OBJECTIVES: This study uses the recently developed Network-Episode Model (NEM) to examine the nature and correlates of utilization among Puerto Ricans reporting mental health problems. The NEM highlights two issues: (1) examining the patterns or combinations of lay and formal use that individuals employ and (2) reformulating how the availability and content of social networks influences patterns of care. METHODS: Using data from the 1989 Mental Health Care Utilization Among Puerto Ricans Study (probability sample of 1,777 individuals living in low-income areas of the island), the authors focus on the patterns and correlates of use for 365 Puerto Ricans reporting service use for mental health problems in the previous year. RESULTS: A combination of clustering and multinomial logit techniques indicates that there are six unique care patterns. Two patterns include the use of mental health providers and are associated with different contingencies. CONCLUSIONS: In general, patterns of use are shaped by age, education, gender, and illness severity. Larger, more supportive networks decrease the use of patterns of care that include formal health care providers, and decrease direct entry into the mental health sector. These results are in line with NEM's predictions for lower class populations and help clarify inconsistencies in previous research on social networks. The implications of this perspective for health services research and treatment are discussed.
OBJECTIVES: This study uses the recently developed Network-Episode Model (NEM) to examine the nature and correlates of utilization among Puerto Ricans reporting mental health problems. The NEM highlights two issues: (1) examining the patterns or combinations of lay and formal use that individuals employ and (2) reformulating how the availability and content of social networks influences patterns of care. METHODS: Using data from the 1989 Mental Health Care Utilization Among Puerto Ricans Study (probability sample of 1,777 individuals living in low-income areas of the island), the authors focus on the patterns and correlates of use for 365 Puerto Ricans reporting service use for mental health problems in the previous year. RESULTS: A combination of clustering and multinomial logit techniques indicates that there are six unique care patterns. Two patterns include the use of mental health providers and are associated with different contingencies. CONCLUSIONS: In general, patterns of use are shaped by age, education, gender, and illness severity. Larger, more supportive networks decrease the use of patterns of care that include formal health care providers, and decrease direct entry into the mental health sector. These results are in line with NEM's predictions for lower class populations and help clarify inconsistencies in previous research on social networks. The implications of this perspective for health services research and treatment are discussed.