Research
Open Access

Predictors of fatal and nonfatal overdose after prescription of opioids for chronic pain: a systematic review and meta-analysis of observational studies

Li Wang, Patrick J. Hong, Wenjun Jiang, Yasir Rehman, Brian Y. Hong, Rachel J. Couban, Chunming Wang, Corey J. Hayes, David N. Juurlink and Jason W. Busse
CMAJ October 23, 2023 195 (41) E1399-E1411; DOI: https://doi.org/10.1503/cmaj.230459

This article has a correction. Please see:

Abstract

Background: Higher doses of opioids, mental health comorbidities, co-prescription of sedatives, lower socioeconomic status and a history of opioid overdose have been reported as risk factors for opioid overdose; however, the magnitude of these associations and their credibility are unclear. We sought to identify predictors of fatal and nonfatal overdose from prescription opioids.

Methods: We systematically searched MEDLINE, Embase, CINAHL, PsycINFO and Web of Science up to Oct. 30, 2022, for observational studies that explored predictors of opioid overdose after their prescription for chronic pain. We performed random-effects meta-analyses for all predictors reported by 2 or more studies using odds ratios (ORs) and 95% confidence intervals (CIs).

Results: Twenty-eight studies (23 963 716 patients) reported the association of 103 predictors with fatal or nonfatal opioid overdose. Moderate- to high-certainty evidence supported large relative associations with history of overdose (OR 5.85, 95% CI 3.78–9.04), higher opioid dose (OR 2.57, 95% CI 2.08–3.18 per 90-mg increment), 3 or more prescribers (OR 4.68, 95% CI 3.57–6.12), 4 or more dispensing pharmacies (OR 4.92, 95% CI 4.35–5.57), prescription of fentanyl (OR 2.80, 95% CI 2.30–3.41), current substance use disorder (OR 2.62, 95% CI 2.09–3.27), any mental health diagnosis (OR 2.12, 95% CI 1.73–2.61), depression (OR 2.22, 95% CI 1.57–3.14), bipolar disorder (OR 2.07, 95% CI 1.77–2.41) or pancreatitis (OR 2.00, 95% CI 1.52–2.64), with absolute risks among patients with the predictor ranging from 2–6 per 1000 for fatal overdose and 4–12 per 1000 for nonfatal overdose.

Interpretation: We identified 10 predictors that were strongly associated with opioid overdose. Awareness of these predictors may facilitate shared decision-making regarding prescribing opioids for chronic pain and inform harm-reduction strategies

Systematic review registration: Open Science Framework (https://osf.io/vznxj/)

Chronic pain affects 20% of the population worldwide1–5 and is commonly managed with opioids. A 2021 systematic review of 60 observational studies found that opioids are prescribed for 27% of adults living with chronic pain, with higher prevalence of prescribing in North America than in Europe.6 In 2018, 13% of people in Canada (aged ≥ 15 yr) reported use of an opioid analgesic in the past year.7 Opioid use is associated with serious harms, including addiction, and nonfatal and fatal overdose.8

From January to September 2022, 5360 deaths were attributed to opioid toxicity in Canada.9 Although determining the relative contribution of prescribed and illicit opioids is complex, a study of 2910 opioid-related deaths in Ontario, Canada, found that, in 2016, one-third of those who died had an active opioid prescription and more than 75% had been dispensed an opioid within 3 years of death.10 Prescription opioid use has been associated with illicit drug use. A cohort study of 59 804 adults in British Columbia, Canada, found that patients who were prescribed opioids for noncancer pain were 8 times more likely to start injection drug use than opioid-naïve patients.11

Several systematic reviews have explored predictors for fatal and nonfatal opioid overdose following prescription for chronic pain.12–17 The most consistently reported associations with opioid overdose were higher doses of opioids, mental health comorbidities, co-prescription of sedatives, lower socioeconomic status and a history of opioid overdose (Appendix 1, eTable 1, available at www.cmaj.ca/lookup/doi/10.1503/cmaj.230459/tab-related-content). However, these reviews have important limitations, including lack of statistical pooling of measures of association,12,13,15–17 inadequate assessment of risk of bias,12,15,16 outdated searches,12–17 a focus on select populations such as older adults12 or people who have been incarcerated,13 and failure to evaluate the overall certainty of evidence.12–17 We sought to identify predictors of fatal and nonfatal overdose after prescription of opioids for chronic pain that addresses these limitations.

Methods

In conducting our systematic review and meta-analysis we followed the reporting of Meta-analysis Of Observational Studies in Epidemiology (MOOSE) statement,18 as well as guidance for systematic review and meta-analysis of prognostic studies,19 and registered our protocol at Open Science Framework (https://osf.io/vznxj/).

Data sources and searches

A medical research librarian performed database-specific electronic searches of MEDLINE, Embase, CINAHL, PsycINFO and Web of Science from inception to Oct. 30, 2022, without language restrictions (Appendix 1, Section 1). We screened reference lists of all eligible studies and 6 previous reviews12–17 for additional studies.

Study selection

We included cohort or case–control studies that explored, in an adjusted analysis, predictors for fatal or nonfatal opioid overdose after prescription of opioids for chronic noncancer or cancer-related pain lasting 3 or more months. Eligible studies had to provide explicit statements that they followed a patient population in which at least 80% were prescribed opioids for chronic pain, for which the date of the first prescription for opioids was known, and fatal or nonfatal overdose was an outcome that was assessed.

We excluded studies that enrolled palliative care patients or exclusively patients who had previously had an opioid overdose. Studies were also ineligible if they included, in all available models, significant associations with variables collected after baseline because these variables may be a result, rather than a cause, of opioid overdose. When study populations overlapped by more than 50% between articles, we included only data in the largest study. We contacted authors to clarify eligibility or to acquire missing data. Four pairs of reviewers independently screened the titles and abstracts of identified citations and the full texts of potentially eligible studies.

Data extraction and risk of bias assessment

Using standardized, pilot-tested data extraction forms and a detailed instruction manual (available at https://osf.io/vznxj/), 4 pairs of reviewers extracted data and assessed the risk of bias from all eligible studies, independently and in duplicate. We collected information regarding study and patient characteristics, and measures of association for all reported predictors. Reviewers addressed discrepancies through discussion, or adjudication by a third reviewer (L.W.), when necessary.

We assessed risk of bias by evaluating the representativeness of the study population, validity of outcome assessment, loss to follow-up and whether predictive models were adjusted, at minimum, for age, sex, substance use disorder and any other comorbid mental illness. We modified the assessment criteria from the Users’ Guides to the Medical Literature (Appendix 1, Section 2).20

Data analysis

We used the κ statistic to measure inter-rater agreement for full-text screening.21

We pooled the prevalence of nonfatal or fatal overdose using random-effects models with a Freeman–Tukey Double Arcsine transformation to stabilize the variance.22

When possible, we pooled all predictors associated with opioid overdose that were reported by at least 2 studies as odds ratios (ORs) and 95% confidence intervals (CIs). When studies provided adjusted relative risks (RRs) or hazard ratios (HRs), we pooled them with ORs, given the low baseline risk of fatal (1 in 100023) and nonfatal (2 in 100024) overdose.25 We performed random-effects models using the DerSimonian–Laird method for all meta-analyses of 3 or more studies, and fixed-effects models when pooling 2 studies.26 We pooled predictors for opioid overdose in general, as we did not find credible subgroup effects between fatal and nonfatal opioid overdose.

