Table 2.
Outcomes at 15 months
| Usual care group (N=749) | Intervention group (N=797) | Adjusted difference in means (95% CI); p value | ||
|---|---|---|---|---|
| Primary outcome* | ||||
| Unadjusted mean EQ-5D-5L (SE) | 0·504 (0·012) | 0·533 (0·012) | 0·00 (−0·02 to 0·02); p=0·93† | |
| Secondary outcomes | ||||
| Illness burden | ||||
| Self-rated health of good or better‡ | 230/631 (36%) | 242/642 (38%) | 0·84§ (0·67 to 1·05); p=0·13 | |
| Mean Bayliss measure of illness burden16 | 18·4 (12·9); n=590 | 16·7 (11·6); n=598 | −0·64¶ (−1·54 to 0·27); p=0·17 | |
| Mean HADS anxiety score17 | 6·3 (4·8); n=624 | 5·8 (4·7); n=629 | −0·24¶ (−0·57 to 0·08); p=0·15 | |
| Mean HADS depression score17 | 6·8 (4·6); n=625 | 6·1 (4·6); n=630 | −0·01¶ (−0·33 to 0·30); p=0·94 | |
| Treatment burden | ||||
| Mean Multimorbidity Treatment Burden Questionnaire score18 | 15·0 (17·1); n=626 | 12·9 (15·0); n=625 | −0·46¶ (−1·78 to 0·86); p=0·49 | |
| Mean eight-item Morisky Medication Adherence Score19 | 6·6 (1·3); n=749 | 6·7 (1·2); n=797 | 0·06¶ (−0·05 to 0·17); p=0·27 | |
| Median number of different drugs prescribed in past 3 months | 11·0 (8·0–15·0); n=736 | 11·0 (8·0–15·0); n=774 | 1·02‖ (0·97 to 1·06); p=0·46 | |
| Patient-centred care | ||||
| Mean PACIC score20 | 2·5 (0·9); n=512 | 2·8 (1·0); n=524 | 0·29¶ (0·16 to 0·41); p<0·0001 | |
| Mean CARE doctor score21 | 37·5 (10·0); n=601 | 40·2 (9·7); n=617 | 1·20¶ (0·28 to 2·13); p=0·0109 | |
| Mean CARE nurse score21 | 38·5 (9·5); n=462 | 40·8 (8·9); n=535 | 1·11¶ (0·03 to 2·19); p=0·044 | |
| Patients reporting they almost always discuss the problems most important to them in managing their own health‡ | 153/599 (26%) | 256/612 (42%) | 1·85§ (1·44 to 2·38); p<0·0001 | |
| Patients reporting that support and care is almost always joined-up‡ | 173/603 (29%) | 257/614 (42%) | 1·48§ (1·18 to 1·85); p=0·0006 | |
| Patients reporting being very satisfied with care‡ | 236/608 (39%) | 345/614 (56%) | 1·57§ (1·19 to 2·08); p=0·0014 | |
| Patients reporting having a written care plan, health plan, or treatment plan** | 91/623 (15%) | 141/623 (23%) | 1·97†† (1·32 to 2·95); p=0·0010 | |
Data are n/N (%), median (IQR), or mean (SD), unless otherwise indicated; treatment effects are presented as adjusted odds ratios, beta-coefficients, or incidence rate ratios (see footnotes). Use of the Morisky Medication Adherence Score is protected by US Copyright laws. Permission for use is required. A licence agreement is available from Donald E Morisky, MMAS Research LLC, 14725 NE 20th St, Bellevue, WA 98007, USA, or from dmorisky@gmail.com. HADS=Hospital And Depression Score.
Means calculated with multiple imputation by chain equations.
Intracluster correlation coefficient was 0·00 (95% CI 0·00–0·00).
Ordinal variable, dichotomised for ease of presentation; full details of question and responses available in appendix.
Adjusted odds ratio from multi-level ordinal logistic regression.
Beta-coefficients; analyses are adjusted multi-level linear regression models.
Incidence rate ratio from adjusted multi-level Poisson regression model.
Not prespecified in the statistical analysis plan; responses of “don't know” were treated as not having a care plan.
Adjusted odds ratio from multi-level logistic regression.