Socio-cultural determinants of child mortality in southern Peru: including some methodological considerations.

Among Amerindian children living at high altitude in the Andes in southern Peru, high child mortality rates have been reported in the literature, especially in the perinatal and neonatal period. We compared mortality rates in children calculated from retrospective survey data in 86 rural families from 2 Aymara and 3 Quechua peasant communities living at the same level of altitude (3825 m) in southern Peru. Relations between land tenure, socio-cultural factors and child mortality were studied, and methodological considerations in this field of interest are discussed. Checks on consistency of empirical data showed evidence for underreporting of neonatal female deaths with birth order 3 and more. Perinatal (124 vs 34 per 1000 births) and infant mortality (223 vs 111 per 1000 live births) was significantly higher in Aymara compared with Quechua children, but no difference was found after the first year of life. A short pregnancy interval was associated with an elevated perinatal and infant mortality rate, and a similar albeit insignificant association was found with increased maternal age. Amount of land owned and birth order were not related with child mortality. Although levels of maternal education are generally low in both cultures, a consistent decline in infant and child mortality was found with the amount of years mothers had attended school. However, the results suggest a U-shaped relationship between the amount of years of parental education and perinatal mortality in offspring. Late fetal and early neonatal mortality were particularly high in one Aymara community where mothers were found to have more years of education. Infanticide, a known phenomenon in the highlands of the Andes, is discussed in relation with the findings of the study. Although maternal and child health services are utilized by the majority of families in 4 of 5 study communities, 43 of 51 mothers under the age of 45 years reported that they delivered their last baby in the absence of traditional midwives or official medical supervision.

Differences in physical growth of Aymara and Quechua children living at high altitude in Peru.

Physical growth of Amerindian children living in two Aymara and three Quechua peasant communities in the Andean highlands of southern Peru (altitude 3,810-3,840 m) was studied, taking into account differences in the microclimate, agronomic situation, and sociodemographic variables. Anthropometric measurements were taken in 395 children aged under 14 years of age in a sample of 151 families in these communities, who were surveyed for sociodemographic variables as well. Data on the land system were available for 77 families. In comparison with reference populations from the United States (NCHS) and The Netherlands, stature, weight, head circumference, and midupper arm circumference (but not weight for stature) in the sample children were reduced. Growth retardation increased after the age of 1 year. Stature and weight in the present sample were very similar compared with previously published data on growth of rural Aymara children living near Lake Titicaca in Bolivia. Head circumference, midupper arm circumference, and weight for stature were significantly larger in Aymara children compared with Quechua children. Land was significantly more fragmented in Aymara compared with Quechua families, but amount of land owned was not different. Perinatal and infant mortality was elevated in Aymara vs. Quechua communities. Most families in Aymara communities used protected drinking water. One Quechua community had a severe microclimate, grim economic outlook, and weak social cohesion. Children in this community showed significant reductions in weight and midupper arm circumference compared with their peers in the other communities. We conclude that (presumably nutritionally mediated) intervillage and Aymara-Quechua differences in childhood physical growth existed in this rural high-altitude population in Peru and were associated with microclimate and the village economy, sociodemographic factors, and differences in the land system.

Relationship of insulin clearance and secretion to insulin sensitivity in non-diabetic Mexican Americans.

The antecedents of type II diabetes, while still controversial, are thought to involve decreased insulin sensitivity and compensatory hypersecretion of insulin. Mexican Americans have a three-fold excess risk of type II diabetes and non-diabetic Mexican Americans are characterized by hyperinsulinaemia and insulin resistance. Few data exist, however, on whether there are defects in insulin secretion and/or clearance in this population. We examined insulin sensitivity, secretion and clearance using combined insulin and C-peptide measurements analysed by the minimal model technique of Bergman and colleagues in 10 non-obese, normoglycaemic Mexican Americans and 11 age, sex and obesity-matched non-Hispanic whites. Mexican Americans had significantly decreased insulin sensitivity (SI 4.06 s. 7.56, P = 0.017), higher first phase insulin secretion (1.03 nM vs. 0.72 nM) and decreased insulin clearance (0.099 vs. 0.161) than non-Hispanic whites. Thus, normal Mexican Americans have higher rather than lower insulin secretion suggesting that lower insulin sensitivity may be an early defect in this ethnic group. In addition, they have reduced insulin clearance. Moreover, insulin sensitivity and insulin clearance were positively correlated. We thus speculate that decreased insulin clearance may represent a further autoregulatory mechanism in addition to increased insulin secretion to compensate for decreased insulin sensitivity.

Diminished insulin sensitivity and increased insulin response in nonobese, nondiabetic Mexican Americans.

The contributions of diminished insulin sensitivity and decreased insulin response to the development of non-insulin-dependent diabetes mellitus (NIDDM) remain controversial. Nondiabetics in high-risk populations for NIDDM, including Pima Indians and Mexican Americans, are characterized by obesity and hyperinsulinemia relative to nondiabetics in the lower-risk white population. However, it is not clear to what extent diminished insulin sensitivity in the high-risk groups reflects obesity per se or is an inherent characteristic of these groups. Insulin sensitivity and secretion were determined in 10 nonobese, normoglycemic Mexican Americans (mean body mass index [BMI], 23.8 kg/m2) and 11 normoglycemic non-Hispanic whites (mean BMI, 22.5 kg/m2) using the intravenous glucose tolerance test (IVGTT) and the minimal model approach of Bergman et al. Age, BMI, sum of skinfolds, and the ratio of waist-to-hip circumference (WHR) were similar in both ethnic groups. Mexican Americans had decreased insulin sensitivity compared with non-Hispanic whites (4.06 +/- 0.72 in Mexican Americans v 7.56 +/- 1.13 in non-Hispanic whites, P = .017). The areas under the C-peptide and insulin curves were significantly greater in Mexican Americans than in non-Hispanic whites (P less than .01), suggesting greater insulin secretion in the former. This study provides evidence for diminished insulin sensitivity and increased insulin response in young, nonobese, normoglycemic Mexican Americans.