OBJECTIVES/SPECIFIC AIMS: Research overview: Providing patient-centered care is increasingly a top priority in the U.S. healthcare system.1,2 Hospitals are required to publicly report patient-centered assessments, including results from the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) patient satisfaction surveys.3 Furthermore, clinician and hospital reimbursements are partially determined by performance on patient satisfaction measures.3 Consequently, hospitals and clinicians may be incentivized to improve patient satisfaction scores over other important outcomes.4 Paradoxically then, the pursuit of patient-centered care may lead clinicians to fulfill patient requests for unnecessary and potentially harmful treatments.5 Opioid prescribing during hospitalizations may be particularly affected by clinicians’ seeking to optimize patient satisfaction scores.6,7 Satisfaction with pain care is an important predictor of overall patient satisfaction in the HCAHPS surveys,8,9 and clinicians report increased pressure to fulfill patient requests for immediate pain-relief.10,11 Therefore, clinicians may prescribe opioids to avoid receiving lower patient satisfaction scores.12,13 Furthermore, clinicians lack clear guidance on opioid prescribing for some populations, including non-surgical inpatients, who represent almost half of all hospitalizations.14 To reduce clinicians’ incentive to prescribe opioids as a means of achieving patient satisfaction, the Center for Medicare and Medicaid Services (CMS) temporarily removed questions related to patient satisfaction with pain care from the clinician and hospital reimbursement formulas beginning in 2018.15 Importantly, prior research16-20 has not rigorously tested the hypothesis implied by the CMS policy change: that certain opioid prescribing practices in inpatient pain care are associated with higher patient satisfaction. Objectives: The purpose of this study was to evaluate the association between the receipt/dose of opioids during non-surgical hospitalizations and patient satisfaction measured by the HCAHPS survey. METHODS/STUDY POPULATION: Methods/Study Population: We conducted a pooled cross-sectional study of adults (18 and older) with non-surgical hospitalizations within the 11-hospital healthcare system in a Midwestern state from 2011-2016. Data were extracted from electronic health records and linked to HCAHPS patient satisfaction surveys. We estimated the propensity score for receipt of any opioids during hospitalization and separately the receipt of high dose opioids (≥100 morphine milligram equivalent [MME]) based on patient, encounter, and facility characteristics for all hospitalizations with complete data. We used nearest neighbor matching to construct two matched samples to minimize selection bias and confounding by indication. We used a standardized difference threshold of < 0.1 as an indication of the balance between matched groups. Outcomes were compared with a test on the equality of proportions using large-sample statistics. All analysis was performed in STATA 14.0 analytical software. Main outcomes: We analyzed four dependent variables. Two pain-specific patient satisfaction variables were derived from the responses to the following survey questions: 1) “During this hospital stay, how often your pain was well controlled? (pain control)” and 2) “During this hospital stay, how often did the hospital staff do everything they could to help you with your pain? (pain help)”, with 4-point Likert scale responses ranging from “Never” to “Always.” We also used two global satisfaction measures derived from the responses to the following survey questions: 1) “Using any number from 0 to 10, where 0 is the worst hospital possible and 10 is the best hospital possible, what number would you use to rate this hospital during your stay (overall patient satisfaction)?” and 2) “Would you recommend this hospital to your friends and family (willingness to recommend a hospital)? (4-point scale of “Definitely Yes” to “Definitely No”). Because the responses are not normally distributed, and the response options are truncated, we dichotomized each of these questions following previously published approaches8 and CMS methodology3 (e.g. “always” vs. all other responses or “9 or 10 rating” vs. all others). RESULTS/ANTICIPATED RESULTS: Results: Among 17,691 patients who reported that they needed pain medications during hospitalization in their HCAHPS survey, 43.7% (n=7,735) received opioids. Among the matched sample (n=8,848), 55% were female, 90% were white, 9% were black, 74% were emergency admissions, 29% had a circulatory diagnosis, 92% were discharged home, and the average pain score ranged from 0.2 to 7.1 during the hospital stay. Compared to matched patients hospitalized but did not receiving opioids, those who received opioids did not significantly differ in their rating of pain help (75% of patients without opioids rated that they always received help for their pain versus 75% of patients with opioids; p=.78), pain control (55% of patients without opioids reported that their pain was well controlled versus 54% on opioids; p=.93), willingness to recommend the hospital (69% of patients without opioids reported that they would definitely recommend a hospital versus 71% with opioids; p=.16) and overall rating of their care (47% of patients without opioids rated their hospitalization as 10 versus 46% on opioids; p=.22). DISCUSSION/SIGNIFICANCE OF IMPACT: Discussion: We found no evidence that receipt of opioids is associated with patient satisfaction, including at doses. To our knowledge, this is the first study that used propensity score matching to examine the association between inpatient opioid prescribing practices and patient satisfaction. Furthermore, our sample is unique in the inclusion of patients hospitalized for non-surgical indicators over a five year period in the multi-hospital healthcare system in a Midwestern state. Our findings add to the existing literature which has shown contradictory associations between opioid prescribing and patient satisfaction.16-22 Specifically, few studies that looked at surgical inpatients showed a lack of association between patient satisfaction16,18 and opioid prescribing, whereas others showed that receipt of opioids was associated with lower patient satisfaction.17-20 Our findings may imply that satisfaction with pain care may be achieved without administering opioids to non-surgical inpatients. Alternatively, satisfaction with pain care may not be influenced by opioid prescribing for non-surgical inpatients. Future research should further examine the association between opioid prescribing and patient satisfaction among non-surgical inpatients on a national scale to get a better understanding of the relationship between certain pain care practices and patient satisfaction.
Improvement work in mental healthcare: an example from Region Jönköping County, Sweden