Costs and quality of life for psoriatic patients at different degrees of severity in southern Sweden – a cross-sectional study

ARTICLE

ejd.2011.1635

Auteur(s) : Ola Ghatnekar¹ og@ihe.se, Anders Ljungberg², Lars-Erik Wirestrand³, Åke Svensson⁴

¹ The Swedish Institute for Health Economics, Box 2127, 220 02 Lund, Sweden

² Abbott Scandinavia AB, Solna, Sweden

³ Department of Dermatology, Kristianstad Hospital, Sweden

⁴ Department of Dermatology, University of Lund, Malmö, Sweden

Reprints: O. Ghatnekar

About 2-3% of the Swedish population are estimated to suffer from psoriasis, which translates into 200 to 300 thousand persons [1]. The management of psoriasis depends on the severity [2] but a clinical definition of mild, moderate and severe psoriasis may not always be clear-cut as it also depends on patient-reported severity [3]. In Sweden, topical treatment with D-vitamin derivate, steroids and emollients as well as ultraviolet light therapy is common for mild to moderate psoriasis. For severe psoriasis, traditional systemic treatment consists of methotrexate, cyclosporine or acitretin, whereas biologic systemic treatment consists of adalimumab, etanercept, infliximab or ustekinumab. The management of patients with psoriasis is mostly performed in outpatient care, either by general physicians or by specialists in dermatology. Inpatient care due to psoriasis is rare.

Published information on the cost-of-illness for psoriasis including biological treatment is scarce. In Switzerland, the cost of out-of-pocket expenses and outpatient visits borne by third-party payers in the year 2005 amounted to 1,141€ to 7,957€ per patient and year, depending on severity [4]. In the Netherlands the annual cost per patient rose from 10,100€ to 17,700€ after the introduction of biological drugs [5]. Although not including biological treatments, the annual societal cost (direct and indirect costs) in Germany was estimated at between 2,866€ and 6,709€ per patient per year, for the years 2002 and 2003/04 [6, 7]. In Italy, the corresponding cost was 8,372€ in the year 2003/04 [8]. Apart from differences in costing perspectives, patient characteristics and health care structures, the variability in these estimates also depends on the treatment patterns and especially the introduction and use of biological drugs.

The purpose of this study was to estimate the cost of psoriasis care from a societal perspective and quality of life during one month in a defined Swedish patient population. We also wanted to assess the association between costs and quality of life, measured with the dermatology life quality index (DLQI) and the EuroQol EQ-5D index, for different degrees of severity, measured with the psoriasis area and severity index (PASI).

Methods

Patient enrolment

We performed a prevalence-based prospective recruitment of patients visiting the dermatology clinics at Malmö University Hospital (tertiary level) and Kristianstad Hospital (secondary level), both in the Southern health care region in Sweden (catchment population 1.7 million). From the 1^st week of September in 2009 patients were asked to participate in the study in conjunction with their ordinary visit to a dermatologist at the clinic. The enrolment period continued until 150 to 200 consecutive patients were recruited, of whom at least 20 patients should have severe psoriasis, defined as PASI>10 and/or be using systemic treatment. As the clinical severity at the point of recruitment could be lower than the underlying severity due to treatment effects, the treatment strategy was thought to work as a proxy for severity. Patients meeting the inclusion/exclusion criteria were scheduled for a follow-up visit with a nurse approximately 1 month later. These criteria were: age≥18 years, diagnosis of psoriasis (ICD-10: L40), written informed consent at recruitment, no problems understanding written Swedish, no severe psychiatric disorder, no alcohol or substance abuse. The study was granted ethics review board approval in Lund, Sweden (Dnr 2009/330).

Study activities

At recruitment the patients received questionnaires to be completed at home before the scheduled follow-up appointment 1 month later and reviewed together with the nurse. One questionnaire captured information on the patient's quality of life, level of education, smoking habits, level of workforce participation, sick leaves and reduced productivity at work due to psoriasis during the past month. Two quality of life instruments were used: one disease specific questionnaire, the Dermatology Life Quality Index (DLQI) ranging from 0 (not affected) to 30 (extremely affected), and one generic questionnaire allowing comparisons with other diseases/conditions, the EuroQol-5D (EQ-5D) ranging from 1 (perfect health) to 0 (dead) [9, 10]. The responses to the EQ-5D were converted to utility scores using the UK social tariff developed by Dolan [11].

