Are patients who use a generalist physician healthier than those who seek specialty care directly?

BACKGROUND: American health care consumers want the option of seeing specialists whenever they wish, but given this option, do they in fact use it without consideration of their health status? This paper reports on a cross-sectional analysis that compares the demographics and health status of fee-for-service Medicare enrollees who exhibited four different patterns of physician access.
METHODS: The Medicare Beneficiary Survey data from 1998 were used. Subjects ages 65 and older were categorized into one of four groups: those with no physician claim, those who saw a generalist only, those who saw a specialist only, and those who saw both. Age, income, education, health status, level of impairment, and disease burden for the four patient groups were compared using ANOVA. Urban/rural status, race, ethnicity, mortality rates, and gender for the four patient groups were compared using chi-square. A predictive model using mutinomial logistic regression was created.
RESULTS: Twelve percent of subjects saw no physician in 1998, 11.6% saw a generalist only, 14.2% visited a specialist only, and 62.1% visited both types of physicians. Subjects who saw both physician types had significantly worse health status and more chronic diseases than the other groups. Subjects who saw generalists only or specialists only had intermediate levels of health status and disease burden that were not significantly different from each other. Subjects who saw a specialist only were the most affluent and highly educated group. Subjects who saw no physician had the best health status and the fewest chronic diseases of all subject groups. Urban residents were more likely to visit some type of physician than were rural residents and were more likely to see a specialist only. Regional differences were noted, with New England showing the highest rates of specialist only use. DISCUSSION: As expected, the healthiest subjects were least likely to visit any health care provider. Subjects with the worst health status were likely to access both generalists and specialists for their care. Subjects who visited a specialist only had higher incomes, more education, and urban residence but no difference in health status when compared to subjects who visited a generalist only.