Skip navigation

Shared Care For Diabetes - A Systematic Review

Medical writer Dr Trisha Greenhalgh was recently commissioned by North Thames RHA to prepare a comprehensive review of shared care systems for diabetes. The research, which has just been published as a book by the Royal College of General Practitioners [1], was performed according to standard methodology for systematic reviews produced by the UK Cochrane Centre in Oxford.


An extensive search of manual and electronic databases and personal communication with workers in the field revealed five RCTs, two comparative and three longitudinal, as well as 12 published descriptive studies and around 30 unpublished schemes in the UK. In her report Dr Greenhalgh highlights the main conclusions from the published trials, suggests areas for further research and makes some general recommendations for the implementation of new shared care initiatives.

Main findings

  1. Both the randomised and non-randomised trials demonstrated clearly that structured care by GPs with an interest in diabetes and supported by an enthusiastic specialist liaison team produces comparable, and occasionally superior, levels of care to those provided in hospital. Unstructured care by disinterested and unsupported GPs is ineffective and wasteful of resources.
  2. Structured care comprises systematic recall of patients, allocation of protected time, and adherence to a standard management protocol.
  3. The three RCTs whose findings supported shared care had two common features: a centralised prompting system that recalled patients for appointments, and some form of structured checklist for the GP.
  4. Established shared care schemes fell into two broad categories: firstly centralised, hospital-based and consultant-led, and secondly decentralised, community based and multidisciplinary. In all published successful schemes an enthusiastic key individual or close-knit steering group was clearly identifiable: the team was usually, but not always, led by a consultant diabetologist.
  5. Three common features of successful district-wide shared care schemes were:
  6. an extensive planning phase in which objectives were carefully defined and the facilities, expertise and commitment of individual general practices were assessed.
  7. locally-developed written guidelines for diabetic management.
  8. a well-developed outreach service, with a highly trained nurse-facilitator who could advise on practical problems, maintain enthusiasm and enable `fast tracking' of patients for specialist review.
  9. Maintaining standards was found to depend on regular audit, as has been shown before with chronic disease management programmes in general practice.


Like all good reviews, the work is not only useful as a description of what is known now , but is also a springboard for recommendations and new research to improve effectiveness in the future . The chief recommendations in Dr Greenhalgh's report are:-
  1. The central tenets of any shared care scheme are the "three Rs" of chronic disease management - registration, recall and regular review.
  2. Setting up a shared care scheme is essentially an exercise in change management: success depends on ownership by the participants. GPs should be involved in standard-setting and strategic planning at the outset.
  3. Extensive preliminary research is essential. Heterogeneity of needs and resources in general practice requires diversity of solutions.
  4. Care should be shared rather than shifted. An efficient and effective outreach service will bring out the best in each primary care team.
  5. An audit cycle with clear, relevant and measurable objectives should be included in the scheme from the outset. Locally developed, district-wide audit guidelines will enable individual performance to be measured against common standards.

Future research

Possible areas for future research include:-

  • longer-term follow-up of patients in trials already published or underway.
  • exploration of the educational needs of the primary health care team and ways of meeting those needs.
  • development of a simple audit dataset for use at the primary-secondary care interface.
  • analysis of the practical problems associated with information transfer on large, centralised electronic databases.
  • exploration of ways to involve less innovative practices in shared care systems.
  • comparison of cost-effectiveness of different shared care systems.

An excellent and readable report with solid and workable advice.
Peter Richardson, R&D Manager, North Thames RHA


1: PM Greenhalgh. Shared care for diabetes: a systematic review. RCGP Occasional Paper 67. London, Royal College of General Practitioners, October 1994.

previous or next story in this issue