Inpatient Opioid Receipt and Care Experiences for Vaginal Delivery

ABSTRACT

Objectives: To examine the relationship between care experiences and inpatient opioid receipt during and after delivery for women hospitalized for vaginal delivery (VD).

Study Design: We used a pooled cross-sectional design with inverse probability weighting to examine the association between inpatient opioid receipt and care experiences of women hospitalized for VD at a single health care system in a Midwestern state. We used 4 Hospital Consumer Assessment of Healthcare Providers and Systems scores (2 pain care items and 2 global items) as measures of care experiences of women hospitalized for VD.

Methods: We used 4 inverse probability-weighted logit regressions to estimate the relationship between inpatient opioid receipt and each patient care experience measure. In supplementary analyses, we used the same inverse probability-weighted methods to estimate the relationship between receipt of opioids and patient care experience measures in 3 patient subgroups based on mean patient-reported pain score during hospitalization (no pain, mild pain, moderate pain).

Results: We found no relationship between inpatient opioid receipt and inpatient pain care experiences. As an exception, we found that women hospitalized for VD were 5 (95% CI, 2-8) percentage points more likely to rate the hospital as 10 (“the best hospital possible”) during hospitalizations in which an opioid was received. We also found higher overall ratings of the hospital among hospitalized women who reported mild pain if they received an opioid (marginal effects = 0.05; 95% CI, 2-8 percentage points).

Conclusions: Receipt of opioids may not be a significant determinant of the pain-specific patient care experiences of women hospitalized for VD.

Am J Manag Care. 2022;28(7):e248-e254. https://doi.org/10.37765/ajmc.2022.89183

Takeaway Points

• We used a pooled cross-sectional design with inverse probability weighting to examine the association between inpatient opioid receipt and care experiences of women hospitalized for vaginal delivery (VD) at a single health care system in a Midwestern state.
• We found no relationship between inpatient opioid receipt and inpatient pain care experiences during hospitalizations for VD. As an exception, hospitalized patients who received opioids gave the hospital higher overall ratings.
• Clinicians should weigh the benefits of prescribing opioids against nonopioid alternatives based on an unclear relationship between the receipt of opioids and improved patient care experience.

Vaginal delivery (VD) is the most common reason for hospitalization in the United States.¹ The American College of Obstetricians and Gynecologists recommends that hospital care for women having a VD should involve addressing pain during and after giving birth with different nonpharmacological approaches and analgesics, including opioids.² Due to widespread availability and patient familiarity, opioids are widely used during VD (ie, intrapartum), with estimates ranging from 25% to 60% of hospitalizations.^3-6 Furthermore, at least a quarter of hospitalized women receive opioids after VD⁷ (ie, post partum) and fill at least 1 outpatient opioid prescription.⁸ Given the annual number of VDs nationally, opioid exposure during hospitalization is a key risk factor for chronic opioid use, misuse, and dependency for women of reproductive age.^9-11 Importantly, substantial variation in opioid use during hospitalizations for VD exists throughout the country.^6,12 Experts argue that some of the observed variations in opioid use may be partially attributable to clinicians’ prescribing behaviors.⁷

Clinicians’ opioid prescribing behaviors may be influenced by concerns about the patient care experience.⁶ The patient care experience measures collected through the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys are used to determine a portion of hospital and clinician reimbursements from CMS.¹³ Women use quality metrics, including HCAHPS, to choose a hospital for planned procedures, such as VD.^14-16 As a result, some experts argue that clinicians may feel pressure to prescribe opioids, rather than nonopioid analgesics, to hospitalized patients to achieve higher patient care experience scores.^17,18

Importantly, we lack conclusive evidence on whether receipt of opioids during hospitalizations for VD influences the patient care experience.^19-22 Women planning childbirth may compare different hospital services and the publicly available HCAHPS patient care experience data to make informed decisions before seeking obstetrical hospital care.^14-16 Furthermore, given that women are particularly concerned about pain during and following VD,^23,24 they may pay additional attention to the hospital’s HCAHPS pain care experience scores. For these reasons, clinicians may perceive extra pressure to prescribe opioids to women with VD to maintain and/or improve the hospital’s reputation and financial standing and to attract future patients. Importantly, women who receive opioids during hospitalizations for a VD are at a higher risk for inpatient complications and subsequent opioid abuse or dependence.^7,20,25 Thus, it is crucial to understand the relationship between inpatient opioid receipt and patient experience following VD as a critical step in minimizing potential downstream opioid harms.

