1 Organon Laboratories, Cambridge Science Park, Milton Road, Cambridge CB4 0FL, 2 Centre for Health Economics, University of York, Heslington, York YO10 5DD, UK and 3Quathos, Heemraadserf 32, 3 991 KA Houten, The Netherlands
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: cost-effectiveness/HMG/IVF/rFSH/uFSH
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In 1995, recombinant FSH (rFSH) was introduced in the UK and was shown to be more biopotent than urinary FSH (uFSH) (Out et al., 1995; Bergh et al., 1997
). Ongoing pregnancy rates per started cycle, including frozenthawed embryo replacements, were significantly higher using rFSH as compared with uFSH (Out et al., 1995
). In addition, a recent Cochrane meta-analysis showed that the use of rFSH is associated with a 3.7% absolute increase in fresh clinical pregnancy rate as compared with uFSH (Daya and Gunby, 2000
). However, its wide-spread availability was hampered by concerns about its relatively high acquisition costs as compared with urinary gonadotrophins (Meniru, 1999
). In the current study, a cost-effectiveness analysis was undertaken to determine whether the additional acquisition costs and higher pregnancy rate associated with rFSH provides value for money.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Design
A decisionanalytic model (Figure 1) was used to estimate the direct medical cost and associated outcome per woman undergoing IVF treatment. Decision models are widely used in a variety of different scientific disciplines to assess the relative value of different decision options (Weinstein and Fineberg, 1980
). Typically, in economic evaluations, models are used to evaluate treatments where relevant clinical trials have not been conducted or where trials did not collect appropriate economic data (Buxton et al., 1997
) In these scenarios, decisionanalytic models can be used to synthesize the best available data from all available sources. Decision analysis uses a decision tree to represent the various decision options and the outcomes associated with these different options. Although evidence exists regarding the clinical effectiveness of the gonadotrophins in IVF, there has not been a prospective evaluation of the cost-effectiveness of these interventions. The decision tree was created using Data version 3.5 (Tree Age Software, Williamstown, MA, USA).
|
After the initial treatment allocation, patients who became pregnant directly (pregnant-fresh) or via a subsequent thawed embryo transfer (pregnant-frozen) were considered to have a successful outcome and thus did not enter further cycles of the model. Patients who did not become pregnant during the initial cycle either received a further cycle of IVF treatment or dropped out for non-medical considerations. This process continued until all women had received a maximum of three cycles of treatment. Women move through the model according to the probabilities of the events occurring (Table I).
|
For the purpose of this study it was assumed that the probability of becoming pregnant over the first three cycles of IVF treatment is constant, as indicated by the summary of product characteristics of Puregon® (Summary of Product Characteristics, Puregon). Additionally, the chance of stopping IVF treatment due to non-medical reasons (financial or emotional considerations) was assumed to be 20% after the first cycle and 40% after the second cycle. The differential drop-out was considered by the Expert panel to reflect clinical practice in the UK. Ongoing pregnancy rates after uFSH were considered to be equal to those of uFSH-HP (O'Dea and Banks, 1996) and at least equal to HMG (Daya et al., 1995
; Agrawal et al., 2000
; RCOG, 2000
).
Although the clinical trial was not specifically designed to collect health economics data, resource utilization data was obtained during the trial in the form of the daily doses of administered gonadotrophins. Drugs used during the various IVF stages and possible hospitalizations due to ovarian hyperstimulation syndrome (OHSS) were based on the average UK IVF treatment pattern, as derived from interviews with the Expert panel. All relevant drug costs including rFSH, uFSH-HP and HMG were obtained directly from the UK list prices. Table II presents the unit costs of resources at 1999 prices and the average prices for IVF procedures as used in 20 IVF clinics in the UK. For Puregon®, the NHS tariff price per IU has been applied, i.e. £0.403 per IU; for Metrodin-HP® this was £0.248 per IU and for Menogon® £0.132 per IU.
|
Analyses
The decisionanalytic model was used to estimate both the cost and cost-effectiveness of the alternative treatment strategies. The robustness of the results was explored using an extensive series of sensitivity analyses.