At the follow-up visit, information from both the medical record and the patient was retrieved, including patient age, gender, type of psoriasis, body mass index (BMI), co-morbidities, current treatment, if the patient had visited any other medical professionals due to the psoriasis during the past month, mode and length of transport for psoriasis care, and how much time the patient spent every day on topical treatment application. The estimated number of outpatient visits during the last 12 months was also reported in order to provide an estimate for how far the results for the selected time period could be generalised in an annual perspective, as the intensity of disease and treatment activity may vary. Of course, this kind of question cannot match the validity of prospective data collection, but may however be indicative. In addition, the nurse documented the patient's PASI-score at this visit. In order to increase the conformity in PASI assessment both within and between the study sites, all staff involved had a half day course in PASI assessment prior to the enrolment. In addition, patients were regarded as having severe psoriasis if they received systemic drug treatment.

Costs

The one month resource information from the questionnaires was used to calculate the societal cost for psoriasis. Hence, we included direct medical costs (outpatient visits, hospitalisations, pharmaceuticals, phototherapy visits, intravenous administration, and naturopathic preparations), direct non-medical costs (transport to psoriasis related care) and indirect costs, i.e. time spent on topical application and transport (leisure time), sick leave (absenteeism), reduced productivity at work (presenteeism) and early retirement due to psoriasis. The value of the productivity loss was approximated with national averages for wages including pay-roll taxes matched by age and sex (human capital approach) [12]. The time spent on topical application and transport was evaluated according to the Swedish Transport Agency's calculation on leisure time [13]. All unit costs for the cost calculations were taken from published literature and Swedish official sources in the year 2009. Swedish Krona (SEK 1=EUR 0.094=USD 0.13), see table 1.

Table 1 Unit costs, Euro 2009.

*Prescribed psoriasis drugs/moisturisers: methotrexate, etanercept, adalimumab, infliximab, calcipotriol, betamethasone, diclofenac, ibuprofen, hydroxyzine, sulfasalazine, ketoprophen, risedronic acid, codeine/paracetamole, cyclosporine, acitretin, folic acid.

**Over the counter drugs: skin creams, analgesics, shampoo, D-vitamin, etc.

***Naturopathic preparations: different aloe vera formulations, shampoo, creams, etc.

Analysis

Costs were accumulated at the individual patient level and divided by the number of months (30.4 days) between the first and second visits in order to estimate the monthly cost per patient. Patients were analysed according to current treatment strategy as topical treatment only (TT), topical treatment plus ultraviolet light therapy (LT), traditional systemic treatment (TST) and biologic systemic treatment (BST). All statistical analyses were performed in SPSS for Windows, version 12.0.2 (SPSS Inc.). Non-parametric Mann-Whitney U-test with a two-sided significance level P<0.05 was used to determine significant differences between the means.

In order to investigate the determinants for quality of life and for costs, two linear regression analyses were performed with DLQI and costs as dependent variables, respectively. Cases with missing observations were excluded from the analysis. Independent variables in Table 2 (patient characteristics) were subject to a stepwise forward variable inclusion on the basis of coefficient significance level (P<0.05). All dichotomous variables were coded as 1 if the condition was present and 0 otherwise. A Kolmogorov-Smirnov test and P-P plot were used to test for log-normally distributions. When the dependent variable was logarithmically transformed, the deviation from unity of the independent exponentiated variable coefficient (Exp(β)) should be interpreted as the relative effect the independent variable has on the dependent variable. As any resource-related variables per definition were part of the dependent variable, they were excluded from the model.

Table 2 Patient characteristics in the year 2009, N=164

*Education level is measured on a 4-grade scale where 1 is the older elementary school level and 4 is university training.

** Other co-morbidity: Asthma (2), allergy (3), AV-block, breast cancer, slipped disc, EP (2), gout, hypertrophic cardiomyopathy, hypothyroidism (2), COPD (2), Non-Hodgkin's lymphoma, aortic coarctation surgery, thyroid cancer, polyneuropathy, Restless Legs Syndrome, spine surgery, ulcerative colitis (2).