The purpose of the current study is to examine the relationship between inpatient opioids used during and after VD and patient care experience in a 5-year longitudinal panel of women experiencing VD at 8 hospitals within a single health care system in a Midwestern state. Policy makers may use our findings to inform future refinements to hospital reimbursement. Likewise, clinicians may leverage our conclusions to inform decision-making to address pain care needs of women hospitalized for VD.

METHODS

We conducted a pooled cross-sectional study of hospitalizations for VD from 2011 to 2016. For this study, an external research institution linked the data from the electronic health records (EHRs) with HCAHPS patient experience scores for each hospitalization for VD. Thus, our data included only EHR data from VD that also had HCAHPS scores. The external research institution deidentified linked data and released them for our analysis. Approximately 7% of our observations had repeated hospitalizations for childbirth; we included those repeated hospitalizations in our sample to improve overall generalizability and not limit the sample to hospitalizations for women who are giving birth for the first time.

HCAHPS Patient Experience Data

We used the results of the HCAHPS patient care experience survey to capture women’s experiences with hospital care during hospitalization for a VD. The HCAHPS is a standardized, publicly reported survey that contains 18 core questions about patient care experiences with hospital care.¹³ The survey covers a wide variety of topics, including questions on patient care experiences with pain care, the overall rating of the hospital, and the likelihood of recommending the hospital to family or friends. Patients receive the HCAHPS survey between 48 hours and 6 weeks after hospital discharge. Per the CMS requirements, the HCAHPS survey is the sole assessment of patient care experiences with hospital care.

EHR Data

We used EHR data from 8 hospitals in a single health care system in a Midwestern state to extract patient-level and encounter-level data (eg, receipt of opioids and nonopioid medications, epidural anesthesia, patient demographic information).

Study Sample

Our study sample consisted of women (18 years and older) who were hospitalized for a VD (Table 1). We excluded from our sample women who underwent surgery such as cesarean delivery using codes from the standardized International Classification of Diseases, Ninth Revision (ICD-9) and Tenth Revision (ICD-10). We also excluded women hospitalized for a VD who responded “no” to the HCAHPS gate question about pain, thus indicating not being in pain during their hospitalization, as pain care, including opioid administration, would not be applicable to them. The institutional review board of our research institution deemed this study as exempt.

Dependent Variables

Our dependent variables included HCAHPS questions that are specific to patient care experience with pain care, such as “During this hospital stay, how often was your pain well controlled?” (coded as pain control) and “During this hospital stay, how often did the hospital staff do everything they could to help you with your pain?” (coded as pain help). Each question has 4-point scale response options (“never,” “sometimes,” “usually,” “always”). The HCAHPS poses pain-specific questions only when a patient reports having pain during hospitalizations. Patient care experience with the hospital and overall willingness to recommend a hospital were assessed with responses to 2 separate questions on the survey. To assess the overall rating of the hospital, patients were asked, “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?” To assess the patient’s willingness to recommend the hospital, patients were asked, “Would you recommend this hospital to your friends and family?” using a 4-point scale (response options: “definitely yes,” “probably yes,” “probably not,” and “definitely no”). We dichotomized each of these questions following previously published approaches^26-28 and CMS methodology.¹³ We compared “always” vs all other responses, a “10” rating vs all other numerical ratings, and “definitely recommend” vs other responses.

Independent Variable

Our independent variable was a dichotomous indicator (yes/no) of whether a patient received opioid medications during a hospitalization for a VD. To include all possible opioid medications in our analysis, we used a list based on CMS guidelines, previous literature, and consultations with clinicians and pain specialists (eAppendix A [eAppendices available at ajmc.com]). The list of opioid medications included commonly used opioids for inpatient pain management with various potential routes of administration (eg, oral, intravenous). We also included opioids that can be used in combination with other analgesic medications for epidural analgesia during VD. We excluded hospitalizations with missing information on receipt of opioids.

