Linking individual medicare health claims data with work-life claims and other administrative data

BMC Public Health

Table 3 Determinants of inpatient and outpatient visits (Sample 1)

***p < 0.01, **p < 0.05, *p < 0.1
Three OLS models and three Poisson models are presented for each of the two outcome variables (inpatient and outpatient visits). These analyses are performed on repeated cross-section data. Because there are repeat measures for individuals, a set of dummies for individuals is necessary. These controls also remove variation caused by static individual characteristics (e.g., some people go to the doctor more often than others). In addition, standard errors are clustered by individual to account for serial correlation. Otherwise, model 1 is completely unadjusted. The model 2 includes an age polynomial, because there are increases in medical care consumption that occur as individuals age. Model 3 includes a measure of whether the individual retired

ISSN: 1471-2458