Results

From the first week of September until the last week of December 2009, 184 patients met the inclusion criteria. A total of 164 patients completed the follow-up visit (89%). Drop-outs were more frequently recruited in Malmö and with more female sex than completers (P<0.05), all other patient characteristics were not statistically significant. Patient characteristics for completers are presented in table 2. The distribution of both sex and place of recruitment were equal, with a wide variation in age (19 to 86 years), and the vast majority (74%) had plaque psoriasis. According to the patients’ charts and self-reported co-morbidities, an average of 1.6 co-morbidities were registered and patients were slightly overweight on average (BMI=27). Almost half (53%) the study population experienced problems with joints, 31 patients were diagnosed with Psoriasis Arthritis (PsA) (table 2). Some differences in patient characteristics between sites were (Malmö vs Kristianstad): prevalence of PsA (25% vs 13%; P=0.054), age (49 vs 55; P<0.05), female sex (57% vs 42%; P<0.05), and lower PASI score (4.6 vs 6.5; P<0.05). The proportion of patients in the different treatment strategies was TT: 34%, LT: 24%, TST: 25%, and BST: 16%.

Based on the current treatment of patients, the mean PASI score was 5.66. A DLQI score of 7.65 indicated a moderate impairment in the patient's quality of life, which was in line with the EQ-5D index score of 0.71, indicating a loss of 0.29 years from a year with full health. A great variation in all three measures was seen, Table 2.

During the study period, 50% of the patients were working full- or part-time, 28% were retired due to age and 3% retired due to sickness; 8% were on temporary sick leave for any reason. The remaining patients (11%) were either on parental leave, students or unemployed. Among the 113 non-retired patients, six patients reported they had quit previous employment or had reduced working time due to psoriasis (mean 50% part-time). During the one month follow-up period, six patients had on average 6.5 days of temporary sick leave due to psoriasis and 58 patients reported reduced productivity while at work, with on average 14.7 days at a productivity level of 86.4%.

The mean total cost per patient month amounted to 994€, of which 22% was indirect costs, mainly from absenteeism (31€), presenteeism (78€) and early retirement (52€) (table 3). The main direct cost drivers were outpatient visits, light therapy and biological drugs. During the 1-month follow-up, patients attended on average 4.5 outpatient visits, including light therapy, for a total cost of 471€. However, a patient's estimated annual outpatient visits amounted to 117€, which was four times lower. One erythrodermic patient experienced a 10-day psoriasis-related inpatient stay after recruitment. Traditional systemic drugs (methothrexate, acitretin, and cyclosporine) were prescribed to 26% of the patients and accounted for 0.5% of total costs. Biological drugs (etanercept, adalimumab or infliximab) were prescribed to 16% of the patients and accounted for 20% of total costs. Seven patients did not use any moisturisers or topical treatment and one did not have any active treatment at all.

Table 3 Cost per patient-month for psoriasis (N=164; Euro 2009)

*Sick leave (absenteeism), reduced productivity at work (presenteeism) and early retirement.

Patients with TT only, i.e. moisturisers and corticosteroids with or without calcipotriol, cost 369€ per month (figure 1). Forty per cent of the societal cost for using moisturisers or topical treatment came from leisure time consumed when applying them to the skin. Total cost for all other treatment strategies were significantly more costly than TT and apart from the BST this holds true for both direct and indirect costs. Patients treated with light therapy were 11 years younger (P<0.05), had fewer comorbidities (1.05; P<0.05) and had on average 9.8 therapy sessions.

Both TST and BST patients had a higher prevalence of PsA than TT patients (38% and 37%, respectively vs 5%; P<0.00). Indirect costs accounted for 40% of total TST costs, mainly due to productivity losses (377€). Finally, drug acquisition and administration costs accounted for 78% of the direct cost for BST patients. Indirect costs were lower in this group compared to TST patients (p=0.04).

There was no statistically significant difference in total costs between men and women, although the use of calcipotriol/betamethasone was higher for men (43€ vs 28€; P=0.01), while indirect costs for topical treatment application were higher for women (54€ vs 35€; P=0.04). Although not statistically significant, 67% of the patients receiving biological drugs were male (P=0.135). In spite of the higher PASI-score for men (6.8 vs 4.5; P<0.00), they had a higher EQ-5D-weight than women (0.74 vs 0.68; P=0.01).

The main differences in total costs between Malmö (1,278€) and Kristianstad (754€) were attributable to the use of light therapy (146€; P<0.05) and biological treatment (172€; P<0.05), which together accounted to 318€. In addition, production losses were higher in Malmö than in Kristianstad, adding another 196€ in difference (P<0.05).

