Critique of Systematic ReviewsRebecca Ngassa
Date of submission
Systematic reviews are held as the highest level of evidence in research and are considered the cornerstone of evidence-based practice (EBP). A Systemic Research Review (SRR) addresses one of more research questions using identification, evaluation and integration towards relevant findings. Because research is constant and there may be conflicting findings, it is often necessary to conduct SRRs to close the gaps on some of the findings. Polit and Beck (2017) define systematic review as “a rigorous synthesis of research findings on a particular research question, using systematic sampling and data collection procedures and formal protocol” (p. 746). Sickle cell disease is a genetic disorder that results in chronic anemia, pain, disability, organ damage, increased risk for infection and early death. This paper will compare key components of two systematic reviews addressing “Medication adherence among pediatric patients with Sickle Cell Disease” (SCD), A Systematic Review” by Walsh, Cutrona, Kavanagh, Crosby, Malone, Lobner, and Bundy (2014) and “Systematic and Meta-Analytic Review” by Loiselle, Lee, Szulczewski, Drake, Crosy, and Pai (2016). Major Key components such as problem identification, literature search and figure, study quality evaluation, synthesis of evidence, conclusion of evidence, and literature table will be analyzed.
Critique of Problem Identification
Loiselle et al. (2016) splendidly stated that pediatric medication nonadherence leads to additional disease complications including death, increased emergency department visits, increased inpatient hospitalization, and increased financial burden. Using the Pediatric-Management Model as theoretical basis, the authors examined how patient family demographics, medical, and psychological factors along with healthcare system levels can negatively influence adherence and outcomes. The association of adherence rates by medication type and assessment method was examined between SCD medication adherence, demographic, psychological, and medical factors through meta-analysis.
Walsh et al. (2014) presented a concise and clear picture of their research problem and objective of study. Adherence was defined as “the extent to which a patient is taking his medication as prescribed by his healthcare providers” (p. 1176). They concisely stated that poor adherence increased risks for complications, increased health care costs, and decreased medication efficacy. The objective of this study was to describe adherence rates of varying medication types, identification of nonadherence to medication and patient characteristics, and showcase the effects of moderate adherence and nonadherence on clinical outcomes. In this study, no theoretical basis for the research problem were discussed. Whereas the background by Loiselle et al (2016) included theoretical basis, and thoroughly discussed the research problem behind the study. In contrast, Walsh et al’s (2014) concisely discussed the research question. To my opinion this aspect is redundant as the discussion of theory would have been an important and positive addition to their research, but this important aspect is lacking in both researches.
Critique of Literature Search and Figure
Databases searched by Loiselle et al. (2016) included PubMed; CINAHL, ERIC, Ovid, PsychINFO, EBSCOhost, EBMAR, and Scopus. Search dates included were all dates for PubMed and 1982 through February 2015 for the other databases. Eight key terms used for search were: medication compliance, sickle cell disease, adolescent, infant, child, treatment compliance, sickle cell anemia, and medication adherence. There was no information on the person performing the literature search but an informative and easy to read PRISMA four-phase flow diagram was used to show study selection details. This visualized the initial article search through screening and inclusion. Exclusion criteria was discussed in the text portion of the review and shown visually in the flow diagram.
Article search for Walsh et al. (2014) included systematic review of Cochrane, Scopus, Embase, Web of Science, PubMed, the World Health Organization Global Health Library from the year 1940 to February 2012. The key words used for search were not specifically mentioned in text, but the authors included specific medication names that are used for treatment, specific diseases pertaining to this study, and different terms for adherence. The search was conducted by a professional librarian without search restrictions. The flow diagram was present and visualized the process of initial abstract identification to inclusion into the final literature review. Exclusion criteria were shown on the flow diagram and were discussed in text.
Based on my review, both authors performed an extensive search using key words and a flow diagram. Loiselle et al. (2016) provided more detailed information in text and a quick reference via the flow diagram. Walsh et al (2014) on the other hand excluded abstracts for being “not English” and according to Khan, Kunz, Kleijnen, and Antes (2013), searches should not be language restricted.
