Search

Objectives: To determine the age specific prevalence of impairment and disability relating to hemiplegic stroke in one rural area of Tanzania., Methods: During the yearly house to house census of the study population of 148 135 (85 152 aged 15 and over) in August 1994, specific questions were asked to identify those who might be disabled from stroke. People thus identified were subsequently interviewed and examined by one investigator. In those in whom the clinical diagnosis of stroke was confirmed a more detailed interview and examination relating to risk factors and recovery was carried out., Results: One hundred and eight patients, 61 men and 47 women, were identified with a median age of 70 (range 18-100). Median age at first stroke was 65 years. The age specific rates in this study were lower than previous studies in developed countries. All were cared for at home although 23 (21%) were bedbound., Conclusions: Although prevalence of impairment and disability related to stroke in this population as a whole was low this is mainly explained by the age structure, with less than 6% being aged 65 and over. Age standardised rates for stroke with residual disability were about half those found in previous studies in developed countries. Death from stroke in Africa may be higher but data are limited. With the demographic transition stroke is likely to become a more important cause of disability in sub-Saharan Africa.

Background: Most data for stroke mortality in sub-Saharan Africa are hospital based. We aimed to establish the contribution of cerebrovascular disease to all-cause mortality and cerebrovascular disease mortality rates in adults aged 15 years or more in one urban and two rural areas of Tanzania., Methods: Regular censuses of the three surveillance populations consisting of 307,820 people (125,932 aged below 15 years and 181,888 aged 15 or more) were undertaken with prospective monitoring of all deaths arising in these populations between June 1, 1992 and May 31, 1995. Verbal autopsies were completed with relatives or carers of the deceased to assess, when possible, the cause of death., Findings: During the 3-year observation period 11,975 deaths were recorded in the three surveillance areas, of which 7629 (64%) were in adults aged 15 years or more (4088 [54%] of these in men and 3541 [46%] in women). In the adults, 421 (5.5%) of the deaths were attributed to cerebrovascular disease, 225 (53%) of these in men and 196 (47%) in women. The yearly age-adjusted rates per 100,000 in the 15-64 year age group for the three project areas (urban, fairly prosperous rural, and poor rural, respectively) were 65 (95% CI 39-90), 44 (31-56), and 35 (22-48) for men, and 88 (48-128), 33 (22-43), and 27 (16-38) for women, as compared with the England and Wales (1993) rates of 10.8 (10.0-11.6) for men and 8.6 (7.9-9.3) for women., Interpretation: We postulate that the high rates in Tanzania were due to untreated hypertension. Our study assessed mortality over a single time period and therefore it is not possible to comment on trends with time. However, ageing of the population is likely to lead to a very large increase in mortality from stroke in the future.

Insulin sensitivity was assessed using a glucose-insulin infusion test in 15 newly diagnosed non-obese hypertensive black Tanzanians with normal glucose tolerance and in 15 normotensive control subjects matched for age, sex, and body mass index. The steady-state blood glucose and metabolic clearance rate of glucose (MCR) were used as measures of insulin sensitivity. The mean MCR (glucose) was significantly reduced (7.12+/-0.57 versus 9.50+/-0.69 micromol/kg per minute; P<.05) and mean steady-state blood glucose was significantly elevated (5.0+/-0.3 versus 3.7+/-0.3 mmol/L; P<.01) in subjects with hypertension compared with the normotensive group. For all subjects there was a significant inverse correlation between MCR (glucose) and systolic (P=.003) and diastolic (P=.005) blood pressure; and a positive correlation was found between fasting serum insulin levels and systolic (P=.005) and diastolic (P=.004) blood pressure. These observations were independent of body mass index and serum lipid levels. These data indicate a strong association between insulin mediated glucose uptake and blood pressure in this population of normal weight untreated urban Africans.

Objective: To define the potential role of proinsulin-like molecules as risk factors for cardiovascular disease., Research Design and Methods: Fasting concentrations of proinsulin, des-31,32-proinsulin, and insulin, and of insulin 2 h after a 75-g glucose load, were measured in 1,034 nondiabetic europid subjects and 257 south Asian subjects and related to prevalent coronary heart disease (Minnesota-coded electrocardiographic criteria or ischemic chest pain). In 137 south Asian subjects, the fasting concentrations were related to incident coronary heart disease over a 6.5-year follow-up., Results: The standardized odds ratios for prevalent coronary heart disease were as follows: fasting insulin, 1.29 (1.11-1.49), P = 0.0006; 2-h insulin, 1.25 (1.08-1.45), P = 0.003; proinsulin, 1.23 (0.99-1.53), P = 0.058; and des-31,32-proinsulin, 1.32 (1.03-1.69), P = 0.026. The odds ratios were similar in the two ethnic groups. These relationships became insignificant when controlling for age, sex, and BMI. The standardized odds ratios for incident coronary heart disease were as follows: fasting insulin, 0.99 (0.63-1.55), P = 0.97; proinsulin, 1.13 (0.72-1.78), P = 0.59; and des-31,32-proinsulin, 1.00 (0.61-1.63), P = 1.00., Conclusions: We have found similar relationships between concentrations of proinsulin-like molecules and prevalent coronary heart disease, as are observed for insulin in these nondiabetic subjects, although these molecules comprise only approximately 10% of all insulin-like molecules. It appears biologically implausible that these relationships represent cause and effect.

The aim of this study was to determine the contribution of diabetes mellitus to all-cause mortality and diabetes mortality rates in adults 15 years and above living in one urban and two rural areas of Tanzania (Dar es Salaam, Hai and Morogoro Rural Districts). The three surveillance populations comprised 307,912 persons. Prospective monitoring of all deaths between 1 June 1992 and 31 May 1995 was carried out. Cause of death was determined by verbal 'autopsy' conducted with relatives of the deceased. In total, 4299 deaths were recorded in children (aged < 15 years) and 8054 in adults. In children there were no reported deaths associated with diabetes (due to or in children with diabetes). The adult male mortality rates associated with diabetes were 34, 30, and 15 per 100,000 per year in Dar es Salaam, Hai and Morogoro Rural Districts respectively. The figures in women were 21, 18, and 4 per 100,000 per year, respectively. The percentages of all adult male deaths associated with diabetes were 2.6%, 2.1% and 0.7% respectively. In women the percentages were 1.7%, 1.8%, and 0.2% respectively. Acute metabolic complications, infection, and stroke each accounted for approximately 30% of all diabetic deaths. Thus diabetes mortality rates varied between the three surveillance areas, being lowest in the poorest rural area. Rates were higher in men in all areas. While care is required in the comparison of mortality rates between countries, it was noteworthy that Tanzania, a country with a low diabetes prevalence, had diabetes mortality rates which were higher than or comparable to rates in Mauritius and the United States. Most patients died from preventable causes, indicating a need for improved case-management of diabetic emergencies as well as better detection and treatment of hypertension.

