Search results
Eighteen papers93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,117 reporting on 16 studies93,94,95,96,97,99,100,102,103,104,105,106,107,108,109,117 were included in the barriers and facilitators review. One of the studies117 was also included in the effects review (Review 1). The PRISMA flow chart is reported in Figure 1. Four studies62,63,64,65,122 were identified in the update search; none of these studies identified any additional barriers or facilitators in addition to those reported in the papers identified in the original search, suggesting that data saturation had been reached. Consequently, these additional studies were not incorporated into this review.
Study characteristics
Table 21 outlines the included study characteristics. All studies were published between 2015 and 2022, of which, eight studies were published between 2020 and 2022. Studies were undertaken in three countries, predominantly in USA. Eleven studies were conducted in the USA, four studies in Australia and one study in the UK.
| Author, date | Country | Aim | Study design | Theoretical approach (if reported) | Method of data collection (relevant to our review) | Method of data analysis (relevant to our review) | Setting | Perspective | Sample |
|---|---|---|---|---|---|---|---|---|---|
| Benintendi 202193 | USA | To describe the ways in which well-intentioned taper initiatives impacted people living with chronic pain, a population already experiencing a well-documented and significant burden of stigma. | Qualitative | Framework of structural stigma used to guide analysisa | 41 interviews | Thematic analysis (deductive and inductive) | Primary care clinics at Boston Medical Center in Boston, MA, and the High Dose Opioid Tapering Initiative, pain management clinic, addiction treatment services, and primary care clinics of the University of Michigan Health System (UMHS, also calledMichigan Medicine) in Ann Arbor, MI | Patients | N = 41 patients who had undergone an opioid taper in the last 2 yearsAge range 28–7618 males, 23 females |
| Firemark 202194 | USA | To identify factors that influence or interfere with referrals by primary care providers (PCPs) to a pharmacist-led telephone-based programme to assist patients undergoing opioid tapering | Qualitative | RE-AIM implementation frameworkb | 35 interviews | Thematic analysis (inductive) | An integrated healthcare system (HCS) in Oregon and Southwest Washington | Provider | N = 35 20 PCPs, 15 STORM staff (12 pharmacists, 3 non-pharmacists; 9 current STORM staffs, 6 former STORM staffs)15 males, 20 females |
| Frank 201696 | USA | To explore patients’ perspectives on opioid tapering. | Qualitative | Health Belief Model | 24 semistructured interviews | Thematic analysis (deductive and inductive) | Primary care clinics affiliated with three health systems in Denver, Colorado: (1) an academic medical centre, (2) an urban, safety-net medical centre and (3) a Veterans Affairs medical centre | Patients | N = 24 patients, of which 6 currently on opioid medications without tapering, 12 currently tapering chronic opioid therapy (COT), 6 discontinued COT within the past 3 yearsMean age 52 (range 31–73)11 males, 13 females |
| Giannitrapani 201897,98 | USA | To understand providers’ perceptions of barriers to reducing opioid use and improving the use of non-pharmacological pain management therapies for chronic pain | Qualitative | Fortney, Burgess123 dimensions of accessc | Nine focus groups | Thematic analysis (deductive)Post hoc analysis of barriers and facilitators | Two academically affiliated CA Medical Centres, California and Oregon and associated community-based outpatient clinics | Providers | N = 60 primary care providers, registered nurses, licensed practical nurses, clerks, psychologists and social workers.Mean age: NRMale/female: NR |
| Henry 201999 | USA | To characterise patients’ experiences with opioid tapering and identify communication strategies that are likely to foster productive, patient-centred discussions of opioid tapering. | Qualitative | Health Belief Model | Focus groups (n = 4) and interviews | Thematic analysis (inductive) | 13 different primary care clinics within the University of California, Davis Health System located in the greater Sacramento area | Patients | N = 21 patients took part in focus groups, of which 14 had completed tapering, 4 were currently tapering and 3 had been recommended to taperN = 7 patients also interviewed, of which 4 had completed tapering, 2 were currently tapering, and 1 had been recommended to taperMean age 58.210 males, 11 females |
| Henry 2019b100,101 | USA | To identify patient statements about opioids that indicate potential openness to tapering opioids or trying non-opioid pain treatments | Qualitative | – | 86 video-recordings of clinical visits | Thematic analysis (inductive) | Two academic primary care clinics at the University of California, Davis Medical Center in Sacramento, California | Patients and providers | N = 86 patients, of which 8 were currently undertaking tapering and 49 physiciansPatientsMean age: 59.631 males, 55 femalesPhysiciansMean age: 29.612 males, 37 females |
| Kennedy 2018102 | USA | To explore PCPs’ experiences discussing and implementing opioid tapering with patients on long-term opioid therapy | Qualitative | – | Focus groups (n = 6) | Thematic analysis (inductive) | Six academically affiliated primary care clinics in university, urban safety net, and Veterans Health Administration medical centres in Colorado | Providers | N = 40, of which 34 physicians, 3 physician assistants/nurse practitioners, 3 others (nurses/clinical pharmacist)Mean age: 4021 male, 19 female |
| Kuntz 202195 | USA | To identify factors from patients and pharmacists, that were cited as hindering or contributing to successful opioid tapering | Mixed methods | – | Qualitative: InterviewsQuantitative: Baseline and follow-up data collected from patients, records and pharmacy dispensing data | Qualitative: Thematic analysis (inductive)Quantitative: Multilevel hierarchical modelling to evaluate predictors of a successful taper, defined as a 50% decrease from baseline opioid use during the 12 months following the index date. | Kaiser Permanente Northwest (KPNW) an integrated health system in northwest Oregon and southwest Washington | Qualitative: Patients and providersQuantitative: Patients | Qualitative: N = 25 patients who had undergone STORM tapering and N = 12 former and current STORM pharmacistsPatientsMean age: 585 males, 20 femalesPharmacistsMean age NR2 males, 10 femalesQuantitative:N = 1384 patientsAge:24% 21–49 years50% 50–64 years26% ≥ 65 years37% male, 63% female |