Critique of Study Quality Evaluation
While both authors did adequate study evaluation, Loiselle et al. (2016) was superior to Walsh et al. (2014) in the sense that, three authors were used to collect and standardize data versus two authors. The Theory of Pediatric Self-Management Model was used to guide how correlates were selected and structured as opposed to the absence of theory discussion. All abstracts were reviewed by the first author and any discrepancies were resolved by consensus in comparison to rereview. Inclusion and exclusion criteria were discussed in both studies and shown in the flow diagrams. Result reliability, however, was assessed in Walsh et al. (2014) using the Newcastle-Ottowa Quality Assessment Scale. According to Polit and Beck (2017) the use of this tool increases study validity and reliability.
Critique of Synthesis of Evidence
According to Khan et al (2013), data analysis is a study of quality and effects, characteristics, and statistical methods for comparing studies and combining their effects through meta-analysis. Of the forty-nine studies that reported medications as a primary or secondary aim, twenty-two studies were included in the systematic review by Loiselle et al. (2016). Meta-analysis was performed by using a 22-item quality rating rubric and two independent coders rated the studies. Rating agreement was determined using Cohen’s kappa. Any coding disagreement was resolved by the first author. Inclusion criteria for meta-analysis was any study with association between medication adherence and a demographic from the Pediatric-Management Model.
As compared to Walsh et al. (2014) ninety-one articles were screened for inclusion and twenty-four studies were included in the review. There was no meta-analysis performed., and according to Polit & Beck (2017), “for systematic integration of quantitative evidence, meta-analysis offers a simple advantage: objectivity” (p. 648).
Critique of Conclusion of Evidence
Both authors did an excellent job in interpreting and summarizing their findings. Limitations and potential biases were discussed in each study. Impacts of the studies were discussed with regards to possible interventions for increasing medication adherence and improved patient health outcomes. Each author reiterated their research problem and reinforced how their data could impact medication adherence.
Study findings by Loiselle et al (2016) showed that adherence is likely to be higher when methods other than self-reporting are used. Provider attention is particularly important when they are prescribing penicillin and iron chelators because adherence is lower than other prescribed medications. Adolescents are of greater risk of adherence problems. Barriers to medication regimens should be assessed and given interventions to increase adherence.
Walsh et al. (2014) stated that route of administration influenced adherence. Injected antibiotics had better adherence than oral. It was also identified that moderate adherence might have some benefit than poor adherence, more objective measures for monitoring adherence should be used and multicomponent interventions are more effective than educational interventions alone.
Critique of Literature Review Table
Both authors included literature review tables that were informative, complete, and present. Any abbreviations or variables used where defined. Literature table comparison found Walsh et al (2014) to be formatted to ease data facilitation, it was less clustered, and was visually more appealing.
However, Loiselle et al. (2016) used numerical and statistical data displayed throughout three separate tables. Meta-analysis tables displayed clear results including confidence interval (CI) and effect size. A Forest plot of correlations and CIs was included to graphically display results. Walsh et al. (2014) did not perform a meta-analysis so there is no comparison with regards to meta-analysis by Loiselle et al. (2016). Conclusively, according to both research, medication adherence is distinctive among pediatric patients with SCD. Various interventions such as routine monitoring of medication, support to avoid and prevent mistakes and provide education to improve understanding of risks and benefits of medication adherence are necessary.
Khan, K. S., Kunz, R., Kleijnen, J., & Antes, G. (2014, March). Five steps to conducting a systematic review. Journal of The Royal Society of Medicine, 96, 118-121.Loiselle, K., Lee, J. L., Szulczewski, L., Drake, S., Crosby, L. E., & Pai, A. L. (2016). Systematic and meta-analytic review: Medication adherence among pediatric patients with sickle cell disease. Journal of Pediatric Psychology, 41(4), 406-418. https://doi.org/10.1093/jpepsy/jsv084Polit, D. F., & Beck, C. (2017). Nursing research: Generating and assessing evidence for nursing practice (10 ed.). Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins.Walsh, K. E., Cutrona, S. L., Kavanagh, P. L., Crosby, L. E., Malone, C., Lobner, K., & Bundy, D. G. (2014, December). Medication adherence among pediatric patients with sickle cell disease: A systematic review. Pediatrics, 134(6), 1175-1183. https://doi.org/10.1542/peds.2014-0177