When eligible studies explored associations between individual types of opioids versus all other types of opioids and overdose,27,28 we considered only the comparison of fentanyl versus non-fentanyl opioids, given concerns regarding an elevated risk of overdose with fentanyl because of its high potency.10

To evaluate the relationship of morphine equivalent dose with opioid overdose, we performed a 2-stage, random-effects, dose–response meta-analysis.29,30 We also performed a 1-stage dose–response meta-analysis as a sensitivity analysis.31 When associations for age were reported as categorical data, we converted into continuous data.32–34

We considered relative associations to be large when the pooled OR was 2.0 or higher, or 0.5 or less, and small-to-trivial when the pooled OR was greater than 0.5 and less than 2.0.

We explored the consistency of associations between our pooled results and studies reporting the same predictors that were not possible to pool (e.g., the authors reported the association was significant, but with no accompanying data). We deemed predictors as promising if they were not amenable to meta-analysis but single studies reported a highly significant (p ≤ 0.001) and large association (OR ≥ 2.0 or OR ≤ 0.5) with a study population of at least 1000 patients.34

To avoid overestimating associations, we imputed an OR of 1 for predictors that were excluded in adjusted analyses because of nonsignificance.34,35 We calculated the absolute risk for each predictor amenable to meta-analysis to facilitate interpretation using baseline risks of 1 in 100023 for fatal overdose and 2 in 100024 for nonfatal overdose. We used Stata statistical software version 17.0 (StataCorp) for all analyses.

Small study effects

When at least 10 studies were included in a meta-analysis,33,36 we assessed publication bias by funnel plots and the Egger test.37

Subgroup analyses, meta-regression and sensitivity analyses

We evaluated heterogeneity using forest plots36 and τ2 for all random-effects models. We conducted a priori subgroup analyses for factors that we hypothesized would have a larger association with opioid overdose among chronic noncancer versus cancer-related pain; fatal versus nonfatal overdose; intentional versus unintentional overdose; high versus low risk of bias on a component-by-component basis; current versus previous substance use disorder; comorbid mental health disorders versus none; co-prescription of benzodiazepines versus no benzodiazepine prescription; and tobacco use disorder versus current tobacco use. We conducted subgroup analyses if each subgroup contained at least 2 studies. We assessed the credibility of significant effects using Instrument for Assessing the Credibility of Effect Modification Analyses (ICEMAN) criteria.38

We performed sensitivity analyses by excluding studies for which we imputed an OR of 1, converted categorical data to continuous data or derived measures of association from the RR or HR, and studies that did not exclusively enroll patients with chronic pain.

Certainty of evidence

We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to summarize the certainty of evidence.36 We rated down for imprecision if the 95% CI included both a small-to-trivial and large association with opioid overdose. Given the number of predictors considered in our review, we focused our main presentation of findings on predictors that showed large relative associations (OR ≥ 2 or ≤ 0.5) with opioid overdose supported by moderate- or high-certainty evidence.

Ethics approval

We did not seek ethics approval for this systematic review and meta-analysis of published data.

Results

We reviewed 2918 citations and included 28 unique studies in our analyses (21 cohort studies23,24,39–57 and 7 case-control studies27,28,58–62). We also identified 7 studies (4 cohort63–66 and 3 case-control studies67–69) with overlapping populations that reported unique predictors (Figure 1 and Appendix 1, Section 3 and eTable 1). Inter-rater agreement for full-text screening was near perfect (κ = 0.89). We contacted 64 authors; 2542,44,46,50,52,56,61,69–86 of 33 authors clarified eligibility, and 1024,46,47,51,55,63,65–67,78 of 31 authors provided additional data for meta-analysis.

Figure 1:
Figure 1:

Flow diagram of study selection. *Twenty-eight studies with the largest sample size and longest follow-up were included in our primary analysis; among these, 3 articles reported 2 separate cohorts.46,53,58 †Seven studies63–69 included overlapping study populations with 4 studies55,56,58,60 included in our primary analysis.

Study characteristics

The 28 studies enrolled a total of 23 963 716 patients (52% female) with a median of the mean age of 52 years (interquartile range [IQR] 47 to 57). Twenty-four studies were conducted in the United States, 3 were conducted in Canada and 1 was conducted in the United Kingdom. Twenty-one studies23,24,40–47,50–53,55–57,59–62 included only patients with chronic noncancer pain, and 727,28,39,48,49,54,58 included patients with either chronic noncancer or cancer-related pain. Twenty-two studies enrolled patients with previous or current substance use disorder (median proportion 9%, IQR 4%–13%),23,24,27,28,39–43,46,48–51,53–55,57,58,60–62 and 3 studies excluded patients with comorbid substance use disorder.45,52,56 Twenty-three studies included patients with comorbid mental illness (median proportion 31%, IQR 20%–41%),23,24,27,28,40–42,45–49,51–58,60–62 and 5 studies exclusively recruited veterans.27,40,51,55,58 All studies used administrative databases. The median sample size was 43 885 (IQR 11 186–203 353) (Table 1 and Appendix 1, eTable 2).

View this table:
Table 1:

Characteristics of 28 eligible studies

Risk of bias

Twenty-five studies (89%) were at high risk of bias for at least 1 criterion (Appendix 1, eTable 3). Thirteen studies (46%) included samples that were not representative of the target study population, because all patients were veterans,27,40,51,55,58 more than 60% were patients with comorbid mental illness,41,48,61 all patients were aged 65 years or older,62 more than 50% of patients were disabled44 or patients were prescribed a high daily opioid dose (≥ 50 mg54,56 or ≥ 90 mg morphine equivalent).49 Seventeen studies (61%) were unable to exclude illicit opioid overdoses,24,27,39–42,46,48–50,52–57,61 which may compromise the validity of the outcome measure. Seventeen studies (61%) reported adequately adjusted regression models.23,24,27,28,40,41,44–46,49,52,53,55,58,60,61 Only 3 studies reported loss to follow-up (all < 20%)27,40,50 (Appendix 1, eTable 3).

Predictors of fatal and nonfatal overdose

Moderate-certainty evidence showed the pooled prevalence of fatal opioid overdose after prescription for chronic pain was 1.3 per 1000 (95% CI 0.6–2.3 per 1000) for fatal overdose and 3.2 per 1000 (95% CI 2.0–4.7 per 1000) for nonfatal overdose (Appendix 1, eFigure 1A, eFigure 1B and eTable 4). A total of 103 predictors associated with opioid overdose were reported, among which 72 were amenable to meta-analysis.

Opioid prescribing predictors

High-certainty evidence from 14 studies involving 1 315 173 patients showed a linear dose–response relationship with opioid overdose (Appendix 1, eFigure 2A and eTable 5). The association was small at a 50-mg morphine equivalent dose/day (OR 1.69, 95% CI 1.50–1.90; Appendix 1, eFigure 2B and eTable 6) and large at 90 mg (OR 2.57, 95% CI 2.08–3.18; Figure 2 and Table 2), with an absolute risk 2.6 per 1000 for fatal overdose and 5.1 per 1000 for nonfatal overdose at a 90-mg morphine equivalent dose/day.

Figure 2:
Figure 2:

Association of prescribed opioid dose (per 90-mg morphine equivalent dose/d increase) with risk of fatal or nonfatal opioid overdose. Bohnert et al., 2011a is for 36 803 patients with chronic cancer pain; Bohnert et al., 2011b is for 111 759 patients with chronic noncancer pain. Note: CI = confidence interval, OR = odds ratio.

