INTRODUCTION — Opioids may be required for treating either acute or chronic pain. When clinicians prescribe opioids, both the clinician and the patient should understand the associated risks of undesired long term use, misuse, and/or opioid use disorder. This topic will discuss the incidence of and risk factors for long term opioid use and misuse after prescription of opioids for pain. Management of acute and chronic pain and the appropriate use of opioids are discussed separately.
●(See "Management of acute pain in opioid naïve adults in the ambulatory setting".)
●(See "Use of opioids in the management of chronic non-cancer pain".)
Epidemiology of prescription drug misuse and opioid use disorder is also discussed separately.
●(See "Prescription drug misuse: Epidemiology, prevention, identification, and management".)
DEFINITIONS
●Long term opioid use – There is no accepted definition of new long term opioid use, and definitions used in the literature vary widely [1-3]. Consequently, the reported incidence of long term opioid use after an initial opioid prescription varies widely as well. The lack of a standardized definition causes difficulty interpreting study results and lack of clarity on patient risks and risk factors. Definitions used in the literature vary from very liberal (eg, filling one opioid prescription within a specific time period after initial prescription), to more restrictive or stringent (eg, continuous or repeated opioid use over a specific time period after an initial prescription). Database studies commonly define persistent use as the number of opioid prescriptions at or during a specified period of time, which may not reflect actual consumption. The reasons for an opioid prescription refill (eg, pain or a nonprescribed reason, storage for later use, or diversion) are typically unknown.
A commonly used definition for long term opioid use is greater than three months of uninterrupted and frequent use of opioids [1], focused on the time period from 90 days to 1 year after the initial prescription [3]. Examples of studies of the variability of definitions include the following:
•In a population based database study of approximately 163,000 opioid naïve surgical patients in Canada, 66 percent of whom filled an opioid prescription in the week after surgery, 29 different definitions found in the literature were applied to estimate long term opioid use [2]. Estimates of persistent opioid use varied by over 100 fold (range 0.01 percent to 14.7 percent), depending on the definition used. Levels of agreement among definitions were greater among more stringent definitions of long term opioid use.
•Another systematic review of studies of long term use after opioid prescription (34 studies) found 41 unique definitions for long term opioid use; 36 percent of studies required a cumulative duration of opioid use of three months [1].
The definitions described above typically apply to patients who are opioid naïve at the time of initial prescription. In one population based study of surgical patients, persistent use in patients who were taking opioids preoperatively was defined as any increase in their average daily opioid prescribed in the 90 days prior to surgery, measured from 90 to 365 days postoperatively [4]. Similar to other database studies, the reasons for increased postoperative opioid prescriptions were unknown.
●Opioid dependence – Opioid dependence is defined as a state in which abrupt cessation or rapid reduction of an opioid after long term use results in physical and affective withdrawal.
●Opioid misuse – Opioid misuse is defined as any use outside of the manner and intent for which it was prescribed. This includes overuse, use to get high, diversion (sharing or selling to others), having multiple prescribers or nonprescribed sources of the opioid. (See "Prescription drug misuse: Epidemiology, prevention, identification, and management", section on 'Prescription drug misuse'.)
●Opioid use disorder – Opioid use disorder (OUD) is a psychiatric disorder, with specific diagnostic criteria (table 1). Opioid misuse is a necessary but not sufficient criterion for OUD. (See "Opioid use disorder: Epidemiology, clinical features, health consequences, screening, and assessment", section on 'Diagnosis'.)
IMPLICATIONS OF LONG TERM OPIOID USE — It is accepted that long term use of opioids should be avoided when possible for most patients. Though the majority of patients who receive an opioid prescription for acute pain do not go on to use opioids long term, the consequences for those who do can be significant, and even lethal. Long term opioid use results in opioid tolerance (requiring increasing doses for similar effect), dependence, sleep disturbances, endocrine abnormalities [5], constipation, immunosuppression, hyperalgesia, neuropsychological deficits, and mood changes including depression [6,7]. Long term use also predisposes to opioid misuse (including use of illicit opioids [8]) and opioid use disorder, which can lead to overdose and opioid-related death [9,10]. These issues are discussed separately.
