Skip to content
  • Blog
  • 23rd May 2023

How can behavioural insights tackle current NHS demand as efficiently as possible?

Part 2 of a 5 part series on how the role of behavioural insights in reducing backlog in the UK's National Health Service

Yesterday, we outlined the enormous challenge facing NHS Trusts and other healthcare organisations in reducing the elective care backlog.

Today, we’re sharing thoughts on how we might tackle current demand as efficiently as possible, focusing on:

  1. Spreading demand by highlighting live availability
  2. Optimising pathways to ensure clinicians have the right information to make the best decisions
  3. Increasing system productivity
  4. Ensuring patients attend appointments, and
  5. Ensuring patients are optimally prepared for procedures

We cannot change the fact that sicker patients may need more complicated care, there may be some ways to promote productivity within the healthcare system

1. Spread demand by highlighting live availability

Data from the King’s Fund in September 2021 showed that the wait for elective care was not spread equally throughout the country: those living in the most deprived areas were 1.8 times more likely to wait over a year for care compared to those living in the least deprived areas.

Behavioural insights could help reduce this problem: in a trial with NHS England and NHS Digital, we found that highlighting hospitals with limited capacity on online referral platforms used by GPs reduced referrals to services with the longest waiting times by 20-38%. A similar intervention could be introduced to increase the likelihood that someone living in one of the most deprived areas of the country might receive care faster. 

2. Optimise pathways to ensure clinicians have the right information to make the best decisions 

Ensuring that clinicians have the right information (at the right time) on which to base their clinical decisions will help services tackle their backlog as efficiently as possible.

This might encompass a range of situations, including supporting primary care staff to decide which kind of a referral is required e.g. a non-urgent referral versus an urgent cancer referral (or indeed whether to refer or not – this is discussed later under reducing avoidable demand in the short-term).

This might also include ensuring the necessary investigations have been conducted before the referral to secondary care is made (eg ensuring the relevant blood and faecal tests have been conducted ahead of referral to gastroenterology), and making sure the necessary information about the relevant tests and results is shared with secondary care staff.

Reviewing existing pathways and processes (eg for ordering tests) to identify ways to make these more efficient (eg eliminating pop-ups/built-in checks that are not useful and simply slow the process down, or updating them to make them impactful), would likely be beneficial.

Primary care staff are also facing immense pressures, and so interventions to optimise pathways should make patient care and appropriate investigation easier for primary care staff, rather than giving them yet more responsibilities.

The behavioural science literature, relating to healthcare and other policy areas, points to a range of examples for how this could be achieved. For example, simple checklists, which in the past have improved surgical outcomes, could be introduced to make it easier for primary care clinicians to decide whether or not to refer a patient (and to which service), determine if other investigations need to be conducted prior to referral, and ensure that the required data is subsequently shared with secondary care.

Ensuring that any checklists or relevant guidance (ideally the most up-to-date guidance, presented in a way that is easy and quick to understand) are linked at the relevant point in the primary care clinician’s documentation system could improve use of and adherence to these tools (in the past, removing tiny points of friction has led to behaviour change eg linking taxpayers directly to a form rather than the webpage which contained the form increased tax repayments). 

3. Increase system productivity

While NHS England’s plan to tackle the backlog aims to deliver 30% more elective activity by 2024/25 than before the pandemic, the number of patients treated from NHS waiting lists in 2022 was 5% lower than in 2019.

A number of reasons for this have been proposed, including difficulty discharging patients, patients being more unwell and so requiring more complicated and lengthier treatment, and reduced productivity (potentially driven by eg lower staff morale, increased infection control measures, and/or lower manager to clinical staff ratios).

While we cannot change the fact that sicker patients may need more complicated care, there may be some ways to promote productivity within the healthcare system. For example, work conducted by Monitor in 2015 found that uptake of nine good practices could lead to productivity gains of 13-20% in elective care.

These practices included: stratifying patients by risk and creating low-complexity pathways for low-risk patients; increasing throughput in theatres by measuring, communicating and managing the number of cases per theatre session; using enhanced recovery practices (including optimising analgesia, hydration and post-operative mobilisation); and providing virtual follow-up for low risk patients.

While these are process-level changes (which may or may not already be implemented in different trusts), incorporation of findings from the behavioural science literature into the design of these interventions could maximise their impact on productivity. 

For example, the behavioural science literature has demonstrated the importance of which option is set as the default (eg in 2012 the UK started automatically enrolling employees into pension schemes – moving from an opt-in system to opt-out – which saw a 29% increase in the number of people saving for retirement). Changing the default option does not preclude the alternative behaviour, but it increases the hassle involved in doing it (known as the ‘friction cost’ in behavioural science) and implies the recommended choice.

