The burden, epidemiology, costs and treatment for Duchenne muscular dystrophy: an evidence review

The remit of our review was to identify, collate and describe contemporary evidence of epidemiology (prevalence and mortality), burden (severity and progression), illness costs (direct and indirect) and treatment patterns (pharmacological and other) of Duchenne muscular dystrophy. Current guidelines were also scrutinised for the latest treatment recommendations.

Searches were carried out from 2005 to June 2015 in 10 databases to identify information on the epidemiology, prevalence and burden of DMD. Guideline searches were undertaken to identify management and treatment of DMD. A pragmatic internet search was also carried out to look for sources to support evidence gaps in prevalence of DMD. Additionally, email alerts and RSS feeds were set up to ensure the latest research was not missed. Further details of searching methods including example search strategies can be found in Additional file 1: Appendix 1.

The main Embase strategy was independently peer reviewed by a second Information Specialist, using the Canadian Agency for Drugs and Technologies in Health (CADTH) checklist [9].

Titles and abstracts identified through electronic database and web searching were independently screened by two reviewers (drawn from a team of SR, RL, AH, MB, WJ) in order to determine whether they met the criteria for inclusion in the review. During this initial phase of the screening process any references which obviously did not meet the inclusion criteria were excluded. Full paper copies were obtained for all of the remaining references. These were then independently examined in detail by two reviewers (drawn from the team above working in pairs). All papers excluded at this second stage of the screening process were documented along with the reasons for exclusion. With respect to both screening stages, any discrepancies between reviewers were resolved through discussion or the intervention of a third reviewer (SR or NA).

Details are reported in Additional file 2: Appendix 2. In summary, aside from prevalence studies, where the general population (or subsets thereof) was of interest, we included all studies which described the population as DMD, even if details on diagnostic methods were missing. We excluded any studies which only reported on mixed populations (e.g. included Becker Muscular Dystrophy (BMD) or other forms of non-Duchenne Muscular Dystrophy).

Epidemiology and burden of disease outcomes of interest were: point prevalence, birth prevalence, demographic characteristics, clinical characteristics of the disease, mortality, incidence/prevalence of comorbidities and progression of the disease.

Quality of life (QoL) outcomes of interest were: the impact of the disease on quality of life (of patient and caregiver) as measured using a generic and disease specific or symptom specific measures.

Cost of illness outcomes of interest included patient and caregiver costs.

We also sought information about current treatment guidelines and treatment patterns.

Case studies were only included where evidence gaps could remain after consideration of other study types. Countries of interest included those in European Union (EU), South America, North America, Japan and Turkey (following advice from content experts at BioMarin Pharmaceuticals). For guidelines, countries of interest were restricted to EU countries and North America.

The years of interest were 2005 to 2015 inclusive. Due to the large number of papers retrieved and in order to concentrate on the most recent evidence, we decided to focus on records from 2010 onwards. Where evidence gaps existed, we sought records from earlier dates…

Studies were not limited by language or publication status (unpublished or published).

Data extraction was performed by two reviewers independently (drawn from a team of SR, RL, AH, MB, WJ). Any discrepancies were resolved through discussion or through the intervention of a third reviewer (SR or NA). Exemplar data extraction sheets are presented in Additional file 3: Appendix 3.

Two reviewers (drawn from a team of SR, RL, AH, MB, WJ) independently assessed each of the studies using a recommended tool, STROBE [10]. Any discrepancies were resolved through discussion or the intervention of a third reviewer (SR or NA).

Results are presented in Additional file 4: Appendix 4.

Our review found a temporal trend from using both genetic testing and muscle biopsy towards only using genetic testing to identify cases of DMD. Population characteristics in earlier studies may therefore be different to those completed more recently.

