Postoperative Opioid Prescribing and the Pain Scores on Hospital Consumer Assessment of Healthcare Providers and Systems Survey

In 2012, the Centers for Medicare & Medicaid Services (CMS) implemented the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) to capture key elements of patient satisfaction, including pain management. HCAHPS surveys are administered to patients 48 hours to 6 weeks after discharge, and scores are used to determine hospital payments.¹ However, patients complete surveys during a time when many are filling postdischarge opioid prescriptions. This timing has raised concerns that HCAHPS measures could inadvertently incentivize clinicians to overprescribe opioids after discharge to ensure satisfactory ratings and reimbursement.^2,3 Citing these concerns, CMS announced it will remove pain management from its determination of hospital payments beginning in 2018, even though little is known regarding the potential correlation between HCAHPS scores and postdischarge opioid prescribing.³ We sought to evaluate the association between HCAHPS pain measures and postoperative opioid prescribing in surgical patients, which accounts for nearly 40% of surgical prescriptions.⁴

The University of Michigan institutional review board approved this study and waived the requirement for informed consent. Insurance claims from the Michigan Value Collaborative, a quality improvement collaborative funded by Blue Cross Blue Shield of Michigan, were examined. We identified patients with continuous enrollment who underwent orthopedic, general, gynecologic, cancer, cardiac, and vascular surgery from 2012 through 2014. To minimize confounding from hospital-level variation attributable to postoperative complications, patients with hospital admissions greater than 30 days or subsequent procedures within 30 days were excluded. Opioid prescriptions were obtained from pharmacy claims and converted to oral morphine equivalents (OMEs). HCAHPS data from corresponding years (2012-2014) were obtained from Hospital Compare.⁵

In this hospital-level study, the exposure variable was the mean quantity of opioids (OMEs) filled during the 30 days after discharge. Primary outcome measures were the HCAHPS pain management and pain dimension scores during the corresponding year. Pain management scores are defined as the percentage of patients who reported their pain was “always” well controlled.¹ Pain dimension scores (range, 0-10) are calculated by comparing a hospital’s pain management score to national benchmarks and are added to other hospital quality measures to determine payments.¹

We used hospital-level, mixed linear models for repeated measures to determine the covariate-adjusted association of postoperative opioid prescribing with HCAHPS pain measures. Postoperative opioid prescribing was log-transformed to satisfy the normality assumption of the model. We adjusted for hospital-level aggregates of patient-level variables (case mix, comorbidities, preoperative long-term opioid use, patients with no opioid prescription fills) and year of evaluation. HCAHPS data were adjusted for patient attributes before release.¹ Opioid prescribing was risk-adjusted for case mix and preoperative long-term opioid use, and adjusted for reliability using empirical Bayes techniques. Analysis was conducted using SAS (SAS Institute), version 9.4. Two-sided P values less than .05 were considered statistically significant.

In this cohort, 31 481 patients underwent surgery at 47 hospitals in Michigan. Table 1 shows HCAHPS pain management and pain dimension scores with hospitals stratified into quintiles of opioid prescribing. In unadjusted comparisons, the first and fifth quintiles had no significant differences in pain management scores (mean [95% CI]: first quintile, 69.5 [66.7-71.7]; fifth quintile, 69.1 [67.2-71.4]) or pain dimension scores (mean [95% CI]: first quintile, 1.9 [1.5-2.0]; fifth quintile, 1.4 [0.9-1.9]). Mixed linear models revealed postoperative opioid prescribing was not correlated with pain management or pain dimension scores after adjusting for covariates (Table 2). Gynecologic procedures were associated with improved pain management and pain dimension scores.

In this study, postoperative opioid prescribing was not correlated with HCAHPS pain measures. The study examined a subset of patients used to generate HCAHPS scores and was limited to a single payer in Michigan. Nonetheless, surgical patients are a key contributor to HCAHPS scores, and opioids account for almost 40% of surgical prescriptions.⁴ Given the growing evidence demonstrating postoperative opioid prescriptions exceed patient requirements,⁶ these findings suggest reducing opioid prescriptions may not worsen HCAHPS scores and hospital reimbursement in Michigan. Moreover, these results may also inform policy makers in the current decision to remove pain management from determination of hospital payments.

Accepted for Publication: March 2, 2017.

Corresponding Author: Jennifer F. Waljee, MD, MPH, MS, Section of Plastic Surgery, University of Michigan Health System, 2130 Taubman Center, SPC 5340, 1500 E Medical Center Dr, Ann Arbor, MI 48109-5340 (filip@med.umich.edu).

Author Contributions: Drs Lee and Waljee had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lee, Hu, Brummett, Englesbe, Waljee.

Acquisition, analysis, or interpretation of data: Lee, Hu, Brummett, Syrjamaki, Dupree, Waljee.

Drafting of the manuscript: Lee, Brummett, Waljee.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Lee, Hu, Syrjamaki, Waljee.

Obtained funding: Brummett, Dupree, Englesbe.

Administrative, technical, or material support: Brummett, Syrjamaki, Dupree, Waljee.

Supervision: Brummett, Dupree, Englesbe, Waljee.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Brummett reports consulting for Tonix, receiving research funding from Neuros Medical, and holding a patent for peripheral perineural dexmedetomidine. Mr Syrjamaki reports receiving grant support from Blue Cross Blue Shield of Michigan (BCBSM) for his role in the Michigan Value Collaborative. Dr Dupree reports receiving grant support from BCBSM for his role in the Michigan Value Collaborative. Dr Waljee reports serving as an unpaid consultant for 3M Health Information systems. No other disclosures were reported.

Funding/Support: Drs Brummett, Englesbe, and Waljee receive funding from the Michigan Department of Health and Human Services. This work was supported by National Research Service award 5T32 CA009672-23 from the National Cancer Institute (Dr Lee); grant 1K08 HS023313-01 from the Agency for Healthcare Research and Quality (Dr Waljee); the American College of Surgeons (Dr Waljee); and the American Foundation for Surgery of the Hand (Dr Waljee); grant R01 AR060392 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (Dr Brummett); grants R01 DA038261-05 and 1R01 DA038261-01A1 from the National Institute on Drug Abuse (Dr Brummett); grants K23 DA038718-04 and P50 AR070600-05 CORT from the National Institutes of Health’s (NIH’s) Department of Health and Human Services (Dr Brummett); and the Michigan Genomics Initiative from University of Michigan School Dean’s Office (Dr Brummett).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: Support for the Michigan Value Collaborative is provided by BCBSM as part of the BCBSM Value Partnerships program; however, the opinions, beliefs and viewpoints expressed by the authors do not necessarily reflect those of BCBSM or any of its employees. The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Michigan Department of Health and Human Services.

Additional Contributions: We thank Justin B. Dimick, MD, MPH (University of Michigan), for his assistance with critical revision of the article. He did not receive compensation for his contribution.