●(See "Prevention and management of side effects in patients receiving opioids for chronic pain".)
Opioid dependence is a predictable response to chronic opioid use, but can also occur after acute exposure to opioids. A degree of opioid dependence occurs in some patients after even a single dose [11]. Minor opioid withdrawal symptoms can occur when opioids are discontinued after only several days of use; severity of withdrawal symptoms increases with higher doses and longer duration of opioid therapy [12]. Patients receiving long-term opioid therapy for chronic pain may report worsening pain as a result of opioid withdrawal. Thus, avoidance of withdrawal symptoms could be one reason for continued opioid use beyond the duration needed for pain.
LONG TERM OPIOID USE
Rate of long-term opioid use after prescription for acute pain — Data regarding new long term opioid use after an initial prescription come mostly from retrospective database studies, and are therefore subject to possible confounding and deficiencies in data collection. Reported rates vary depending on outcome definitions and the studied patient population, including the geographic location (eg, United States versus other countries). Globally, opioid analgesic consumption has decreased in North America but remains much higher in Canada and the United States compared to other countries worldwide. This geographic disparity should be considered when considering the rates of long-term opioid use after prescribing for acute pain indications [13]. A number of studies have found that persistent opioid use occurs in approximately 6 percent of patients who receive an opioid, though results range from 0.37 percent to over 15 percent [4,14-22], and as high as 27 percent in high risk patients [23]. The rate of long term use varies with patient risk factors and the characteristics of the initial opioid prescription, as discussed below. (See 'Risk factors for long term opioid use after acute pain' below.)
Surgical patients — Many studies of long term opioid use have involved surgical patients, as surgery is among the most common indications for initiating opioids [24,25]. Risk of persistent opioid use may be higher after some types of surgery. (See 'Type of surgery' below.)
Large database studies of surgical patients have reported widely varying rates of long term opioid use (0.4 to 6.5 percent), depending on the patient population, the type of surgery performed, and the way in which long term use was defined. Small single center studies have reported rates of 10 to 14 percent for opioid use at 90 days after surgery [26,27]. Patient risk factors for long term opioid use are discussed below. (See 'Risk factors for long term opioid use after acute pain' below.)
Examples of relevant studies include the following:
●In a United States insurance database study including over 36,000 surgical patients ages 18 to 64 who filled an initial opioid prescription around the time of surgery, the rate of persistent opioid use was 5.9 percent in patients who underwent minor surgery, versus 6.5 percent of patients who underwent major surgery [14]. The study identified a cohort of patients who did not have surgery and assigned a fictitious surgery date. In these patients the rate of new persistent opioid use was 0.4 percent. New persistent opioid use was defined as an opioid prescription filled between 90 and 180 days after surgery.
●In a population based retrospective cohort study of over 39,000 opioid naïve patients aged 66 years or older who had major elective surgery at a Canadian acute care hospital, 3 percent of patients who received opioids in the early postoperative period (within 90 days of surgery) had prolonged opioid use, defined as filling one or more opioid prescriptions from 91 to 180 days after surgery [16]. In a follow-up study of this same cohort, the rate of ongoing opioid use fell from approximately 2 percent at 180 days to 0.4 percent at 365 days [18].
●In a meta-analysis of 33 studies of surgical patients (>1,900,000 patients) in the United States, the pooled rate of persistent opioid use at three months after surgery in opioid naïve patients was 1.2 percent (95% CI 0.4–3.9 percent) [28].
●In a database study of 340,000 patients who underwent ambulatory surgery in Ontario, Canada, overall 13 percent of patients had persistent opioid use after ambulatory surgery; the incidence was much higher in patients who were taking opioids preoperatively [4]. Persistent opioid use occurred in 6.7 percent of opioid naïve patients, in 42.3 percent of patients who had received <60 morphine milligram equivalents (MME) per day in the 90 days prior to surgery, and in 89.5 percent of patients who had received ≥60 MME per day in the week prior to surgery. In patients who were taking opioids preoperatively, persistent use was defined as any increase from the average MME daily dose they used in the 90 days prior to surgery, measured at 90 to 365 days after surgery. The reasons for such increases, and whether the patients consumed the prescribed opioids, were unknown.
