Disclosures Funded, in part, by a research grant from American - PDF document

APNA 29th Annual Conference Session 4016: October 31, 2015 Motivational Interviewing within Compliance/Adherence Therapy to Improve Outcomes of People with Severe Mental Illness: A Meta-Analysis Choochart Wong-Anuchit, PhD(Candidate), RN Joanne Kraenzle Schneider, PhD, RN Mary M. Krieger, MLIS, RN Disclosures Funded, in part, by a research grant from American Psychiatric Nursing Foundation Conducted in partial fulfillment of a 3-manuscript, non-traditional doctoral dissertation at Saint Louis University Presentation Objectives • Understand the value of meta-analytic research • Interpret meta-analytic results to apply in clinical practice • Recognize how to implement motivational interviewing counseling in clinical practice Wong-Anuchit 1

APNA 29th Annual Conference Session 4016: October 31, 2015 Research Synthesis Defined • Research synthesis, research review, systematic review • No consensus in definition • Research synthesis – evaluative review of research • Primary focus and goal – Integrate empirical research to: • Create generalizations (implies limits) • Attend to relevant theories, research, conflicts, and central issues Meta-Analysis: A Brief Introduction Primary source: Borenstein, M., Hedges, L. V., Higgins, P. T., & Rothstein, H. R. (2009). Introduction to Meta-analysis. West Sussex, UK: John Wiley & Sons. Vote Counting “The process of counting the number of studies that are statistically significant and comparing this with the number that are not statistically significant.” “Vote counting is never a valid approach.” Borenstein et al. (2009) page 255 Wong-Anuchit 2

APNA 29th Annual Conference Session 4016: October 31, 2015 Logic of Vote Counting • Significant finding = effect ≠ • Nonsignificant finding effect But • No statistical significance might mean low statistical power Meta-Analysis Defined  Meta-analysis is the quantitative synthesis of primary studies that are essential for informing evidence-based practice and policy.  A meta-analytic study can provide an overall effect size, provide insights for clinical practice related to outcomes, and point to areas for future research. Meta-Analysis Defined • Used as a synonym for research synthesis • Includes quantitative procedures to statistically combine the results of studies Wong-Anuchit 3

APNA 29th Annual Conference Session 4016: October 31, 2015 Meta-analysis • Synthesizes data from a series of studies • Can often identify the true effect • If true effect is substantial and consistent, nonsignificant findings are due to low power. • Statistically sound to test the null hypothesis • Can assess magnitude of effect and consistency across primary studies How Meta-Analysis Works Individual studies Summary effect • Effect size (ES) • Effect size (ES) • Precision • Precision • Study weight • p -value • p -values Basic points • Compute ES and variance for each study; then weighted mean of each ES • Weighted mean: Assign more weight to more precise studies Individual Studies • Effect size (ES): A value that reflects the magnitude of the treatment effect; strength of a relationship between 2 variables (square in fig.) o ES based on ratios – a ratio of 1.0 = no difference between groups o ES based on mean difference – 0.0 = no difference • Precision: Confidence Interval (CI) o Primarily driven by sample size • Study weights: Good precision, higher weight o Solid square sizes reflect weight • p -value: Under .05 if 95% CI does not include null value Adapted from Borenstein et al. (2009) page 4 Wong-Anuchit 4

APNA 29th Annual Conference Session 4016: October 31, 2015 Summary Effect  Effect size: The weighted mean of the individual effects (bottom line in figure) – depends on assumptions  Fixed-effect model: All studies share the same true effect true effect size, summary is common ES  Random-effects model: True ES vary from study to study, summary is estimate of mean of ES distributions.  Precision: variance, SE, CI: diamond width in fig.  p -value: Reflects the magnitude of summary ES and the volume of information on which estimate is based. Adapted from Borenstein et al, (2009) page 4 Why Meta-analysis?  Provides context  Provides power  Provides the magnitude of effect  Summary effect size  Capable of examining inconsistency across studies Wong-Anuchit 5

APNA 29th Annual Conference Session 4016: October 31, 2015 Research Background • Motivational interviewing (MI) is a popular intervention. • Therapeutic behavioral change occurs by engaging individuals to argue for change (Miller & Rollnick, 2013) • MI is a purposive counseling style to resolve ambivalence for change. Research Background • Motivational interviewing (MI) • Widely used with compliance/adherence therapy • To improve adherence in people with severe mental illness. • No published meta-analysis • Examining MI effectiveness in people with severe mental illness. Motivational Interviewing  Client-centered, directive approach for enhancing intrinsic motivation to change by exploring and resolving ambivalence Wong-Anuchit 6

APNA 29th Annual Conference Session 4016: October 31, 2015 Motivational Interviewing  Client-centered or person-centered-focuses on person’s interests and concerns  Consciously directive—to resolve ambivalence  Method of communication—not a set of techniques  Elicits intrinsic motivation for change  Exploring and resolving ambivalence—has to be in person’s inherent interest Purpose  To synthesize studies that test the effects of motivational interviewing interventions on three primary outcomes reflecting treatment adherence: o Symptoms o Medication attitude o Functioning Method: Literature Search • Comprehensive database search • Identify eligible studies • Publication years: through the end of December 2014 PsycINFO o Ovid MEDLINE o SCOPUS, CINAHL o Web of Science o Cochrane Library o Sociological Abstracts (ProQuest) o Social Work Abstracts (EBSCO Host) o ProQuest Dissertations and Theses o Full Text without date restriction o Wong-Anuchit 7

APNA 29th Annual Conference Session 4016: October 31, 2015 Method: Statistics  Using random-effects models, summary effect sizes with 95% confidence intervals were computed using Hedges’ g.  Two-tailed p -values for the Z tests were used to test the null hypotheses that the mean effects were equal to zero. Results  Retrieved and coded primary studies  Studies (s) = 16  Adults with severe mental illness  At least 5 participants per group  Researchers used motivational interviewing (MI) within compliance/ adherence therapy. Results  MI within compliance/adherence therapy significantly improved psychotic symptoms  g = –0.353  Confidence interval: –0.662, –0.044; p = .025  s = 15, N = 1153  Global functioning  g = 0.653  CI: –0.004, 1.310; p = .051  s = 6, N = 385 Wong-Anuchit 8

APNA 29th Annual Conference Session 4016: October 31, 2015 Results  MI within compliance/adherence therapy significantly improved psychotic symptoms  No improvement in attitude  g = 0.175  Confidence interval: –0.046, 0.395; p =.120  s = 14, N = 1058  All studies were significantly heterogeneous warranting follow-up moderator analyses. Conclusion • Motivational interviewing (MI) within compliance/adherence therapy • Improves symptoms and functioning, • Informs psychiatric-mental health nurses about using MI in clinical practice to improve treatment adherence outcomes. Conclusion • Future research to provide additional insight • Moderator analyses of • Participant • Design • Intervention characteristics Wong-Anuchit 9

APNA 29th Annual Conference Session 4016: October 31, 2015 References Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta analysis. West Sussex, UK: John Wiley & Sons. Cooper, H. M. (2010). Research synthesis and mata-analysis: A step-by-step approach (4 th ed.). Washington, DC: SAGE Publications. Miller, W. R., & Rollnick, S. (2013). Motivational Interviewing: Helping people change (3 rd ed.). New York, NY: Guilford Press.  Wong-Anuchit 10