Analytical Approach

We used a propensity score approach to account for variables potentially correlated with patients’ receipt of opioids and their care experience. Propensity score approaches attempt to reduce confounding by indication based on the fact that clinicians may factor in patient characteristics when ordering opioids for women undergoing VD. Propensity scores make the distribution of observed covariates similar between treated (received opioids) and untreated (did not receive opioids) observations, resulting in both measured and unmeasured covariates being balanced between treatment groups. Thus, by using a propensity score approach we did not need to balance unmeasured covariates.^29,30 The EHR data that we used for our analysis allowed us to measure several observed variables associated with the receipt (or not) of opioids. Thus, we have approximately unbiased estimates of the association between opioid receipt and patient experience.

First, we estimated a propensity score based on the receipt of an opioid during hospitalization for a VD. To develop the propensity score, we used logistic regression that regressed the opioid receipt variable on patient, encounter, and facility characteristics (Table 2). Patient-level characteristics included age (18-24, 25-34, and 35-44 years), race (White, African American, Asian, other), and residential location (urban, rural). Encounter-level characteristics included length of stay (continuous variable measured in days), discharge diagnosis, secondary diagnosis of mental health disorder, Charlson Comorbidity Index (CCI) score, receipt of epidural anesthesia during VD, and year (2011-2016). We also included maximum, minimum, and mean pain score during hospitalization. We grouped the ICD-9/ICD-10 discharge diagnosis codes using the Clinical Classifications Software single-level definitions. We calculated the CCI score based on the ICD-9/ICD-10 secondary diagnosis codes and diagnosis-related groups using Healthcare Cost and Utilization Project comorbidity software version 3.7.27 (Healthcare Cost and Utilization Project). We also included a binary variable capturing whether the hospitalization included an epidural anesthesia during VD. Facility-level dummy variables were used to capture time-invariant characteristics of each of the 8 health care facilities.

Next, we converted the propensity scores into the inverse probability of treatment weights (IPWs). The IPWs allowed us to use all observations in our data set, as opposed to ones that matched only on certain characteristics, thus improving the overall generalizability of our study.³¹ The IPWs are defined as the inverse of the propensity score for the “exposed group” (hospitalized women receiving opioids). The “unexposed group”—hospitalized women not receiving opioids—is defined as the inverse of 1 minus the propensity score. To reduce the influence of outliers, inverse probabilities were winsorized at the 90th percentile. We compared unweighted and weighted descriptive characteristics for hospitalizations with inpatient opioid receipt compared with no receipt, using standardized differences greater than 10 as a measure of imbalance, because the standardized difference is independent of sample size (eAppendix B). We also compared patient experience scores between hospitalizations that included epidural vs no epidural and by type of medication used in the epidural (opioid vs no opioid) (eAppendix C).

Finally, we ran 4 separate logit regressions to estimate the relationship between inpatient opioid receipt and each patient care experience measure, where the IPWs are used for weighting the final models. We present the marginal effects (ME), the absolute change in the probability of the outcome attributable to receipt of opioids, at the mean level of all covariates. The ME is interpreted as the likelihood that patient experience measures increased as a result of opioid receipt. To account for repeated measures, we used robust clustering of standard errors in our models.

To account for the potential relationship between receipt of opioids and reported pain during hospitalization, we examined the difference in patient care experience measures by receipt of opioids based on pain numeric rating scale (NRS) cut points. First, we calculated 3 additional propensity scores based on the mean pain score during hospitalization using NRS cut points: (1) no pain (mean pain score = 0), (2) mild pain (mean pain score = 1-3), and (3) moderate pain (mean pain score = 4-6).³² Next, we converted the propensity scores into IPWs. Finally, we ran 3 separate logistic regressions to estimate the relationship between receipt of opioids and patient care experience measures within the 3 matched groups created based on NRS cut points (Table 3). We considered statistical significance at the conventional α = 0.05 level. All analyses were performed in Stata 14.0 analytical software (StataCorp).

