I have a first class honours degree from Newcastle and a PhD from Manchester. I worked for Beecham’s Research Laboratories for one year, and Marconi Space and Defence for two years, before joining the University of Manchester in 1981. Up until 2000, I was statistician at the Hester Adrian Research Centre (HARC), in the School of Education. HARC was a Centre dedicated to research for people with learning disabilities, and whilst there I developed strong links with other units in Education, including Special Education and the Centre for Human Communication and Deafness, and developed a number of collaborative projects on services for Deaf people, and on children with speech and language difficulties.
I moved to the Centre for Primary Care in 2000 and became leader of the statistic team in 2004. I head a group of four statisticians who provide support to all research projects and post-graduate students in the Centre.
I am currently involved in international collaborations with researchers at RAND, California; the University of Nijmegen in the Netherlands; the University of Minnesota; and a Pan-European research group that includes France, Germany, the Netherlands, Finland, Spain and others. Current collaborations in the UK include projects with researchers based at Plymouth, Edinburgh, Durham, Liverpool John Moores, and York; previous collaborations include Birmingham, Dundee, Keele and Cardiff.
Within the University of Manchester I have strong connections with the School of Medicine Biostatistics Unit.
I also work with staff in the School of Nursing, Midwifery and Social Work, as a statistical advisor on the design and analysis of projects and as a co-applicant on research bids. Projects with staff in the SNMSW have focused on services for disabled adults and children, quality of patient care, and factors in post-natal depression. Much of my own recent work has focused on the role of nurses in the provision of primary care.
I have personally supported more than thirty PhD and Masters students in the statistical aspects of their theses. I am currently supervising one doctoral student.
I welcome enquiries from students wishing to study for a higher research degree (PhD, MD) in the organisation and delivery of primary health care services. Ideas for new doctoral research include:
Social network analysis (quantitative) studies of how health care professionals work together, or work with other agencies (eg social services) to provide care to patients with multiple needs.
Making statistics relevant to policy-makers. Researchers and academics tend to focus on the truth or falsity of research findings, policy makers are more interested in choosing the best option from a range of possibilities. Ways are needed for bridging the gap between the two.
Studies in primary care that have a strong statistical or quantitative methodology component.
More information about postgraduate research in the Primary Care research group of the School of Medicine.
Fellow of the Royal Statistical Society
Lead Statistician and senior researcher at the Centre for Primary Care in the Institute for Population Health
I am a statistician and quantitative researcher, and lead the statistical support team at the Centre for Primary Care. I am involved in the large majority of the Centre’s projects, usually playing a major role in research design, leading the analysis, and contributing to publications. I also run my own projects and contribute to the overall research strategy of the Centre. In addition, I work on relevant issues of quantitative methodology that arise during the course of our work. My current areas of research include:
Use of Primary Care Databases in research
Primary Care databases (PCDs) collate data from the electronic records of patients registered with large numbers of family practices. These databases typically contain detailed data on all primary care consultations for millions of patients spanning two or more decades, making them one of the largest and most detailed sources of patient data globally. At CPC we have a number of both methodological and applied projects using PCDs. Our methodological work is centred on issues around the validity and replicability of research findings obtained from analysing these databases; our applied projects have included using PCDs to assess the impact of incentivising primary care and the quality of the care provided to certain patient groups, such as patients with serious mental illness and with diabetes.
Quality of care
The Quality Theme at NPCRDC is concerned with the quality of the care provided to patients in primary care, including issues of clinical performance, equity, access and satisfaction. Some of the key studies here include a pilot study of UK primary care clinical indicators, which informed the New GP Contract; a longitudinal study of quality in a representative sample of UK practices; and analysis of national QOF (Quality and Outcomes Framework) data on quality and exception reporting at GP practices. Methodological work in this area includes a comparison of methods for combining indicators into composite measures of quality (with RAND, USA) and the optimum number of indicators required in a composite (with Nijmegen University).
We have conducted a number of randomised trials and other studies of interventions aimed at improving patients' abilities to self-manage a chronic condition, and have developed a `whole systems approach' to self-management (the “WISE” approach). Projects include a national evaluation of the Expert Patients Programme, the government initiated programme to provide self care support in the NHS, and investigation of the factors associated with successful self-management.
Meta-analysis is a method for combining the results from a number of different studies. After applying meta-analysis to studies of advanced nursing roles with the University of Nijmegen, I became interested in some important unanswered methodological questions that we subsequently investigated using computer simulations. One project was a comparison of different methods for meta-analysis when study numbers are small and the effects non-normally distributed. Another on-going project concerns methods for adressing heterogeneity of effect across studies in a meta-analysis..