Costs and health-related quality of life of patients with cystic fibrosis and their carers in France

JCF-01129; No of Pages 8

Journal of Cystic Fibrosis xx (2015) xxx – xxx www.elsevier.com/locate/jcf

Original Article

Costs and health-related quality of life of patients with cystic fibrosis and their carers in France Karine Chevreul a,b,c,⁎, Karen Berg Brigham a,d , Morgane Michel a , Gilles Rault e , BURQOL-RD Research Network 1 b

a URC Eco Ile de France, DRCD, Hôtel Dieu, 75004 Paris, France Université Paris Diderot, Sorbonne Paris Cité, ECEVE, UMRS 1123, 75010 Paris, France c Inserm, ECEVE, U1123, 75010 Paris, France d Université Paris Est, 94010 Créteil, France e CRCM de Roscoff, Centre de Perharidy, 29684 Roscoff cedex, France

Received 2 June 2014; revised 20 November 2014; accepted 20 November 2014 Available online xxxx

Abstract Background: Our goal was to provide data on the economic burden and health-related quality of life (HRQoL) associated with cystic fibrosis (CF) in France. Methods: A retrospective cross-sectional study was carried out on adults and children with CF, who completed an anonymous questionnaire regarding their socio-demographic characteristics, healthcare consumption and presence of a carer. Costs were calculated with a bottom-up approach, and HRQoL was assessed using EQ-5D. Results: 82 adults and 158 children were included. The total average annual cost of CF was €29,746 per patient. Total costs were higher in adults than in children and increased with disease duration. The average utility was lower in adults (0.667 vs. 0.783 in children, p = 0.0015). The HRQoL of carers was also affected (0.742 and 0.765 for carers of adults and children with CF, respectively). Conclusions: Our study highlights the burden of CF in terms of costs and decreased HRQoL for both patients and carers. © 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Keywords: Cystic fibrosis; Health-related quality of life; Costs and cost analysis; Economic burden

1. Background The prevalence of cystic fibrosis (CF) is estimated at 12.6/ 100,000 in Europe [1], and in France, the neonatal screening programme reports a prevalence of 1 in 4,754 births for the period 2002–2012 (roughly 21/100,000), with great variations among regions [2]. As is often the case with rare diseases – defined by the ⁎ Corresponding author at: URC Eco, AP–HP, Hôtel Dieu, 1 Place du Parvis Notre Dame, 75004 Paris, France. Tel.: +33 1 40 27 41 48; fax: + 33 1 40 27 41 41. E-mail address: [email protected] (K. Chevreul). 1 Members are listed in Appendix 1.

European Commission as disorders affecting fewer than 5 in 10,000 people [3] – there is no curative treatment for CF. Patient lifespans are greatly shortened, with a life expectancy of roughly 40 years [4], and studies have found that quality of life is also affected [5–7]. Despite its low prevalence, CF may have a considerable impact on healthcare system expenditures in terms of direct healthcare costs and lost productivity [8–10], as orphan drugs are often very expensive and CF's symptoms often appear at a young age. The project “Social economic burden and health-related quality of life (HRQoL) in patients with rare diseases in Europe” (BURQOL-RD) was carried out under the European Commission's 2nd Programme of Community Action in the

http://dx.doi.org/10.1016/j.jcf.2014.11.006 1569-1993/© 2014 European Cystic Fibrosis Society. Published by Elsevier B.V. All rights reserved. Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

2

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

Field of Public Health to explore these issues. Its goal was to develop and test a tool that could quantify the burden of different rare diseases. In order to do so, it examined the socio-economic burden and HRQoL of patients suffering from ten rare diseases – including CF – as well as their carers in eight European countries [11]. We present here the results of the BURQOL-RD project regarding the HRQoL of French patients with CF and their carers and the average cost of the disease, including direct healthcare costs, direct non-healthcare costs (both formal and informal) and indirect costs. 2. Methods 2.1. Patients A retrospective cross-sectional study was carried out between September 2012 and May 2013. Patients with CF were recruited on a strictly voluntary basis through the CF reference centre of Nantes–Roscoff, the French CF Society and the patient association Vaincre la Mucoviscidose. Patients who consented to participate were directed to the BURQOL-RD website and asked to complete an anonymous questionnaire regarding their socio-demographic characteristics (age, sex, education level, marital status, employment etc.), healthcare consumption in the previous six months (drugs, medical devices, consultations, tests etc.), impact of the disease on their employment and presence of any formal or informal carers. For children under 18, the questionnaire could be completed by a parent or other carers. Responses were considered valid if at a minimum the questions regarding resource utilisation to compute direct healthcare costs were completed. The patient's principal carer was asked to respond to a separate questionnaire detailing their characteristics and time spent caring for the person with CF. 2.2. Costs Costs were estimated from a societal perspective using a bottom-up approach and 2012 prices as a reference. The questionnaire included detailed information on the resources used by each patient within the previous six months (drug consumption, hospitalisations, consultations, hours of care needed by the patient etc.). Each resource was then valued with its unit cost (see below), and total costs were computed by multiplying each resource's cost by the number of units consumed. An annual average cost per patient was then calculated. It was only assessed for this one year, as it was not the goal of the study to extrapolate costs to a longer-term time horizon.