View this table:
Table 2:

Evidence profile of predictors with large associations with fatal or nonfatal overdose following opioid prescription for chronic pain

Moderate- to high-certainty evidence showed large associations between opioid overdose and 3 or more prescribers (OR 4.68, 95% CI 3.57–6.12), 4 or more dispensing pharmacies (OR 4.92, 95% CI 4.35–5.57) and prescription of fentanyl versus other opioids (OR 2.80, 95% CI 2.30–3.41). The absolute risks ranged from 2.8 to 4.9 per 1000 for fatal overdose, and from 5.6 to 9.8 per 1000 for nonfatal overdose (Appendix 1, eFigures 2C–2E and Table 2).

Moderate- to high-certainty evidence showed small-to-trivial increased risks of opioid overdose with long- versus short-acting opioid formulations; number of naloxone prescriptions; as-needed and regularly scheduled versus scheduled administration alone; and longer versus shorter duration of opioid use. The absolute risks ranged from 1.0 to 1.9 per 1000 for fatal overdose, and from 2.0 to 3.8 per 1000 for nonfatal overdose (Appendix 1, eFigure 2F, eFigure 2G and eTable 6).

Co-prescription predictors

Moderate-certainty evidence suggested small associations between opioid overdose and co-prescription of benzodiazepines, anticonvulsants, sedatives or muscle relaxants, with the absolute risks ranging from 1.3 to 1.8 per 1000 for fatal overdose, and from 2.6 to 3.6 per 1000 for nonfatal overdose (Appendix 1, eFigures 3A–3D and eTable 7).

Psychological predictors

We found a credible subgroup effect between current versus previous substance use disorder (p = 0.01; Figure 3 and Appendix 1, Section 4); therefore, we reported results for these predictors separately. Moderate- to high-certainty evidence showed large associations between opioid overdose and current substance use disorder (OR 2.62, 95% CI 2.09–3.27; Figure 3), any mental health diagnosis (OR 2.12, 95% CI 1.73–2.61; Figure 4), depression (OR 2.22, 95% CI 1.57–3.14; Appendix 1, eFigure 4A), and bipolar disorder (OR 2.07, 95% CI 1.77–2.41; Appendix 1, eFigure 4B). The absolute risks ranged from 2.1 to 2.6 per 1000 for fatal overdose, and from 4.1 to 5.2 per 1000 for nonfatal overdose (Table 2).

Figure 3:
Figure 3:

Association of substance use disorder (SUD) and risk of fatal or nonfatal opioid overdose. Test of interaction = 0.01 for current versus history of SUD. Bohnert et al., 2011a is for 36 803 patients with chronic cancer pain; Bohnert et al., 2011b is for 111 759 patients with chronic noncancer pain. Glanz et al., 2018a is for 42 828 patients from the derivation site cohort; Glanz et al., 2018b is for 10 708 patients from the validation site cohort. Note: CI = confidence interval, OR = odds ratio.

Figure 4:
Figure 4:

Association of comorbid mental illness and risk of fatal or nonfatal opioid overdose Test of interaction = 0.67. Bohnert et al., 2011a is for 36 803 patients with chronic cancer pain; Bohnert et al., 2011b is for 111 759 patients with chronic noncancer pain. Glanz et al., 2018a is for 42 828 patients from the derivation site cohort; Glanz et al., 2018b is for 10 708 patients from the validation site cohort. Qeadan et al., 2021a is for 76 141 patients with chronic pain syndrome; Qeadan et al., 2021b is for 792 956 patients with low back pain. Note: CI = confidence interval, OR = odds ratio.

Moderate-certainty evidence suggested smaller associations with psychotic disorders, tobacco use or tobacco use disorder, history of substance use disorder and anxiety. The absolute risks ranged from 1.3 to 1.6 per 1000 for fatal overdose, and from 2.6 to 3.2 per 1000 for nonfatal overdose (Appendix 1, eFigures 4C–4E and eTable 8).

Medical predictors

Moderate- to high-certainty evidence showed large associations between opioid overdose and history of overdose (OR 5.85, 95% CI 3.78–9.04; absolute risk 5.9 per 1000 for fatal overdose, 11.7 per 1000 for nonfatal overdose; Appendix 1, eFigure 5A) and pancreatitis (OR 2.00, 95% CI 1.52 to 2.64; absolute risk 2.0 per 1000 for fatal overdose, 4.0 per 1000 for nonfatal overdose; Appendix 1, eFigure 5B) (Table 2).

Moderate- to high-certainty evidence showed small-to-trivial associations with heart failure, hemiplegia or paraplegia, renal disease, liver disease, chronic obstructive pulmonary disease, cancer, hypertension, diabetes, injury or acute pain, emergency department visit and higher Charlson Comorbidity Index scores. The absolute risks ranged from 1.1 to 1.7 per 1000 for fatal overdose, and from 2.3 to 3.3 per 1000 for nonfatal overdose (Appendix 1, eTable 9).

Sociodemographic predictors

Moderate- to high-certainty evidence showed small-to-trivial associations between opioid overdose and public or no insurance versus private insurance, White race or ethnicity versus other racial or ethnic groups, younger age, male sex, unmarried status and geographical region. The absolute risks ranged from 1.1 to 1.8 per 1000 for fatal overdose, and from 2.1 to 3.6 per 1000 for nonfatal overdose (Appendix 1, eFigures 6A–6F and eTable 10).

Predictors not amenable to pooling

The results from studies that reported predictors that we subjected to meta-analysis but whose data could not be included were consistent with our pooled analyses (Appendix 1, eTable 11). We were unable to pool 31 predictors that were each reported by a single study, of which 3 met our criteria as promising for future study, namely opioid tapering, opioid discontinuation (Appendix 1, eTable 12) and traumatic brain injury (Appendix 1, eTable 13). Twenty-three of 31 predictors were consistently not associated with opioid overdose (Appendix 1, eTable 14).

Additional analyses

No additional subgroup analysis or meta-regression were judged as credible according to ICEMAN criteria (Appendix 1, eTable 15). Our sensitivity analyses found no important differences in results whether we incorporated missing data for nonsignificant predictors, pooled different measures of associations (OR, RR or HR), included studies that did not exclusively enroll patients with chronic pain or converted categorical data on age to continuous data (Appendix 1, eTable 16). We detected no evidence of publication bias among predictors reported by at least 10 studies (Table 2 and Appendix 1, eFigure 7 and eTables 6–10).

Interpretation

In this systematic review of observational studies involving nearly 24 million patients receiving opioids for chronic pain, we pooled data on 72 predictors; of these, moderate- to high-certainty evidence showed large relative associations of opioid overdose with previous opioid overdose, current substance use disorder, depression, bipolar disorder, any mental health diagnosis, pancreatitis, 3 or more opioid prescribers, 4 or more dispensing pharmacies, prescribing 90-mg morphine equivalents or more, or prescription of fentanyl. We explored 31 additional predictors that could not be statistically pooled, and preliminary evidence suggested that opioid tapering or discontinuation strategies and traumatic brain injury warrant additional study.

