Trends of long-term opioid therapy and subsequent discontinuation among people with chronic non-cancer pain in UK primary care: A retrospective cohort study

Abstract

Objectives

To examine trends of opioid use, focusing on long-term opioid therapy (L-TOT) and its discontinuation among people with chronic non-cancer pain (CNCP).

Design and setting

Retrospective cohort study using UK Clinical Practice Research Datalink Aurum data.

Population

Incident opioid users (no opioid use in the prior year) with CNCP between 01/01/2009–31/12/2019. Among them, we identified L-TOT users (≥3 opioid prescriptions within 90 days, or total ≥90 supply days within the first year, excluding the initial 30 days) and L-TOT discontinuers (no opioid use for ≥180 days following a L-TOT).

Main outcome measures

Yearly rates of incident opioid users (over CPRD-registered patients), L-TOT users (over incident opioid users), and L-TOT discontinuers (over L-TOT users) were calculated. Annual counts of each group were fitted using segmented negative binomial regression models with an offset considering their corresponding denominators from 2009 to 2019, excluding 2014 due to policy changes in that year.

Results

Among 2,839,161 incident opioid users, 11.4% (n = 324,877) transitioned into L-TOT users within one year, of which 4.8% (n = 15,484) discontinued. Between 2009–2013, rates of L-TOT users significantly declined by 2.6% (incidence rate ratio: 0.974; 95% confidence interval: 0.971 to 0.978) per annum, followed by a significant step change in 2015 (1.026, 1.009 to1.044), and a significant annual increase of 2.4% (1.024, 1.019 to 1.029) from 2015 to 2019, compared to the 2009–2013 trend. The annual rates of L-TOT discontinuers remained stable from 2009 to 2013 (0.987, 0.971 to 1.002), followed by a non-significant step change (0.990, 0.916 to 1.070) in 2015, and a significant decrease in slope by 2.6% (0.974, 0.951 to 0.998) per annum during 2015–2019, relative to 2009–2013.

Conclusions

L-TOT has plateaued since 2015, accompanied with an accelerated decrease in discontinuation rates, suggesting ongoing reliance on opioids for CNCP management, despite increased awareness regarding L-TOT associated risks. This is likely due to the limited availability of other effective pharmacological options and non-pharmacological alternatives, and challenges in their accessibility.

Introduction

Opioid prescribing for chronic non-cancer pain (CNCP) has increased markedly in the United Kingdom (UK) over the past two decades [1,2], raising particular concerns for these people. Due to the complexity of pain management, CNCP patients are more likely to transition to long-term opioid therapy (L-TOT), which in turn exposes them to increased risks of dependency, addiction, bone fractures and opioid-related death [3–5]. In response, several national prescribing guidelines [6–9] in the UK have recommended reducing opioid prescribing, thereby to mitigate L-TOT and its associated harms. In fact, prior to the implementation of these guidelines, growing awareness among healthcare professionals about the adverse effects of opioids, especially in their long-term use, had already led to a levelling-off in opioid prescribing since 2014 [3,10]. Regulatory interventions, such as the 2014 reclassification of tramadol as a Schedule III controlled drug [11], which imposed stricter prescribing and dispensing controls of tramadol, and the earlier 2013 national Controlled Drugs (Supervision of Management and Use) Regulations [12], which aimed at improving the monitoring and use of controlled drugs such as opioids, may have further promoted this trend. These changes have marked an important turning point, calling for further examination of how the trend of opioid prescribing has changed over time. Such an analysis will help assess whether current practices align with recommendations, particularly for L-TOT and its subsequent discontinuation in UK primary care, where opioids are predominantly prescribed [13].

Current evidence on the trends of L-TOT primarily came from the United States (US) and focused on the general population [14]. In the UK, although some studies have reported such data [2,15], they were limited to specific conditions, such as musculoskeletal pain or broader non-cancer pain (not CNCP), and were restricted to earlier periods, up to 2013. Therefore, there is a gap in understanding the trends of L-TOT beyond this period, especially within the CNCP population, who often require long-term pain management and face greater risks from L-TOT than the general population [16,17]. L-TOT discontinuation is essential in reducing potential opioid-related harms. Understanding its level and trend over a recent period will help to identify how well policy and clinical practice have affected the management of L-TOT in UK primary care. However, current evidence regarding such information is lacking, and the reported discontinuation rates derived from US-based studies ranging widely from ~7% to 72% [18–23], depending on the definitions and methodologies used. These findings may not be directly applicable to the UK context due to the differences in healthcare systems, prescribing practices, and patient demographics, which highlight the need for research in this area.

This study aimed to address these gaps by examining patterns and trends of opioid use among people with CNCP in UK primary care, with a special focus on L-TOT and its discontinuation. By analysing these trends, this study sought to provide insights to inform future policies and clinical practice, ultimately promoting safer opioid prescribing and improving the management of CNCP in UK primary care.

Methods

Study design and population

In this retrospective cohort study, we accessed data from the Clinical Practice Research Datalink (CPRD) Aurum (the June 2021 version) on 21^st December, 2023, a database of anonymized UK primary care electronic health records, covering 20% of the national population (approximately 13.3 million active registrants) [24]. Among all registered patients in the CPRD Aurum (S1 Fig), we included those aged ≥18 years on the index date (the first opioid prescription following a 12-month opioid-free period) with a diagnosis of CNCP (code lists are provided in S1 Table), and received ≥1 opioid prescription between 1^st January 2009 and 31^st December 2019. CNCP diagnosis could occur up to 6 months after or any time before the index date. Patients with a cancer diagnosis, except for non-melanoma skin cancer (NMSC), in the previous ten years were excluded due to different drug utilisation patterns and different guidelines for opioid prescribing. We did not exclude patients with NMSC as it is not typically associated with chronic pain or long-term opioid use. This is consistent with standard practice in CNCP research [2,25–28], and avoids unnecessary exclusion of relevant patients. Additionally, NMSC is often underreported in UK primary care records, making it an unreliable exclusion criterion [29]. To be eligible, patients were required to have been registered with the CPRD practice for at least 12 months, and have available linked data for both the Index of Multiple Deprivation (IMD) 2019 and the Hospital Episode Statistics Admitted Patient Care database (HES APC).

The study protocol was approved by the CPRD’s Independent Scientific Advisory Committee (protocol number: 23_002909, available at https://www.cprd.com/approved-studies/predictors-discontinuation-or-reduction-long-term-opioid-therapy-and-its). CPRD also has ethical approval from the Health Research Authority to support research using anonymized patient data (research ethics committee reference 21/EM/0265). Individual patient consent was not required as all data were deidentified. This study was reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement (S2 Table) [30].

