- Email: [email protected]
Cost-effectiveness analysis in the treatment of heavy menstrual bleeding in Spain
European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Cost-effectiveness analysis in the treatment of heavy menstrual bleeding in Spain J. Calaf a,*, I. Lete b, I. Canals c, C. Crespo d,e, B. Espino´s f, I. Cristo´bal g,1 a
Gynaecology and Obstetrics Department, Santa Creu i Sant Pau Hospital, Barcelona, Spain Gynaecology Department, University Hospital of Araba, Vitoria, Spain c Medical Department, Bayer HealthCare, Barcelona, Spain d Department of Statistics, University of Barcelona, Barcelona, Spain e Oblikue Consulting, Barcelona, Spain f Market Access Department, Bayer HealthCare, Barcelona, Spain g Gynaecology and Obstetrics Department, Sanitas La Zarzuela Hospital, Madrid, Spain b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 June 2014 Received in revised form 8 October 2014 Accepted 22 October 2014
Objective: To compare the effectiveness and costs associated with first-line medical treatments for chronic heavy menstrual bleeding (HMB) in Spain. Study design: A cost-effectiveness analysis was conducted comparing the levonorgestrel-releasing intrauterine system (LNG-IUS) with the estradiol valerate/dienogest multiphase oral contraceptive (E2V/DNG), combined oral contraceptives (COC) and progestins (PROG). Study patients were fertile women diagnosed with HMB who initially wished to remain fertile. A Markov model based on reported clinical data and the opinion of a panel of experts was used. The time horizon of the analysis was 5 years. The analysis was conducted from the perspective of the Spanish National Health System (NHS), discounting both costs (s 2013) and future effects at an annual rate of 3%. One-way sensitivity analyses and probabilistic sensitivity analysis were performed to test the robustness of the results. Results: In the analysis at 5 years, the LNG-IUS was associated with a gain of 0.67, 2.22, and 3.53 symptoms free months (SFM) compared with E2V/DNG, COC and PROG, respectively. LNG-IUS contributed more quality-adjusted life months (QALM) than the other treatment alternatives (+1.74 vs. E2V/DNG, +3.33 vs. COC +3.53 vs. PROG). First-line LNG-IUS treatment resulted in savings of s 583, s 988, and s 1891 vs. E2V/DNG, COC and PROG, respectively. These cost benefits, coupled with the greater clinical benefits in terms of SFM and QALM, show that LNG-IUS is the dominant option (less costly and more effective). Conclusion: LNG-IUS is the medical treatment of choice and cost-saving option for the control of HMB in Spain. ß 2014 Published by Elsevier Ireland Ltd.
Keywords: Levonorgestrel-releasing intrauterine system Estradiol valerate/dienogest Heavy menstrual bleeding Cost-effectiveness Spain
Introduction In recent years, efforts have been made to reach a consensus definition of heavy menstrual bleeding (HMB), both in * Corresponding author at: Obstetrics and Gynaecology Service, Santa Creu i Sant Pau Hospital, Sant Antoni Ma Claret Avenue, 167, 08025 Barcelona, Spain. Tel.:+34 935537043; fax: +34 935537057 E-mail address: [email protected] (J. Calaf). 1 Expert panel: Andeyro M. (Hospital La Zarzuela de Madrid), Cancelo MJ. ˜ ete (Hospital Universitario de Guadalajara), Cano A. (Hospital Peset. Valencia), Can ˜ a), Correa M. ML. (Hospital de Toledo, Toledo), Cobian F. (Clinica Cobian, A Corun (Hospital Universitario de Canarias), Doval JL. (Complejo Hospitalario Universitario de Ourense), Gutie´rrez J. (Hospital Virgen Macarena, Sevilla), Haimovich S. (Hospital del Mar, Barcelona), Herna´ndez A. (Hospital La Paz, Madrid), Mendoza N. (Clı´nica Margen). http://dx.doi.org/10.1016/j.ejogrb.2014.10.034 0301-2115/ß 2014 Published by Elsevier Ireland Ltd.
international organizations [1–3] and in Spain, in the framework of the Spanish Society of Gynaecology and Obstetrics (SEGO) consensus documents on menorrhagia [4]. According to the definition of the International Federation of Gynaecology and Obstetrics (FIGO) on abnormal bleeding, HMB is defined as excessive menstrual blood loss (>80 ml) that interferes with normal physical, emotional or social activity and worsens the quality of life of women [2,4]. There are no objective Spanish data on the prevalence of HMB, but globally it is estimated that between 8% and 27% of women suffer this disorder [4]. Despite generally being benign, HMB is common among fertile women, has important health implications, is a frequent complaint in primary care and generates a large number of gynecology referrals [4]. HMB also has important social
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
implications, as it affects the quality of life, usually due to the possible development of anemia, which is objectively associated with HMB [4]. The alternatives proposed by the SEGO as first-line treatment in women diagnosed with HMB without contraindications to or rejection of hormones, are the levonorgestrel-releasing intrauterine system (LNG-IUS) and the newly approved regimen estradiol valerate/dienogest multiphasic oral contraceptive (E2V/DNG). Both alternatives are highly effective, achieving a reduction in bleeding of up to 94% and 89%, respectively [4] and are clearly superior in reducing bleeding to other alternatives such as other combined oral contraceptives (COC), progestins (PROG), tranexamic acid and nonsteroidal anti-inflammatory drugs, which achieve reductions in bleeding ranging between 35% and 68% [4]. Their priority use is recommended vs. surgery (hysterectomy or ablation/endometrial resection), which should only be used in the case of treatment failure or contraindication to medical treatment [4]. LNG-IUS and E2V/DNG are relatively non-aggressive and help to preserve fertility, an essential factor that should be taken into account when patient and physician agree on a treatment consensus. The objective of this study was to analyze the effectiveness of current medical alternatives in the first-line treatment of chronic HMB in Spain.
25
The base case was conducted from the perspective of the Spanish National Health System (NHS), considering only direct medical costs and the ex-factory price of drug treatment. All costs were updated to 2013 according to the Consumer Price Index published by the National Statistics Institute [6]. Costs and clinical outcomes were discounted at an annual rate of 3% [7]. Model structure The model simulates a hypothetical cohort of fertile women diagnosed with HMB who wished to preserve their fertility but, that at the time of starting treatment, did not wish to become pregnant. Thus, the health states of the model were defined combining HMB control and birth control (Fig. 1). All HMB patients entering the model initiate first line treatment with one of the options assessed. Patients remain on the same therapeutic alternative until there is a failure to control HMB or failure of birth control. In order to reflect all main events that could happen during the treatment, including pregnancy and abortion, treatment evaluations were made every 6 months, throughout the time horizon of the model (5 years). Those patients who fail controlling HMB (Fig. 2) or birth control (Table 1) switch to other alternative as a second-line therapy, according to current clinical practice in Spain.