| Langford 2020103 | Australia | To explore the perspectives of healthcare professional stakeholders on the challenges associated with opioid deprescribing andfactors to be considered in the development of opioid deprescribing guidelines | Qualitative | – | Focus groups (n = 2) and interviews | Thematic analysis (inductive) | Australian National Deprescribing Network conference (focus groups), Healthcare professional’s place of work (interviews) | Providers | N = 20 (focus groups) of which 2 GPs, 1 general physician, 1 geriatrician, 1 rheumatologist, 3 clinical pharmacologists, 12 clinical pharmacistsN = 11 (interviews) of which 2 GPs, 1 anaesthetist, 2 geriatricians, 1 addiction specialist, 2 clinical pharmacists, 3 registered nurses |
| Magee 2021*104 | Australia | To examine patients’ use of mobile technologies in health care, interest in using mHealth support, preferences for the form and content of mHealth support, and potential barriers to and facilitators of engagement with mHealth support for opioid tapering. | Mixed methods | Survey and interviews | Private primary care practice in regional New South Wales and a public tertiary pain clinic in a metropolitan area in Australia | Patients | N = 21 patients currently or planning to taperAge range: 29–8310 males, 11 females | ||
| Matthias 2017105 | USA | To understand communication processes related to opioid tapering, to identify best practices and opportunities for improvement. | Qualitative | – | 95 audio-recordings of clinical visits and interviews | Thematic analysis (inductive) | Four of 9 primary care clinics at an academic, safety-net hospital serving primarily low-income patients | Patients and providers | N = 37 patients, 9 PCPsPatients (tapering status not reported)Mean age: 5812 males, 25 femalesPCPsMean age: 451 males, 8 females |
| McNeilage 2022*106 | Australia | To identify patterns in individuals’ experience of tapering and explore whether patient characteristics, readiness to taper, opioid tapering self-efficacy, or psychosocial context were related to tapering trajectory | Longitudinal study (Qualitative data) | – | 173 Semistructured phone interviews with 21 patients. Patients interviewed an average of 8 (range 2–12) over mean duration 12 weeks (range 2–20) | Phase 1: Between subject thematic analysis (inductive)Phase 2: Global ratings of tapering readiness and self-efficacy before commencement of taper (low, moderate, high)Phase 3: A within-subject longitudinal analysis of participants’ experience of pain, distress, interference (low, moderate, high), and withdrawal symptoms (yes/no) together with changes in opioid dose (oral morphine equivalent daily dose) and psychosocial context | A tertiary public pain clinic in a metropolitan research hospital and a private primary care practice in a regional area | Patients | N = 21 patients currently taperingAge range: 29–8310 males, 11 females |
| Quinlan 2020107 | UK | To investigate the extent of the psychological comorbidity of patients with chronic pain and long-term opioid use and their hopes and fears before embarking on opioid tapering. | Mixed methods | Un-named theory used to inform analysis | Survey (Questionnaire) | Qualitative: Thematic analysis (inductive and deductive)Quantitative: Descriptive statistics | Opioid clinic in the Pain Management Centre of the Oxford University Hospitals NHS Foundation Trust | Patients | N = 49/60 patients just before commencing tapering |
| Westanmo 2015117 | USA | To describe processes and outcomes of a health system quality improvement initiative designed to reduce opioid-related harms | Non-patient cohort | – | Survey (Questionnaire) | Frequency analysis | A Veterans Affairs healthcare system in Minneapolis | Providers | N = 34/46 responded before initiative, 31/48 responded after initiative |
| White 2020108 | Australia | To evaluate patient and provider perspectives regarding the opioid-tapering intervention Assess, Inform, Manage and Monitor. | Mixed methods | Behaviour Change Wheeld | Semistructured telephone interviews | Thematic analysis | Two general practices in New South Wales, Australia (ranked in bottom 30% of Australian local government areas) | Patients and providers | N = 6 patients (unclear tapering status)N = 4 providers, of which 1 GP, 1 practice nurse, 1 dietitian, 1 pharmacist |
| Wu 2019109 | USA | To evaluate the effect of this prescribing policy, with secondary aims to gain insight into the patient experience of undergoing opioid tapering and to generate hypotheses for further study | Mixed methods | SEM | Semistructured telephone/face to face interviews | Grounded theory thematic analysis (inductive) | Solano County Family Health Services | Patients | N = 6 patients (3 actively tapering, 3 discontinued)Mean age 55.8, Age range: 53–614 males, 2 females |
Of the 16 studies, 10 were qualitative,93,94,96,97,99,100,102,103,105,106 5 were mixed methods95,104,107,108,109 and 1 was quantitative.117 Qualitative data were extracted from all five mixed-methods studies and quantitative data were extracted from two mixed-methods studies.95,107
Qualitative methods used included interviews in 11 studies,93,94,95,96,99,103,104,105,106,108,109 focus groups in 4 studies,97,99,102,103 audio-recorded clinical visits in 1 study,105 video-recorded clinical visits in 1 study100 and open-ended survey in 2 studies.104,107 Quantitative methods used included surveys in two studies107,117 and data collected from patients, records and pharmacy dispensing in one study.95
Nine studies94,95,97,100,103,104,105,117 explored barriers to, and facilitators of, opioid tapering in chronic non-cancer pain as part of their study aims. Two studies96,102 reported results on barriers and facilitators more widely by examining patient/provider experiences of opioid tapering. In five studies,93,99,106,108,109 barriers and facilitators were inferred from results that described wider patient/provider experiences of opioid tapering.
Thirteen studies93,95,96,97,99,100,102,104,105,106,107,108,117 declared no conflicts of interest and one study109 did not report whether there were any conflicts of interest. Two studies reported a conflict of interest. Firemark et al.94 reported that ‘8 authors received grant funding from the National Institute on Drug Abuse of the National Institutes of Health as part of the grant that is funding this study’ and Langford et al.103 declared ‘non-financial support from Pfizer Australia, outside the submitted work’.
Population characteristics
Participants were sampled from primary care clinics,93,96,99,100,102,104,105,106,108 pain management clinics,93,107 VA medical centres,102,117 other multiple/integrated healthcare systems (e.g. pain addiction treatment services, pharmacy, community-based outpatient clinics, public tertiary pain clinic, family health services)93,94,95,97,104,106,109 and via a deprescribing conference.103
Seven studies focused on patient perspectives,93,96,99,104,106,107,109 five studies94,97,102,103,117 focused on healthcare professional’s perspectives and four studies95,100,105,108 included both patients and healthcare professional perspectives. No studies explored views from a carer perspective.
