Street ball, swim team and the sour cream machine: a cluster analysis of out of school time participation portfolios.

Adolescents spend only a fraction of their waking hours in school and what they do with the rest of their time varies dramatically. Despite this, research on out-of-school time has largely focused on structured programming. The authors analyzed data from the Educational Longitudinal Study of 2002 (ELS:2002) to examine the out-of-school time activity portfolios of 6,338 high school sophomores, accounting for time spent in school clubs and sports as well as 17 other activities. The analytical sample was balanced with respect to sex and racially and ethnically diverse: 49% female, 67% White, 10% Latino, 10% African American, and 6% Asian and Pacific Islander. Approximately 76% of the sample attended public schools, 30% were in the highest socioeconomic quartile, and 20% were in the lowest socioeconomic quartile. The authors identified five distinct out-of-school time activity portfolios based on a cluster analysis. The demographic profiles of students by portfolio type differed significantly with respect to sex, race/ethnicity, socioeconomic status, school type and location. Students by portfolio type also differed significantly in terms of measures of academic success, school behavior, victimization and perceptions of school climate, controlling for covariates. These findings underscore the importance of more complex considerations of adolescents' out-of-school time.

Community-based screening for cardiovascular disease and diabetes using HbA1c.

BACKGROUND: Persons from inner-city immigrant and mixed-ethnic communities are known to be at high risk for cardiovascular disease and diabetes. Such communities may also be underserved for preventive medical care. The authors hypothesized that hemoglobin A1c (HbA1c) could be used as a screening test for a community-based program to detect new cases of diabetes and persons at risk for diabetes and cardiovascular disease. METHODS: Screenings took place in churches, group homes, shelters, community centers, and street corners of the Bronx. Screening data included history of diabetes, age, ethnicity, body mass index, blood pressure, lipid panel, random glucose, and HbA1c. Data were analyzed for number of cases of new diabetes (HbA1c > or =7%), for patients at risk for diabetes (HbA1c 6%-6.99%), and for associations between HbA1c and other variables. The effect of location of screening and self-reported ethnicity on outcome variables was also analyzed. RESULTS: Seven hundred four persons were screened in 25 different sessions. HbA1c and lipid profile were obtained on 539 persons, which formed the cohort for analysis. Mean HbA1c for the cohort was 6.00%. Thirty-two percent of the cohort had HbA1c of more than 6%, and 11.4% had HbA1c of more than 7%. Excluding known diabetics (13% of cohort), 24% had HbA1c of more than 6%, and 3.4% had HbA1c of more than 7%. HbA1c was significantly correlated with total cholesterol, triglycerides, low-density lipoprotein, systolic blood pressure, body mass index, and age; in all cases, correlation coefficients were higher with HbA1c than with random glucose. In addition, significantly higher cardiovascular disease risk factors were found in persons with HbA1c of more than 6%; 6% may be a threshold value for the metabolic syndrome. Mean HbA1c was higher in persons from the South Bronx (which has a higher poverty rate) than the North Bronx (6.08% v 5.74%, p=0.013). There were no statistically significant differences between self-reported ethnic groupings. CONCLUSIONS: There was a high prevalence of undiagnosed diabetes, and of patients at risk for diabetes, in this community setting. Community-based screening can be used as a method for identifying high percentages of patients at risk for diabetes or with undiagnosed diabetes in an inner city, immigrant, mixed-ethnic population.