The objective was to assess changes over time in prescriptions filled for nonopioid analgesics for older postoperative patients in the immediate postdischarge period. The authors hypothesized that the number of patients who filled a nonopioid analgesic prescription increased during the study period.
The authors performed a population-based cohort study using linked health administrative data of 278,366 admissions aged 66 yr or older undergoing surgery between fiscal year 2013 and 2019 in Ontario, Canada. The primary outcome was the percentage of patients with new filled prescriptions for nonopioid analgesics within 7 days of discharge, and the secondary outcome was the analgesic class. The authors assessed whether patients filled prescriptions for a nonopioid only, an opioid only, both opioid and nonopioid prescriptions, or a combination opioid/nonopioid.
Overall, 22% (n = 60,181) of patients filled no opioid prescription, 2% (n = 5,534) filled a nonopioid only, 21% (n = 59,608) filled an opioid only, and 55% (n = 153,043) filled some combination of opioid and nonopioid. The percentage of patients who filled a nonopioid prescription within 7 days postoperatively increased from 9% (n = 2,119) in 2013 to 28% (n = 13,090) in 2019, with the greatest increase for acetaminophen: 3% (n = 701) to 20% (n = 9,559). The percentage of patients who filled a combination analgesic prescription decreased from 53% (n = 12,939) in 2013 to 28% (n = 13,453) in 2019. However, the percentage who filled both an opioid and nonopioid prescription increased: 4% (n = 938) to 21% (n = 9,880) so that the overall percentage of patients who received both an opioid and a nonopioid remained constant over time 76% (n = 18,642) in 2013 to 75% (n = 35,391) in 2019.
The proportion of postoperative patients who fill prescriptions for nonopioid analgesics has increased. However, rather than a move to use of nonopioids alone for analgesia, this represents a shift away from combination medications toward separate prescriptions for opioids and nonopioids.
Consensus guidelines increasingly recommend the use of nonopioid analgesic medications as a component of postoperative acute pain management after hospital discharge
The epidemiology of postoperative pain management prescribing and medication dispensing among older adults undergoing surgery remains unclear
Administrative data from 278,366 surgical patients in Ontario, Canada, aged 66 yr or older between 2013 and 2019 demonstrated that 22% (n = 60,181) of patients filled no opioid prescription, 2% (n = 5,534) filled a nonopioid only, 21% (n = 59,608) filled an opioid only, and 55% (n = 153,043) filled some combination of opioid and nonopioid
The percentage of patients who filled a nonopioid prescription within 7 days postoperatively increased from 9% (n = 2,119) in 2013 to 28% (n = 13,090) in 2019
However, the overall percentage of patients who filled prescriptions for both an opioid and nonopioid remained constant over time: 76% (n = 18,642) in 2013 to 75% (n = 35,391) in 2019
This represents a shift away from combination medications toward separate prescriptions for opioids and nonopioids
Opioids are the mainstay of treatment for moderate to severe pain in the acute setting. However, numerous studies have demonstrated that use for acute pain is associated with an increased risk of morbidity, long-term dependence, and opioid use disorder.1,2 As a result, state and national opioid-prescribing guidelines have been developed that focus on opioid-sparing prescribing in patients with acute and chronic pain, in both surgical and nonsurgical settings.3–6 Among older patients, opioid use may be associated with additional morbidity and mortality. Common comorbidities in older adults, including frailty, sensory impairment, coronary artery disease, dementia, and osteoporosis, may increase opioid-related risk of confusion, falls, sedation, and respiratory depression.7–9
The perioperative period is a time when many patients receive an initial exposure to opioids. To limit excess opioid prescribing in the perioperative period among older adults, physicians are turning to alternative analgesic strategies to manage pain in surgical settings.10,11 When effectively implemented, a tailored approach to postoperative opioid prescribing that includes nonopioid analgesics as opioid-sparing agents may reduce the length of hospitalization, reduce opioid use, and improve the quality of care delivery among both opioid-naive patients and individuals with opioid use disorder.12–14
Although a recent decrease in opioid prescribing has been observed in both the surgical and general populations,15–17 recent trends in nonopioid analgesic prescribing after surgery have not been well described. Therefore, using population-level data, we sought to determine the percentage of older postoperative adults who filled nonopioid analgesic prescriptions, as well as the type of medication, in the immediate postdischarge period. We hypothesized that the number of patients dispensed a nonopioid analgesic had increased during the study period.