Of the 12 studies that involved a patient perspective, 5 studies93,95,96,99,109 included patients who had tapered/discontinued opioids, 6 studies96,99,100,104,106,109 included patients actively tapering, 1 study107 included patients about to commence tapering and 4 studies96,99,100,104 included patients who had not tapered. In two studies, the tapering status of the patient was not reported105 or was unclear.108
Intervention characteristics
Twelve studies93,96,97,99,100,102,103,105,106,107,109,117 reported on barriers and facilitators around tapering in general (i.e. not on any specific tapering intervention). Four studies94,95,104,108 reported on barriers and facilitators in the context of a specific tapering intervention. Two studies94,95 reported on the Support Team Onsite Resource for Management of Pain (STORM) programme, STORM is a phone-based, pharmacist-led programme providing tailored care to patients tapering off opioids and supporting primary care physicians who support them. White et al.108 reported on a multimodal opioid-tapering intervention, ‘Assess Inform Manage Monitor’ involving general practitioners (GPs), nurses, community pharmacists, psychologists, dietitians and exercise physiologists. Magee et al.104 assessed potential barriers to and facilitators of mobile health technologies (mHealth) for providing opioid-tapering support.
Phenomena of interest
Fifteen studies93,94,95,96,97,99,100,102,103,104,105,106,107,108,109 reported on both barriers and facilitators to opioid tapering, while one study reported on barriers only.117
Theoretical frameworks
Seven theoretical frameworks were used across eight studies93,94,96,97,99,107,108,109 to guide methods and analyses. Two studies96,99 used the Health Belief Model,60,108 used the Behaviour Change Wheel,123 Wu et al.109 used SEM, Benintendi et al.93 used a framework of structural stigma,124 Firemark et al.94 used the RE-AIM implementation framework,125 Giannitrapani et al.97 used the dimensions of access framework123 and Quinlan et al.107 used an unnamed theoretical framework.
Quality assessment
Tables 22 and 23 outline the quality assessment of the included studies. Overall, the quality of 13 studies reporting qualitative data was moderate to high and 2 studies had major limitations. The CASP qualitative item most consistently overlooked was consideration of the relationship between participants and researchers. Five studies97,99,100,107,109 only somewhat addressed data analysis, and four of these studies97,100,107,109 also only somewhat addressed ethical issues. One study104 somewhat addressed data collection considerations. Two studies94,103 clearly addressed all CASP qualitative items. In Giannitrapani et al.,97 the study design (based on a post hoc analysis of a RCT) was not appropriate to address the aim of the research and only somewhat addressed a clear statement of findings. Wu et al.109 did not clearly state their findings. The single cohort study117 was of low quality as it did not identify or account for all confounding factors in the analysis.
| Study | 1 Was there a clear statement of the aims of the research? | 2 Is a qualitative methodology appropriate? | 3 Was the research design appropriate to address the aims of the research? | 4 Was the recruitment strategy appropriate to the aims of the research? | 5 Was the data collected in a way that addressed the research issue? | 6 Has the relationship between researcher and participants been adequately considered? | 7 Have ethical issues been taken into consideration? | 8 Was the data analysis sufficiently rigorous? | 9 Is there a clear statement of findings? | Overall quality assessment |
|---|---|---|---|---|---|---|---|---|---|---|
| CASP Qualitative Quality Assessment Criteria | ||||||||||
| Benintendi 202193 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| Firemark 202194 | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No limitations |
| Frank 201696 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| Giannitrapani 201897,98 | Yes | Yes | No | Yes | Yes | Can’t tell | Somewhat | Somewhat | Somewhat | Major limitations |
| Henry 2019a99 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Somewhat | Yes | Few limitations |
| Henry 2019b100,101 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Somewhat | Somewhat | Yes | Few limitations |
| Kennedy 2018102 | Yes | Yes | Yes | Yes | Yes | Somewhat | Yes | Yes | Yes | Few limitations |
| Kuntz 202195 a | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| Langford 2020103 | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Few limitations |
| Magee 2021,104,a | Yes | Yes | Yes | Yes | Somewhat | Can’t tell | Yes | Yes | Yes | Few limitations |
| Matthias 2017105 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| McNeilage 2022106 | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| Quinlan 2020,107,a | Yes | Yes | Yes | Yes | Yes | Somewhat | Somewhat | Somewhat | Yes | Few limitations |
| White 2020,108,a | Yes | Yes | Yes | Yes | Yes | Can’t tell | Yes | Yes | Yes | Few limitations |
| Wu 2019,109,a | Yes | Yes | Yes | Yes | Yes | Can’t tell | Somewhat | Somewhat | No | Major limitations |
| Study | 1 Did the study address a clearly focused issue? | 2 Was the cohort recruited in an acceptable way? | 3 Was the exposure accurately measured to minimise bias? | 4 Was the outcome accurately measured to minimise bias? | 5a Have the authors identified all important confounding factors? | 5b Have they taken account of the confounding factors in the design and/or analysis? | 6 Follow-up long enough? | Overall quality assessment |
|---|---|---|---|---|---|---|---|---|
| CASP Cohort Quality Assessment Criteria | ||||||||
| Westanmo 2015117 | Yes | Yes | Yes | Yes | No | No | Yes | Major limitations |
Confidence in the review findings
Confidence in the review findings derived using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) CERQual approach are shown in the Summary of Findings and Evidence Profile in Appendix 6 (see Table 38). We had high confidence in 21 subthemes, moderate confidence in 13 subthemes, low confidence in 7 subthemes and very low confidence in 4 subthemes. The main reasons for downgrading were concerns about data adequacy and relevance.
Barriers to opioid tapering in adults with chronic non-cancer pain
Overall, we identified eight barriers to opioid tapering in adults with chronic non-cancer pain. (Table 38) in Appendix 6 outlines the key themes and subthemes with supporting quotes and confidence judgements. Figure 4 highlights the barriers and facilitators to opioid tapering within multiple levels of the healthcare system (i.e. at the individual, interpersonal, organisational and environmental levels) according to the SEM.60
FIGURE 4 Barriers and facilitators to opioid tapering in patients with chronic non-cancer pain according to the SEM (Review 3). (Figure not reproduced here; see the original article.)
