Table 2.
Longitudinal analysis of association of recent readmission and cumulative readmissions with changes in physical functioning (N=566) to 18 months post-hospitalization.
| aAcute and Longer-term Associations Estimated Separately | bAcute and Longer-term Associations Estimated Simultaneously | |
|---|---|---|
| SAGES Physical Function Composite | Mean Difference (95% CI) |
Mean Difference (95% CI) |
| Hospitalized since last study visit | −2.8 (−3.8, −1.8) |
−1.4 (−2.5, −0.2) |
| Cumulative readmissions to date | ||
| 0 | Referent | Referent |
| 1 | −2.1 (−3.2, −1.004) |
−1.5 (−2.6, −0.4) |
| 2+ | −4.7 (−6.2, −3.3) |
−4.0 (−5.6, −2.4) |
|
| ||
| Instrumental Activities of Daily Living; one or more dependencies |
cRelative Risk (95% CI) |
Relative Risk (95% CI) |
| dHospitalized since last study visit | 1.5 (1.4, 1.7) |
1.2 (0.99, 1.4) |
| Cumulative readmissions to date | ||
| 0 | Referent | Referent |
| 1 | 1.5 (1.3, 1.8) |
1.4 (1.2, 1.7) |
| 2+ | 2.0 (1.7, 2.5) |
1.8 (1.5, 2.3) |
|
| ||
| Activities of Daily Living; one or more dependencies | Relative Risk (95%CI) |
Relative Risk (95%CI) |
| Hospitalized since last study visit | 2.1 (1.4, 3.0) |
1.3 (0.9, 2.0) |
| Cumulative readmissions to date | ||
| 0 | Referent | Referent |
| 1 | 1.9 (1.3, 2.8) |
1.6 (1.0, 2.5) |
| 2+ | 4.1 (2.3, 7.2) |
3.3 (1.7, 6.4) |
|
| ||
| MOS SF-12 Physical Component Summary | Mean Difference (95% CI) |
Mean Difference (95% CI) |
| Hospitalized since last study visit | −2.3 (−3.2, −1.4) |
−1.5 (−2.6, −0.4) |
| Cumulative readmissions to date | ||
| 0 | Referent | Referent |
| 1 | −1.2 (−2.2, −0.3) |
−0.7 (−1.7, 0.4) |
| 2+ | −3.3 (−4.6, −1.9) |
−2.4 (−4.0, −0.9) |
Analyses based on 503 hospital readmissions. MOS SF-12 – Medical Outcomes Study Short Form-12; CI – Confidence Interval. Models adjust for participants’ baseline age, sex, race/ethnicity, years of education, body mass index, Charlson comorbidity index, and surgery type. The time trend is allowed to vary freely in a nonlinear fashion from baseline through the 1 and 2 month visits, and assumed to follow a linear course thereafter, consistent with exploratory analyses (Figure 1).
Recent hospitalization and cumulative history of hospitalization considered in separate models
Recent hospitalization and cumulative history of hospitalization considered in one combined model (i.e. controlling for one another)
Multiplicative increase in the probability of ADL or IADL dependency vis-à-vis referent category
Referent: no hospitalization at current visit