HIV infection and breast-feeding: policy implications through a decision analysis model.

OBJECTIVES: (1) To develop a comprehensive decision analysis model to compare mortality associated with HIV transmission from breast-feeding with the mortality from not breast-feeding in different populations and (2) to perform sensitivity analyses to illustrate critical boundaries for guiding research and policy. METHODS: Using a decision tree, mortality rates were estimated for all children, children born to mothers infected during pregnancy, and children born to mothers who were uninfected at delivery. Given various assumptions about child mortality rates, relative risks of mortality among children who are not breast-fed compared with those who are (R), rates of HIV transmission from breast-feeding, HIV prevalence, and HIV incidence, scenarios were created and sensitivity analysis used to delineate critical boundaries. RESULTS: Our model shows that only in situations where R is approximately < or = 1.5 and HIV incidence/prevalence is high (prevalence > 10%, incidence > 5%) would universal breast-feeding result in equal or higher mortality compared with non-breast-feeding. Among populations in many developing countries, where there is a high relative risk of mortality if breast-feeding is not practiced, if R > 3, overall mortality is almost always lower among children who are breast-fed, even by HIV-infected mothers. In situations where maternal HIV status is known, the decision whether to breast-feed is largely dependent on the magnitude of additional mortality risk if the child is not breast-fed. The model illustrates the importance of distinguishing between population and individual recommendations. CONCLUSIONS: Based on available data, the model supports current World Health Organization and Centers for Disease Control recommendations on HIV infection and breast-feeding. Given the importance of breast-feeding and the global impact of HIV infection, more research is needed, especially to clarify the range of HIV transmission rates from breast-feeding and to expand specific assessments of relative risks for different areas of the world.

PIP: HIV/AIDS specialists have developed and applied 3 different scenarios to a comprehensive decision analysis model to estimate mortality rates for children of mothers infected with HIV during pregnancy and for children of mothers who were not infected with HIV during delivery. Scenario I represents Central Africa where HIV prevalence and incidence are high. Some scenario I assumptions are HIV prevalence in pregnant women of 30% and proportion of initially uninfected women who become infected after delivery during lactation (d) of 6%. Scenario II is a population where HIV epidemic is rather recent (e.g., some parts of Asia). Its assumptions are HIV prevalence of 5%, and s is 2%. Scenario III symbolizes high-risk populations in North America and Western Europe (HIV prevalence and s = 1%). The scenarios also consider child mortality rates and relative risks (RRs) of mortality of breast fed children and those who were not breast fed. Universal breast feeding would effect equal or higher mortality than non-breast feeding, when the RR of mortality is no more than 1.5 and HIV prevalence/incidence is high (high prevalence = 10% and high incidence = 5%). In developing countries, where the RR of mortality is high if children are not breast fed (RR 3), breast fed children have almost always lower child mortality than those who are not breast fed, regardless of HIV infection status. The decision to breast feed when the HIV status is known depends greatly on the degree of an additional mortality risk if an infant is not breast fed. The model substantiates WHO and CDC recommendations: HIV-positive women in the UK and the US should not breast feed, while those in developing countries with high RR of child mortality should breast feed. Additional research would define the range of HIV transmission rates from breast feeding and increase specific assessments of RRs for various parts of the world.

Phase IV study of the injection Norigest in Pakistan.

A field study of the injectable contraceptive, norethisterone enanthate (NET-EN), was conducted in family planning clinics in Sind and Punjab provinces of Pakistan, to determine the acceptability and feasibility of providing NET-EN in government family planning clinics staffed by Family Welfare Visitors (FWVs). A total of 2147 women were recruited to the study, of whom approximately three-fourths had never previously used contraception. The overall discontinuation rate at one year was 78 per 100 women; the most common reason for discontinuation was bleeding disturbances, including amenorrhea, although returning to the clinic too late for an injection also accounted for a substantial proportion of the discontinuations. Given adequate training, FWVs were shown to be capable of providing NET-EN in family planning clinics, including managing the bleeding disturbances common with this method of contraception. No pregnancies were reported, demonstrating that the method is highly effective when used in a usual family planning clinic situation.

Introduction of the injectable contraceptive NET-EN into family planning clinics in Bangladesh.