Materials and methods
Data collection
A cost-effectiveness analysis using a Markov model of 6monthly cycles was performed to simulate the evolution of a cohort of women in reproductive age, with HMB initiating first-line medical treatment. The model compared costs and effects of different treatment patterns beginning from first line therapy until surgery and over a 5 year horizon (the time until LNG-IUS replacement is required). For each treatment strategy, the model estimated the difference in costs, symptom-free months (SFM), surgery-free months (SuFM) and quality-adjusted life months (QALM). Comparison of the incremental results between treatment with LNG-IUS and other alternatives (E2V/DNG, COC and PROG) was made by calculating the incremental cost-effectiveness ratio (ICER), which was subsequently compared with the accepted cost-effectiveness threshold in Spain of s 2500/QALM [5].
We conducted a systematic review of the literature in PUBMED, Cochrane and DARE to locate clinical studies or reviews (published in English and Spanish) that evaluated HMB and birth control efficacy, health related quality of life (HRQoL) and economic consequences of HMB and birth control. The search strategy included the following key words: heavy menstrual bleeding, abnormal uterine bleeding, menorragia, hypermenorrhea, excessive menstrual loss, DUB, dysfunctional uterine bleeding, progestogen, progesterone, levonorgestrel, intrauterine-device-medicated, IUS, LNG IUS, mirena, progestasert, birth control, oral contraceptives. A panel of 14 clinical experts in HMB treatment, representative of the different Spanish Autonomous Communities (see Expert
Fig. 1. Structure of the model.
26
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
Fig. 2. Probabilities of treatment switching after HMB failure. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest. Source: expert opinion (routine clinical practice).
Panel) was consulted in all phases of the study. Two rounds of questionnaires were used to validate data collection and to estimate the parameters that were not identified in published literature. Moreover, model structure, inputs, assumptions performed and results, were agreed and validated in an Advisory Board.
Treatment options The choice of study comparators reflects Spanish clinical practice. Therefore, the model compared costs and effects of treatment sequences initiated with LNG-IUS, E2V/DNG, or the most frequently used COC or PROG in Spain.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
27
Table 1 Probabilities of treatment switching after contraception failure. Change treatment to
Patients treated with COC
COC PROG LNG-IUS E2V/DNG Surgery
PROG
LNG-IUS
E2V/DNG
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
0.0 0.0 80 0.0 20
0.0 0.0 90 2.5 7.5
0.0 0.0 80 0.0 20
0.0 0.0 90 2.5 7.5
0.0 0.0 60 15 25
0.0 0.0 70 25 5
2.5 0.0 2.5 30 65
2.5 0.0 17.5 25 55
Source: expert opinion (routine clinical practice). COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Treatment efficacy data
Utilities
The efficacy parameters considered in the model were therapeutic success in controlling HMB and therapeutic success in birth control. After reviewing the quality of the studies, the results were weighted to obtain a more-reliable measure of joint probability. Finally, for the LNG-IUS option we weighted the results of three studies [8–10], for the E2V/DNG option two studies [11,12], for the COC option one study [13] and for PROG the study by Irvine et al. [14] (Fig. 3). The long-term efficacy of E2V/DNG, COC and PROG was obtained by extrapolating the short-term efficacy data observed from the pattern of change in efficacy of LNG-IUS (Fig. 3). The probabilities of successful birth control for all treatments were also estimated by a literature review [15,16] (Table 2). The proportion of resections (20%) and hysterectomy (80%) among surgical events and the proportion of women who changed treatment after contraceptive failure according to whether this resulted in pregnancy (25%) or abortion (75%) (Table 1), reflect common clinical practice based on the experts experience.
Utility values, which measure the preference of the patient to be in a specific health state, with a range of 0 to 1, where 1 represents the perception of being in perfect health, were used to quantify the impact of HMB on the quality of life and to calculate the QALMs. A utility value of 0.95 was applied to the control of HMB health states, 0.55 to the failure of control of HMB health states and 0.65 to patients derived to surgery [17].
Treatment switching To our knowledge, there are no Spanish publications that describe routine clinical practice after HMB and birth control failure. Therefore, the probabilities of switching treatment after HMB (Fig. 2) or birth control (Table 1) failure represent the routine clinical practice based on the experts experience (see ‘‘Data Collection’’).
Resource use and costs The total cost of the management of HMB was estimated considering: the costs of the therapeutic alternatives for the control of HMB, the costs associated with surgical procedures, the cost of the consequences of contraception and the cost of the annual follow up of patients (Table 3). The base case analysis was performed with the ex-factory price of treatment alternatives [18] which were reimbursed by the NHS at the moment of the analysis, considering a 60% contribution of the NHS and applying the appropriate discount according to Royal Decree-Law 8/2010 (Royal Decree 8/2010). All resource use and unit cost data [19] (Table 3) were agreed with the panel of experts. Sensitivity analysis To assess the influence of the uncertainty of the model on the study results, and to validate their robustness, sensitivity analyses were conducted. Two one-way sensitivity analyses were
Fig. 3. Efficacy of treatments in controlling HMB. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest; data in gray: data referenced directly from Gupta [8], Endrikat [9], Middleton [10] (LNG-IUS); Fraser [11], Fraser [12] (E2V/DNG), Blumenthal [13] (COC) and Irvine et al. [14] (PROG; resection). Other data: data extrapolated from the pattern of change of efficacy of LNG-IUS.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
28
Table 2 Treatment efficacy in birth control. Therapy
Therapeutic success of birth control (%)
LNG-IUS COC E2V/DNG PROG Resection
99.30% at 1 year(1) 98.75% at 1 year(1) 98.75% at 1 year(1) 85% at 1 year(2) 100% at 1 year(2)
Source: (1) Mansour et ?al. [15]; (2) assumptions from Lete et al. [16]. COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Table 3 Unit costs. Resource
Treatment alternatives for HMB control LNG-IUS (unit) COC (per cycle)a E2V/DNG (per cycle) PROG (per cycle)b Condoms (per cycle) Cost of surgical procedures Hysterectomy (per procedure) Analogues (per procedure) Pre-operative resection (per procedure) Resection (per procedure)
Cost (s 2013) NHS (base case)
Social (sensitivity analysis)
s 167.74 s 1.92 s0 s 12.12 s0
s 220.00 s 62.62 s 101.07 s 24.72 s 31.59
s 162.32
Cost of the consequences of contraception Pregnancy (per procedure) Abortion (per procedure) Cost of follow up (annual)
s 4493.64 s 410.53 s 173.07 s 2194.04 s 2339.21 s 1301.58 s 71.55
Source: CGCOF [18] (drug costs); Gisbert 2013 (unit costs), Expert panel review (use of resources). COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrelreleasing intrauterine system; E2V/DNG: estradiol valerate/dienogest; NHS: Spanish National Health System. a The weighted mean cost of the most commonly used COC regimens used in Spain. b Mean cost of the most commonly used oral progestins in Spain (Primolut NOR, Utrogestan and Progeffix).
