Differences in Healthcare Expenditure, Health-Related Quality of Life, Perceived Quality, Medication Access, and Provider Respect Between Opioid and Non-Opioid Users Among a National Sample of Community-Based Older United States Adults with Pain, 2015

Abstract

Introduction: Increasing numbers of older adults, coupled with increasing use of opioids and opioid-related deaths, presents considerable healthcare challenges in the United States (US). This study compared the healthcare expenditure, health-related quality of life (HRQoL), perceived healthcare quality, access to medications, and perceived respect shown by provider between older US adults (≥50 years) with pain who use opioid medications and those who do not use opioid medications. Methods: This study employed a retrospective, cross-sectional database design using Medical Expenditure Panel Survey (MEPS) data from 2015. Eligible participants were those alive for the calendar year, aged 50 years or older, and reported having chronic pain in the past four weeks. The key independent variable was opioid use status (opioid users were defined as those who had a Multum Lexicon therapeutic class code of 60 or 191; the remainder was deemed non-opioid users). Hierarchical linear regression models were constructed to assess healthcare expenditures (inpatient, outpatient, office-based, emergency room, prescription medications, other, and total), HRQoL (short form 12 version 2 physical component summary [SF-12v2-PCS], short form 12 version 2 mental component summary [SF-12v2-MCS], and Kessler 6 [K6] scores), and perceived quality between opioid users and non-opioid users, adjusting for appropriate covariates. Hierarchical logistic regression models were constructed to assess access to medications (unable to receive prescription medications, delayed receiving prescription medications) and perceived respect shown by provider between opioid users and non-opioid users, adjusting for appropriate covariates. National estimates were obtained by adjusting for the complex survey design of MEPS. An alpha level of 0.05 was set a priori for all analyses. All analyses were conducted using SAS version 9.4. This study was deemed exempt from the human subject protection program. Results: The study cohort consisted of 4,759 subjects - 1,525 opioid users and 3,234 non-opioid users. The weighted total number of non-institutionalized adults alive aged 50 or older with pain in the US in the year 2015 was 50,898,592 - 16,757,516 (32.9%) were opioid users and 34,141,076 (67.1%) were non-opioid users. Opioid use was associated with all personal characteristics (p<0.05), except gender and race. After adjustment for predisposing, enabling, need, personal health practices, and external environmental factors, opioid users had 61% greater outpatient expenditure (β=0.477, p<0.0001), 69% greater office-based expenditure (β=0.524, p<0.0001), 14% greater emergency room expenditure (β=0.131, p=0.0045), 63% greater prescription medication expenditure (β=0.486, p<0.0001), 29% greater other healthcare expenditure (β=0.251, p=0.0002), 105% greater total healthcare expenditure (β=0.718, p<0.0001), and 15% greater perceived healthcare quality scores (β=0.154, p=0.0286). After adjustment for predisposing, enabling, need, personal health practices, and external environmental factors, older US adults (≥50 years) with pain in the past four weeks who perceived their provider showed them respect were 1.4 (95% CI = 1.0, 1.8) times more likely to be opioid users than non-opioid users. There was no significant difference between opioid users and non-opioid users for inpatient expenditures, SF-12v2-PCS scores, SF-12v2-MCS scores, K6 scores, or access to medications (p>0.05). Conclusions: This study estimated that the number of older US adults (≥50 years) with pain in 2015 was approximately 51 million, and that the estimated prevalence of opioid use among older US adults (≥50 years) with pain in 2015 was approximately 17 million. Opioid use was associated with all personal characteristics, except gender and race. Adjusted healthcare expenditures were greater among opioid users compared to non-opioid users for all categories of expenditure, except inpatient expenditures. Adjusted perceived healthcare quality measures and perceived respect shown by healthcare provider measures were greater among opioid users compared to non-opioid users. Adjusted health-related quality of life measures and access to prescription medication measures were not associated with opioid use status. Future research is warranted to investigate reasons why some of these findings exist, and to explore these variables in greater depth, over longer periods of time, and in additional populations.