Two studies reported birth prevalence and five studies reported point prevalence (see Additional file 4: Appendix 4 Table A8 for characteristics and Table 1 for results). Quality of study reporting was assessed in Additional file 4: Appendix 4 (Table A1 and Table A2). Both birth prevalence studies were judged to be of medium quality but lacked adequate description of study participants [2, 3]. Two of the point prevalence studies were judged to be of medium quality but again lacked an adequate description of study participants [12, 13]. The remaining three studies were judged to be of low quality [14‐16]. Romitti [14] did not report an adequate description of the study design, nor did it fully describe the eligibility criteria or study participants. Mah [15] failed to adequately describe the eligibility criteria, outcomes or study participants. Bladen [16] also failed to provide adequate descriptions of eligibility criteria and study participants. Thus poor reporting makes it very difficult to assess possible changes in the way DMD has been defined over time.

Of the studies reporting birth prevalence, one USA study by Mendell [2] carried out a study on newborn screening for DMD in one of the four main birthing hospitals in Ohio. Creatine kinase (CK) levels in newborn screening blood spots were measured followed by genetic analysis. The authors suggest this approach minimises false-positive testing. Birth prevalence was reported as 15.9 per 100,000 newborn males. A second study, Moat [3] reported on a newborn blood spot screening programme for DMD over a 21 year period in Wales, UK . Again, CK levels in newborn screening blood spots were measured followed by genetic analysis/muscle biopsy and adjusted for false negatives and cases identified where parents declined to participate in screening. Birth prevalence was reported as 19.5 per 100,000 new-born males.

Of the studies reporting point prevalence, a study by Bladen [16] reported on TREAT-NMD, a worldwide network for neuromuscular diseases which supports new therapies for patients . The network was reported to have many functions including clinical and epidemiological research. From the figure presented, the number of patients per country in the national DMD registry can be estimated and the point prevalence calculated. For France, USA, UK and Canada the point prevalence of DMD was calculated as 10.9, 1.9, 2.2 and 6.1 per 100,000 males, respectively. Mah [15] reported on a population based study of dystrophin mutations in Canada. Of the 773 individuals with dystrophinopathy as confirmed by genetic testing (97%), muscle biopsy (2%), or family history (1%), 529 had DMD. Point prevalence of DMD was reported as 10.3 per 100,000 males aged 0–24 in Canada based on the 2006 consensus. Rasmussen [13] reported on children with neuromuscular disorders from a region of South Eastern Norway. Diagnosis was confirmed by genetic testing and/or muscle biopsy. The point prevalence of DMD was 16.2 per 100,000 males under 18 years of age in this region reported on 1^st July 2005. Romitti [14] presented population based prevalence estimates for DMD and BMD in 6 US states based on the Muscular Dystrophy Surveillance, Tracking, and Research Network (MD STARnet) as established by the Centers for Disease Control and Prevention. Diagnosis of DMD was based on symptoms and age at onset, creatine kinase value, results of dystrophin mutation analysis testing, muscle biopsy reports, and family history. Point prevalence of DMD was 10.2 per 100,000 males aged 5–24 in 2010.

Few studies reported the prevalence of DMD in relation to an all age male population. However, we did find one study by Norwood [12] which reported a detailed population study of patients with genetic muscle disease in northern England. Although outside of our inclusion criteria, it presented most recent data on total population (without age restriction). The point prevalence was reported as 8.3 per 100,000 males on 1^st August 2007 based on 124 cases identified through genetic testing and muscle biopsy.

We identified three studies reporting information on the survival of DMD patients [17‐19]. Quality of study reporting was assessed in Additional file 4: Appendix 4 (Table A3). Both Rall [19] and Kieny [20] were found to be of medium quality but neither provided an adequate description of study participants. Passamano [18] was assessed as low quality as it failed to adequately describe study design, outcomes or study participants and it was unclear if the study population was representative of the target population.