Regression analyses

The determinants of quality of life measured by DLQI indicated that a 1 unit increase in PASI from the mean would increase the DLQI score by 0.29, i.e. lower QoL, see Regression 1 in table 4. Likewise, a 0.1 increase in QALY-weight would reduce the DLQI score by 0.77, whereas each extra minute per day for topical treatment application would increase the DLQI score by 0.08. All three aspects of reduced labour productivity led to QoL impairments and a patient with, for example, both temporary sick leave and presenteeism increased the DLQI-score by 7.75 points.

Table 4 Regression results of DLQI and logarithmically transformed costs

Note: All variables in Table 2 were included as independent variables in the stepwise forward regression but only variables with coefficient significance P<0.05 are presented here;

Exp(b)-1 %=100*(exp(Coefficient)-1) percent change in costs.

In Regression 2 in table 4, we explored the association between costs and some independent factors, while controlling for biological treatment. As the assumption was that both higher PASI and DLQI scores would affect costs, we constructed a dichotomous variable for PASI ≥12 in order to attain a statistically significant parameter estimate (P<0.05). A unit increase in DLQI would increase costs by 4.5% whereas a PASI-score of 12 or above would increase costs by 39%. On the other hand, patients recruited in Kristianstad had 47% lower costs than patients treated in Malmö. Smokers and patients with elementary schooling also incurred lower costs than patients who never smoked or had higher schooling. However, as this form of education was dismantled during the 1970s, the variable also captured an age-effect apart from the education-effect itself.

Discussion

This study presents the results from a prevalence-based prospective study on psoriatic patients in southern Sweden. The monthly cost per patient amounted to 994€. Of this, 26% were costs for light therapy, including transportation and time from work. Outpatient costs were 23% and indirect costs in terms of production losses and lost leisure time were 22%. We saw a great variability in costs depending on the level of treatment intensity, ranging from 369€ for topical treatment only to 1,709€ for patients treated with biological drugs in addition to topical and light therapy treatment.

In comparison to other psoriasis studies, our results indicate an almost double cost per year in current prices, although the included cost items differed and relative prices may have changed over time [4, 6-8]. In addition, comparisons between health care systems may not be straight forward due to differences in treatment patterns, organisational structures and unit prices. Furthermore, due to the introduction of new treatment alternatives, e.g. biological drugs, comparisons with older studies can be difficult.

As the results of this study were based on resources consumed during one month in fall/winter 2009 when psoriasis tends to flare, the patients were asked to estimate the number of visits during one year. Using these estimates, i.e. four times lower costs for health care visits, would result in an annual cost of 7,300€ which was closer to the above mentioned references. Concerning other resources than health care visits, it is reasonable to think that the use of moisturizers would be reduced during the sunnier season. Another important cost driver was productivity losses and it is uncertain how seasonal variation affects productivity. Maybe the seasonal aspect influenced the analysis of relationship between costs and clinical variables and patient characteristics which we thought would be discernable. Although DLQI and only a severity score of PASI>12 and a few patient characteristics turned out as significant covariates, it seemed like the explanatory factors for costs are complex or perhaps due to a too small patient sample. The relationship between PASI and DLQI has also been shown to be difficult to establish [14], indicating that the quality-of-life domains are poorly measured with PASI.

The quality-of-life measured with the EQ-5D instrument indicated an average QALY-weight of 0.71. The corresponding figure for the general population in Sweden at a mean age of 52 has been estimated at 0.84 in the year 1998 [15]. Without controlling for other patient characteristics, the average QALY in psoriatics would then be 0.13 QALYs less than in the general population. This corresponds to almost 1.6 months per year of perfect health lost among the surveyed patients compared to the general population. Similar quality of life reductions among psoriatic patients compared to the general population have been shown in other countries with different techniques and instruments [16-18]. In line with results from Schmitt and Ford, we found a relationship between productivity loss and DLQI although reduced quality of life due to unemployment may not be uniquely attributed to the psoriasis [19].

Unit costs were taken from official sources and should ideally reflect the opportunity cost. This may not, however, always be the case for several reasons. First, unit prices from the Southern Sweden Health Care Region sometimes apply an indexation of unit prices/tariffs from previous years and may not fully reflect the actual cost of providing the services as the content and input prices of services may have changed. Second, the same unit price is used for both Malmö and Kristianstad although differences may exist in reality. However, this source is the best available and is relevant for the study sample. Due to the plethora of different over-the-counter drugs (OTC) and naturopathic preparations, we assumed a fixed amount of SEK100 per month for these two cost items.