RESULTS

Our initial sample included 6665 hospitalizations of women who had a VD. First, we excluded 619 hospitalizations (9.3%) in which women reported not being in pain during their hospitalization by answering “no” to the HCAHPS gate question about pain. Second, we excluded 787 (11.8%) observations due to missing data on patient care experience scores (n = 720) or responses to the HCAHPS gate question about pain (n = 67). We assessed for potential missingness using cross-tabulations with significance levels and found no evidence to suspect missingness of data not at random. Thus, our final sample included 5259 unique hospitalizations. Approximately 7% of the women in our sample were hospitalized 2 or more times for VD during the study period (2011-2016).

The mean age of women in our sample was 29 years (Table 1). The majority (86%) were White. More than two-thirds (72%) of hospitalizations involved the receipt of opioids during the stay. Similarly, approximately 70% of hospitalizations involved receipt of nonopioid pain medication. The mean pain score for an entire hospitalization was 2.1 (on a scale from 0 to 10). The vast majority of women (4590; 87.3%) were in mild pain during hospitalization using NRS cut points. Approximately two-thirds of the sample were always satisfied with their pain control, received help to address pain, and stated that they would recommend the hospital to their family and friends. Less than half of the sample (45%) gave a “10” overall rating to the hospital. The pain-related patient experience scores for hospitalizations that included epidural vs no epidural and by type of medication used in the epidural (opioid vs no opioid) were similar. Thus, we found no evidence that the choice of epidural vs no epidural, and whether a woman received opioid via epidural or another mode, affected pain-related patient scores within our sample (eAppendix C).

Associations Between Receipt of Opioids and Patient Care Experience Measures

We observed a statistically significant association between opioid receipt during hospitalization and overall rating of the hospital (ME = 0.05; 95% CI, 0.01-0.08) (Table 2). Specifically, women hospitalized for VD were 5 percentage points more likely to rate the hospital as “10” overall if they received an opioid during the hospitalization. We found no statistically significant relationships between receipt of opioids and other patient care experience measures used in our analysis. None of the supplementary analyses performed on 3 propensity score–matched groups by NRS cut points yielded statistically significant associations (Table 3). As an exception, women who had mild pain during hospitalization were 5 percentage points more likely to rate the hospital as “10” overall if they received an opioid during the hospitalization (ME = 0.05; 95% CI, 0.01-0.08).

DISCUSSION

We conducted a pooled cross-sectional study to examine whether receiving opioids during hospitalization for a VD is associated with 4 patient care experience measures (2 pain care–specific measures and 2 global measures) using inverse probability weighting to account for potential biases resulting from confounding. We found no relationship between inpatient opioid receipt and inpatient pain care experiences. Our results are similar to those of prior literature that demonstrated lack of or inconsistent relationships between inpatient opioid use and patient care experiences.^{21,22,26,33-36} Thus, despite concerns raised by legislators, clinician lobbying groups, and senior health officials,^17,18 our findings indicate that receipt of opioids may not be a significant determinant of the pain-specific patient care experiences of women hospitalized for VD. If one is to assume that receipt of opioids during hospitalization should affect patient care experience, we should have observed the largest discrepancy between patient experience scores of women who received opioids during hospitalization vs no opioid. Given that we found no evidence of the relationship between opioid receipt and pain care–specific care experience scores, it is possible that positive patient care experiences with pain care, measured by HCAHPS, can be similarly accomplished with opioids or with nonopioid alternatives.

There are several potential explanations of our findings. First, our study focused on women hospitalized for VD who did not undergo any surgical procedures to assist with childbirth. Thus, our patient population may be markedly different from women who had cesarean deliveries or other major surgical interventions related to delivery. Women who had cesarean deliveries are more likely to experience pain during hospitalization from the procedures performed. Thus, the receipt of opioids may be a more critical factor in driving patient care experience in women with surgical procedures performed compared with women without surgical interventions during childbirth. Second, given that the mean pain score during hospitalization in our sample was relatively low (2.1), receipt of opioids may not be a factor driving patient experience with pain care. Finally, it is plausible that other pain care treatments, such as nonpharmacological therapies (eg, massage),36 were used to address relatively mild levels of pain among women undergoing VD and adequately controlled their pain.

