infection (without diagnosing individual infections in the population) requires that the therapy administered be safe, highly effective, inexpensive, and associated with minimal side effects. A mass treatment approach requires that the intervention be acceptable to the community concerned and that the number of infections prevented be sufficient to justify the expense and any possible risks (World Health Organization, 1986). For many years, putting silver nitrate or antibacterial eyedrops in the eyes of newborn infants as a prophylaxis against ophthalmia neonatorum caused by gonorrhea has met these criteria (Brunham, Holmes, and Embree, 1990). This intervention ranks among the most cost-effective in terms of preventing serious morbidity (preventable blindness) at exceptionally low cost (Laga, Meheus, and Piot, 1989). Another mass treatment strategy involves the treatment of populations with high STD prevalence with antibiotics known to be effective against pathogens prevalent in those communities. To be effective at the population level, mass treatment interventions require high treatment

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions compliance and coverage and need to take into account local migration patterns and sexual networks. After considering these factors, STD mass treatment interventions have generally focused on treating either specific subpopulations known to have high STD rates (such as sex workers and migrant laborers) or entire communities with high STD rates. A "selective mass treatment" program among female sex workers (introduced as an addition to a long-established screening program) in the Philippines had a strong initial effect on the prevalence of gonorrhea, but this effect dissipated after a few months because of high rates of reinfection (Holmes et al., 1996). Recently, several authors have suggested that "epidemiologic" STD control may also be a worthwhile HIV intervention strategy (Wawer et al., 1995; Cates, Rothenberg, and Blount, 1996). The potential for bacterial and parasitic infections of the reproductive tract to augment the transmission of HIV (Wasserheit, 1992; Clottey and Dallabetta, 1993) makes this an attractive strategy in areas where HIV is highly endemic. One STD mass treatment trial is currently being conducted among communities in Rakai, Uganda. Twenty-six villages have been randomly selected to receive the intervention: mass treatment of all consenting individuals aged 15-59 every 9 to 10 months with antibiotics, in addition to an intensive health education and condom distribution campaign. In 26 control villages, the population receives the health education and condom distribution campaign, but not mass treatment for STDs; people will be referred for STD treatment based on their symptoms or positive serologic test results for syphilis (Wawer, 1995). The sustainability of this type of mass treatment intervention has been questioned, however, given the high costs of the treatment regimens involved (Science, 1995). The trial may also provide further information concerning the nature of STD/HIV synergy in both asymptomatic and symptomatic populations. When comparing a mass treatment approach with other STD treatment strategies, one should consider the estimated rates of reinfection and the feasibility of providing adequate clinic-based STD services within a population. In practice, a mass treatment approach may be best used as an initial, one-time intervention to lower overall STD prevalence, in conjunction with the establishment of adequate STD diagnostic and treatment services to sustain the reduction in STD prevalence over time. Tertiary Prevention The third set of interventions is tertiary prevention, minimizing the impact of complications of infection. The main components of tertiary prevention are clinical management of septic abortion, alarm and transport for ectopic pregnancies, the management of infertility, and cervical

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions cancer screening. As noted above, these complications are a major source of reproductive morbidity and mortality. For many of these conditions, the associated disability costs are fairly high, but so are the perceived costs of intervention. Consequently, the threshold to intervene is often quite high and a vicious cycle has developed. Because interventions have rarely been attempted, proven models for successful program development in resource-poor settings are lacking, and inexperience fuels programmatic complacency. Strengthening the clinical management of women who go to health facilities with complications of septic abortion is a priority area for intervention. Effective programs have the potential to avert many of the deaths that occur each year as a result of unsafe abortion (McLaurin, Hord, and Wolf, 1990). A number of programs for training providers in the use of appropriate technology, such as manual vacuum aspiration, have been developed and implemented. Of course, efforts to manage the complications of septic abortion must be accompanied by concerted efforts to prevent septic abortion in the first place. Strengthening postabortion contraceptive services is an important strategy. Another life-threatening complication of some RTIs is ectopic pregnancy. When this occurs, it typically presents as a medical emergency requiring urgent surgical intervention. Successful clinical management of this uncommon, but serious, condition will depend largely on the availability of appropriate diagnostic and transport systems to ensure safe maternity (see Chapter 4). This is particularly important for women in rural areas who must travel considerable distances to surgical facilities. The management of infertility and the development of screening programs to detect and manage cervical neoplasms are examples of other areas where interventions are needed to manage the complications of reproductive tract infection. Recently, the WHO Special Programme on Human Reproduction has developed a manual on "simplified infertility management" for developing countries (Rowe et al., 1993). A number of pilot projects for cervical cancer screening have also been recently started (Blumenthal et al., 1994; AVSC International and Program for Appropriate Technology in Health, 1994). To date, however, little data are available concerning cost-effective interventions to reduce the impact of these complications. RECOMMENDATIONS We present first our general recommendations for the promotion of infection-free reproductive health. Specific recommendations are then discussed in three areas: immediate priorities for existing reproductive health programs, including those programs initially established either

