Publicatie

‘Australische medische schapenvacht’ ter preventie van decubitus: meta-analyse met individuele patiëntendata toont effectiviteit.

Mistiaen, P., Jolley, D.J., McGowan, S., Hickey, M.B., Spreeuwenberg, P., Francke, A.L. ‘Australische medische schapenvacht’ ter preventie van decubitus: meta-analyse met individuele patiëntendata toont effectiviteit. Nederlands Tijdschrift voor Geneeskunde: 2011, 155(18), p. A 3034.
Read online
OBJECTIVE: To assess the effectiveness of the Australian Medical Sheepskin (AMS) on the prevention of sacral pressure ulcers based on combined data from 3 previously published trials. DESIGN: Meta-analysis of randomised clinical trials (RCTs). METHOD: The effectiveness of these specially treated sheepskins from the Merino sheep was studied in three RCT's (2 involving Australian hospital patients and 1 involving Dutch nursing-home patients). In all 3 trials patients were randomized to 1 of 2 groups: a control group, who received standard care, and an intervention group, who, in addition to standard care, had at least one AMS placed under the sacral area when in bed. The total research population consisted of 1,281 patients from 11 institutions with 45 nursing wards. Data from the 3 RCT's were pooled and analysed: (a) with conventional meta-analysis based on the published effect sizes and (b) with multilevel logistic regression analysis based on the combined individual patient data for each trial separately and for the combined dataset. The primary outcome measure was the incidence of sacral pressure ulcers. RESULTS: The incidence of sacral pressure ulcers was 12.2% in the control group versus 5.4% in the intervention group. The odds-ratio was 0.37 (95% CI: 0.17-0.77) with conventional meta-analysis, and 0.35 (95% CI: 0.23-0.55) with multilevel logistic regression analysis on the combined individual patient data. CONCLUSION: Both analysis methods confirm that the AMS is effective in preventing sacral pressure ulcers. The confidence interval in the multilevel logistic regression analysis on individual patient data was almost 50% smaller than in the conventional meta-analysis, and gave a more reliable picture of the actual effect.(aut. ref.)