The cost analysis was undertaken from the perspective of the IVF provider clinic, whether this was a NHS or private clinic. The analysis considered direct medical costs of the IVF procedures for women up to 12 weeks of pregnancy only. No attempt was made to model either direct non-medical (e.g. travel costs) or indirect costs related to the women, partner or the future of the baby.
In the cost-effectiveness analysis, the average cost per ongoing pregnancy and incremental costs per additional pregnancy were calculated over three cycles. The average cost per ongoing pregnancy was estimated by dividing the cost of the intervention by its effectiveness, and was calculated by the following equation: cost per IVF treatment/chance of ongoing pregnancy. The incremental cost-effectiveness analysis was based on the additional costs that one treatment imposes over another with the additional effect it produces and is considered a more appropriate comparison between relevant alternative treatment strategies (Torgerson and Spencer, 1996). The critical issue here is whether the additional cost (incremental cost) associated with the more expensive treatment is worth paying for the additional benefits (incremental benefits) that arise with a more effective treatment. The decision rules developed to address this issue focus on the incremental cost per additional unit of outcome of one treatment (A), relative to another (B). In the context of this study, this is calculated as the difference in costs between two therapies and the difference in clinical pregnancy rates between two therapies: (Cost ACost B)/(Outcome AOutcome B). Incremental cost-effectiveness ratios were calculated for rFSH compared with HMG and for rFSH compared with uFSH-HP.
In the sensitivity analysis, the robustness of the study outcome was tested by varying critical key parameters expected to either strongly influence the results or to be based on uncertain assumptions. A tornado analysis was run to identify the parameters that most strongly influenced the results. Additionally, where a degree of uncertainty existed (e.g. data derived from expert opinion and observational data), regarding the baseline values used in the model, a series of univariate analyses were run to determine the robustness of the results to plausible changes in each of these parameters (Briggs and Sculpher, 1995). The incremental cost per additional pregnancy for rFSH relative to uFSH-HP and HMG was re-estimated under each alternative scenario. The sensitivity analysis investigated the impact of the price of an IVF procedure on the results by varying the price range from £7502500. This was the price range observed in 20 IVF clinics in the UK. Secondly, the effect of a lower chance of pregnancy after the first cycle was tested by assuming a relative decrease in pregnancy rate of 7% for cycle 2 and 12% for cycle 3. These figures were derived from the Human Fertilisation and Embryology Authority (HFEA) database, as previously reported (Templeton et al., 1996
). This database includes data from 52 507 IVF cycles gathered in the period April 1991 to April 1994 in the UK. Thirdly, the effects of varying the dose from between 30% above and 30% below the baseline average doses found in the trial (rFSH range: 14972779 IU; uFSH-HP and HMG range: 16703101 IU) were investigated (Expert panel).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cost and cost-effectiveness analyses
The cost per IVF treatment with rFSH was £5135. Costs for uFSH-HP and HMG per IVF attempt were £4806 and £4202 respectively. These figures relate to the average cost of IVF treatment over three cycles, accounting for patients who became pregnant in earlier cycles or dropped out for non-medical reasons.
The average cost per ongoing pregnancy using rFSH was lower and hence more favourable (£8992) compared with both uFSH-HP (£10834) and HMG (£9472).
The relative cost-effectiveness of rFSH compared with uFSH-HP and HMG shows that the incremental cost per additional pregnancy for rFSH compared with uFSH-HP was £2583, and for rFSH compared with HMG £7321.
Using just one treatment cycle, the cost per ongoing pregnancy was £8992 in the rFSH group, £10743 in the uFSH group and £9381 in the HMG group.
Sensitivity analyses
Firstly, the impact of the price of an IVF procedure on the results was investigated with the price range observed in 20 IVF clinics in the UK. The cost per IVF procedure reported by the sample varied from £7502500 with an embryo transfer and £4001500 when no embryo transfer occured. The resulting range of the incremental cost per additional pregnancy with embryo transfer was £21882879 and £69267617 compared with uFSH-HP and HMG respectively. Without embryo transfer the relevant range was £22742876 relative to uFSH-HP and £70127614 relative to HMG.