Importantly, due to the concerns about prescribing opioids to improve patient care experience scores, in 2018 CMS released revised pain care questions; these, rather than the questions that we used in our analysis, are currently posed to patients.³⁷ Given that the HCAHPS survey is used nationally to assist with CMS hospital reimbursement decisions, future studies should reexamine how inpatient opioid receipt is associated with patient experience using the new set of pain care questions introduced by CMS.

Notably, we found that receipt of opioids during hospitalizations for VD was significantly associated with a higher overall rating of care. It is plausible that inpatient opioid receipt serves as a proxy indicator for staff’s attentiveness to women’s pain treatment needs during hospitalization for a VD. Our supplementary analyses support this notion, as women with mild levels of pain who received opioids during hospitalizations had a higher likelihood of rating their overall hospital stay more favorably. It is plausible that adequate pain control with opioids was particularly important for women with mild levels of pain. It is also possible that some other factor, not captured in our analysis, is influencing the relationship between inpatient opioid use and overall patient rating of the hospital. Finally, the significant relationship between receipt of opioids and overall rating of the hospital may be driven by a distinct set of expectations of women hospitalized for VD compared with other patient populations. Women hospitalized for a VD are generally younger and healthier than other hospitalized patients. Furthermore, given the planned nature of hospitalization for VD, they have ample time to prepare for hospitalization as opposed to patients hospitalized unexpectedly with likely a different expectation about pain care.³⁸ Finally, pregnant women increasingly expect to have access to a wide array of pain management approaches to adequately address their pain care needs during hospitalization for VD.^39,40 Thus, future research should examine the potential mechanisms through which opioid use may be associated with care experiences of women hospitalized for VD.

Limitations and Strengths

Our study has some limitations. First, we used pooled cross-sectional data; thus, our findings can be interpreted only as associations. We utilized propensity score adjustment to minimize possible selection bias due to confounding by indication, but we recognize that this methodology cannot completely eliminate the influence of such confounding. For example, women may have different inclinations to respond to the HCAHPS survey based on their level of pain during hospitalization (ie, those in more pain may be more likely or less likely to complete the survey). However, the HCAHPS survey is used nationally to assist with CMS hospital reimbursement decisions. Thus, additional research is needed to account for the potential bias resulting from confounding by indication and for omitted variable bias when examining the relationship between receipt of opioid and patient care experiences. Second, we included data from a single Midwestern state, which may limit the generalizability of our results. Furthermore, our study sample was mostly White, making it potentially more difficult to apply our results to other racial/ethnic minority groups. Given that women belonging to minority groups have higher rates of pregnancy complications that might be related to pain and are more likely to report higher pain scores, additional research is needed to examine the relationship between inpatient opioid use and care experiences among racial/ethnic minority patients.^41,42 Third, our data did not include specific information about the dose or duration of opioids received during hospitalizations. Subsequently, we were unable to estimate whether patient care experiences varied with the dose or duration of opioids received. Fourth, we were unable to access information about patients’ prior exposure to opioids, which might affect patient experiences with pain care during hospitalization. Finally, we were not able to include physician characteristics in the analysis. Given recent evidence demonstrating that trainee physicians are less likely to prescribe opioids at discharge for women who had VD,⁴³ additional research is needed to understand how provider characteristics may influence the relationship between inpatient opioid receipt and patient care experiences.

Despite these limitations, our study has several strengths. We used the inverse probability weighting approach to address potential bias resulting from unaccounted confounding factors. Likewise, our sample includes an understudied yet significant patient population. With more than 3 million VDs occurring in the United States every year, our findings assist with clarification of an ongoing debate: Are opioids overprescribed to women hospitalized for VD and, if so, does that lead to persistent use, misuse, and dependency?^9-11 Finally, our findings may have implications for policy makers as subsequent revisions to hospital reimbursement calculations based on patient experience scores are proposed.⁴⁴

CONCLUSIONS

We did not observe significant relationships between inpatient opioid receipt and patient experience, except for the overall rating of hospital care. Our findings have direct implications for clinicians as they weigh the benefits of prescribing opioids vs nonopioid alternatives, based on an unclear relationship between the receipt of opioids and improved patient care experience. Our findings also suggest that hospital-based clinicians are likely already providing care for women undergoing VD that goes beyond simply administering opioids during their stay. Likewise, policy makers may use our findings in informing revisions to hospital reimbursements given that the receipt of opioids is not clearly linked to better patient care experiences.