Opioid measures

All patients were followed from the index date until the earliest occurrence of death, transfer out of the general practice, the last data collection date of the general practice, a cancer diagnosis, or the end of the study (31 December 2019). Incident opioid users were defined as having no opioid prescription record for ≥12 months prior to the date of the next prescription. L-TOT users were defined as receiving ≥ 3 opioid prescriptions within a 90-day period, or a sum of opioid supply days lasting ≥ 90 days within the first year of follow-up, excluding the initial 30 days [2]. L-TOT discontinuers were defined as no opioid use for ≥180 days following an episode of L-TOT. The choice of a 180-day discontinuation interval is to ensure that discontinuation was an intentional treatment decision rather than factors that may delay prescription refills, such as geographic relocation, practice transfers, or extended inpatient hospitalizations. The definitions of L-TOT users and L-TOT discontinuers are shown in S2 Fig.

All nonparental opioid analgesics (code list is provided in S3 Table) were included, except for methadone and sublingual buprenorphine, as their primary indication is for opioid addiction rather than pain management. Opioid drug exposure data were processed using the logic of a previously published drug preparation algorithm [2,31]. Details of the relevant decision process are provided in S3 Fig. Opioid doses were converted to oral morphine milligram equivalents (MME) using standardized conversion factors [32].

Covariates

Baseline patient characteristics (i.e., age, sex) were collected on the index date. Ethnicity data was obtained through the linkage with HES APC data [33]. Regions were classified as North England (North East, North West, Yorkshire & the Humber), Midland (East Midlands, West Midlands), South England (South West, South Central, South East Coast, East of England), and London. Socioeconomic status was measured using the IMD 2019 [34] and categorized into quintiles (Q1-least deprived, Q5-most deprived).

Comorbidities were measured by the Charlson comorbidity index (CCI) score and categorized into three grades: low (0–2); medium (3–4); and high (≥5) [35]. The presence of other conditions, including substance use disorders (SUDs), alcohol dependence, anxiety, depression, schizophrenia, osteoarthritis, rheumatoid arthritis, and epilepsy were identified within five years prior to the index date. We also identified concurrent drug use including benzodiazepines, gabapentinoids, antidepressants, Z-drugs, muscle relaxants or non-steroidal anti-inflammatory drugs (NSAIDs) within five years prior to the index date. We applied this 5-year look-back period to minimize misclassification of prior exposures and comorbidities, based on evidence [36–39] suggesting that shorter windows (e.g., 1–2 years) may lead to substantial misclassification, particularly for intermittent treatments and diagnoses.

Statistical analysis

Descriptive statistics were used to present patient baseline characteristics, including demographics, socioeconomic status, lifestyle (i.e., drinking and smoking status), comorbidities, and concurrent drug use. For categorical variables (i.e., ethnicity, IMD, drinking status, and smoking status), missing data were categorized as a separate group of “unknown” and included in the regression models. The yearly rates of incident opioid users, L-TOT users, and L-TOT discontinuers were calculated using below formula:

Annual counts of each group (2009–2019) were fitted using segmented negative binomial regression models, with an offset term based on the natural logarithm of their corresponding denominators: CPRD registered patients for incident opioid users, incident opioid users for L-TOT users, and L-TOT users for discontinuers. The year 2014 was selected as the breakpoint based on previous studies [10,40] and structural break analysis (S4 Table). However, we a-priori decided to exclude the 2014 data from our main analyses to avoid introducing potential bias from a partial effect of the tramadol reclassification policy in that year [11].

All models included: 1) a year variable indicating the yearly trend from 2009 to 2013 (β1); 2) a binary indicator variable differentiating two time periods (coded ‘0’ for the period of 2009–2013 and ‘1’ for the period of 2015–2019), reflecting the step (level) change (β2) between the observed in 2015 and the projected in 2015 (a linear projection obtained based on the 2009–2013 data); and 3) an interaction term between year and the indicator variable, quantifying the annual slope change (β3) during 2015–2019 compared to the 2009–2013 slope. No subgroup or interaction analyses were conducted. All data analyses were conducted with Stata/MP 18.0.

Sensitivity analysis

To assess the robustness of our results, we conducted two sensitivity analyses: 1) including the year 2014. 2) we performed Poisson regression as an alternative to negative binomial regression.

Result

Patient characteristics

This study included 2,839,161 incident opioid users between 1^st January 2009 and 31^st December 2019, and the median follow-up duration was 6.06 years (interquartile range: 2.74 to 10.66) for the cohort. Among these incident opioid users, 11.4% (n = 324,877) transitioned to L-TOT users within one year, of whom 4.8% (n = 15,484) subsequently discontinued in the following one year (S1 Fig). Table 1 summarized the demographics, comorbidities, concurrent medication use, and opioid dosage and formulation for L-TOT users and L-TOT discontinuers. In both groups, over 78% were white, and females constituted more than 57%. Missing data for variables such as smoking status, IMD, ethnicity, and drinking status ranged from 0.1% to 30% and were handled by including a separate “unknown” category in the analysis (Table 1). L-TOT discontinuers were generally younger (mean age: 55.58 ± 18.87 years) and had a lower average daily dose (18.80 ± 11.32 MME mg/day). Compared to L-TOT users, they also had higher proportions of anxiety (35.26%), depression (43.46%), lower CCI scores (73.88% had a score of 0–2), lived in London (15.38%), used weak opioids (82.83%) or short-acting opioids (96.76%), and had concurrent use of NSAIDs (63.87%), and benzodiazepines (22.35%).

[Figure omitted. See PDF.]

Trends for incident opioid users, L-TOT users and L-TOT discontinuers

Tables 2 and 3 and Fig 1 illustrated the trends of incident opioid users, L-TOT users, and L-TOT discontinuers. Between 2009 and 2013, the yearly rates of incident opioid users showed a flat trend (incidence rate ratio: 1.001, 95% confidence interval: 0.988 to 1.014). Although there was a non-significant step change in 2015 (0.988, 0.928 to 1.052), their trend from 2015 to 2019 showed a significant annual decrease of 8.7% (0.913, 0.897 to 0.930), compared to the slope during 2009–2013.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

For L-TOT users, during 2009–2013, the yearly rates exhibited a significant decline of 2.6% (0.974, 0.971 to 0.978) per annum, followed by a statistically significant step change in 2015 (1.026, 1.009 to 1.044), and an annual increase of 2.4% thereafter from 2015 to 2019 (1.024, 1.019 to 1.029), relative to the trend from 2009 to 2013. In summary, the trend for L-TOT users decreased initially but plateaued in recent years.

The annual rates of L-TOT discontinuation showed a relatively stable trend from 2009 to 2013, with a non-significant annual decrease of 1.3% (0.987, 0.971 to 1.002), followed by a non-significant step change in 2015. This downward trend continued but became significant during 2015–2019, with a decrease of 2.6% per annum (0.974, 0.951 to 0.998) compared to the slope between 2009 and 2013. In summary, L-TOT discontinuation began to decrease significantly only from 2015 onwards.