Theme B1: Lack of knowledge, beliefs and poor understanding about opioid use, tapering and perceived risks associated with opioid use
At an individual level, both providers and patients described a perceived or actual lack of knowledge and understanding among patients around tapering and the risks and side effects associated with long-term opioid use (subtheme B1.1, high confidence).93–95,98,99,101,102,104–106 Generally, patients were aware of the risks and side effects of opioid use but did not associate these risks with their own opioid use. For example, patients either viewed risk of addiction as being associated with illegal drug behaviour, had been on opioids for a long time without overdosing or they attributed their side effects to other causes. These beliefs along with a misunderstanding of tapering (often viewed as stopping opioids abruptly or completely, rather than a gradual reduction in opioid use) resulted in a general lack of understanding around why some patients need to taper and, ultimately, to an unwillingness to taper.
Patients described opioids as playing an important role in their pain management (subtheme B1.2, moderate confidence). For some patients, the fact that opioids were thought to be effective in controlling their pain and, in some cases, was the only way to control their pain, led to contentment about their current opioid use or a feeling of dependence on opioids. Patients considered that the current benefits from opioids (e.g. reduced pain) outweighed the risk of opioid harms in the future. For those who considered tapering opioids, some patients believed they needed, or expected, an alternative treatment (pharmaceutical or non-pharmaceutical) to opioids to manage their pain (subtheme B1.3, moderate confidence).
Providers also believed that they lacked the knowledge and skills to taper patients (subtheme B1.4, moderate confidence). From a ‘knowledge’ perspective, providers described not knowing ‘how to taper’ due to a lack of adequate opioid taper training, as well as a lack of awareness or understanding around the resources available to support opioid tapering.
Theme B2: Fears and negative expectations of, or beliefs around, opioid tapering
At the individual level, both patients and providers identified fears, or expectations of, an uncertain or negative outcome as a significant barrier to opioid tapering (subtheme B2.1, high confidence).93–96,98,101–107,116 Both patients and providers feared that patients may experience an adverse health outcome from tapering, for example, a fear of a return of, or increase in pain, worse quality of life, loss of function or opioid withdrawal symptoms. Patient expectations of an uncertain or negative outcome were partly driven by past bad experiences of opioid tapering. Provider expectations of an uncertain or negative outcome centred on the worsening of the patient–provider relationship (e.g. feeling threatened during patient interactions, patient discontinues health care or changes provider) and was largely based on previous experiences.
Patients and providers were pessimistic about the effectiveness of non-opioid medication options in managing pain or improving quality of life (subtheme B2.2, low confidence). In addition, providers felt some of the alternative analgesics on offer would not be suitable for specific patients with contraindications.
There was an expectation among providers that opioid tapering was challenging for both the patient and the provider, requiring substantial effort on both parts (subtheme B2.3 moderate confidence). Observations were largely made based on provider tapering experience. Providers described difficulty in changing existing patient preferences and expectations around using opioids to manage pain and felt that some patients were hard to motivate. Patient interactions were described as emotionally exhausting or draining, leading some providers to avoid tapering discussions altogether.
Theme B3: Strained patient–provider relationship
At an interpersonal level, both patients and providers described a lack of a trust in the patient–provider relationship as a barrier to opioid tapering (subtheme B3.1, moderate confidence).92–94,98,101,102,104–108 Patients described a lack of trust of providers in several ways. First, patients were particularly distrustful of their GPs when they had been on opioids for a long time, and now were suddenly being asked to taper. Conversely, some patients trusted their GPs, but did not trust other healthcare professionals. In these instances, a lack of trust arose from a lack of understanding of either the individual’s need to taper or the role of pharmacists and others in the tapering process. In addition, a lack of trust arose when patients felt that their providers were not being honest with them about their motives to taper. Some providers, on the other hand, did not trust patients to share their pain-related symptoms or opioid-related side effects in the belief that patients feared having their opioids withdrawn. Consequences of a lack of trust between patients and providers include providers being unable to accurately assess a patient for tapering and patients feeling betrayed and choosing to discontinue health care with their provider.
Patients also described feeling stigmatised or judged by providers who they felt labelled them as ‘addicts’ or as being dependent on opioids (subtheme B3.2, high confidence). As a result, patients felt their pain was disbelieved or invalidated by providers. In one instance, this experience of being disbelieved and stigmatised particularly affected a patient who felt marginalised due to their race.
When providers were, or were perceived to be, unsupportive, this was considered as a patient-reported barrier to tapering (subtheme B3.3, moderate confidence). Patients felt providers were unsupportive in the way they responded, for example, that they lacked empathy, did not listen to them or were inflexible, and in a proactive way, for example, they did not seek to understand patient needs during tapering discussions, did not explain the rationale for tapering or provide advice or guidance on managing a taper. This led to patients feeling abandoned by providers.
Theme B4: Lack of patient motivation or involvement in tapering process
Motivation to taper was considered a barrier at individual and wider environmental levels.92–94,98,104,105 Barriers to tapering arose when motivations and goals for tapering were lacking or not driven by patient need (subtheme B4.1, moderate confidence). Patients were less receptive to tapering when providers did not consider their needs and were motivated to taper by external institutional pressure. On the other hand, providers were not motivated to offer tapering to patients they thought were not open to tapering for fear of failure.
A lack of patient involvement in tapering decisions and processes was a patient-reported barrier to tapering (subtheme B4.2, low confidence) at the interpersonal level. Patients did not feel empowered to taper if they lacked choice or control over the taper or if they perceived a power imbalance between the provider and patient over tapering decisions.
Theme B5: Burden of tapering
Both patients and providers described several sources of burden of tapering relating to the individual and environmental level. First, they described the emotional burden (subtheme B5.1, high confidence).93–96,98,101–105,108 For patients, a poor emotional status either prior to or during tapering due to withdrawal symptoms was considered detrimental to tapering success, with increased stress making their pain worse. For some providers, tapering was an overwhelming experience for them. They described trying to balance competing demands and expectations around the need to balance safe prescribing, with a need to maintain patient satisfaction made decisions around tapering opioids more difficult.
