Table 2. Interrupted Time-Series Analysis of Monthly Fall-Related Injury Among MA and Commercially Insured Enrollees (Comparison Group) From 2010 to 2015.
Group by agea | Before coverage (2010–2012) | After coverage (2013–2015) | Before vs after coverage | ||||||
---|---|---|---|---|---|---|---|---|---|
Fall-related injury per 10 000 person-days (95% CI)b | Rate (95% CI)c | P valuec | Fall-related injury per 10 000 person-days (95% CI)b | Rate (95% CI)c | P valuec | Rate change (95% CI) | P value | P value for MA vs commercial rate change | |
65-69 y | |||||||||
MA (n = 3 547 378) | 1.34 (1.31 to 1.37) | 0.00 (−0.01 to 0.02) | .76 | 1.38 (1.33 to 1.42) | 0.06 (0.03 to 0.09) | .001 | 0.05 (0.02 to 0.09) | .002 | .06 |
Comparison (n = 815 567) | 1.44 (1.41 to 1.48) | −0.02 (−0.04 to 0.00) | .05 | 1.40 (1.34 to 1.45) | −0.02 (−0.05 to 0.02) | .33 | 0.01 (−0.03 to 0.04) | .76 | |
70-74 y | |||||||||
MA (n = 1 668 561) | 1.50 (1.47 to 1.54) | 0.02 (0.00 to 0.04) | .10 | 1.56 (1.52 to 1.59) | 0.05 (0.02 to 0.07) | <.001 | 0.03 (0.00 to 0.06) | .08 | .76 |
Comparison (n = 211 269) | 1.62 (1.58 to 1.67) | 0.00 (−0.02 to 0.03) | .91 | 1.57 (1.50 to 1.64) | 0.04 (0.00 to 0.08) | .09 | 0.04 (−0.01 to 0.09) | .13 | |
75-79 y | |||||||||
MA (n = 1 211 833) | 1.84 (1.80 to 1.89) | 0.03 (0.00 to 0.05) | .07 | 1.95 (1.90 to 1.99) | 0.06 (0.03 to 0.09) | .001 | 0.03 (−0.01 to 0.07) | .13 | .92 |
Comparison (n = 107 596) | 1.98 (1.91 to 2.06) | −0.03 (−0.07 to 0.02) | .24 | 2.02 (1.90 to 2.13) | 0.00 (−0.07 to 0.07) | .99 | 0.03 (−0.06 to 0.11) | .51 | |
≥80 y | |||||||||
MA (n = 1 439 617) | 3.20 (3.15 to 3.24) | 0.04 (0.01 to 0.06) | .006 | 3.30 (3.23 to 3.37) | 0.16 (0.11 to 0.20) | <.001 | 0.12 (0.07 to 0.17) | <.001 | .04 |
Comparison (n = 118 640) | 3.12 (2.99 to 3.25) | 0.04 (−0.04 to 0.11) | .36 | 3.51 (3.39 to 3.63) | 0.03 (−0.05 to 0.11) | .47 | −0.01 (−0.11 to 0.10) | .91 |
Abbreviation: MA, Medicare Advantage.
Sample size reflects total number from 2010 to 2015.
Data from the first month of the period.
All rate, rate changes, and P values reported were obtained from regression models that included fixed effects for month. The rate presents the change per month in the injury rate.