Materials and Methods
Study Design and Data Sources
We performed a population-based, retrospective cohort study using linked health administrative data at ICES in Ontario, Canada. In Ontario, universal access to healthcare services is provided by a single payer and provider, with prescription medication benefits provided to all individuals older than 65 yr. Data from all health encounters is captured by ICES, an independent research institute, through specific health administrative databases. Provincial and national administrative databases are linked deterministically using unique, encoded identifiers that are assigned to each individual. These datasets were linked using unique encoded identifiers and analyzed at ICES. For details regarding the databases and their validation, see Supplemental Content 1 (http://links.lww.com/ALN/C973).
Importantly, although the Narcotic Monitoring System includes any outpatient opioid prescription dispensed in Ontario irrespective of insurance coverage for all ages, the Ontario Drug Benefit captures all prescriptions dispensed for eligible individuals in outpatient settings, including long-term care facilities, community homes, or homes for special care. All patients older than 65 yr are eligible for the drug benefit plan, which provides access to greater than 4,000 medications for a yearly deductible ($100 Canadian Dollars) and copay per prescription ($2 Canadian Dollars to $6.11 Canadian Dollars) that varies based on socioeconomic factors, residing in long-term care facility, or receiving community care services. Medications such as nonsteroidal anti-inflammatory drugs (NSAIDs) or acetaminophen can be obtained over the counter. However, once the yearly deductible ($100 Canadian Dollars) is paid, the Ontario drug benefit plan will cover the cost of these drugs for a copayment price of $2 to $6.11 Canadian Dollars if they are prescribed by a physician. Most pharmacies do not charge patients the $2 copayment; therefore, prescriptions are free for lower-income seniors, although higher-income seniors pay up to $4.11 Canadian Dollars per prescription, which is typically a lower price than purchasing medications over the counter.18 Physicians are not directly financially incentivized to prescribe medications. Personal identification documentation is required to fill an opioid prescription. These characteristics did not change during the study period.19
The study was exempted from review by the institutional review board at Sunnybrook Health Sciences (Toronto, Canada). Informed consent was waived because data were deidentified. Reporting followed Reporting of Studies Conducted using Observational Routinely Collected health data statement for Pharmacoepidemiology (RECORD-PE) guidelines.20
Participants and Exposure
Using Canadian Classification of Intervention codes, we identified adults aged 66 yr or older, who underwent 1 of 15 surgical procedures between July 1, 2013, and March 30, 2020. The index date was defined as the date of admission (in- patient surgery) or date of surgery (out-patient surgery). The procedures and associated codes are listed in Supplemental Content 2 (http://links.lww.com/ALN/C974), and the validity and reliability of these codes has been confirmed through reabstraction studies.21 The surgical procedures were chosen because they represent multiple specialties, including general surgery, cardiac surgery, otolaryngology, urology, obstetrics and gynecology, and orthopedics and are among the most commonly performed in Ontario.22
We excluded patients if they (1) had invalid provincial health insurance; (2) a surgical procedure in the 90 days before their index surgical procedure; (3) received palliative care services in the preceding 90 days (defined as one or more billing codes for palliative care services) as subsequent analgesic use was possibly for end-of-life care23 ; (4) had a prolonged hospitalization defined as a length of stay greater than 14 days; or (5) were discharged to in-patient rehabilitation, because their opioid dispensing is not recorded in outpatient prescription databases. Multiple admissions for the same patient were included, if the subsequent admission met inclusion criteria without any associated exclusion criteria.