These three European long term retrospective cohort studies have each traced patients over a minimum of 30 years. All three studies (one each from Italy, France and Germany) reported median survival between 24 and 26 years. In the French study by Kieny [17, 20], median survival (calculated using the Kaplan-Meir model) was reported as 25.8 years for patients born between 1955 and 1969 and 40.9 years for patients born after 1970; the authors suggested that this difference was linked to greater availability of ventilator assistance through tracheotomy in the later birth cohort. In an Italian study by Passamano [18], the percentage overall mortality was assessed for patients when aged 20 and 25 born in either the 1960s, 1970s or 1980s. The study found that for those born in the 1960s, 76.7% would have died by the age of 20 and 86.5% by the age of 25; for those born in the 1970s, 46% would have died by the age of 20 and 69.4% by the age of 25; for those born in the 1980s, 40.2% would have died by the age of 20 and 50.8% by the age of 25. A study in Germany by Rall [19] also looked at patients born in the 1970s and found that median survival was 24 years, although this finding was sensitive to diagnosis method in that subjects with only a clinical diagnosis (as opposed to molecular testing) had a higher (67%) chance of reaching 24 years. Details of studies reporting mortality are set out in Additional file 4: Appendix 4 Table A9 with results in Table 2.

Forty-seven studies reported some information relating to the severity of DMD and/or its progression and 27 of these reported that they identified cases using genetic testing. The quality of reporting of these studies was recorded in Table A4. Three studies were judged to be of high quality [21‐23] satisfying all the criteria. The remainder of the studies were medium and low quality, typically not providing adequate information on study participants leading to uncertainty as to whether the population was representative of the target population.

Study characteristics are set out in Additional file 4: Appendix 4 Table A10. Considerable variation in the methods used to measure severity was identified. We found considerable heterogeneity between studies in terms of criteria used to assess ambulatory status, wheelchair usage, mobility, scoliosis, cardiac and respiratory function and intelligence. Two studies reported the distribution of general severity i.e. a summary measure of disease status in the DMD population, both of which were based on ambulatory status as described by Bushby et al [6]. There was a far higher percentage of the German population [23] in the most severe equivalent category i.e. late non-ambulatory and non ambulatory with confinement (stages 4 and 5): 47.6% versus 35.8% in the US [24]. It is not clear why this should be the case, although it should be noted that the German study was of better quality and was published more recently. Results are set out in Additional file 4: Appendix 4 Table A11.

Ten studies reported cross-sectional data on loss of ambulation, either as the percentage who have already lost ambulation or mean age at loss of ambulation. Percentage loss of ambulation varied from 32.6% in a study of the whole DMD population across four continents (Bello [25]) to 56.4% in a Japanese study (Nakamura [26]) although age of participants was reported in neither study. The relationship with age was shown clearly in a US study by Mayer [27] in that there was no loss of ambulation before age 8 years and progressive loss until age 16–18 years, after which loss was 100%. One French study (Martigne [28]) reported a mean age of ambulation loss of 10 years. Unsurprisingly similar results were found in six studies reporting wheelchair use, four as percentage [29‐32] and two as time to first use [19, 33]. Percentage use and time to use were unsurprisingly similar to those for loss of ambulation.

Eight studies reported mean six minute walking distance (6MWD plus, in some cases, other measures of mobility. 6MWD varied from 288.7 m reported in Pane [34] for those able to walk less than 350 m and aged at least 7 years to 428.7 m in those able to walk at least 350 m in the same age group. The only international study had an estimate of 361.1 m for those at least 5 years old (McDonald [22]).

The mean time to climb four stairs was varied between 2.5 s for those less than 7 years old to 6.6 s for those at least 7 years old in the international study (McDonald [22]). In the same study, mean 10 m run/walk time was 4.8 and 7.1 s respectively.

The percentage of the DMD population with scoliosis was reported in four studies from three countries and one multinational study. The percentage of the DMD population with scoliosis varied between 3.9% in a Japanese study of males with no age restrictions (Nakamura [26]) and 52.1% in a French study of boys ranging between 6 and 19 years old (Khirani [35]). The variation with stage of disease was shown in a multinational study in that the percentage was lowest at 16.6% in those in the early ambulatory stage (median age 7.2 years) and highest at 77.6% in the late non-ambulatory stage (median age 19.9) (Janssen [36]).

