Points for pain: waiting list priority scoring systems

May be the way forward, but we need to learn more about their effects

Doctors have long worried that the British government's emphasis on the number of people on waiting lists, and the time they spend there, obscures the need to treat patients according to clinical urgency. This concern has been voiced most recently in a report from the BMA,1 2 though others have gone further and pointed to the futility of pursuing policies to reduce, or even abolish, waiting lists.3 4 The BMA warns that additional funds earmarked for reducing NHS waiting lists and waiting times will provide an incentive for operating on large numbers of minor cases, leaving more urgent cases and potentially cost effective treatments to wait. The danger with such initiatives is that they provide only temporary relief and do not address the underlying problem of ensuring that waiting lists operate as an efficient and equitable non-price rationing mechanism. The BMA paper argues in favour of priority scoring systems, such as those developed for elective health care in New Zealand, Canada, and Sweden. 5 6 7 The success of such systems seems, however, to be mixed.

In New Zealand an evaluation of the generic surgical priority criteria at Auckland Hospital showed wide variation and poor agreement between the surgeons' clinical judgment in assessing priority and the score patients obtained on the priority score.8 In Sweden a central register established to ensure guaranteed maximum waiting times for cataract surgery found that centres using formal priority scoring systems were more successful in adhering to maximum waiting time guarantees than centres without such systems.7

In the United Kingdom local authorities use priority scoring systems for allocating public housing.9 Such systems assess the relative priority of individuals or families based on current housing conditions, overcrowding, presence of dependent children, and medical or welfare circumstances. These allocation systems have proved controversial so that assessment of their efficiency and equity in the public housing sector (as yet not systematically evaluated) is essential to the debate as to hether waiting list priority scoring systems offer a way forward for the NHS. In particular, priority scoring systems used for allocating public housing differ between local authorities, as does the availability of public housing, leading to differences in waiting times for families in similar circumstances (V Burholt, personal communication).

In the United Kingdom consultants have always prioritised their waiting lists according to broad categories—urgent, soon, and routine. Pilot experiments with priority scoring systems for managing NHS waiting lists have been led by individual clinicians. 10 11 At Guy's Hospital, for example, the top 22conditions on a general surgical waiting list were ranked according to their expected net quality adjusted life year (QALY) gain per unit of bed and theatre resource.11 At Salisbury and Carmarthen patients were initially ranked according to points awarded based on the following criteria: rate of progress of disease, pain or distress, disability or dependence on others, loss of occupation, and time already waited. Both approaches led to clustering of conditions, which posed difficulties for preparing balanced theatre lists. This problem has been overcome at Carmarthen through the introduction of a patient initial quotient to determine whether a patient should be placed on a waiting list, and an algorithm to reflect time waited, which has led to a more balanced case mix on prioritised waiting lists (B Davies, personal communication).

The main arguments in favour of introducing priority scoring systems are that they make the management of waiting lists transparent; the criteria by which priority is given to patients are explicit; and they should lead to patients being treated in order of clinical need, rather than according to arbitrary maximum waiting time guarantees. They also make it possible to set minimum thresholds of clinical need for referral onto waiting lists.

However, priority scoring systems also raise a host of philosophical, technical, and managerial questions. Should scoring systems be condition or specialty specific or, could a set of generic or common criteria be applied across several or all clinical specialties? What clinical and social criteria should be used to decide the relative priority of patients? Who among the many interested groups—consultants, general practitioners, health authority commissioners, patients, and the general public—should be asked to decide on such criteria and their relative weight?

Families seeking public housing have long had to accept that their case for housing or re-housing must be weighed against those of other families on local authority housing lists. Would the British public accept the introduction of explicit priority scoring systems in the NHS?

It has been argued that priority scoring systems would solve the clinical dilemma faced by consultants currently trying to operate according to clinical urgency and also meet maximum waiting time guarantees.11 They may do this by providing a transparent and explicit indication of need. However, thexplicit measurement of need may also lead to professional disharmony if aggregated priority or need scores are consistently different across clinical specialties and used as evidence for the reallocation of beds and theatre time. This might, in theory, improve the efficiency and equity of the management of waiting lists but would be dependent on confidence in interscorer consistency—that is, that a particular patient would be given the same or similar priority score by different doctors.

One of the most serious issues limiting the potential benefits of priority scoring systems is the potential for “gaming” by doctors, patients, and their families. Priority scoring systems would cease to discriminate constructively between high and low priority cases if sympathetic or harassed general practitioners or consultants—or patients wise to the system—exaggerated the case for priority. However, introduction of enforceable contracts linking broad urgency categories to a gradient of maximum waiting times in Victoria's public health system in Australia did not appear to lead to evidence of gaming in terms of the regrading of patients to meet maximum waiting time guarantees.12

As an alternative to a market system where price rations access to health care, as in the United States, there is much merit in using waiting lists as a rationing mechanism for elective health care if the waiting lists are managed efficiently and fairly. Priority scoring systems may offer the NHS an opportunity for policies that promote the treatment of patients in order of clinical need—and thus promote clinical effectiveness. Nevertheless, before their widespread introduction we need to evaluate their dynamic effects over time on case mix, distribution of waiting times, and patterns of resource use. This will involve looking within the “black box” of NHS waiting list management to find out far more about the beliefs and behaviour of those involved in the delivery and receipt of elective health care in the NHS.