Following the liberalisation of medical practice in Tanzania since the early 1990's, and the introduction of user fees in public hospitals in 1993, a household survey evaluated utilisation of health care in Dar es Salaam. A sample of 6,589 inhabitants was interviewed in April 1995 by means of a two-stage cluster sampling technique. Of the respondents, 32% reported some use of health care within the previous two weeks. Among these respondents, 35% had used government health services, 41% had used private services and self-treatment was chosen by 27%. The user patterns identified reveal that adults aged 15-49 years used government health service least often. Use of government services clearly decreased as the level of education, socioeconomic class and wealth status of the zone of residence of the ill person increased. Conversely in the study sample, there was an apparent tendency for people with a high level of education or belonging to a rich socio-economic class to use private facilities more often. The data also indicate that already after two years the private sector plays an important role in providing medical care and that a two-tier system of health care delivery is developing. In order to render the private sector complementary to public services, there is need for a coherent policy on legislation, development, regulation and control of private sector health services as well as a monitoring system to reinforce the policies.

Objective: To measure age and sex specific mortality in adults (15-59 years) in one urban and two rural areas of Tanzania., Design: Reporting of all deaths occurring between 1 June 1992 and 31 May 1995., Setting: Eight branches in Dar es Salaam (Tanzania's largest city), 59 villages in Morogoro rural district (a poor rural area), and 47 villages in Hai district (a more prosperous rural area)., Subjects: 40,304 adults in Dar es Salaam, 69,964 in Hai, 50,465 in Morogoro rural., Main Outcome Measures: Mortality and probability of death between 15 and 59 years of age (45Q15)., Results: During the three year observation period a total of 4929 deaths were recorded in adults aged 15-59 years in all areas. Crude mortalities ranged from 6.1/1000/year for women in Hai to 15.9/1000/year for men in Morogoro rural. Age specific mortalities were up to 43 times higher than rates in England and Wales. Rates were higher in men at all ages in the two rural areas except in the age group 25 to 29 years in Hai and 20 to 34 years in Morogoro rural. In Dar es Salaam rates in men were higher only in the 40 to 59 year age group. The probability of death before age 60 of a 15 year old man (45Q15) was 47% in Dar es Salaam, 37% in Hai, and 58% in Morogoro; for women these figures were 45%, 26%, and 48%, respectively. (The average 45Q15s for men and women in established market economies are 15% and 7%, respectively.), Conclusion: Survivors of childhood in Tanzania continue to show high rates of mortality throughout adult life. As the health of adults is essential for the wellbeing of young and old there is an urgent need to develop policies that deal with the causes of adult mortality.

Little is known of the natural history of blood pressure (BP) levels in diabetic patients from sub-Saharan Africa. BP levels were therefore recorded in such patients in Dar es Salaam, Tanzania, over 2, 5, and 7 years. Hypertension was found in 5% of insulin-treated diabetes mellitus (IDDM) and 29.2% of non-insulin-dependent diabetes mellitus (NIDDM) patients at presentation with diabetes. Hypertension developed in a further 2 IDDM (3.7%) and 27 NIDDM (15.6%) patients at 2 years, and in 3 IDDM (13.0%) and 9 NIDDM (9.8%) patients at 5 years. Seven NIDDM (18.4%) patients had developed hypertension by 7 years. In NIDDM patients with normal BP initially, the mean systolic BP rose from 131 to 141 mmHg (P < 0.001) 2 years later (n = 146); from 131 to 138 mmHg (P < 0.001) for those followed for 5 years (n = 82); and from 131 to 138 mmHg (P < 0.05) for those followed for 7 years (n = 31). The mean diastolic BP was 83 mmHg initially and 84 mmHg (NS) for those followed for 2 years (n = 146). There was no observed rise in mean diastolic BP at 5 or 7 years of follow-up. In IDDM patients without hypertension, only the systolic BP rose significantly by 5 years, from 124 to 132 mmHg (P < 0.001; n = 20). These changes were independent of age, sex, body mass index, and proteinuria.(ABSTRACT TRUNCATED AT 250 WORDS)

Previous studies have suggested an association between polymorphisms in the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Most of the studies so far have been performed in Caucasoid populations. We have investigated 418 random IDDM patients and 422 healthy control subjects from three different ethnic groups; Tanzanian blacks, Norwegian Caucasians and Japanese orientals. Our data suggest that polymorphisms in the insulin gene region confer susceptibility to IDDM in Caucasians, and that a similar tendency though not statistically significant is observed among Tanzanian blacks, while no significant contribution is seen among Japanese orientals. We further demonstrate that the disease-associated genotype INS +/+ confers susceptibility independently of HLA class II alleles associated with IDDM. Compared to the contribution of particular HLA-DQ alleles in IDDM susceptibility, the additional risk conferred by the insulin gene region polymorphism is, however, small. Genotyping of the insulin gene region will therefore most probably not be a useful tool in the prediction of IDDM.

In 1987, as part of the World Health Organization's Inter-Health Programme, we carried out a noncommunicable diseases survey in six rural villages in Tanzania. Each women completed a questionnaire, part of which was concerned with obstetric history, and underwent a physical examination. Our findings of unacceptably high perinatal mortality and abortion rates are described below.

A study of the prevalence of schistosomiasis was carried out in 253 school children in Melela, Tanzania, one year after a single dose of praziquantel, 40 mg/kg body weight. The cure rate was 90.4%. However the new incidence estimate was 21.2% in children who were initially negative. The use of reagent strips in urine tests for the detection of urinary schistosomiasis was also evaluated. The presence of blood in the urine was highly sensitive (> 96%) and specific (87%). The sensitivity of proteinuria was less, but it was highly specific (94%). This study shows that chemotherapy will have to be combined with other measures to achieve lasting benefits and raises the question as to how often the population should be treated. It also confirms the value of urine test strips as an indirect diagnostic test for urinary schistosomiasis in an endemic area. Further follow-up is necessary to make useful predictions concerning incidence and reinfection rates in the community. This will also help in deciding how often chemotherapy should be given to the population.

Study Objective: To assess the level of cardiovascular risk factors in young people in sub-Saharan Africa living in rural and urban settings., Design: Cross sectional survey of the population aged 15 to 19 years., Setting: Eight rural Tanzanian villages in three regions, and two districts in Dar es Salaam., Participants: 664 males and 803 females in rural villages and 85 males and 121 females in the city. Response rates for total population were 74% to 94% in the rural areas and 60% in the city., Measurements and Results: Measurements included blood pressure, body mass index, serum lipids, and blood glucose concentrations (fasting and two hours after 75 g glucose). Blood pressure was slightly but significantly higher in young women than in young men (115/67 mmHg versus 113/65 mmHg) and increased significantly with age. Only 0.4% subjects had blood pressure greater than 140 and/or 90 mmHg. There were no urban-rural differences. Body mass index was higher in females (mean (SD) 20.3 (2.8) kg/m2) than males (18.5 (2.1)). Overweight was found in only 0.6% at age 15 years but 5.4% at age 19 years. Serum cholesterol concentrations were low at 3.5 mmol/l in males and 3.7 mmol/l in females. Only 7% had values above 5.2 mmol/l. The highest concentrations were found in the city and in Kilimanjaro, the most prosperous rural region. Serum triglycerides were 1.0 (0.5) mmol/l in males and 1.1 (0.5) mmol/l in females, and were highest in the city dwellers. Diabetes was rare (0.28% males, 0.12% females) but impaired glucose tolerance was present in 4.7% and 4.1% respectively. Drinking alcohol was equally prevalent in males and females, reaching 30% at age 19 years. Only 0.4% of females smoked compared with 7.3% of males. Smoking was commoner in rural areas that in the city., Conclusions: Several risk factors for cardiovascular disease were found in Tanzanian adolescents, but levels were much lower than in studies reported from developed nations. The challenge is to maintain these low levels as the population becomes more urbanised and more affluent.

A community-based survey was used to assess the prevalence of risk factors for coronary heart disease (CHD) in rural Tanzanians. In all, 8581 subjects (3705 men, 4876 women) aged > or = 15 years in eight villages in three regions in rural Tanzania representing a range of socioeconomic deprivation were studied. The main outcome measures were serum cholesterol and triglyceride level, blood pressure and prevalence of dyslipidaemia, hypertension, smoking, overweight, impaired glucose tolerance (IGT) and diabetes; as well as ECG changes. Mean serum cholesterol levels in men were 4.2, 3.4 and 3.7 mmol/l, and in women 4.4 3.6 and 3.9 mmol/l in Kilimanjaro, Morogoro and Mara regions respectively. In Kilimanjaro region 17.4% of men and 19.0% of women had values above 5.2 mmol/l compared with only 5.0% and 6.7% in Morogoro region and 4.8% and 6.9% respectively in Mara region. Systolic and diastolic blood pressures increased with age in both men and women in all three regions with the most marked increase in Kilimanjaro region and the smallest rise in Mara region. Mean age-adjusted values were highest in Kilimanjaro region (124/75 mm Hg, and 125/76 mm Hg in men and women respectively) and lowest in Mara region (120/70 mm Hg in men and 118/68 mm Hg in women). Hypertension was found in 6.6% of men and 7.5% of women in Kilimanjaro region, 3.3% and 4.7% in Morogoro, and 2.6% and 3.4% in Mara region. Cigarette smoking was found in 42.6% of men in Kilimanjaro region, 28.2% of Morogoro region and 8.6% in Mara region. Less than 4% of women smoked in all three regions.(ABSTRACT TRUNCATED AT 250 WORDS)

Objective: To ascertain the annual incidence of diabetes requiring treatment with insulin in children and adolescents aged 0-19 years in Dar es Salaam, Tanzania, during a 10 year period from 1 January 1982 to 31 December 1991., Design: Prospective registration at a major urban hospital of all patients with newly diagnosed diabetes who were resident in Dar es Salaam., Setting: Muhimbili Medical Centre, Dar es Salaam, Tanzania., Patients: 86 patients: 45 male, 41 female., Results: The annual incidence of juvenile diabetes for both sexes was 1.5 per 100,000 population aged 0-19 years (95% confidence interval 1.3 to 1.7). Incidence per 100,000 population per year increased with age: 0.6 (0.0 to 0.13) in the age group 0-4 years, 0.5 (0.3 to 0.7) at 5-9 years, 2.2 (1.8 to 2.6) at 10-14 years, and 3.4 (2.9 to 3.9) at 15-19 years., Conclusion: Juvenile diabetes mellitus is fairly rare in sub-Saharan Africa. If environmental factors such as infection and material deprivation were important determinants of insulin dependent diabetes in Africans, as they may be in Europeans, much higher rates would have been expected unless genetic factors possibly exert a protective role. The eightfold greater incidence in African Americans than in Tanzanians may be related to greater genetic admixture in African Americans with people from countries in Europe with a high incidence.

Objective: To investigate the relation between undernutrition and diabetes., Design: Survey of glucose tolerance in rural Tanzania., Setting: Eight villages in three widely separated regions of Tanzania., Subjects: 8581 people aged 15 and above: 3705 men and 4876 women., Main Outcome Measures: Oral glucose tolerance, body mass index, height, and low haemoglobin and cholesterol concentrations., Results: In the eight villages 42.7-56.9% of all men and 30.0-45.2% of all women had a body mass index below 20 kg/m2; the lowest quintile was 18.2 kg/m2 in men and 18.6 kg/m2 in women. The prevalence of diabetes did not change significantly from the lowest to the highest fifths of body mass index in men (lowest 1.6% (95% confidence interval 0.8% to 2.9%) v highest 1.3% (0.7% to 2.5%)) or women (1.1% (0.6% to 2.1%) v 0.5% (0.2% to 1.2%)). In men and in women prevalence of impaired glucose tolerance was greater in the lowest fifths of height (8.2% (6.3% to 10.6%), and 11.1% (9.2% to 13.3%)) respectively and body mass index (9.6% (7.5% to 12.1%), and 8.4% (6.7% to 10.5%)) than in the highest fifths (impaired glucose tolerance 4.7% (3.4% to 6.5%); and 5.1% (3.9% to 6.7%); body mass index 5.1% (3.7% to 7.0%), and 7.7% (6.2% to 9.6%)., Conclusion: Rates of diabetes were not significantly associated with low body mass index or height, but overall rates were much lower than those in well nourished Western populations. Increased impaired glucose tolerance in the most malnourished people may reflect the larger glucose load per kilogram weight. The role of undernutrition in the aetiology of diabetes must be questioned.

Objective: To test the hypothesis that consumption of cassava with liberation of cyanide causes diabetes in malnourished individuals., Research Design and Methods: Glucose tolerance was assessed in two rural communities in Tanzania; in one (Nyambori), the main source of calories was cassava; and in the other (Uswaa), cassava was rarely eaten. Undernutrition was prevalent in both communities. The people of Nyambori were known to have high dietary cyanide exposure for many years from consumption of insufficiently processed cassava. Of the 1435 people in Nyambori greater than or equal to 15 yr old, 1067 (74%) were surveyed, and 1429 of 1472 (97%) eligible subjects in Uswaa were surveyed. All had 75-g oral glucose tolerance tests and measurement of BMI. Plasma and urine thiocyanate and blood cyanide also were measured in some subjects., Results: Mean +/- SD plasma and urine thiocyanate levels in Nyambori were 296 +/- 190 and 497 +/- 457 microM (n = 204), respectively, compared with 30 +/- 37 and 9 +/- 13 microM, respectively, in Uswaa (n = 92) (P less than 0.001 for all differences). The mean blood cyanide level in Nyambori was elevated (1.4 [range 0.1-30.2] microM; n = 91). The prevalence of diabetes in the cassava village (Nyambori) was 0.5% compared with 0.9% in Uswaa (NS). The prevalence of IGT was similar in the two villages in the 15- to 34- and the 34- to 54-yr-old age-groups; but in those greater than or equal to 55 yr old, IGT was higher in Nyambori (17.4 vs 7.2%, P = 0.029). Mean fasting and 2-h blood glucose levels were slightly higher in Nyambori village after adjusting for age, sex, and BMI (4.5 vs. 4.2 and 5.0 vs. 4.4 mM, respectively)., Conclusions: High dietary cyanide exposure was not found to have had a significant effect on the prevalence of diabetes in an undernourished population in Tanzania. Cassava consumption is thus highly unlikely to be a major etiological factor in so-called MRDM, at least in East Africa.

Objective: To estimate the costs of diagnosis and treatment of diabetes in Tanzania., Design: Costs estimated from the reported and recorded experience of patients with newly presenting diabetes in 1989-90 and of diabetic patients first seen in 1981-2., Setting: Muhimbili Medical Centre, Dar es Salaam., Subjects: 464 patients (315 men and 149 women). 262 patients diagnosed during 1 September 1989-31 August 1990 (group 1) and 202 during 1 June 1981-31 August 1982 (group 2)., Results: The average annual direct cost of diabetes care in 1989-90 was $287 for a patient requiring insulin and $103 for a patient not requiring insulin. Purchase of insulin accounted for 68.2% ($156) of the average annual outpatient costs for patients requiring insulin. For patients not requiring insulin the cost of oral hypoglycaemic drugs and treatment of chronic complications and infections accounted for 42.5% ($29.3) and 48.8% ($33.7) of costs respectively. Cost of outpatient care of diabetic patients for the whole of Tanzania was estimated at $2.7m, *75,128 (32.2%) of which was for insulin. Doctors' and nurses' costs accounted for 0.2% of total costs of outpatient care. The annual direct inpatient care costs were estimated at $1.25m. Around 0.2% of the Tanzanian population aged 15 years and over used the equivalent of 8% of the total government health expenditure, which was $47,4088,382., Conclusion: Diabetes places a severe strain on the limited resources of developing countries. If African patients with diabetes have to pay for their treatment most will be unable to do so and will die.

Between 1 June 1986 and 31 August 1987 all 47 pregnant diabetic Tanzanian women attending Muhimbili Medical Centre, Dar es Salaam were seen and managed by a small team of interested physicians and obstetricians. Of the 50 pregnancies there were 44 (88%) live births, five (10%) perinatal deaths, and one (2%) spontaneous abortion. One child was born with a serious congenital abnormality. All five perinatal deaths were seen in women who presented late in pregnancy (three) or had poor blood glucose control (two). In 10 (36%) of the 28 pregnancies in the 25 patients with insulin-requiring diabetes, insulin requirements decreased greater than 8 U during the course of pregnancy. The present study suggests that with close supervision of the pregnant diabetic patient in Africa perinatal mortality rates approaching those seen in developed countries can be achieved, despite the lack of home blood glucose monitoring. Rates could be further reduced if medical services were more widely available, and if all patients were educated on the importance of early presentation in pregnancy.

Quinine is widely used for the treatment of severe and complicated malaria, although resistant strains of Plasmodium falciparum may occur. The drug has been incriminated as a cause of hypoglycaemia in some malaria patients. To determine if quinine has untoward metabolic effects during treatment of severe and complicated malaria we have studied the effects of quinine on blood glucose and intermediary metabolites, serum insulin, C-peptide, plasma glucagon and non-esterified fatty acids in 97 children with severe malaria in Dar es Salaam. All patients responded clinically. No patient developed hypoglycaemia while on quinine therapy given as 10 mg/kg in 10 ml/kg of 5% dextrose infused over 4 h every 8 h. Endogenous insulin secretion, as reflected by C-peptide levels, increased after 4 h but insulin levels did not change significantly. Blood lactate, 3-hydroxybutyrate, plasma non-esterified fatty acids and plasma glucagon all fell appropriately during treatment. We conclude that quinine, when administered at the recommended dose and rate, does not disrupt blood glucose homeostasis, and is still the drug of choice for severe and complicated malaria in children.