Assigning all low-risk patients to virtual follow-up by default (while still enabling clinicians to opt back into in-person follow-up where clinically indicated) might be an effective way of encouraging clinician uptake of this practice. Equally, communicating the default option to low-risk patients might help overcome any patient hesitation when it comes to virtual follow-up.

For example, a US study examining end-of-life care preferences in seriously ill patients asked these patients to indicate on a form whether they would prefer comfort-oriented care or life-extending care. 61% of participants selected comfort care when neither option was pre-selected on the form.

In contrast 77% selected comfort care when this option was pre-selected (with a tick in a box), while only 43% selected this option when the box for life-extending care was pre-checked. If patients are offered choice in the type of post-procedure follow-up they receive, pre-checking boxes to indicate the default option may be a powerful way of increasing patient uptake of virtual follow-up.

Initiatives to improve productivity will require considerable time input from already-busy clinicians and hospital managers. A behavioural insights approach could also help increase uptake of good practice. For example, the use of checklists could facilitate (and therefore reduce the resources required for) the adoption of these interventions.

4. Ensure patients attend appointments

In 2020-2021, 5.6 million NHS outpatient appointments were missed. This number does not include appointments cancelled either by the patient or the hospital. Without prior warning of non-attendance it is likely that these appointment slots cannot be refilled and so go to waste.

Ensuring that patients attend their scheduled appointments will help tackle the backlog as soon as possible. This is in keeping with a document recently published by NHS England, highlighting the importance of reducing outpatient appointment non-attendance. In the past, behavioural insights has reduced non-attendance simply by altering the wording of a pre-existing appointment reminder text message sent to patients.

In this trial, conducted with Imperial College London, the Department of Health and Social Care (DHSC) and Public Health England (PHE), BIT found that highlighting the cost of missing a hospital appointment reduced non-attendance by 24%. Scaling this intervention to trusts which have not yet implemented it (or considering what wording might work best in their local context) could help reduce non-attendance from today.

5. Ensure patients are optimally prepared for procedures

For certain healthcare procedures, patients must complete preparatory steps in order for their procedure to be able to go ahead. For example, patients must fast before an operation and do several things to empty their bowels ahead of a colonoscopy.

However, these preparatory actions are not always done optimally, which can lead to cancellation of the procedure. For example, a 2021 systematic review identified that patients not following pre-operative instructions led to around 2.5% of operations being cancelled across the 78 included studies while French data indicates that poor bowel preparation is the leading cause (2%) of colonoscopy failure.

At best, failed colonoscopy means a wasted appointment; at worst it can mean worrying lesions are not seen and biopsied, which in turn can lead to increased future demand on the health service and greater distress for patients. Processes and patient information for pre-procedure preparation will vary depending on the procedure and across locations.

However, a wide variety of tools from the behavioural science literature, evidenced to change behaviour, could be used to promote proper adherence with preparatory steps. For example, making sure information provided to patients is as clear as possible, with calls to actions that are straightforward to follow, may promote adherence. Exploring the barriers to suboptimal adherence will help inform targeted and effective interventions in the local context.

While we acknowledge that many other factors lead to a greater proportion of operations cancelled (including lack of time, bed space, and change in patient status), applying behavioural insights to ensure patients are optimally prepared (and attend for their surgeries) is likely to be a low-cost, quick way of avoiding cancellation and therefore getting through surgical waiting lists more rapidly. 

While not fasting for long enough may mean that an operation cannot go ahead, there is evidence which suggests that reducing excessive fasting (ie fasting for too long) can lead to improved anaesthetic recovery. Promoting optimal pre-procedure preparation might therefore have positive implications for the elective backlog and for patient outcomes.

Similarly, optimising adherence to ‘prehabilitation’ activities (eg physical activity, nutritional support) ahead of an operation can reduce postoperative complications and length of hospital stay, and therefore reduce further demand on the health service. The preoperative period could also be a key moment to change other health behaviours: for example, there is some evidence that preoperative smoking cessation interventions can not only reduce postoperative complications but can also increase smoking cessation rates one year after the operation. Changing behaviours at this timely moment could therefore reduce longer-term demand on the health service.

Finally, behaviourally informed interventions could help clinicians decide where the gold standard of treatment would not be appropriate. For example, some breast cancer patients are not able to have radiotherapy, in which case a mastectomy might be more appropriate than a lumpectomy. Recognising this in advance might reduce instances in which procedures or investigations do not have the desired outcomes or are unable to go ahead.

The Behavioural Insights Team provides organisations with support applying behavioural insights to their local challenges. In the past, support provided by BIT has ranged from light-touch advisory support to leading full-scale randomised controlled trials. If you would like to discuss anything in this series of blogs in more detail or if you would like to explore how BIT might be able to support you to apply behavioural insights to start reducing your healthcare backlog, please get in touch: