Adolescent health, risk behaviours, and mortality

Despite adolescents (10- 19 year olds) having the lowest mortality compared to older and younger age groups in the country, the mortality rate of this age-group is increasing. Prominent broad- underlying and immediate causes of death have varied very little over the period. What remains unclear is the determinants of these causes of death among South African adolescents. This paper is a study of the determinants of adolescent mortality in South Africa. Data from the South African Death Registry is used (2006- 2009). Cause-specific mortality rates and proportional mortality ratios are used. Logistic regression is then used to infer the odds of adolescent mortality by specific demographic and socioeconomic characteristics. Results paper include that occurrences of mortality due to ‘Tuberculosis’ and ‘ill- defined and unknown causes’ are high among adolescents. Further this paper has found that female adolescents are at greater odds of natural or disease mortality than males. However, females are at lower odds of unnatural and violence related causes of death. Also, marriage decreases the odds of mortality.

The proportion of adolescent deaths (9%) due to pregnancy or during child birth to total maternal mortality is very high in India. Although some studies on local scale are available, no study as yet has explored the factors affecting maternal health care utilization in adolescent mothers at national level. Using the data from third wave of District Level Household Survey (2007 -08), this paper examines the factors associated with the utilization of maternal healthcare services among adolescent women (aged 13 -19 years) in urban India. Bi-variate analyses including chi -square test was used to determine the difference in proportion, and multilevel binary logistic regression models were applied to understand the net effect of predictor variables on selected outcomes. About 23% of all adolescent mothers in India availed full ANC; 70% had a safe delivery and 64% had postnatal care in 2007 -08.After controlling for a number of factors, we find that the women, who are rich, older, highly educated, exposed to mass media , were more likely to utilize prenatal and natal care. These groups may be small portion of population but contribute significantly to maternal and child mortality, therefore, it is imperative to focus on such vulnerable groups in the policy.

While certain mass media exposure is associated with the presence of problem behaviors, little is known about the longitudinal impact of exposure to internet on the development of cigarette smoking and alcohol misuse in adolescents. Due to the pervasiveness of internet use among adolescents, this paper examines internet exposure and its linkage with future adolescent cigarette/alcohol use. It is also argued that this longitudinal relationship may differ by gender; an attempt is made to delineate plausible internet exposure revealed by boys versus girls in cigarette versus alcohol use. Data are taken from the Taiwan Youth Project, a cohort sample who did not have cigarette/alcohol use experience at age 16 with a 4-year follow-up. Measures of internet exposure include 3 aspects: where to use, why to use, and how often to use. To incorporate conventional determinants with peer, individual, and family characteristics measured at age 16 or younger, multivariate regressions are employed to estimate models in odds of development of cigarette smoking and alcohol misuse by age 20. Results confirm that in addition to the above determinants, adolescent internet exposure contributes to the development of cigarette smoking and alcohol misuse and its relationships vary by gender.

Adolescent risk taking behaviours have become important policy issues in South Africa which is faced with a myriad of social and health problems among youth. HIV prevalence rates are high, adolescent rates are among the highest in the world, substance and alcohol abuse are high among teenagers and rape cases perpetuated by boys to girls in these age groups have increased. In response to these issues, a study on Risk Behaviour Among School Adolescents was undertaken.

A multi-cluster sampling was used to collect data covering 7 schools and 1065 students in Grades 9 and 11. Considering the sensitivity of the questions, need for honesty responses and the high literacy levels of the group, self-completion questionnaires were used. Logistic regressions were used to model the effects of ever had sex, frequency of sex and sexual partners. For this analysis these were reduced to binary outcomes.

Parental control and peer pressure have emerged as the strongest predictors of teenage sexual activities. Also, the roles of the family and religious institutions as protective factors have been established. Peer pressure emerged as the most important risk factor. One important observation is that youth who have accurate knowledge are more likely to take protective action.