Sensitivity analysis

The sensitivity analysis including the year 2014 data showed that the pre-post 2014 trends of yearly rates for incident opioid users and L-TOT users maintained consistent with the main analyses (S5 Table). However, the annual L-TOT discontinuation rates in the pre-2014 trend became significant (0.980, 0.969 to 0.992), whereas the post-2014 slope change turned to be non-significant (0.981, 0.960 to 1.003). However, despite these changes in significance, the overall trend indicated a steady decline in discontinuation rates throughout the study period from 2009 to 2019. Results from Poisson regressions were consistent with those obtained from the negative binominal regressions (S6 Table).

Discussion

In this population-based retrospective cohort study, we analyzed the trends of incident opioid users, L-TOT users and L-TOT discontinuers from 2009–2019, using 2014 as the change point. Between 2009–2013, rates of incident opioid users remained stable but showed a significant annual decline from 2015 onward, highlighting the effectiveness of the 2014 tramadol reclassification [11] and other regulatory interventions [12] in reducing opioid initiations. However, the trend for L-TOT users plateaued since 2015, despite a previously downward trend between 2009–2013. This stabilisation can be attributed, at least partly, to the sharp reduction in incident opioid users, which resulted in a smaller pool of patients eligible for transitioning to L-TOT. At the same time, this shift highlights a persistent reliance on opioids for managing CNCP in certain patient groups, for whom, L-TOT may still be perceived as essential despite the known risks and ongoing policy interventions.

Regarding L-TOT discontinuation, the overall rate remained low at 4.8%, indicating substantial barriers to discontinuation and many patients being exposed to the associated risks. Factors contributing to this may include physiological dependence, fear of withdrawal symptoms, limited access to effective non-opioid pain management strategies, and clinicians’ hesitancy to initiate reductions [41–43]. Over time, L-TOT discontinuation rates steadily declined, with a marked drop after 2015. This decline likely reflects changes in both the numerator (fewer patients discontinuing L-TOT) and the denominator (a changing L-TOT user population). With fewer new patients initiating opioids, the remaining L-TOT population may increasingly consist of individuals with more complex, treatment-resistant conditions. These patients are less likely to discontinue opioids due to their greater reliance on L-TOT for managing severe or persistent pain. Additionally, the impact of regulatory measures and policies may have disproportionately affected new opioid initiations rather than existing L-TOT users, contributing to the observed trends.

Comparison with other studies

Our study identified an 11.4% rate of L-TOT within the CNCP population, which is comparable to the 14.6% prevalence reported in prior research using CPRD GOLD data among non-cancer pain patients from 2016 to 2021 [2]. Both studies applied the same definition of L-TOT; however, the slightly lower prevalence in our findings may be explained by the differences in the code lists used to define pain conditions.

In terms of trends, our findings regarding the significant decrease in L-TOT rates from 2009 to 2013, align with an observational study using CPRD GOLD data [15]. That study reported a stabilisation in annual long-term opioid incidence (defined as receiving ≥2 opioid prescriptions within 90 days) among patients with musculoskeletal pain from 2009 to 2011, followed by a slight decline by 2013. While there are differences in study populations and definitions of L-TOT, the trends are generally comparable.

In contrast, data from the NHS Business Services Authority (BSA)’s medication safety indicators dashboard showed a 54% increase in the number of individuals using oral or transdermal opioids for longer than three months in England between 2015 and 2021 [44]. This finding was contradictory to our study, which observed a plateau in L-TOT rates during the similar timeframe. The discrepancies may be explained by differences in the definition of L-TOT and study design, i.e., the NHS BSA data does not specify its study population and likely includes a broader demographic beyond CNCP patients. However, despite these differences, both our study and the NHS BSA data indicated that L-TOT use has not significantly declined in recent years. This persistent trend again, highlighted the ongoing challenges in stopping long-term opioid prescribing.

Focusing on L-TOT discontinuation, current evidence on its prevalence is predominantly derived from US-based studies, with reported rates varying widely from ~7% to 72%, often among veterans with CNCP or populations receiving high-dose chronic opioid therapy (>90 MME mg/day) [18–23]. Our study identified a lower L-TOT discontinuation rate of 4.8% in the CNCP population, who were prescribed with lower daily doses (mean ~18 MME mg/day). This conservative opioid prescribing pattern reflected a cautious approach taken by general practitioners (GPs) in the UK, aligning with current clinical guidelines [45], which recommend minimising opioid doses not to exceed 50 MME mg/day whenever possible. However, it is important to note that differences in study populations, settings, and definitions of both L-TOT and discontinuation limit the direct comparability of our findings with those from the US. Nevertheless, this low discontinuation rate emphasized two challenges: first, an ongoing reliance on opioids for managing CNCP in UK primary care despite guidelines [7,46] discouraging them as the first-line treatment; second, the significant difficulties mentioned above in stopping L-TOT among this population.

To date, no UK-based studies have reported the trends of L-TOT discontinuation. The only evidence came from a US study [18], which reported an increase in annual L-TOT discontinuation rates from 5.7% in 2012 to 8.5% in 2017 among Medicare beneficiaries with non-cancer pain. While this upward trend contrasted with our observed declining-to-flat L-TOT discontinuation rates, these results were not comparable due to variations in definitions, study populations, healthcare systems, and pain conditions. Nevertheless, both studies highlighted the reliance on L-TOT, as noted by the US authors in their paper, that the majority of patients remained on L-TOT every year.

Unlike the US, where opioid monitoring programmes [47] and proactive opioid taper decision tools [48] are often available, the UK currently lacks such tailored interventions, to support opioid discontinuation in primary care. Bridging this gap will require the development of evidence-based, patient-centred approaches, along with alternative pain management strategies that align with current clinical guidelines. Such efforts are critical to overcoming the barriers to L-TOT discontinuation and reducing opioid reliance in the management of CNCP.

Strengths and limitations of this study

This study has several strengths. First, to our knowledge, this is the first study using CPRD Aurum data to analyze trends in opioid use, particularly focusing on the L-TOT and its subsequent discontinuation among people with CNCP in UK primary care. Previous research has not addressed the trends in L-TOT discontinuation, making this study an important addition to the literature. Second, CPRD Aurum database provided comprehensive, high-quality data [49], and the large and nationally representative sample ensured the robustness and generalisability of our findings to the UK population. Finally, the code lists used in this study were developed and validated by a research team member (CM), which ensured the methodological rigor.

