Exploring the relationship between language, postoperative pain, and opioid use

BACKGROUND

Racial and ethnic disparities in pain management are well documented. Differences in pain assessment and management by language have not been studied in the postoperative setting in gynecologic surgery.

OBJECTIVE

This study aimed to investigate the association between language and immediate postoperative pain management by comparing pain assessments and perioperative opioid use in non-English speakers and English speakers.

STUDY DESIGN

This was a retrospective cohort study comparing perioperative outcomes between non–English-speaking patients and English-speaking patients who had undergone a gynecologic oncology open surgery between July 2012 and December 2020. The primary language was extracted from the electronic medical record. Opioid use is expressed in oral morphine equivalents. Proportions are compared using chi-square tests, and mean values are compared using 2-sample t tests. Although interpreter services are widely available in our institution, the use of interpreters for any given inpatient-provider interaction is not documented.

RESULTS

Between 2012 and 2020, 1203 gynecologic oncology patients underwent open surgery, of whom 181 (15.1%) were non-English speakers and 1018 (84.9%) were English speakers. There was no difference between the 2 cohorts concerning body mass index, surgical risk score, or preoperative opioid use. Compared with the English-speaking group, the non–English-speaking group was younger (57 vs 54 years old, respectively; P<.01) and had lower rates of depression (26% vs 14%, respectively; P<.01) and chronic pain (13% vs 6%, respectively; P<.01). Although non–English-speaking patients had higher rates of hysterectomy than English-speaking patients (80% vs 72%, respectively; P=.03), there was no difference in the rates of bowel resections, adnexal surgeries, lengths of surgery, intraoperative oral morphine equivalents administered, blood loss, use of opioid-sparing modalities, lengths of hospital stay, or intensive care unit admissions. In the postoperative period, compared with English-speaking patients, non–English-speaking patients received fewer oral morphine equivalents per day (31.7 vs 43.9 oral morphine equivalents, respectively; P<.01) and had their pain assessed less frequently (7.7 vs 8.8 checks per day, respectively; P<.01) postoperatively. English-speaking patients received a median of 19.5 more units of oral morphine equivalents daily in the hospital and 205.1 more units of oral morphine equivalents at the time of discharge (P=.02 and P=.04, respectively) than non–English-speaking patients. When controlling for differences between groups and several factors that may influence oral morphine equivalent use, English-speaking patients received a median of 15.9 more units of oral morphine equivalents daily in the hospital cohort and similar oral morphine equivalents at the time of discharge compared with non–English-speaking patients.

CONCLUSION

Patients who do not speak English may be at risk of undertreated pain in the immediate postoperative setting. Language barrier, frequency of pain assessments, and provider bias may perpetuate disparity in pain management. Based on this study's findings, we advocate for the use of regular verbal pain assessments with language-concordant staff or medical interpreters for all postoperative patients.