●A systematic review found four retrospective insurance claims database studies (486,000 patients) that reported rates of persistent opioid use after cesarean delivery in previously opioid naïve patients [29]. Rates of persistent opioid use ranged from 0.12 to 2.2 percent. Persistent opioid use was defined as ≥2 opioid prescriptions filled within the first year after cesarean delivery, in addition to other criteria that varied among the studies.
Other patient populations
●Traumatic injuries – Reported rates of persistent opioid use after trauma vary widely, and range from 6 to >35 percent [20,30-32]. Opioid use remote from injury may be significantly higher in patients who sustain severe trauma, with persistent post-injury or postsurgical pain [32].
In a prospective observational study that analyzed persistent opioid use in a representative sample of over 36,800 patients from the Medical Expenditure Panel Survey in the United States, approximately 4700 patients reported an injury over a 2 year period; 9 percent of injured patients used an opioid at least once and 15.6 percent became persistent opioid users [20]. Compared with those who did not report an injury, injured patients were 40 percent more likely to become persistent opioid users. In an adjusted analysis, injury increased the odds of persistent opioid use (adjusted odds ratio 1.4, 95% CI 1.1-1.9). Persistent use was defined as opioid use in at least two 4 month study epochs (ie, at least 8 months). Conclusions from this study are limited by its very broad definition of injury, not limited to severe or painful injuries.
●Acute pain in the emergency department (ED) – Reported rates of persistent opioid use in patients who receive an opioid prescription for acute pain in the ED vary from 1 to approximately 17 percent, with variable definitions of persistent use [33-36]. Examples of relevant studies include the following:
•In a prospective cohort study of 484 opioid-naive patients who were prescribed opioids for acute pain at discharge from two academic EDs, 21 percent filled at least two opioid prescriptions in the six months after ED discharge, and one percent filled ≥6 prescriptions in the six months after discharge [33].
•In a retrospective study of 23,381 opioid-naïve Washington Medicaid beneficiaries receiving an opioid prescription within one day of ED discharge, 4.4 percent exhibited persistent opioid use (defined as ≥1 opioid prescription filled in every calendar quarter) in the 12 months after discharge [34].
•In a retrospective single institution study of 4800 opioid naïve patients treated for an acute painful condition in an academic ED in Colorado, patients who filled an opioid prescription were at higher risk of recurrent opioid use compared with patients who did not receive a prescription (17 versus 10 percent, adjusted odds ratio 1.8 [95% CI 1.3-2.3]) [35]. Persistent use was defined as filling an opioid prescription within 60 days of the first anniversary of the initial ED visit.
•In a retrospective cohort study of opioid naive Medicaid recipients with ED visits in the state of Washington, over 23,300 of whom received an opioid prescription at ED discharge, 13.7 percent went on to use opioids long term or receive high risk opioid prescription refills within 12 months, compared with 3.2 percent of patients who received no opioids at the ED visit [34].
●Dental procedures – Retrospective database studies have reported persistent opioid use in 1.3 to 3.2 percent in patients who fill opioid prescription at the time of dental procedures [37,38]. Persistent opioid use was defined as filling ≥1 opioid prescription during the year after the procedure.
●Medical conditions – Reported rates of persistent opioid use after hospital admission for medical conditions are similar to other clinical settings, at 4 to approximately 8 percent [39,40].
•In a retrospective study of almost 200,000 opioid naïve patients who were admitted to the hospital (70 percent for medical conditions), 5.9 percent of patients who received opioids during admission were still using opioids at 90 days after discharge, compared with 3 percent of those who received no inpatient opioids [39]. At 365 days, 7.7 percent of patients who received inpatient opioids were using them, compared with 4.3 percent of patients who received no inpatient opioids.
•In a national database cohort study of over 200,000 Swedish patients who were admitted to the intensive care unit, 4.13 percent of previously opioid naïve patients who received opioids in the ICU continued using opioids chronically, defined as at least one prescription in the first 90 days, as well as at least one prescription in 90 to 180 days after discharge [40].