extra-billing beyond the statutory tariffs by certain doctors. Regarding hospitalisation costs, a daily tariff was derived from the National Hospital Cost Study using weighted disease-related groups [16]. Finally distances for ground transport costs were approximated with the average distance for access to healthcare previously estimated in France [17], with variable costs per kilometre depending on the type of transport. Unit costs are presented in Appendix 2. 2.2.2. Direct non-healthcare formal costs Direct non-healthcare formal costs included professional care (either at home by a professional carer or in institutions run by social services) and non-healthcare transport. Transport costs were calculated in a manner similar to healthcare transport costs (described above). Home-based care costs were valued using the French mean wage of a home carer, while average daily tariffs were used for services provided in institutions. 2.2.3. Direct non-healthcare informal costs Informal carers were defined as people providing unpaid help to a family member or friend requiring care. We used the replacement cost method to value the hours dedicated to providing care, which assumes that a professional carer would have to be hired if the informal carer did not provide these services [18]. The amount of time informal carers spent helping the person with CF was recorded in the questionnaire, and we used the average hourly wage of a health aide to calculate the cost. 2.2.4. Indirect costs We calculated the indirect costs for productivity loss and early retirement as a result of the disease. We took a human capital approach (HCA) to convert the data regarding time away from work into monetary units [19]. This method uses the average earnings of a worker as a proxy for labour productivity losses, and we based our estimations on the average annual French gross wages by gender provided by the National Institute of Statistics and Economics Studies (INSEE) [20]. While the friction cost approach (FCA) [21] which takes into account the compensation of a short-term absence by colleagues or the patient himself on his return or of a long-term absence by the hiring of a new employee could also have been used, HCA was preferred because it is more widely used in this context, and updated and country specific data are lacking for FCA. Finally, patients working in ‘sheltered workshops’ are paid on average 55% of the legal minimum wage, and their productivity loss cost was calculated using the French mean gross salary minus the wage received. 2.3. Health-related quality of life

2.2.1. Direct healthcare costs Direct healthcare costs consisted of drugs, medical tests, consultations, hospitalisations, medical devices and healthcare transport. The unit cost for drugs, medical devices and medical tests was taken from the Statutory Health Insurance (SHI) tariffs [12–15]. The cost of a consultation was calculated for each specialty, using an average consultation cost which accounted for

The HRQoL of patients with CF and their carers was assessed using the EQ-5D-5L questionnaire [22] along with a visual analogue scale (VAS) to rate their perceived global health on that day (between 0 and 100). The EQ-5D is a generic tool intended for self-reporting adults which assesses five dimensions of health (mobility, self-care, everyday activities,

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

pain/discomfort and anxiety/depression) with a five-level scale. The health states given by each combination of answers are associated with a utility generally between 0 and 1, although it may be negative for health states deemed worse than death. No value set yet exists to transform EQ-5D-5L health states into utilities, therefore a mapping from the French EQ-5D-3L value set was performed to obtain utilities. In the case of children, this questionnaire has been used in children in different studies [23–26] and studies have also shown that parent-proxy ratings are both feasible and valid, at least in children with attention deficit-hyperactivity disorder [27,28]. In our study, it was only answered for patients over 5 years old, either by the patient if they were old enough (10 out of 92) or by their legal representative. Patients were also asked to complete the Barthel index [29], which measures a person's ability to perform activities of daily living and mobility, with lower scores indicating a higher degree of dependence. Finally, carers also completed the Zarit burden interview (22-item version), which measures their subjective burden, with higher scores revealing a heavier burden [30]. 2.4. Statistical analysis Mean and standard deviation (SD) were calculated for continuous variables, and percentages were computed for categorical variables. Median and interquartile range (IQR) were also calculated for the costs and HRQoL variables. Differences in costs and HRQoL were assessed overall and for different patient characteristics (age, gender, disease duration and level of dependence). Wilcoxon tests and Kruskal–Wallis tests were conducted at a 0.05 significance level to compare the results in the different sub-groups. Analyses were performed using Excel version 2007 and SAS version 9.3 (SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Patients Four hundred fifty-nine people with CF started the questionnaire, but only 240 (52.3%) were included in the analysis (82 adults and 158 children). A flowchart is presented in Appendix 3. There was no difference between the overall population of respondents and those who were included in the analysis in terms of age, gender distribution, age at diagnosis, the proportion who required a carer, and marital and employment status for adults and type of schooling for children (Appendix 4). Moreover, when compared to the 6145 patients of the French Cystic Fibrosis Data Registry (a registry certified by the national committee of rare disease registries whose population is composed of people with CF followed in the care centres participating in the registry) [31], we found that while our population was younger and had a lower age at diagnosis, its gender distribution and marital status were similar, as were the median number of hospitalisations and the prescribed treatments (Appendix 5).

3

Table 1 Characteristics of the study population.

