Evidence-Based Interventions – Where next?

The proposed extension to the Evidenced-Based Interventions Programme (EBI) to include a further 31 procedures and diagnostic tests is proving to be predictably controversial.

Amid the usual talk of NHS rationing and patients being denied essential treatment, the Academy of Medical Royal Colleges (AMRC) consultation  concluded in July 2020.

The AMRC engagement document suggests that clinicians are providing inappropriate care to the tune of over 350,000 spells and that post-Covid-19 waiting list pressures add urgency to dealing with this issue.

One of the problems with EBI is the need to view interventions within the context of a clinical pathway. This generally leads to the application of clinical criteria to decide whether a particular treatment is appropriate. Some argue that is nothing more than the exercise of clinical judgement and that clinicians should be left to decide what’s best for their patients. The counter-argument says that evidence-based best practice should be universal and ensures resources are targeted where they have the greatest impact.  

As always, there is more than one way to view the situation and often the best way forward is to ask questions of what is being presented. For example:

We may or may not accept the proposed guideline that ‘Troponin testing should only be used in cases where a clinical diagnosis of acute coronary syndrome is suspected or for prognostic purposes when pulmonary embolism is confirmed.’  However, how confident can we be that it translates into an activity reduction opportunity of 229,114?

We might also ask:

• How do we explain to patients what constitutes a ‘minimally symptomatic hernia?’

• Arthroscopic repair of meniscal tears might not be appropriate, but exactly how many are performed for patients with concurrent ligament or tendon damage?

• How robustly can we apply risk stratification to identify those needing colonoscopies?

• What assurance or audit mechanisms do we have in place to assess clinical pathways and conservative treatments prior to consideration of surgery?

Put simply, we need better information.

Accurate coding and counting are obviously crucial to policing the system. The AMRC coding summary tables go so far as to quote recommended SQL code. Unfortunately, clinical criteria do not lend themselves readily to automated reporting. Clinical indications and treatment history are the most obvious examples of information that cannot be flagged in a SUS dataset. The consultation document highlights outpatient diagnosis coding as another problem area.

NHS Clinical Commissioners acknowledge that ‘the detail of coding needs to be reviewed following the analysis of all consultation responses, and then worked through in specific demonstrator sites with both commissioners and providers.’

Given the increasing scope of EBI, and the complexity of clinical criteria applied, it seems vital that any such review include consideration of how best to monitor whole-pathway compliance.     

We should consider the evidence carefully so as to truly understand how patients are being treated across the country before coming to a decision. We await the outcome of the consultation with interest.