Some limitations need to be acknowledged. First, while the CPRD Aurum database provides comprehensive information on prescribed opioids, it does not capture over-the-counter opioids such as codeine and dihydrocodeine, which may potentially underestimate total opioid utilisation in the study population. Second, the availability of prescription records does not guarantee that patients actually used the medications as prescribed, as adherence cannot be directly observed. This may introduce misclassification bias, a limitation inherent to studies relying on CPRD data and prescription records. Third, the findings may not be fully generalisable to other healthcare settings or populations with different prescribing practices or guidelines. Fourth, although the analysis focused on trends before and after 2014, these trends may have been influenced by other concurrent national policies on analgesic prescribing and the exclusion of this year may limit the ability to fully assess the impact of policy change on prescribing practices. Fifth, the regression analysis did not account for all potential confounding factors (e.g., pain severity, marital status, healthcare insurance, etc.), which would provide a more complete picture of opioid use and its effects. Sixth, selection bias may be present due to regional variability in practice participation within the CPRD Aurum database. For example, the underrepresentation of certain regions (e.g., the North East [1.5%]) could limit the generalisability of findings to those areas [50]. Lastly, we focused on discontinuation rather than dose reductions. Many L-TOT users may reduce their doses without fully stopping, which could potentially underestimate efforts to reduce opioid prescribing in primary care.

Clinical implications

This study has important clinical implications for L-TOT discontinuation and the management of CNCP in UK primary care. Incident opioid prescribing for CNCP has decreased in recent years, reflecting the impact of policies and growing awareness of opioid-related risks. Although these efforts have contributed to a reduction in L-TOT use, achieving a consistent decline remains challenging. It is concerning that 11.4% of individuals still transitioned to L-TOT, highlighting the difficulties faced by clinicians and patients in stopping or reducing opioid use [51]. These challenges include the lack of safe and effective pharmacological alternatives, the complex nature of chronic pain, and barriers that prevent patients from accessing or engaging in non-pharmacological pain treatments that they find acceptable, such as physical therapy, exercise, or cognitive-behavioral therapy. Often these treatments and therapies need adapting for marginalized groups.

The decline in L-TOT discontinuation rates further suggested that patients are struggling with stopping L-TOT. Alongside carefully assessing the necessity of opioid prescriptions, and limiting duration of prescriptions, our results suggested GPs should aim to identify patients who are likely to succeed in discontinuation. To support GPs in this, clear discontinuation protocols need to be developed at a local and/or national level [6]. Furthermore, adopting a holistic pain management approach, addressing both physical, social and psychological aspects of CNCP could complement these efforts to improve patient outcomes and promote safer opioid use [52].

Conclusions

In conclusion, this study provided valuable insights into the trends of L-TOT use and its discontinuation for CNCP in UK primary care. The decreasing-to-flat trend of L-TOT users and the consistent declining trend of L-TOT discontinuers suggested an ongoing reliance on opioids for CNCP management despite policy impacts and growing awareness of L-TOT associated risks, which likely due to the lack of safe and effective pharmacological alternatives, the complexity of chronic pain management, and factors making it difficult for patients to access or engage in acceptable non-pharmacological pain treatments. These findings highlighted the need for sustained efforts to promote safer prescribing and discontinuation practices. Future research should identify key factors that contribute to successful L-TOT discontinuation, barriers which prevent discontinuation, and evaluate the effectiveness of opioid reduction strategies.

Supporting information

S1 Fig. Flowchart of identification of incident opioid users, L-TOT users and L-TOT discontinuers from 2009–2019.

https://doi.org/10.1371/journal.pone.0326604.s001

(DOCX)

S2 Fig. Definition of L-TOT users and L-TOT discontinuers.

https://doi.org/10.1371/journal.pone.0326604.s002

(DOCX)

S3 Fig. Decisions made for opioid drug preparation to derive daily dose.

https://doi.org/10.1371/journal.pone.0326604.s003

(DOCX)

S1 Table. CNCP code list.

https://doi.org/10.1371/journal.pone.0326604.s004

(DOCX)

S2 Table. STROBE Statement—Checklist of items that should be included in reports of cohort studies.

https://doi.org/10.1371/journal.pone.0326604.s005

(DOCX)

S3 Table. Opioid product code list.

https://doi.org/10.1371/journal.pone.0326604.s006

(DOCX)

S4 Table. Results of structural break analysis using annual number of opioid users between 2000–2020.

https://doi.org/10.1371/journal.pone.0326604.s007

(DOCX)

S5 Table. Incidence rate ratios of negative binominal regression models for yearly rates of incident opioid users, L-TOT users and L-TOT discontinuers from 2009 to 2019.

https://doi.org/10.1371/journal.pone.0326604.s008

(DOCX)

S6 Table. Incidence rate ratios of Poisson regression models for yearly rates of incident opioid users, L-TOT users and L-TOT discontinuers during 2009–2013 and 2015–2019.

https://doi.org/10.1371/journal.pone.0326604.s009

(DOCX)

Acknowledgments

We gratefully acknowledge Professor Daren Ashcroft and the NIHR Greater Manchester Patient Safety Research Collaboration, for covering the data access costs. We also thank all contributing patients and general practitioners for making anonymized data available for research.

References

1. 1. Bailey J, Gill S, Poole R. Long-term, high-dose opioid prescription for chronic non-cancer pain in primary care: an observational study. BJGP Open. 2022;6(4):BJGPO.2021.0217. pmid:35728819

* View Article

* PubMed/NCBI

* Google Scholar

2. 2. Jani M, Birlie Yimer B, Sheppard T, Lunt M, Dixon WG. Time trends and prescribing patterns of opioid drugs in UK primary care patients with non-cancer pain: A retrospective cohort study. PLoS Med. 2020;17(10):e1003270. pmid:33057368

* View Article

* PubMed/NCBI

* Google Scholar

3. 3. Chen T-C, Knaggs RD, Chen L-C. Association between opioid-related deaths and prescribed opioid dose and psychotropic medicines in England: a case-crossover study. Br J Anaesth. 2021;127(5):789–97. pmid:34419240

* View Article

* PubMed/NCBI

* Google Scholar

4. 4. Dunn KM, Saunders KW, Rutter CM, Banta-Green CJ, Merrill JO, Sullivan MD, et al. Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med. 2010;152(2):85–92. pmid:20083827

* View Article

* PubMed/NCBI

* Google Scholar

5. 5. Bedson J, Chen Y, Ashworth J, Hayward RA, Dunn KM, Jordan KP. 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(5):908–22. pmid:30620116

* View Article

* PubMed/NCBI

* Google Scholar

6. 6. Faculty of Pain Medicine of the Royal College of Anaesthetists. Tapering and stopping. Available at: https://fpm.ac.uk/opioids-aware-structured-approach-opioid-prescribing/tapering-and-stopping

* View Article

* Google Scholar

7. 7. National Institute for Health and Care Excellence NICE Guideline 193. Chronic pain (primary and secondary) in over 16s: assessment of all chronic pain and management of chronic primary pain (NG193). 2021. Available at: https://www.nice.org.uk/guidance/ng193

* View Article

* Google Scholar

8. 8. Oxford University Hospitals, NHS foundation trust. Guidance for opioid reduction in primary care 2017. Available at: https://www.ouh.nhs.uk/services/referrals/pain/documents/gp-guidance-opioid-reduction.pdf.