The exposure of interest was time, defined by fiscal year (April 1 to March 31) when the surgical procedure was performed.
Preoperative Opioid Requirements
Opioid prescriptions filled were identified using drug identification numbers in the Narcotic Monitoring System and included oral formulations, buccal strips, and transdermal patches of the most frequently prescribed outpatient opioids: morphine, hydromorphone, fentanyl, codeine, oxycodone, tramadol, and meperidine.24 We did not include hydrocodone and normethadone, because they are used solely as cough suppressants in Ontario. We stratified patients based on their preoperative opioid use. Chronic opioid use was defined as having one of the following in the in the year before surgery: (1) 10 or more filled prescriptions of opioids, (2) 120 days of cumulative opioid prescriptions filled, or (3) any filled prescription for buprenorphine or methadone.25 Opioid naive was defined as no prescription filled within 90 days before a surgical procedure and not meeting criteria for chronic opioid use. Intermittent opioid use was defined as a patient not meeting criteria for chronic opioid use or being opioid naive.
Anticipated Postsurgical Pain
The surgical procedures included have varying degrees of postsurgical pain and likely analgesic requirements. Therefore, we classified each surgical procedure into one of three categories using procedure-specific, opioid-prescribing recommendations developed by the Michigan Opioid Prescribing Engagement Network.10 We classified procedures as “low,”, “medium,” or “high” expected pain if 5 to 10, 15 to 20, or 25 to 50 oxycodone immediate-release 5-mg tablets were recommended, respectively (Supplemental Content 2, http://links.lww.com/ALN/C974). The Michigan Opioid Prescribing Engagement Network recommendations were informed by expert opinion, published studies, and patient-reported opioid consumption data obtained through a Quality Improvement Network of more than 40 hospitals.10
Outcome
The primary outcome was one or more filled prescriptions for a nonopioid analgesic in the first 7 days postdischarge (days 0 to 6), where day 0 was defined as either the hospital discharge date or the date of surgery, if there was no hospitalization. The primary outcome did not include nonopioid analgesics incorporated into combination opioid/nonopioid drugs (such as oxycodone/acetaminophen). Each nonopioid analgesic that was chosen is known to be useful in the management of acute, chronic, or neuropathic pain.26 Nonopioid analgesics assessed included: (1) gabapentin or pregabalin; (2) acetaminophen; and (3) nonsteroidal anti-inflammatory drugs. Secondarily, we also evaluated drugs that are primarily used in the management of chronic pain together as a single category, including (1) tricyclic antidepressants, including amitriptyline and nortriptyline; (2) serotonin norepinephrine reuptake inhibitors (venlafaxine and duloxetine); (3) muscle relaxants (cyclobenzaprine, baclofen); and (4) nabilone. Drugs were identified based on their drug identification number in the Ontario drug benefit database; details regarding included drugs can be found in Supplemental Content 3 (http://links.lww.com/ALN/C975). A new nonopioid prescription was defined as a prescription filled within a specific class of medication with no prescriptions in that same class of medications in the 30 days before surgery/hospitalization.
We evaluated the following secondary outcomes: (1) the type(s) of nonopioid analgesic of the first prescription filled. If two nonopioid analgesic classes (i.e., acetaminophen and an NSAID) were dispensed on the same day or within days 0 to 6 from surgery, this contributed only one nonopioid analgesic prescription filled postoperatively; however, they contributed separately to each drug class’s count. (2) Filled opioid prescriptions within the first 7 days. (3) The total morphine equivalent dose of all opioids dispensed in the first prescription between days 0 and 6, including multiple prescriptions that were filled on the same day. Milligram morphine equivalents (MMEs) were calculated using established conversion measures.3 (4) The percentage of patients who filled different combinations of analgesics, including: (a) an opioid; (b) a nonopioid; (c) a combination opioid/nonopioid (such as oxycodone/acetaminophen); (d) both an opioid and nonopioid (as separate prescriptions); and (e) both a combination opioid/nonopioid and a nonopioid.