Mean age, years (SD) Mean age at diagnosis, years (SD) Female, n (%) Marital status Single, n (%) Married/in long-term relationship, n (%) Employment status Employed, n (%) Employed but currently on sick leave, n (%) Unemployed and looking for work, n (%) Unemployed, on disability pension, n (%) Homemaker n (%) Others, n (%) Schooling Regular school, no accommodation, n (%) Regular school, personalised support, n (%) Specialised school, n (%) Day-care, n (%) Others, n Reported sick days due to CF, n (%) Need for a carer, n (%) Use of a carer, hours per week a (SD) Professional carer Informal carer a

Adults n = 82

Children n = 158

28.6 (8.1) 4.4 (6.5) 47 (57.3)

8.5 (4.8) 0.8 (1.8) 67 (42.4)

42 (51.2) 38 (46.3)

NA NA

46 (56.1) 7 (8.5) 10 (12.2) 4 (4.9) 5 (6.1) 10 (12.2)

NA NA NA NA NA NA

NA NA NA NA NA 27 (32.9) 8 (9.8)

106 (67.1) 27 (17.1) 1 (0.6) 1 (0.6) 23 (14.6) NA 54 (34.2)

0 (0) 39.3 (24.8)

6.4 (5.4) 44.2 (33.1)

For patients requiring care hours.

The mean age of the population was 15.4 years (SD = 11.3), 52.5% of respondents were male and the average disease duration was 13.4 years (SD = 10.1). The adult population had a mean age of 28.6 years and was predominantly female (Table 1). The majority (64.6%) was employed, with over half (58.5%) reporting disease-related consequences for their work including a third who required sick leave averaging 63.5 days. Fewer than 10% required a carer, and in all cases the principal carer was an informal carer. The paediatric population had a mean age of 8.6 and was predominantly male (Table 1). Two-thirds (67.1%) went to school with no special accommodations, while 17.1% required personalised support at school. Over one-third had a carer at home, who most often was an informal carer (88.9%). 3.2. Costs The total average annual cost of CF was €29,746 per patient (Table 2). Direct healthcare costs accounted for over half, with drugs and hospitalisations representing the largest shares. Direct non-healthcare formal costs were estimated at €4,512, with social services accounting for 92.6% of that cost. Direct non-healthcare informal costs were similarly high (€4,827). Excluding children, indirect costs were high (€10,408 per adult patient). Total costs were significantly higher in adults than in children: €40,595 vs. €24,116, p = 0.0017 (Table 3). This was also true for direct healthcare costs (€23,968 vs. €13,157, p = 0.0040), with adults having higher drug, test and hospitalisation costs,

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

4

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

Table 2 Mean and median costs (in 2012 Euros) and health-related quality of life of patients with CF and their carers.

Direct HC costs Drugs Medical tests Consultations Hospitalisations Medical devices HC transport Direct non-HC costs (formal) Professional carer Social services Non-HC transport Direct non-HC costs (informal) Main informal carer Other informal carers Total direct costs Productivity loss (patients) b Early retirement (patients) b Total indirect costs b Total annual costs Utility (patients) VAS (patients) a b

Mean (SD)

Median (IQR a)

16,851 (21,088) 6,589 (7,009) 497 (496) 2,565 (1,736) 6,172 (18,303) 998 (1,872) 30 (253) 4,512 (13,303) 87 (759) 4,177 (13,336) 249 (528) 4,827 (15,165) 3,794 (10,863) 1,033 (5,855) 26,190 (31,712) 9,194 (13,451) 1,213 (6,265) 10,408 (14,057) 29,746 (36,119) 0.730 (0.231) 71.90 (18.53)

9,895 (17,280) 4,259 (14,119) 352 (393) 2,716 (2,078) 974 (4,383) 238 (620) 0 (0) 120 (924) 0 (0) 0 (0) 72 (156) 0 (0) 0 (0) 0 (0) 14,923 (23,682) 1,008 (14,926) 0 (0) 2,025 (14,926) 17,434 (26,712) 0.798 (0.317) 75.00 (30.00)

IQR: interquartile range. Adults only.

while children had higher consultation costs. For direct non-healthcare formal costs no difference between adults and children was found. However, direct non-healthcare informal costs were significantly higher in children (€6,704 vs. €1,211,

Table 3 Mean costs (in 2012 Euros) and health-related quality of life of patients and their carers by patient age group.

