Family Practice Vol. 16, No. 5, 543-544
© Oxford University Press 1999
Book Reviews |
Comparative and cross-cultural health research.
Unit for Evidence-Based Practice and Policy University College London and Royal Free School of Medicine, London
J Øvretveit. (196 pages, £19.95.) Radcliffe Medical Press Ltd, 1998. ISBN 1-85775-274-0.
"Research proves it—French are dirtier than us!" announced a front-page headline in the Sun newspaper a few months ago. The claim was based on two separate pieces of market research that appeared to show that the French consumed 40 per cent fewer bars of soap per head of population than the British.
As John Øvretveit argues in this important new book, making comparisons is both instinctive and fashionable. Discerning and reflecting on differences between ourselves and others is part of the way we learn, and helps us make sense of and improve our world. The availability of a wealth of quantitative and ethnographic data is expanding exponentially on, for example, disease prevalence, beliefs and attitudes, service structure, staffing levels, resource allocation, activity and ‘outputs’, costs, training, satisfaction, and so on. As the tabloid editors well know, invidious comparisons between different individuals, regions or countries can now be produced at the touch of a button. We are right to view the conclusions of amateur researchers in this field with a high degree of scepticism.
Comparative health research is operationally defined (page 6) as "research that creates empirical or explanatory knowledge about health, health services or health systems, by making comparisons using scientific methods that are appropriate for the subject studied and for the purpose of the research". Such research has long been recognized to be conceptually difficult and fraught with methodological challenges. A sound theoretical framework for comparing health status, health services and health policies in different cultures and contexts is long overdue and must draw appropriately on a range of different academic disciplines. As Donald Berwick says in the introduction, Øvretveit is by nature a builder of bridges. He offers dozens of examples of good (and less good) health research, undertaken by transdisciplinary teams and informed variously by basic medical science, psychology, linguistics, mathematics, anthropology, sociology, business administration, marketing, management science and informatics.
As a first principle, Øvretveit suggests that prospective research teams draw a simple diagram of what they plan to investigate. They need to distinguish the ‘whole object’ (e.g. a population, an organization, a health policy, etc.) from the ‘characteristic’ of that object to be compared (e.g. prevalence of a particular disease, a training programme within the organization, the allocation of funds, etc.). They should then site the ‘whole object’ within its economic, political, sociocultural and other contexts, which may affect the compared item or alter its meaning and interpretation. Next, they should select an appropriate research design from a broad menu of three—descriptive (survey, case-study or audit), epidemiological (retrospective case-control, cohort) or experimental (before-and-after, prospective case control or randomized controlled trial), and, if appropriate, mix both quantitative (measurement) and qualitative (interpretive) techniques within their chosen design.
The meat of this book is in the chapters on the practicalities of getting the question right, matching the preferred design to both the research question and the kind of data it is possible to collect, doing the fieldwork, making the comparison (including controlling for confounders) and presenting the results in such a way as to maximize their impact. Øvretveit includes useful appendices with definitions, further reading, and a critical appraisal checklist for evaluating published comparative health research. He ends with a short checklist for policymakers: (a) Has this research shown a difference—and is it valid? (b) What is the cause or explanation? (c) Can we do anything about it? (d) Is it a political priority to do something about it? That, inevitably, will be the bottom line.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||