Patient Characteristics
Baseline patient characteristics included age and sex on the index date, surgery type, income quintile, rural dwelling, preoperative opioid use as defined earlier, preoperative nonopioid use, and comorbidities. Income quintile was derived from the median income of the associated neighborhood where the patient lived in the Canadian census dissemination area.27 Rural residence was defined as a patient who resided in a community with a population of less than 10,000. Preoperative nonopioid analgesic use was defined as one or more prescriptions for a nonopioid analgesic filled within 90 days before the surgical procedure. Comorbidities, namely the Charlson comorbidity score, were identified using standard methods28 based on the International Classification of Diseases (10th Revision, Canada) codes in the year preceding surgery. Although 0.2% of patients had missing data for Neighborhood Income Quintile, this characteristic was not used in our models, and all patients were included in each analysis.
Statistical Analysis
For each fiscal year of surgery, we described baseline characteristics of patients and the analgesic prescriptions. These were summarized using percentages (95%CI), means ± SD, and medians (interquartile ranges). These were first summarized overall, then by procedure. For these unadjusted comparisons over time, we used the Cochran-Armitage trend test or Jonckheere-Terpstra test for unadjusted analysis of categorical variables and generalized linear regression for continuous variables. We assessed the annual change in the incidence of filled opioid and nonopioid prescriptions using logistic regression and opioid dose using linear regression. We accounted for specific patient characteristics that, based on both biologic plausibility and previous literature, may be associated with filling an analgesic prescription postoperatively, including age, sex, income quintile, preoperative opioid use, and Charlson comorbidity score.29–31 Finally, we adjusted for surgery type, because the volume of each surgical procedure changed over time, and each surgical procedure may be associated with varying degrees of pain and postoperative analgesic requirements.
Among patients who filled a new prescription for a nonopioid analgesic, we described the percentage of patients prescribed each type of nonopioid prescription class (NSAIDS, acetaminophen, gabapentin or pregabalin, and chronic pain adjuncts, including tricyclic antidepressants, serotonin-norepinephrine reuptake inhibitors, muscle relaxants, and nabilone). We then describe the percentage of patients being prescribed either a combination opioid/nonopioid (such as oxycodone/acetaminophen) or prescriptions for an opioid and nonopioid adjunct (such as hydromorphone and a separate prescription for naproxen).
To determine whether the dose of opioids prescribed varied depending on the presence or absence of a nonopioid analgesic prescription, we assessed prescriptions in the most recent year of data (2019). We chose to assess the last year of data to avoid the complication of shifting prescribing practices over time. We calculated the total MME dose for the opioid prescribed if patients filled a prescription for (1) an opioid alone, (2) a combination opioid/nonopioid, (3) an opioid prescription filled concurrently with a nonopioid analgesic, or (4) a combination opioid/nonopioid filled along with a nonopioid analgesic. Mean opioid dose was adjusted for age, sex, income quintile, preoperative opioid use, surgery type, and comorbidities.
Sensitivity Analyses
Multiple post hoc sensitivity analyses were performed. First, to limit misclassification of our primary outcome, we increased the definition of when a patient was considered analgesic naive before surgery to 90 days. Second, it is possible that patients may have filled prescriptions for analgesics intended for postoperative pain management preoperatively rather than postdischarge.32 Therefore, we changed the window for the primary outcome and each secondary outcome to allow patients to fill an analgesic prescription within the first 30 days preoperatively until 6 days postoperatively.
Statistical tests were 2-tailed, with significance defined as P < 0.05. A data analysis and statistical plans were written, date-stamped, and recorded in the investigators’ files before data were accessed. Analyses were performed in SAS version 9.4 (SAS Institute, USA).