Direct HC costs Drugs Medical tests Consultations Hospitalisations Medical devices HC transport Direct non-HC costs (formal) Professional carer Social services Non-HC transport Direct non-HC costs (informal) Main informal carer Other informal carers Total direct costs Productivity loss (patients) Early retirement (patients) Total indirect costs Total annual costs Utility (patients) VAS (patients)

Adults mean (SD)

Children mean (SD)

p

23,968 (29,707) 8,277 (7,684) 754 (648) 2,344 (2,004) 11,445 (27,066) 1,127 (2,111) 21 (118) 5,008 (15,011) 0 (0) 4,806 (15,033) 202 (378) 1,211 (6,425) 998 (5,088) 213 (1,680) 30,187 (37,518) 9,194 (13,451) 1,213 (6,265) 10,408 (14,057) 40,595 (46,187) 0.667 (0.251) 65.56 (19.98)

13,157 (13,471) 5,712 (6,485) 364 (326) 2,679 (1,575) 3,435 (10,485) 930 (1,739) 35 (301) 4,255 (12,367) 132 (933) 3,851 (12,403) 273 (591) 6,704 (17,806) 5,245 (12,652) 1,459 (7,085) 24,116 (28,142) NA NA NA 24,116 (28,143) 0.783 (0.200) 77.12 (15.49)

0.0040 0.0191 b 0.0001 0.0069 0.0026 0.5613 0.8292 0.4289 0.1043 0.3183 0.3083 0.0005 0.0005 0.0404 0.3713 – – – 0.0017 0.0015 b 0.0001

p = 0.0005). The cost repartition among the different cost items for the two populations is presented in Appendix 6. We found no cost difference between men and women with CF. Disease duration was associated with higher total annual costs as well as higher direct healthcare costs, particularly drug, medical testing, hospitalisation and medical device costs (Table 4). While there was no difference in formal direct non-healthcare costs, informal direct non-healthcare costs decreased over time due to the link between disease duration and age, with children requiring more informal care than adults. A higher level of dependence as revealed by a lower Barthel index score was associated with higher informal costs as well as higher costs for the main professional carer but had no effect demonstrated on direct healthcare costs, indirect costs or total costs (Table 5). 3.3. Health-related quality of life HRQoL data was available for 166 patients (75 patients and 91 children) and 40 carers (of 6 adults and 34 children). Out of the 92 children, 10 answered the questionnaire themselves and a parent answered for the others. The average utility for a patient with CF in our study was 0.730, and the average VAS was 71.90 (Table 2). Utility was lower in adults than in children (0.667 vs. 0.783, p = 0.0015) (Table 3). The average Barthel index reflected a mild dependence among patients (97.55, SD = 4.32) and was similar for both adults and children: 98.08 (SD = 2.84) and 97.11 (SD = 5.19), respectively, p = 0.52. The HRQoL of carers was similarly affected, with a utility of 0.761 and a VAS of 78.95, and there was no difference if they looked after an adult or a child (utility of 0.742 and 0.765 respectively, p = 0.7904, and VAS of 76.67 vs. 79.35, p = 0.97). The Zarit burden for carers was mild to moderate (27.11, SD = 16.21), and unlike the utility it was significantly higher for individuals caring for a child: 29.24 (SD = 15.93) vs. 13.67 (SD = 11.40), p = 0.03. We found no gender-based difference in HRQoL for patients or carers. However, patients with a longer disease duration had a significantly worse HRQoL (lower utility and VAS, Table 4), and the same was true for carers of patients with a longer disease duration (lower utility only: 0.738 for a disease duration of 0–9 years, 0.819 for 10–19 years, 0.839 for 20– 29 years and 0.356 after 30 years, p = 0.0498). Higher levels of dependence were linked to a lower HRQoL as well, although this was only seen in patients' VAS and not in their utility score (Table 5). This did not hold true for carers, who also did not have a significantly higher Zarit score when patients had a higher level of dependence (26.14 vs. 21.2, p = 0.26). As expected, we found a strong correlation between patient utility and patient VAS (r = 0.60, p b 0.0001). There was also a strong correlation between patient utility and both carer utility (r = 0.56, p = 0.003) and carer VAS (r = 0.69, p b 0.0001). The correlation between patient utility and Barthel index was weak but significant (r = 0.25, p = 0.0015) and there was no correlation between patient utility and Zarit burden. Regarding

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

5

Table 4 Mean costs (in 2012 Euros) and health-related quality of life of patients and their carers by disease duration.

Direct healthcare costs Drugs Medical tests Consultations Hospitalisations Medical devices Healthcare transport Direct non-healthcare costs (formal) Professional carer Social services Non-healthcare transport Direct non-healthcare costs (informal) Main informal carer Other informal carers Total direct costs Productivity loss (patients) a Early retirement (patients) a Total indirect costs a Total annual costs Utility (patients) VAS (patients) a

Disease duration 0–9 years Mean (SD) n = 108

Disease duration 10–19 years Mean (SD) n = 65

Disease duration 20–29 years Mean (SD) n = 38

Disease duration ≥ 30 years Mean (SD) n = 19

p

11,465 (9,093) 5,442 (6,349) 366 (255) 2,735 (1,603) 2,133 (3,956) 775 (1,473) 14 (57) 3,914 (12,604) 193 (1,125) 3,506 (12,634) 215 (346) 7,361 (20,111) 5,418 (13,416) 1,943 (8,473) 22,740 (26,224) 7,463 (8,617) 0 (0) 7,463 (8,617) 23,016 (26,313) 0.783 (0.166) 74.50 (17.28)

