This week sees the publication of a new Atlas of Variation, focusing on diagnostic services (in England) for the first time.  Readers familiar with the Atlases of Variation will be aware the atlases are designed to highlight variations in services and to encourage commissioners and providers to ask questions to understand if the variations are unwarranted, in other words, cannot be explained by differences in demographics, incidence or prevalence.

Why variation matters

The focus on diagnostic services is valuable for commissioners as there is significant variation which results in a “more disorderly health service” for people in need of a diagnosis.  Variation can be due to a range of factors including:

• differences in units of measurement and reference ranges for test results
• referral rates, leading to under-testing and over-testing
• inconsistent coding
• configuration of services
• productivity of services, for example, due to technologies and operating hours
• cost and pricing

For commissioners, the implications of variation can be far-reaching – for example, late diagnosis of disease can lead to poorer outcomes and an increase in admissions.  However, the report acknowledges the problem of over-resourcing which can increase risk of harm: “it is difficult, therefore, to determine the ‘right’ rate of testing for any test i.e. the ‘right’ rate to give the optimum balance of benefit to harm and cost.”

What can we learn from the atlas?

Data is provided on 77 indicators across the 5 main specialties (imaging, endoscopy, physiological diagnostics, pathology, genetic testing) in map and graph format with commentary on the context and magnitude of variation along with recommendations for action.  Unfortunately, much of the data is presented at PCT level due to the time period covered by the key datasets; for CCGs operating to the same boundaries as former PCTs, this won’t present a problem but other CCGs may need additional analysis to understand.

One example of variation is the rate of barium enema procedures which shows over a 1000 fold variation (172 fold if the PCTs with the 5 highest and lowest rates are excluded); it is suggested this is mainly due to insufficient capacity for colonoscopy.

Commentary

Commissioners can identify local variations using the interactive version of the atlas online and are recommended to:

• further explore the variations in their local health economy , working with local providers to review activity rates and equity of access to understand where variations are unwarranted and need attention
• work with providers to review/develop evidence-based pathways and referral guidelines for diagnostic services
• request analysis and audit to identify over or under-referral and to match level of activity to local need
• work with providers to improve data collection and accuracy of coding and reporting, to improve understanding of incidence and prevalence
• review timeliness of key service interventions e.g. stroke assessment (a case study on stroke service reconfigurations in London and Greater Manchester is provided)
• require providers to participate in national audits, e.g. SSNAP (stroke) [http://www.strokeaudit.org/], TARN (head injury) [https://www.tarn.ac.uk/Home.aspx], and quality initiatives e.g. Improving Quality in Physiological Diagnostic Service Scheme
• ensure services are compliant with NICE Quality Standards and Guidance, where available
• ensure use of recommended tools, e.g. the Productivity and Planning Assessment Tool for endoscopy services
• act as a liaison between GPs and service providers to develop referral guidelines (using NICE tools where available), assess the relative value of services (e.g. shifting resources from gastroscopy to colonoscopy and/or flexible sigmoidoscopy) and understand where diagnosis may be safely and more conveniently delivered in a community setting
• understand future demand and available capacity for key services, particularly where there may be unmet needs, e.g. audiology assessment
• explore patient flow where time from referral to assessment compares poorly to peers
• review funding models to avoid “perverse incentives” e.g. block contract versus payment by results

There are a number of recommendations specific to particular populations or conditions, which are too numerous to go into here, for example:

• developing networked provision for paediatric endoscopy
• agreeing frequency of HbA1c testing
• provision of quality assured spirometry (accredited training) for COPD

Commissioners will be interested in the impact of  variations on outcomes they are being monitored against.  It will be critical to understand what is needed for local populations and to assess if current provision is likely to lead to desired outcomes.  Diagnostic services are also critical to preventive strategies, where diagnostic data may encourage lifestyle changes.

The evidence base for diagnostics is not as strong as for treatment interventions; where there is a lack of national guidance, there may be a need for commissioners to coordinate professionals to develop local guidance.

 

Link

NHS Right Care (2013), The NHS Atlas of Variation in Diagnostic Services.  Available from:  http://www.rightcare.nhs.uk/index.php/atlas/diagnostics-the-nhs-atlas-of-variation-in-diagnostics-services/.