Of the 6 previous systematic reviews that have explored predictors for opioid overdose,12–17 only 1 conducted meta-analysis for the single predictor of opioid dose.14 This review pooled 7 cohort studies and reported a larger association than our review (RR 4.28 for > 100 v. ≤ 100 morphine equivalent dose/d); however, they included unadjusted data that may overestimate associations.14 Clinical guidelines recommend against use of high doses (e.g., ≥ 90-mg morphine equivalent dose/d) when starting a trial of opioids for chronic pain management and recommend approaching patients on high doses to consider tapering to reduce potential harms, including overdose.8,87,88 However, forced or aggressive tapering of opioids or stopping opioids may increase risk of overdose and death.89,90 Our review found conflicting evidence from 6 studies,48,55,56,64–66 with 2 reporting that tapering or stopping decreased risk of overdose55,65 and 4 reporting no association or increased risk (Appendix 1, eTable 12).48,56,64,66 One source of this variability may be how tapering is approached. Emerging evidence suggests that voluntary, supported opioid tapering may help most people who are prescribed high-dose opioid therapy for chronic pain to safely reduce their dose.91,92 A clinical trial of 608 patients prescribed strong opioids to manage chronic noncancer pain, randomized to usual care or education and support for tapering, found that 29% of patients in the intervention arm had stopped opioids at 1 year, compared with 7% of patients in the usual care arm, but with no effect on perceived pain interference with daily life activities.93

Our review, which included 22 studies that were not considered by previous reviews,28,39,41,43–45,49–51,53–57,59,60,63–67,69 quantified large associations between opioid overdose and higher opioid dose,12,14–16 prescription of fentanyl,15 current substance use disorder, 12,15,16 mental health disorders,12 depression16,17 and previous opioid overdose.15 Previous systematic reviews have qualitatively summarized these associations, but with conflicting results.12,13,15–17 Further, we also found moderate- to high-certainty evidence for 33 additional predictors that were not reported by previous reviews.

We found large associations between opioid overdose and multiple prescribers or dispensing pharmacies, both of which have been linked to diversion and opioid use disorder.94,95 A cross-sectional study of nearly 1.5 million opioid prescriptions found that prescriber and pharmacy shopping accounted for 0.6% of dispensed medications.96 Prescription drug monitoring programs have been developed to address this issue, but their effectiveness in reducing opioid-related harms is uncertain.97,98 We found a large association between opioid overdose and pancreatitis, which may be a surrogate for alcohol use disorder.99 We found a smaller positive association between opioid overdose and number of naloxone prescriptions, which is likely a surrogate for patients at higher risk for opioid overdose.100

The opioid crisis has generated interest in identifying patients at higher risk of addiction or overdose and has led to the development of several screening tools; however, these instruments have either not been validated or have shown poor psychometric properties.46,101–103 Our findings suggest that awareness of, and attention to, several patient and prescription characteristics, may help reduce the risk of opioid overdose among people living with chronic pain.

Evidence alone is insufficient for clinical decisions regarding management of chronic pain, which also requires consideration of individual patient values. A systematic review found that people living with chronic pain place less value on the possibility of addiction versus the possibility of important pain relief.104 Our findings should prove helpful for conveying risks of overdose to patients when deciding whether to start a trial of opioids for chronic pain, and will facilitate evidence-based, shared decision-making.

Limitations

We were unable to pool data for predictors from studies that used different measures. Results from these studies were, however, consistent with results from studies amenable to pooling. Although we did not find credible subgroup effects for fatal versus nonfatal overdose, intentional versus unintentional overdose or chronic cancer versus noncancer pain, our analyses may have been underpowered as most eligible studies reported mixed types of opioid overdose and enrolled patients with both cancer and noncancer chronic pain.

Although opioid overdose is a serious outcome, we defined an important increase in risk as at least twice the baseline risk (i.e., OR ≥ 2) given that fatal and nonfatal overdoses are uncommon events. We reported all absolute measures of association and 95% CIs for predictors to facilitate use of alternate thresholds. Studies eligible for our review used administrative data to identify opioid-related overdose or death; however, this approach has shown more than 80%–90% positive predictive value.105,106 We did not assess the validity of predictor assessment; however, information on predictors was abstracted from either medical records, pharmacy, insurance or other administrative databases. As such, they are at low risk of bias for ascertainment of exposure.107

No opioid-conversion method for calculating the morphine equivalent dose is universally accepted, and competing approaches can yield important differences;108 however, 14 studies of more than 1 million patients provided high-certainty evidence for an association between higher doses of prescribed opioids and increased risk of overdose. Finally, we evaluated all included study models for whether they adjusted for age, sex, substance use disorder and any other comorbid mental illness; however, residual or unmeasured confounding may have affected our findings.

Conclusion

In this meta-analysis of observational studies of patients prescribed opioids for chronic pain, moderate- to high-certainty evidence showed large associations of fatal and nonfatal overdose with a history of opioid overdose, depression, bipolar disorder, a mental health diagnosis, current substance use disorder, pancreatitis, multiple opioid prescribers or dispensing pharmacies, prescription of 90-mg morphine equivalents or higher and prescription of fentanyl. Awareness of these predictors may facilitate shared decision-making regarding prescribing opioids for chronic pain and may inform harm-reduction strategies.

Acknowledgements

The authors thank Dr. Norman Buckley, Department of Anesthesia, McMaster University, for providing suggestions on clarification of opioid types and use of methadone for pain or for substance use disorder, and Dr. Alka Kaushal, Department of Family Medicine, McMaster University for literature screening. They thank Dr. Matthew Miller, Department of Health Sciences and Epidemiology, Northeastern University; Drs. Almut G. Winterstein and Yan Li, Department of Pharmaceutical Outcomes and Policy, College of Pharmacy, University of Florida; Dr. Joseph A Boscarino, Center for Health Research, Geisinger Clinic; Dr. Barbara J. Turner, Gehr Center for Health Systems Science and Innovation, Keck School of Medicine; Dr. Yuanyuan Liang, Department of Epidemiology and Public Health, University of Maryland School of Medicine; Dr. Alicia Agnoli, Department of Family and Community Medicine, University of California; Dr. Fusheng Wang, Department of Biomedical Informatics, Stony Brook University; Dr. Elizabeth Oliva, Veterans Affairs Program Evaluation and Resource Center, Veterans Affairs Office of Mental Health and Suicide Prevention; Dr. Krista Highland, Defense and Veterans Center for Integrative Pain Management, Department of Anesthesiology, Uniformed Services University; and Dr. Bradley C. Martin, Division of Pharmaceutical Evaluation and Policy, College of Pharmacy, University of Arkansas for Medical Sciences, for clarifying issues regarding their studies. They thank Dr. Tara Gomes, Li Ka Shing Knowledge Institute; Dr. Barbara J. Turner, Gehr Center for Health Systems Science and Innovation, Keck School of Medicine, University of Southern California; Dr. Michael Von Korff, Kaiser Permanente Washington Health Research Institute; Dr. John Bedson, Arthritis Research UK Primary Care Centre, Research Institute for Primary Care & Health Sciences, Keele University; Dr. Ying Chen, Biostatistics & Epidemiology, Department of Health and Environmental Sciences, Xi’an Jiaotong-Liverpool University; and Dr. Bradley C. Martin, Division of Pharmaceutical Evaluation and Policy, College of Pharmacy, University of Arkansas for Medical Sciences, for providing additional data. No financial compensation was provided to any of these individuals. We regret to inform readers that Dr. Boscarino has recently deceased.

Footnotes

  • Competing interests: Corey Hayes reports support from the National Institute on Drug Abuse Clinical Trials Network (5UG1DA013732-23S3). David Juurlink is an unpaid member of Physicians for Responsible Opioid Prescribing. David Juurlink is also a member of the American College of Medical Toxicology. David Juurlink has received payment for lectures from Texas Tech University Health Sciences Centre and payment for expert testimony regarding analgesics, including opioids. He has received travel support for presentations and scientific meetings from the Canadian Institutes of Health Research, Stanford University and Texas Tech University Health Sciences Centre. Jason Busse is the Principal Investigator for the update of the 2017 Canadian Opioid Guideline, which is funded by Health Canada. No other competing interests were declared.