Cardiac function or percentage with cardiomyopathy was reported in six studies. Mean shortening fraction varied from 21.2% in the whole DMD population (left ventricular shortening fraction) from the US study by Ashwath [37] to 35% for those at least 10 years old from a US study by Thomas [38]. This variation appears to be consistent with age and thus disease stage. Percentage with cardiomyopathy varied similarly from 21% in younger boys (mean age (sd); 7.2 (2) years) (Thomas [38]) to 57.3% in a DMD population aged 10 years or more (Ashwath [37]). One study showed inter-country variation from 41.9% in Denmark to 52.4% in the UK in adult populations (Rodger [39]). It is not known whether demographic differences between populations may explain this difference.

Respiratory function, whether measured by percentage on assisted ventilation, time to introduction of ventilation, percentage of predicted Forced Expiratory Volume (ppFEV1) or percentage of predicted Forced Vital Capacity (ppFVC), was reported in 14 studies. The percentage of all DMD patients on assisted ventilation varied very widely from 0% in a Brazilian study of boys (mean age 11 years) by de Moura [29],0.7% in a multinational European study (mean age 13 years) by Vry [40] and 22% in a Japanese study (mean age not reported but most individuals described as less than 20 years old) (Nakamura [26]). Variation by disease progression was shown in the US study by Mayer [27] with a gradual decline from 126.6% predicted forced vital capacity (FVC) in the under 6 years age group to 7.3% predicted FVC in those aged 20 to 22 years.

Only one study (n = 4) by Khirani [35] in France reported the annual change in percent predicted Forced Vital Capacity (ppFVC). They found a 4.9% decline in respiratory function for patients with a mean age at baseline of 11.6 years. The percentage on assisted ventilation and age to start of ventilation after long term follow-up were reported in three moderate sized studies (no distinctions were made between night-time, daytime or continuous ventilation). After a mean follow-up of 18.3 years Martigne [28] found that 20% of study participants in France (mean age at baseline 13.0 years) were on assisted ventilation and the mean age for start of assisted ventilation was 16.8 years. In another French study, Kieny [17] claimed a follow up of 30 years, although the mean follow-up duration was not reported, and reported a much higher percentage of participants on assisted ventilation (65%) with a median age at start assisted ventilation of 20.1 years. The percentage on ventilation was also found to have increased from 60% before 1970 to 83% during and after 1970 with a decrease in the age at start of assisted ventilation from 20.1 to 18.3 years. The median age at start of assisted ventilation was essentially the same (i.e. 20 years) as that reported in the German study by Rall [19]. The authors of this study suggested that prolonged survival of DMD patients born after 1970 was directly associated with increased use of ventilation with tracheotomy especially when performed early. This was the only study to have made such a claim.

Two studies reported measures of intelligence, one reporting a mean score of 86.4 (compared to a mean score of 107.7 for a non-DMD group) on the Wechsler Intelligence Scale for Children-Revised in boys between 6 and 12 years (Lorusso [31]) and one reporting a mean score of 89.5 (compared to 100 for a non-DMD group) on the Bayley III cognitive composite instrument is in boys younger than 3 years from the US (Connolly [41]).

Progression of DMD can be measured in various ways. Those that are reported in the literature include changes in ambulatory status, ambulatory capacity (including 6MWD), and respiratory function.

Five studies followed up ambulatory status for those with DMD. Three of these studies began with those that were ambulant, two that followed up for 3 years in Italy by Pane [34] and in Italy and Belgium by Pane [42] and one that followed up for 7 years in the UK by Ricotti [43]. A study by Mah [44] followed up ambulant and non-ambulant boys for 1 year; this was an international study where information was not reported by country. Finally, Soderpalm [45] followed up anyone with DMD regardless of ambulatory status for 4 years in western Sweden.

