Care for people with haemophilia (PWH) has improved much over the last two decades leading to near normal lives for those receiving early regular prophylaxis with clotting factor concentrates (CFC). Yet, there are significant limitations of those practices. In the absence of a well-defined optimal prophylaxis protocol, there are wide variations in practices with a two to threefold difference in doses. In those parts of the world where there are constraints on the availability of CFC, episodic replacement remains the norm for most patients even though it is evident that this does not change the natural history of the disease over a wide range of doses. Suitable prophylactic protocols therefore need to be developed wherever possible at these doses. Finally, there are only limited data on long-term outcomes in haemophilia from anywhere in the world. The practice of documenting specific outcomes as part of the regular evaluation of PWH needs to be established and the appropriate instruments used to assess them. Definitions of clinical events and endpoints of interventions in clinical studies are being developed to help such data collection. The correlations between different replacement therapy protocols and specific outcomes will help define what is best at different dose levels. Such data will allow better health planning and treatment choices throughout the world.
There is much to celebrate regarding the care of PWH today [1, 2]. The concepts of early diagnosis followed by regular factor replacement therapy (‘prophylaxis’) with clotting factor concentrates (CFC) to prevent bleeds and joint damage that were established over four decades ago  changed the lives of those who could benefit from it. With recombinant CFC (rCFC) adding to the pool of plasma-derived CFC (pCFC) nearly two decades ago, there was further impact on the care of PWH around the world [4, 5]. For those in developed countries and with access to recombinant products, higher doses could be instituted for replacement therapies, allowing more intensive prophylaxis from an early age with much better outcomes with regard to preservation of musculoskeletal function . As a result of rCFC becoming the standard of care in the developed world, pCFC became more accessible to PWH in other parts of the world, with improvements in their care [7, 8]. Compared to how lives of PWH were quarter of a century ago, there was now the possibility of some living almost normally and many more with much less pain, disability or early loss of life .
While these successes are very significant, a closer look reveals that many aspects of care of PWH remain unresolved and have not received their due attention. Not only is early prophylaxis not universal, even where there is access to abundant CFC, but also different models of replacement therapies have not been systematically evaluated for their safety and efficacy. Optimal prophylaxis protocols remain undefined. The situation of course is much worse with regard to effective models of care where access to CFC remains restricted. Furthermore, in both circumstances, there are very little data on the long-term musculoskeletal outcome in a disease where the predominant manifestation is bleeding into muscles and joints .
This review will discuss the lacunae in defining effective and cost-efficient replacement therapy protocols in different circumstances, describe some of the efforts being taken to address them and make suggestions for the way forward.
CFC replacement therapy
Unrestricted access to CFC
Most developed countries have had relatively unrestricted access to CFCs for over two decades. Yet, early prophylaxis is not universal in many of these countries for several reasons. Apart from healthcare system-related access issues, there is also the lack of motivation of the families, difficulties with venous access, other logistic difficulties and fear of inhibitors [11-13]. Even with nearly a third of patients developing inhibitors [14, 15], significant data on the epidemiology of inhibitor development and the contribution of environmental factors to it have mostly come from one group [16, 17]. The total quantity of CFC used in the developed countries has nearly doubled over the last decade with addition of new patients, better survival of older PWH, increasing intensity of doses and prophylaxis extending to adults as well as immune tolerance induction for those with inhibitors (Fig 1). However, this has not been matched with proper data on outcomes to show its full benefits.
While observational data with prophylaxis collected over several decades at two centres (Malmo, Sweden and Utrecht, the Netherlands) has established its role in reducing bleeding and maintaining near normal musculoskeletal status, it is important to recognize that the two worked with different philosophies – the former aiming to maintain >1% circulating factor level at all times and the latter targeting the clinical avoidance of bleeding. This resulted in the total doses at Malmo being nearly double those at Utrecht [3, 18]. Both centres have increased their total annual doses, with greater availability of CFC and evolving philosophy of intensified replacement therapy, but the proportions remain similar. Recently performed randomized controlled studies have confirmed the obvious superiority of prophylaxis over episodic treatment in preventing bleeds and therefore improving long-term outcomes [19, 20]. While all these approaches used fairly fixed dose protocols, investigators in Canada attempted to individualize requirements with escalating doses based on individual bleeding patterns . Such approaches, however, need to be carefully designed so as not to allow too many joint bleeds before escalating replacement. No major study has been attempted to prospectively compare different prophylaxis protocols.