Risk factors for long term opioid use after acute pain
Patient factors — Patient characteristics that have been consistently associated with increased risk of new long term opioid use after prescription for pain include prior opioid use, tobacco, alcohol, or other substance use, psychiatric disorders (eg, depression, anxiety), and chronic pain conditions [4,14,17-19,26,28,41-44]. A systematic review of observational studies of long term opioid use after surgery or trauma (35 studies, 1,398,182 patients) found a 22.2 times greater incidence of long term opioid in patients with prolonged opioid use before the event, compared with patients without prior opioid use (58.8 percent versus 2.6 percent) [43].
Literature on the risks associated with age and sex is conflicting. In a meta-analysis of studies of opioid use after surgery, age <50 (2 studies) and female sex (14 studies) were associated with increased risk of new long term opioid use, pooled odds ratios 1.83 and 1.16, respectively [28]. In contrast, in a database study of long term opioid use in over 18,000,000 surgical and nonsurgical opioid naïve patients, male sex and age >50 were among the identified risk factors [17].
In one retrospective study of opioid naïve adolescent surgical patients who received an opioid prescription after surgery, a family history of long term opioid use was associated with increased risk of one or more opioid refills between 91 and 180 days after surgery (4.1 percent versus 2.4 percent, adjusted odds ratio 1.54, 95% CI 1.39-1.71) [45].
Most patients who take opioids prior to surgery continue opioids postoperatively, often at a higher dose than they were taking preoperatively [4,28,46-48].
Surgery — It is unclear whether increased risk of opioid use after surgery is the result of persistent postoperative pain, other factors related to surgery, including intraoperative opioid use, or misuse for conditions unrelated to surgery (eg, sleep, anxiety). Continued opioid use is more prevalent in the early months after surgery and declines over time [4,18].
Type of surgery — Some operations are associated with a higher risk of postoperative opioid use. In most studies, operations associated with greater postoperative pain (eg, major orthopedic, spine, open abdominal and thoracic surgery) are associated with highest risks of long term opioid use [16-19,49]. However, in the meta-analysis described above, rates of persistent opioid use were similar after minor versus major surgery [28].
In a retrospective United States health claims database study including approximately 1,800,000 opioid naïve patients, of whom approximately 642,000 had surgery (1 of 11 common operations), the incidence of chronic opioid use in the first year after surgery ranged from 0.12 percent after cesarean delivery to 1.4 percent after total knee arthroplasty, compared with 0.14 percent in nonsurgical patients [17]. Chronic opioid use was defined as having filled ≥10 prescriptions or >120 days’ supply of an opioid in the first year of surgery, excluding the first 90 days. The definition of chronic opioid use was the same in nonsurgical patients, for whom a randomly assigned fictitious surgery date was used. While the authors used statistical adjustment (multivariable models) to adjust for potential confounders, it is possible that unobserved differences existed between the surgical and non-surgical cohorts, limiting accuracy of results.
Prescription related risk factors — The choice of opioid, dose, and duration of initial opioid use in the opioid naïve patient may influence risk of long-term opioid use [22,50-52]. Observational data suggest that use of long acting or extended-release opioids (rather than immediate release short acting preparations), tramadol rather than other short acting opioids, increasing total prescription dose, and prescriptions longer than 5 to 7 days are associated with increased risk of long term use. However, there are no randomized trials confirming a causal association and existing studies are highly susceptible to confounding.
●Choice of opioid – Use of long acting or extended release opioids or tramadol for initial opioid prescription may be associated with increased risk of long term opioid use, based on observational data [22,50,51,53,54]. The 2022 Centers for Disease Control (CDC) Guidelines for opioid prescribing recommended against the use of long acting opioids for acute pain, but this recommendation was based on the potential for increased risk of overdose, not the risk of long term use [55]. (See 'Prescription related risk factors' below.)
•In an insurance database study of initial opioid prescriptions in 1,300,000 previously opioid naïve patients, the probability of continued opioid use was highest in patients who received long acting opioids (27.3 percent at 1 year, 20.5 percent at 3 years), compared with short acting opioids (8.9 percent at one year, 5.3 percent at 3 years) [53]. Use of tramadol was also associated with a higher risk of persistent opioid use, compared with other short acting opioids.
