Impact of Shared Decision Making on Clinical, Humanistic, and Economic Outcomes Among Patients With Pain

Abstract

Introduction: Pain is a leading global public health issue and poses significant clinical, humanistic, and economic burden for patients. Chest pain is one of the most costly and common reasons why patients visit the emergency department in the United States. Clinicians must assess treatments and processes of care options while incorporating patients’ values and preferences. Since individual preferences vary, shared decision making (SDM) and use of decision aids (DAs) may be a useful process to promote patient-provider partnership and reach a preferable decision. The purpose of this study is to: describe findings from studies reporting impact of SDM/DAs in terms of clinical, economic, and/or humanistic outcomes among patients with pain; quantitatively synthesize outcomes data from identified studies to evaluate the effects of SDM/DAs in patients with pain; and, to evaluate the cost-effectiveness of the Chest Pain Choice decision aid (CPC-DA) in low-risk chest pain patients presenting to the emergency department. Methods: This study utilized the systematic review and meta-analysis and decision analytic cost-effectiveness analysis study designs. Medline, Embase, CINAHL, PsychINFO, and Cochrane library from inception to June 18, 2020. Additional sources of data included trial registries (World Health Organization, ClinicalTrials.gov), reference lists of included articles and relevant systematic reviews identified in the search. Studies comparing SDM and/or DA with usual care among patients with pain were included. Two investigators independently screened the articles identified in the search for inclusion, including title/abstract and full text review, and extraction. To compare the effect of shared decision making and/or use of decision aids on patient outcomes, in published literature in Objective 1, the following was conducted: data extraction into Excel, use of random effects model, calculation of the standardized mean difference for each study, and development of a forest plot to pool the findings. The Comprehensive Meta-Analysis software was used for the meta-analysis, when there were greater than three studies reporting data for a particular outcome. A decision analytic model was developed to compare costs and effectiveness of the CPC-DA and usual care from a payer perspective. Results from a multicenter pragmatic parallel randomized controlled trial (RCT) in six emergency departments in the United States were used for probability of admission to the observation unit versus outpatient follow up and management and 30-day outcomes. This trial compared the effectiveness of SDM (CPC-DA) with usual care in choice of admission for observation and further cardiac testing or for referral for outpatient evaluation in patients with possible acute coronary syndrome. Other published studies were used to populate per unit costs of management and 30-day outcomes probability of cardiac adverse event. One-way deterministic sensitivity analysis and probabilistic sensitivity analysis using Monte Carlo simulation were conducted. Results: Of 7806 sources identified in the search strategy, 28 studies published from 2000-2020 were included in the systematic review. Most of the studies had a RCT design (n=17), were conducted in the US (n=22), and were related to chest pain, stable coronary artery disease (CAD) /angina, symptomatic CAD (n=11) disease area. Reported outcomes included decision conflict scale (n=9), knowledge (n=9), patient satisfaction (n=11), acceptability of decision aid patient acceptability (n=3), preparedness for decision making scale (n=2), observing patient involvement scale (n=2), trust in physician scale (n=3), impact on pain (n=5), opioid use, and healthcare resource use (n=3). Data from a total of 14 studies was pooled for meta-analyses of six outcomes: decisional conflict scale, decisional conflict scale – chest pain, knowledge, satisfaction with decision or decision making, satisfaction with decision scale, satisfaction with treatment or care. SDM was associated with a moderate reduction in decisional conflict scale (standardized difference in means -0.213; 95% CI -0.345, -0.080; p=0.002; I2=64.571), decisional conflict scale – chest pain (standardized difference in means -0.355; 95% CI -0.589, -0.121; p=0.003; I2=72.664), and increase in knowledge (standardized difference in means 0.274; 95% CI 0.121, 0.428; p=0.000; I2=75.843). There was no statistically significant association with satisfaction with decision or decision making (standardized difference in means 0.131; 95% CI -0.019, 0.281; p=0.087; I2 =49.045), satisfaction with decision scale (standardized difference in means 0.161; 95% CI -0.043, 0.366; p=0.123; I2=14.536), satisfaction with treatment or care (standardized difference in means -0.058; 95% CI -0.340, 0.225; p=0.689; I2=70.756). In the cost-effectiveness analysis, the CPC-DA group yielded an expected cost of $3,867.51 and effect of 0.59, while the usual care group had an expected cost of $4,616.16 and effect of 0.40. The CPC-DA group was dominant with an incremental cost of -$748.65, incremental effect of 0.187, and ICER of -$3,997.68 per admission averted to ED-OU. These findings were robust to sensitivity analyses. Conclusions: The results of the systematic review and meta-analyses provide evidence that SDM and/or DAs are associated with some favorable impact on outcomes among patients with pain, including moderate reduction in decisional conflict and increase in knowledge. Use of SDM and/or DAs may promote better understanding of therapeutic options and promote patient-clinician agreement, which may have favorable downstream effects for better clinical prognosis and management of pain patients. This study contributes to evidence in important emerging areas of research in SDM and particularly in pain. The CPC-DA, designed to facilitate an informed decision between patients and clinicians to either be admitted to the ED-OU or have an outpatient follow up for further cardiac testing and evaluation, can be cost-effective compared with usual care among low-risk chest pain patients presenting to the ED.