FormalPara Key points

Economic evaluations in haemophilia are increasing in number and quality, providing information on the cost effectiveness of different treatment options in patients with and without inhibitors. However, the variability in the methods used and the differences in the treatments under assessment often make the identification of general trends difficult.

Only immune tolerance induction showed consistent results in its cost-saving profile compared with the treatment with bypassing agents. While prophylaxis with recombinant Factor VIII seems a cost-effective option, compared with on demand, this observation depends on several factual variables, including (1) annual background bleeding rate, (2) relative efficacy, (3) dosing regimens and (4) the price of recombinant Factor VIII.

The introduction of new treatments is changing haemophilia management and making new economic assessments based on solid evidence and methods mandatory. In light of the evolution of the haemophilia health economic literature, future analyses based on short-term time spans and intermediate outcomes will neither match current scientific standards nor fulfil the requirements of policy-making bodies.

1 Introduction

Haemophilia is a rare congenital bleeding disorder characterised by gene abnormalities leading to defective or missing clotting Factor VIII (FVIII), called haemophilia A (HA), and Factor IX (FIX), called haemophilia B (HB), with a variable impact both from clinical and quality-of-life points of view [1,2,3]. The prevalence of HA is approximately five times that of HB [4], with a worldwide frequency estimated at one per 5000–7000 male births [5].

Patients affected by severe HA or HB present with repeated, spontaneous and post-traumatic bleeding episodes that seriously interfere with their everyday lives. In particular, recurrent haemarthrosis (bleeding into the joint cavities) leads to serious deterioration of joint structures, with a consequent reduction in function and atrophy of the associated skeletal muscles [6]. In the last decade, there has been a continuous improvement in the treatment of patients with haemophilia, thanks to the availability of recombinant concentrates characterised by high efficacy and safety, and to the widespread adoption of prophylaxis as the replacement therapy regimen [7]. Indeed, patients with haemophilia can be treated on demand (OD) (i.e. following a bleed), or prophylactically to prevent bleeding and the consequent deleterious effects on joint status in the first instance. Increasing evidence supports the clinical and quality-of-life benefits over OD treatment of the adoption of prophylaxis, started before the age of 3 years, prior to two joint bleeds (‘primary prophylaxis’), but also after the onset of serial bleeding or established joint damage (‘secondary prophylaxis’ or ‘tertiary prophylaxis’, respectively) [8]. Starting prophylaxis early in life and after very few joint bleeds is associated with better joint outcomes [9].

The major complication in the treatment of patients with haemophilia is the development of inhibitory antibodies to FVIII or FIX, which occurs in approximately 30% of patients with HA and in approximately 3% of patients with HB [10, 11]. These antibodies reduce therapy effectiveness even to zero, neutralising the clotting activity and requiring the use of immune tolerance induction (ITI) treatment, a bypassing agent in prophylaxis or an OD regimen [12]. Immune tolerance induction therapy consists of providing a factor concentrate (FVIII for HA and FIX for HB) regularly at a high dose until the body is trained to recognise the treatment product without reacting to it [13]. When ITI is successful, the inhibitors disappear and the patient’s response to factor concentrates (FVIII or FIX) returns to normal. When ITI fails, the persistence of inhibitors at a high titre precludes the standard replacement treatment with FVIII/FIX concentrates and requires the use of bypassing agents in prophylaxis or an OD regimen, making patient management challenging. Indeed, the efficacy of bypassing agents, i.e. activated prothrombin complex concentrates (aPCC) and recombinant activated FVII (rFVIIa), needed to overcome the haemostatic interference of the inhibitor, is not comparable to that of factor concentrates [13]. Further, the treatments for patients with haemophilia with inhibitors are more expensive than the treatment for patients without inhibitors [13,14,15].

