OBJECTIVE -Many African-American women are affected by diabetes and its complications, and culturally appropriate lifestyle interventions that lead to improvements in glycemic control are urgently needed. The aim of this qualitative study was to identify culturally relevant psychosocial issues and social context variables influencing lifestyle behaviors-specifically diet and physical activity-of southern African-American women with diabetes.RESEARCH DESIGN AND METHODS -We conducted 10 focus group interviews with 70 southern African-American women with type 2 diabetes. Group interviews were audiotaped and transcripts were coded using qualitative data analysis software. A panel of reviewers analyzed the coded responses for emerging themes and trends.RESULTS -The dominant and most consistent themes that emerged from these focus groups were 1) spirituality as an important factor in general health, disease adjustment, and coping; 2) general life stress and multi-caregiving responsibilities interfering with daily disease management; and 3) the impact of diabetes manifested in feelings of dietary deprivation, physical and emotional "tiredness," "worry," and fear of diabetes complications.CONCLUSIONS -Our findings suggest that influences on diabetes self-management behaviors of African-American women may be best understood from a sociocultural and family context. Interventions to improve self-management for this population should recognize the influences of spirituality, general life stress, multi-caregiving responsibilities, and the psychological impact of diabetes. These findings suggest that family-centered and church-based approaches to diabetes care interventions are appropriate.

BACKGROUND: Studies of adolescent smoking suggest that the causes of smoking initiation may differ for Blacks and Whites. METHODS: Correlates of smoking initiation were examined among 1,277 nonsmokers, ages 12-14, who completed questionnaires in their homes. The analyses examined relationships between smoking initiation and 11 explanatory variables using logistic regression with the combined sample and with Black and White samples. RESULTS: Over two years, 24 percent of Whites and 14 percent of Blacks started to smoke. Whites were more likely to start smoking at age 12 and Blacks at age 14. Having a best friend who smoked increased the odds of initiating smoking over twofold for Whites but had no effect on the odds of smoking for Blacks. CONCLUSIONS: Whites initiate smoking earlier than Blacks and are more likely to be influenced by friend behavior.

This paper presents a theoretical framework for conceptualizing alcohol-related disparities experienced by young Blacks in the United States. The framework highlights areas of risk and opportunity as they relate to the development of alcohol use and alcohol-related problems. In this paper, life course development theory serves as a guide for identifying a critical period in the development of alcohol-related disparities and it serves to guide the identification of opportunities to prevent or attenuate this health outcome. We also highlight concepts from ecosocial theory, resilience theory, and prevention science that advance our understanding of risk and protective factors for the social problems that young Blacks experience related to alcohol use. We conclude with suggestions for designing studies that range from etiology to preventive interventions. We also recommend methodologies that allow for more nuanced understandings of the etiology and prevention of alcohol use and alcohol-related problems experienced by young Blacks than have been available to date.

' Images of Smoking and Smokers SYNOPSTSThe authors used qualitative and quantitative data to identify and interpret specific images teens have about smoking and smokers. Qualitative data were collected in 1996 from 793 teenagers participating in 125 focus groups at eight different sites across the Unrted States. lYost focus groups were homogeneous with respect to gender, ethnicity, and smoking status. Ages ranged from l2to l8 years, and about halfofthe parlicipants were female. The majority of participants (62%) were white and African American, the remainder (38%) were Hispanic, American Indian, and Asian/Pacific lslander. Groups were comprised of smoking and nonsmoking teens. Focus group activities were used to elicit image-related discussions about attitudes, beliefs, and perceptions of smoking. Investigators identifled seven consistent and distinctive image themes:Appeoronce (smoking is dirty and unattractive), Activity (nonsmokers have busy, active lives), Drugs ond sex (smokers aie substance abusers and are sexually active), Rebe//lon (smokers belong to rebellious groups), Affect (smokers are depressed, angry, and stressed-out) /n controi (nonsmokers have self-control and are independent), and Pdde (nonsmokers are proud of themselves, their families, and their heritage).A large scale, multi-site qualitative research approach can increase understanding of teen smoking. The identification of distinctive images of smoking can help researchers develop more sophisticated models of the processes of teen smoking than currently extst.PUBLIC HEALl'H REPORTS ' 2OO1 SUPPLEMENT 1 ' VOLUME 116

Between one-fifth and one-third of patients hospitalized on general medicine wards experience significant depressive symptoms during their hospitalization. This study employed 71 general medical in-patients and examined the relative association of illness/hospitalization characteristics, patient characteristics and environmental characteristics with in-patient medical depression. Multiple regression results indicated that in-patient medical depression was related to pre-hospitalization depression and social functioning, patient perception of physician supportiveness and patient perception of illness-related life-disruption. None of the objective illness/hospitalization variables related to depression while in the hospital. These results are interpreted with regard to several current theories in medical psychology including a life-stress model emphasizing the ability of prior disorder to predict subsequent disorder, a social interaction model focusing on the effects of physicians' supportive behaviour on patients' emotional adjustment in the hospital, and models of illness that stress cognitive appraisal in determining illness-related mood and behaviour.