•In another insurance claims database cohort study of approximately 358,000 previously opioid naïve patients who received postoperative opioid prescriptions, prescription of tramadol was associated with slightly increased adjusted risk of long term opioid use compared with other short acting opioids [56]. When a strict definition of chronic opioid use was applied (starting within 180 days, lasting ≥90 days, and including either ≥10 refills or ≥120 day supply), the unadjusted rates of persistent use were 0.6 percent for patients who received tramadol, 0.5 percent for other short acting opioids, and 2.31 for long acting opioids. The adjusted risk ratio for persistent use after receiving tramadol versus other short acting opioids was 1.47 (95% CI 1.25-1.69), and for long acting opioids was 1.18 (95% CI 1.02-1.35). Conclusions are limited by the usual limitations of retrospective data and lack of information on the indication for prescription of specific opioids.
●Prescription dose and duration – Increasing total dose and duration of opioid prescription may be associated with increased risk of long term use [22,23,53,56,57]. Similar to the data on the choice of opioid described above, this is based on observational data, and causation has not been proven. However, randomized trials are not likely to be performed. Most guidelines recommend and some states mandate limiting the initial prescription for acute pain to ≤7 days, and ideally ≤3 days. (See "Management of acute pain in opioid naïve adults in the ambulatory setting", section on 'Opioid dose and duration'.)
The following studies have not controlled for the indication for opioid prescription (eg, pain severity, diagnosis of pain condition) or other factors that may be independent risk factors for long term opioid use.
•In a 2020 review of 13 retrospective studies including over 13,000,000 patients with musculoskeletal injuries, opioid prescription for >7 days (5 studies) and higher morphine milligram equivalents (MME) dose (6 studies) were associated with increased risk of long term opioid use, based on varied definitions of long term use [23]. Meta-analysis was not possible, and the overall quality of evidence was judged to be of low certainty.
•In the insurance database study discussed above, the rate of continued use of opioids at one year after an initial prescription was 6 percent for patients who received at least one day of opioids, increasing to 13.5 percent for patients who received an initial prescription for ≥8 days, and 14 percent for patients who received a prescription refill [53].
•In a study of data from the Prescription Drug Monitoring database in the state of Oregon for approximately 537,000 previously opioid naïve patients who received an opioid prescription, total prescription dose <120 MME (eg, approximately 16 oxycodone 5 mg tablets, 15 hydromorphone 2 mg tablets) during the initiation month was associated with 2 percent risk of long term opioid use, and rose steadily to 7 percent in patients who received 280 to 400 MME in the first month [22]. Long term use was defined as having ≥6 opioid prescription fills within one year of initial prescription.
•In another insurance database study described above, initial prescription dose ≥300 MME was associated with increased risk of long term opioid use by any definition, with adjusted odds ratios of 1.1 to 1.6 compared with doses <300 MME [56].
OPIOID MISUSE AND OPIOID RELATED ADVERSE EVENTS
Rate of opioid misuse or opioid related adverse events after opioid prescription — The risks of opioid misuse and opioid-related adverse events relate to both the initial prescription and the effects of long term opioid use. In addition, prescription of opioids carries the risk of diversion, which raises the possibility of misuse and overdose in individuals other than the prescription recipient. (See "Prescription drug misuse: Epidemiology, prevention, identification, and management", section on 'Misuse'.)
The risk of misuse, opioid use disorder, and overdose after an initial prescription for acute pain is likely low for patients without risk factors. Much of the existing literature consists of retrospective or database studies, and may or may not include prescreening and exclusion for patients with risk factors for substance misuse. Examples of studies on this issue include the following:
●Any opioid initiation for pain – In a study of data from the 2016 to 2017 National Surveys on Drug Use and Health in the United States, over 86 million patients were prescribed any opioid (most for pain), and 12 percent of them misused them [58].