In a study of boys in Italy (mean age 8.2 years), Mazzone [46] found a 3% loss in ambulation at 1 year follow-up. Over 3 years, the percentage who lost ambulation varied from 5.2% for those who with baseline 6MWD of ≥ 350 m and ≤ 7 years old to 64% with baseline 6MWD of <350 m and were ≥ 7 years old, reported in the follow up study by Pane [34] . In this study, loss of ambulation after 3 years, for combined sub groups, was 29%. In a study of DMD in Italy and Belgium, loss of ambulation was reported as 2.1% after 1 year in those who could originally walk ≥100 m (mean age at baseline 7.9 years) (Pane [42]) Mah [44] reported that, after 1 year, the percentage loss of ambulation increased from 43% at baseline to 57.1% at follow-up (mean age 12.0 at baseline). In a Swedish study, Soderpalm [45] for 18 patients aged between 2 and 19 years reported proportions of non-ambulant increasing from 22 to 50% over a 4-year mean follow up period. Therefore, it appears that participants in the Mah study may have been at an earlier stage in the disease. Median age for loss of ambulation was estimated at between 12 and 14 years by Ricotti [47] in the UK. Results are set out in Table 3 with changes in 6MWD in Table 4.

We found no evidence on the impact of specific mutations on severity/disease progression.

We found 18 studies reporting treatment patterns for DMD patients. One study was assessed as high quality [23], 12 were assessed as medium and five as low (see Additional file 4: Appendix 4 Table A5) [4, 35, 39, 40, 47]. Typically low and medium quality studies neglected to report a description of the study participants or outcomes.

This section reports on different treatment regimens reported in different studies. Ideally this would be linked to experiences in terms of clinical outcomes; however, this is not possible because of the considerable variation in reporting of outcomes as well as heterogeneity of populations considered. Fourteen studies reported levels of corticosteroid usage, with a further four studies reporting on different aspects of care. International variations in use of corticosteroids, scoliosis surgery, ventilation and physiotherapy were found.

Usage of corticosteroids was found to vary by ethnicity with 67.6% of white American DMD patients having this treatment as opposed to only 40.5% of black American DMD patients (Fox, [48]). Characteristics of the studies providing evidence for corticosteroid use are set out in Additional file 4: Appendix 4 Table A12 with variation in use percentages shown in Fig. 2.

We found a number of studies reporting uptake of non-drug therapy. Uptake of scoliosis surgery was reported in one French study reporting that 52% of DMD patients underwent this surgery between 2001 and 2011 (Khirani [35]).

Another French study by Kieny [17] was the only one to report on uptake of ventilation. The prime focus of the study was to assess life expectancy over the period 1981 to 2011 and the study therefore relates to DMD at all ages. Kieny and others have suggested that ventilator assistance, mostly through tracheotomy, prolongs life expectancy. Although the numbers of cases were fairly low, only 27.9% of patients born prior to 1970 underwent tracheostomy, whereas this proportion had risen to 47.8% for patients born after 1980 (an even higher percentage was recorded for patients born between 1970 and 1980 at 58.5%).

Rodger [39] reported the most extensive results for non-pharmacological management of DMD patients. The study not only compared and contrasted treatment uptake in Germany, UK, Denmark and Eastern Europe but also did so separately for children and non-ambulatory adults. Use of physiotherapy services was particularly high in Germany and Denmark with 93.2 and 87.8% of non-ambulatory adults receiving this service, respectively. The comparative percentage for Eastern Europe was 51.3% with the UK only achieving 21.4%. A similar picture was evident for services received by children with DMD where utilisation was 91.8 and 93.3% in Germany and Denmark respectively but only 73.5 and 55% in Eastern Europe and UK respectively. These findings should also be considered alongside the reported weekly utilisation of these services, where the UK also has the lowest levels.

Rodger [39] also reported comparative experience of regular assessments/check-ups for non-ambulatory adults with DMD from UK, Denmark and Germany. The greatest variation was evident for 6 monthly lung function assessment with 45.2% seen in UK but only 7% in Denmark, 6 monthly cardiac function assessment with 33.8% uptake in Germany but only 9.5% in UK and hospital-based planned check-ups with 67.3% in Germany and 25.3% in Denmark. This points to considerable heterogeneity of care patterns from country to country.