The prevalence of Impaired Glucose Tolerance (IGT) and diabetes mellitus was studied in a migrant Hindu Indian community in Dar es Salaam, Tanzania. Using 1985 WHO criteria, 75 g oral glucose tolerance tests (OGTT) were performed on 1147 (583 men; 564 women) subjects aged 15 years and over. The age-standardized prevalence of IGT in men and women was 15.2 and 17.2%, and that of diabetes was 9.1% (6.5% known; 2.6% previously undiagnosed) and 9.0% (3.7% known; 5.3% previously undiagnosed), respectively. Diabetes was present in 12.9% of men and 12.8% of women aged 35 years and above. The overall age- and sex-standardized prevalence of IGT was 16.2% and of diabetes 9.1%. The major risk factors associated with diabetes in both men and women were age, family history of diabetes, and physical inactivity, and in women body mass index (BMI). Age and BMI contributed to the higher frequency of IGT in women as compared with men while in men, age was the only contributory factor. Despite overweight and obesity being more frequent in women than men, age-standardized prevalence rates of IGT and diabetes were similar between men and women. The decreased prevalence of diabetes in men and women performing moderate/heavy physical activity was independent of age and BMI.

Objectives: To seek differences in the prevalence of diabetes mellitus and other coronary heart disease risk factors, and to identify factors associated with these differences within a Hindu Indian community., Design: Population based cross sectional survey., Setting: Dar-es-Salaam, Tanzania., Subjects: Of 20 Hindu subcommunities categorised by caste in Dar-es-Salaam, seven were randomly selected. 1147 (76.7%) of 1495 subjects aged 15 or over participated., Main Outcome Measures: Blood glucose concentrations (fasting and two hours after oral glucose loading), serum total cholesterol and serum triglyceride concentrations, blood pressure, and height and weight., Results: The subcommunities differed substantially in socioeconomic characteristics and lifestyle. Overall, 9.8% of subjects (109/1113) had diabetes, 17.0% (189/1113) impaired glucose tolerance, 14.5% (166/1143) hypertension, and 13.3% (151/1138) were obese. The mean fasting blood glucose concentration was 4.9 mmol/l, the blood glucose concentration two hours after oral loading (75 g) 6.0 mmol/l, the total cholesterol concentration 4.9 mmol/l, the serum triglyceride concentration 1.4 mmol/l, and body mass index (weight/height: kg/m2) 24.3. Systolic and diastolic blood pressures were 121 and 77 mm Hg respectively. There were important intercommunity differences even after standardisation for age, sex, and body mass index--for example, in mean fasting blood glucose concentration (range 4.5 (Jains) to 5.9 mmol/l (Patels)), serum total cholesterol concentration (range 4.5 (Jains) to 6.2 mmol/l (Suthars)), systolic blood pressure (range 110 (Limbachias) to 127 mm Hg (Bhatias)), and prevalences of diabetes (range 3.4% (3/87 Limbachias) to 18% (20/111 Navnats)) and hypertension (range 5.7% (5/87 Limbachias) to 19.4% (43/222 Bhatias). Variables which showed significant linear correlation with subcommunity variations were entered into a multiple regression model. Intercommunity variations persisted. The Limbachia and Jain communities had the lowest prevalence of and mean values for coronary heart disease risk factors and the Bhatia and Patel communities had the highest., Conclusions: In this series intercommunity variations in disease and risk factors might have been related to genetic, dietary, socioeconomic, and lifestyle differences but could not be explained by the characteristics studied. Studies of Indian subcommunities are warranted to confirm and extend these descriptive findings and explore the genetic basis of diabetes. Communities of Indian origin should not be perceived as homogeneous.

During a study of diabetes prevalence in six rural Tanzanian communities, a repeat oral glucose tolerance test (OGTT) was carried out in 514 subjects greater than or equal to 15 yr of age within 1 wk of an initial 75-g OGTT. In 498 subjects, blood glucose was measured 2 h after the glucose load on both occasions, and in 175 subjects, fasting blood glucose measurement was also repeated. Of the 498 subjects, 245 had normal glucose tolerance in the first test and were selected at random for further testing; 223 subjects had impaired glucose tolerance (IGT), and 30 had diabetic values. Diabetes and IGT were diagnosed on the basis of the 2-h blood glucose values. In the second test, 241 (98.4%) of the 245 subjects with normal tolerance continued in this category and 4 (1.6%) showed IGT. Of the 223 with IGT in the first test, 171 (76.2%) reverted to normal on the second test, 7 (3.1%) had diabetic values, and 45 (20.2%) persisted with IGT. Of the 30 subjects diagnosed as diabetic in the first test, 8 (26.7%) remained with diabetic values, 11 (36.7%) had IGT, and 11 (36.7%) were normal. Based on the second test, the population-prevalence rates of diabetes and IGT would have been 0.5 and 3.3% vs. 1 and 7.6% based on the first test. There was a significant downward trend in the mean 2-h blood glucose values in all three diagnostic groups. Regression toward the mean could not account for the downward shift in blood glucose values observed on retesting.(ABSTRACT TRUNCATED AT 250 WORDS)

There is still controversy concerning the reference ranges for glucose tolerance tests in pregnancy. The WHO has recommended the universal use of the 75 g oral glucose load with 2-h post-load values of greater than 6.7 mmol l-1 to be considered impaired glucose tolerance (IGT) in the non-pregnant, and equivalent to gestational diabetes in the pregnant. Some data are available for pregnant Caucasians but little information is available for other ethnic groups. Oral glucose tolerance tests (75 g) have therefore been performed in 189 pregnant women in rural Tanzania. Mean fasting blood glucose values were 4.0 mmol l-1 in non-pregnant women, and 3.7, 3.5, and 3.3 mmol l-1 in pregnant women in the first, second, and third trimesters, respectively. Two-hour OGTT values were 4.7 mmol l-1, and 4.6, 4.5, and 4.2 mmol l-1 while the upper limit of normal values (mean + 2SD) were 7.1 mmol l-1, and 6.8, 6.8, and 6.1 mmol l-1. The 2-h glucose levels are therefore close to WHO recommendations but lower than those reported for Caucasians. By contrast with reports for Caucasians, glucose tolerance did not deteriorate during pregnancy. The prevalence of diabetes and IGT was zero in the pregnant group.