The patients were recruited at two hospital based specialist clinics and may therefore not be representative of all psoriatic patients, as many psoriasis patients may be treated at the primary care level. The difference in level of specialisation between the two hospitals was, however, an attempt to cover some variation in patient characteristics and treatment alternatives. The difference in use of biological treatment between sexes could be the result of some fertile women avoiding this alternative if they are planning a pregnancy. It could also be a reflection of women having better managed psoriasis with topical treatment and light therapy, as reflected in the amount of time spent on topical application and light sessions, reducing the clinical grounds for prescribing biological drugs. As a comparison, the Swedish register for the systemic treatment of psoriasis, PsoReg, reported 60 per cent males on systemic treatment [20]. At registration, males had also in general higher PASI and quality-of-life scores than women. Gender differences in Swedish psoriasis care have also been studied by Uttjek et al. with similar results to ours, but also describing differences in psoriasis care expectations between genders [21]. However, as the regression analysis showed, gender did not turn out to be a statistically significant determinant for either quality of life or costs. Hence, there may be other factors, not measured in this study, that could explain differences in gender, or our patient sample was too small. What we did see was that middle-aged patients with lower education levels and smokers had lower costs, possibly indicating socio-economic differences in patient demand and/or treatment. We would otherwise expect costs to increase among smokers, as they tend to have more severe psoriasis [22].

The costs for two other autoimmune conditions, Crohn's disease (CD) and rheumatoid arthritis (RA) have been estimated at 721€ for 4 weeks (2007 prices) and 1,002€ per month (2004 prices), respectively [23, 24]. Here, the main cost drivers were indirect costs, 65% in CD and 55% in RA. Although the use of biological treatments is more prevalent today than when these studies were performed, their use has been shown to offset, at least partially, some of the indirect costs, which is why we can argue that the comparison still has some bearing [25]. This indicates that the cost of psoriasis care could be on par with other autoimmune conditions.

One limitation with this prevalence study is the cross-sectional design of the current treatment where the treatment strategy was used as a proxy for disease severity. The sample included a mix of patients with well- and less well controlled psoriasis and patients who recently may have changed treatment where the full effect has not yet been attained. This could result in costs that may not reflect the long-term clinical results. It would therefore be interesting to follow cohorts from the change of treatment in terms of costs and quality of life, i.e. an incidence approach. Ideally they would be followed during a year to cover the seasonal variations.

Conclusion

The one month cost-of-illness for a psoriasis patient during fall/winter amounted to almost 1,000€/month, with great variations depending on treatment strategy. Despite a 1,190€ higher drug cost for biological- compared to traditional systemically treated patients, total cost per month differed by 623€ because of offsets from improved productivity. In addition, a trend towards lower severity and reductions in OP and topical treatment was seen. The association between costs and severity was complex, perhaps due to the selected study period and the study design.

Disclosure

Financial support: this study was sponsored by Abbott Scandinavia AB, Sweden. They facilitated the management of the study but were not involved in the data collection or analysis. Conflict of interest: Anders Ljungberg is an employee at Abbott Scandinavia AB.

References

1. Mallbris L, Akre O, Granath F, et al. Increased risk for cardiovascular mortality in psoriasis inpatients but not in outpatients. Eur J Epidemiol 2004; 19(3):225-30.

2. Pathirana D, Ormerod AD, Saiag P, et al. European S3-guidelines on the systemic treatment of psoriasis vulgaris. J Eur Acad Dermatol Venereol 2009 Oct;23 Suppl 2:1-70.

3. Mrowietz U, Kragballe K, Reich K, et al. Definition of treatment goals for moderate to severe psoriasis: a European consensus. Arch Dermatol Res. 2011 Jan;303(1):1-10.

4. Navarini AA, Laffitte E, Conrad C, et al. Estimation of cost-of-illness in patients with psoriasis in Switzerland. Swiss Med Wkly 2010 Feb 6;140(5-6):85-91.

5. Driessen RJ, Bisschops LA, Adang EM, Evers AW, Van De Kerkhof PC, De Jong EM. The economic impact of high-need psoriasis in daily clinical practice before and after the introduction of biologics. The British journal of dermatology 2010 Jun;162(6):1324-9.

6. Berger K, Ehlken B, Kugland B, Augustin M. Cost-of-illness in patients with moderate and severe chronic psoriasis vulgaris in Germany. J Dtsch Dermatol Ges 2005 Jul;3(7):511-8.