Thirteen studies reported either HRQoL (see Additional file 4: Appendix 4 Table A13) or utilities (see Additional file 4: Appendix 4 Table A14). Parent proxy scores, where collated, were similar to directly elicited values. The quality of reporting of HRQoL studies is presented in Additional file 4: Appendix 4 Table A6. There are 2 high quality studies [21, 23], eight medium quality studies and three low quality studies [49‐51]. Typically low and medium quality studies neglected to report a description of the study participants, outcomes or eligibility criteria.

The most frequently used tool for measuring HRQoL was the PedsQL, used in five studies. Three of these studies were conducted in the USA, the largest (n = 406) by Uzark [52] in four age groups. Bendixen [53] used two age groups (cut-off at 10 years) and Lim [50] for a cohort of boys and their parents (as proxies). Henricson [21] focused on those who were ambulatory, whereas Schreiber-Katz [23] covered the whole DMD population and subdivided by stage according to ambulatory ability. All studies provided at least the total score and two studies, by Schreiber-Katz [23] and Uzark [52] compared the DMD individual value with one elicited from the parent as proxy. Henricson [21] also used PODCI alongside PedsQL.

One study by Pangalila [54] in adults only, in the Netherlands, compared two different instruments, SF-36 and World Health Organization Quality of Life instrument (WHOQOL-BREF) as well as reporting the Fatigue Severity Score (0 to 5 scale) and Hospital Anxiety and Depression Scale (HADS) (0 to 21 scale). Simon [55] reported the Life Satisfaction Index for Adolescents (LSI-A) in four age groups of boys in Brazil, Baiardini [56] reported the Children Health Questionnaire - Parent Form 50 in Italian boys, Bendixen [51] reported the CAPE (0 to 5 scale) in the US and Canada, de Moura [29] reported the Autoquestionnaire Qualité de vie Enfant Imagé (AUQEI) in Brazil and Houwen-van Opstal [57] reported the KIDSCREEN-52 physical domain in the Netherlands. Interestingly, Pangalila [54] concluded that adults with DMD are ‘generally satisfied with their overall quality of life.’ Also, there was little variation according to stage as shown in both Houwen-van Opstal [57] on the KIDSCREEN instrument and Simon [55] on the LSI-A instrument with no clear trend.

Two papers reported utility values for DMD patients. One large multinational study Landfeldt [58] (n = 770) estimated Health Utilities Index (HUI) values (from both the boys and parents perspective) in four countries, Germany, Italy, the UK and the US. In this study the average for all DMD boys was 0.48 (considerably lower than perfect health (HUI = 1)) but inter-country variation was not large (from 0.43 in the UK to 0.52 in Italy). The other study, reporting utility values, Pentek [49], used EuroQol – 5 Dimensions – 5 Levels (EQ-5DL) for 57 boys with DMD in Hungary. However this study was judged of low quality reporting according to STROBE criteria, largely because of uncertainty surrounding representativeness of those evaluated.

We found one high quality study [23], one medium quality [58] and one low quality [32] study providing evidence of the cost of illness for DMD. The main features of these studies are set out in Additional file 4: Appendix 4 Table A15. Taken together, they represent a strong source of evidence of costs accrued at different stages of the condition and across different countries.

The high quality study assessed the burden of illness for German DMD patients and caregivers in 2013 [23]. This study provided an assessment of cost for customised severity groupings (based on Bushby et al [6]), thereby enabling better understanding of the costs associated with progression of the condition. For Schreiber-Katz [23], 2013 total direct medical costs ranged from 4,420 Euro (€) for (stage 1) patients to 68,968 Euro (€) for (stage 5) patients i.e. nearly a 16 fold increase. It is perhaps worth noting that the annual cost of hospitalisation represented between 7% (stage 4) and 14% (stage 1) of all direct costs in the Schreiber-Katz [23] study. Further subdivision of direct costs was provided including a detailed breakdown according to service headings (as opposed to staff group headings). The key variation in unit costs associated with progression relate to provision of medical aids costs for (stage 5) which were 104.5 times more than for (stage 1) and costs of respiratory management costs for (stage 5) which were 923.7 times more than for (stage 1). Results are set out in Table 5.

