Barriers to management of opioid withdrawal in hospitals in England: a document analysis of hospital policies on the management of substance dependence

The United Kingdom (UK) has the largest reported population of people who use non-prescribed opioids in Europe. Latest available English surveillance data report over 250,000 people using heroin or other illicit opioids in 2016/17 [1], with 140,294 in treatment for opioid dependence in 2020/21 [2]. The rate of hospital admission in this population is several times greater than people of the same age in the general population [3], with most admissions relating to long-term health conditions, injuries and bacterial infections [3‐7]. Delays to seeking treatment are common, most admissions are unplanned [4, 8, 9], and many result in discharge against medical advice [10‐15]. Patients who leave hospital prematurely are more likely to be readmitted and have higher all-cause mortality [11‐14]. Hospital discharge is a particularly risky time for people who use heroin as they may leave hospital in an unfamiliar neighbourhood and use drugs in public places, often while unwell and with reduced opioid tolerance, which are all risk factors for overdose. A recent study in England demonstrated one in fourteen fatal opioid overdoses happen shortly after hospital discharge and the risk in this period is four times higher than usual [16].

Qualitative research with people who inject drugs in London, UK, has found that fear of opioid withdrawal is a primary barrier to timely hospital presentation and completion of inpatient care for injecting-related infections [9]. This is corroborated by other studies demonstrating that opioid withdrawal in hospital can cause severe distress, with patients often self-discharging to collect community opioid prescriptions or use illicit opioids [17, 18]. Other factors contributing to late presentation and discharge against medical advice include inadequate pain management, competing priorities precluding hospital attendance and treatment completion, and stigmatising attitudes toward people who use illicit drugs among hospital staff [9, 17‐20].

In the UK, opioid substitution therapy (OST)—the prescription of methadone or buprenorphine—is recommended for opioid dependence [21]. This is well evidenced to reduce drug-related health harms including HIV [22], hepatitis C [23], and mortality [24]. Continuity of OST provision between community and inpatient settings is crucial to prevent opioid withdrawal, reduce discharge against medical advice and facilitate effective treatment of presenting medical issues [10, 25, 26]. National guidance recommends the development of local pathways to ensure the prompt prescription of OST in hospital [27]. Access may, however, be limited by concerns about the safety of OST, particularly in combination with illicit drugs and other medications when the patient’s tolerance may be unknown [27]. Hospitals are directed to identify a list of critical medicines, for which administration should not be omitted or delayed. Until recently, this categorisation has been supported by the 2010 Specialist Pharmacy Service Omitted and Delayed Medicines Tool (ODMT)  [28]. The ODMT categorised risk depending on the potential consequences for patients not receiving medications. We sought to understand if and how hospital policies contributed to OST delay by first, assessing OST categorisation in the few publicly available hospital critical medicines lists and in the 2010 ODMT, and second, requesting and reviewing policies on substance dependence management from hospital trusts, the organisational unit of England’s National Health Service (NHS). An initial outcome of this work was to inform the reclassification of drugs for substance dependence management in the ODMT. This collaborative process with the Specialist Pharmacy Service contributed towards a more comprehensive review of the ODMT in 2020 and its subsequent discontinuation to make way for a more holistic, patient-centred and context-specific framework.

‘Well thought-out protocols and guidance for how hospital staff can respond to people who may have problems from their use of drugs or alcohol, which address the full pathway from before admission to the point of discharge, will support better outcomes for the patient and clinicians and can reduce the likelihood of re-presentation’ [27].

We found that hospital policies relating to opioid dependence varied widely and often precluded timely OST. Many policies did not follow national guidance. Of the 135 NHS trusts who replied to our requests, 37 (27%) indicated that as of January 2020, they did not have a local policy in place informing the management of opioid withdrawal and/or OST prescription. People who receive community OST prescriptions and/or are dependent on illicit opioids are a marginalised population facing multiple barriers to accessing healthcare. This is critical to address given high and rising rates of morbidity and mortality among people who use drugs and the individual suffering this represents. Recognised is the role of stigma in dissuading people who use drugs from accessing care, with multiple studies illustrating discriminatory attitudes and practices toward this group in medical settings [20, 42, 43]. Others have identified how institutional policies may perpetuate or challenge stigma and discriminatory practice [44, 45]. Our study, however, is the first to comprehensively review a national census of hospital substance dependence policies, crucial for understanding and transforming healthcare inequities and outcomes for the most marginalised.