In the absence of data comparing different doses for prophylaxis, varying doses are used, based on individual experiences or conviction of what is best. There is therefore great heterogeneity in replacement therapy protocols with regard to time for initiation as well as the dose and frequency of administration, even within the same healthcare environment . The irony of these practices is perhaps most obvious in Western Europe, a zone with relative socioeconomic parity in healthcare and where considerable efforts have been made to standardize care . Data collected by the annual global survey of the WFH and a European Hemophilia Consortium survey have shown for some years now that the CFC use in these countries varied from less than 3 IU per capita to more than 7 IU per capita . There is hardly an example of another disease where there is such variation in doses of a drug used for its treatment. National registries and databases, such as the one in the UK, that have included data on CFC use, have shown that even within the same country, there can be >twofold differences between regions or centres . As these practices have not been combined with any systematic data on outcomes, it has not been possible to judge the relative merits of the different protocols .
This lack of prospective comparative data is often attributed to the relative rarity of haemophilia and the small number of patients at most centres. This is perhaps only partly true because two centres from countries with relatively small populations have collected basic outcome data such as the annual bleeding rates (ABR) and joint scores (clinical and radiological) systematically over a period of decades and taught much to the world. Certainly this could have been done elsewhere as well. We continue to learn from their experiences. A recent comparison of these data have shown that with prophylaxis starting in Sweden at about 1 year of age and an average annual dose of ~4000 IU kg−1 year−1, there were about 2.5 joint bleeds over 5 years compared with prophylaxis starting at about 4.5 years in Netherlands with an average annual dose of ~2000 IU kg−1 year−1 but nearly 10 joint bleeds per PWH . At 24 years of age, this resulted in slightly worse joint scores for patients in the Netherlands but no difference in activities. However, the total annual cost was 66% higher in Sweden. Extrapolated, this meant an extra US $91 000 for every bleed avoided. These are very important conclusions because it allows informed choices to be made. We must also recognize their limitations. The most striking issue is the age of starting prophylaxis which is a well-recognized predictor of long-term outcome . If prophylaxis had been started earlier by about 2 years of age, would the outcome on the lower dose protocol have been the same? It is indeed possible that the differences may have been even less significant. If more centres had collected similar data, there would have been much more data on the correlation between different doses and outcomes. Better informed choices could have been made then regarding treatment options within the dose range used at these centres – about 1500–5000 IU kg−1 year−1.
Therefore, the art of replacement therapy, even after 50 years of practicing it, is far from optimal. This needs to be addressed. It is obvious that the studies most needed are prospective comparisons between different prophylaxis protocols balancing as many variables as possible. While this is unlikely to find commercial sponsorship, why this has not being done with support from healthcare funds defies logic given the fact that >90% of the cost of care is for CFC. Healthcare providers as well as patients should support such studies so that the quality of care is more strongly grounded and therefore better protected. This is important not only for the developed countries in how they would practice prophylaxis but even more so for those developing countries that are now beginning to initiate prophylaxis programmes.
Restricted access to CFC
Where access to CFC continues to be restricted, there are many more difficult issues to be addressed with regard to replacement therapy. Many of these countries have only identified 25–50% of the expected number of PWH in their populations. Identifying the rest and accurately diagnosing them is a major task . With regard to the treatment of those already identified, while support from governments is increasing in many countries, there are challenges related to regular procurement and distribution of CFCs so as to allow access to those who need it [28, 29]. Introduction of the concepts of prophylaxis and making that acceptable to PWH and their families is also a big task. While many of these issues need to be addressed at the bureaucratic, social and educational levels, there is an urgent need to develop suitable models of replacement therapy that are practical and effective in those circumstances .
Until about 5 years ago, CFC replacement in most developing countries was episodic only. Much of this was done not at home but in hospitals, usually only after large bleeds. Given the wide socioeconomic diversity in the region of the developing world, the doses used varied between <100 IU kg−1 year−1 to about 1500 IU kg−1 yr−1, depending on availability. Data on long-term outcome were limited and showed generally poor results [31, 32]. This has been confirmed in a recent prospective observational study  which showed that ABR and joint damage did not improve over a wide range of doses from 100–2000 IU kg−1 year−1 given as episodic replacement. These data therefore showed that episodic CFC replacement over a wide range of doses do not meaningfully alter the bleeding profile and joint damage in severe haemophilia. Good long-term outcome therefore cannot be expected from such treatment protocols.
How should healthcare providers in these countries decide how much CFC to provide in their health budgets for haemophilia? More importantly, what outcomes can they expect with what they will provide? This becomes increasingly relevant in countries that have access to about 1000–2000 IU kg−1 year−1 but think that they cannot implement prophylaxis as per current models? This would include many developing countries as is evident from the WFH annual global survey  including some of the larger ones such as Russia, Brazil and South Africa [34, 35].