Glucose tolerance was assessed, according to WHO diagnostic criteria, in 506 consecutive African patients admitted with sputum-positive pulmonary tuberculosis to the tuberculosis wards of Muhimbili Medical Centre, Dar es Salaam. Nine (1.8%) patients were known to have diabetes. Following a 75 g oral glucose tolerance test (OGTT) diabetes was diagnosed in a further 25 (4.9%) patients giving an overall crude diabetes prevalence rate of 6.7%. Impaired glucose tolerance (IGT) was present in 82 (16.2%) subjects. A repeat OGTT was carried out in the 25 patients found to have diabetic values after the first test. Eight (28%) of the 25 patients reverted to normal glucose tolerance after the second test, 6 (24%) to IGT, and 11 (48%) remained with blood glucose values in the diabetic range, giving a crude diabetes prevalence rate of 4%. In a survey of glucose tolerance, using similar methodology, in 693 members of an urban community in Dar es Salaam the prevalence rates of diabetes and IGT were 0.9% and 8.8% respectively. Diabetes was therefore at least four times as common in the tuberculosis patients (p less than 0.001), and IGT twice as frequent (p less than 0.0001). This study confirms the relationship between diabetes and tuberculosis. Those caring for patients with tuberculosis should be aware of the increased prevalence of diabetes in their patients, since failure to diagnose the problem may adversely affect prognosis.

Oral glucose tolerance was studied following a 75 g glucose load in 108 (82.4%) of 131 male and 110 (79.1%) of 139 female members of a Hindu subcommunity aged 15 years and over in Dar es Salaam. One year later, the glucose tolerance tests were repeated in 93 (86.1%) and 93 (84.5%) of the 108 male, and 110 of the female subjects, respectively. In the first survey, 25 (26.9%) of the 93 male and 24 (25.8%) of the 93 female subjects had impaired glucose tolerance (IGT), 6 (6.4%) and 15 (16.1%), respectively, had diabetes mellitus; and 62 (66.7%) and 54 (58.1%), respectively, had normal glucose tolerance. In the repeat survey, of the 93 male and 93 female subjects, 8 (8.6%) and 7 (7.5%) had IGT, 4 (4.3%) and 10 (10.8%) had diabetes; and 81 (87.1%) and 76 (81.7%) were normal, respectively. Of the 21 subjects diagnosed as having diabetes in the first survey, 13 (61.9%) continued to show diabetic glucose tolerance; 4 (19%) IGT and 4 (19%) had normal glucose tolerance with no gender difference. One (1.6%) of the 62 male subjects and none of the 54 females with normal glucose tolerance in the first survey progressed to IGT, while the remainder retained normal glucose tolerance. Diabetes and IGT rates in both surveys were higher for the older than the younger persons. A significant fall in mean fasting and post-75 g blood glucose levels, and in mean systolic and diastolic pressure levels was observed between the first and second surveys in both genders. There was, however, no significant difference in body mass index (BMI), serum total cholesterol and triglyceride levels between surveys, suggesting that major dietary changes had not taken place. Male subjects who showed persistent IGT had significantly lower mean level of body mass index (kg/m2) than subjects who reverted to normal, whilst for the whole group those who had persistent IGT were older. It is tempting to speculate that these changes were due to community action. However, in view of the lack of change in weight and lipids and similar results in other communities in Tanzania when retested at 1 week, further studies are needed to establish the significance of the findings.

Addison's disease is rarely diagnosed in most African countries although tuberculosis, one of its major causes, is a widespread problem. In this study adrenal function was assessed using the Synacthen test in 50 patients with chronic pulmonary tuberculosis admitted to hospital in Dar es Salaam. Sixteen patients (32%) had an impaired response. Two had subnormal basal cortisol levels, one of whom had a normal response to Synacthen. There was no significant difference between the patients with an impaired cortisol response and those with a normal response with respect to frequency of non-specific symptoms, weight loss and body mass index. The mean supine and erect diastolic blood pressures were, however, significantly lower in those with an impaired cortisol response compared to the normal cortisol response group (64 mm Hg vs 74 mm Hg supine (P less than 0.01), and 62 mm Hg vs 73 mm Hg erect (P less than 0.005]. Basal and one-hour plasma cortisol levels correlated significantly with systolic and diastolic blood pressure, and correlated negatively with duration of tuberculosis and diastolic blood pressure. These findings are consistent with reported observations in Zulu patients with pulmonary tuberculosis, and suggest that impaired adrenal function may contribute to morbidity and even mortality among patients with tuberculosis in Africa. Adrenal hypofunction should be considered in any tuberculosis patient with hypotension and poor response to chemotherapy.

During a 1-7 year follow-up of 1250 African diabetic patients at Muhimbili Medical Centre, Dar es Salaam, 68 (5.4%) were known to have developed pulmonary tuberculosis and two (0.2%) spinal tuberculosis. In 18 (25.7%) of the 70 patients tuberculosis was diagnosed prior to the diagnosis of diabetes, and in 32 (45.7%) after the diagnosis of diabetes. In 20 (28.6%) patients tuberculosis and diabetes were diagnosed at approximately the same time with most presenting with symptoms of diabetes. The prevalence of tuberculosis was greater in the young, in those with a low body mass index (BMI), in patients with insulin-dependent diabetes mellitus compared to those with non-insulin-dependent diabetes mellitus (9.0% vs 2.7%) and in those whose diabetes was poorly controlled. Seventeen (24.3%) of the 70 patients are known to have died. Pulmonary tuberculosis may progress rapidly and a high index of suspicion is required. It remains a significant cause of morbidity and mortality among diabetic patients in developing countries.

