Table 4. Rates and Trends in Outcome Measures Among High-Quality Home Health Agencies, 2013-2019.
Outcome | Parametera | Interpretation | Patient quality of care | Patient satisfaction | Both star ratings | |||
---|---|---|---|---|---|---|---|---|
Estimate (SE) | P value | Estimate (SE) | P value | Estimate (SE) | P value | |||
Hospitalization rate during spellb | β1 | Preexposure trend | −0.06 (0.08) | .51 | −0.07 (0.07) | .42 | −0.07 (0.09) | .49 |
β2 | Postexposure level change | −0.32 (0.17) | .16 | −0.35 (0.15) | .10 | −0.18 (0.18) | .39 | |
β3 | Postexposure trend change | 0.87 (0.09) | .00 | 0.56 (0.08) | .01 | 0.94 (0.10) | .00 | |
β1 + β3 | Postexposure trend | 0.81 (0.05) | <.001 | 0.49 (0.04) | <.001 | 0.87 (0.05) | <.001 | |
Hospitalization rate within 30 d of dischargeb | β1 | Preexposure trend | 0.12 (0.08) | .22 | 0.13 (0.08) | .19 | 0.10 (0.08) | .30 |
β2 | Postexposure level change | −0.01 (0.16) | .94 | −0.10 (0.17) | .61 | −0.01 (0.17) | .95 | |
β3 | Postexposure trend change | 0.04 (0.09) | .71 | −0.03 (0.09) | .78 | 0.03 (0.09) | .80 | |
β1 + β3 | Postexposure trend | 0.15 (0.05) | .00 | 0.10 (0.05) | .03 | 0.12 (0.05) | .01 | |
Timely initiation of carec | β1 | Preexposure trend | −0.57 (0.19) | .06 | −0.65 (0.22) | .06 | −0.45 (0.27) | .19 |
β2 | Postexposure level change | 1.03 (0.41) | .09 | 2.31 (0.48) | .02 | 1.23 (0.58) | .12 | |
β3 | Postexposure trend change | −0.97 (0.23) | .02 | −0.79 (0.26) | .06 | −1.02 (0.32) | .05 | |
β1 + β3 | Postexposure trend | −1.54 (0.12) | <.001 | −1.44 (0.14) | <.001 | −1.47 (0.17) | <.001 | |
Improvement in ambulation | β1 | Preexposure trend | 2.64 (0.51) | .01 | 3.05 (0.52) | .01 | 2.65 (0.51) | .01 |
β2 | Postexposure level change | 2.65 (1.10) | .09 | 3.38 (1.12) | .06 | 2.81 (1.09) | .08 | |
β3 | Postexposure trend change | −0.72 (0.60) | .32 | −0.65 (0.62) | .37 | −0.73 (0.60) | .31 | |
β1 + β3 | Postexposure trend | 1.93 (0.32) | <.001 | 2.40 (0.33) | <.001 | 1.92 (0.32) | <.001 | |
Improvement in bed transferring | β1 | Preexposure trend | 3.14 (0.84) | .03 | 3.67 (0.87) | .02 | 3.20 (0.81) | .03 |
β2 | Postexposure level change | 3.54 (1.82) | .15 | 3.44 (1.87) | .16 | 3.54 (1.74) | .14 | |
β3 | Postexposure trend change | 0.17 (1.00) | .87 | 0.30 (1.02) | .79 | 0.13 (0.95) | .90 | |
β1 + β3 | Postexposure trend | 3.32 (0.53) | <.001 | 3.96 (0.55) | <.001 | 3.32 (0.51) | <.001 | |
Improvement in bathing | β1 | Preexposure trend | 1.60 (0.27) | .01 | 2.05 (0.34) | .01 | 1.68 (0.27) | .01 |
β2 | Postexposure level change | 1.35 (0.58) | .10 | 2.11 (0.74) | .06 | 1.60 (0.59) | .07 | |
β3 | Postexposure trend change | 0.37 (0.32) | .33 | 0.31 (0.40) | .49 | 0.28 (0.32) | .45 | |
β1 + β3 | Postexposure trend | 1.97 (0.17) | <.001 | 2.37 (0.22) | <.001 | 1.96 (0.17) | <.001 |
Parameters were estimated from an interrupted time series regression model of Y = α + β1T + β2X + β3XT + ε, where t = linear time and X = post–star ratings period. The pre–star ratings period was from January 2013 to December 2015 and the post–star ratings period was from January 2016 to December 2019.
Hospitalization rate variables included inpatient hospitalizations, readmissions, and outpatient observational visits.
Timely initiation of care refers to home health spells that started within 2 days of an inpatient discharge date.