In Reply Dr Fiscella notes that there was a negative correlation between discharges for ambulatory care–sensitive conditions and life expectancy at the county level, as reported in the supplemental appendix in our study. Fiscella correctly notes that this finding is worthy of further investigation.
We were concerned about placing too much weight on this one finding because other measures of access to preventive care did not display such an association with longevity, including the share of people with a primary care visit, adequate testing for persons with diabetes, and appropriate mammography screening. In contrast, correlations with measures of healthy behavior were more robust across alternative proxies.