Secondly, the effects of a lower chance of pregnancy after the first cycle were tested by assuming a relative decrease in pregnancy rate of 7% for cycle 2 and 12% for cycle 3. These figures were derived from the HFEA database, as previously reported (Templeton et al., 1996). Using declining chances of pregnancy in cycles 2 and 3, a cumulative pregnancy rate of 54.8% for rFSH and 42.0% for both uFSH-HP and HMG was estimated. This reduction in outcomes translated through into higher average cost-effectiveness ratios for all the drugs. The average cost per ongoing pregnancy for rFSH was £9410, for uFSH-HP it increased to £11477 and for HMG it increased to £10035. The incremental cost-effectiveness ratio for rFSH over uFSH-HP marginally improved to £2043, and for rFSH over HMG the ratio improved to £7156.
Thirdly, the effects of varying the dose from between 30% above and 30% below the baseline average doses (rFSH range: 14972779 IU; uFSH-HP and HMG range: 16703101 IU) was investigated to explore the implications of potential dose variations considered relevant by the Expert panel. Due to the lack of reliable evidence regarding the clinical outcomes associated with these doses, the impact of this dose variation was explored only in relation to the overall costs of each treatment. The impact of changing the dose was explored by varying the cost of the treatment (± 30%) according to the dosage given. The model was re-estimated using these minimum and maximum dose estimates. The resulting incremental cost per additional pregnancy for rFSH ranged from £14393723 relative to uFSH-HP and £47579883 relative to HMG.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The sensitivity analyses demonstrated that the results of the model were relatively robust to variations in the assumptions of the baseline model. The incremental cost per additional pregnancy was insensitive to either wide variations in the cost of IVF services in both NHS and private settings or changes in the assumption that the probability of becoming pregnant was constant over three cycles. The average cost per ongoing pregnancy remained lower with rFSH compared with uFSH-HP and HMG in both these scenarios.
The results of the sensitivity analysis were most sensitive to variations in the dose level assumed in the baseline model. Using the minimum dose level significantly improved the incremental cost-effectiveness of rFSH in comparison with both uFSH-HP and HMG. However, at the maximum dose level (2675 IU for rFSH and 3000 IU for HMG) the incremental cost-effectiveness for rFSH may no longer be as favourable relative to uFSH-HP and HMG. It can be questioned whether such high total doses of rFSH are, on average, a relevant treatment option. In a recent study, an increase from 1727 to 2701 IU of rFSH did not significantly enlarge the number of retrievable oocytes (Out et al., 2000).
It is possible that probabilities for the various outcomes in subsequent cycles will change. No literature data which could be used to estimate the extent of these changes are available. Whether this will influence the outcome of the current study including three treatment cycles depends on the likelihood that these probabilities change differently for rFSH than for uFSH or HMG. It seems reasonable to assume that any changes in probabilities are due to patient characteristics and not to the gonadotrophins used. Moreover, a single cycle analysis also showed a higher cost-effectiveness for rFSH compared with urinary gonadotrophins.
The favourable incremental cost-effectiveness of rFSH lies in its small but significantly higher ongoing pregnancy rate after inclusion of frozenthawed embryo replacements (Out et al., 1995; RCOG, 2000
). It might be argued that this higher pregnancy rate only reflects the results of the cryoprogramme in a limited number of women, and that with longer follow-up such a difference in favour of rFSH may disappear. However, a detailed 2.5 year follow-up on the fate of the remaining frozen embryos after the original study (unpublished observations), identified another 33 rFSH-treated women with a total of 154 embryos thawed for transfer, compared with 16 women in the uFSH group with a total of 62 embryos thawed. An additional 10 pregnancies were found in the rFSH thawed embryos and one in the uFSH group. This shows the consistency of the original observations.
