Table 4.
Random effects meta-analysis for outcomes stratified by study characteristics in 17 trials of inpatient rehabilitation specifically designed for geriatric patients. Figures are odds ratios (for functional improvement) or relative risks (for nursing home admissions and mortality) with 95% confidence intervals
| Study characteristics | Short term (at discharge) | Longer term (at 3-12 month follow-up) | |||||
|---|---|---|---|---|---|---|---|
| Functional improvement | Nursing home admission | Mortality | Functional improvement | Nursing home admission | Mortality | ||
| Type of intervention programme: | |||||||
| General geriatric rehabilitation | 1.34 (1.12 to 1.60) | 0.53 (0.33 to 0.86) | 0.76 (0.54 to 1.06) | 1.02 (0.86 to 1.21) | 0.90 (0.71 to 1.13) | 0.88 (0.75 to 1.04) | |
| Orthopaedic geriatric rehabilitation | 2.33 (1.62 to 3.34), P=0.04* | 0.72 (0.56 to 0.91) | 0.66 (0.42 to 1.04) | 1.79 (1.24 to 2.60), P=0.01* | 0.79 (0.61 to 1.02) | 0.77 (0.61 to 0.96) | |
| Mean age of study population (years): | |||||||
| ≤80 | 1.88 (1.19 to 2.97) | 0.42 (0.27 to 0.64) | 1.01 (0.47 to 2.16) | 1.26 (0.87 to 1.82) | 0.82 (0.58 to 1.16) | 0.85 (0.62 to 1.16) | |
| >80 | 1.74 (1.05 to 2.88) | 0.75 (0.58 to 0.96), P=0.045* | 0.68 (0.50 to 0.92) | 1.45 (1.08 to 1.94) | 0.88 (0.72 to 1.07) | 0.82 (0.70 to 0.95) | |
| Intervention programme: length of hospital stay after randomisation in intervention group: | |||||||
| ≤21 days | 2.38 (1.53 to 3.70) | 0.67 (0.42 to 1.09) | 0.59 (0.29 to 1.22) | 1.43 (1.00 to 2.05) | 0.99 (0.69 to 1.42) | 0.84 (0.67 to 1.05) | |
| >21 days | 1.52 (1.08 to 2.13) | 0.61 (0.45 to 0.83) | 0.76 (0.56 to 1.04) | 1.30 (0.95 to 1.78) | 0.81 (0.67 to 0.98) | 0.85 (0.73 to 0.99) | |
| Outpatient follow-up after inpatient rehabilitation for patients in intervention group: | |||||||
| Yes | — | — | — | 1.49 (0.93 to 2.39) | 0.76 (0.64 to 0.92) | 0.84 (0.61 to 1.16) | |
| No/NR | — | — | — | 1.29 (1.00 to 1.66) | 0.88 (0.69 to 1.12) | 0.82 (0.71 to 0.96) | |
| Length of follow-up for outcome evaluation (months): | |||||||
| ≤6 | — | — | — | 1.44 (0.94 to 2.21) | 0.82 (0.67 to 1.00) | 0.83 (0.66 to 1.05) | |
| >6 | — | — | — | 1.32 (0.99 to 1.76) | 0.90 (0.68 to 1.19) | 0.85 (0.73 to 0.99) | |
| Intervention programme quality: intervention included initial multidimensional geriatric assessment: | |||||||
| Yes | 1.75 (1.31 to 2.35) | 0.60 (0.46 to 0.78) | 0.75 (0.57 to 0.99) | 1.36 (1.07 to 1.71) | 0.81 (0.69 to 0.94) | 0.86 (0.77 to 0.97) | |
| No/NR | NA | 0.84 (0.54 to 1.29) | 0.40 (0.13 to 1.18) | NA | 1.23 (0.75 to 2.02) | 0.90 (0.61 to 1.33) | |
| Intervention programme quality: intervention included patient assignment (to therapies and goal setting): | |||||||
| Yes | 1.81 (1.23 to 2.67) | 0.54 (0.37 to 0.78) | 0.68 (0.39 to 1.19) | 1.32 (0.90 to 1.92) | 0.88 (0.62 to 1.26) | 0.83 (0.68 to 1.00) | |
| No/NR | 1.80 (0.96 to 3.37) | 0.71 (0.53 to 0.94) | 0.74 (0.53 to 1.03) | 1.40 (1.02 to 1.90) | 0.86 (0.71 to 1.04) | 0.85 (0.71 to 1.02) | |
| Methodological trial quality: concealed randomisation: | |||||||
| Yes | 1.61 (1.21 to 2.13) | 0.61 (0.45 to 0.83) | 0.75 (0.56 to 1.02) | 1.45 (1.02 to 2.06) | 0.78 (0.68 to 0.91) | 0.90 (0.79 to 1.03) | |
| No/NR | 2.91 (1.36 to 6.24) | 0.66 (0.40 to 1.07) | 0.54 (0.24 to 1.22) | 1.29 (0.99 to 1.68) | 1.05 (0.72 to 1.54) | 0.77 (0.60 to 0.98) | |
| Methodological trial quality: data assessment by an independent assessor: | |||||||
| Yes | 1.39 (1.17 to 1.65) | 0.59 (0.33 to 1.06) | 0.63 (0.31 to 1.26) | 1.34 (0.97 to 1.85) | 0.81 (0.68 to 0.95) | 0.92 (0.80 to 1.06) | |
| No/NR | 2.43 (1.47 to 4.00) | 0.63 (0.47 to 0.85) | 0.72 (0.48 to 1.07) | 1.40 (0.97 to 2.03) | 0.93 (0.66 to 1.30) | 0.76 (0.62 to 0.93) | |
NR=not reported, NA=not applicable (no studies in this category).
*Exact P values for significant results in meta-regression analyses (P<0.05). If result of meta-regression was not significant (P≥0.05), no P value is listed.