Research on patient compliance in developing countries.

Patient compliance with health care provider advice is a subject of particular importance in developing countries, one that has not been sufficiently studied. This article begins by explaining why this type of research is important and describing the various methodologies used to examine patient compliance in developed nations--notably self-reporting by the patient, collateral reporting by friends or relatives, pill or bottle counting, review of the patient's clinic attendance, review of clinical outcomes, and use of biochemical tracers. The applicability of each of these methods to scenarios commonly found in developing countries is then considered, and the results of the limited compliance research performed in developing countries are described. In general, it is suggested that self-reporting, collateral reporting, and pill or bottle counting can be employed effectively in many Third World situations. The article also notes the importance of standardizing research procedures, provides logistic advice about applying various compliance research methods in Third World conditions, and points up the need to validate the accuracy of the methods used.

PIP: Research is needed on patient compliance with health care provider advice in developing countries because 1) the consumption of pharmaceuticals is on the rise in developing countries due to the expansion of health care services and successful marketing of drugs by multinational pharmaceutical companies, 2) pharmaceuticals consume a large share of developing countries' limited health care budgets, 3) patient compliance plays an important role in improving patient health and enabling health care systems to be effective, 4) poor medical outcomes resulting from noncompliance may prompt dissatisfaction with the health care delivery system and cause deterioration of the patient-physician relationship, and 5) lack of compliance with pharmaceutical regimens leads to the storage of leftover medicine in homes, thus increasing the risk of accidental poisoning and misuse of pharmaceuticals. The authors explain how self-reporting by the patient, collateral reporting by friends or relatives, pill or bottle counting, review of the patient's clinic attendance, review of clinical outcomes, and use of biochemical tracers are used in developed countries to assess patient compliance. The applicability of each of these methods to scenarios commonly found in developing countries is then considered and the results of the limited compliance research performed in developing countries described. It is generally suggested that self-reporting, collateral reporting, and pill or bottle counting can be used effectively in many Third World situations. The authors stress the importance of standardizing research procedures, provide logistic advice about applying various compliance research methods in the Third World, and note the need to validate the accuracy of the methods used.

A multicenter comparative trial of triphasic and monophasic, low-dose combined oral contraceptives.

A comparative multicenter clinical trial of two combined oral contraceptives (OCs) was conducted at clinics located in the Sudan, Sri Lanka, Chile, the Dominican Republic and Ecuador. The trial was designed to determine if there were differences in efficacy, safety and acceptability between a triphasic and a low-dose monophasic OC. This report includes analysis of 1088 women. At each center, subjects were randomly allocated to one of the two OCs. Follow-up visits were scheduled at 1, 4, 8 and 12 months after admission. There were two accidental pregnancies attributed to user failure reported during the study period; one in the triphasic group and one in the monophasic group. Adverse experiences were mainly minor with headaches and dizziness being the most common complaints; frequency of reports was similar in both groups. Cycle control was good in both groups with women in the triphasic group reporting fewer complaints of intermenstrual bleeding. Both OCs were safe and effective.

PIP: Researchers compared the efficacy, safety, and acceptability of a triphasic oral contraceptive (OC), Triquilar, with those of a monophasic OC, Lo-Femenal, among 1088 women attending clinics in Chile, the Dominican Republic, Ecuador, Sri Lanka, and the Sudan. Both OCs contained levonorgestrel and ethinyl estradiol. 90% of women in each group exhibited good user compliance. Only 1 unplanned pregnancy occurred in each group, and both pregnancies were attributed to user failure. The gross cumulative efficacy rates at 11 months were 0.3/100 woman-years for the triphasic OC and 0.2/100 woman-years for the monophasic OC. The continuation rate at 11 months was lower for Lo-Femenal than it was for Triquilar (80.8% vs. 84.6%), but the difference was not significant. The leading side-effect-related reason for discontinuation in both groups was headache. Another key reason for OC discontinuation in both groups was personal reasons, such as planning a pregnancy. Most women in both groups did not have menstrual complaints (78.8% for the Triquilar group and 77.1% for the Lo-Femenal group). Intermenstrual bleeding rates were low (7.6% for the Triquilar group and 9% for the Lo-Femenal group). Significant intercenter differences for women reporting intermenstrual bleeding and side effects (e.g., headaches) existed (p .05). Women from both groups at the clinic in the Sudan always had lower reports of intermenstrual bleeding than those at the other clinics. In fact, no woman discontinued OC use because of intermenstrual bleeding in the Sudan. Women in Sri Lanka reported fewer side effects than those in other countries, suggesting they could better tolerate OCs than the other women. This multicenter study's findings indicates that both Triquilar and Lo-Femenal are effective and safe. In addition, they exhibit good cycle control.