19,456 (19,516) 8,079 (7,224) 473 (516) 2,693 (1,616) 6,904 (17,065) 1,241 (2,089) 65 (464) 6,058 (14,037) 0 (0) 5,684 (14,105) 374 (815) 4,859 (10,873) 4,505 (10,428) 354 (1,389) 30,373 (32,692) 5,917 (8,412) 0 (0) 5,917 (8,412) 31,012 (33,728) 0.712 (0.256) 74.61 (16.74)

27,594 (36,337) 8,656 (7,573) 824 (678) 2,626 (2,128) 14,214 (35,707) 1,243 (2,428) 31 (169) 4,705 (14,292) 0 (0) 4,574 (14,294) 131 (246) 48 (296) 48 (296) 0 (0) 32,347 (39,161) 9,100 (13,865) 873 (5,381) 9,973 (14,313) 42,320 (49,479) 0.702 (0.202) 65.77 (18.13)

23,968 (27,162) 7,103 (7,899) 884 (641) 2,025 (1,792) 12,662 (19,919) 1,279 (2,169) 15 (46) 4,523 (16,230) 0 (0) 4,197 (16,265) 326 (606) 2,415 (10,527) 1,619 (7,058) 796 (3,469) 30,906 (42,214) 13,518 (17,249) 3,491 (10,458) 17,009 (17,529) 47,915 (52,970) 0.589 (0.303) 58.82 (20.96)

0.0008 0.0118 b 0.0001 0.0736 0.0024 0.0150 0.9649 0.7621 0.1245 0.4813 0.3377 0.0165 0.0161 0.0645 0.1904 0.4738 0.3446 0.1254 0.0084 0.0498 0.0037

Adults only.

the correlation between patient utility and costs, it was consistently significant and negative but weak (r between − 0.17 for direct formal non-healthcare costs and − 0.32 for direct healthcare costs, with a coefficient of − 0.40 for total costs, p b 0.0001).

4. Discussion We found that CF has a significant impact on the quality of life of patients regardless of age and is associated with an average total annual cost of €29,746.

Table 5 Mean costs (in 2012 Euros) and health-related quality of life of patients and their carers by patient dependency level (Barthel index score). Mild dependence Mean (SD) n = 150 Direct healthcare costs Drugs Medical tests Consultations Hospitalisations Medical devices Healthcare transport Direct non-healthcare costs (formal) Professional carer Social services Non-healthcare transport Direct non-healthcare costs (informal) Main informal carer Other informal carers Total direct costs Productivity loss (patients) c Early retirement (patients) c Total indirect costs c Total annual costs Utility (patients) VAS (patients) a b c

20,260 (24,870) 7,765 (7,322) 555 (530) 2,529 (1,735) 8,291 (22,392) 1,082 (1,920) 39 (317) 4,531 (13,652) 10 (127) 4,226 (13,702) 295 (619) 2,564 (8,332) 2,270 (7,496) 293 (1,581) 27,355 (33,030) 8,966 (13,955) 1,421 (6,766) 10,387 (14,652) 32,202 (38,682) 0.737 (0.228) 72.10 (18.82)

a

Moderate dependence Mean (SD) n = 13 16,034 (10,583) 7,599 (7,474) 681 (717) 3,291 (2,595) 2,997 (4,436) 1,438 (2,182) 27 (99) 12,730 (21,104) 280 (1,010) 12,154 (21,272) 296 (491) 10,950 (15,271) 10,582 (15,163) 368 (804) 39,714 (37,111) 10,735 (9,371) 0 (0) 10,735 (9,371) 42,191 (38,099) 0.605 (0.235) 67.08 (13.39)

b

p

0.9390 0.8734 0.5460 0.3020 0.5636 0.3044 0.9896 0.0921 0.0267 0.2548 0.1551 0.0008 0.0007 0.0213 0.2907 0.5358 0.7161 0.6947 0.2963 0.1346 0.0307

Mild dependence: Barthel index N 91. Moderate dependence: Barthel index ≤ 90. Adults only.

