|
Health care provider-delivered adherence promotion interventions: a meta-analysis |
Wu YP, Pai AL |
|
|
CRD summary This review concluded that health care provider-delivered adherence interventions for children with chronic illness and their families appeared to be effective, particularly immediately after intervention. The differences across trials, the small-to-medium effect sizes, and the potential for bias, make it difficult to determine the reliability of the authors' conclusions, which should be interpreted with caution. Authors' objectives To assess the effectiveness of health care provider-delivered interventions to improve treatment adherence among children with a chronic illness. Searching Six electronic databases, including PubMed, CINAHL, and Scopus, were searched in 2012, for relevant studies in English. The search strategy was presented. The reference lists from relevant reviews and articles were manually screened. Study selection Eligible for inclusion were randomised controlled trials (RCTs) of paediatric interventions to increase adherence to primary self-managed behavioural interventions for a chronic illness. Participants had to be patients younger than 18 years old or their family. At least half of the intervention content or time had to be delivered by at least one health care provider. Trials had to provide sufficient data to calculate the effect size. Trials were excluded if they were delivered at schools or summer camps or more than half of the intervention was provided by a mental health professional. Most of the included trials were of youths with asthma, others were of youths with diabetes or obesity. Some trials included parents. Just over half of the participants were male, with a mean age of seven years. Interventions targeted behaviour, such as taking medication, symptom monitoring, diet, or physical activity. They varied across trials, including providing families with strategies to manage treatment, education, or pillboxes or calendars for self-monitoring. Generally there were seven sessions or contacts with the health care providers. Settings varied, with delivery to individual patients or families, or groups. Most interventions were not explicitly based on a theory, such as the Health Belief Model. Most comparators were treatment as usual; others were an alternative active intervention, or an attention placebo (education booklet). Outcomes were measured using various methods; patient or parent reports, pharmacy records, or biological assay. Two reviewers independently screened trials for inclusion. Assessment of study quality Trial quality was assessed using the Cochrane risk of bias tool, except the criterion for blinding of patients. It was unclear how many reviewers were involved in the assessment. Data extraction Means and standard deviations were extracted before and after intervention, and at follow-up, to calculate effect sizes. It was unclear how many reviewers extracted the data. Methods of synthesis A random-effects model was used to combine the data and calculate a summary effect size (Cohen's d) after treatment and at follow-up. Where trials reported more than one effect size, the average was calculated and used in the analyses. Further analyses were conducted, excluding interventions involving a behavioural health provider, and of interventions based on theory versus those not based on theory. Analyses were performed by type of adherence behaviour after treatment and at follow-up. Prespecified subgroup analyses were performed based on intervention type, number of health care providers, setting, format, and adherence outcome measure. Statistical heterogeneity was assessed using Cochran's Q. Publication bias was assessed using Rosenthal's fail-safe N. Results of the review Thirty-five RCTs (4,616 participants; mean 125) were included in the review. Nine trials met all the risk of bias criteria. The main issues were with unclear allocation concealment. The results were fully reported. After treatment, participants receiving the health care provider intervention showed greater adherence than other interventions, with a medium effect size (Cohen's d=0.49, 95% CI 0.32 to 0.66; 35 RCTs). At follow-up, the treatment effect was small (Cohen's d=0.32, 95% CI 0.10 to 0.54; 10 RCTs). There was evidence of significant statistical heterogeneity for both analyses, but this was not investigated further. Effect sizes after treatment differed across the adherence behaviour measured, ranging from small (d=0.11) to large (d=0.61). Other findings were reported in the review. Rosenthal's fail-safe N suggested that 2,643 unpublished studies would be needed to reduce the overall effect size. Authors' conclusions Health care provider-delivered interventions for children with a chronic illness and their families appeared to improve adherence, with the greatest gains immediately after intervention, but further research was needed. CRD commentary The review question and supporting inclusion criteria were clearly stated. Six databases were searched for relevant articles, but as this was restricted by language, potentially relevant trials may have been missed. The authors formally assessed publication bias using Rosenthal's fail-safe N, but this method is not recommended by The Cochrane Collaboration and the findings should be interpreted with caution. Trial quality was assessed using appropriate criteria; most trials were at some risk of bias and this was not considered in the analyses. Two reviewers assessed trials for inclusion in the review, but the authors did not explicitly state whether this was the case for quality assessment and data extraction. Reviewer error and bias cannot be ruled out. The evidence was fairly extensive, but patient and intervention details were summarised rather than reported for each trial. Given the variability across trials, the methods used to analyse the data seem appropriate, but the main results were presented only as a forest plot. The authors attempted to investigate statistical heterogeneity, but the underlying causes were not identified. The authors stated that the time between before and after treatment, and follow-up varied across trials, but the extent of variation was unclear. The authors acknowledged several additional limitations to the evidence, including a focus on children with asthma and younger children, and variations in methods and outcome measures, some of which could overestimate the effects. The findings suggested improvements in adherence with health care provider-delivered interventions, but these improvements reduced over time, which the authors acknowledged. The differences across trials, the small-to-medium effect sizes, and the uncertainty in review methods, make it difficult to determine the reliability of the authors' conclusions. These conclusions should be interpreted with caution. Implications of the review for practice and research Practice: The authors stated that health care provider-delivered interventions could be more effective for certain adherence behaviour, such as taking medication. Other behaviour, such as making dietary and physical activity changes might require more intensive interventions or delivery of interventions by multidisciplinary providers. Research: The authors stated that research was needed to focus on testing interventions for other specific chronic illnesses, including an attention control or active comparator, and assessing outcomes using more robust methods. Research should address adherence issues in older youths, and the effects of multidisciplinary interventions. The authors made further recommendations for research. Funding Funded by the National Institutes of Health, USA. Bibliographic details Wu YP, Pai AL. Health care provider-delivered adherence promotion interventions: a meta-analysis. Pediatrics 2014; 133(6): e1698-e1707 Indexing Status Subject indexing assigned by NLM MeSH Adolescent; Child; Chronic Disease /psychology /therapy; Female; Follow-Up Studies; Health Promotion /methods; Humans; Male; Medication Adherence /psychology; Nurse-Patient Relations; Patient Care Team; Patient Compliance /psychology; Physician-Patient Relations; Professional-Patient Relations; Randomized Controlled Trials as Topic; Self Care /psychology; Treatment Outcome AccessionNumber 12014030964 Date bibliographic record published 27/06/2014 Date abstract record published 09/09/2014 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. |
|
|
|