performed: one considering the social perspective instead of the NHS and the other by extending the time horizon of the model to 10 years, considering the continuation of the treatment with LNG-IUS and its re-insertion after 5 years. For the social perspective the retail price of all treatment alternatives was considered (Table 3) [18].
Following economic evaluation recommendations, in order to test the uncertainty around the model variables and the robustness of the results a probabilistic sensitivity analysis (PSA) was performed using a Monte Carlo simulation (1000 iterations) on modeled variables using a Log-Normal distribution for costs, a Normal distribution for utility scores and a Beta distribution for probabilities [20].
Results Clinical benefits Both the results at 6 months and at 5 years showed that initiating treatment with LNG-IUS was the most effective of the HMB treatment options considered in terms of SFM and QALM (Figs. 4 and 5). At five years, treatment continuity was 72.3% for LNG-IUS, 31% for E2V/DNG, 15% for COC and 1% for PROG. The improvements in effectiveness and quality of life were associated with the fact treatment failures with COC, PROG and E2V/DNG permitted switch to more effective treatments such as LNG-IUS. This resulted in similar effectiveness at 5 years between treatment options (86% for LNG-IUS, 84% for E2V/DNG, 82% for COC and 80% for PROG), while at 6 months the differences in maintaining control of HMB were higher (83.2% for LNG-IUS, 72% for E2V/DNG, 61.2% for COC, and 22% for PROG). The greater effectiveness of LNG-IUS allowed patients to benefit from less aggressive treatment, which increased surgery-free survival. In addition, patients treated with LNG-IUS were referred for surgery later than patients treated with other alternatives (5 SuFM vs. E2V/DNG, 8.4 SuFM vs. COC and 13.63 SuFM vs. PROG). Economic benefits At 6 months, the mean cost per patient was s 205, s 325, s 416 and s 796 for the therapeutic lines initiated with LNG-IUS, E2V/ DNG, COC and PROG, respectively (Fig. 6). The cost difference between LNG-IUS and the other therapies increased gradually after the first 6 months of treatment, resulting in a difference of between s 120 and s 583 compared with E2V/DNG and between s 210 and s 988 compared with COC and between s 590 and s 1891 compared with PROG, at 6 months and 5 years, respectively (Fig. 6). Incremental cost-effectiveness analysis The analysis showed that the dominant therapeutic line was that initiated with LNG-IUS, leading to a mean cost per patient lower than any of the alternatives and providing greater effectiveness in terms of QALM (Table 4).
Fig. 4. Symptoms free month (SFM) at (a) 6 months and (b) 5 years. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
29
Fig. 5. Quality-adjusted life months (QALM) gained at (a) 6 months and (b) 5 years. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Fig. 6. Total cost of the treatment alternatives (5 years). COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest.
Sensitivity analysis The one-way sensitivity analysis showed similar results to the base case and confirmed that changing the analysis perspective from NHS to the social and increasing the time horizon to 10 years resulted in LNG-IUS continuing to be the dominant option. The probabilistic sensitivity analysis also confirmed the dominance of LNG-IUS in all simulations vs. COC and PROG and in 98% of cases vs. E2V/DNG (Fig. 7). Comments The results of this study indicate that initiating treatment for HMB with LNG-IUS, in addition to preserving fertility and delaying surgery need, saves money and provides clinical advantages and better quality of life than other therapeutic alternatives. Despite the initial cost of LNG-IUS insertion, this is compensated by its high
effectiveness in controlling HMB, leading to a smaller proportion of patients requiring surgery, which is more aggressive for patients and more costly for the healthcare system. These results are reinforced by the analysis at 10 years and the sensitivity analysis, which showed results similar to those of the base case. This study can be considered as an update of the analysis by Lete et al. [16] which compared the cost and effectiveness of LNG-IUS vs. COC and PROG and concluded that LNG-IUS was cost-effective compared to the other treatment alternatives. E2V/DNG was not included as it was not indicated for the treatment of HMB at the time of the study. The present study therefore responds to the need to assess LNG-IUS in the current scenario for the treatment of HMB, where E2V/DNG is also a major player and has shown greater efficacy than other oral treatments, including COC and PROG [11,12]. Unlike the study by Lete et al. [16] in which the efficacy of all the HMB treatment alternatives was held constant, the present analysis considered the change in efficacy over time on the basis of
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
30
Table 4 Analysis of cost effectiveness (5 years). COST LNG-IUS E2V/DNG COC PROG
s 1531 s 2114 s 2518 s 3421
Cost difference
s 583 s 988 s 1891
QALM 49.57 47.83 46.24 44.18
QALM difference
ICER vs. LNG-IUS
1.74 3.33 5.39
Dominant Dominant Dominant
ICER: incremental cost-effectiveness ratio; COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest; QALM: quality-adjusted life months.
The results of this analysis demonstrate that initiating treatment with LNG-IUS should be the treatment of choice for HMB as it is a cost-saving option compared with the other options studied. Acknowledgements This study was funded by Bayer Hispania, the manufacturer of LNG-IUS and E2V/DNG, and developed by Oblikue Consulting on its behalf. All experts received honoraria for participating in the model elaboration and validation Advisory Boards. None of them received economic consideration for developing this article. Canals and B. Espino´s are employed by Bayer HealthCare, the sponsor of the study. C. Crespo was employed by Oblikue Consulting at the time of the execution and analysis of the study, which received financing from Bayer HealthCare to carry out the study.