Tapering was also physically burdensome (subtheme B5.2, low confidence) requiring a sustained effort from both patients and providers. Patients expended energy trying to regulate their behaviour to manage their pain and often found it difficult or were unable to regulate their behaviour. Providers described tapering as time-intensive, referring to both the time it took to have tapering discussions with patients and then to help them to manage pain with non-pharmacological alternatives.
Patients also described the health burden of tapering (subtheme B5.3, high confidence) with some experiencing an increase in pain, withdrawal symptoms or loss of function.
Theme B6: Opioid-tapering initiatives are not tailored to incorporate patient need or preference
Organisational-level barriers were observed when the design and delivery of opioid-tapering initiatives were not tailored to, or did not consider, patient needs and preferences (subtheme B6.1, very low confidence).92,98,101–103,105,108 One study reported on patient-reported potential barriers to an app-based opioid-tapering initiative.104 Study patients felt the content of the text messages sent to support the self-management of their pain were impersonal, or they disliked having to respond to text messages. Issues of accessibility were raised that included the patient not having the means to access taper support (e.g. no/poor internet access) or having poor vision and, therefore, not being able to read the text messages. For providers, the fact that tapering support did not include methods of improving patient motivation was considered a barrier. A failure to incorporate patient preferences around opioid-tapering implementation activities resulted in an individual-level barrier arising because of a lack of patient experience of, or confidence in, using the mode of tapering support to self-manage their pain.104
Furthermore, a potential consequence of the poor design and delivery of opioid-tapering initiatives was the production of unintended consequences (subtheme B6.2, high confidence). Patients felt that the constant monitoring of opioid use (e.g. via pill counts or urine testing) reinforced perceptions of stigma around addiction, reduced patient autonomy and increased mistrust. On some occasions, this resulted in patients changing provider, discontinuing health care or seeking alternative opioid supplies.
Theme B7: Unsupportive health-system-related tapering environmental context
Studies reported a range of interpersonal-, organisational- and environmental-level barriers relating to the availability of, and access to, tapering resources.92,93,96,98,101–103,108 Above all, a lack of access to resources and training inhibited providers from initiating a taper (subtheme B7.1, high confidence). Providers complained of a lack of availability of alternative pain management therapies to help taper opioids, and even when alternatives were available, they described a lack of timely access to them, or had concerns that the resources would be discontinued or were understaffed. This resulted in providers referring only their most challenging patients for fear of overwhelming the service. Barriers relating to affordability and accessibility (e.g. transportation, setting) prevented patients accessing tapering resources. Providers also felt there was a lack of adequate tapering training available to them.
Even when resources were available, some providers felt that their attempts to taper were hindered by ineffective interprofessional collaboration (subtheme B7.2, very low confidence). Providers discussed a sense of disconnect between primary care and the acute sector over the management of patients on long-term opioids for chronic pain. Prescribers in the inpatient setting did not see opioid reduction as their responsibility; rather, they were more concerned with managing acute pain. Therefore, GPs felt they had less control over opioid reduction. PCPs felt their lack of understanding around tapering initiatives was compounded when other healthcare professionals involved in tapering patients were not co-located within their healthcare settings.
External pressures on providers (subtheme B7.3, moderate confidence) due to workload or pressure to demonstrate productivity to leadership meant providers felt that they lacked time to help patients taper opioids.
Both patients and providers felt that institutional polices, guidelines and initiatives did not support them to taper (subtheme B7.4, moderate confidence). Providers felt current guidance inadequately addressed opioid discontinuation. When it was addressed, there was an underlying assumption that all patients would follow the same tapering trajectory, thereby failing to account for fluctuations in patients’ pain and the need for tapering initiatives to be individualised. Patients, on the other hand, felt marginalised by the wider context, specifically policies around the US opioid epidemic which they felt unfairly targeted them (i.e. labelled them as ‘opioid addicts’).
Theme B8: Patient experiences poor or unexpected life circumstances
At an individual level, patients’ social networks and social support played an important part in preventing patients from accessing a taper.92,94,98,105 Patients described the absence of a social support (i.e. friends or family) as a barrier to opioid tapering (subtheme B8.1, low confidence). This was sometimes due to the side effects of tapering (e.g. keeping a partner awake when a patient struggled to sleep) or a lack of understanding from the patient’s spouse. Others felt socially isolated; some had a desire to hide away during a taper, or their family and friends lived far away.
Personal social responsibilities (i.e. work, money and family commitments) also impeded the ability of patients to taper opioids (subtheme B8.2, high confidence).
Personal events occurring outside of opioid tapering experienced by patients also impacted on a patient’s desire/ability to continue with a taper (subtheme B8.3, high confidence). Patients described experiencing psychological stress or health-related issues due to reasons not associated with tapering as a barrier to taper, for example, death of family members or receiving a cancer diagnosis.
Facilitators of opioid tapering in adults with chronic non-cancer pain
Overall, we identified eight facilitators of opioid tapering in adults with chronic non-cancer pain. Appendix 6 outlines the key themes and subthemes with supporting quotes and confidence judgements.
Theme F1: Patient recognises the negative impact of opioids, understands what tapering is and perceives/experiences the benefits of tapering
At the individual level, patient recognition of the negative impact of opioid use on themselves and/or on others (e.g. on family) facilitated the initiation of a taper (subtheme F1.1, high confidence).94,95,98,99,101,102,104–107 For example, this occurred when patients associated bad side effects with opioid use, were concerned about possible dependence or addiction, spoke of not feeling in control or were aware of the impact their opioid use was having on those close to them. Patients recognised why they needed to taper when providers tailored the rationale to taper to individual circumstances.
Patients were more willing to taper opioids when they understood what tapering was and why they themselves needed to taper (subtheme F1.2, very low confidence). Patients who understood that tapering was a gradual, dynamic process, rather than an abrupt cessation of opioids were more open to tapering. Experience of a taper was a factor associated with a successful outcome.106
Patient knowledge about tapering alternatives and beliefs around tapering effectiveness were also important (subtheme F1.3, very low confidence). Patients were optimistic about tapering effectiveness when they could see the value of referral for non-medication behavioural treatment.