* View Article

* Google Scholar

9. 9. National Health Service (NHS). Opioid prescribing for chronic pain. Available at: https://www.england.nhs.uk/south/info-professional/safe-use-of-controlled-drugs/opioids/

* View Article

* Google Scholar

10. 10. Curtis HJ, Croker R, Walker AJ, Richards GC, Quinlan J, Goldacre B. Opioid prescribing trends and geographical variation in England, 1998-2018: a retrospective database study. Lancet Psychiatry. 2019;6(2):140–50. pmid:30580987

* View Article

* PubMed/NCBI

* Google Scholar

11. 11. Advisory Council on the Misuse of Drugs. Advisory council on the Misuseof Drugs Consideration of Tramadol. 2013. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/144116/advice-tramadol.pdf.

* View Article

* Google Scholar

12. 12. Department of Health. Controlled drugs (supervision of management and use) regulations 2013: information about the regulations. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/214915/15-02-2013-controlled-drugs-regulation-information.pdf.

* View Article

* Google Scholar

13. 13. Zin CS, Chen LC, Knaggs RD. Changes in trends and pattern of strong opioid prescribing in primary care. Eur J Pain. 2014;18(9):1343–51. pmid:24756859

* View Article

* PubMed/NCBI

* Google Scholar

14. 14. Mojtabai R. National trends in long-term use of prescription opioids. Pharmacoepidemiol Drug Saf. 2018;27(5):526–34. pmid:28879660

* View Article

* PubMed/NCBI

* Google Scholar

15. 15. Bedson J, Chen Y, Hayward RA, Ashworth J, Walters K, Dunn KM, et al. Trends in long-term opioid prescribing in primary care patients with musculoskeletal conditions: an observational database study. Pain. 2016;157(7):1525–31. pmid:27003191

* View Article

* PubMed/NCBI

* Google Scholar

16. 16. Stephens KA, Ike B, Baldwin L-M, Packer C, Parchman M. Challenges and Approaches to Population Management of Long-Term Opioid Therapy Patients. J Am Board Fam Med. 2021;34(1):89–98. pmid:33452086

* View Article

* PubMed/NCBI

* Google Scholar

17. 17. American Academy of Addiction Psychiatry. Use of opioids in the treatment of chronic, non-malignant pain. Available at: https://www.aaap.org/wp-content/uploads/2018/07/USE-OF-OPIOIDS-IN-THE-TREATMENT-OF-CHRONIC-revised1.pdf.

* View Article

* Google Scholar

18. 18. Neprash HT, Gaye M, Barnett ML. Abrupt discontinuation of long-term opioid therapy among medicare beneficiaries, 2012–2017. Journal of General Internal Medicine. 2021;36:1576–83.

* View Article

* Google Scholar

19. 19. Nataraj N, Strahan AE, Guy Jr GP, Losby JL, Dowell D. Dose tapering, increases, and discontinuity among patients on long-term high-dose opioid therapy in the United States, 2017–2019. Drug and Alcohol Dependence. 2022;234:109392.

* View Article

* Google Scholar

20. 20. Vanderlip ER, Sullivan MD, Edlund MJ, Martin BC, Fortney J, Austen M, et al. National study of discontinuation of long-term opioid therapy among veterans. Pain. 2014;155(12):2673–9. pmid:25277462

* View Article

* PubMed/NCBI

* Google Scholar

21. 21. Minegishi T, Garrido MM, Stein M, Oliva EM, Frakt AB. Opioid Discontinuation Among Patients Receiving High-Dose Long-Term Opioid Therapy in the Veterans Health Administration. J Gen Intern Med. 2020;35(Suppl 3):903–9. pmid:33145683

* View Article

* PubMed/NCBI

* Google Scholar

22. 22. Stein BD, Sherry TB, O’Neill B, Taylor EA, Sorbero M. Rapid discontinuation of chronic, high-dose opioid treatment for pain: prevalence and associated factors. Journal of General Internal Medicine. 2022:1–7.

* View Article

* Google Scholar

23. 23. Hallvik SE, Johnston K, Geddes J, Leichtling G, Korthuis PT, Hartung DM. Identifying opioid dose reductions and discontinuation among patients with chronic opioid therapy. Pharmacoepidemiol Drug Saf. 2021;30(3):395–9. pmid:32844498

* View Article

* PubMed/NCBI

* Google Scholar

24. 24. CPRD Aurum June 2021 Release Notes. Available at: https://www.cprd.com/sites/default/files/2022-02/2021-06%20CPRD%20Aurum%20Release%20Notes.pdf.

* View Article

* Google Scholar

25. 25. Sullivan MD, Von Korff M, Banta-Green C, Merrill JO, Saunders K. Problems and concerns of patients receiving chronic opioid therapy for chronic non-cancer pain. Pain. 2010;149(2):345–53. pmid:20334974

* View Article

* PubMed/NCBI

* Google Scholar

26. 26. Starrels JL, Becker WC, Weiner MG, Li X, Heo M, Turner BJ. Low use of opioid risk reduction strategies in primary care even for high risk patients with chronic pain. J Gen Intern Med. 2011;26(9):958–64. pmid:21347877

* View Article

* PubMed/NCBI

* Google Scholar

27. 27. Rogers KD, Kemp A, McLachlan AJ, Blyth F. Adverse selection? A multi-dimensional profile of people dispensed opioid analgesics for persistent non-cancer pain. PLoS One. 2013;8(12):e80095. pmid:24312456

* View Article

* PubMed/NCBI

* Google Scholar

28. 28. Martin BC, Fan M-Y, Edlund MJ, Devries A, Braden JB, Sullivan MD. Long-term chronic opioid therapy discontinuation rates from the TROUP study. J Gen Intern Med. 2011;26(12):1450–7. pmid:21751058

* View Article

* PubMed/NCBI

* Google Scholar

29. 29. Newlands C, Currie R, Memon A, Whitaker S, Woolford T. Non-melanoma skin cancer: United Kingdom National Multidisciplinary Guidelines. J Laryngol Otol. 2016;130(S2):S125–32. pmid:27841126

* View Article

* PubMed/NCBI

* Google Scholar

30. 30. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. pmid:18064739

* View Article

* PubMed/NCBI

* Google Scholar

31. 31. Lunt M, Pye S, Movahedi M. DrugPrep (STATA): an algorithm to transform RAW CPRD prescriptions data into formatted analysis-ready drug exposure data.(version v2. 0.0). Zenodo. 2018.

