Use of inpatient services by location, age range, and length of time since diagnosis.

DM diabetes mellitus, IO insufficient observations to calculate the statistic, N number of participants, NGT normal glucose tolerance.ap-value ≤0.10 for age.bp-value ≤0.05 for age.cp-value ≤0.01 for duration of diabetes.dp-value ≤0.05 for overall DM vs. NGT difference.ep-values not calculated for this category because results were derived in part from means instead of person-level data.fp-values not calculated for this category; p-values calculated for components within the category. General notes1) Data based on use of services during the preceding 90 days, as reported by the participant, except mean length of stay, which was based on the most recent hospitalization during the preceding year.2) Difference was defined as the rate for participants with DM minus the rate for participants with NGT.3) Ratio was defined as the rate for participants with DM divided by the rate for participants with NGT.4) Mean number of admissions among those admitted, preceding 90 days was not broken down by number of admissions because multiple hospitalizations occurred too infrequently to yield meaningful results.5) Mean annual admissions per person were determined by multiplying self-reported admissions per person from preceding 90 days by 4.0.6) Mean inpatient days per person were determined by multiplying mean annual admissions/person by mean length of stay.7) P-values were intentionally not calculated for the days per person and admissions per person data because these values were calculated from group means.8) Results as displayed are unadjusted except for the NGT values in the column labeled “All Participants/Adjusted.” In that column, adjusted values for participants with NGT were directly standardized to the age-sex distribution of participants with DM.9) All p-values were calculated from multivariable regression models using continuous measures of age and, for participants with DM, years since diabetes was diagnosed.