Currently, cost is one of the most important aspects in haemophilia care. Numerous studies have analysed the healthcare direct costs of patients with haemophilia [13, 15,16,17,18,19,20,21,22]. The costs of factor concentrates amount to 90% of the total healthcare direct costs of haemophilia care [23]. The scientific and political debate regarding the high cost of haemophilia treatment, its effects and the different use across different countries is increasing [24, 25]. A recent review [26] assessed the quality of reporting in more recent economic evaluations in haemophilia; focusing on the common methodological deficiencies and proposing standards for conducting and reporting future economic evaluations. However, there are still unresolved important questions on the economic evaluations of haemophilia treatments, such as:

  1. 1.

    Is it cost effective to provide recombinant factor prophylaxis for severe HA or HB without an inhibitor, or is OD therapy a more cost-effective option?

  2. 2.

    Is the provision of prophylactic therapy for adults with severe HA or HB cost effective?

  3. 3.

    Is ITI for patients with inhibitors cost effective?

  4. 4.

    Is prophylaxis for patients with inhibitors cost effective?

  5. 5.

    Will new recombinant factor products (which have an extended half-life) or non-factor replacement strategies (e.g. a recombinant, humanised, bispecific monoclonal antibody that bridges activated FIX and Factor X to restore the function of missing activated FVIII) offer good value compared with existing recombinant products?

In light of such crucial questions, we examined cost-effectiveness analyses conducted on treatment options in haemophilia, focusing on their results and their strengths and limitations. In the discussion, we identify some possible issues associated with the economic evaluations of new treatment options for patients with haemophilia, considering such new treatments are now coming onto the market [24, 25].

2 Methods

A systematic literature review was performed to collect and critically review the health economic evidence on the different treatments of patients affected by haemophilia. Our study updates a previous review of economic evaluations of haemophilic treatments options by Drummond et al. covering the years from 2008 until 2015 [26]. While the review by Drummond and colleagues focussed on methodological issues related to reporting economic evaluations in haemophilia, our review focuses on the assessment of empirical results. Studies published earlier than 2008 were not included given the focus of our review on modern treatments, and the lack of consensus on standardised reporting prior to that year [27]. To update Drummond et al.’s review, original studies and analyses published between November 2015 and September 2017 were searched for in PubMed and EMBASE, using “cost effectiveness” OR “economic evaluation” OR “cost analysis” OR “cost utility” OR “cost–benefit” OR “economic analysis” OR “pharmaco economic” OR “economic near model” OR “decision model” OR “economic study” OR “cost-effectiveness” OR “cost-analysis” OR “cost-utility” OR “cost–benefit” OR “pharmaco-economic” OR “decision-model” OR “economic-study” OR “cost” AND “haemophilia” OR “haemophilia” OR “factor VIII deficiency” OR “factor 8 deficiency” as keyword research terms.

To maximise retrieval of all pertinent papers, we applied medical subject headings (‘MeSH’ terms), or keyword searches when at all appropriate, and for the sake of completeness, we reviewed the reference lists of all relevant studies and reviews on the topic to identify additional studies of interest. Two members of the review team (AL, MM) examined the studies in a three-step process: (1) we selected the articles based on the title and abstract, (2) we retrieved and reviewed the potentially relevant articles selected based on the abstract in step 1, and (3) we analysed articles that met the inclusion criteria and included them in the review. Disagreement between the two reviewers was resolved by consensus of a third party (PAC).

To be included in the review, an article had to summarise findings in English, and compare treatments in patients with haemophilia, reporting information on costs and effects (e.g. cost-effectiveness study, and cost-utility study). Papers reporting information on treatment costs only (e.g. cost-of-illness study) and articles that did not assess treatment options but focused on other aspects of haemophilia management, such as diagnostic tests or surgical operations, and differences in access to healthcare services were excluded. The quality of identified studies was assessed by two reviewers (PAC and LSD) using the Consolidated Health Economic Evaluation Reporting Guidelines checklist [28]. Further, for each included study, we extracted the following data using a standardised form: aim, design, time horizon, perspective, subject characteristics, treatments, clinical parameters, results including costs, outcomes, incremental cost-effectiveness ratios (ICERs) and authors’ conclusions.