Similar results were obtained when the model was used in Italy (Mantovani et al., 1999) and The Netherlands (van Loon, 1998
). However, from a UK perspective the applicability of the results obtained from the models used in both these countries are limited, because they only consider the comparison between rFSH and uFSH-HP. The incorporation of an additional treatment arm, HMG, which has the lowest acquisition cost, is critical to enable decision-makers to determine the relative cost-effectiveness of all relevant and available treatments in the UK. In addition, the model is further differentiated through the inclusion of patient drop-out due to non-medical reasons, which are considered to be an accurate representation of the real world situation in the UK. The results hold from the perspective of the clinic, whether this is a NHS or private provider. From the patient perspective the results are still the same, as self-funded patients would have to pay for all drug treatments and procedures. Over three cycles the cost per unit of outcome (ongoing pregnancy) is lower with rFSH than with HMG or uFSH-HP.
The decision rule used to determine the relative cost-effectiveness of alternative gonadotrophin treatments may have different implications dependent upon whether the available budget for IVF treatment is fixed. Within the framework of a constrained budget (NHS), the decision to adopt a more expensive treatment could result in a lower number of couples receiving IVF treatment. However, since rFSH has the lowest cost per ongoing pregnancy, this treatment option would still produce a higher number of total pregnancies despite the lower number of couples treated. This analysis assumes the underlying objective of the provider of IVF services is to maximize the expected number of pregnancies. From the perspective of a NHS provider, it would still be possible to treat the same number of couples by re-allocating resources from elsewhere within the total fertility budget or from non-fertility services. In this instance, the incremental cost per additional unit of outcome (pregnancy) would have to be comparable with the incremental cost per additional unit of outcome associated with alternative treatment strategies within these other services. Due to the highly specific outcomes associated with IVF, we can only undertake these external comparisons with other fertility services with a similar outcome objective. The incremental cost per additional unit of outcome associated with rFSH appears extremely favourable in comparison with a variety of alternative fertility treatment services, for example in comparison with the use of IVF versus surgery for tubal factor infertility (moderate: £2402711773; severe £1086712474) and stimulated intra-uterine insemination versus no treatment for male factor infertility (£8546) (Philips et al., 2000).
The assumptions used in the model were conservative and in disadvantage of the rFSH arm. Firstly, pregnancy rates between uFSH-HP and HMG were considered to be equal (Agrawal et al., 2000) whereas evidence exists that these rates are higher using FSH-only preparations (Daya et al., 1995
; RCOG, 2000
).
Secondly, the OHSS incidence between rFSH and uFSH in the trial was slightly but not significantly higher with rFSH (3.2 versus 2.0%). Nevertheless, costs were put in the model because they are known to be high due to hospitalization, frequently in intensive care units.
Thirdly, the route of administration was not taken into account. IVF treatment protocols for rFSH and uFSH-HP allow s.c. injections that may be self-administered by the patient at home and may result in a minimal impact on indirect costs. With an impure preparation as HMG, the i.m. route only is allowed and will generally be nurse-administered within the clinic. Consequently, for HMG there may be associated non-health care costs (travel costs for additional clinic visits) and indirect costs. The likely result of this would be to lower the incremental cost per additional pregnancy for rFSH relative to HMG, therefore improving the cost-effectiveness of rFSH further.
The RCOG tertiary guidelines state that, `because of the small differences in outcome, additional factors should be considered when choosing a gonadotrophin regimen, including patient acceptability, costs and drug availability' (RCOG, 2000).
This study demonstrates the importance of assessing both differences in cost and differences in outcome together and not in isolation. Although the differences in acquisition cost were large, the relative cost-effectiveness of rFSH appeared favourable in comparison with both uFSH-HP and HMG.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Notes |
---|
* At the time of the study, Dr David Sykes was Health Economist and Dr Henk J.Out was Medical Director, both with Organon Laboratories, Cambridge, UK.