7. Sohn S, Schoeffski O, Prinz J, et al. Cost of moderate to severe plaque psoriasis in Germany: a multicenter cost-of-illness study. [Comparative Study Multicenter StudyResearch Support, Non-U.S. Gov’t] Dermatology 2006; 212(2):137-44.

8. Colombo G, Altomare G, Peris K, et al. Moderate and severe plaque psoriasis: cost-of-illness study in Italy. Ther Clin Risk Manag 2008 Apr;4(2):559-68.

9. The EuroQol Group. EuroQol--a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy 1990 Dec;16(3):199-208.

10. Finlay AY. Quality of life measurement in dermatology: a practical guide. Br J Dermatol 1997 Mar;136(3):305-14.

11. Dolan P. Modeling valuations for EuroQol health states. Med Care 1997 Nov;35(11):1095-108.

12. Statistics Sweden. Wage and salary statistics. 2009.

13. Hjort C, Tenskog M. The Swedish transport agency's socio-economic calculations. Borlänge: The Swedish Transport Agency2006. Report No.: 2006:127.

14. Schafer I, Hacker J, Rustenbach SJ, Radtke M, Franzke N, Augustin M. Concordance of the Psoriasis Area and Severity Index (PASI) and patient-reported outcomes in psoriasis treatment. Eur J Dermatol 2010 Jan-Feb;20(1):62-7.

15. Burstrom K, Johannesson M, Diderichsen F. A comparison of individual and social time trade-off values for health states in the general population. Health Policy 2006 May;76(3):359-70.

16. Schmitt J, Meurer M, Klon M, Frick KD. Assessment of health state utilities of controlled and uncontrolled psoriasis and atopic eczema: a population-based study. Br J Dermatol 2008 Feb;158(2):351-9.

17. Badia X, Mascaro JM, Lozano R. Measuring health-related quality of life in patients with mild to moderate eczema and psoriasis: clinical validity, reliability and sensitivity to change of the DLQI. The Cavide Research Group. Br J Dermatol 1999 Oct;141(4):698-702.

18. Wahl A, Loge JH, Wiklund I, Hanestad BR. The burden of psoriasis: a study concerning health-related quality of life among Norwegian adult patients with psoriasis compared with general population norms. J Am Acad Dermatol 2000 Nov;43(5 Pt 1):803-8.

19. Schmitt JM, Ford DE. Work limitations and productivity loss are associated with health-related quality of life but not with clinical severity in patients with psoriasis. Dermatology 2006; 213(2):102-10.

20. Schmitt-Egenolf M. Swedish registry for systemic psoriasis treatment Annual report September 2010. Umeå: Public health and clinical medicine, Umeå Universitet2010.

21. Uttjek M, Dufaker M, Nygren L, Stenberg B. Psoriasis care consumption and expectations from a gender perspective in a psoriasis population in northern Sweden. Acta Derm Venereol 2005; 85(6):503-8.

22. Armstrong AW, Armstrong EJ, Fuller EN, Sockolov ME, Voyles SV. Smoking and pathogenesis of psoriasis: a review of oxidative, inflammatory and genetic mechanisms. The British journal of dermatology 2011 Jul 20.

23. Mesterton J, Jonsson L, Almer SH, Befrits R, Friis-Liby I, Lindgren S. Resource use and societal costs for Crohn's disease in Sweden. Inflamm Bowel Dis 2009 Apr 30;15(12):1882-90.

24. Jacobsson LT, Lindroth Y, Marsal L, Juran E, Bergstrom U, Kobelt G. Rheumatoid arthritis: what does it cost and what factors are driving those costs? Results of a survey in a community-derived population in Malmo, Sweden. Scand J Rheumatol 2007 May-Jun;36(3):179-83.

25. Augustsson J, Neovius M, Cullinane-Carli C, Eksborg S, van Vollenhoven RF. Patients with rheumatoid arthritis treated with tumour necrosis factor antagonists increase their participation in the workforce: potential for significant long-term indirect cost gains (data from a population-based registry). Ann Rheum Dis 2010 Jan;69(1):126-31.

26. Southern Regional Health Care Committee. Regional Tariffs and Fees for the Southern Health Care Region in 2008. 2009.

27. The Swedish Association of the Pharmaceutical Industry (LIF). FASS (Pharmaceutical Specialties in Sweden). Stockholm 2009.