3 Results

The literature search identified 136 records from bibliographic database searches and supplementary searching (e.g. reference/citation checking and additional references). After title and abstract screening, 35 records were considered to be potentially relevant and, after full-text screening, 26 studies were considered eligible for inclusion (Fig. 1). Tables 1, 2, 3 summarise these studies.

Fig. 1
figure 1

Flowchart of the selection process

Full size image
Table 1 Studies on patients without inhibitors
Full size table
Table 2 Studies on patients with inhibitors [excluded immune tolerance induction (ITI)]
Full size table
Table 3 Studies on patients with inhibitors assessing immune tolerance induction (ITI)
Full size table

The majority of economic evaluations reported information on many items required by the Consolidated Health Economic Evaluation Reporting Guidelines checklist (see Table 1 of the Electronic Supplementary Material). The quality of the studies seems to have improved in the last few years, and this improvement is more evident if we compare the quality of the studies retrieved in our review with the quality of the studies published between 1996 and 2008 and assessed by Miners [29].

Within the included studies, 14 were cost-effectiveness analyses, seven were cost-utility analyses, while nine reported information on the cost and effectiveness of treatments, making it possible to estimate their cost effectiveness. Most studies were conducted in Europe (n = 13, 50.0%) and in North American countries (n = 5, 19.2%) or in both continents (n = 3, 11.5%), followed by Asia (n = 4, 15.4%) and Latin America countries (n = 1, 3.8%). The majority of studies used the third-party payer’s perspective for measuring costs, while three studies applied a societal perspective, although cost inputs were not always consistent with the perspective taken. Nine studies included drug costs only, while 15 studies included drug costs, direct medical costs (e.g. hospitalisation, visits/examinations/check-ups, tests, rehabilitation and management of bleeding) and non-medical costs (e.g. travelling costs). Only two studies measured indirect costs (e.g. productivity losses and parents’ lost workdays).

The articles retrieved were divided based on the target population: (1) patients without inhibitors and (2) patients with inhibitors. Eleven (42.3%) of the included studies assessed treatments for patients with haemophilia without inhibitors [1, 21, 30,31,32,33,34,35,36,37,38], while 15 (57.7%) studies assessed treatments for patients with inhibitors [13, 14, 39,40,41,42,43,44,45,46,47,48,49,50,51]. The details of the studies included in each group and the treatments assessed are reported below.

3.1 Patients Without Inhibitors

On-demand treatments were compared with one or more prophylaxis regimens in nine of 11 studies conducted in patients without inhibitors; while only two studies compared different prophylaxis regimens without including ON treatment (Table 1).

3.1.1 Comparisons of Prophylaxis vs. On-Demand Treatment in Haemophilia A and B

All studies except one included only direct costs using the insurance or National Health Service point of view. These studies used different outcomes to assess the cost effectiveness of prophylaxis vs. OD treatment: five studies used quality-adjusted life-years (QALYs), four studies used number of bleedings avoided, and one study used both QALYs and number of bleedings avoided (Table 1). The majority of the studies assessed the cost effectiveness using a Markov model, while three studies used direct evidence from either a clinical trial or a retrospective cohort study.

Only one study reported prophylaxis as a cost-saving treatment compared with OD treatment, using the UK National Health Service point of view [34]. In the same multi-country study, primary prophylaxis was more effective and costly compared with OD treatment when the analysis was conducted using the US or Swedish third-party payer’s perspective, with an ICER of US$68,108 and SEK484,888 per QALY gained, respectively [34]. The other eight studies evaluated prophylaxis as the most effective and more expensive treatment. Whether prophylaxis is a cost-effective strategy mainly depends on the willingness to pay for an incremental unit of effect. Furthermore, other factors critically influencing results were type of factor used (e.g. recombinant or plasma-derived FVIII), the treatment efficacy (number of bleedings associated with or avoided by the investigated strategies), the choice of outcome indicators (e.g. bleeding events avoided or absence of joint damage, or QALYs) and the time horizon.