Author Affiliations: Indiana University School of Medicine–Terre Haute (RD), Terre Haute, IN; Indiana University Richard M. Fairbanks School of Public Health (JB, OM), Indianapolis, IN; University of Florida (CAH), Gainesville, FL; Indiana University School of Medicine (MB, AK), Indianapolis, IN.

Source of Funding: National Institutes of Health KL2 UL1TR001108 Indiana Clinical and Translational Science Institute.

Author Disclosures: Dr Harle has received grants from the Agency for Healthcare Research and Quality and National Institute of Drug Abuse to Indiana University relating to pain care, and he has a patent pending on a data and information technology system for opioid prescribing risk assessment. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (RD, JB, CAH, MB, OM); acquisition of data (OM); analysis and interpretation of data (RD, JB, CAH, MB, OM); drafting of the manuscript (RD, MB, AK, OM); critical revision of the manuscript for important intellectual content (RD, JB, CAH, MB, AK, OM); statistical analysis (JB, OM); obtaining funding (OM); administrative, technical, or logistic support (RD); interpretation of data (AK); and supervision (RD, OM).

Address Correspondence to: Robin Danek, MPH, Indiana University School of Medicine–Terre Haute, 1433 N 6 ½ St, Terre Haute, IN 47802. Email: rdanek@iupui.edu.

REFERENCES

1. Martin JA, Hamilton BE, Osterman MJK, Driscoll AK, Drake P. Births: final data for 2017. Natl Vital Stat Rep. 2018;67(8):1-50.

2. American College of Obstetricians and Gynecologists. Obstetric Care Consensus No. 2: levels of maternal care. Obstet Gynecol. 2015;125(2):502-515. doi:10.1097/01.AOG.oooo460770.99574.9f

3. Sprawson E. Pain in labour and the intrapartum use of intramuscular opioids—how effective are they? Br J Midwifery. 2017;25(7):418-424. doi:10.12968/bjom.2017.25.7.418

4. Smith LA, Burns E, Cuthbert A. Parenteral opioids for maternal pain management in labour. Cochrane Database Syst Rev. 2018;6(6):CD007396. doi:10.1002/14651858.CD007396.pub3

5. Wee MYK, Tuckey JP, Thomas PW, Burnard S. A comparison of intramuscular diamorphine and intramuscular pethidine for labour analgesia: a two-centre randomised blinded controlled trial. BJOG. 2014;121(4):447-456. doi:10.1111/1471-0528.12532

6. Mourad M, Chen L, Landau R, et al. Hospital variation in administration of oral opioid medication during vaginal delivery hospitalizations. Am J Obstet Gynecol. 2018;218(1 suppl):S95-S96. doi:10.1016/j.ajog.2017.10.112

7. Badreldin N, Grobman WA, Yee LM. Inpatient opioid use after vaginal delivery. Am J Obstet Gynecol. 2018;219(6):608.e1-608.e7. doi:10.1016/j.ajog.2018.08.039

8. Prabhu M, Garry EM, Hernandez-Diaz S, MacDonald SC, Huybrechts KF, Bateman BT. Frequency of opioid dispensing after vaginal delivery. Obstet Gynecol. 2018;132(2):459-465. doi:10.1097/AOG.0000000000002741

9. Osmundson SS, Min JY, Grijalva CG. Opioid prescribing after childbirth: overprescribing and chronic use. Curr Opin Obstet Gynecol. 2019;31(2):83-89. doi:10.1097/GCO.0000000000000527

10. Osmundson SS, Wiese AD, Min JY, et al. Delivery type, opioid prescribing, and the risk of persistent opioid use after delivery. Am J Obstet Gynecol. 2019;220(4):405-407. doi:10.1016/j.ajog.2018.10.026