Results
Of the 1,309,753 admissions for one of the selected surgical procedures, there were 278,366 surgical admissions and 237,882 unique patients aged 66 yr or older included in the study (fig. 1; median age 72 [IQR 69 to 77]; 55% women). Patient characteristics changed during the study period (table 1). Through time, an increasing number of the patients who underwent surgery were older, opioid naive, and more likely to undergo knee or hip joint replacements (table 1; table 2).
Overall, in the cohort, 22% (n = 60,181) of patients filled no analgesic prescriptions, 2% (n = 5,534) filled a nonopioid prescription only, 21% (n = 59,608) filled an opioid prescription only, and 55% (n = 153,043) filled some combination of opioid and nonopioid.
Nonopioid Analgesic Prescriptions Filled after Surgery
The percentage of patients filling a prescription for a nonopioid analgesic within 7 days postoperatively increased with time (n = 2119 [9%] in 2013 and n = 13,090 [28%] in 2019; fig. 2). The yearly incidence of new nonopioid analgesic prescriptions filled remained similar after post hoc sensitivity analyses (Supplemental Content 4, http://links.lww.com/ALN/C976). Stratified by degree of anticipated pain after surgery, the overall pattern also remained the same, with those who underwent moderately painful procedures showing the most variability over time (fig. 2C; Supplemental Content 5, http://links.lww.com/ALN/C977).
Nonopioid Analgesic Type
Overall, the percentage of patients filling prescriptions for each nonopioid analgesic increased over time with the largest increase for acetaminophen (fig. 3A). Among procedures with low anticipated pain, patients filled prescriptions for both acetaminophen and NSAIDs more frequently after 2016; however, overall percentages remained low (fig. 3B and Supplemental Content 6, http://links.lww.com/ALN/C978). A similar pattern was seen among patients undergoing high-pain procedures (fig. 3D). Among procedures with moderate anticipated pain, the percentage of patients who filled prescriptions for acetaminophen and NSAIDS was more stable over time (fig. 3C). For specifics of individual procedures, see Supplemental Content 6 (http://links.lww.com/ALN/C978).
Prescriptions for Opioids Filled after Surgery
The percentage of patients filling an opioid prescription within 7 days postoperatively increased during the study period until 2017 and then began to decline (table 3). This was consistent across most surgical procedures. However, patients undergoing both hip (n = 3,131/3,716; 84%) and knee (n = 7,437/8,439; 88%) arthroplasty filled the highest percentage of opioid prescriptions postoperatively in 2013, which persisted across the entire study period. Results remained similar after post hoc sensitivity analyses that increased the window for perioperative filled opioid prescriptions to include the 30 days before surgery (Supplemental Content 4, http://links.lww.com/ALN/C976).
Combinations of Opioid and Nonopioid Analgesics
The percentage of patients who filled a prescription for a combination opioid/nonopioid decreased during the study period (n = 12,939 [53%] in 2013 to n = 13,453 [28%] in 2019), but the percentage of patients who filled separate prescriptions for both an opioid and nonopioid analgesic increased (fig. 4). Therefore, the total percentage of patients who filled prescriptions for either a combination opioid/nonopioid or separate prescriptions for opioids and nonopioids remained constant through time (76% in 2013 to 75% in 2019; fig. 4 and Supplemental Content 7, http://links.lww.com/ALN/C979).
When patients filled both an opioid and nonopioid analgesic prescription, they were most often dispensed an opioid and acetaminophen (n = 13,942; 5%). If 2 nonopioid analgesics were dispensed alongside an opioid, they were most often acetaminophen and an NSAID (n = 5,615; 2%; Supplemental Content 8, http://links.lww.com/ALN/C980). Almost all patients who filled prescriptions for gabapentin or pregabalin (n = 8,564), which are known to have sedative properties, were coprescribed an opioid (n = 7,810; 91%).