●Opioid initiation for acute pain
•In a retrospective database study of 1,015,116 surgical patients who had no known history of opioid misuse or ongoing opioid use, 56 percent received postoperative opioids, and misuse (defined as a diagnosis code for opioid use disorder or overdose) was identified in 0.6 percent of patients after surgery [54].
•A Swedish registry based cohort study evaluated substance related morbidity (ie, diagnosis of or death from non-tobacco substance use disorder or overdose, pharmacotherapy for alcohol use disorder, or conviction of a substance-related crime) in patients aged 13 to 29 years who received an opioid prescription [59]. Opioid recipient outcomes were compared with a matched cohort who received no analgesics, a cohort who received a new prescription for nonsteroidal antiinflammatory drugs (NSAIDs), and with twin or other multiple birth siblings. Receipt of an opioid prescription was associated with a 30 to 40 percent increase in the risk of substance related morbidity or mortality. The adjusted incidence of morbidity within five years was 6.2 percent for opioid recipients, versus 4.9 percent for NSAID recipients.
•In a retrospective cohort study of 304,780 United States Veterans undergoing surgery, persistent postoperative opioid use, defined as any opioid prescription filled between 90 and 365 days postoperatively, was independently associated with an increased hazard of developing both opioid use disorder (hazard ratio [HR] 1.88, 95% CI 1.8-1.9) and overdose (HR 1.78, 95% CI 1.7-1.9) regardless of whether patients were preoperative opioid users or not [60].
●Chronic opioid therapy – Reported rates of opioid misuse or aberrant drug related behaviors in patients on chronic opioid therapy vary widely from 0.6 to over 40 percent, partly depending on whether the studied patients were screened for opioid related risk prior to prescription. Whether the use of screening tools reduces the incidence of misuse remains to be determined. This issue is discussed separately. (See "Use of opioids in the management of chronic non-cancer pain", section on 'Risk assessment tools'.)
Risk factors for prescription opioid misuse and opioid related adverse events
Patient factors — Demographic and biopsychosocial risk factors for opioid misuse after an opioid prescription overlap with the risk factors for long term use. In a meta-analysis of 65 studies of risk of misuse of opioids prescribed for acute or chronic pain, factors associated with increased risk of opioid misuse included age <40 years, male sex, history of or current illicit or legitimate substance use, current use of benzodiazepines, and mental health disorders [61]. (See "Prescription drug misuse: Epidemiology, prevention, identification, and management", section on 'Risk factors'.)
In a meta-analysis of 28 observational studies of predictors of opioid overdose after opioid prescription for chronic pain, patient factors associated with increased risk of overdose included a history of overdose, substance use disorder, any mental health disorder, depression, bipolar disorder, and pancreatitis [62]. Absolute risks among patients with predictors were 2 to 6 per 1000 for fatal overdose, and 4 to 12 per thousand for nonfatal overdose, with odds ratios from 2 to 5.9 compared with no risk factors.
Prescription related risk factors — Higher doses and longer duration of prescriptions are associated with increased risk of opioid misuse and adverse events. Guidelines from the Centers for Disease Control suggest limiting the doses for chronic opioid prescription to <90 morphine milligram equivalents (MME) per day when possible, due to increased risk of opioid related mortality above that dose. (See "Opioid tapering for patients with chronic pain", section on 'Daily opioid dose with risk exceeding benefit'.)
In the meta-analysis described above, prescription related factors associated with increased risk of opioid overdose included higher opioid dose, three or more opioid prescribers, four or more dispensing pharmacies, and prescription of fentanyl [62].
●Choice of opioid – The type of opioid initially prescribed may influence the risk for opioid misuse. Hydromorphone and oxycodone may be associated with increased risk, compared with other commonly used immediate release opioids [54]. The 2016 Centers for Disease Control (CDC) Guidelines for opioid prescribing recommended against the use of long acting opioids for acute pain, based on increased risk of overdose compared with immediate release formulations [63]. This recommendation was based on a single propensity score-adjusted analysis that suggested an increased risk of overdose with long acting opioid formulations [64]. However, this finding was not reproduced in similar comparative studies [65,66].
●Larger quantities of opioids prescribed – Prescribing higher doses of opioids and for longer durations for both acute and chronic pain may be associated with increased risk of opioid misuse.