The study by Landfeldt [58] provided comparative direct costs for Germany, UK, US and Italy in terms of staff groupings as well as service headings. Most inter-country variation in large spend categories was found for physio/OT where spend in the US was 4.5 times that in Italy, psychology where spend in the US was 14.4 times that in either Italy or Germany, specialist physicians where spend in the US was 21.9 times that in Italy and visits to healthcare professionals where spend in the US was seven times that in Italy. The study by Larkindale [32] reported on medical costs but without comparative data or comparable cost categorisation.

Indirect costs were also quantified in each of the three cost of illness studies and expressed in terms of non-service costs and co-payments (see Table 6). Schreiber-Katz [23] assessed the costs of time off work and the impact on parents for each of the DMD stages in their study. They observed that (stage 5) patients had costs which were 2.5 times higher than (stage 2) patients in terms of costs of time off work and that (stage 1) patients had costs which were three times higher than (stage 5) patients in terms of impact on parents. Interestingly, taken together, these two forms of indirect costs represent a much greater cost than direct costs for (stage 1) patients (13,078 Euro (€) as opposed to 4,220 Euro (€)) and also a greater cost for (stage 5) patients (32,907 Euro (€) as opposed to 22,989 Euro (€)). The Landfeldt [58] study provided inter-country comparisons of indirect costs. There was broad similarity in terms of time off work and income loss. However, US funding mechanisms explain the relative high cost of insurance premiums. Loss of leisure time was costed as higher in Germany than in the other countries [58]. Information on co-payments was also provided as part of the Landfeldt [58] study. Italy has the highest co-payments of all four countries in each of the categories considered. No study estimated cost of lost productivity due to reduced life expectancy.

Social care costs were assessed in all three cost of illness studies for both services and equipment/adaptations – see Tables 7 and 8. The Schreiber-Katz [23] study provided evidence that informal carer and social care costs are positively associated with severity as measured by Stage of DMD. The costs of informal care time were 5.4 times higher for stage 5 patients than for stage 1. The costs of social care are also 5.4 times higher. In the international comparative study, Landfeldt [58] found broad similarities in the costs of informal carer time but the costs of home help; personal assistants etc. were notably higher in UK than in comparator countries. Information from Schreiber-Katz [23] on travel/car adaptations did not suggest a relationship with DMD progression and the comparative data provided by Landfeldt [58] suggested broad similarity in spend on equipment costs between UK, US and Germany with spend in Italy being noticeably lower.

Finally, out-of-pocket expenses were considered in the Landfeldt [58] study as detailed in Table 9. These clearly represent a considerable cost of illness and there was a degree of similarity between countries.

Three key sources were identified in respect to guidelines for the treatment of DMD (see Table 10). In 2010, recommendations were made to consider glucocorticoids, including Deflazacort and Prednisone, as first line therapies for DMD patients of 2 years and over whose condition was not improving (Bushby [6]). Glucocorticoid therapy is highly recommended for patients of 6 years and over to slow the decline in muscle strength and function. It is also recommended that patients, in particular those with pre-existing risk factors, are monitored for side effects such as weight gain, growth retardation, bone demineralisation and increased fracture risk. Supplementary guidance for respiratory management of DMD patients was also published. Birnkrant [59] produced supplementary guidance for respiratory management of DMD patients which recommended equipment, procedures, tests, and diagnostic evaluations, emphasising the assessment of hypoventilation and the identification of specific thresholds for forced vital capacity (FVC), peak cough flow, and maximum expiratory pressure. More recently, results of an international collaboration were published (Kinnett [11]). These guidelines highlight the importance of a multidisciplinary approach to the care of DMD patients, addressing the primary and secondary manifestations of the condition including use of corticosteroids, coronary care, pulmonary care, physical therapy, surgical considerations and psychosocial care.