Pre-diabetes: Should we screen for it and if so, how and what should we offer to those who screen positive?
The problem
Diabetes prevalence is 4-7% in the UK and 9.4% in Newham (APHO 2010), where the population is vulnerable because of genetic risk and environmental deprivation. Preventing or delaying type 2 diabetes is a local and national priority.University College London Partners (UCLP) has funded a pilot ‘pre-diabetes’ intervention pathway for patients with dysglycaemia in selected Newham GP practices. To inform this work, we have developed a new approach to review the literature, an ‘impact orientated systematic review’. This describes current best practice on screening for pre-diabetes, which interventions are effective, whilst addressing barriers to implementation.
The approach
We characterized the setting of Newham, its geography, demographics, patterns of disease, risk and local services. We are engaged with steering group members, ensuring this review is relevant to their commissioning needs. We used Medline and Embase to retrieve 2,868 abstracts of which 104 abstracts pertaining to screening and 101 abstracts regarding interventions were extracted for full paper review. 26 abstracts from external sources were also included.Quantitative data on screening tests is tabulated to compare test accuracy, prevalence and progression rates. Quantitative data from intervention trials will be analysed to assess their short, medium and long-term effectiveness. Primary care stakeholders have been consulted to ascertain the feasibility of implementing trial interventions locally. Qualitative studies exploring barriers in diabetes prevention will be evaluated.
Findings
Four different tests were identified, each of which had strengths and limitations in different contexts. For example the OGTT, the gold standard test for diagnosing diabetes and impaired glucose tolerance, is impractical. In comparison, the HbA1c is a more practical test but has been shown to not be a sensitive screening tool for pre-diabetes. It has been suggested that the accuracy of this test could be improved if used in people at high risk of developing diabetes and in conjunction with the fasting blood glucose. The HbA1c identifies a different cohort of pre-diabetic patients to the OGTT with a degree of overlap; an important consideration when deciding which cohort should be offered interventions. HbA1c point of care (POC) testing is not a validated screening tool. Lifestyle interventions are effective at reducing diabetes prevalence and delaying diabetes onset by approximately 3-4 years. However, real world multi-level barriers may affect the implementation and impact of these interventions. Culturally acceptable schemes, incorporating cognitive therapy, could help address these. However, the economic feasibility of delivering multi-faceted community interventions is to be evaluated.
Consequences
This review has used a ‘co-creation’ approach, where researchers and stakeholders have come together to generate research questions and knowledge to inform healthcare policy. These findings are being implemented in the Newham pre-diabetes 2015-16 pilot design. If feasible and cost effective, these recommendations may be commissioned more widely.
Credits
- Eleanor Barry, University of Oxford, Oxford, UK
- Samantha Roberts, Barts Health NHS Trust, London, UK
- Shanti Vijayaraghavan
- Trisha Greenhalgh, Barts Health NHS Trust, London, UK