Perhaps unsurprisingly, an improved or unchanged health status (e.g. decrease/no change in pain, improvements in functioning) or minimal/no withdrawal symptoms while tapering (subtheme F1.4, high confidence) were facilitators of opioid tapering.
Theme F2: Ability to initiate, manage/cope with taper
The ability of patients and providers to manage and cope with a taper were reported at the individual level.93,94,96,98,101–105,108 Being psychologically and physically prepared to taper (subtheme F2.1, high confidence) was described as an enabling factor by patients. Patients understood that tapering may lead to an increase in their pain or withdrawal symptoms but were accepting of the consequences and were emotionally stable.
Self-efficacy, a patient’s belief in their ability to manage a taper (subtheme F2.2, moderate confidence), was described as a facilitator to successful opioid tapering. Once patients recognised that tapering would require effort on their behalf, they adopted pharmaceutical and/or non-pharmaceutical strategies (including behavioural changes such as planning their day more effectively) to help them manage their pain.
From a provider perspective, an awareness of tapering resources and training (subtheme F2.3, moderate confidence) was seen to facilitate initiation of tapering discussions with patients and/or referral of patients for tapering support.
Providers also described being able, confident in their own ability to taper and being proactive (subtheme F2.4, low confidence) as enabling factors. Provider confidence in their ability to taper increased when patients were open to tapering or when interactions with patients went well. Providers described being proactive in identifying opportunities to initiate a taper, choosing their moment when patients were experiencing serious adverse effects from opioid use or when they inherited a patient from another provider.
Theme F3: Clear, consistent, supportive and timely communication in preparing patients to taper
Interpersonal-level factors describing the need for timely communication when initiating and during opioid tapering to manage patient expectations were highlighted by both patients and providers (subtheme F3.1, high confidence).93,94,101–104 Providers spoke of a need to discuss opioid tapering at the same time as opioids are prescribed, thereby managing patient expectations, preparing them for tapering and easing provider burden around the need to initiate a taper discussion in the future. Providers also highlighted the need for timely and responsive reminders about the tapering support available to them. Patients, on the other hand, described a preference for communications from providers to be responsive to patient needs rather than disrupting their day unnecessarily.
Clear, consistent, tailored communication from providers (subtheme F3.2, high confidence) was also a facilitator to successful tapering. Providers tried to focus their conversations on patient-relevant side effects and were consistent with their messages during tapering discussions over time. Patients sought reassurance from providers that first, they would not be abandoned and, second, that what they were experiencing during tapering was normal. Patients suggested this could be achieved at an organisational level by providers tailoring the information sent to the patients to the stage of tapering. As a result, effective provider communication increased patient trust in providers.
Theme F4: Strong, open, supportive patient–provider relationship
At the interpersonal level, both patients and providers discussed the importance of a strong, open and supportive patient–provider relationship as a facilitator to successful opioid tapering.94,95,98,99,101,104,105,107 Mutual honesty and trust between patient and provider (subtheme F4.1, moderate confidence) facilitated the initiation of, and engagement with, opioid tapering.
Patients and providers also described the characteristics of a supportive provider (subtheme F4.2, high confidence) as being someone who is non-judgemental, empathetic, flexible and responsive to patient needs, and who seeks to listen to and understand patients’ concerns and offers guidance on managing a taper.
Theme F5: Decision to initiate and manage taper involves patient and is driven by patient need
A patient-centred shared decision-making approach to the initiation and management of opioid tapering was discussed by both patients and providers as a facilitator of opioid tapering.92,94,98,101,102,104–108 At the individual level, patients who were motivated to taper (e.g. had experienced a serious AE on opioids, risked being excluded from family life or felt their lives revolved around opioids) or who had set themselves personal goals (e.g. wanted better health/quality of life or to be more involved in family life) (subtheme F5.1, high confidence) were more likely to achieve a successful taper. In addition, patients were more willing to taper when providers were motivated by individual patient need. Providers described being more willing to taper if they believed patients were open to tapering or requested to taper.
At an interpersonal level, patient involvement in tapering decisions was also considered important (subtheme F5.2, high confidence), both at the point of initiating a taper and throughout the taper process, with both patients and providers discussing the need for patients to be offered options and control over a taper.
Theme F6: Taper initiative is tailored to incorporate patient need/preferences
Patients and providers both mentioned the benefits of organisational-level facilitators when a tailored approach to tapering opioids is required (subtheme F6.1, high confidence).92–94,98,101–105,108 For example, providers sought to schedule the timing of a taper initiation to coincide with patient having some ‘down time’. Providers also recognised the need to adjust the dose and/or speed of the taper based on patient response to maintain patient engagement.
Providers discussed the importance of adopting a holistic, multimodal and multidisciplinary approach to tapering (subtheme F6.2, moderate confidence). Providers discussed the need to include non-opioid analgesic alternatives to opioids and pain management strategies and involve pharmacists, physiotherapists, psychologists and others to help patients manage their pain, withdrawal symptoms, comorbidities and emotional health. Providers particularly appreciated support from other healthcare professionals when dealing with more complex tapering patients.
In studies that examined facilitators of a specific tapering intervention, patients often expressed preferences for the design and delivery of taper support (subtheme F6.3, low confidence). Patients expressed preferences around the content of taper support and the delivery mode (e.g. face-to-face vs. mobile technologies), suggesting that active patient involvement in the tapering process is an important enabler.
Theme F7: Supportive health system environment
At the organisational and wider environmental levels, a supportive health system environment, which provided tapering resources and guidance, was essential to enable opioid tapering in chronic pain patients.93–96,101–103,105,107 Accessibility to resources was often discussed by both providers and patients as a facilitator to opioid tapering (subtheme F7.1, high confidence). Providers stressed the need not only for access, but also for timely access to alternative treatments and multidisciplinary support in managing patients’ chronic pain. The co-location of providers and multidisciplinary teams offering taper support was suggested as a solution. Involving a multidisciplinary team also helped to address provider workload pressures that could have prevented them from initiating a taper. Providers also recognised that adequate funding was necessary to enable accessibility. From a patient perspective, accessibility related to out-of-hours access to tapering support, regular contact with the same provider and low cost were important facilitators.