References
Fig. 7. Probabilistic sensitivity analysis (1 year). COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest; QALM: quality-adjusted months.
new evidence. Our results are consistent with other international studies comparing the clinical and economic benefits of using LNGIUS or other alternatives [21–23]. Additionally, this study is the first one to evaluate the differential value of LNG-IUS vs. E2V/DNG. The present study does not compare therapies but essentially compares different treatment patterns. Therefore, the results show that the treatment pattern initiated with LNG-IUS provided greater clinical and economic benefits. This is important in the long-term because less-effective therapies benefit from the clinical results of the most effective therapies used as second or third line treatments. For direct comparison, the net impact of each could be determined separately, although this does not reflect routine clinical practice. Thus, this study represents an advance in the economic evaluation of treatments for HMB as it reflects the current management in Spain and, depending on current clinical practice, could be extrapolated to other settings. The potential limitations of this study are derived from the lack of information in the literature, which resulted in various assumptions being made. The diagnosis of HMB is subjective, and there is no uniformity in the criteria of the therapeutic success of HMB in the studies used to determine effectiveness in controlling HMB. In addition, the same efficacy in birth control was assumed for all treatment options in the group of COC and E2V/DNG [15,16]. Likewise, the model did not consider discontinuation of treatment for HMB. Another limitation was that conditioned by the lack of quality of life data from Spanish HMB patients. The utilities for each health state were obtained from the international literature [17]. However, we believe that the limitations and assumptions of the study have been compensated by always being based on data validated by clinical experts and by having been analyzed in depth in the various sensitivity analyses, which showed that the uncertainty associated with these constraints did not change the results of the model.
[1] Fraser IS, Critchley HO, Munro MG, Broder M. Can we achieve international agreement on terminologies and definitions used to describe abnormalities of menstrual bleeding? Hum Reprod 2007;22(3):635–43. [2] Munro MG, Critchley HO, Broder MS, Fraser IS, FIGO Working Group on Menstrual Disorders. FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age. Int J Gynaecol Obstet 2011;113(1):3–13. [3] National Collaborating Centre for Women’s and Children’s Health. Heavy menstrual bleeding guideline; 2007, Available from: http://www.nice. org.uk/ nicemedia/pdf/CG44FullGuideline.pdf [Cited 29.05.14]. [4] Consensus documents on menorrhagia SEGO; 2013, Available from: http:// www.sego.es [Cited 29.05.14]. [5] Sacrista´n JA, Oliva J, Del Llano J, Prieto L, Pinto JL. What is an efficient health technology in Spain? Gac Sanit 2002;16(4):334–43. [6] INE (National Statistics Institute). Consumer Price Index; 2013, Available from: http://www.ine.es [Cited 29.05.14]. ˜ anzas F, et al. Spanish recommendations on [7] Lo´pez-Bastida J, Oliva J, Anton economic evaluation of health technologies. Eur J Health Econ 2010;11: 513–20. [8] Gupta J, Kai J, Middleton L, Pattison H, Gray R, Daniels J, ECLIPSE Trial Collaborative Group. Levonorgestrel intrauterine system vs. medical therapy for menorrhagia. N Engl J Med 2013;368(2):128–37. [9] Endrikat J, Shapiro H, Lukkari-Lax E, Kunz M, Schmidt W, Fortier M. A Canadian: multicentre study comparing the efficacy of a levonorgestrel-releasing intrauterine system to an oral contraceptive in women with idiopathic menorrhagia. J Obstet Gynaecol Can 2009;31(4):340–7. [10] Middleton LJ, Champaneria R, Daniels JP, et al. Hysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system (Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from individual patients. Br Med J 2010;341:c3929. [11] Fraser IS, Parked S, Mellinger U, Machlitt A, Serrani M, Jensen J. Eur J Contracept Reprod Health Care 2011;16:258–69. [12] Fraser IS, Jensen J, Schaefers M, Mellinger U, Parke S, Serrani M. Normalization of blood loss in women with heavy menstrual bleeding treated with an oral contraceptive containing estradiol valerate/dienogest. Contraception 2012;86(2):96–101. [13] Blumenthal PD, Trussell J, Singh RH, Guo A, Borenstein J, Dubois RW. Costeffectiveness of treatments for dysfunctional uterine bleeding in women who need contraception. Contraception 2006;74:249–58. [14] Irvine GA, Campbelle-Brown MB, Lumsden MA, Heikkila¨ A, Walker JJ, Cameron IT. Randomised comparative trial of the levonorgestrel intrauterine system and norethisterone for the treatment of idiopathic menorraghia. Br J Obstet Gynaecol 1998;104:492–8. [15] Mansour D, Inki P, Gemzell-Danielsson K. Efficacy of contraceptive methods: a review of the literature. Eur J Contracept Reprod Health Care 2010;15(Suppl. 2):S19–31. [16] Lete I, Cristo´bal I, Febrer L, Crespo C, Arbat A, Herna´ndez FJ. Economic evaluation of the levonorgestrel-releasing intrauterine system for the treatment of dysfunctional uterine bleeding in Spain. Eur J Obstet Gynecol Reprod Biol 2011;154(1):71–80. [17] Sculpher M. A cost-utility analysis of abdominal hysterectomy vs. transcervical endometrial resection for the surgical treatment of menorrhagia. Int J Technol Assess Health Care 1998;14:302–19. [18] General Council of Pharmaceutical Associations of Spain Database (CGCOF); 2013. Available from: http://www.portalfarma.com [Cited 29.05. 14]. [19] Gisbert R, Brosa M. Spanish Health Costs Database eSALUD (2014). Barcelona: Oblikue Consulting, S.L.; 2014, Available from: www.oblikue.com/bddcostes [Cited 29.05.14].
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31 [20] Briggs AH. Handling uncertainty in cost-effectiveness models. Pharmacoeconomics 2000;17(5):479–500. [21] You JH, Sahota DS, MoYuen P. A cost-utility analysis of hysterectomy, endometrial resection and ablation and medical therapy for menorrhagia. Hum Reprod 2006;21:1878–83. [22] Clegg JP, Guest JF, Hurskainen R. Cost-utility of levonorgestrel intrauterine system compared with hysterectomy and second generation endometrial
31
ablative techniques in managing patients with menorrhagia in the UK. Curr Med Res Opin 2007;23:1637–48. [23] Ganz ML, Shah D, Gidwani R, Filonenko A, Su W, Pocoski J. The cost-effectiveness of the levonorgestrel-releasing intrauterine system for the treatment of idiopathic heavy menstrual bleeding in the United States. Value Health 2013;16(2):325–33.