The important question therefore is whether these countries should continue to practice episodic treatment or move to the best form of prophylaxis that is practical at the quantities available to them. Prophylactic CFC replacement therapy aims at reducing the number of days a PWH is at risk of spontaneous haemorrhage. In a PWH with severe disease, the ‘time at risk’ of bleeding can be considered 100% without any replacement therapy. Now if his factor level is raised to >1%, taken as a marker of successful replacement therapy, then even at 10 IU kg−1 dose−1 given twice a week, a severe PWH reduces this ‘time at risk’ by ~33% (taking a t½ of ~8 h for FVIII). As clinical efficacy often lasts beyond levels being maintained above 1%, it is likely that this reduction in risk time may be even greater. If this is enhanced to 10 IU kg−1 three times per week, this will then start reaching reductions in the ‘time at risk’ of ~60%. If paradigms could be changed completely and find practical and convenient ways found to administer CFC once a day then even with doses as low 5 IU kg−1 day−1 one could maintain >1% at all times with an annual dose well below 2000 IU kg−1. All the evidence suggests that any prophylaxis that reduces ABR should help improve long-term outcomes. After all, even in Western countries, prophylaxis had started with lower doses and they had already noted improvements in their patients before reaching current doses. There are also limited recent data that CFC doses as low as 10 IU kg−1 two to three times/week reduce joint bleeding in patients who previously received episodic replacement . Such prophylaxis programmes need to be systematically initiated in different countries .
Assessment of outcome
Finally, everywhere in the world there is a need to assess outcomes with whatever replacement protocols that are followed. This has been a relatively ignored subject in the field around the world and needs to change, not only because modern medicine attempts to work on evidence but also because there is greater cost alertness from healthcare providers everywhere and high-cost diseases such as haemophilia are more likely to come under the scanner . It is vital therefore that assessment of relevant outcomes with appropriate tools becomes part of the care of PWH. This field has significantly advanced in the last 10 years as well .
Traditionally, the ABR into joints and other sites has been a simple and predictive indicator of long-term outcome in haemophilia with regard to joint disease and overall musculoskeletal status. This continues to be used as a surrogate marker of both disease severity before intervention and an index of the efficacy of treatment provided. For standardizing bleeding assessment from other sites, tools have been developed in the past few years . These have so far been used mainly to evaluate those conditions where bleeding is more skin and mucosal. There utility in haemophilia and related rarer bleeding disorders needs evaluation. The WFH has attempted to review and summarize the potential of the most relevant of these tools on a website to make them more easily accessible to the community for their use and comments. While data on ABRs are relatively easy to collect, there can be errors in a patient's assessments and reporting. It is important therefore to always combine this with assessment of joints.
Joints – clinical and radiological assessment
The Hemophilia Joint Health Score is gradually replacing the WFH clinical score as a validated tool for the clinical assessment of joints [40, 41]. It is an adaptation of the previous clinical score but has the advantages of having included muscle power and gait and removing some of the more ambiguous measurements such as axial deformity and introducing newer aspects such as strength and gait. It has been found to be useful by several groups and its use is increasing but still confined mostly to research studies. It needs to become part of the annual assessment of PWH along with ABRs .
Similarly, for radiological assessment of joints, the plain X-ray Pettersson score is being substituted by more sensitive techniques such as magnetic resonance imaging to pick up early joint damage. This is indeed now considered the gold standard . However, the currently recommended protocol for data acquisition is very elaborate and could take 2–3 h for scanning alone for all six joints. This may even require some children to be anaesthetized. These challenges along with its high cost have prevented its wide use in clinical care. To overcome some of these issues, more recently ultrasound has been used to evaluate joints [44, 45]. This is more practical as it is much more accessible and can be easier to perform in children. Both these approaches need to be tested more widely to decide their final position in the clinical management of PWH.
To evaluate the functional capacity of PWH, two instruments have been developed to assess activities. The first of these is the Hemophilia Activities List (HAL) – a self-administered questionnaire which assesses different domains of common activities . A paediatric version has also been developed – the pedHAL . These instruments have been found to be useful in several studies in Western settings . Its construct validity in other socioeconomic environments remains to be tested. Also, as this is a self-assessed questionnaire, it will need to be validated in different languages for different parts of the world. Another instrument for assessing activities is the Functional Independence Score in Haemophilia . This is a performance-based tool related mainly to tasks of daily living assessed by simulating them in the clinic. While healthcare personnel need to be trained in its use, language-related issues for PWH are mostly avoided. This instrument has also been successfully used in several countries that do not have early replacement therapy [50-52] but has limited utility among those PWH who have minimal joint disease because of a ‘ceiling’ effect in them with almost all of them getting a maximum score. A more challenging version therefore needs to be developed.