In the study by Castro Jaramillo et al., when primary prophylaxis was provided throughout life using recombinant FVIII (rFVIII), the additional cost per QALY gained vs. OD treatment was US$91,147, compared with US$54,995/QALY when plasma-derived FVIII was considered [35]. The use of different outcomes is mainly associated with the time horizon adopted in the analysis: studies using a short time horizon usually estimated the cost per bleed avoided, while studies with longer time horizons generally provided estimates in terms of cost per QALY gained [35]. Gringeri et al. reported an ICER of €7537 per bleeding event avoided comparing primary prophylaxis with OD treatment directly estimated within a randomised controlled trial following patients up to 7 years (median 4 years) after randomisation [1]. Risebrough et al. simulated the 5-year cost and outcomes in a cohort of patients with haemophilia from the age of 1 year, estimating an ICER of Can$3192 per joint bleeding avoided and of $Can244,085 per target joint avoided comparing escalation-dose prophylaxis with OD treatment. In the same study, when QALY was assessed as an outcome, the ICER was Can$542,938 per QALY gained [30]. In the only study conducted in patients with HB, Polack et al. reported an ICER of €22,605 per bleeding avoided comparing prophylaxis with OD treatment, comparing a mixed treatment (50% OD and 50% prophylaxis) provided with two types of FIX: recombinant vs. plasma derived [21].

3.1.2 Comparing Prophylaxis Regimens

In the analyses conducted on patients without inhibitors, three studies compared different prophylaxis regimens. All studies used only direct costs and most of these used the QALY as the outcome (Table 1). Risebrough et al. compared a standard-dose primary prophylaxis with an escalation-dose prophylaxis, a regimen that starts with a low dose and frequency of infusions, which is increased (adjusted), if a patient experiences bleeding events. In this study, the standard-dose primary prophylaxis was more effective and costly compared with escalation-dose prophylaxis with an ICER of Can$9046 per bleeding avoided and an ICER of Can$ > 1,000,000 per QALY gained [30]. In the study by Colombo et al., three prophylaxis regimens were simulated, with primary and secondary prophylaxis proving to be cost-effective options compared with a “hybrid strategy” [33].

Finally, Iannazzo et al. compared standard prophylaxis with a prophylactic regimen based on individual patients’ pharmacokinetics in patients with HA, aged between 10 and 65 years. The pharmacokinetic-driven prophylaxis was a cost-saving option compared with standard prophylaxis, and this result was achieved through the optimal reallocation of rFVIII units used in the haemophilic population based on individual patients needs [37]. As reported by Iannazzo et al. [37], pharmacokinetic-driven prophylaxis proved to be a cost-saving approach to treat children with haemophilia, compared with standard prophylaxis, also in USA. Pharmacokinetic-driven prophylaxis decreased the cost of therapies and visits. The amount saved amounted to €8986 per patient/year [38].

3.1.3 Patients with Inhibitors

The included studies on patients with haemophilia with inhibitors reported heterogeneity in investigated treatments using different products and regimens (dose and frequencies) within the studies, making the results difficult to compare. On-demand treatments with different products (bypassing agents and rFVIII) were compared in six studies; while three studies compared prophylaxis regimens with OD treatment (Table 2). Immune tolerance induction treatments were assessed in six studies; in these studies, different ITI protocols were compared with each other and with OD treatment and/or prophylaxis with bypassing agents.

3.2 On-Demand Treatments with Different Products

Six studies investigated the cost and effectiveness associated with the rFVIIa compared with aPCC, for the management of bleeding episodes in adults and children with haemophilia and inhibitors. In three studies, rFVIIa proved more effective than aPCC in managing bleeding [40, 42, 43], while another three studies found aPCC to be more effective [39, 41, 45]. However, there was considerable heterogeneity in the outcomes considered, limiting the generalisability of results. As far as costs are concerned, we found conflicting evidence. According to Salaj and colleagues [42], You and colleagues [40], and Jimenez-Yuste and colleagues [43], rFVIIa was cost saving, while according to the groups of Hay and Zhou [41] and Steen Carlsson et al. [39], aPCC was less expensive than rFVIIa.