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

6

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

Ours is the first study to simultaneously assess the burden of CF in terms of both HRQoL and complete societal costs in a general population of patients. Previous studies focused largely on direct healthcare costs or populations from randomized controlled trials or large database claims in calculating their cost estimates. Our bottom-up approach allowed us to include costs such as those related to informal care and transport that cannot be captured by a top-down approach, which only allocates a portion of a known total expenditure to the disease [32]. Two other French studies have studied the direct healthcare cost of CF (excluding transport costs) in small groups of patients treated in reference centres in the early 2000s. The first [33] found a total annual direct cost of €16,189 per patient (€15,665 in 2012€ using the CSMB price index which reflects the price evolution of healthcare consumption [34]), which is similar to our study (€16,851), as was the second study [35] in the first year of their analysis (mean annual cost of €16,474 per patient in 2000, or €15,940 in 2012€). Costs were higher in 2003 (€22,725, or €21,989 in 2012€), which was partially attributed by the authors to an increase in the use of expensive medications (dornase alfa and tobramycin). Since 2003, the price of these two drugs has decreased by 10–15%, which may explain our lower costs since nearly two-thirds of our patients used dornase alpha and one-quarter used tobramycin, and very few used newer, more expensive drugs (for example only one was prescribed inhaled aztreonam). Despite those similarities, our findings differ from those of the French statutory health insurance (SHI). Indeed, the SHI reported a mean direct healthcare cost of €22,454 per patient with CF in 2009 for the 5,536 patients who had declared their CF under the SHI chronic disease scheme that provides 100% coverage [36,37]. This would correspond to €22,319 in 2012 €, while our estimate of direct healthcare costs was only €16,851. This can be due to either of two limits of our study, the first being the retrospective collection of healthcare resource consumption which may be associated with a risk of recall bias, especially in the case of direct healthcare resources. Indeed, patients may not remember precisely how many times they used a resource over a six-month period, which in turn may lead to approximations or a misallocation of costs [32] — here an underestimation of their healthcare consumption by our sample. This difference may also be a consequence of fewer severe cases in our sample than the general CF population in France, as it has been shown that CF costs increase with disease severity [9,38]. Unfortunately, disease severity (lung function, nutritional status…) could not be assessed with our questionnaire. We made attempts to objectify the generalizability of our sample by comparing its characteristics to those of patients present in the French Cystic Fibrosis Data Registry [31] and found that while our population was a little younger – as was the mean age at diagnosis –, their other characteristics were similar. However, we again lacked an appropriate proxy for disease severity, something future studies should rectify. International studies have also studied the costs of CF, although most focused only on direct healthcare costs and very few included informal costs and indirect costs. Direct

healthcare costs appear to vary greatly between countries [8–10,40], ranging from US$15,571 in Australia [10] in 2003– 2005 (€14,134 in 2012€ after applying OECD's prices and purchasing power parities (PPP) conversion rates [41]) to €38,869 in Germany [9] in 2004 (€45,293 in 2012 PPP). It is difficult to compare those results with our own as they cover different populations (patients in a clinical trial, followed at one hospital or covered by private insurances) and include different cost items in their cost calculations. If the Australian study found an allocation of expenditure consistent with our own, with the majority of the cost attributable to hospitalisations and drugs, American studies [8,40] had a much larger part attributed to drug costs (57% and 85% of direct costs vs. 39% in our study). This is not surprising, as drugs are notoriously more expensive in the US than in Europe. The German study which reported the highest cost also found that drug costs accounted for 81% of total costs [9], but in this case the difference may be explained by patient characteristics. Indeed, the authors reported that direct medical costs ranged from €3,876 (€4,517 in 2012 PPP) for patients with a mild form of the disease (no Pseudomonas aeruginosa infection) to €88,096 (€102,656 in 2012 PPP) for patients with severe disease manifestations (pulmonary hypertension and global respiratory insufficiency). Our population may therefore include patients with less severe CF, which would result in lower costs. Only two of the aforementioned studies also looked at the indirect costs of CF, and both found much lower costs than our estimate of €10,408 (25.6% of total costs in adults): €2,491 (€2,903 in 2012 PPP, 6% of total costs) in Germany [9] and $1,180 (€2,281 in 2012 PPP, 4% of total costs) in the US [8]. However, neither included costs related to early retirement. To the best of our knowledge, no other study has examined the cost of informal care in CF, which we estimated at €4,827 (16.2% of the total cost). Given the high cost, informal care should be included in future studies examining the economic burden of CF. Finally, the use of the EQ-5D in our study is both a weakness and a strength. Most studies examining HRQoL in CF have used disease-specific instruments such as the Cystic Fibrosis Questionnaire [6,7], while only one other study has reported utilities derived from EQ-5D [5]. The advantage of the EQ-5D is that it is a generic tool which allows researchers to elicit country-specific utilities which can then be used in costeffectiveness analysis and decision-making, but the questionnaire was developed for an adult population and is meant to be self-reporting. However, other studies have used the EQ-5D in children [23–26] and parent-proxy ratings have also been shown to be both feasible and valid [27,28]. We have found that patients' utility (0.667 for adults and 0.783 for children) was lower than the reference value of 0.83 (± 0.15) for the French-speaking general adult population [39]. It decreased significantly with disease duration and with a more severe dependence score. Our results appear to be consistent with the other study using EQ-5D, which found a mean utility of 0.85 for adults (over 16 years old) without pulmonary exacerbation, 0.79 for adults with mild exacerbations and 0.60 for adults with severe exacerbations. There was a strong correlation between

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

patient utility and patient VAS, carer utility and carer VAS. Researchers wanting to use those scores for modelling purposes should therefore be aware of this and include only the most relevant variable in their models. On the other hand, the correlation between patient utility and Barthel index was weak and there was no correlation between patient utility and Zarit burden; those two indexes may therefore be included along with patient utility among potential explanatory variables should the need arise.