Glycaemic status on hospital admission was compared in 97 children with severe falciparum malaria (36 with cerebral malaria) and 89 children with other serious illnesses (32 in coma; 57 with acute pneumonia, not in coma). The frequency of hypoglycaemia (blood glucose below 2.2 mmol/l) did not differ significantly between malarial and control patients (5.2% vs 11.2%) nor between the comatose (11.1% vs 18.8%) and conscious (1.6% vs 7.0%) malarial and control subgroups. Compared with normoglycaemic patients, hypoglycaemic patients had appropriately low serum insulin (3.0 vs 8.2 mU/l) and C-peptide (0.13 vs 0.42 mmol/l) and high plasma non-esterified fatty acids (1.42 vs 0.83 mmol/l). Hypoglycaemia, the level of consciousness, and death were all significantly associated with the time since the last meal. Hypoglycaemia is not a specific complication of malaria but is found in severely ill fasted children, resulting from glycogen depletion and perhaps impaired hepatic gluconeogenesis. It should be sought in all severely sick children. A single bolus dose of glucose may not be enough to correct it.

The prevalence of islet cell antibodies (ICA and CF-ICA) together with other organ-specific auto-antibodies was investigated in 122 newly presenting black Tanzanian diabetic patients in Dar es Salaam. ICA were found in three (8.6%) IDDM patients and five (6.8%) insulin-requiring NIDDM patients; six of the eight were also CF-ICA positive. Altogether 22% of patients showed one or more positive autoantibody result but there was no clustering of response, and no association of ICA with other antibodies except for two NIDDM subjects who showed one other positive result. There were no differences between insulin-requiring (IDDM) and NIDDM subjects or between younger (less than 30 years) and older patients. We conclude that there is no major association between diabetes and islet cell antibodies in black Tanzanians.

Objective: To study the clinical characteristics of newly diagnosed diabetic patients in tropical Africa., Design: Prospective study of all newly diagnosed diabetic patients registered at a major urban hospital between 1 June 1981 and 31 May 1987., Setting: Muhimbili Medical Centre, Dar es Salaam, Tanzania., Patients: 1250 Patients: 874 men, 376 women., Results: 272 (21.8%) Patients had diabetes requiring insulin, 825 (66.0%) had diabetes not requiring insulin, and 153 (12.2%) had diabetes of uncertain type. Most patients (1103, 88.2%) presented with the classic symptoms of diabetes. The peak time of presentation of diabetic patients requiring insulin was at age 15 to 19 years. Male manual workers and peasant farmers with diabetes not requiring insulin presented at a significantly older age and had a lower body mass index than sedentary office workers. Forty six (18.1%) of the patients requiring insulin diabetes and 111 (14.4%) not requiring insulin had first degree relative with diabetes. Twenty seven per cent of patients were underweight (body mass index less than 20 kg/m2) and 14.6% were obese (body mass index greater than 30 kg/m2). Hypertension was diagnosed in 211 (26.7%) of 791 patients not requiring insulin. Nine (3.3%) of those requiring insulin may have had the protein deficient type of diabetes related to malnutrition. The fibrocalculous variety of diabetes related to malnutrition was not observed., Conclusions: Newly presenting diabetic patients in Tanzania with diabetes requiring insulin are older at presentation than those in Britain; most diabetic patients present with diabetes not requiring insulin and a smaller proportion of Tanzanian patients are obese. Most have a lower socioeconomic state than diabetic patients in Britain. There are often delays in diagnosis in Tanzania, and there is a higher incidence of death shortly after presentation.

Objective: To determine the clinical course of diabetes mellitus in tropical Africa., Design: Continuing care and follow up until 31 March 1989 of all newly diagnosed diabetic patients registered at one hospital between 1 June 1981 and 31 May 1987., Setting: Muhimbili Medical Centre, Dar es Salaam, Tanzania., Subjects: 1250 Newly diagnosed diabetic patients seen over a six year period. 272 (21.8%) Had diabetes requiring insulin, 825 (66.0%) diabetes not requiring insulin, and 153 (12.2%) diabetes of uncertain type., Main Outcome Measures: Survival rates during each year of follow up., Results: 205 (16.4%) Patients were known to have died, 126 (61.5%) in hospital and 79 (38.5%) in the community. At least a further 71 patients were likely to have died. The five year survival rates (95% confidence intervals) for patients with diabetes requiring and not requiring insulin were 71% (62% to 80%) and 84% (80% to 89%) respectively for known deaths and 60% (51% to 69%) and 82% (77% to 86%) respectively for known plus probable deaths. 49 (3.9%) Patients died at the time of presentation. Severe diabetic ketoacidosis and infection were responsible for most deaths in patients with diabetes requiring insulin. Infection was responsible for 24% of deaths in patients with diabetes not requiring insulin and was the main cause of death in the group with uncertain type of diabetes. Cardiovascular and renal causes were responsible for 24% of hospital deaths of patients with diabetes not requiring insulin. Diabetes requiring insulin, young age, and ketonuria at presentation were associated with a significantly worse five year survival on multivariate analysis. On univariate analysis underweight, female sex, low educational background, and manual occupations were additional factors with a worse prognosis., Conclusion: Diabetes in sub-Saharan Africa is, in many patients, a serious disease with a poor prognosis. Most deaths, however, are due to preventable causes. More effort is therefore required to increase public awareness of diabetes and to improve patient detection, management, and follow up.

The prevalence of diabetes and impaired glucose tolerance has been determined in an Asian Muslim community in Dar-es-Salaam, Tanzania. Two-h oral glucose (75 g) tolerance tests were performed on 1049 subjects over 14 years old, who were fasting, from a random sample of families. The overall age and sex-adjusted prevalence of diabetes was 7.1% (4.4% known, 2.7% previously undiagnosed) with a steady increase from 0.8% at 15-24 years and 3.0% at 25-34 years, to 24.9% for 65 years and over. Impaired glucose tolerance (IGT) rates ranged from 11.4% (15-24 years) to 22.3% (over 64 years). The overall age-adjusted prevalence of IGT was 21.5%. The mean body indices (BMIs) were 24.3 and 26.4 for males and females, respectively, but age-adjusted diabetes rates were similar in the two sexes (7.0% and 7.6%, respectively). Diabetes and IGT were commoner in those with BMI greater than 25 only in the older age groups. Diabetes and IGT were commoner in those with a family history of diabetes. Increasing parity was also associated with a higher diabetes prevalence. Diabetes and IGT are thus common in Asians in Tanzania, in contrast to the indigenous community. Rates are indeed higher than in most other immigrant Asian communities.

