Summary of findings 1. Implementation interventions compared with no intervention for promoting uptake of evidence‐based practices in stroke rehabilitation.
| Implementation interventions compared with no intervention for promoting uptake of evidence‐based practices in stroke rehabilitation | ||||||
|
Patient or population: healthcare professionals providing stroke rehabilitation Settings: inpatient, outpatient or community rehabilitation Intervention: any implementation intervention Comparison: control (no intervention) | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No of participants (studies, clusters) | Quality of the evidence (GRADE) | Comments | |
| Assumed risk | Corresponding risk | |||||
| No intervention | Implementation intervention | |||||
| Quality of care: healthcare professional adherence to EBP at 12 months | Study population stroke patients | RR 1.19 (0.53 to 2.64) | 1455 patient participants (2 trials, 39 clusters) | ⊕⊝⊝⊝ Very lowc,d | We are uncertain about the estimate of healthcare professional adherence to EBP at 12 months as the certainty of the evidence is very low. | |
| 5%a | 6% (2.7% to 13.2%) | |||||
| 33%b | 39.3% (17.5% to 87.1%) | |||||
| Patient adherence to recommended treatment: number of outdoor journeys per week at 6 months | 7.4 | 7.9 (5.6 to 10.2) | Adjusted MD 0.5 (–1.8 to 2.8) | 100 participants (1 trial, 21 clusters) | ⊕⊕⊝⊝ Lowe | Implementation interventions in stroke rehabilitation may lead to little or no difference in patient adherence to recommended treatment at 6 months, compared with no intervention. |
| Patient health status: HRQoL (EQ‐5D: –0.59 to 1, higher score better) at up to 6 months | 0.58f | 0.59 (0.56 to 0.63) | MD 0.01 (–0.02 to 0.05) | 1242 participants (2 trials, 65 clusters) | ⊕⊕⊕⊝ Moderateg | Implementation interventions in stroke rehabilitation probably lead to little or no difference in patient HRQoL at up to 6 months compared with no intervention. 4 trials assessed HRQoL using EQ‐5D but data from only 2 trials could be pooled. Findings across studies appeared consistent. |
| Patient health status: ADL (Barthel Index, 0–20, higher score = better) at up to 6 months | 15.8f | 16.09 (15.64 to 16.53) | MD 0.29 (–0.16 to 0.73) | 1272 participants (2 trials, 65 clusters) | ⊕⊕⊕⊝ Moderateg | Implementation interventions in stroke rehabilitation probably lead to little or no difference in patient function (ADL) at up to 6 months compared with no intervention. 4 trials assessed ADL using the Barthel Index but data from only 2 trials could be pooled. Findings across studies appeared consistent. |
| Patient health status: psychological well‐being (GHQ‐12, 0–36, higher score = worse) at up to 6 months | 14.9h | 14.69 (14.36 to 15.40) | SMD –0.02 (–0.54 to 0.50) | 1273 participants (2 trials, 65 clusters) | ⊕⊕⊝⊝ Lowi | Implementation interventions in stroke rehabilitation may lead to little or no difference in patient psychological well‐being at up to 6 months compared with no intervention. 3 trials reported psychological well‐being using different measures, data from 2 trials could be pooled. Findings across studies appeared consistent. |
| Health professional outcomes | No studies reported this outcome | |||||
| *The basis for the assumed risk (e.g. the mean control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ADL: activities of daily living; CI: confidence interval; EBP: evidence‐based practice; EQ‐5D: EuroQol 5‐dimension health state measure; GHQ‐12: General Health Questionnaire‐12; HRQoL: health‐related quality of life; MD: mean difference; SMD: standardised mean difference; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate. | ||||||
a Adherence estimated from the control group value at 12 months in McCluskey 2016.
b Adherence estimated from the control group value at 12 months in Power 2014.
c Downgraded one level due to serious risk of bias; lack of blinding of personnel in both trials, outcome assessors not blinded and incomplete outcome data in one trial. d Downgraded two levels due to very serious imprecision; 95% confidence intervals wide. e Downgraded two levels due to very serious imprecision; suboptimal information size (one study with 100 participants) and 95% confidence intervals wide. f Estimated from the control group value at six months in Forster 2015. g Downgraded one level due to serious imprecision; suboptimal information size. h Estimated from the control group value on the GHQ‐12 at six months in Forster 2015. i Downgraded two levels due to serious indirectness; differences in outcome measures and uncertainty in whether the outcomes are assessing the same health issue.