While part of the heterogeneity in results may be attributable to different patient characteristics (e.g. patients with mild-moderate vs. severe haemophilia), outcome parameters and treatment algorithms, based on evidence on the funding bodies, publication bias may have played a role in some of the papers. Many analyses used a single study or assumptions to estimate the dose/frequency and/or efficacy for the different products assessed. The selection of the study and the assumptions could have been influenced by various aspects, including the funding bodies, showing more favourable results for one product instead of another and creating a publication bias.

3.3 Prophylaxis vs. On-Demand Treatments

Of the three studies assessing the cost and effects of prophylaxis in patients with inhibitors, all studies compared aPCC-based prophylaxis with an OD strategy with aPCC or rFVIIa. [14, 44, 46]. Two studies were based on decision analytical models while one study was conducted within a randomised clinical trial (Table 2). Two studies compared aPCC prophylaxis with rFVIIa OD using a 1-year time horizon, reporting prophylaxis with aPCC as the less expensive treatment option [44, 46]. These studies were originally cost analyses, but also reported a clinical outcome, associating prophylaxis with aPCC with a lower number of bleeding episodes.

In a clinical trial, Lessinger et al. compared aPCC, infused prophylactically at a target dose of 85 U/kg of body weight (three times per week), with aPCC OD therapy at a target dose of 85 U/kg, used for bleeding episodes. The trial reported a significantly lower number of bleedings associated with 6 months of prophylaxis, with a cost per bleeding avoided of US$35,565 [14].

3.4 Immune Tolerance Induction Regimens vs. On-Demand and/or Prophylaxis Treatments with Bypassing Agents

Eradication of the inhibitor through ITI is generally recommended as the first treatment option, particularly in children with high-responding inhibitors because it provides the prospect of efficacious FVIII replacement therapy and the feasibility of prophylaxis with consequent preservation of joint status and quality of life [13].

Six studies evaluated the costs and effectiveness of different ITI protocols. Differences between studies were found with regard to methodology: (1) different time horizons, from 1 year to lifetime [47]; (2) different outcome measures, with QALYs, life-years saved, bleeding events, hospital stays and frequency of patients with inhibitor eradication; (3) different ITI regimes, e.g. low-dose ITI, high-dose ITI, risk-assessment ITI, Bonn ITI and Malmo ITI; mixed regimens including OD treatment with rFVIIa or aPCC followed by high- or low-dose ITI. Despite such methodologic differences, ITI was always less expensive and more effective compared with the alternative, i.e. OD or prophylaxis with bypassing agents. Within the different ITI protocols assessed, low-dose ITI and risk-assessment ITI seemed to have the most favourable cost-effectiveness profile [51].

Rocino et al. reported that the high cost of ITI treatment was counterbalanced by a high rate of success in the treatment of patients with inhibitors, which is associated with an additional gain in life expectancy and health-related quality of life reported by patients without inhibitors [13]. However, a low-dose ITI may be more economically convenient than a high-dose ITI when it is combined with aPCC prophylaxis. A synergistic effect of ITI with aPCC reflected a potential to reduce morbidity by lowering the risk for breakthrough bleeds [51]. Odeyemi and Danø [47] applied a cost-minimisation technique to compare the economic impact of using aPCC or rFVIIa to manage bleeding that occurred during the time between the appearance of the inhibitor and the start of ITI (pre-ITI), to improve the patients’ probability of achieving success with a low-dose ITI, instead of a high-dose ITI. The authors concluded that rFVIIa reported lower treatment costs (− £188,405) compared with aPCC, suggesting the management of bleeding pre-ITI with rFVIIa to increase the use of low-dose ITI and reduce costs.

