The outcomes of haemophilia and its treatment: why we need a core set

'If you can't measure it, you can't manage it.' This ubiquitous phrase – often wrongly attributed to management guru Peter Drucker 1 – probably originated with Lord Kelvin's famous address in May of 1883 2. It expresses the importance of having a strong understanding of any phenomenon that one might want to change, and a benchmark against which to know whether that change occurred. This sentiment is certainly one that is shared by many in the haemophilia world. I believe that, as a community, we are at the point where a common, parsimonious and efficient set of standardized health outcome measures – a core set – should be developed for haemophilia care and research. There has been much work done in recent years to provide more, and better, standardized measures of health outcome. As a community, we have developed sophisticated measures of factor levels, inhibitors, joint examination, musculoskeletal imaging, physical function, overall health status/health-related quality of life (HRQL) and more 3. These measures, or tools, work well (i.e. they are valid, highly reliable and responsive to change). It is my belief that we are at a juncture, a time when our energies would be better placed in standardizing a core set, rather than continuing to develop more outcome tools, or tweak our existing ones. What do I mean by outcome measures? A standard dictionary definition is the 'determination and evaluation of the results of an activity, plan, process, or program and their comparison with the intended or projected results' 4. A more familiar medical definition is 'a measure of the quality of medical care, the standard against which the end result of the intervention is assessed' 5. In general medicine, until recent times, mortality was the most important, and often the only, health outcome that was considered when guiding care 6. However, medical care is so powerful and complex now, that we unarguably need more sophisticated tools 7. Similarly, a strong argument can be made that, for haemophilia care, we need more tools than simply mortality, bleeding rates, factor usage and infection rates. (Although mortality is still increased 8-10, bleeding is a good predictor of joint damage and HRQL, factor use accounts for 90% or more of haemophilia costs 11, and infection remains a major cause of morbidity and mortality.) There is now a long history, over several decades, of conceptual models that can guide our choice of health outcome measures. For example, Bergner in the 1980s proposed a conceptual model of health status 12, and Wilson and Cleary proposed their influential model of health-related quality of life in the mid-90s 13. More recently, in haemophilia, many have embraced the World Health Organization's International Classification of Function, Disability and Health (ICF) 14, 15. The ICF model suggests that health can be conceived as a health condition (in our case, haemophilia) that influences (and is influenced by) one's structure and function (anatomy and physiology), activities (instrumental activities of daily life) and participation (participation in social roles). These in turn influence each other, and are all influenced by personal and environmental factors 16. Groups like the International Prophylaxis Study Group (IPSG) and individual researchers have developed tools, specifically for haemophilia, to measure many of the ICF domains. The disease process, and anatomy and physiology are well measured by factor and inhibitor levels, as well as the pharmacokinetics of clotting factor, and standardized scoring of joint examination 17, and imaging studies 18, 19. Activity limitation is measured, validly and reliably, by the Functional Independence Score for Haemophilia (FISH) 20 and the Haemophilia Activities List (HAL) 21, along with its childhood version 22. A number of tools have been validated as good measures of overall health status/HRQL 23. When do we need to measure all of these things? Probably never. I believe that in our clinics, and for research, we only need to measure that which answers the question at hand. Why then, is there a need for a core set? My experience as a rheumatologist (and former co-chair of the American College of Rheumatology (ACR) subcommittee in charge of things such as core sets) is that there is great value in a common (but practical) set of standardized measurements. The ACR, in partnership with its European counterpart, has developed and promoted several core sets for the different rheumatic diseases – for both adult and paediatric patients. This work began formally in 1993, with a core set for rheumatoid arthritis 24. In 1997 a core set for childhood arthritis was proposed 25. The ACR, and other groups, have developed core sets of health outcome measures for a wide variety of rheumatic diseases including systemic lupus erythematosus and idiopathic inflammatory myopathies. These rheumatology core sets have enabled clinicians and researchers to have practical, and reasonably detailed, evaluations of their patients, to understand the nature of the subjects in published research and to develop criteria for meaningful response to treatment 26 and disease remission 27 (used in the clinic, in research and increasingly in reimbursement decisions). Why should the haemophilia community embrace core sets? I can think of several related reasons. Research: Clinical: Advocacy: For a core set to succeed, it must be practical (i.e. a parsimonious set of highly important measures) while being reasonably comprehensive (i.e. not only cover what is important to treaters but also to families and patients – so called, 'patient-centred outcomes'). Work towards establishing an efficient set of important outcome measures has already begun 3, and I believe it to be of the utmost importance to the haemophilia community. Dr. Feldman has received research grant support from Bayer/Talecris, Baxter/Shire, and Novo Nordisk.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.