There is growing concern among primary care physicians that the DSM system does not accurately reflect mental health problems seen in primary care. Primary care patients are frequently judged by their primary care physician to be psychologically distressed and functionally impaired; yet they do not meet DSM criteria for a mental disorder. However, studies using the DSM system are cited as evidence that primary care physicians fail to recognize mental health problems, do not optimally treat those they recognize, and need to refer such patients to mental health professionals. The purpose of this pilot study is to test the methods and estimate statistical parameters needed for conducting a longitudinal primary care-based study assessing the accuracy and clinical validity of the DSM classification system in primary care settings. This study will measure recruitment and dropout rates using three follow-up strategies. It will also determine optimal timing for qualitative interviews and whether there are differences based upon baseline """"""""distress"""""""". This longitudinal primary care based pilot study will follow 100 Hispanic and non-Hispanic white patients over a 6-month period. Excluding patients with DSM disorders using a modified SCID screening module or those on psychotropics at baseline, 50 distressed and 50 non-distressed patients matched on the number of chronic medical problems will undergo baseline assessment including the Brief Social Desirability Scale. Baseline and monthly interviews will include identification of common DSM disorders via the SCID, measurement of levels of anxiety, depression, and somatization via the SCL-R, assessment of distress via the GHQ-12, assessment of recent acute illnesses, measurement of functional status using the SF-36, identification of life stresses using the Holmes and Rahe scale and the Daily Hassles Scale, measurement of hope and support via the Herth Hope Index and the DUSOCS respectively, and assessment of alcohol use via the AUDIT. Monthly follow-up interviews will occur using one of three follow-up strategies. Additionally, 16 balanced qualitative interviews will be performed to determine optimal timing of interviews and assess validity of quantitative assessments. Analysis of temporal relationships will rely upon latent growth curve analysis. In addition, the study will identify which follow-up strategy produces the highest follow-up rate and whether baseline """"""""distress"""""""" is an important inclusion criterion. This pilot study will also assess response bias via method of interview, levels of symptomatology corresponding to classification thresholds, and the relationship between social desirability and variation over time.