4 Discussion

Although haemophilia is a rare condition, several economic evaluations of haemophilia treatments have been performed in the last decade. Not surprisingly, a relevant number of studies were conducted in patients with inhibitors, considering the high acquisition price of bypassing agents. All studies in patients without inhibitors, except one, were carried out on HA treatments, possibly reflecting the relative epidemiologic and economic burden of the condition. As reported in other studies, the variability in the methods used and the differences in the treatments under assessment often make the identification of general trends difficult [26, 29, 52]. This aspect is also reflected in our review, which only provides a clear answer to the question: “Is ITI for patients with inhibitors cost effective?”. In our study, only ITI in patients with inhibitors showed consistent and concordant results on its cost-saving or dominant profile, compared with the treatment with bypassing agents. However, when we tried to answer the question: “Is it cost-effective to provide recombinant factor prophylaxis for severe HA or HB without inhibitors, or is OD therapy a more cost-effective option?”, prophylaxis with rFVIII seemed a cost-effective option, compared with OD treatment. However, this observation depends on several factual variables, including (1) annual background bleeding rate, (2) relative efficacy, (3) dosing regimens and (4) price of rFVIII [30, 31, 33, 34, 50]. A difference in values and assumptions made for these variables created a significant variability in the cost of OD and prophylaxis treatment reported in the studies; which seems not related to the time of publication. When we analysed four studies conducted in the same country (Italy), the cost of rFVIII prophylaxis was different even within studies published in the same year. In 2011, the study by Colombo et al. reported a 1-year prophylaxis cost per patient of €23,738 (regimen: 30 IU kg-1 2.5 times a week; €0.68 per IU) [33], while Gringeri et al. reported a cost of €79,668 (regimen: 25–40 IU kg-1 two or three times a week; €0.75 per IU) [1]. In 2016, Iannazzo et al. reported a cost of €265,859 (regimen: 30 IU kg-1 every 48 h; €0.65 per IU) [37] and Coppola et al. a cost of €29,006 (regimen: 25 IU kg-1 three times a week; 0.65 per IU) [36].

However, the uncertainty related to the results of these analyses is even more influenced by methodological variables, including: (1) the use of a short time horizon (e.g. 1 year, 6 months and even a single bleed) that is not able to detect the effect of a treatment for the entire life of a patient with haemophilia; (2) the use of outcomes with little generalisability (e.g. bleed avoided, target joint and pain intensity); (3) the use of a clinical effect related to a single study, conducted in a small sample of patients; (4) the use of non-standardised doses and the frequency of clotting factor infusion; (5) poor definition of the uncertainty related to the results and the parameters, and (6) differences in the cost and outcome discount rates.

Further, another issue to consider in the retrieved studies is the potential risk of publication bias that can lead to publication of only favourable results. Within the included studies, 19 of 26 (73%) analyses were funded by industry (Tables 1, 2, 3), raising the potential risk of publication bias. Unfortunately, the heterogeneity of the outcomes used in the evaluations make the assessment of publication bias very difficult. The use of a more standardised outcome in future analysis could make it possible to better assess this aspect.

Our observations are in line with those reported by Drummond et al. regarding the quality of reporting methods of economic evaluations in haemophilia [26]. The improvement in the quality of evaluations can be observed in some of the recently published studies included in our analyses, potentially positively influenced by consensus initiatives [27, 53]. Although the methodological quality seems to be improving, weaknesses still exist, and a fruitful collaboration between clinical experts, budget holders and health economists is indispensable to best guide clinicians and decision makers. In fact, in patients without inhibitors, the initial economic evaluations conducted in the 1990s and in the first part of the millennium were usually short-term analyses comparing prophylaxis with OD treatment using cost per bleeding avoided as an outcome [29], with few exceptions such as the paper published in this journal in 2002 by Miners et al. [54].

More recent evaluations have evolved and become more sophisticated in different ways. First, the time horizon has been significantly extended, with more recent analyses adopting a lifetime approach, as natural for a genetic, chronic and lifelong disease like haemophilia. Second, outcomes of choice have moved away from bleedings avoided and moved towards QALYs, an evolution made possible thanks to the increasing availability of long-term evidence on outcomes, particularly on the association between frequency and severity of bleedings, development of haemophilic arthropathy and the global health status of patients, including utility. Third, the concept of prophylaxis itself has changed from standard regimens towards a personalised approach based on the individual patient’s pharmacokinetic and social expectations (e.g. physical exercise and sport) [37].