32. 32. Nielsen S, Degenhardt L, Hoban B, Gisev N. A synthesis of oral morphine equivalents (OME) for opioid utilisation studies. Pharmacoepidemiol Drug Saf. 2016;25(6):733–7. pmid:26693665

* View Article

* PubMed/NCBI

* Google Scholar

33. 33. Herbert A, Wijlaars L, Zylbersztejn A, Cromwell D, Hardelid P. Data Resource Profile: Hospital Episode Statistics Admitted Patient Care (HES APC). Int J Epidemiol. 2017;46(4):1093–1093i. pmid:28338941

* View Article

* PubMed/NCBI

* Google Scholar

34. 34. Morse S. Index of multiple deprivation (IMD)(UK). Encyclopedia of Quality of Life and Well-Being Research. Springer. 2024. p. 3440–3.

35. 35. Huang Y-q, Gou R, Diao Y-s, Yin Q-h, Fan W-x, Liang Y-p Charlson comorbidity index helps predict the risk of mortality for patients with type 2 diabetic nephropathy. Journal of Zhejiang University Science B. 2014;15:58–66.

* View Article

* Google Scholar

36. 36. Riis AH, Johansen MB, Jacobsen JB, Brookhart MA, Stürmer T, Støvring H. Short look-back periods in pharmacoepidemiologic studies of new users of antibiotics and asthma medications introduce severe misclassification. Pharmacoepidemiol Drug Saf. 2015;24(5):478–85. pmid:25601142

* View Article

* PubMed/NCBI

* Google Scholar

37. 37. Gilbertson DT, Bradbury BD, Wetmore JB, Weinhandl ED, Monda KL, Liu J, et al. Controlling confounding of treatment effects in administrative data in the presence of time-varying baseline confounders. Pharmacoepidemiol Drug Saf. 2016;25(3):269–77. pmid:26608680

* View Article

* PubMed/NCBI

* Google Scholar

38. 38. Preen DB, Holman CDJ, Spilsbury K, Semmens JB, Brameld KJ. Length of comorbidity lookback period affected regression model performance of administrative health data. J Clin Epidemiol. 2006;59(9):940–6. pmid:16895817

* View Article

* PubMed/NCBI

* Google Scholar

39. 39. Jivraj NK, Scales DC, Gomes T, Bethell J, Hill A, Pinto R, et al. Evaluation of opioid discontinuation after non-orthopaedic surgery among chronic opioid users: a population-based cohort study. Br J Anaesth. 2020;124(3):281–91. pmid:32000975

* View Article

* PubMed/NCBI

* Google Scholar

40. 40. Jenkins DA, Yimer BB, Benitez-Aurioles J, Peek N, Lunt M, Dixon W. Trends for opioid prescriptions and the impact of the COVID-19 pandemic among patients with rheumatic and musculoskeletal diseases between 2006 and 2021. Rheumatology. 2023.

* View Article

* Google Scholar

41. 41. Mackey K, Anderson J, Bourne D, Chen E, Peterson K. Benefits and Harms of Long-term Opioid Dose Reduction or Discontinuation in Patients with Chronic Pain: a Rapid Review. J Gen Intern Med. 2020;35(Suppl 3):935–44. pmid:33145689

* View Article

* PubMed/NCBI

* Google Scholar

42. 42. Frank JW, Lovejoy TI, Becker WC, Morasco BJ, Koenig CJ, Hoffecker L. Patient outcomes in dose reduction or discontinuation of long-term opioid therapy: a systematic review. Annals of Internal Medicine. 2017;167(3):181–91.

* View Article

* Google Scholar

43. 43. Nevedal AL, Timko C, Lor MC, Hoggatt KJ. Patient and Provider Perspectives on Benefits and Harms of Continuing, Tapering, and Discontinuing Long-Term Opioid Therapy. J Gen Intern Med. 2023;38(8):1802–11. pmid:36376623

* View Article

* PubMed/NCBI

* Google Scholar

44. 44. In figures: the rise in long-term opioid use. Available at: https://pharmaceutical-journal.com/article/feature/in-figures-the-rise-in-long-term-opioid-use

* View Article

* Google Scholar

45. 45. Olsen Y. The CDC Guideline on Opioid Prescribing: Rising to the Challenge. JAMA. 2016;315(15):1577–9. pmid:26978227

* View Article

* PubMed/NCBI

* Google Scholar

46. 46. National Health Service. Primary Care Guidelines for the Management Of Chronic Non-Cancer Pain in Adults. Available at: https://medicines.bedfordshirelutonandmiltonkeynes.icb.nhs.uk/wp-content/uploads/2023/01/modified-JPC-pain-guidelines-1-1.pdf

* View Article

* Google Scholar

47. 47. Pain Management, Opioid Safety, and PDMP (PMOP). Available at: https://www.va.gov/PAINMANAGEMENT/PMOP/index.asp

* View Article

* Google Scholar

48. 48. U.S. Department of Veterans Affairs. Pain Management Opioid Taper Decision Tool. Available at: https://www.pbm.va.gov/PBM/AcademicDetailingService/Documents/Pain_Opioid_Taper_Tool_IB_10_939_P96820.pdf

* View Article

* Google Scholar

49. 49. Herrett EL, Thomas SL, Smeeth L. Validity of diagnoses in the general practice research database. Br J Gen Pract. 2011;61(588):438–9. pmid:21722461

* View Article

* PubMed/NCBI

* Google Scholar

50. 50. Wolf A, Dedman D, Campbell J, Booth H, Lunn D, Chapman J, et al. Data resource profile: Clinical Practice Research Datalink (CPRD) Aurum. Int J Epidemiol. 2019;48(6):1740–1740g. pmid:30859197

* View Article

* PubMed/NCBI

* Google Scholar

51. 51. Berna C, Kulich RJ, Rathmell JP, editors. Tapering long-term opioid therapy in chronic noncancer pain: evidence and recommendations for everyday practice. Mayo clinic proceedings. Elsevier; 2015.

