Towards a standardised method of patient prioritisation that accounts for clinical harm

Problem

It is widely accepted that COVID-19 will be endemic in the UK for at least 18–24 months and will continue to consume significant health and social care capacity and resources.¹ As we move from a pandemic emergency response towards a new normal, we need a way of prioritising the huge backlog of patients requiring elective care. The number of referral to treatment (RTT) patients waiting to start treatment at the end of June 2020 was 3.9 million patients and it is estimated that waiting lists could increase by more than 100% to over 9 million over the next 12 months.^2,3 Of those on the waiting list in June, 50,536 patients had been waiting more than 52 weeks.

During COVID-19 every effort has been made to review the surgical ‘to be scheduled’ lists, and outpatient appointment lists have been reviewed to identify those patients who need emergency treatment. However, trusts often have limited records of these reviews, which are often done on spreadsheets, and even less understanding of how some patients are re-prioritised.

Traditional methods for prioritising elective care in the UK are unidimensional, based on time waited, eg RTT or the 62-day cancer standard. They rely on retrospective reviews once the patient has already waited or breached a performance target, and do not account for future projected waiting. They can drive a preoccupation with targets rather than clinical need. Furthermore, they do not allow for the re-prioritisation of patients with the same diagnostic code but different clinical requirements.

The patient safety evidence base suggests that this is not the safest approach to prioritisation and can result in inadvertent harm to patients or clinical deterioration. In the UK, studies have started to demonstrate the value of more patient-centric approaches to prioritisation of elective procedures by using data such as Patient Reported Outcome Measures (PROMs).⁴ Examples from health systems internationally demonstrate the benefit of more dynamic approaches to clinical risk prioritisation in a range of settings. For instance, studies from New Zealand provide evidence of how risk prioritisation can take account of a more diverse dataset, including quality-of-life information.⁵ Studies from Italy go further to suggest the need to engage with waitlists in a fluid and responsive way to prioritise and re-prioritise elective patients.⁶ Despite the growing evidence on this topic, a recent systematic review reveals a lack of conclusive evidence about best practice in clinical risk prioritisation for elective patients, providing part of the impetus for this study.⁷

Potential solution

The Royal College of Surgeons of England (RCS), the British Society of Gastroenterology and the Joint Advisory Group produced guidance on the prioritisation of patients for surgery and endoscopy respectively at the start of the pandemic.^8,9 The principle underpinning their guidance is a time-based assessment of the urgency of provision. Imperial College Healthcare NHS Trust (ICHT) has developed a novel framework that extends this principle to all aspects of elective clinical care and enables clinicians to take actual or potential clinical harm into account.

Following a series of discussions with clinical colleagues and lay partners, who were concerned that the NHSE guidance lacked the sensitivity to reflect individual patients’ needs, a prototype matrix was developed and tested with further stakeholders and at the ICHT Clinical Reference Group. This testing resulted in further iteration of the matrix and its application, and the development of standardised harm definitions. While development of the matrix was driven by the COVID-19 pandemic, it was developed using evidence-based clinical priorities and is applicable at any time of constrained resources and in any elective care setting.

The matrix (Fig 1), combined with standardised clinical priority categories (Priority 1–4, or P1–4) and standardised definitions of clinical harm (Table 1), provides a resource to guide decision making and support the generation of ‘outcome priorities’, with corresponding actions. The matrix enables rapid quantification of individual patient care requirements, thus enabling clinicians to target more accurately those patients with the greatest need and those who would see the greatest benefit. It can be used at the level of the trust or system to generate patient tracking lists (PTLs).

• Download figure
• Open in new tab
• Download powerpoint

Fig 1.

The standardised matrix, combining priority and clinical harm. PTL = patient tracking list.

Once generated, the outcome priorities and actions can be adjusted according to COVID-19 vulnerability and additional patient needs. This modification was added following consultation with patients and lay partners. The presence or absence of vulnerability does not alter the clinical priority, nor the level of clinical harm. However, it will alter the balance between the risk of treatment during the pandemic and the risk of deferral, the recommended management options, the place of treatment, the consent process and the scheduling of follow-up. Ultimately, the decision about whether to proceed or defer will need to be made by the clinician responsible for the patient and the patient themselves. Patients must be appropriately consented, and any discussions or decisions must be clearly documented. See Box 1 for examples of how the matrix can be used operationally.

The framework redefines waiting times, not according to a time waited per se (RTT), but rather by clinical harm (actual or potential). However, the length of time an individual has already waited will be important in reviewing harm. In addition, as capacity or resource is expanded, the length of waiting will become an important feature in prioritising P3 and P4 patients. The framework will also enable a significant number of patients waiting for routine procedures or investigations that will not alter prognosis, organ function or functional status to be re-prioritised within a priority category, move priority category and/or potentially be discharged to primary care with appropriate safety-netting.

Operationalising this approach may seem daunting. For example, at ICHT the RTT Patient Tracking List comprises almost 60,000 pathways and there are estimated to be several hundred thousand non-RTT pathways. However, implementation at ICHT is progressing rapidly due to the robust and innovative implementation plan, centred around four key elements (Box 2). The matrix is being used to embed a culture change in how all patients waiting for investigations, outpatient appointments or treatment are managed. It is being rolled out in phases, starting with surgical orders (Phase 1). Phase 2 will cover all diagnostic orders and Phase 3 will cover all other outpatient activity.

The framework builds on the RCS clinical priorities, with effective safety mechanisms so that patients are not ‘lost’ within scheduling processes and are re-prioritised as required based on individual factors such as harm or vulnerability to COVID-19.