Providers also appreciated the availability of tapering guidelines and policies to support them to initiate taper discussions with patients (subtheme F7.2, high confidence). However, there was also some recognition that the existence of guidelines alone is not sufficient and that a guideline implementation strategy would be beneficial. Providers also felt that guidelines should recognise the need for a tailored, multimodal approach to tapering opioids.
Theme F8: Life circumstances
Patients’ individual-level life circumstances played an important role in facilitating successful opioid tapering.94,95,98,101,103,105 Both patients and providers highlighted the benefits of a patient social support system such as family, friends, other tapering patients and providers in providing moral support and encouragement (subtheme F8.1, high confidence).
Patients who had social responsibilities were more open to opioid tapering (subtheme F8.2, low confidence). Providers noted that patients to whom family were important were more receptive to tapering discussions. Patients themselves also highlighted family responsibilities (e.g. need to care for sick relatives) as reasons for embarking on an opioid taper.
Mapping barriers and facilitators to the Theoretical Domains Framework
Individual, interpersonal, organisational and environmental barriers and facilitators to the successful tapering of opioids in chronic non-cancer patients were mapped to all 14 domains of the TDF (see Table 39 in Appendix 7). At the individual level, patient barriers were mapped to 12 TDF domains and patient facilitators to 12 TDF domains. Patient barriers were not identified for ‘Social/professional role and identity’ and ‘Intentions’, while patient facilitators were not identified for ‘Social/professional role and identity’ and ‘Optimism’.
Provider barriers were mapped to seven TDF domains and facilitators to seven TDF domains. Provider barriers were not identified for ‘Skills’, ‘Reinforcement’, ‘Intentions’, ‘Goals’, ‘Environmental context and resources’, ‘Social influences’, ‘Behavioural regulation’ while facilitators were not identified for ‘Social/professional role and identity’, ‘Beliefs about consequences’, ‘Intentions’, ‘Memory, attention, and decision processes’, ‘Environmental context and resources’, ‘Emotion’ or ‘Behavioural Regulation’.
At the interpersonal level, barriers mapped to four TDF domains (‘Skills’, ‘Beliefs about capabilities’, ‘Social influences’, ‘Emotions’) and facilitators to eight TDF domains (‘Skills’, ‘Beliefs about capabilities’, ‘Reinforcement’, ‘Intentions’, ‘Memory, attention, and decision processes’, ‘Environmental context and resources’, ‘Social influences’, ‘Emotions’).
At the organisational level, barriers were mapped to three TDF domains (‘Beliefs about consequences’, ‘Environmental context and resources’, ‘Social influences’) and facilitators to two TDF domains (‘Environmental context and resources’, ‘Social influences’). At the environmental level, barriers were mapped to four TDF domains (Social/professional role and identity’, ‘Beliefs about consequences’, ‘Reinforcement’, ‘Environmental context and resources’) and facilitators to one TDF domain (‘Environmental context and resources’).
Mapping barriers and facilitators to the conceptual model of healthcare access
Figure 5 maps the barriers and facilitators to opioid tapering that influence the realisation of access from both a patient and provider perspective to Levesque et al.’s modified conceptual model of healthcare access.55,56
FIGURE 5 Conceptual model of healthcare access with domains of barriers and facilitators of tapering opioids in adults with chronic pain (Review 3). (Figure not reproduced here; see the original article.)
From a patient perspective, most factors influencing healthcare access were predominantly individual level and focused around three dimensions of healthcare access: ‘ability to perceive’ (related to a patient’s health literacy and beliefs about tapering),55 ‘ability to seek’ (related to personal autonomy and ability to choose to seek care)55 and ‘ability to engage’ (related to patient participation and involvement in decision-making around tapering).55 Patients reported fewer organisational-/environmental-level barriers around the dimensions of access, ‘ability to reach’ (related to personal mobility and freedom to access tapering) and ‘ability to pay’ (related to ability to afford tapering, e.g. the ability to take time off work to attend tapering appointments).
Most provider factors influencing healthcare access were concentrated around the dimension, ‘ability to be appropriate’ (related to the fit between a patient’s tapering need and services offered) and were a mix of individual-, organisational- and environmental-level factors. However, a number of barriers and facilitators were also evenly spread across another three dimensions: ‘ability to approach and be approachable’ (related to the fact that patients know that they need to taper),55 ‘ability to accept or be acceptable’ (related to cultural and social factors that determine whether tapering is acceptable and appropriate55) and ‘ability to be available and to accommodate’ (related to the existence of tapering support and its ability to be reached in a timely manner).55
Complementary patient and provider barriers were noted. Both patients and providers talked about the need for trust and honesty when seeking to taper (‘ability to accept or be acceptable’/’ability to seek’), accessibility of tapering resources (‘ability to be available and to accommodate’/ability to reach) and the need for patient involvement during a taper (‘ability to be appropriate’/’ability to engage’).
Summary
To our knowledge, this is the first systematic review of barriers and facilitators to focus specifically on opioid tapering in patients with chronic non-cancer pain. Eighteen papers reported in 16 studies were included in a mixed-methods synthesis. Most included studies were conducted in the USA. This is not surprising given the extent of the opioid epidemic127 and calls from the Center of Disease Control and Prevention128 to limit the use of high-dose opioids. Over half the studies included in this review were published within the last 2 years, which reflects the growing concern around long-term opioid use in patients with chronic non-cancer pain and the current lack of suitable solutions to the problem. Overall, eight barriers and eight facilitators of opioid tapering in chronic non-cancer pain patients were identified across multiple levels of the healthcare system (i.e. at the individual, interpersonal, organisational and environmental levels, see Figure 4) and across multiple TDF domains.
Individual-level factors acting as barriers to opioid tapering were predominant and focused around patient and provider beliefs (subtheme B1.1, high confidence, subthemes B1.2/B1.4, moderate confidence), expectations of tapering (sub-theme B2.1, high confidence, subtheme B2.2, low confidence, subtheme B2.3 moderate confidence), a lack of patient motivation (subtheme B4.1, moderate confidence), tapering burden (subthemes B5.1/b5.3, high confidence, subtheme B5.2, low confidence) and the influence of social factors (subtheme B8.1, low confidence, subthemes B8.2/B8.3, high confidence). In addition, most patient barriers and facilitators clustered around the first and last dimensions of Levesque et al.’s55 conceptualisation of healthcare access framework (see Figure 5). Crucial to the success of opioid tapering therefore, is both a patient and provider’s willingness to taper and the ability of a patient to maintain a taper.