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Cost-effectiveness analysis in the treatment of heavy menstrual bleeding in Spain J. Calaf a,*, I. Lete b, I. Canals c, C. Crespo d,e, B. Espino´s f, I. Cristo´bal g,1 a
Gynaecology and Obstetrics Department, Santa Creu i Sant Pau Hospital, Barcelona, Spain Gynaecology Department, University Hospital of Araba, Vitoria, Spain c Medical Department, Bayer HealthCare, Barcelona, Spain d Department of Statistics, University of Barcelona, Barcelona, Spain e Oblikue Consulting, Barcelona, Spain f Market Access Department, Bayer HealthCare, Barcelona, Spain g Gynaecology and Obstetrics Department, Sanitas La Zarzuela Hospital, Madrid, Spain b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 June 2014 Received in revised form 8 October 2014 Accepted 22 October 2014
Objective: To compare the effectiveness and costs associated with first-line medical treatments for chronic heavy menstrual bleeding (HMB) in Spain. Study design: A cost-effectiveness analysis was conducted comparing the levonorgestrel-releasing intrauterine system (LNG-IUS) with the estradiol valerate/dienogest multiphase oral contraceptive (E2V/DNG), combined oral contraceptives (COC) and progestins (PROG). Study patients were fertile women diagnosed with HMB who initially wished to remain fertile. A Markov model based on reported clinical data and the opinion of a panel of experts was used. The time horizon of the analysis was 5 years. The analysis was conducted from the perspective of the Spanish National Health System (NHS), discounting both costs (s 2013) and future effects at an annual rate of 3%. One-way sensitivity analyses and probabilistic sensitivity analysis were performed to test the robustness of the results. Results: In the analysis at 5 years, the LNG-IUS was associated with a gain of 0.67, 2.22, and 3.53 symptoms free months (SFM) compared with E2V/DNG, COC and PROG, respectively. LNG-IUS contributed more quality-adjusted life months (QALM) than the other treatment alternatives (+1.74 vs. E2V/DNG, +3.33 vs. COC +3.53 vs. PROG). First-line LNG-IUS treatment resulted in savings of s 583, s 988, and s 1891 vs. E2V/DNG, COC and PROG, respectively. These cost benefits, coupled with the greater clinical benefits in terms of SFM and QALM, show that LNG-IUS is the dominant option (less costly and more effective). Conclusion: LNG-IUS is the medical treatment of choice and cost-saving option for the control of HMB in Spain. ß 2014 Published by Elsevier Ireland Ltd.
Keywords: Levonorgestrel-releasing intrauterine system Estradiol valerate/dienogest Heavy menstrual bleeding Cost-effectiveness Spain
Introduction In recent years, efforts have been made to reach a consensus definition of heavy menstrual bleeding (HMB), both in * Corresponding author at: Obstetrics and Gynaecology Service, Santa Creu i Sant Pau Hospital, Sant Antoni Ma Claret Avenue, 167, 08025 Barcelona, Spain. Tel.:+34 935537043; fax: +34 935537057 E-mail address: [email protected] (J. Calaf). 1 Expert panel: Andeyro M. (Hospital La Zarzuela de Madrid), Cancelo MJ. ˜ ete (Hospital Universitario de Guadalajara), Cano A. (Hospital Peset. Valencia), Can ˜ a), Correa M. ML. (Hospital de Toledo, Toledo), Cobian F. (Clinica Cobian, A Corun (Hospital Universitario de Canarias), Doval JL. (Complejo Hospitalario Universitario de Ourense), Gutie´rrez J. (Hospital Virgen Macarena, Sevilla), Haimovich S. (Hospital del Mar, Barcelona), Herna´ndez A. (Hospital La Paz, Madrid), Mendoza N. (Clı´nica Margen). http://dx.doi.org/10.1016/j.ejogrb.2014.10.034 0301-2115/ß 2014 Published by Elsevier Ireland Ltd.
international organizations [1–3] and in Spain, in the framework of the Spanish Society of Gynaecology and Obstetrics (SEGO) consensus documents on menorrhagia [4]. According to the definition of the International Federation of Gynaecology and Obstetrics (FIGO) on abnormal bleeding, HMB is defined as excessive menstrual blood loss (>80 ml) that interferes with normal physical, emotional or social activity and worsens the quality of life of women [2,4]. There are no objective Spanish data on the prevalence of HMB, but globally it is estimated that between 8% and 27% of women suffer this disorder [4]. Despite generally being benign, HMB is common among fertile women, has important health implications, is a frequent complaint in primary care and generates a large number of gynecology referrals [4]. HMB also has important social
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
implications, as it affects the quality of life, usually due to the possible development of anemia, which is objectively associated with HMB [4]. The alternatives proposed by the SEGO as first-line treatment in women diagnosed with HMB without contraindications to or rejection of hormones, are the levonorgestrel-releasing intrauterine system (LNG-IUS) and the newly approved regimen estradiol valerate/dienogest multiphasic oral contraceptive (E2V/DNG). Both alternatives are highly effective, achieving a reduction in bleeding of up to 94% and 89%, respectively [4] and are clearly superior in reducing bleeding to other alternatives such as other combined oral contraceptives (COC), progestins (PROG), tranexamic acid and nonsteroidal anti-inflammatory drugs, which achieve reductions in bleeding ranging between 35% and 68% [4]. Their priority use is recommended vs. surgery (hysterectomy or ablation/endometrial resection), which should only be used in the case of treatment failure or contraindication to medical treatment [4]. LNG-IUS and E2V/DNG are relatively non-aggressive and help to preserve fertility, an essential factor that should be taken into account when patient and physician agree on a treatment consensus. The objective of this study was to analyze the effectiveness of current medical alternatives in the first-line treatment of chronic HMB in Spain.
25
The base case was conducted from the perspective of the Spanish National Health System (NHS), considering only direct medical costs and the ex-factory price of drug treatment. All costs were updated to 2013 according to the Consumer Price Index published by the National Statistics Institute [6]. Costs and clinical outcomes were discounted at an annual rate of 3% [7]. Model structure The model simulates a hypothetical cohort of fertile women diagnosed with HMB who wished to preserve their fertility but, that at the time of starting treatment, did not wish to become pregnant. Thus, the health states of the model were defined combining HMB control and birth control (Fig. 1). All HMB patients entering the model initiate first line treatment with one of the options assessed. Patients remain on the same therapeutic alternative until there is a failure to control HMB or failure of birth control. In order to reflect all main events that could happen during the treatment, including pregnancy and abortion, treatment evaluations were made every 6 months, throughout the time horizon of the model (5 years). Those patients who fail controlling HMB (Fig. 2) or birth control (Table 1) switch to other alternative as a second-line therapy, according to current clinical practice in Spain.
