The Continuous Improvement Methodology

The Counterweight Programme has broken new ground by providing the first evidence based programme for effective and cost-effective weight management in Primary Care. The way this has been done was also highly innovative, using a research process which is still unfamiliar to many, called ‘Continuous Improvement Methodology, rather than the more familiar Randomised Controlled Trial (RCT).

Treatment methods in Medicine and Health Care are rightly expected to be safe and efficacious – ie they do what they say they will. To prove this, the standard approach for evidence based medicine is an RCT, sometimes using a ‘cluster-randomised’ approach, where the practices are randomised to give different treatments, each to large numbers of patients. RCTs specifically test for safety and efficacy, but other research methods are usually needed to examine and evaluate ‘effectiveness ’, when the treatment is to be used in real-life, non-specialist, non-research settings. Efficacious treatments which come out of RCTs well, may then prove to be unacceptable, ineffective, or too expensive to sustain, in a real-life setting. Observational and Qualitative research is needed to explore complicated decisions and behaviours when doctors are faced with tricky funding and reimbursement plans, and patients’ priorities are heavily influenced by non-medical factors, by family and occupational pressures and by media reports. Health Economic research can use data from RCTs, but if it is to relate to true costs in realistic settings it has to be informed by other non-RCT research methods. Even if a treatment is effective and cost-effective, it will fail unless it can be made sustainable within politically driven strategies and cross-governmental policies, for which a research understanding of political sciences is needed. These are all elements of Translational Research within Medicine. Continuous Improvement Methodology underpins the entire process.

It is certainly possible to design RCTs or Cluster Randomised Trials (CRTs) to compare the addition a of a new treatment, in a realistic setting, to the plethora of existing and changing pressures and priorities in routine practice, but such trials need to be enormous because of he variability between real people and over time, so very slow and very expensive. Another problem arises if such a trial is completed, in that by the time a trial is finished and reports, other new bright ideas usually arise, suggesting an alternative treatment, for which the whole process needs to be repeated. Controlled trials thus suffer a limitation that they are designed to provide definitive results, but the world seldom remains static for those results to be applied.

A more serious problem for Randomised Trial design is that the proposed new treatment has to be compared with something else – ‘Usual Treatment’ or ‘Conventional Treatment’. If there is in fact no established usual treatment, then RCT design cannot be used, other than by comparing a new treatment to a group receiving no treatment at all. On ethical grounds it is unacceptable to give no treatment when a treatment exists, or to give a treatment known to be poor.

The Medical RC, which has provided us with guidelines for RCTS and CRTs, and for Complex Interventions, has not yet provided guidance on methodology for programme development in this unusual situation where there is no existing management for comparison. That was the case for Obesity and Weight Management. There was no existing service or model of good practice so we could not do any kind of RCT or CRT, and patients would not accept a long period of randomisation to no-treatment or to an inferior programme.

So at the very beginning of Counterweight we looked for the best ‘evidence-founded’ and theory-driven’ model to use and adopted the 'Closed-Loop Audit' system, (i.e. repeated cycles of audit and feedback, to improve the model in an iterative way. From the beginning, Counterweight offered the best service that our existing evidence and theories could support, and with experience and continuous evaluation the service will be improved. It started as a ‘best bet’, and is now firmly supported by evidence. That is what is called Continuous Improvement Methodology (or Continuous Process Improvement).in industry to develop products and keep them marketable in competitive world. The Morris Minor, with its model unchanged for 30 years could not compete with Toyota when they produced a new improved model every few months. Toyota used the business principle of ‘kaizen’, which is Japanese for ‘continuous Improvement’.

As well as in various other businesses (e.g. Microsoft, Insurance services) Continuous Improvement Methodology has been applied in various service developments including healthcare, social services and community policing, but it has not been much applied to developing healthcare services before

The first 8 years of service development and evaluation has provided, in the current Counterweight Programme, the necessary starting point for future Continuous Improvement, for which a range of research methods will be used. A key feature is the central data collection and analysis within Counterweight, allowing us to identify trends in results and to direct research questions towards areas of need. Already, qualitative research has been completed to tease out the reasons why some practices find it easier to implement Counterweight than others, and why some patients do better than others. The lessons learned will lead to small improvements in the programme, and results will be monitored and audited continuously. When new ideas arise for weight management, rather than setting out to recruit and run a RCT or CRT independently, with the risk that its results will not translate well into a sustainable routine service, it is now possible to save enormously on the costs and time needed by running RCTs within the framework of Counterweight. There are already trained staff and ongoing recruitment in real-life routine practice settings, with the current Counterweight Programme providing a ready-made control arm to which possible improvements can be compared. So the whole obesity and weight management research process can be made much cheaper and quicker now we have Counterweight running

There are already plans for RCTs of this kind, e.g. to assess whether a family-based approach, involving partners and children, might improve outcomes, whether the weight gain of children can be modified by Counterweight, whether locating Counterweight in workplaces rather than health centres can enhance its results, whether Counterweight can be adapted for patients with psychoses or learning difficulties, and whether an enhanced programme can achieve greater weight loss and maintenance for diabetic and pre-diabetic patients.