•In a retrospective insurance database study of over 65,000 opioid naïve patients who underwent cesarean delivery, risk of serious opioid-related events (ie, persistent opioid use, OUD diagnosis, opioid overdose, or opioid-related death) was increased compared with patients without an opioid prescription, only in patients who received a total dose of ≥100 MME [67]. Patients who received <10 tablets of oxycodone 5 mg were not at increased risk of serious opioid related events, compared with patients without an opioid prescription.
•Among veterans receiving opioids prescriptions for at least three consecutive months for chronic noncancer pain, a high daily opioid dose (MME ≥200 mg per day) is associated with opioid misuse [68].
•In patients prescribed opioids for cancer pain, a morphine equivalent daily dose greater than 50 mg is independently associated with opioid misuse [69].
•In a retrospective study of 800 United States veterans who received prescription opioids, prescription for 30 to 180 days was associated with increased risk of opioid misuse, compared with opioid prescription for <30 days [70].
•In an insurance database study of over 1,000,000 opioid naïve patients who underwent surgery, 56 percent of whom filled a postoperative opioid prescription, opioid misuse occurred in 0.6 percent of patients [54]. Total duration of opioid use was the strongest predictor of misuse, with each additional week of opioid use increasing risk of misuse by approximately 20 percent. Total daily dose was associated with clinically important increase in risk only with opioid use beyond two weeks. In patients who were still using opioids at nine weeks, doses of 100 to 150 MME per day were associated with a 2.4 fold increase in misuse compared with patients taking 50 to 60 MME per day, and a 12-fold increase compared with patients who took <20 MME per day.
●Absence of an opioid agreement plan – Having the patient sign an opioid agreement form prior to initiating chronic opioid therapy may modestly reduce the risk of misuse, though the literature is limited and of low quality. A systematic review of the literature found four poor- to fair-quality observational studies that compared patients who completed treatment agreements with matched or historical controls without treatment agreements. Opioid misuse was decreased modestly (7 to 23 percent) with the use of a treatment agreement, with or without urine drug testing [71]. In a subsequently published retrospective chart review of 800 veterans on chronic opioid therapy for non-cancer pain, half of whom signed an opioid care plan, having a signed opioid care plan was associated with reduced risk of adverse drug related behavior (overall risk 22 percent, odds ratio for signed agreement 0.81) [71].
Predicting future opioid misuse and opioid use disorder — Clinicians should attempt to assess for opioid related risks before prescribing opioids, recognizing that screening is not very sensitive or specific for future opioid misuse or use disorder [72]. A personal history of substance use disorder, certain mental health diagnoses (eg, personality disorder), and concomitant prescription of certain psychiatric medications (eg, atypical antipsychotics) appear useful for identifying patients at higher risk for misuse, though the available literature is weak [72].
Although a number of tools have been developed for the prediction of risk (eg, the Opioid Risk Tool [ORT], Screener and Opioid Assessment for Patients with Pain [SOAPP], revised SOAPP), they are all based on low-quality studies and there are few high-quality studies that have assessed the diagnostic accuracy of these instruments for predicting opioid misuse (form 1 and table 2 and form 2) [72]. (See "Use of opioids in the management of chronic non-cancer pain", section on 'Risk assessment tools'.)
Many guidelines and most states require that clinicians check the Prescription Drug Monitoring Program database prior to prescribing an opioid, to detect undisclosed prescriptions for controlled substances. (See "Prescription drug misuse: Epidemiology, prevention, identification, and management", section on 'Prescription monitoring programs'.)
Urine drug screening is not typically used prior to opioid prescription for acute pain, but is routine (and in some states mandated) prior to prescription for chronic pain. (See "Urine drug testing for patients with chronic pain".)
IMPLICATIONS FOR PRACTICE — The majority of patients who are prescribed opioids for acute pain do not go on to chronic opioid use or opioid misuse. However, the consequences of long term opioid use, misuse and abuse are significant and potentially lethal. Opioids should be prescribed only when necessary, as an adjunct to other multimodal analgesic therapies (eg, nonopioid analgesics, regional anesthesia techniques as appropriate, nonpharmacologic therapies), and at the lowest effective dose for the shortest duration necessary. (See "Management of acute pain in opioid naïve adults in the ambulatory setting".)