The prevalence of diabetes mellitus and impaired glucose tolerance (IGT) was assessed by use of WHO diagnostic criteria in 6299 Africans aged 15 years and above living in six villages in Tanzania. 0.87% (1.1% male, 0.68% female) had diabetes and 7.8% (6.9% male, 7.7% female) had IGT. Prevalence rates were 1.1% and 8.4%, respectively, when age-adjusted to the USA population. Only 7 (13.5%) of the 53 individuals with diabetes had been known to have the disorder; 34 (74%) of the other 46 were symptom-free. Mean age was 54 (SD 20) for diabetic subjects and 37 (17) years for the whole population. Diabetes and IGT rates did not differ significantly between villages despite geographical, socioeconomic, and dietary differences. Diabetes rates increased modestly with age and body mass index (BMI). Fasting blood glucose (FBG) levels did not rise significantly with age but correlated positively with systolic blood pressure (BP) and negatively with haemoglobin concentration (Hb) and BMI. The 2 hour post-glucose load blood glucose values correlated positively with age, sex, and systolic BP and negatively with Hb. Diabetes is less prevalent in rural Africa than in developed countries, even when age has been corrected for. This difference is probably related to body weight, diet, and exercise.

The prevalence of abnormal serum total thyroxine (T4) and triiodothyronine (T3) concentrations were determined in 1,206 in-patients in two associated psychiatric hospitals. The biochemical pattern of primary hypothyroidism occurred in five females and one male (prevalence 0.5 per cent), but in only one patient was the diagnosis clinically obvious. Eight patients (all female) were clinically hyperthyroid (prevalence 0.7 per cent), of whom six were previously undiagnosed. There was no evidence that phenothiazines or benzodiazepine therapy had any significant effect on thyroid hormone levels. The small differences in thyroid hormone levels between psychiatric diagnostic groups could be explained by differences in age distribution.

Pituitary function studies have been carried out in eight male and eight female patients with Down's syndrome who were suspected of having hypothyroidism secondary to pituitary deficiency. In all cases, the response of Human Growth Hormone to hypoglycaemia was adequate and the response of Lutenising Hormone and Follicle Stimulating Hormone to the injection of Lutenising Hormone and Follicle Stimulating Hormone was not signfiicantly greater than that in age and sex-matched institutionalised controls. Therefore there is no evidence that the relative hypothyroidism in these patients is due to pituitary hypofunction, nor do the results provide evidence of pituitary deficiency in Down's syndrome generally.

Two patients remained clinically euthyroid following treatment for thyrotoxicosis with antithyroid drugs in spite of persistently elevated thyroid radioiodine uptakes not suppressable by exogenous triiodothyronine. Perchlorate discharge tests showed a defect in the intrathyroidal organification of iodine. Circulating levels of thyroxine were normal. From our study of 105 patients treated for thyrotoxicosis with antithyroid drugs, apparent remission of thyrotoxicosis by this mechanism might occur in up to 2 percent of patients.

The thyroid status of 82 institutionalized adults with Down's syndrome has been assessed. Compared to age and sex matched control subjects, these patients had significantly lower mean total serum thyroxine (T4) and triiodothyronine (T3) concentrations (T4; 69.1+/-22.2 nmol/1; (mean+/-SD) vs. 100.1+/-19.1, P less than 0.001; T13; 1.61+/-0.47 nmol/1 vs. 1.76+/-0.34, P less than 0.025), lower free thyroxine index (FTI), (FTI; 66.1+/-22.4 vs. 95.1+/-20.2, P less than 0.001), and higher basal serum thyrotrophin (TSH) concentrations (TSH; 7.6+/-10.7 mU/1 vs. 3.8+/-1.5, P less than 0.001). These changes were not related to age or sex. Abnormalities in one or more test of thyroid function were demonstrated in at least 38 (46%) of the 82 patients. Two main patterns of abnormality were defined: 1) subnormal T4, FTI and elevated basal TSH levels (primary hypothyroidism) in 13 (16%). All seven of the 13 patients in whom TRH tests were performed showed the expected exaggerated TSH response, and seven out of the 13 patients (54%) had positive thyroid antibodies, 2) Subnormal T4, subnormal or low normal FTI, and basal TSH levels within the normal range in 18 (22%). The mean basal TSH concentration was, however, significantly higher than in patients with normal T4 and FTI levels, suggesting a minor degree of thyroid failure. Only two of the 18 patients (11%) had positive thyroid antibodies. Of the 17 patients in the group tested, 13 showed a normal TSH response to TRH, three an exagerrated response (all females), and one had an impaired response. Other patterns of abnormal thyroid function were observed occasionally: one female patient had biochemical T3 toxicosis; another had the biochemical pattern of subclinical hypothyroidism, four patients with normal basal T4, FTI and TSH levels showed an exaggerated TSH response to TRH and one patient had an impaired response. These data indicate that htyroid dysfunction, in particular hypothyroidism, is common in adults with Down's syndrome, though specific tests are usually required to make the diagnosis. The general reduction in thyroid function in Down's syndrome may be due to impaired development of the thyroid gland. However, frank chemical hypothyroidism may occur only when thyroiditis is superimposed on preexisting diminished thyroid reserve.

1 Carbimazole was administered to nine hyperthyroid patients, and blood samples were taken at various time intervals for analysis of carbimazole, and its metabolite methimazole.2 A technique was developed for the measurement of methimazole in serum using a High-Pressure Liquid Chromatograph, which could detect nanogram quantities of this metabolite.3 After a single oral dose, the patients' blood levels appeared to fall into two groups, either those with a maximum concentration of methimazole between 30 and 60 min, or those whose maximum was 2 to 3 hours.4 The results would suggest that there could be a correlation between the high level of methimazole in serum and the high thyroxine concentration found in some patients.

1. Classical and high-pressure liquid chromatographic separations were devised for the separation and isolation of the metabolite 3-methyl-2-thiohydantoin from the urine of patients receiving carbimazole orally. 2. 3-methyl-2-thiohydantoin was identified by comparing its absorption and mass spectral properties with authentic material. 3. 3-methyl-2-thiohydantoin was also detected in the plasma of patients receiving methimazole intravenously.