  • This article has been peer reviewed.

  • Contributors: Li Wang and Jason Busse conceived and designed the study. Li Wang, Patrick Hong, Wenjun Jiang, Yasir Rehman, Brian Hong, Rachel Couban and Chunming Wang acquired the data. Li Wang carried out the statistical analysis. Li Wang, Corey Hayes, David Juurlink and Jason Busse interpreted the data. Li Wang and Jason Busse drafted the manuscript. All of the authors revised it critically for important intellectual content, gave final approval of the version to be published and agreed to be accountable for all aspects of the work.

  • Funding: This study was supported by a grant from Health Canada’s Substance Use and Addictions Program. Corey Hayes was supported by Veterans Affairs Health Services Research and Development under a Career Development Award-2 grant. Jason Busse is supported, in part, by a Canadian Institutes of Health Research Canada Research Chair in the prevention and management of chronic pain.

  • Data sharing: Details of the characteristics of the included studies were shared in the supplementary materials. The study protocol is available from the Open Science Framework (https://osf.io/vznxj/). Statistical codes and data can be obtained from the corresponding author.

  • Accepted September 18, 2023.

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) licence, which permits use, distribution and reproduction in any medium, provided that the original publication is properly cited, the use is noncommercial (i.e., research or educational use), and no modifications or adaptations are made. See: https://creativecommons.org/licenses/by-nc-nd/4.0/