Clinicians should screen patients for risk factors for long term opioid use. They should discuss with patients the risks of physiologic dependence, misuse, addiction, and overdose, side effects of opioids, the importance of additional or alternative treatments to reduce the required dose, and should also discuss the availability of naloxone.
Risk assessment and shared decision making when initiating opioid therapy for chronic pain are discussed separately. (See "Use of opioids in the management of chronic non-cancer pain".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic pain management" and "Society guideline links: Acute pain management".)
SUMMARY AND RECOMMENDATIONS
●Implications of long term opioid use – Long term opioid use results in dependence, tolerance, and opioid side effects (eg, sleep disturbances, endocrine abnormalities, constipation, immunosuppression, hyperalgesia, neuropsychological deficits, and mood changes including depression). Long term opioid use also predisposes patients to opioid misuse and opioid use disorder. (See 'Implications of long term opioid use' above.)
●Long term opioid use – Definitions of long term opioid use varies widely; a commonly used definition is greater than three months of uninterrupted and frequent use of opioids in the period from 90 days to 1 year after initial prescription.
•Rate of long term opioid use – A number of retrospective studies have found persistent opioid use in approximately 6 percent of patients who receive a prescription, though the reported range is from 0.37 to over 15 percent, and as high as 27 percent in high risk patients. These widely reported rates of persistent use vary with the study definition of persistent use. (See 'Rate of long-term opioid use after prescription for acute pain' above.)
•Risk factors
-Patient factors associated with increased risk of long term use include prior tobacco, alcohol, or other substance use, psychiatric disorders (eg, depression, anxiety), prior long term opioid use, and chronic pain conditions. (See 'Patient factors' above.)
-Surgery that is associated with greater postoperative pain (eg, major orthopedic, spine, open thoracic or abdominal surgery) may be associated with higher risk of long term opioid use. (See 'Type of surgery' above.)
-Increasing total dose and duration of opioid prescription (>5 to 7 days) may be associated with increased risk of long term use. Use of long acting or extended-release opioids or tramadol may be associated with increased risk of long term opioid use. However, these associations are based on observational data that may be subject to confounding. (See 'Prescription related risk factors' above.)
●Opioid misuse
•Rate of opioid misuse – The rates of opioid misuse, opioid use disorder, and overdose after initial prescription are unclear, but likely low in patients without risk factors. Reported rates in retrospective database studies without clear data on risk factor screening range from 0.6 to over 40 percent. (See 'Rate of opioid misuse or opioid related adverse events after opioid prescription' above.)
•Risk factors
-Patient risk factors for opioid misuse may include age <40 years, male sex, history of or current illicit or legitimate substance use, current use of benzodiazepines, and mental health disorders. Screening for patient risk factors is not very sensitive or specific. (See 'Patient factors' above and 'Predicting future opioid misuse and opioid use disorder' above.)
-Prescribing higher doses of opioids and for longer durations for both acute and chronic pain may be associated with increased risk of opioid misuse. Hydromorphone and oxycodone may be associated with increased risk compared with other commonly used immediate release opioids. (See 'Prescription related risk factors' above.)
●Practice implications
•Prior to prescribing opioids, clinicians should screen patients for risk factors for long term opioid use and misuse. (See 'Patient factors' above and 'Patient factors' above.)
•Opioids should be prescribed only when necessary, as an adjunct to other multimodal analgesic therapies (eg, nonopioid analgesics, regional anesthesia techniques as appropriate, nonpharmacologic therapies), and at the lowest effective dose for the shortest duration necessary. (See "Approach to the management of acute pain in adults" and "Management of acute pain in opioid naïve adults in the ambulatory setting".)
•Clinicians should discuss with patients the risks of physiologic dependence, addiction, and overdose, side effects of opioids, the importance of additional or alternative treatment to reduce the required dose, and should also discuss the availability of naloxone. (See 'Implications for practice' above.)
Do you want to add Medilib to your home screen?