|
Implementing a hospital guideline on pneumonia: a semi-quantitative review |
Cortoos P J, Simoens S, Peetermans W, Willems L, Laekeman G |
|
|
CRD summary This review aimed to evaluate different guideline implementation interventions for improving the treatment of community-acquired pneumonia in a hospital setting. The authors concluded that describing which interventions are most successful is unlikely to be correct without taking other hospital-specific factors into account. Given the limitations in the evidence and review, these conclusions appear appropriately cautious, though non-specific. Authors' objectives To quantify the impact of different guideline implementation interventions to improve the treatment of community-acquired pneumonia (CAP) in a hospital setting. Searching PubMed, the Cochrane Library, the Cochrane EPOC Register, EMBASE and CINAHL were searched from inception to March 2006; the search terms were reported. In addition, the bibliographies of retrieved articles were examined for further relevant studies. Study selection Studies were considered eligible if they were hospital-based trials evaluating the effect of guidelines on compliance with care processes, clinical and/or economic outcomes in the treatment of CAP, and they also described their implementation interventions. Where the intervention was not clear, authors were contacted for additional details. The interventions evaluated in the selected studies included educational interventions and outreach, reminders, local opinion leader(s), audit and feedback, multi-participation, external steering and structural interventions.
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection. Assessment of study quality The methodological quality of the included studies was assessed according to criteria developed by the Cochrane Collaboration. These criteria addressed the potential for a range of biases: selection, performance, attrition and detection.
Two reviewers appear to have independently conducted the validity assessment, with any disagreements resolved by consensus. Data extraction The proportional change before and after the implementation of an intervention was calculated for every outcome. Then, a median of these changes was calculated for studies as a whole and for different outcome types of interest (care processes, clinical or economic outcomes). For the minority of controlled trials, these calculations were only carried out for the intervention arm of the trial, to allow comparison across studies. In addition, the presence or absence of audit and feedback, a local opinion leader, or quality improvement organisation was taken into account.
Two independent reviewers extracted the data. Methods of synthesis Correlations between the number or type of interventions and outcome variables were calculated using two-tailed Spearman's rho. For the binary variables, a two-tailed Mann-Whitney U test was used. Sensitivity analyses were conducted, both with and without outlying values. Results of the review A total of 30 publications from 27 different primary studies were included in the review.
The most frequently used interventions were educational meeting (63%; 17 studies) and distribution of written material (78%; 21 studies), or a combination of the two interventions (52%; 14 studies).
With one exception, all studies showed no change or an improvement (up to 103.6%) following implementation interventions. For processes of care and clinical outcomes, changes were highly variable. In general, improvements were seen for mortality, length of stay, economic outcomes and antibiotic guideline compliance.
There was no significant correlation between the total number of interventions and the median improvement on all outcome variables. Improvement in length of stay was significantly lower when audit and feedback were used (p=0.003; n=20), but most analyses showed no significant impact of implementation interventions on process outcomes, clinical and economic improvements, improvement of guideline concordance, length or stay or mortality. No difference was seen with or without the inclusion of outliers.
A significant positive correlation was detected between increased improvement of process outcomes and clinical outcomes (Spearman's rho 0.452; p=0.04, n=22). There was no significant correlation between improvement in compliance and median improvement on clinical outcome variables (Spearman's rho 0.551; p=0.16, n=8). Authors' conclusions Other hospital-specific factors are likely to have a higher impact on the rate of improvement than the implementation interventions alone. Describing which interventions are most successful is unlikely to be correct without taking these hospital-specific factors into account. CRD commentary This review was based on a question that was broadly defined in terms of the participants, interventions, study designs and outcomes of interest. A wide range of electronic databases and study bibliographies was searched for relevant studies. However, the search appears to have been limited to published studies and it is not clear if there were any language restrictions; the potential for publication and language biases cannot, therefore, be ruled out. In addition, the authors did not describe any efforts to minimise the potential for errors or bias in the study selection process. The validity of the included studies was assessed according to established criteria, and some aspects of this assessment were incorporated into the synthesis. Given the heterogeneity of the included studies, it seems appropriate not to have undertaken a meta-analysis. However, the method used meant that a very diverse set of outcomes was grouped together, precluding the ability to draw very specific conclusions. In addition, only the intervention arms of controlled studies contributed to the analysis, thereby reducing the value of incorporating these more rigorous study designs. The authors' conclusions therefore appear appropriately cautious in view of the limitations described, though it is unclear whether more specific conclusions may have been drawn in the absence of these limitations. Implications of the review for practice and research Practice: The authors did not state any implications for practice.
Research: The authors stated that future research should focus on how to identify and define hospital-specific factors and how to adapt the intervention to these factors. Bibliographic details Cortoos P J, Simoens S, Peetermans W, Willems L, Laekeman G. Implementing a hospital guideline on pneumonia: a semi-quantitative review. International Journal for Quality in Health Care 2007; 19(6): 358-367 Indexing Status Subject indexing assigned by NLM MeSH Community-Acquired Infections /therapy; Critical Pathways /standards; Guideline Adherence; Humans; Inservice Training; Pneumonia /therapy; Practice Guidelines as Topic; Quality Assurance, Health Care /methods AccessionNumber 12007004160 Date bibliographic record published 30/09/2008 Date abstract record published 23/12/2008 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|