11. Zhu Y, Huybrechts KF, Desai RJ, et al. Prescription opioid use after vaginal delivery and subsequent persistent opioid use and misuse. Am J Obstet Gynecol MFM. 2021;3(2):100304. doi:10.1016/j.ajogmf.2020.100304

12. Becker NV, Gibbins KJ, Perrone J, Maughan BC. Geographic variation in postpartum prescription opioid use: opportunities to improve maternal safety. Drug Alcohol Depend. 2018;188:288-294. doi:10.1016/j.drugalcdep.2018.04.011

13. CMS, HHS. Medicare program; hospital inpatient value-based purchasing program. final rule. Fed Regist. 2011;76(88):26490-26547.

14. Hebert LE, Freedman L, Stulberg DB. Choosing a hospital for obstetric, gynecologic, or reproductive healthcare: what matters most to patients? Am J Obstet Gynecol MFM. 2020;2(1):100067. doi:10.1016/j.ajogmf.2019.100067

15. Maurer M, Firminger K, Dardess P, Ikeler K, Sofaer S, Carman KL. Understanding consumer perceptions and awareness of hospital-based maternity care quality measures. Health Serv Res. 2016;51(suppl 2):1188-1211. doi:10.1111/1475-6773.12472

16. Maurer M, Carman KL, Yang M, Firminger K, Hibbard J. Increasing the use of comparative quality information in maternity care: results from a randomized controlled trial. Med Care Res Rev. 2019;76(2):208-228. doi:10.1177/1077558717712290

17. Alford DP. Opioid prescribing: can the art become more science? J Gen Intern Med. 2013;28(10):1253-1254. doi:10.1007/s11606-013-2493-2

18. Liebschutz JM, Alford DP. Safe opioid prescribing: a long way to go. J Gen Intern Med. 2011;26(9):951-952. doi:10.1007/s11606-011-1797-3

19. Lee JS, Hu HM, Brummett CM, et al. Postoperative opioid prescribing and the pain scores on Hospital Consumer Assessment of Healthcare Providers and Systems Survey. JAMA. 2017;317(19):2013-2015. doi:10.1001/jama.2017.2827

20. Herzig SJ, Rothberg MB, Cheung M, Ngo LH, Marcantonio ER. Opioid utilization and opioid-related adverse events in nonsurgical patients in US hospitals. J Hosp Med. 2014;9(2):73-81. doi:10.1002/jhm.2102

21. Maher DP, Wong W, Woo P, et al. Perioperative factors associated with HCAHPS responses of 2,758 surgical patients. Pain Med. 2015;16(4):791-801. doi:10.1111/pme.12651

22. Nota SPFT, Spit SA, Voskuyl T, Bot AGJ, Hageman MGJS, Ring D. Opioid use, satisfaction, and pain intensity after orthopedic surgery. Psychosomatics. 2015;56(5):479-485. doi:10.1016/j.psym.2014.09.003

23. Phibbs CS, Mark DH, Luft HS, et al. Choice of hospital for delivery: a comparison of high-risk and low-risk women. Health Serv Res. 1993;28(2):201-222.

24. Gibson E. Women’s expectations and experiences with labour pain in medical and midwifery models of birth in the United States. Women Birth. 2014;27(3):185-189. doi:10.1016/j.wombi.2014.05.002

25. Jarlenski M, Bodnar LM, Kim JY, Donohue J, Krans EE, Bogen DL. Filled prescriptions for opioids after vaginal delivery. Obstet Gynecol. 2017;129(3):431-437. doi:10.1097/AOG.0000000000001868

26. Mazurenko O, Blackburn J, Bair MJ, Kara AY, Harle CA. Receipt of opioids and patient care experiences among nonsurgical hospitalized adults. Health Serv Res. 2020;55(5):651-659. doi:10.1111/1475-6773.13556

27. Mazurenko O, Menachemi N. Use of foreign-educated nurses and patient satisfaction in U.S. hospitals. Health Care Manage Rev. 2016;41(4):306-315. doi:10.1097/HMR.0000000000000077