Materials and methods
Data collection
A cost-effectiveness analysis using a Markov model of 6monthly cycles was performed to simulate the evolution of a cohort of women in reproductive age, with HMB initiating first-line medical treatment. The model compared costs and effects of different treatment patterns beginning from first line therapy until surgery and over a 5 year horizon (the time until LNG-IUS replacement is required). For each treatment strategy, the model estimated the difference in costs, symptom-free months (SFM), surgery-free months (SuFM) and quality-adjusted life months (QALM). Comparison of the incremental results between treatment with LNG-IUS and other alternatives (E2V/DNG, COC and PROG) was made by calculating the incremental cost-effectiveness ratio (ICER), which was subsequently compared with the accepted cost-effectiveness threshold in Spain of s 2500/QALM [5].
We conducted a systematic review of the literature in PUBMED, Cochrane and DARE to locate clinical studies or reviews (published in English and Spanish) that evaluated HMB and birth control efficacy, health related quality of life (HRQoL) and economic consequences of HMB and birth control. The search strategy included the following key words: heavy menstrual bleeding, abnormal uterine bleeding, menorragia, hypermenorrhea, excessive menstrual loss, DUB, dysfunctional uterine bleeding, progestogen, progesterone, levonorgestrel, intrauterine-device-medicated, IUS, LNG IUS, mirena, progestasert, birth control, oral contraceptives. A panel of 14 clinical experts in HMB treatment, representative of the different Spanish Autonomous Communities (see Expert
Fig. 1. Structure of the model.
26
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
Fig. 2. Probabilities of treatment switching after HMB failure. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest. Source: expert opinion (routine clinical practice).
Panel) was consulted in all phases of the study. Two rounds of questionnaires were used to validate data collection and to estimate the parameters that were not identified in published literature. Moreover, model structure, inputs, assumptions performed and results, were agreed and validated in an Advisory Board.
Treatment options The choice of study comparators reflects Spanish clinical practice. Therefore, the model compared costs and effects of treatment sequences initiated with LNG-IUS, E2V/DNG, or the most frequently used COC or PROG in Spain.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
27
Table 1 Probabilities of treatment switching after contraception failure. Change treatment to
Patients treated with COC
COC PROG LNG-IUS E2V/DNG Surgery
PROG
LNG-IUS
E2V/DNG
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
Pregnancy (%)
Abortion (%)
0.0 0.0 80 0.0 20
0.0 0.0 90 2.5 7.5
0.0 0.0 80 0.0 20
0.0 0.0 90 2.5 7.5
0.0 0.0 60 15 25
0.0 0.0 70 25 5
2.5 0.0 2.5 30 65
2.5 0.0 17.5 25 55
Source: expert opinion (routine clinical practice). COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Treatment efficacy data
Utilities
The efficacy parameters considered in the model were therapeutic success in controlling HMB and therapeutic success in birth control. After reviewing the quality of the studies, the results were weighted to obtain a more-reliable measure of joint probability. Finally, for the LNG-IUS option we weighted the results of three studies [8–10], for the E2V/DNG option two studies [11,12], for the COC option one study [13] and for PROG the study by Irvine et al. [14] (Fig. 3). The long-term efficacy of E2V/DNG, COC and PROG was obtained by extrapolating the short-term efficacy data observed from the pattern of change in efficacy of LNG-IUS (Fig. 3). The probabilities of successful birth control for all treatments were also estimated by a literature review [15,16] (Table 2). The proportion of resections (20%) and hysterectomy (80%) among surgical events and the proportion of women who changed treatment after contraceptive failure according to whether this resulted in pregnancy (25%) or abortion (75%) (Table 1), reflect common clinical practice based on the experts experience.
Utility values, which measure the preference of the patient to be in a specific health state, with a range of 0 to 1, where 1 represents the perception of being in perfect health, were used to quantify the impact of HMB on the quality of life and to calculate the QALMs. A utility value of 0.95 was applied to the control of HMB health states, 0.55 to the failure of control of HMB health states and 0.65 to patients derived to surgery [17].
Treatment switching To our knowledge, there are no Spanish publications that describe routine clinical practice after HMB and birth control failure. Therefore, the probabilities of switching treatment after HMB (Fig. 2) or birth control (Table 1) failure represent the routine clinical practice based on the experts experience (see ‘‘Data Collection’’).
Resource use and costs The total cost of the management of HMB was estimated considering: the costs of the therapeutic alternatives for the control of HMB, the costs associated with surgical procedures, the cost of the consequences of contraception and the cost of the annual follow up of patients (Table 3). The base case analysis was performed with the ex-factory price of treatment alternatives [18] which were reimbursed by the NHS at the moment of the analysis, considering a 60% contribution of the NHS and applying the appropriate discount according to Royal Decree-Law 8/2010 (Royal Decree 8/2010). All resource use and unit cost data [19] (Table 3) were agreed with the panel of experts. Sensitivity analysis To assess the influence of the uncertainty of the model on the study results, and to validate their robustness, sensitivity analyses were conducted. Two one-way sensitivity analyses were
Fig. 3. Efficacy of treatments in controlling HMB. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest; data in gray: data referenced directly from Gupta [8], Endrikat [9], Middleton [10] (LNG-IUS); Fraser [11], Fraser [12] (E2V/DNG), Blumenthal [13] (COC) and Irvine et al. [14] (PROG; resection). Other data: data extrapolated from the pattern of change of efficacy of LNG-IUS.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
28
Table 2 Treatment efficacy in birth control. Therapy
Therapeutic success of birth control (%)
LNG-IUS COC E2V/DNG PROG Resection
99.30% at 1 year(1) 98.75% at 1 year(1) 98.75% at 1 year(1) 85% at 1 year(2) 100% at 1 year(2)
Source: (1) Mansour et ?al. [15]; (2) assumptions from Lete et al. [16]. COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Table 3 Unit costs. Resource
Treatment alternatives for HMB control LNG-IUS (unit) COC (per cycle)a E2V/DNG (per cycle) PROG (per cycle)b Condoms (per cycle) Cost of surgical procedures Hysterectomy (per procedure) Analogues (per procedure) Pre-operative resection (per procedure) Resection (per procedure)
Cost (s 2013) NHS (base case)
Social (sensitivity analysis)
s 167.74 s 1.92 s0 s 12.12 s0
s 220.00 s 62.62 s 101.07 s 24.72 s 31.59
s 162.32
Cost of the consequences of contraception Pregnancy (per procedure) Abortion (per procedure) Cost of follow up (annual)
s 4493.64 s 410.53 s 173.07 s 2194.04 s 2339.21 s 1301.58 s 71.55
Source: CGCOF [18] (drug costs); Gisbert 2013 (unit costs), Expert panel review (use of resources). COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrelreleasing intrauterine system; E2V/DNG: estradiol valerate/dienogest; NHS: Spanish National Health System. a The weighted mean cost of the most commonly used COC regimens used in Spain. b Mean cost of the most commonly used oral progestins in Spain (Primolut NOR, Utrogestan and Progeffix).
performed: one considering the social perspective instead of the NHS and the other by extending the time horizon of the model to 10 years, considering the continuation of the treatment with LNG-IUS and its re-insertion after 5 years. For the social perspective the retail price of all treatment alternatives was considered (Table 3) [18].