References

  1. ↵
    1. Dahlhamer J,
    2. Lucas J,
    3. Zelaya C,
    4. et al
    . Prevalence of chronic pain and high impact chronic pain among adults — United States, 2016. MMWR Morb Mortal Wkly Rep 2018;67:1001–6.
    OpenUrlCrossRefPubMed
    1. Schopflocher D,
    2. Taenzer P,
    3. Jovey R
    . The prevalence of chronic pain in Canada. Pain Res Manag 2011;16:445–50.
    OpenUrlCrossRefPubMed
    1. Sá KN,
    2. Moreira L,
    3. Baptista AF,
    4. et al
    . Prevalence of chronic pain in developing countries: systematic review and meta-analysis. Pain Rep 2019;4:e779.
    OpenUrl
  4. Chronic pain in Australia. Canberra (AU): Australian Institute of Health and Welfare; 2020. Available: https://www.aihw.gov.au/reports/chronic-disease/chronic-pain-in-australia/summary (accessed 2022 Mar. 15).
  5. ↵
    1. Breivik H,
    2. Collett B,
    3. Ventafridda V,
    4. et al
    . Survey of chronic pain in Europe: Prevalence, impact on daily life, and treatment. Eur J Pain 2006;10:287–33.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Wertheimer G,
    2. Mathieson S,
    3. Maher CG,
    4. et al
    . The Prevalence of opioid analgesic use in people with chronic noncancer pain: systematic review and meta-analysis of observational studies. Pain Med 2021;22:506–17.
    OpenUrl
  7. ↵
    Pain relief medication containing opioids, 2018. Ottawa: Statistic Canada; 2019. Cat no 82-625-x. Available: https://www150.statcan.gc.ca/n1/pub/82-625-x/2019001/article/00008-eng.htm (accessed 2023 June 16).
  8. ↵
    1. Busse JW,
    2. Craigie S,
    3. Juurlink DN,
    4. et al
    . Guideline for opioid therapy and chronic noncancer pain. CMAJ 2017;189:E659–66.
    OpenUrlFREE Full Text
  9. ↵
    Health Infobase. Opioid- and stimulant-related harms in Canada. Ottawa: Government of Canada; 2023. Available: https://health-infobase.canada.ca/substance-related-harms/opioids-stimulants/ (accessed 2023 June 16).
  10. ↵
    1. Gomes T,
    2. Khuu W,
    3. Martins D,
    4. et al
    . Contributions of prescribed and non-prescribed opioids to opioid related deaths: population-based cohort study in Ontario, Canada. BMJ 2018;362:k3207.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Wilton J,
    2. Abdia Y,
    3. Chong M,
    4. et al
    . Prescription opioid treatment for non-cancer pain and initiation of injection drug use: large retrospective cohort study. BMJ 2021;375:e066965.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Zullo AR,
    2. Danko KJ,
    3. Moyo P,
    4. et al
    . Prevention, diagnosis, and management of opioids, opioid misuse, and opioid use disorder in older adults. Rockville (MD): Agency for Healthcare Research and Quality; 2020.
  13. ↵
    1. van Draanen J,
    2. Tsang C,
    3. Mitra S,
    4. et al
    . Socioeconomic marginalization and opioid-related overdose: a systematic review. Drug Alcohol Depend 2020;214:108127.
    OpenUrl
  14. ↵
    1. Adewumi AD,
    2. Hollingworth SA,
    3. Maravilla JC,
    4. et al
    . Prescribed dose of opioids and overdose: a systematic review and meta-analysis of unintentional prescription opioid overdose. CNS Drugs 2018;32:101–16.
    OpenUrlPubMed
  15. ↵
    1. Elzey MJ,
    2. Barden SM,
    3. Edwards ES
    . Patient characteristics and outcomes in unintentional, non-fatal prescription opioid overdoses: a systematic review. Pain Physician 2016;19:215–28.
    OpenUrlPubMed
  16. ↵
    1. King NB,
    2. Fraser V,
    3. Boikos C,
    4. et al
    . Determinants of increased opioid-related mortality in the United States and Canada, 1990–2013: a systematic review. Am J Public Health 2014;104:e32–42.
    OpenUrlCrossRefPubMed
  17. ↵
    1. van Draanen J,
    2. Tsang C,
    3. Mitra S,
    4. et al
    . Mental disorder and opioid overdose: a systematic review. Soc Psychiatry Psychiatr Epidemiol 2022;57:647–71.
    OpenUrl
  18. ↵
    1. Stroup DF,
    2. Berlin JA,
    3. Morton SC,
    4. et al
    . Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Riley RD,
    2. Moons KGM,
    3. Snell KIE,
    4. et al
    . A guide to systematic review and meta-analysis of prognostic factor studies. BMJ 2019;364:k4597.
    OpenUrlFREE Full Text
  20. ↵
    1. Randolph AG,
    2. Cook DJ,
    3. Guyatt G
    . Prognosis. In: Guyatt G, Rennie D, O’Meade, et al. editors. JAMA Evidence. Users’ Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice (3rd edition). McGraw Hill; 2015:421–9.
  21. ↵
    1. Landis JR,
    2. Koch GG
    . The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Freeman MF,
    2. Tukey JW
    . Transformations related to the angular and the square root. Ann Math Stat 1950;21:607–11.
    OpenUrlCrossRef
  23. ↵
    1. Kaplovitch E,
    2. Gomes T,
    3. Camacho X,
    4. et al
    . Sex differences in dose escalation and overdose death during chronic opioid therapy: a population-based cohort study. PLoS One 2015;10:e0134550.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Dunn KM,
    2. Saunders KW,
    3. Rutter CM,
    4. et al
    . Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med 2010;152:85–92.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Grant RL
    . Converting an odds ratio to a range of plausible relative risks for better communication of research findings. BMJ 2014;348:f7450.
    OpenUrlFREE Full Text
  26. ↵
    1. Dettori JR,
    2. Norvell DC,
    3. Chapman JR
    . Fixed-effect vs random-effects models for meta-analysis: 3 points to consider. Global Spine J 2022;12:1624–6.
    OpenUrl
  27. ↵
    1. Zedler B,
    2. Xie L,
    3. Wang L,
    4. et al
    . Risk factors for serious prescription opioid-related toxicity or overdose among Veterans Health Administration patients. Pain Med 2014;15:1911–29.
    OpenUrl
  28. ↵
    1. Nadpara PA,
    2. Joyce AR,
    3. Murrelle EL,
    4. et al
    . Risk factors for serious prescription opioid-induced respiratory depression or overdose: comparison of commercially insured and Veterans Health Affairs populations. Pain Med 2018;19:79–96.
    OpenUrl
  29. ↵
    1. Orsini N,
    2. Li R,
    3. Wolk A,
    4. et al
    . Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software. Am J Epidemiol 2012;175:66–73.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Greenland S,
    2. Longnecker MP
    . Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol 1992;135:1301–9.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Crippa A,
    2. Discacciati A,
    3. Bottai M,
    4. et al
    . One-stage dose-response meta-analysis for aggregated data. Stat Methods Med Res 2019;28:1579–96.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Bucher HC,
    2. Guyatt GH,
    3. Griffith LE,
    4. et al
    . The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials. J Clin Epidemiol 1997;50:683–91.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Higgins J,
    2. Thomas J,
    3. Chandler J,
    4. et al
    . Cochrane handbook for systematic reviews of interventions. Version 6.3. Oxford: Cochrane Collaboration; updated Feb. 2022. Available: https://training.cochrane.org/handbook (accessed 2022 Mar. 30).
  34. ↵
    1. Wang L,
    2. Guyatt GH,
    3. Kennedy SA,
    4. et al
    . Predictors of persistent pain after breast cancer surgery: a systematic review and meta-analysis of observational studies. CMAJ 2016;188:E352–61.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Gelman AH,
    2. Hill J
    . Chapter 25. Missing-data imputation. In: Gelman AH, Hill J, editors. Data analysis using regression and multilevel/hierarchical models. New York: Cambridge University Press; 2006:529–44.
  36. ↵
    1. Iorio A,
    2. Spencer FA,
    3. Falavigna M,
    4. et al
    . Use of GRADE for assessment of evidence about prognosis: rating confidence in estimates of event rates in broad categories of patients. BMJ 2015;350:h870.
    OpenUrlFREE Full Text
  37. ↵
    1. Egger M,
    2. Davey Smith G,
    3. Schneider M,
    4. et al
    . Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Schandelmaier S,
    2. Briel M,
    3. Varadhan R,
    4. et al
    . Development of the Instrument to assess the Credibility of Effect Modification Analyses (ICEMAN) in randomized controlled trials and meta-analyses. CMAJ 2020;192:E901–6.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Hartung DM,
    2. Middleton L,
    3. Haxby DG,
    4. et al
    . Rates of adverse events of long-acting opioids in a state Medicaid program. Ann Pharmacother 2007;41:921–8.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Miller M,