28. Elliott MN, Lehrman WG, Goldstein EH, et al. Hospital survey shows improvements in patient experience. Health Aff (Millwood). 2010;29(11):2061-2067. doi:10.1377/hlthaff.2009.0876

29. Rosenbaum PR, Rubin DB. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am Stat. 1985;39(1):33-38. doi:10.2307/2683903

30. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399-424. doi:10.1080/00273171.2011.568786

31. Austin PC. The performance of different propensity-score methods for estimating differences in proportions (risk differences or absolute risk reductions) in observational studies. Stat Med. 2010;29(20):2137-2148. doi:10.1002/sim.3854

32. Hjermstad MJ, Fayers PM, Haugen DF, et al; European Palliative Care Research Collaborative. Studies comparing Numerical Rating Scales, Verbal Rating Scales, and Visual Analogue Scales for assessment of pain intensity in adults: a systematic literature review. J Pain Symptom Manage. 2011;41(6):1073-1093. doi:10.1016/j.jpainsymman.2010.08.016

33. Bot AGJ, Bekkers S, Arnstein PM, Smith RM, Ring D. Opioid use after fracture surgery correlates with pain intensity and satisfaction with pain relief. Clin Orthop Relat Res. 2014;472(8):2542-2549. doi:10.1007/s11999-014-3660-4

34. Maher DP, Woo P, Wong W, Zhang X, Yumul R, Louy C. Perioperative factors associated with Hospital Consumer Assessment of Healthcare Providers and Systems responses of total hip arthroplasty patients. J Clin Anesth. 2016;34:232-238. doi:10.1016/j.jclinane.2016.03.047

35. Maheshwari K, Cummings KC III, Farag E, Makarova N, Turan A, Kurz A. A temporal analysis of opioid use, patient satisfaction, and pain scores in colorectal surgery patients. J Clin Anesth. 2016;34:661-667. doi:10.1016/j.jclinane.2016.07.005

36. Jones L, Othman M, Dowswell T, et al. Pain management for women in labour: an overview of systematic reviews. Cochrane Database Syst Rev. 2012;2012(3):CD009234. doi:10.1002/14651858.CD009234.pub2

37. Steinberg C. CMS replaces pain management questions on patient hospital satisfaction surveys. Mass Pain Initiative. September 2017. Accessed July 16, 2021. https://masspaininitiative.org/files/CMS_Replaces_HCAHPS_Pain_Management_Questions.pdf

38. Stranges E, Wier LM, Elixhauser A. Complicating conditions of vaginal deliveries and Cesarean sections, 2009. Healthcare Cost and Utilization Projectstatistical brief No. 131. Agency for Healthcare Research and Quality. May 2012. Accessed June 22, 2018. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb131.pdf

39. Lally JE, Murtagh MJ, Macphail S, Thomson R. More in hope than expectation: a systematic review of women’s expectations and experience of pain relief in labour. BMC Med. 2008;6:7. doi:10.1186/1741-7015-6-7

40. Yurashevich M, Carvalho B, Butwick AJ, Ando K, Flood PD. Determinants of women’s dissatisfaction with anaesthesia care in labour and delivery. Anaesthesia. 2019;74(9):1112-1120. doi:10.1111/anae.14756

41. Grobman WA, Bailit JL, Rice MM, et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network. Racial and ethnic disparities in maternal morbidity and obstetric care. Obstet Gynecol. 2015;125(6):1460-1467. doi:10.1097/AOG.0000000000000735

42. Mossey JM. Defining racial and ethnic disparities in pain management. Clin Orthop Relat Res. 2011;469(7):1859-1870. doi:10.1007/s11999-011-1770-9

43. Badreldin N, Grobman WA, Chang KT, Yee LM. Patient and health care provider factors associated with prescription of opioids after delivery. Obstet Gynecol. 2018;132(4):929-936. doi:10.1097/AOG.0000000000002862

44. Pfeifer GM. Is it time to revise HCAHPS? Am J Nurs. 2019;119(11):14. doi:10.1097/01.NAJ.0000605292.00285.d5