52. 52. Faculty of Pain Medicine of the Royal College of Anaesthetists. Opioid Optimisation Guidance for Pain Medicine Specialists. Available at: https://fpm.ac.uk/sites/fpm/files/documents/2024-11/Opioid%20Optimisation%20-%20Nov%2024_0.pdf

* View Article

* Google Scholar

Citation: Cai Q, Huang Y-T, Allen T, Morris C, Grigoroglou C, Kontopantelis E (2025) Trends of long-term opioid therapy and subsequent discontinuation among people with chronic non-cancer pain in UK primary care: A retrospective cohort study. PLoS One 20(6): e0326604. https://doi.org/10.1371/journal.pone.0326604

About the Authors:

Qian Cai

Roles: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Affiliation: Division of Informatics, Imaging & Data Sciences, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom

ORICD: https://orcid.org/0000-0002-6052-5664

Yun-Ting Huang

Roles: Conceptualization, Methodology, Resources, Supervision, Writing – review & editing

Affiliations: Division of Informatics, Imaging & Data Sciences, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom, Centre for Pharmacoepidemiology and Drug Safety, Division of Pharmacy and Optometry, University of Manchester, Manchester, United Kingdom

Thomas Allen

Roles: Conceptualization, Methodology, Resources, Supervision, Validation, Writing – review & editing

Affiliation: Centre for Health Economics, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom

ORICD: https://orcid.org/0000-0002-2972-7911

Charlotte Morris

Roles: Conceptualization, Resources, Validation, Writing – review & editing

Affiliations: Division of Population Health, Health Services Research & Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom, National Institute for Health and Care Research (NIHR) School for Primary Care Research, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom, NIHR Greater Manchester Patient Safety Research Collaboration, University of Manchester, Manchester, United Kingdom

Christos Grigoroglou

Roles: Conceptualization, Methodology, Resources, Supervision, Validation, Writing – review & editing

Affiliation: Centre for Health Economics, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom

Evangelos Kontopantelis

Roles: Conceptualization, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – review & editing

Affiliations: Division of Informatics, Imaging & Data Sciences, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom, National Institute for Health and Care Research (NIHR) School for Primary Care Research, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom

ORICD: https://orcid.org/0000-0001-6450-5815

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

References

1. Bailey J, Gill S, Poole R. Long-term, high-dose opioid prescription for chronic non-cancer pain in primary care: an observational study. BJGP Open. 2022;6(4):BJGPO.2021.0217. pmid:35728819

2. Jani M, Birlie Yimer B, Sheppard T, Lunt M, Dixon WG. Time trends and prescribing patterns of opioid drugs in UK primary care patients with non-cancer pain: A retrospective cohort study. PLoS Med. 2020;17(10):e1003270. pmid:33057368

3. Chen T-C, Knaggs RD, Chen L-C. Association between opioid-related deaths and prescribed opioid dose and psychotropic medicines in England: a case-crossover study. Br J Anaesth. 2021;127(5):789–97. pmid:34419240

4. Dunn KM, Saunders KW, Rutter CM, Banta-Green CJ, Merrill JO, Sullivan MD, et al. Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med. 2010;152(2):85–92. pmid:20083827

5. Bedson J, Chen Y, Ashworth J, Hayward RA, Dunn KM, Jordan KP. 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(5):908–22. pmid:30620116

6. Faculty of Pain Medicine of the Royal College of Anaesthetists. Tapering and stopping. Available at: https://fpm.ac.uk/opioids-aware-structured-approach-opioid-prescribing/tapering-and-stopping

7. National Institute for Health and Care Excellence NICE Guideline 193. Chronic pain (primary and secondary) in over 16s: assessment of all chronic pain and management of chronic primary pain (NG193). 2021. Available at: https://www.nice.org.uk/guidance/ng193

8. Oxford University Hospitals, NHS foundation trust. Guidance for opioid reduction in primary care 2017. Available at: https://www.ouh.nhs.uk/services/referrals/pain/documents/gp-guidance-opioid-reduction.pdf.

9. National Health Service (NHS). Opioid prescribing for chronic pain. Available at: https://www.england.nhs.uk/south/info-professional/safe-use-of-controlled-drugs/opioids/

10. Curtis HJ, Croker R, Walker AJ, Richards GC, Quinlan J, Goldacre B. Opioid prescribing trends and geographical variation in England, 1998-2018: a retrospective database study. Lancet Psychiatry. 2019;6(2):140–50. pmid:30580987

11. Advisory Council on the Misuse of Drugs. Advisory council on the Misuseof Drugs Consideration of Tramadol. 2013. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/144116/advice-tramadol.pdf.

12. Department of Health. Controlled drugs (supervision of management and use) regulations 2013: information about the regulations. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/214915/15-02-2013-controlled-drugs-regulation-information.pdf.

13. Zin CS, Chen LC, Knaggs RD. Changes in trends and pattern of strong opioid prescribing in primary care. Eur J Pain. 2014;18(9):1343–51. pmid:24756859

14. Mojtabai R. National trends in long-term use of prescription opioids. Pharmacoepidemiol Drug Saf. 2018;27(5):526–34. pmid:28879660

15. Bedson J, Chen Y, Hayward RA, Ashworth J, Walters K, Dunn KM, et al. Trends in long-term opioid prescribing in primary care patients with musculoskeletal conditions: an observational database study. Pain. 2016;157(7):1525–31. pmid:27003191

16. Stephens KA, Ike B, Baldwin L-M, Packer C, Parchman M. Challenges and Approaches to Population Management of Long-Term Opioid Therapy Patients. J Am Board Fam Med. 2021;34(1):89–98. pmid:33452086

17. American Academy of Addiction Psychiatry. Use of opioids in the treatment of chronic, non-malignant pain. Available at: https://www.aaap.org/wp-content/uploads/2018/07/USE-OF-OPIOIDS-IN-THE-TREATMENT-OF-CHRONIC-revised1.pdf.

18. Neprash HT, Gaye M, Barnett ML. Abrupt discontinuation of long-term opioid therapy among medicare beneficiaries, 2012–2017. Journal of General Internal Medicine. 2021;36:1576–83.

19. Nataraj N, Strahan AE, Guy Jr GP, Losby JL, Dowell D. Dose tapering, increases, and discontinuity among patients on long-term high-dose opioid therapy in the United States, 2017–2019. Drug and Alcohol Dependence. 2022;234:109392.

20. Vanderlip ER, Sullivan MD, Edlund MJ, Martin BC, Fortney J, Austen M, et al. National study of discontinuation of long-term opioid therapy among veterans. Pain. 2014;155(12):2673–9. pmid:25277462

21. Minegishi T, Garrido MM, Stein M, Oliva EM, Frakt AB. Opioid Discontinuation Among Patients Receiving High-Dose Long-Term Opioid Therapy in the Veterans Health Administration. J Gen Intern Med. 2020;35(Suppl 3):903–9. pmid:33145683

22. Stein BD, Sherry TB, O’Neill B, Taylor EA, Sorbero M. Rapid discontinuation of chronic, high-dose opioid treatment for pain: prevalence and associated factors. Journal of General Internal Medicine. 2022:1–7.

23. Hallvik SE, Johnston K, Geddes J, Leichtling G, Korthuis PT, Hartung DM. Identifying opioid dose reductions and discontinuation among patients with chronic opioid therapy. Pharmacoepidemiol Drug Saf. 2021;30(3):395–9. pmid:32844498

24. CPRD Aurum June 2021 Release Notes. Available at: https://www.cprd.com/sites/default/files/2022-02/2021-06%20CPRD%20Aurum%20Release%20Notes.pdf.