    2. Barber CW,
    3. Leatherman S,
    4. et al
    . Prescription opioid duration of action and the risk of unintentional overdose among patients receiving opioid therapy. JAMA Intern Med 2015;175:608–15.
    OpenUrl
  41. ↵
    1. Ray WA,
    2. Chung CP,
    3. Murray KT,
    4. et al
    . Out-of-hospital mortality among patients receiving methadone for noncancer pain. JAMA Intern Med 2015;175:420–7.
    OpenUrl
  42. ↵
    1. Turner BJ,
    2. Liang Y
    . Drug overdose in a retrospective cohort with non-cancer pain treated with opioids, antidepressants, and/or sedative-hypnotics: interactions with mental health disorders. J Gen Intern Med 2015;30:1081–96.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Garg RK,
    2. Fulton-Kehoe D,
    3. Franklin GM
    . Patterns of opioid use and risk of opioid overdose death among Medicaid patients. Med Care 2017;55:661–8.
    OpenUrlPubMed
  44. ↵
    1. Carey CM,
    2. Jena AB,
    3. Barnett ML
    . Patterns of potential opioid misuse and subsequent adverse outcomes in Medicare, 2008 to 2012. Ann Intern Med 2018;168:837–45.
    OpenUrl
  45. ↵
    1. Chung CP,
    2. Dupont WD,
    3. Murray KT,
    4. et al
    . Comparative out-of-hospital mortality of long-acting opioids prescribed for non-cancer pain: a retrospective cohort study. Pharmacoepidemiol Drug Saf 2019;28:48–53.
    OpenUrl
  46. ↵
    1. Glanz JM,
    2. Narwaney KJ,
    3. Mueller SR,
    4. et al
    . Prediction model for two-year risk of opioid overdose among patients prescribed chronic opioid therapy. J Gen Intern Med 2018;33:1646–53.
    OpenUrl
  47. ↵
    1. Bedson J,
    2. Chen Y,
    3. Ashworth J,
    4. et al
    . Risk of adverse events in patients prescribed long-term opioids: a cohort study in the UK Clinical Practice Research Datalink. Eur J Pain 2019;23:908–22.
    OpenUrlCrossRefPubMed
  48. ↵
    1. James JR,
    2. Scott JM,
    3. Klein JW,
    4. et al
    . Mortality after discontinuation of primary care-based chronic opioid therapy for pain: a retrospective cohort study. J Gen Intern Med 2019;34:2749–55.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Young JC,
    2. Lund JL,
    3. Dasgupta N,
    4. et al
    . Opioid tolerance and clinically recognized opioid poisoning among patients prescribed extended-release long-acting opioids. Pharmacoepidemiol Drug Saf 2019;28:39–47.
    OpenUrl
  50. ↵
    1. El-Akkad SE,
    2. Nolan S,
    3. Fairbairn N,
    4. et al
    . The impact of high-dose opioid prescription on mortality rates among people living with HIV: a retrospective cohort study. Int J Drug Policy 2020;78:102705.
    OpenUrl
  51. ↵
    1. Fonda JR,
    2. Gradus JL,
    3. Brogly SB,
    4. et al
    . Traumatic brain injury and opioid overdose among post-9/11 veterans with long-term opioid treatment of chronic pain. J Head Trauma Rehabil 2020;35:209–17.
    OpenUrl
  52. ↵
    1. Li Y,
    2. Delcher C,
    3. Wei YJJ,
    4. et al
    . Risk of opioid overdose associated with concomitant use of opioids and skeletal muscle relaxants: a population-based cohort study. Clin Pharmacol Ther 2020;108:81–9.
    OpenUrlPubMed
  53. ↵
    1. Qeadan F,
    2. Madden EF
    . Associations between naloxone prescribing and opioid overdose among patients with acute and chronic pain conditions. Addiction 2022;117:457–71.
    OpenUrl
  54. ↵
    1. DiPrete BL,
    2. Ranapurwala SI,
    3. Maierhofer CN,
    4. et al
    . Association of opioid dose reduction with opioid overdose and opioid use disorder among patients receiving high-dose, long-term opioid therapy in North Carolina. JAMA Netw Open 2022;5:e229191.
    OpenUrl
  55. ↵
    1. Hayes CJ,
    2. Koonce RM,
    3. Gressler LE,
    4. et al
    . Association between opioid therapy trajectories and potential opioid-related adverse health events. Pharmacoepidemiol Drug Saf 2022;31:1075–90.
    OpenUrl
  56. ↵
    1. Larochelle MR,
    2. Lodi S,
    3. Yan S,
    4. et al
    . Comparative effectiveness of opioid tapering or abrupt discontinuation vs no dosage change for opioid overdose or suicide for patients receiving stable long-term opioid therapy. JAMA Netw Open 2022;5:e2226523.
    OpenUrl
  57. ↵
    1. Lo-Ciganic WH,
    2. Hincapie-Castillo J,
    3. Wang T,
    4. et al
    . Dosing profiles of concurrent opioid and benzodiazepine use associated with overdose risk among US Medicare beneficiaries: group-based multi-trajectory models. Addiction 2022;117:1982–97.
    OpenUrl
  58. ↵
    1. Bohnert AS,
    2. Valenstein M,
    3. Bair MJ,
    4. et al
    . Association between opioid prescribing patterns and opioid overdose-related deaths. JAMA 2011;305:1315–21.
    OpenUrlCrossRefPubMed
  59. ↵
    1. Gwira Baumblatt JA,
    2. Wiedeman C,
    3. Dunn JR,
    4. et al
    . High-risk use by patients prescribed opioids for pain and its role in overdose deaths. JAMA Intern Med 2014;174:796–801.
    OpenUrl
  60. ↵
    1. Gomes T,
    2. Greaves S,
    3. van den Brink W,
    4. et al
    . Pregabalin and the risk for opioid-related death: a nested case-control study. Ann Intern Med 2018;169:732–4.
    OpenUrlCrossRefPubMed
  61. ↵
    1. Glanz JM,
    2. Binswanger IA,
    3. Shetterly SM,
    4. et al
    . Association between opioid dose variability and opioid overdose among adults prescribed long-term opioid therapy. JAMA Netw Open 2019;2:e192613.
    OpenUrl
  62. ↵
    1. Salkar M,
    2. Ramachandran S,
    3. Bentley JP,
    4. et al
    . Do formulation and dose of long-term opioid therapy contribute to risk of adverse events among older adults? J Gen Intern Med 2022;37:367–74.
    OpenUrl
  63. ↵
    1. Hayes CJ,
    2. Krebs EE,
    3. Hudson T,
    4. et al
    . Impact of opioid dose escalation on the development of substance use disorders, accidents, self-inflicted injuries, opioid overdoses and alcohol and non-opioid drug-related overdoses: a retrospective cohort study. Addiction 2020;115:1098–112.
    OpenUrlCrossRef
  64. ↵
    1. Agnoli A,
    2. Xing G,
    3. Tancredi DJ,
    4. et al
    . Association of dose tapering with overdose or mental health crisis among patients prescribed long-term opioids. JAMA 2021;326:411–9.
    OpenUrlCrossRefPubMed
  65. ↵
    1. Hayes CJ,
    2. Hudson T,
    3. Krebs EE,
    4. et al
    . Association between discontinuing chronic opioid therapy and newly diagnosed substance use disorders, accidents, self-inflicted injuries and drug overdoses within the prescribers’ health care system: a retrospective cohort study. Addiction 2022;117:946–68.
    OpenUrl
  66. ↵
    1. Hayes CJ,
    2. Krebs EE,
    3. Brown J,
    4. et al
    . Impact of transitioning from long-term to intermittent opioid therapy on the development of opioid-related adverse outcomes: a retrospective cohort study. Drug Alcohol Depend 2022;231:109236.
    OpenUrl
  67. ↵
    1. Gomes T,
    2. Mamdani MM,
    3. Dhalla IA,
    4. et al
    . Opioid dose and drug-related mortality in patients with nonmalignant pain. Arch Intern Med 2011;171:686–91.
    OpenUrlCrossRefPubMed
    1. Ilgen MA,
    2. Bohnert AS,
    3. Ganoczy D,
    4. et al
    . Opioid dose and risk of suicide. Pain 2016;157:1079–84.
    OpenUrlCrossRefPubMed
  69. ↵
    1. Gomes T,
    2. Juurlink DN,
    3. Antoniou T,
    4. et al
    . Gabapentin, opioids, and the risk of opioid-related death: a population-based nested case-control study. PLoS Med 2017;14:e1002396.
    OpenUrlCrossRefPubMed
    1. Park TW,
    2. Saitz R,
    3. Ganoczy D,
    4. et al
    . Benzodiazepine prescribing patterns and deaths from drug overdose among US veterans receiving opioid analgesics: case-cohort study. [Online]. BMJ 2015;350:h2698.
    OpenUrlAbstract/FREE Full Text
    1. Boscarino JA,
    2. Kirchner HL,
    3. Pitcavage JM,
    4. et al
    . Factors associated with opioid overdose: a 10-year retrospective study of patients in a large integrated health care system. Subst Abuse Rehabil 2016;7:131–41.
    OpenUrlCrossRef
    1. Liang Y,
    2. Goros MW,
    3. Turner BJ
    . Drug overdose: differing risk models for women and men among opioid users with non-cancer pain. Pain Med 2016;17:2268–79.
    OpenUrl
    1. Cochran G,
    2. Gordon AJ,
    3. Lo-Ciganic WH,
    4. et al
    . An examination of claims-based predictors of overdose from a large Medicaid program. Med Care 2017;55:291–8.
    OpenUrl
    1. Sun EC,
    2. Dixit A,
    3. Humphreys K,
    4. et al
    . Association between concurrent use of prescription opioids and benzodiazepines and overdose: retrospective analysis. BMJ 2017;356:j760.