Our review findings suggest several factors operating at an interpersonal level [i.e. trusting, supportive patient–provider relationship (subtheme F4.1, moderate confidence, subthemes F4.2, high confidence), effective communication of opioid risk so patients can relate to their need to taper (subtheme F3.2, high confidence) and environmental level (provider training (subtheme F2.3, moderate confidence)] that may help to increase patient and provider willingness to taper. Crucially, trust needs to be established early in the patient–provider relationship, before tapering becomes necessary.86 Planning to taper opioids at the point at which they are prescribed (subtheme F3.1, high confidence) may help to alleviate providers’ anxiety over initiating tapering conversations. This finding supports the UK NICE guidelines129 on the withdrawal management of medicines associated with dependence or withdrawal symptoms in adults; this guidance recommends that conversations around risks associated with taking opioids and tapering opioids are initiated before the point at which tapering becomes necessary and, ideally, at the point of prescribing.
Our findings suggest that individual-level barriers, including patient motivation (subthemes B2.3/B4.1 moderate confidence, subtheme F5.1, high confidence), social factors (subtheme B8.1/F8.2, low confidence, subthemes B8.2/F8.1, high confidence) and stigma (subtheme B3.2, high confidence), may play an important role in whether a patient not only seeks to initiate a taper but also remains engaged throughout the taper (see Figure 5). Such barriers may be addressed at the interpersonal and organisational levels, for example, having a supportive, non-judgemental provider (subtheme F4.2, high confidence), use of motivational interview techniques130 and consideration of patient preference in the design/delivery of tapering support (subtheme B6.1, very low confidence) such as in the EMPOWER study,131 may help to ensure that tapering interventions and providers do not unintentionally reinforce perceptions of stigma around addiction (subtheme B6.2, high confidence). In addition, part of successful tapering requires skilful consulting behaviours using techniques such as MI.
There are many opportunities for tapering in NHS GP practices. Repeat prescriptions are updated regularly, but our findings show that these may be missed opportunities due to organisational and environmental barriers including lack of time (subtheme B5.2, low confidence), lack of access to non-pharmacological treatments, especially psychological treatments (subtheme B7.1, high confidence), lack of timely access to deprescribing interventions (subtheme B7.1, high confidence) and ineffective inter-professional collaboration (subtheme B7.2, very low confidence). Many of these organisational- and environmental-level barriers are consistent with those often reported in other studies evaluating barriers to health care.132 Currently in the UK, there is limited provision of services to specifically support patients tapering opioids.129
Our focus on tapering, as opposed to wider deprescribing and monitoring, strengthens the applicability of the identified barriers and facilitators to an opioid-tapering non-cancer pain population and, in some cases, increases confidence in the findings of recent broad systematic reviews on barriers and facilitators to opioid monitoring and deprescribing.26 Furthermore, our review identifies additional factors to those reported by Cross et al.,26 for example, providers should be motivated to taper by individual patient need rather than institutional pressure, tapering initiatives should avoid reinforcing perceptions of stigma around addiction. The influence of patients’ social responsibilities (work, financial, family commitments) and patients’ experiences of psychological stress and/or health-related issues due to reasons not associated with tapering are also key factors.
While several barrier and facilitator themes are common across different healthcare contexts [e.g. the lack of time (subtheme B7.3, moderate confidence) and the need for a trusting patient–provider relationship (subtheme F4.1, moderate confidence)],132,133,134,135,136 there are some contextual factors that make opioid tapering particularly challenging for both patient and providers. First, the way in which a patient’s ability to taper fluctuates depending upon a patient’s health and emotional status (subthemes B5.1/B5.3/F1.4/F2.1, high confidence) and social factors (subtheme B8.1/F8.2, low confidence, subthemes B8.2/B8.3/F8.1, high confidence). Fluctuations of state may arise either as a consequence of tapering- or non-tapering-related events including changes in life circumstances (subthemes B8.2/B8.3/F8.1, high confidence). Therefore, it is essential that tapering is tailored to take account of changes in individual circumstances (subtheme F6.1, high confidence) and that there is timely access to tapering resources, including psychosocial support (subtheme F6.2, moderate confidence). These findings provide qualitative evidence for the NICE guidelines129 which recommend considering a patient’s individual circumstances, available support and timely access, when planning an opioid taper. Second, opioid tapering may be even more challenging for patients on multiple medications when providers need to consider how these additional medications (e.g. anti-depressant, anti-psychotic, non-narcotic pain medications) may affect a patient’s response to tapering or how to taper more than one medication. Finally, there is a need to achieve a balance between not letting patients feel abandoned, for example, by offering more regular reviews with the need to avoid increasing feelings of stigma, for example, through closer monitoring of taper patients.
Progress on addressing health system barriers to opioid tapering can already be seen in the UK where recent NICE guidelines129 recognise the need for an individualised, flexible approach to opioid tapering, trust within the patient–provider relationship, avoidance of terminology that may be interpreted as stigmatising language and consideration of a patient’s circumstance and available support. However, barriers and facilitators to opioid tapering are often complex and context dependent. Given the current wider adverse environmental contexts [coronavirus disease discovered in 2019 (COVID-19) pandemic, current downturn in the worldwide economic situation], the greatest challenges to successful opioid tapering may be the implementation of guidance recommendations by providers and organisations.
Current wider contextual events, such as the COVID-19 pandemic, may impact both individual- and system-level barriers to opioid tapering, thus severely hindering timely initiation of opioid tapering. The pandemic has placed immense pressures on the NHS and health systems worldwide with an increasing backlog of patients waiting to access healthcare treatment. Two studies104,106 in our review referred to the pandemic. One study106 highlighted the potential for it to exacerbate a known barrier to tapering – social isolation. The other study104 highlighted the potential for the pandemic to accelerate the acceptance of new digital health technologies to facilitate opioid tapering. However, in adopting new technologies, practitioners must ensure that they do not widen inequalities in access to tapering support (see Chapter 8).