25. Sullivan MD, Von Korff M, Banta-Green C, Merrill JO, Saunders K. Problems and concerns of patients receiving chronic opioid therapy for chronic non-cancer pain. Pain. 2010;149(2):345–53. pmid:20334974

26. Starrels JL, Becker WC, Weiner MG, Li X, Heo M, Turner BJ. Low use of opioid risk reduction strategies in primary care even for high risk patients with chronic pain. J Gen Intern Med. 2011;26(9):958–64. pmid:21347877

27. Rogers KD, Kemp A, McLachlan AJ, Blyth F. Adverse selection? A multi-dimensional profile of people dispensed opioid analgesics for persistent non-cancer pain. PLoS One. 2013;8(12):e80095. pmid:24312456

28. Martin BC, Fan M-Y, Edlund MJ, Devries A, Braden JB, Sullivan MD. Long-term chronic opioid therapy discontinuation rates from the TROUP study. J Gen Intern Med. 2011;26(12):1450–7. pmid:21751058

29. Newlands C, Currie R, Memon A, Whitaker S, Woolford T. Non-melanoma skin cancer: United Kingdom National Multidisciplinary Guidelines. J Laryngol Otol. 2016;130(S2):S125–32. pmid:27841126

30. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. pmid:18064739

31. Lunt M, Pye S, Movahedi M. DrugPrep (STATA): an algorithm to transform RAW CPRD prescriptions data into formatted analysis-ready drug exposure data.(version v2. 0.0). Zenodo. 2018.

32. Nielsen S, Degenhardt L, Hoban B, Gisev N. A synthesis of oral morphine equivalents (OME) for opioid utilisation studies. Pharmacoepidemiol Drug Saf. 2016;25(6):733–7. pmid:26693665

33. Herbert A, Wijlaars L, Zylbersztejn A, Cromwell D, Hardelid P. Data Resource Profile: Hospital Episode Statistics Admitted Patient Care (HES APC). Int J Epidemiol. 2017;46(4):1093–1093i. pmid:28338941

34. Morse S. Index of multiple deprivation (IMD)(UK). Encyclopedia of Quality of Life and Well-Being Research. Springer. 2024. p. 3440–3.

35. Huang Y-q, Gou R, Diao Y-s, Yin Q-h, Fan W-x, Liang Y-p Charlson comorbidity index helps predict the risk of mortality for patients with type 2 diabetic nephropathy. Journal of Zhejiang University Science B. 2014;15:58–66.

36. Riis AH, Johansen MB, Jacobsen JB, Brookhart MA, Stürmer T, Støvring H. Short look-back periods in pharmacoepidemiologic studies of new users of antibiotics and asthma medications introduce severe misclassification. Pharmacoepidemiol Drug Saf. 2015;24(5):478–85. pmid:25601142

37. Gilbertson DT, Bradbury BD, Wetmore JB, Weinhandl ED, Monda KL, Liu J, et al. Controlling confounding of treatment effects in administrative data in the presence of time-varying baseline confounders. Pharmacoepidemiol Drug Saf. 2016;25(3):269–77. pmid:26608680

38. Preen DB, Holman CDJ, Spilsbury K, Semmens JB, Brameld KJ. Length of comorbidity lookback period affected regression model performance of administrative health data. J Clin Epidemiol. 2006;59(9):940–6. pmid:16895817

39. Jivraj NK, Scales DC, Gomes T, Bethell J, Hill A, Pinto R, et al. Evaluation of opioid discontinuation after non-orthopaedic surgery among chronic opioid users: a population-based cohort study. Br J Anaesth. 2020;124(3):281–91. pmid:32000975

40. Jenkins DA, Yimer BB, Benitez-Aurioles J, Peek N, Lunt M, Dixon W. Trends for opioid prescriptions and the impact of the COVID-19 pandemic among patients with rheumatic and musculoskeletal diseases between 2006 and 2021. Rheumatology. 2023.

41. Mackey K, Anderson J, Bourne D, Chen E, Peterson K. Benefits and Harms of Long-term Opioid Dose Reduction or Discontinuation in Patients with Chronic Pain: a Rapid Review. J Gen Intern Med. 2020;35(Suppl 3):935–44. pmid:33145689

42. Frank JW, Lovejoy TI, Becker WC, Morasco BJ, Koenig CJ, Hoffecker L. Patient outcomes in dose reduction or discontinuation of long-term opioid therapy: a systematic review. Annals of Internal Medicine. 2017;167(3):181–91.

43. Nevedal AL, Timko C, Lor MC, Hoggatt KJ. Patient and Provider Perspectives on Benefits and Harms of Continuing, Tapering, and Discontinuing Long-Term Opioid Therapy. J Gen Intern Med. 2023;38(8):1802–11. pmid:36376623

44. In figures: the rise in long-term opioid use. Available at: https://pharmaceutical-journal.com/article/feature/in-figures-the-rise-in-long-term-opioid-use

45. Olsen Y. The CDC Guideline on Opioid Prescribing: Rising to the Challenge. JAMA. 2016;315(15):1577–9. pmid:26978227

46. National Health Service. Primary Care Guidelines for the Management Of Chronic Non-Cancer Pain in Adults. Available at: https://medicines.bedfordshirelutonandmiltonkeynes.icb.nhs.uk/wp-content/uploads/2023/01/modified-JPC-pain-guidelines-1-1.pdf

47. Pain Management, Opioid Safety, and PDMP (PMOP). Available at: https://www.va.gov/PAINMANAGEMENT/PMOP/index.asp

48. U.S. Department of Veterans Affairs. Pain Management Opioid Taper Decision Tool. Available at: https://www.pbm.va.gov/PBM/AcademicDetailingService/Documents/Pain_Opioid_Taper_Tool_IB_10_939_P96820.pdf

49. Herrett EL, Thomas SL, Smeeth L. Validity of diagnoses in the general practice research database. Br J Gen Pract. 2011;61(588):438–9. pmid:21722461

50. Wolf A, Dedman D, Campbell J, Booth H, Lunn D, Chapman J, et al. Data resource profile: Clinical Practice Research Datalink (CPRD) Aurum. Int J Epidemiol. 2019;48(6):1740–1740g. pmid:30859197

51. Berna C, Kulich RJ, Rathmell JP, editors. Tapering long-term opioid therapy in chronic noncancer pain: evidence and recommendations for everyday practice. Mayo clinic proceedings. Elsevier; 2015.

52. Faculty of Pain Medicine of the Royal College of Anaesthetists. Opioid Optimisation Guidance for Pain Medicine Specialists. Available at: https://fpm.ac.uk/sites/fpm/files/documents/2024-11/Opioid%20Optimisation%20-%20Nov%2024_0.pdf