farmington hospital ratings

Use Hospital-Ratings to find a compare hospital's in your area! We have ratings for hospital's everywhere! Click here

In the second stage, the dependent variable was whether an individual did or did not receive each of the following services within the past two years: mammograms, pap smears, blood pressure checkups, cholesterol level checkups, and flu shots. The probability of receiving a service within the past two years was the Y in the equation shown in the empirical model section. The explanatory variables were predisposing, enabling, need, and the endogeneity correction factors. A probit model was used in the second stage.Because the MEPS used a complex sampling design, person weight, primary sampling unit, and sample stratification were incorporated in the two-stage estimation to obtain nationally representative estimates and correct standard errors. The standard errors for the use of predicted probabilities were also adjusted by the weighting scheme of the MEPS. Computer software STATA[R] version 7 (Stata Corp: College Station, TX) was used to perform the regressions controlling for the complex sampling design.

EMPIRICAL ESTIMATIONEmpirical ModelFollowing Figure 1, two equations were used for the estimation. The dependent variables were an individual''s choice of a usual source of care (a regular doctor, a regular site, or no usual source care), denoted by USC, and preventive service use, Y.Prob(USC) =f(predisposing, enabling, needs, error term I) (1)Y=g(predisposing, enabling, needs, USC, error term II) (2)An endogenous treatment effect model, an extension of a selection model, would be appropriate for analyzing the relationship between having a usual source of care and service use, as described above. Numerous analogies of the relationship between a usual source of care and service use can be found in the economics literature. Individuals purchase insurance because they foresee future use of medical care (Dowd et al. 1988). Workers join a union because they expect higher wages (Svejnar 1980). Consumers who purchase cars are those who travel frequently (Truong and Hensher 1985). Individuals who buy energy efficient appliances are those who use the appliances frequently (Revelt and Train 1997). In this study, a control function estimation was used to operationalize the endogenous treatment effect model.