PIP: In preparation for introducing the injectable contraceptive norethisterone enanthate (NET-EN) into the national family planning program, a field study was conducted in 6 family planning clinics in Bangladesh. 3 of the clinics were located in Dhaka, the capital city, 1 was located in a town approximately 15 kilometers from Dhaka, and the other 2 clinics were located several hundred kilometers from Dhaka. A total of 913 women were chosen to receive injections of NET-EN every 8 weeks for 6 months and every 12 weeks thereafter. The mean age of the subjects was 26.8 years, and the mean number of live births was 3.4. The overall cumulative discontinuation rates were 26.3/100 women at 6 months, 37.3/100 at 12 months, and 42.9/100 at 18 months. The most common reason for discontinuing was a disturbance in bleeding. Heavy and/or prolonged bleeding was the single most frequent reason (6.3/100 women at 12 months), but amenorrhea was not uncommon (5.1/100 women at 12 months), while irregular bleeding or spotting was given as a reason somewhat less frequently (3.9/100 women). 3 women became pregnant during the study, giving a cumulative pregnancy rate of 0.4/100 women at 18 months. A variety of complaints fell within the category of "other medical reasons," the most common being fatigue, headache, and dizziness. The overall discontinuation rates varied markedly by center, as did the reasons for discontinuation. The lowest discontinuation rate of 14.7/100 women at 12 months was seen in the Mohammedpur Fertility Services and Training Center. This rate was considerably lower than that in any other center. The highest discontinuation rate was found by the Bangladesh Association for Voluntary Sterilization -- 52.0/100 women at 12 months. The variation in rate of discontinuation because of bleeding disturbances was most dramatic, with relatively high rates at the Bangladesh Association for Voluntary Sterilization, markedly low rates in the Mohammedpur Fertility Services and Training Centre, and intermediate rates elsewhere. Neither the overall discontinuation rate, nor the individual reasons for discontinuation varied much by age group. The most notable finding regarding discontinuation of NET-EN was the marked difference between centers. This is particularly notable given the lack of any major differences with respect to age, parity, residence, and history of contraception of the subjects. These findings suggest that the specific approach used in each clinic had a considerable impact on continuation rates and acceptance of NET-EN.^ieng

Alcohol consumption among Mexican American and Anglo women: results of a survey along the U.S.--Mexico border.

A household probability survey of 1233 Mexican American women and 798 Anglo women residing along the U.S.-Mexico border was conducted. A higher proportion of abstainers was found among the Mexican Americans than among the Anglos in almost every social and demographic category examined (age, marital status, education and employment status). Because the level of alcohol consumption increased markedly with the years of education completed, almost all of the overall ethnic differences observed could be accounted for by the generally lower level of education among the Mexican Americans. However, ethnic subgroups of Mexican American women reported different levels of alcohol consumption that could not be accounted for by differences in education, suggesting that additional ethnic factors contribute to drinking patterns.

Lung cancer mortality and smoking habits: Mexican-American women.

Lung cancer mortality was reported to be higher among Mexican-American women as compared with Anglo women from 1950 until 1970; however, smoking habits of Mexican-American women have not been adequately described. This study updates lung cancer mortality data in Texas, describes smoking patterns of 1,255 Mexican-American women from a household survey in the four states bordering Mexico, and compares these findings to a reference group of Anglo women residing in the same area. In 1970, lung cancer mortality rates were similar for Mexican-American and Anglo women in Texas; however, by 1974-1976 Mexican-American women in Texas had a 40 per cent lower rate than Anglo women and by 1979, a 49 per cent lower rate. In our 1979 survey results, Mexican-American women reported lower levels of smoking, both in prevalence and amount smoked, as compared with Anglo women. The lower prevalence was reported for all social and demographic categories examined. The relatively low lung cancer mortality rate is most likely due to relatively low levels of cigarette smoking among Mexican-American women. Based on the trend in lung cancer deaths and our survey findings, we would anticipate a continuing low level of lung cancer mortality among Mexican-American women.

Family planning practices among Anglo and Hispanic women in U.S. counties bordering Mexico.

PIP: The US Center for Disease Control surveyed women of reproductive age living on the US side of the Mexican border as to contraceptive use and source of care. It found that overall contraceptive practice is quite similar among married Hispanic and Anglo (white, non-Hispanic) women: 75% of Anglo and 66% of Hispanics use some method. Among never marrieds however, Anglos are twice as likely as Hispanics to use a method. About 22% of Anglo and 12% of Hispanic women are protected from pregnancy by contraceptive sterilization of themselves or partners, the difference almost entirely attributable to a very low incidence of vasectomy among Hispanic males. The pill is the most popular reversible method among both groups. Hispanic women are more likely to go to Planned Parenthood or health department clinics, Anglo women to go to private physicians or clinics. Unmet need is much higher among Hispanics: about 4 times the proportion of married Hispanic women as comparable Anglo women were at risk of unintended pregnancy and were using no method.^ieng