    OpenUrlAbstract/FREE Full Text
    1. Ray BR,
    2. Lowder EM,
    3. Kivisto AJ,
    4. et al
    . EMS naloxone administration as non-fatal opioid overdose surveillance: 6-year outcomes in Marion County, Indiana. Addiction 2018;113:2271–9.
    OpenUrl
    1. Groenewald CB,
    2. Zhou C,
    3. Palermo TM,
    4. et al
    . Associations between opioid prescribing patterns and overdose among privately insured adolescents. Pediatrics 2019;144:e20184070.
    OpenUrlCrossRefPubMed
    1. Moyo P,
    2. Zhao X,
    3. Thorpe CT,
    4. et al
    . Dual receipt of prescription opioids from the Department of Veterans Affairs and Medicare Part D and prescription opioid overdose death among veterans: a nested case-control study. Ann Intern Med 2019;170:433–42.
    OpenUrl
  78. ↵
    1. Chua KP,
    2. Brummett CM,
    3. Conti RM,
    4. et al
    . Association of opioid prescribing patterns with prescription opioid overdose in adolescents and young adults. JAMA Pediatr 2020;174:141–8.
    OpenUrl
    1. Nam YH,
    2. Bilker WB,
    3. DeMayo FJ,
    4. et al
    . Incidence rates of and risk factors for opioid overdose in new users of prescription opioids among US Medicaid enrollees: a cohort study. Pharmacoepidemiol Drug Saf 2020;29:931–8.
    OpenUrl
    1. Smolina K,
    2. Crabtree A,
    3. Chong M,
    4. et al
    . Prescription-related risk factors for opioid-related overdoses in the era of fentanyl contamination of illicit drug supply: a retrospective case-control study. Substance Abuse 2022;43:92–8.
    OpenUrl
    1. Liang Y,
    2. Turner BJ
    . Assessing risk for drug overdose in a national cohort: role for both daily and total opioid dose? J Pain 2015;16:318–25.
    OpenUrlCrossRefPubMed
    1. Deng J,
    2. Hou W,
    3. Dong XY,
    4. et al
    . A large-scale observational study on the temporal trends and risk factors of opioid overdose: real-world evidence for better opioids. Drugs Real World Outcomes 2021;8:393–406.
    OpenUrlPubMed
    1. Oliva EM,
    2. Bowe T,
    3. Manhapra A,
    4. et al
    . Associations between stopping prescriptions for opioids, length of opioid treatment, and overdose or suicide deaths in US veterans: observational evaluation. BMJ 2020;368:m283.
    OpenUrlAbstract/FREE Full Text
    1. Weiner SG,
    2. El Ibrahimi S,
    3. Hendricks MA,
    4. et al
    . Factors associated with opioid overdose after an initial opioid prescription. JAMA Netw Open 2022;5:e2145691.
    OpenUrl
    1. Weiner SG,
    2. Hendricks MA,
    3. El Ibrahimi S,
    4. et al
    . Opioid-related overdose and chronic use following an initial prescription of hydrocodone versus oxycodone. PLoS One 2022;17:e0266561.
    OpenUrl
  86. ↵
    1. Dunham JR,
    2. Highland KB,
    3. Costantino R,
    4. et al
    . Evaluation of an opioid overdose composite risk score cutoff in active-duty military service members. Pain Med 2022;23:1902–7.
    OpenUrl
  87. ↵
    1. Dowell D,
    2. Haegerich TM,
    3. Chou R
    . CDC guideline for prescribing opioids for chronic pain — United States, 2016. JAMA 2016;315:1624–45.
    OpenUrlCrossRefPubMed
  88. ↵
    1. Rosenberg JM,
    2. Bilka BM,
    3. Wilson SM,
    4. et al
    . Opioid therapy for chronic pain: overview of the 2017 US Department of Veterans Affairs and US Department of Defense Clinical Practice Guideline. Pain Med 2018;19:928–41.
    OpenUrl
  89. ↵
    1. Kroenke K,
    2. Alford DP,
    3. Argoff C,
    4. et al
    . Challenges with implementing the Centers for Disease Control and Prevention opioid guideline: a consensus panel report. Pain Med 2019;20:724–35.
    OpenUrl
  90. ↵
    1. Dowell D,
    2. Haegerich T,
    3. Chou R
    . No shortcuts to safer opioid prescribing. N Engl J Med 2019;380:2285–7.
    OpenUrlPubMed
  91. ↵
    1. Darnall BD,
    2. Ziadni MS,
    3. Stieg RL,
    4. et al
    . Patient-centered prescription opioid tapering in community outpatients with chronic pain. JAMA Intern Med 2018;178:707–8.
    OpenUrl
  92. ↵
    1. Ziadni M,
    2. Chen AL,
    3. Krishnamurthy P,
    4. et al
    . Patient-centered prescription opioid tapering in community outpatients with chronic pain: 2- to 3-year follow-up in a subset of patients. Pain Rep 2020;5:e851.
    OpenUrl
  93. ↵
    1. Sandhu HK,
    2. Booth K,
    3. Furlan AD,
    4. et al
    . Reducing opioid use for chronic pain with a group-based intervention: a randomized clinical trial. JAMA 2023;329:1745–56.
    OpenUrl
  94. ↵
    1. Peirce GL,
    2. Smith MJ,
    3. Abate MA,
    4. et al
    . Doctor and pharmacy shopping for controlled substances. Med Care 2012;50:494–500.
    OpenUrlCrossRefPubMed
  95. ↵
    1. Rigg KK,
    2. March SJ,
    3. Inciardi JA
    . Prescription drug abuse & diversion: role of the pain clinic. J Drug Issues 2010;40:681–702.
    OpenUrlCrossRefPubMed
  96. ↵
    1. Chua KP,
    2. Brummett CM,
    3. Conti RM,
    4. et al
    . Assessment of prescriber and pharmacy shopping among the family members of patients prescribed opioids. JAMA Netw Open 2019;2:e193673.
    OpenUrl
  97. ↵
    1. Rhodes E,
    2. Wilson M,
    3. Robinson A,
    4. et al
    . The effectiveness of prescription drug monitoring programs at reducing opioid-related harms and consequences: a systematic review. BMC Health Serv Res 2019;19:784.
    OpenUrlPubMed
  98. ↵
    1. D’Souza RS,
    2. Lang M,
    3. Eldrige JS
    . Prescription Drug Monitoring Program. Stat-Pearls. Treasure Island (FL); 2022.
  99. ↵
    1. Klochkov AKP,
    2. Lim Y,
    3. Sun Y
    . Alcoholic pancreatitis. Treasure Island (FL): Stat-Pearls; updated 2022 May 23.
  100. ↵
    1. O’Brien DC,
    2. Dabbs D,
    3. Dong K,
    4. et al
    . Patient characteristics associated with being offered take home naloxone in a busy, urban emergency department: a retrospective chart review. BMC Health Serv Res 2019;19:632.
    OpenUrl
  101. ↵
    1. Klimas J,
    2. Gorfinkel L,
    3. Fairbairn N,
    4. et al
    . Strategies to identify patient risks of prescription opioid addiction when initiating opioids for pain: a systematic review. JAMA Netw Open 2019;2:e193365.
    OpenUrl
    1. Lo-Ciganic WH,
    2. Huang JL,
    3. Zhang HH,
    4. et al
    . Evaluation of machine-learning algorithms for predicting opioid overdose risk among Medicare beneficiaries with opioid prescriptions. JAMA Netw Open 2019;2:e190968.
    OpenUrl
  103. ↵
    1. Zedler B,
    2. Xie L,
    3. Wang L,
    4. et al
    . Development of a risk index for serious prescription opioid-induced respiratory depression or overdose in Veterans’ Health Administration patients. Pain Med 2015;16:1566–79.
    OpenUrl
  104. ↵
    1. Goshua A,
    2. Craigie S,
    3. Guyatt GH,
    4. et al
    . Patient values and preferences regarding opioids for chronic noncancer pain: a systematic review. Pain Med 2018;19:2469–80.
    OpenUrl
  105. ↵
    1. Gladstone E,
    2. Smolina K,
    3. Morgan SG,
    4. et al
    . Sensitivity and specificity of administrative mortality data for identifying prescription opioid-related deaths. CMAJ 2016;188:E67–72.
    OpenUrlAbstract/FREE Full Text
  106. ↵
    1. Landen MG,
    2. Castle S,
    3. Nolte KB,
    4. et al
    . Methodological issues in the surveillance of poisoning, illicit drug overdose, and heroin overdose deaths in New Mexico. Am J Epidemiol 2003;157:273–8.
    OpenUrlCrossRefPubMed
  107. ↵
    1. Deeks JJ,
    2. Dinnes J,
    3. D’Amico R,
    4. et al
    . Evaluating non-randomised intervention studies. Health Technol Assess 2003;7:iii-x, 1–173.
    OpenUrlCrossRefPubMed
  108. ↵
    1. Rennick A,
    2. Atkinson T,
    3. Cimino NM,
    4. et al
    . Variability in opioid equivalence calculations. Pain Med 2016;17:892–8.
    OpenUrl
Back to top

In this issue

Canadian Medical Association Journal: 195 (41)
CMAJ
Vol. 195, Issue 41
23 Oct 2023

Article tools

Respond to this article
Print
Download PDF
Article Alerts
Email Article
Citation Tools
Predictors of fatal and nonfatal overdose after prescription of opioids for chronic pain: a systematic review and meta-analysis of observational studies
Li Wang, Patrick J. Hong, Wenjun Jiang, Yasir Rehman, Brian Y. Hong, Rachel J. Couban, Chunming Wang, Corey J. Hayes, David N. Juurlink, Jason W. Busse
CMAJ Oct 2023, 195 (41) E1399-E1411; DOI: 10.1503/cmaj.230459
Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Collections

Get the latest content from Canada's leading medical journal, free to all physicians. Click to sign up for CMAJ or JAMC email alerts.

 

View Latest Classified Ads