Tuberculosis, poverty, and "compliance": lessons from rural Haiti.

Tuberculosis (TB) is the leading cause of death among rural Haitian adults, and TB control in Haiti is widely acknowledged to be a failure. The causes of both the endemicity of TB and the failure of attempts to address it are briefly reviewed before data from a study conducted in rural, central Haiti are presented. Members of one group of patients with active TB were given free medical care; members of a second group were given free care as well as financial aid, incentives to attend a monthly clinic, and aggressive home follow-up by trained village health workers. Comparing the two groups shows significant differences in mortality, sputum positivity after 6 months of treatment, persistent pulmonary symptoms after 1 year of treatment, average amount of weight gained, ability to return to work, and cure rate. The roles of human immunodeficiency virus and cultural factors are also examined. When adequate nutrition and access to free care were assured, drug-dependent and patient-dependent factors were shown to be of secondary importance in determining treatment outcome. Based on these data from a small, community-based TB-control project, the authors conclude that high cure rates can be achieved if the primacy of economic causes of TB is acknowledged and addressed.

PIP: Between February 1989 and June 1990, village health workers with the Proje Veye Sante community health program in the central plateau of Haiti enrolled 30 adults with tuberculosis (TB) living around the reservoir in the Peligre basin (sector 1) and 30 TB patients living in villages and towns next to sector 1 (sector 2) to compare 2 TB treatment approaches. The cases (sector 1 patients) received free medical care, home visits, US$30/month for the 1st 3 months (financial aid), nutritional supplements, monthly reminders to visit the clinic, and travel expenses. The controls (sector 2 patients) received only free medical care. 1 case (3.3%) and 2 controls (6.7%) tested positive for HIV. By June 1991, all 30 cases were cured of TB compared with only 56.7% of controls. None of the cases died but 10% of controls died. None of the cases exhibited sputum positivity for acid fast bacilli 6 months after diagnosis yet 13.3% of controls did. 1 year after treatment only 2 cases (6.7%) still had pulmonary symptoms compared with 13 (43.3%) controls. Cases gained more weight on average than controls (10.4 lbs. vs. 1.7 lbs). All but 2 cases (93.3%) were able to return to work after 1 year of treatment while only 14 controls (46.7%) could. Cases made more trips to the clinic and experienced more home visits than the controls (11.4 vs. 5.8 and 37.9 vs. 1.1, respectively). 25 cases (83.3%) and 26 controls (86.7%) did not deny that sorcery may have been responsible for their illness. The results demonstrated that high cure rates can occur under extremely impoverished conditions where hospitals do not exist. The top priority under these conditions should be identification and treatment of patients with active pulmonary TB. TB programs should address nutrition and providing TB patients easy access to drugs. Direct financial aid provides an incentive for TB patients to follow through with treatment which the free treatment alone does not do.

Maternal mortality at Mount Hope Women's Hospital, Trinidad.

A ten-year survey of the magnitude and causes of obstetrical deaths at Mount Hope revealed a maternal mortality rate of 33.3 per 100,000 live births. The leading causes of death were the hypertensive disorders, and the most common identifiable factors were inadequate antenatal care and substandard clinical management.

PIP: Obstetrical deaths at the Mount Hope Women's Hospital, Trinidad, were reexamined from records over 1981-1990, and discussed under the categories poor prenatal care, clinical management or provision of medical facilities. There were 19 obstetrical deaths out of 57,012 live births, giving a maternal mortality rate of 33.3/100,000 in this tertiary care hospital. Most of the deaths occurred in women aged 30-34, para 5 or more. 73.7% were related to hypertension in pregnancy, 8 with severe eclampsia and 6 with eclampsia, and the other 5 were due to placental abruption, postpartum hemorrhage, anesthesia complication, acute fatty liver and amniotic fluid embolism. Cases classified as substandard care included the 14 women with hypertensive disorders, none of whom had antenatal care at this specialized unit. In 3 referral by the practitioner was delayed, and 3 others did not comply. A woman listed under failed medical facilities had massive abruptio placenta, and no fresh blood was available, and another had an anaphylactic reaction to a mismatched blood transfusion. Other avoidable deaths were 3 associated with general anesthesia: one woman having emergency cesarean section for severe pre-eclampsia had anoxia and severe brain damage; another short, obese woman had cardiac arrest during a failed attempt at endotracheal intubation; a third died from aspiration of gastric fluid. The high mortality among women with hypertensive disorders is regrettable, considering lack of referral to this specialized unit, but the prognosis of eclamptics even with expert aggressive treatment is poor. This maternal mortality rate ranks midway between those of developed countries and developing countries. It is about four times that of the U.S.