4.1 Future Directions

New concerns about cost effectiveness and the sustainability of old, new and future treatment options are being raised among clinicians and budget holders dealing with patients with haemophilia. The body of evidence directly or indirectly generated by economic evaluations published in recent years is now being applied to new treatment strategies already on the market or that will be marketed in the near future [55]. For decades, the treatment of haemophilia has been straightforward: simply replace the deficient protein with an available intravenous factor concentrate or use bypassing agents [56]. New categories of haemophilia therapy are now available or in late phases of development: engineered factor concentrates with extended half-lives, gene therapy and non-factor replacement haemostasis strategies. These products can make it possible to address gaps in the current approach to haemophilia management but do not provide a one-size-fits-all opportunity [56]. For example, the extended half-lives of rFVIII and rFIX could simplify the prophylactic regimens in HA and HB without inhibitors, reducing the frequency of infusions and extending the protection from bleeding, making the treatment more tolerable to the patient (and to parents) with a consequent improvement of therapeutic adherence. The use of novel non-replacement products, such as emicizumab, [55, 57] could be an alternative strategy in patients with or without inhibitors but can be used only for prophylaxis regimens and not to manage bleeding. Further, these therapies require much less frequent dosing than most factor products and are administrated subcutaneously rather than intravenously [56]. Gene therapy could offer a definitive cure but is still in the early clinical development stage, and the long-term safety and efficacy issues have already been raised. Finally, all of these treatments should be assessed considering their safety profiles and the possible immunological responses.

However, proper clinical and economic investigation of these strategies is also necessary, considering the different prices of these products and the possible impact of the sustainability of the treatment in patients with haemophilia. Considering the key role of factor concentrate costs in haemophilia healthcare [23], the assessment of the cost effectiveness of new and old products is crucial to understand the possible impact on the management of patients with haemophilia. To perform economic evaluations of these products, reliable data input and assumptions will be necessary, especially for the long-term effects, the expected adherence (reduced number of infusions, and more comfortable administration routes associated with the new products), the impact of (non)adherence in terms of clinical effect, product consumption and patients’ quality of life. In particular, transforming the reduced number of infusions, the lower treatment burden and a better route of administration into an improvement of the utility value will be the crucial challenge to be addressed while we are waiting for the future results of ad-hoc research. Future research on patients with haemophilia treated with the new products should be aimed at providing a real estimation of quality of life and utility impact reported by the patients, and at associating this impact with the different characteristics of these new products: reduced number of infusions, lower treatment burden, better adherence and better route of administration. All the discussed aspects should be taken into account if we want to answer the new questions that have been raised on the proposed value of new recombinant factor products (which have extended half-lives) or non-factor replacement strategies compared with existing recombinant products.

5 Conclusions

Haemophilia is a complex medical and social condition with high treatment costs. As reported in other studies, economic evaluations in this field are increasing in number and quality, highlighting the need for more robust analyses based on reliable methods, assumptions and data inputs [26, 29, 52].

Analysis of the retrieved studies indicates that prophylaxis seems a cost-effective treatment option compared with OD treatment for patients with HA and HB without inhibitors, particularly when more adequate, comprehensive and long-term analyses are considered and, second, when personalised prophylaxis is compared with a one-size-fits-all approach. For patients with inhibitors, ITI is dominant compared with any alternative course of care in which patients are placed on a regimen with bypassing agents and, again, an approach to ITI based on individual risk has the potential to be superior. In patients without inhibitors, we found a less clear picture when different regimens with bypassing agents were compared, as the available analyses either present conflicting comparative evidence on the cost of bleeding only, or express the outcomes in terms of cost per bleeding avoided, i.e. something that is not comparable to other healthcare interventions.

The introduction of new treatments will mandate assessment based on solid evidence. In light of the evolution of the haemophilia health economic literature, future analyses based on a short-term time span and intermediate outcomes will neither match current scientific standards nor fulfil requirements from policy-making bodies. Further, more reliable assessments that translate the new treatment options characteristics (e.g. reduced frequency of infusions, extended protection from bleeding and different routes of administration) in terms of cost-effectiveness profiles are needed to optimise the resources available for the management of patients with haemophilia.