[8]

[9]

[10]

5. Conclusion [11]

Our study highlights the burden of cystic fibrosis in terms of its impact on quality of life for both patients and carers. The disease is also associated with a considerable economic cost, which is dominated by direct healthcare costs, although direct non-healthcare costs represent a major share of total expenditure as well, as do indirect costs in adult patients. Future surveys should, if possible, assess costs in a prospective manner and evaluate patients' severity so as to allow a more precise analysis. Disclosure The authors declare no conflicts of interest

[12]

[13]

[14] [15]

[16]

Acknowledgements [17]

This study was supported by the Social Economic Burden and Health-Related Quality of Life in Patients with Rare Diseases in Europe Project, which received funding from the European Union in the framework of the Health Programme [grant A101205]. The Executive Agency of the European Union is not responsible for any use that may be made of the information contained here.

[18]

[19] [20]

Appendix A. Supplementary data

[21]

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jcf.2014.11.006.

[22]

References

[23]

[1] Prévalence des maladies rares, données bibliographiques. Classement par ordre alphabétique des maladies ou groupes de maladies; 2013. [2] Association française pour le dépistage et la prévention des handicaps de l'enfant. Bilan d'activité 2012; 2013. [3] European Commission — rare diseases. [Internet]. [cited 2013 Oct 22]. Available from: http://ec.europa.eu/health/rare_diseases/policy/index_en.htm. [4] Orphanet: cystic fibrosis. [Internet]. [cited 2013 Oct 22]. Available from: http://www.orpha.net/. [5] Bradley JM, Blume SW, Balp M-M, Honeybourne D, Elborn JS. Quality of life and healthcare utilisation in cystic fibrosis: a multicentre study. Eur Respir J Mar 2013;41(3):571–7. [6] Abbott J, Hurley MA, Morton AM, Conway SP. Longitudinal association between lung function and health-related quality of life in cystic fibrosis. Thorax Feb 2013;68(2):149–54. [7] Hegarty M, Macdonald J, Watter P, Wilson C. Quality of life in young people with cystic fibrosis: effects of hospitalization, age and gender, and

[24]

[25]

[26]

[27]

7

differences in parent/child perceptions. Child Care Health Dev Jul 2009; 35(4):462–8. Dewitt EM, Grussemeyer CA, Friedman JY, Dinan MA, Lin L, Schulman KA, et al. Resource use, costs, and utility estimates for patients with cystic fibrosis with mild impairment in lung function: analysis of data collected alongside a 48-week multicenter clinical trial. Value Health J Int Soc Pharmacoecon Outcome Res Apr 2012;15(2):277–83. Heimeshoff M, Hollmeyer H, Schreyögg J, Tiemann O, Staab D. Cost of illness of cystic fibrosis in Germany: results from a large cystic fibrosis centre. Pharmacoeconomics Sep 1 2012;30(9):763–77. Van Gool K, Norman R, Delatycki MB, Hall J, Massie J. Understanding the costs of care for cystic fibrosis: an analysis by age and health state. Value Health J Int Soc Pharmacoecon Outcome Res Apr 2013;16(2): 345–55. Linertová R, Serrano-Aguilar P, Posada-de-la-Paz M, Hens-Pérez M, Kanavos P, Taruscio D, et al. Delphi approach to select rare diseases for a European representative survey. The BURQOL-RD study. Health Policy Amst Neth Nov 2012;108(1):19–26. Assurance Maladie. Base des Médicaments et Informations Tarifaires. [Internet]. [cited 2014 Jul 31]. Available from: http://www.codage.ext. cnamts.fr/codif/bdm_it/index_presentation.php?p_site=AMELI. Liste des Produits et Prestations. [Internet]. [cited 2013 Oct 22]. Available from: http://www.codage.ext.cnamts.fr/codif/tips/index_presentation.php? p_site=AMELI. Classification Commune des Actes Médicaux. [Internet]. [cited 2013 Oct 22]. Available from: http://www.ameli.fr/accueil-de-la-ccam/index.php. Table Nationale de Codage de Biologie. [Internet]. [cited 2013 Oct 22]. Available from: http://www.codage.ext.cnamts.fr/codif/nabm/index_presentation.php?p_site=AMELI. Agence Technique de l'Information sur l'Hospitalisation. Echelle nationale coûts par GHM. [Internet]. [cited 2013 Oct 23]. Available from: http://www. atih.sante.fr/?id=000370000AFF. Coldefy M, Com-Ruelle L, Lucas-Gabrielli V, Marcoux L. Les distances d'accès aux soins en France métropolitaine au 1er janvier 2007. Volume 1. Rapport Irdes. N°550 ed; Juin 2011. Van den Berg B, Brouwer WBF, Koopmanschap MA. Economic valuation of informal care. An overview of methods and applications. Eur J Health Econ HEPAC Health Econ Prev Care Feb 2004;5(1): 36–45. Hodgson T, Meiners M. Cost-of-illness methodology: a guide to assessment practices and procedures. Milbank Mem Fund Q 1982:429–91. Institut national de la statistique et des études économiques. [Internet]. [cited 2013 Oct 22]. Available from: http://www.insee.fr/fr/. Koopmanschap MA, Rutten FF, van Ineveld BM, van Roijen L. The friction cost method for measuring indirect costs of disease. J Health Econ Jun 1995;14(2):171–89. Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, et al. Development and preliminary testing of the new five-level version of EQ5D (EQ-5D-5L). Qual Life Res Int J Qual Life Asp Treat Care Rehabil Dec 2011;20(10):1727–36. Stolk EA, Busschbach JJ, Vogels T. Performance of the EuroQol in children with imperforate anus. Qual Life Res Int J Qual Life Asp Treat Care Rehabil Feb 2000;9(1):29–38. Polinder S, Meerding WJ, Toet H, Mulder S, Essink-Bot M-L, van Beeck EF. Prevalence and prognostic factors of disability after childhood injury. Pediatrics Dec 2005;116(6):e810–7. Willems DCM, Joore MA, Hendriks JJE, Wouters EFM, Severens JL. Cost-effectiveness of a nurse-led telemonitoring intervention based on peak expiratory flow measurements in asthmatics: results of a randomised controlled trial. Cost Eff Resour Alloc CE 2007;5:10. Willems DCM, Joore MA, Nieman FHM, Severens JL, Wouters EFM, Hendriks JJE. Using EQ-5D in children with asthma, rheumatic disorders, diabetes, and speech/language and/or hearing disorders. Int J Technol Assess Health Care Jul 2009;25(3):391–9. Secnik K, Matza LS, Cottrell S, Edgell E, Tilden D, Mannix S. Health state utilities for childhood attention-deficit/hyperactivity disorder based on parent preferences in the United Kingdom. Med Decis Making Int J Soc Med Decis Making Feb 2005;25(1):56–70.