Tools to assess intensity and quantity of activities in haemophilia need to be defined. Apart from calculating energy requirements for different activities, accelerometers have also been used to actually document activities . A good tool is also needed for the assessment of participation.
Health-related quality of life assessment
Over the past decade, a lot of effort has been directed towards evaluating health-related quality of life (hrQOL) of PWH . There have been several challenges with this approach. Given the nature of this evaluation, these instruments usually contain items in the physical, functional, social, emotional and mental domains apart from those related to the disease and its treatment. Such constructs lead to complex instruments that are a challenge for both patients and those administering it. Language and social context related issues also require its adaptation and validation in different parts of the world [55, 56]. More importantly, what really needs to be evaluated is whether the information collected adds to the management of the individual or in comparison of cohorts beyond what is obtained from the assessment tools described above. hrQOL instruments often tend to be less sensitive to smaller differences in outcome assessments and this can lead to difficult conclusions. In one recent analysis comparing the value of prophylaxis over episodic replacement therapy, it was concluded that there is not enough evidence to justify prophylaxis based on hrQOL data . Therefore, it is not only important to use an appropriate hrQOL instrument but to also ensure that these are not administered in isolation but only as an adjunct to more specific assessment of bleeding, joint status and activities. One will then need to decide whether hrQOL assessment is indeed providing data that has incremental value in the management of these patients or in healthcare planning.
Moving towards a system for creating evidence for haemophilia care
As we attempt to move towards collecting data on outcomes, one of the lacunae in the field is the lack of definitions of specific complications or the response to therapies. In fact, except for the severity of haemophilia and the high- and low-titre inhibitors, there were in the past no other definitions in haemophilia . Even joints and muscle bleeds and the titre at which to consider inhibitors significant are not defined. All these are of vital importance in pivotal studies for assessment of CFC as well as for the long-term assessments of PWH. An ISTH SSC group has recently provided these definitions, along with those for types of prophylaxis and assessment of after acute haemarthrosis and surgical haemostasis . Another group is working on definitions of endpoints in clinical studies . All these will help in better study designs and outcomes data collection.
Another hurdle to significant data collection in the past has been the lack of cooperative groups. This prevented data from being pooled and analysed as large cohorts. Over the past few years, this is changing. There are now several efforts towards cooperative data collection. One of the oldest, is the PEDNET group in Europe, a collaboration of 30 centres in 16 countries . A system of well-defined data collection established nearly 15 years ago, has now led to a very robust database which has allowed some very important conclusions to be drawn, particularly with regard to inhibitors in previously untreated patients as well as other aspects of haemophilia management. The universal data collection process established by the Centres for Disease Control in the USA has also helped collect outcomes data, including inhibitors and some musculoskeletal aspects, from a very large number of PWH . Finally, the more recent European effort through the EUHASS programme is also a very good example of observational data collection that is much more comprehensive and is likely to provide data on the management of haemophilia that would simply not have been possible otherwise .
We need to recognize though that these efforts cover only some parts of the world. Can similar systems be created in other parts of the world? The WFH has been collecting basic information on haemophilia care in all its member countries for over a decade. This process has been better defined and enhanced in the last 2 years. This is truly a remarkable source for further data collection and an opportunity that should be tapped to help all PWH in the world get better care. Towards this end, the WFH is initiating a programme of data collection that will focus on specific questions and find the right centres in the world that can provide such data over a period of time or even as a cross sectional survey.
In conclusion, while we applaud the many advances in the management of haemophilia over the last five decades, we must also recognize that not enough effort has gone into creating strong evidence around the most important aspect of the treatment of this disease – prophylactic replacement therapy. There is lack of good evidence for all the core issues – time for starting, doses and duration and the associated outcomes. Only recently have appropriate instruments been developed for systematic outcomes assessment but now the challenge is to convince all stakeholders to use them. These are not easy tasks and will require considerable motivation, resources and international collaboration to achieve the goals. It is good that we have begun to move in that direction. These efforts must be coordinated and international organizations such as the WFH and ISTH, as well as regional organizations, could play important roles in helping those efforts. Such data will not only help establish evidence-based haemophilia care all over the world, but will also allow for better healthcare planning to be done and informed choices to be made with the resources available.
The data from the WFH Global survey and the figure were kindly provided by Mark Brooker and Aicha Traore from the World Federation of Hemophilia. I am grateful to Mike Makris and Marijke van den Berg for their comments on this article.
AS has received a competitive research grant from and serves on the grants review committee of the Bayer Hemophilia Awards Program. He is a member of the international advisory boards of Bayer HealthCare, Baxter and Novo Nordisk.