Following economic evaluation recommendations, in order to test the uncertainty around the model variables and the robustness of the results a probabilistic sensitivity analysis (PSA) was performed using a Monte Carlo simulation (1000 iterations) on modeled variables using a Log-Normal distribution for costs, a Normal distribution for utility scores and a Beta distribution for probabilities [20].
Results Clinical benefits Both the results at 6 months and at 5 years showed that initiating treatment with LNG-IUS was the most effective of the HMB treatment options considered in terms of SFM and QALM (Figs. 4 and 5). At five years, treatment continuity was 72.3% for LNG-IUS, 31% for E2V/DNG, 15% for COC and 1% for PROG. The improvements in effectiveness and quality of life were associated with the fact treatment failures with COC, PROG and E2V/DNG permitted switch to more effective treatments such as LNG-IUS. This resulted in similar effectiveness at 5 years between treatment options (86% for LNG-IUS, 84% for E2V/DNG, 82% for COC and 80% for PROG), while at 6 months the differences in maintaining control of HMB were higher (83.2% for LNG-IUS, 72% for E2V/DNG, 61.2% for COC, and 22% for PROG). The greater effectiveness of LNG-IUS allowed patients to benefit from less aggressive treatment, which increased surgery-free survival. In addition, patients treated with LNG-IUS were referred for surgery later than patients treated with other alternatives (5 SuFM vs. E2V/DNG, 8.4 SuFM vs. COC and 13.63 SuFM vs. PROG). Economic benefits At 6 months, the mean cost per patient was s 205, s 325, s 416 and s 796 for the therapeutic lines initiated with LNG-IUS, E2V/ DNG, COC and PROG, respectively (Fig. 6). The cost difference between LNG-IUS and the other therapies increased gradually after the first 6 months of treatment, resulting in a difference of between s 120 and s 583 compared with E2V/DNG and between s 210 and s 988 compared with COC and between s 590 and s 1891 compared with PROG, at 6 months and 5 years, respectively (Fig. 6). Incremental cost-effectiveness analysis The analysis showed that the dominant therapeutic line was that initiated with LNG-IUS, leading to a mean cost per patient lower than any of the alternatives and providing greater effectiveness in terms of QALM (Table 4).
Fig. 4. Symptoms free month (SFM) at (a) 6 months and (b) 5 years. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
29
Fig. 5. Quality-adjusted life months (QALM) gained at (a) 6 months and (b) 5 years. COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/DNG: estradiol valerate/dienogest.
Fig. 6. Total cost of the treatment alternatives (5 years). COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest.
Sensitivity analysis The one-way sensitivity analysis showed similar results to the base case and confirmed that changing the analysis perspective from NHS to the social and increasing the time horizon to 10 years resulted in LNG-IUS continuing to be the dominant option. The probabilistic sensitivity analysis also confirmed the dominance of LNG-IUS in all simulations vs. COC and PROG and in 98% of cases vs. E2V/DNG (Fig. 7). Comments The results of this study indicate that initiating treatment for HMB with LNG-IUS, in addition to preserving fertility and delaying surgery need, saves money and provides clinical advantages and better quality of life than other therapeutic alternatives. Despite the initial cost of LNG-IUS insertion, this is compensated by its high
effectiveness in controlling HMB, leading to a smaller proportion of patients requiring surgery, which is more aggressive for patients and more costly for the healthcare system. These results are reinforced by the analysis at 10 years and the sensitivity analysis, which showed results similar to those of the base case. This study can be considered as an update of the analysis by Lete et al. [16] which compared the cost and effectiveness of LNG-IUS vs. COC and PROG and concluded that LNG-IUS was cost-effective compared to the other treatment alternatives. E2V/DNG was not included as it was not indicated for the treatment of HMB at the time of the study. The present study therefore responds to the need to assess LNG-IUS in the current scenario for the treatment of HMB, where E2V/DNG is also a major player and has shown greater efficacy than other oral treatments, including COC and PROG [11,12]. Unlike the study by Lete et al. [16] in which the efficacy of all the HMB treatment alternatives was held constant, the present analysis considered the change in efficacy over time on the basis of
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31
30
Table 4 Analysis of cost effectiveness (5 years). COST LNG-IUS E2V/DNG COC PROG
s 1531 s 2114 s 2518 s 3421
Cost difference
s 583 s 988 s 1891
QALM 49.57 47.83 46.24 44.18
QALM difference
ICER vs. LNG-IUS
1.74 3.33 5.39
Dominant Dominant Dominant
ICER: incremental cost-effectiveness ratio; COC: combined oral contraceptives; PROG: progestins, LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest; QALM: quality-adjusted life months.
The results of this analysis demonstrate that initiating treatment with LNG-IUS should be the treatment of choice for HMB as it is a cost-saving option compared with the other options studied. Acknowledgements This study was funded by Bayer Hispania, the manufacturer of LNG-IUS and E2V/DNG, and developed by Oblikue Consulting on its behalf. All experts received honoraria for participating in the model elaboration and validation Advisory Boards. None of them received economic consideration for developing this article. Canals and B. Espino´s are employed by Bayer HealthCare, the sponsor of the study. C. Crespo was employed by Oblikue Consulting at the time of the execution and analysis of the study, which received financing from Bayer HealthCare to carry out the study.