While NHS trusts are responsible for developing local policies, these are underpinned by reference to national guidelines and toolkits. The responsiveness of the Specialist Pharmacy Service to considering our research findings, then collaborating to review and reframe the categorisation of drugs for substance dependence, illustrates the potential for change more broadly. We made the case that the 2010 ODMT classification of drugs for substance dependence was problematic for two reasons. First, it combined two substance dependencies (alcohol and opioids) with distinct risk profiles into one classification and negated consideration of other dependencies (e.g., benzodiazepines). Second, the risk categorisation (low for delay, medium for omission) did not reflect risks associated with psychological distress (including through fear of withdrawal) and concomitant use of illicit drugs on wards, self-discharge and treatment interruption. In consultation with community stakeholders, clinicians and Public Health England, the Specialist Pharmacy Service worked with us to reconfigure the categorisation of drugs used in substance dependence. From one category (drugs used in alcohol and opioid dependence), three were demarcated (alcohol, opioid, benzodiazepine) with risk severity increased for all (Fig. 4). This process proved to be an interim step in a broader reconsideration of the role of the ODMT by the Specialist Pharmacy Service. It is to be withdrawn, with a view to developing a more context-specific and holistic framework to inform frontline care.

This iterative process of stakeholder consultation and resource refinement indicates an openness by organisations, such as the Specialist Pharmacy Service, to support broader and more substantive revisions to local NHS policies for substance dependence. The reconfiguration, and now potential discontinuation of the ODMT in its current form, paves the way for further changes to improve care for people who use drugs more broadly. The prior designation of medication delay for opioid dependency as no or low risk is likely to have underpinned or reinforced statements common to hospital policies such as ‘opioid withdrawal is not a life-threatening condition, but opioid toxicity is’. This statement emphasises the proximal risk of opioid overdose (toxicity), arguably well managed within a hospital setting, while negating consideration of the broader ways in which opioid withdrawal may constitute a life-threatening situation. This is particularly the case when withdrawal leads to patients self-discharging and using illicit opioids outside of the hospital setting with reduced opioid tolerance related to inadequate OST provision. A recent study showed that people who use illicit opioids are four times more likely to fatally overdose soon after hospital discharge than usual, suggesting a need for better continuity of care [16].

Findings illustrate policy and procedural barriers to OST, including inconsistent application of national guidelines across trusts. Policies highlighted the need for patients to abstain from illicit drugs when in hospital but many recommended OST regimens which would be unlikely to facilitate this without withdrawal symptoms, as opposed to in the community, where OST may be titrated alongside continued heroin use, and more rapidly than some hospital policies recommend. The primary role of OST initiation in hospital is to prevent opioid withdrawal in hospital, regardless of whether it is continued following discharge. We note with concern policies that require patients to demonstrate motivation ‘to change at least some aspects of their drug misuse’ prior to OST initiation. Requirements such as behavioural contracts and for urine sample provision and OST consumption to be observed would not be required for other patient groups receiving opioids as analgesia; illustrating how stigma towards people who use drugs can be structurally embedded within institutional protocols. Broader-based stigma reduction interventions focusing on attitudinal change in hospital settings [46‐48] can only succeed if underpinned by non-pejorative patient-centred policies.

While we have focused on reporting policy barriers to timely OST, guidance with the potential to improve the care of people with opioid dependence and facilitate prompt OST prescription were also identified (see Supplementary file). These, with findings reported in this paper, will inform the development of a standardised policy template for opioid dependence management to be evaluated by our team as part of a broader intervention to improve care for people with opioid dependence in NHS hospital trusts¹. Vitally, ongoing work to improve the care of people who use drugs should draw on the lived expertise of this patient group; none of the policies we reviewed indicated patients had been involved in their development.

We focus on highlighting policy barriers to timely management of opioid withdrawal in acute hospital settings with the aim of transforming and standardising NHS practice. Our review of 101 policies from 86 acute hospital NHS trusts demonstrates variability in approach, including deviation from national guidelines. Policy barriers to timely OST in hospital pose a risk to the safety of patients with opioid dependency. Although opioid withdrawal is commonly framed as ‘low risk’ in relation to opioid overdose, we contend that these events are inextricably linked and as such should be accorded equal priority.