Opioid Dose Associated with Nonopioid Analgesics
Overall, the median MME dose dispensed after surgery across low- and moderate-pain procedures decreased after 2016, but the morphine equivalent dose dispensed for the most painful procedures decreased after 2018 (Supplemental Content 9, http://links.lww.com/ALN/C981). In 2019, patients who filled prescriptions for both opioid and nonopioid analgesics were prescribed an opioid dose higher (mean ± SD, 272 ± 222 MME) than those prescribed an opioid alone (264 ± 287 MME) or a combination opioid/nonopioid (241 ± 190 MME), but the patients who filled both a combination opioid/nonopioid and a separate nonopioid prescription received the highest opioid dose (322 ± 209 MME). However, the higher mean opioid dose was driven by patients undergoing knee replacement, coronary artery bypass graft, and open or laparoscopic colectomy (Supplemental Content 10, http://links.lww.com/ALN/C982 and Supplemental Content 11, http://links.lww.com/ALN/C983). Therefore, after adjusting for measured confounders, including preoperative opioid use and surgery type, patients received a lower average morphine equivalent dose if they filled prescriptions for an opioid and nonopioid separately (adjusted mean MME 238 [95% CI, 233–241]) compared with those who filled a prescription for an opioid alone (adjusted mean MME 277 [95% CI, 273–280]), a combination opioid/nonopioid (adjusted mean MME 260 [95% CI, 257–264]), or a combination opioid/nonopioid and separate nonopioid analgesic (adjusted mean MME 277 [95% CI, 266–286]; Supplemental Content 11, http://links.lww.com/ALN/C983).
Discussion
In this study of greater than 275,000 older adults undergoing surgery, the percentage of patients who filled outpatient nonopioid analgesic prescriptions within 7 days of surgery increased 3-fold from 9 to 28% during a 7-year period. However, this pattern was largely due to a shift away from filling prescriptions for combination opioid/nonopioids toward separate opioid and nonopioid prescriptions, because the overall percentage of patients who received some combination of an opioid and nonopioid did not change over the same time period. Although the total morphine equivalent dose prescribed was lower in 2019 than in 2013 across all procedures, these data suggest that there has not been an increased reliance on nonopioid analgesics for pain postoperatively.
Our findings reveal important information regarding postoperative pain management in older adults. First, and most importantly, we identified that the overall percentage of patients who received any type of nonopioid, either in a combination opioid/nonopioid or as separate opioid and nonopioid prescriptions, remained unchanged during the study period. Furthermore, the percentage of patients prescribed an opioid without a nonopioid adjunct increased until 2018. Given the release of opioid-prescribing guidelines, as well as the litany of calls for change, the limited use of nonopioid adjuncts and continued dependence on opioids for the management of acute postoperative pain remains a significant opportunity for improvement. Among patients undergoing procedures with moderate or high postoperative pain, use of adjunctive acetaminophen or NSAIDs in the immediate postoperative period is associated with reductions in postoperative opioid requirements and related adverse effects.33 A combination of NSAIDS and acetaminophen may confer additional benefit and is associated with less postoperative pain or opioid consumption than opioids alone across a variety of orthopedic and general surgical procedures.13,34 Additionally, patients undergoing procedures associated with low postoperative pain, such as appendectomy and thyroidectomy, may be comfortably treated with opioid-free techniques.35–38 In our study, few patients undergoing these low-pain procedures filled prescriptions for acetaminophen or NSAIDs. Recent surgery-specific analgesic prescribing recommendations such as Michigan Opioid Prescribing Engagement Network 10 and SolvingPain,11 which identify procedures that may be appropriately treated with nonopioid analgesics alone or with multimodal analgesia, can facilitate tailored approaches to pain management. We identified that solely a nonopioid analgesic was dispensed to less than 3% of patients, and 20% of patients did not fill any prescription for analgesia, a portion of whom may be using over-the-counter or previous analgesic prescriptions.
Second, the increased frequency of filled nonopioid analgesic prescriptions corresponded with a decrease in combination opioid/nonopioid prescriptions. Physician decoupling of combination opioids, namely prescribing acetaminophen and a separate opioid, is likely a major driver of our findings. Combination opioid decoupling has multiple potential benefits, particularly in older adults. Physiologic differences in older adults, including depressed renal function, impaired hepatic metabolism, and altered pharmacokinetics, may lead to increased adverse effects with both opioid and nonopioid analgesics.39 Combination opioids typically have a fixed dose of acetaminophen and a corresponding fixed dose of opioid. Patients experiencing a greater degree of pain would only be able to increase their combination opioid dose to the maximum safe daily dose of the included acetaminophen. Conversely, patients with limited pain, who require only nonopioid analgesics, may take an opioid unnecessarily, increasing the risk of opioid-related harm. Additionally, opioids have a dose- dependent increased risk of both delirium and respiratory depression among older adults.40 Therefore, decoupling allows improved titration of opioid dose, aiming to appropriately treat pain while minimizing adverse effect.
Decoupling also allows titration of the dose and duration of nonopioid analgesics to minimize adverse effects. Although acetaminophen remains a first-line agent for the management of mild pain in older adults,8 high-risk patients such as those with preexisting liver disease or alcohol dependence should be prescribed a daily dose 50% lower.8,41 Decoupling would allow a lower acetaminophen dose to be prescribed to high-risk older adults .42 Similarly, nonselective NSAIDS in older adults may be associated with both cardiovascular and renal dysfunction, as well as gastrointestinal bleeding risk.43 Therefore, guidelines have recommended that NSAIDs be used in a highly selected population, for a limited duration.8 Congruently, we demonstrated that less than 3% of patients undergoing coronary artery bypass graft filled prescriptions for NSAIDs. Finally, consistent with guidelines in the management of acute postoperative pain,13 decoupling would allow the provision of a regular, standing, dose of acetaminophen and/or NSAIDS in appropriate patients, without forcing consumption of an opioid.
Third, although only 3% of patients filled a new prescription for gabapentin or pregabalin within 6 days of surgery, almost all these patients also filled an opioid prescription. Concomitant exposure to gabapentinoids alongside opioids can be associated with increased risk of opioid-related adverse events, including respiratory depression, particularly among older adults.44 It is unclear if gabapentinoids offer analgesic benefit, or decreased opioid consumption, when combined with opioids in the perioperative period,45 stressing the importance of risk-benefit assessment when prescribing gabapentinoids postoperatively.
Limitations
Our study has a number of limitations. First, because NSAIDs and acetaminophen can be obtained over the counter, our analysis may underestimate the frequency of nonopioid analgesic dispensed in our patient population. For example, an increasing number of patients filled an opioid prescription alone over the study period, which may represent patients consuming over-the-counter adjuncts. Nevertheless, a similar percentage of patients did not fill any analgesic prescriptions each year, suggesting a continued reliance on prescription analgesics for postoperative pain management. Likewise, these data represent prescriptions filled by patients rather than the number of tablets consumed or the number of prescriptions given by physicians. Therefore, our results likely underestimate the number prescriptions given by physicians and overestimate the dose and volume of analgesics consumed. Additionally, we found that the percentage of patients who did not fill a prescription for analgesia remained stable over time, suggesting that patients were not increasingly relying on nonopioid analgesics available over the counter. Second, due to the nature of health administrative data in Ontario, we could not evaluate changes in pain outcomes, either perioperatively or postoperatively. Similarly, we were unable to measure inpatient analgesic prescribing. Third, because this study was performed with patients in Ontario, it is possible that these data do not represent trends in other jurisdictions. Indeed, significant variability exists in analgesic-prescribing practices across the world, which may be driven by prescribing regulations, limited or inequitable access to essential pain medicines, or cultural differences in pain management and associated opioid use.46 This may limit the external generalizability of our findings. This limitation also extended to the inability to evaluate patients younger than 65 yr due to the lack of comprehensive data on nonopioid prescriptions. However, given that patients older than 65 yr are potentially more at risk of opioid-related adverse events, understanding nonopioid analgesic prescribing in the population remains critical. Finally, because race and ethnicity were not reliably recorded in the datasets used, we were unable to evaluate whether access to appropriate analgesia was equitable across ethnicities.
