Table 4. Estimated odds ratios for the increase in vaccination coverage associated with an increase of one pediatrician per 10 000 children in Japan.

	OR [95% CI]	P value	Adjusteda OR [95% CI]	P value
Measles-Rubella (n = 1733)
	1.022 [1.021–1.024]	<0.001	1.012 [1.010–1.015]	<0.001
DPT (n = 1691)
	1.013 [1.011–1.014]	<0.001	1.019 [1.016–1.022]	<0.001

CI, confidence interval; DPT, diphtheria, pertussis, and tetanus; OR, odds ratio.

^aThe models included the following control variables: total population, a composite index of socioeconomic indicators created from socioeconomic variables for the percent of the population with a college-level education, the percent of white-collar workers, the unemployment rate, and per capita income, the incidence of single motherhood, number of hospital beds per 1000 population, length of roads, crime rate, accident rate, land area, PED_dummy which equals 1 if municipalities had at least one pediatrician and equals 0 otherwise, and an interaction term between PED_dummy and the density of other types of physicians and regional dummy variables for prefectures for DPT and the ones without length of road and the prevalence of single motherhood for Measles-Rubella.