Prescription for Health is a national program of The Robert Wood Johnson Foundation (RWJF) in collaboration with the Agency for Healthcare Research and Quality (AHRQ). A major goal of Prescription for Health is to measure the extent to which comprehensive strategies are effective in changing patient behavior and quality of life relative to four target health risk behaviors: diet, smoking, alcohol use, and physical activity. The funded projects will use a common set of survey instruments to help measure outcomes and draw overarching conclusions across projects. This study will only be analyzing aggregated data and does not have responsibility for recruitment of patients, randomization (if applicable), or interventions. Individual project designs may differ.

The data will be collected by and submitted from up to 10 practice-based research networks over a two-year period. We will: 1) aggregate de-identified data and 2) analyze the aggregated data. We will not identify or recruit study subjects, administer surveys to patients, or conduct any interventions with patients or practices. The individual research networks will submit only de-identified data to a central data repository for subsequent analysis by us. Individual research networks will submit their own protocols for review by their IRBs in order to collect patient-specific data, conduct interventions, and submit de-identified data to our central repository.

By combining data from funded projects we will be able to answer questions that individual projects would not be able to answer. The analysis will use the combined set of data to gain statistical power sufficient to test hypotheses of interest, determine the distribution of responses to the selected measures in patients in primary care practices, assess the impact of types of interventions, compare the impact of different types of interventions, determine if these measures can be used routinely, and conclude how useful, if at all, the selected measures were in Prescription for Health projects. Statistical analysis will include basic descriptive statistics and distributions of scores at baseline and follow-up and more complex hierarchical linear modeling, which seeks to understand effects in multi-site studies (across sites or groups of sites). We will not directly evaluate specific research network interventions, but will assess them as part of larger groups. This study will report only aggregated results and will not name individual primary care practices or research networks.

Inclusion Criteria:

• varies by project

Exclusion Criteria:

• varies by project

Adults, adolescents, and children in primary care practices.