Submitted on May 2, 2000; July 5, 2001
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bergh, C., Howles, C.M., Borg, K. et al. (1997) Recombinant human follicle stimulating hormone (r-hFSH, Gonal-F) versus highly purified urinary FSH (Metrodin HP)results of a randomized comparative study in women undergoing assisted reproductive techniques. Hum. Reprod., 12, 21332139.[Abstract]
Briggs, A.H. and Sculpher, M.J. (1995) Sensitivity analysis in economic evaluation: a review of published studies. Health Econ., 4, 355371.[ISI][Medline]
Buxton, M.J., Drummond, M.F., Van Hout, B.A. et al. (1997) Modelling in economic evaluation: an unavoidable fact of life. Health Econ., 6, 217227.[ISI][Medline]
Daya, S. and Gunby, J. (2000) Recombinant versus urinary follicle stimulating hormone for ovarian stimulation in assisted reproduction cycles (Cochrane Review). In The Cochrane Library, Issue 4, 2000. Update Software, Oxford.
Daya, S., Gunby, J., Hughes, E.G. et al. (1995) Follicle-stimulating hormone versus human menopausal gonadotrophin for in vitro fertilization cycles: A meta-analysis. Fertil. Steril., 64, 347354.[ISI][Medline]
Gunnell, D.J. and Ewings, P. (1994) Infertility prevalence, needs assessment and purchasing. J. Publ. Health Med., 16, 2935.[Abstract]
Jones, H.W. Jr, Out, H.J., Hoomans, E.H.M. et al. (1997) Cryopreservation: the practicalities of evaluation. Hum. Reprod., 12, 15221524.[Abstract]
Mantovani, L.G., Belisari, A. and Szucs, T.D. (1999) Pharmaco-economic effects of in-vitro fertilization in Italy. Hum. Reprod., 14, 953958.
Meniru, G.L. (1999) What are the benefits of recombinant gonadotrophins? Is Puregon a `good' or `super' drug? Hum. Reprod., 14, 14091411.
Out, H.J., Mannaerts, B.M.J.L., Driessen, S.G.A.J. et al. (1995) A prospective, randomized, assessor-blind, multicentre study comparing recombinant and urinary follicle-stimulating hormone (Puregon versus Metrodin) in in-vitro fertilization. Hum. Reprod., 10, 25342540.[Abstract]
Out, H.J., Mannaerts, B.M.J.L., Driessen, S.G.A.J. et al. (1996) Recombinant follicle stimulating hormone (rFSH, Puregon) in assisted reproduction: more oocytes, more pregnancies. Results from five comparative studies. Hum. Reprod. Update, 2, 162171.
Out, H.J., Braat, D.D.M., Lintsen, B.M.E. et al. (2000) Increasing the daily dose of recombinant follicle-stimulating hormone (Puregon®) does not compensate for the age-related decline in retrievable oocytes after controlled ovarian hyperstimulation. Hum. Reprod., 15, 2935.
O'Dea, L. and Banks, K. (1996) Metrodin® and Metrodin XPTM: comparative dynamics of folliculogenesis and the estrogen response. Fertil. Steril., 66 (Suppl. 1), S190.
Philips, Z., Barraza-Llorens, M. and Posnett, J. (2000) Evaluation of the relative cost-effectiveness of treatment for subfertility. Hum. Reprod., 15, 96106.[Medline]
RCOG (1998a) The initial investigation and management of the infertile couple. Royal College of Obstetricians and Gynaecologists.
RCOG (1998b) The management of infertility in secondary care. Royal College of Obstetricians and Gynaecologists.
RCOG (2000) The management of infertility in tertiary care. Royal College of Obstetricians and Gynaecologists.
Templeton, A., Fraser, C. and Thompson, B. (1991) Infertility-epidemiology and referral practice. Hum. Reprod., 6, 13911394.[Abstract]
Templeton, A., Morris, J.K. and Parslow, W. (1996) Factors that affect outcome of in vitro fertilization treatment. Lancet, 348, 14021406.[ISI][Medline]
Torgerson, D.J. and Spencer, A. (1996) Marginal costs and benefits. Br. Med. J., 312, 3536.
Van Loon, J.M. (1998) Cost-effectiveness analysis of Puregon in comparison to Metrodin HP in in vitro fertilisation (IVF) in the Netherlands. Fertil. Steril., 70 (suppl. 1), S351.
Weinstein, M.C. and Fineberg, H.V. (1980) Clinical Decision Analysis. W.B. Saunders Co., Philadelphia.
accepted on October 4, 2001.