Conclusions
In this population-based study of older adults undergoing surgery between 2013 and 2019, we found that an increasing percentage of patients filled a prescription for a nonopioid analgesic, most often acetaminophen. This finding was consistent across multiple surgical procedures. However, our findings represent a shift away from filling prescriptions for combination opioid/nonopioids toward separate opioid and nonopioid prescriptions, because the overall percentage of patients who received some combination of an opioid and nonopioid did not change over the same time period. Patients undergoing most procedures, when prescribed both an opioid and nonopioid analgesic, were prescribed an opioid dose lower than those prescribed an opioid alone. Because multiple surgery-specific prescribing recommendations suggest the use of nonopioid analgesia, increased use of nonopioid adjuncts after surgery represents an opportunity for improvement among perioperative physicians.
Research Support
Funding was provided by the National Institute on Drug Abuse (Bethesda, Maryland; R01 DA042299–01A1 to Drs. Wunsch and Neuman, and R01 DA044293–01A1 to Dr. Bateman). Dr. Wunsch received support from the Canadian Institutes of Health Research Canada Research Chair [Tier 2] (Ottawa, Canada). Drs. Ladha, Wunsch and Wijeysundera are supported by a Merit Award from the Department of Anesthesia at the University of Toronto (Toronto, Ontario, Canada). Dr. Wijeysundera is supported in part by a New Investigator Award from the Canadian Translational Anesthesiology Research at St. Michael’s Hospital and University of Toronto (Toronto, Canada). Dr. Goel is supported by the Early Career Investigator Pain Research Grant Program from the Canadian Pain Society (Markham, Canada) in partnership with Pfizer Canada (Kirkland, Quebec, Canada). This study was supported by ICES (Ontario, Canada), which is funded by an annual grant from the Ontario Ministry of Health and the Ministry of Long-Term Care (Toronto, Canada).
Competing Interests
Dr. Bateman is an investigator on grants to Brigham and Women’s Hospital (Boston, Massachusetts) from the US Food and Drug Administration (Silver Spring, Maryland), Baxalta (Westlake Village, California), GSK (Brentford, United Kingdom), Lilly (Branchburg, New Jersey), Pacira (Parsippany-Troy Hills, New Jersey), and Pfizer (New York, New York) for unrelated work; and a paid consultant to Aetion, Inc (New York, New York), and the Alosa Foundation (Boston, Massachusetts). Dr. Ladha is a co-Principal Investigator of an unrelated observational study of medical cannabis funded by Shoppers Drug Mart. He receives no personal or salary support as part of this relationship. The other authors declare no competing interests.
Supplemental Digital Content
Supplemental Content 1: http://links.lww.com/ALN/C973
Supplemental Content 2: http://links.lww.com/ALN/C974
Supplemental Content 3: http://links.lww.com/ALN/C975
Supplemental Content 4: http://links.lww.com/ALN/C976
Supplemental Content 5: http://links.lww.com/ALN/C977
Supplemental Content 6: http://links.lww.com/ALN/C978
Supplemental Content 7: http://links.lww.com/ALN/C979
Supplemental Content 8: http://links.lww.com/ALN/C980
Supplemental Content 9: http://links.lww.com/ALN/C981
Supplemental Content 10: http://links.lww.com/ALN/C982
Supplemental Content 11: http://links.lww.com/ALN/C983