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions primarily or exclusively to provide family planning services; program gaps, highlighting the need to expand services to those groups currently not reached by any reproductive health services, especially men and young adults; and research priorities. As noted above, data concerning the cost and effectiveness of the many interventions for preventing and treating RTIs and their consequences are extremely limited. With finite resources, difficult choices will need to be made. These choices will be shaped by existing facility infrastructures, health personnel, research capacity, among other factors, and should be the outcome of considered debate in local communities. General Recommendations Policy makers and program managers should design locally relevant and culturally sensitive RTI prevention and treatment programs. The actual content of intervention programs and the challenges posed by their implementation should be accurately monitored and evaluated. The cost and effectiveness of the various intervention strategies to prevent RTIs should be evaluated and compared. Immediate Priorities Much can be done right now by family planning programs to respond to the concerns of their clients and staff about RTIs, STDs, and HIV/AIDS. All clinic staff need to be well-informed about HIV/AIDS so that they can answer their clients' basic questions. In this respect, it is important that staff learn to work through any fears they may have about AIDS or any judgmental attitudes they may harbor toward people with STDs. This will enable them to respond accurately and with sensitivity to those who may be infected or at risk of becoming infected. Clinic staff also should be aware of the symptoms of RTIs so that even if diagnosis and treatment are not available on site, their knowledge can be taken into account when considering the method of family planning most appropriate for each client. Family planning programs should also consider clients' risk of exposure to RTIs in determining protocols for providing various contraceptive methods. Barrier methods, particularly condoms, could be a better option for some clients despite being considered "less effective" contraceptive methods. For some clients, the secondary benefits of RTI prevention may be as important as the primary benefit of contraception. Programs should have well-designed informational materials dealing with RTIs, STDs, and AIDS available for staff and clients. For example, simple, pictorial instructions on how to correctly use

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions and dispose of condoms are essential. All clinicians cannot assume that people know how to use condoms. Many do not, and embarrassment about such ineptitude has been shown in some studies to be a major reason why some people at risk still do not use condoms (Richters, 1994). Program managers should encourage provision of basic RTI treatment services in clinics whenever possible. If services cannot be offered on site, providers should make the effort to learn which testing and treatment services for RTIs, STDs, and AIDS are available in their area and refer clients to these services as appropriate. Clinics and community-based distribution programs should have ample condom supplies for distribution to clients. Moreover, clients should be resupplied as quickly, efficiently, and as unobtrusively as possible. Program Gaps Health systems, particularly those in resource poor settings, often fail in their attempts to recognize or coordinate an effective response to RTI prevention or treatment. This failure occurs for a number of reasons, including weak management information systems, poor logistics and commodity distribution, inadequate provider training, and a narrow policy focus. Periodic disruptions of the supply of condoms and antibiotics are a particularly unfortunate example of this type of programmatic deficiency. Lapses in infection control practices are another common program failure. The overall management of reproductive health services should be strengthened as a means to ensure implementation of interventions known to be effective. The establishment and monitoring of infection prevention standards for clinical services could be seen as a "sentinel" intervention. Infection prevention, consisting of simple measures such as hand washing, appropriate use of gloves, and adequate sterilization of instruments should be a minimum standard for all service delivery. Monitoring services for lapses in infection prevention practices will allow program managers to identify gaps in provider training and motivation, interruptions in supply and logistics systems, and difficulties in coordination between different service elements. If basic infection prevention practices cannot be ensured, it is unlikely that other, more complicated interventions for RTI prevention or treatment will succeed. This overall monitoring may best be accomplished by an explicit attempt to build an alliance with service providers as a means of promoting a general climate of organizational development within service programs. The COPE (client-oriented, provider-efficient) method of self-assessment developed by AVSC International is an excellent example of such an approach (Dwyer and Jezowski, 1995).

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions One of the problems facing program managers in large service delivery systems is the lack of well-defined mechanism for "going to scale" with successful small-scale pilot projects. This is especially a problem in the government sector in developing countries. Larger scale demonstration projects that are developed with the input of all relevant community groups (including government, nongovernmental service organizations, advocacy groups) and attempt to implement a broad range of RTI interventions within the constraints of a typical community's budget would be helpful in gaining experience in this area. The participation of community decision makers is an essential element for the sustainability of these intervention efforts. Another urgent need is to expand the range of clients served by reproductive health programs. Adolescents, men, and the current users of traditional and informal sector services are examples of groups not reached through current programs. There is also a need to work with private sector services to improve the quality of their efforts to prevent and treat RTIs. Expanding services may require relaxing some of the current restrictions on nonphysician providers and changing some cultural assumptions regarding the presumed sexual abstinence of certain groups of women (i.e., adolescent, postpartum, or postmenopausal women). Correcting the current deficiency of trained female service providers through augmented training and support is also needed. Research Priorities Many unanswered questions concerning the optimal approach to achieving infection-free reproductive health would benefit considerably from both basic and applied research. Seven topics should be the main priority areas for future research: local characterization of RTI epidemiology and antibiotic sensitivity—and a low-cost methodology for making such assessments; the relationships between perceptions of reproductive morbidity, syndromic presentation, and biomedical definitions of infection (including documentation of the current patterns of help-seeking behavior and the perceptions of men); sexual behavior and factors influencing decision making concerning the use of prevention technologies, such as condoms and spermicides; development and validation of case management strategies, as well as operations research on all aspects of service organization and delivery; demonstration and pilot projects, focusing on models of integrated services;

OCR for page 40
Reproductive Health in Developing Countries: Expanding Dimensions, Building Solutions product development research on low-cost RTI diagnostics, therapeutics, vaccines, and woman-controlled prevention technologies, such as vaginal microbicides; and documentation of the interactions of RTIs and existing contraceptive technology.