The long term objective of this research is to evaluate and analyze outcomes of primary health care for homeless clients. Patient satisfaction with health care is a critical outcome particularly when considering the homeless population who are known to be underusers of health care (Hodnicki, 1990, Gelburg, 1993). Yet measures of satisfaction with demonstrated validity and reliability in general populations of clients have shown a lack of variability or a skew in ratings with vulnerable populations (Avis, Bond and Arthur, 1997; Larrabee, Ferri and Hartig, 1997; Lindsey, Henly and Tyree, 1997; Rubin et al. 1993). Therefore, the aim of this study is to develop and validate a reliable measure of client satisfaction with primary health care for homeless clients. The study will have two phases. The first phase will use qualitative phenomenologic inquiry methods to explore the meaning of satisfaction with primary health care for homeless clients. The themes identified will be used to construct a quantitative measure of satisfaction with care which will be piloted tested. The second phase will be a descriptive cross sectional comparison of three measures of satisfaction in two samples of homeless clients using primary health care. One hundred homeless subjects from a rural clinic and 100 homeless subjects from an urban clinic will complete a questionnaire consisting of a demographic survey, the Patient Satisfaction Questionnaire III (PSQIII) (Marshall, Hays, Sherbourne and Wells, 1993), the DiTomasso-Willard Patient Satisfaction Questionnaire (DWPSQ) (DiTomasso and Willard, 1991), and the inductively developed measure from the first phase of this study. Data analysis will exam psychometric properties of the three measures with the two samples. Using each satisfaction score as the dependent variable, multiple regression will be used to examine the relative effects of clients' demographic characteristics, primary diagnosis, comorbidities, insurance status, clinic type and scores on other satisfaction measures on the dependent measure. Results of this study could provide a valid and reliable measure of homeless clients' satisfaction with primary health care which could be used in evaluating outcomes of health care programs for this difficult to serve population.