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006

8

K. Chevreul et al. / Journal of Cystic Fibrosis xx (2015) xxx–xxx

[28] Matza LS, Secnik K, Mannix S, Sallee FR. Parent-proxy EQ-5D ratings of children with attention-deficit hyperactivity disorder in the US and the UK. Pharmacoeconomics 2005;23(8):777–90. [29] Mahoney F, Barthel D. Functional evaluation: the Barthel index. Md State Med J Feb 1965;14:61–5. [30] Hérbert R, Bravo G, Préville M. Reliability, validity, and reference values of the Zarit burden interview for assessing informal caregivers of community-dwelling older persons with dementia. Can J Aging 2000: 494–507 [19th ed.]. [31] Vaincre la Mucoviscidose. Ined. French cystic fibrosis registry — annual data report 2012. Paris: Vaincre la Mucoviscidose; Mar 2014. [32] Tarricone R. Cost-of-illness analysis. What room in health economics? Health Policy Amst Neth Jun 2006;77(1):51–63. [33] Horvais V, Touzet S, François S, Bourdy S, Bellon G, Colin C, et al. Cost of home and hospital care for patients with cystic fibrosis followed up in two reference medical centers in France. Int J Technol Assess Health Care 2006;22(4):525–31. [34] Le Garrec M, Bouvet M. Comptes nationaux de la santé 2012. Direction de la recherche, des études, de l'évaluation et des statistiques; 2013. [35] Huot L, Durieu I, Bourdy S, Ganne C, Bellon G, Colin C, et al. Evolution of costs of care for cystic fibrosis patients after clinical guidelines

[36]

[37]

[38]

[39]

[40]

[41]

implementation in a French network. J Cyst Fibros Off J Eur Cyst Fibros Soc Sep 2008;7(5):403–8. Assurance Maladie. Prévalence des ALD au 31/12/2009. [Internet]. [cited 2013 Oct 23]. Available from: http://www.ameli.fr/l-assurance-maladie/ statistiques-et-publications/donnees-statistiques/affection-de-longue-dureeald/prevalence/frequence-des-ald-au-31-12-2009.php. Assurance Maladie. Coût des ALD en 2009 dans la population du régime général. [Internet]. [cited 2013 Oct 23]. Available from: http://www.ameli. fr/l-assurance-maladie/statistiques-et-publications/donnees-statistiques/ affection-de-longue-duree-ald/cout/cout-des-ald-en-2009.php. Eidt-Koch D, Wagner TOF, Mittendorf T, Graf von der Schulenburg J-M. Outpatient medication costs of patients with cystic fibrosis in Germany. Appl Health Econ Health Policy 2010;8(2):111–8. Perneger TV, Combescure C, Courvoisier DS. General population reference values for the French version of the EuroQol EQ-5D health utility instrument. Value Health J Int Soc Pharmacoecon Outcome Res Aug 2010;13(5):631–5. Briesacher BA, Quittner AL, Fouayzi H, Zhang J, Swensen A. Nationwide trends in the medical care costs of privately insured patients with cystic fibrosis (CF), 2001–2007. Pediatr Pulmonol Aug 2011;46(8):770–6. CCEMG - EPPI-Centre Cost, Converter v. 1.2 [Internet]. [cited 2013 Oct 31].

Please cite this article as: Chevreul K, et al, Costs and health-related quality of life of patients with cystic fibrosis and their carers in France, J Cyst Fibros (2015), http:// dx.doi.org/10.1016/j.jcf.2014.11.006