References
Fig. 7. Probabilistic sensitivity analysis (1 year). COC: combined oral contraceptives; PROG: progestins; LNG-IUS: levonorgestrel-releasing intrauterine system; E2V/ DNG: estradiol valerate/dienogest; QALM: quality-adjusted months.
new evidence. Our results are consistent with other international studies comparing the clinical and economic benefits of using LNGIUS or other alternatives [21–23]. Additionally, this study is the first one to evaluate the differential value of LNG-IUS vs. E2V/DNG. The present study does not compare therapies but essentially compares different treatment patterns. Therefore, the results show that the treatment pattern initiated with LNG-IUS provided greater clinical and economic benefits. This is important in the long-term because less-effective therapies benefit from the clinical results of the most effective therapies used as second or third line treatments. For direct comparison, the net impact of each could be determined separately, although this does not reflect routine clinical practice. Thus, this study represents an advance in the economic evaluation of treatments for HMB as it reflects the current management in Spain and, depending on current clinical practice, could be extrapolated to other settings. The potential limitations of this study are derived from the lack of information in the literature, which resulted in various assumptions being made. The diagnosis of HMB is subjective, and there is no uniformity in the criteria of the therapeutic success of HMB in the studies used to determine effectiveness in controlling HMB. In addition, the same efficacy in birth control was assumed for all treatment options in the group of COC and E2V/DNG [15,16]. Likewise, the model did not consider discontinuation of treatment for HMB. Another limitation was that conditioned by the lack of quality of life data from Spanish HMB patients. The utilities for each health state were obtained from the international literature [17]. However, we believe that the limitations and assumptions of the study have been compensated by always being based on data validated by clinical experts and by having been analyzed in depth in the various sensitivity analyses, which showed that the uncertainty associated with these constraints did not change the results of the model.
[1] Fraser IS, Critchley HO, Munro MG, Broder M. Can we achieve international agreement on terminologies and definitions used to describe abnormalities of menstrual bleeding? Hum Reprod 2007;22(3):635–43. [2] Munro MG, Critchley HO, Broder MS, Fraser IS, FIGO Working Group on Menstrual Disorders. FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age. Int J Gynaecol Obstet 2011;113(1):3–13. [3] National Collaborating Centre for Women’s and Children’s Health. Heavy menstrual bleeding guideline; 2007, Available from: http://www.nice. org.uk/ nicemedia/pdf/CG44FullGuideline.pdf [Cited 29.05.14]. [4] Consensus documents on menorrhagia SEGO; 2013, Available from: http:// www.sego.es [Cited 29.05.14]. [5] Sacrista´n JA, Oliva J, Del Llano J, Prieto L, Pinto JL. What is an efficient health technology in Spain? Gac Sanit 2002;16(4):334–43. [6] INE (National Statistics Institute). Consumer Price Index; 2013, Available from: http://www.ine.es [Cited 29.05.14]. ˜ anzas F, et al. Spanish recommendations on [7] Lo´pez-Bastida J, Oliva J, Anton economic evaluation of health technologies. Eur J Health Econ 2010;11: 513–20. [8] Gupta J, Kai J, Middleton L, Pattison H, Gray R, Daniels J, ECLIPSE Trial Collaborative Group. Levonorgestrel intrauterine system vs. medical therapy for menorrhagia. N Engl J Med 2013;368(2):128–37. [9] Endrikat J, Shapiro H, Lukkari-Lax E, Kunz M, Schmidt W, Fortier M. A Canadian: multicentre study comparing the efficacy of a levonorgestrel-releasing intrauterine system to an oral contraceptive in women with idiopathic menorrhagia. J Obstet Gynaecol Can 2009;31(4):340–7. [10] Middleton LJ, Champaneria R, Daniels JP, et al. Hysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system (Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from individual patients. Br Med J 2010;341:c3929. [11] Fraser IS, Parked S, Mellinger U, Machlitt A, Serrani M, Jensen J. Eur J Contracept Reprod Health Care 2011;16:258–69. [12] Fraser IS, Jensen J, Schaefers M, Mellinger U, Parke S, Serrani M. Normalization of blood loss in women with heavy menstrual bleeding treated with an oral contraceptive containing estradiol valerate/dienogest. Contraception 2012;86(2):96–101. [13] Blumenthal PD, Trussell J, Singh RH, Guo A, Borenstein J, Dubois RW. Costeffectiveness of treatments for dysfunctional uterine bleeding in women who need contraception. Contraception 2006;74:249–58. [14] Irvine GA, Campbelle-Brown MB, Lumsden MA, Heikkila¨ A, Walker JJ, Cameron IT. Randomised comparative trial of the levonorgestrel intrauterine system and norethisterone for the treatment of idiopathic menorraghia. Br J Obstet Gynaecol 1998;104:492–8. [15] Mansour D, Inki P, Gemzell-Danielsson K. Efficacy of contraceptive methods: a review of the literature. Eur J Contracept Reprod Health Care 2010;15(Suppl. 2):S19–31. [16] Lete I, Cristo´bal I, Febrer L, Crespo C, Arbat A, Herna´ndez FJ. Economic evaluation of the levonorgestrel-releasing intrauterine system for the treatment of dysfunctional uterine bleeding in Spain. Eur J Obstet Gynecol Reprod Biol 2011;154(1):71–80. [17] Sculpher M. A cost-utility analysis of abdominal hysterectomy vs. transcervical endometrial resection for the surgical treatment of menorrhagia. Int J Technol Assess Health Care 1998;14:302–19. [18] General Council of Pharmaceutical Associations of Spain Database (CGCOF); 2013. Available from: http://www.portalfarma.com [Cited 29.05. 14]. [19] Gisbert R, Brosa M. Spanish Health Costs Database eSALUD (2014). Barcelona: Oblikue Consulting, S.L.; 2014, Available from: www.oblikue.com/bddcostes [Cited 29.05.14].
J. Calaf et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 184 (2015) 24–31 [20] Briggs AH. Handling uncertainty in cost-effectiveness models. Pharmacoeconomics 2000;17(5):479–500. [21] You JH, Sahota DS, MoYuen P. A cost-utility analysis of hysterectomy, endometrial resection and ablation and medical therapy for menorrhagia. Hum Reprod 2006;21:1878–83. [22] Clegg JP, Guest JF, Hurskainen R. Cost-utility of levonorgestrel intrauterine system compared with hysterectomy and second generation endometrial
31
ablative techniques in managing patients with menorrhagia in the UK. Curr Med Res Opin 2007;23:1637–48. [23] Ganz ML, Shah D, Gidwani R, Filonenko A, Su W, Pocoski J. The cost-effectiveness of the levonorgestrel-releasing intrauterine system for the treatment of idiopathic heavy menstrual bleeding in the United States. Value Health 2013;16(2):325–33.