NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Women's and Children's Health (UK). Fertility: Assessment and Treatment for People with Fertility Problems. London (UK): RCOG Press; 2004 Feb. (NICE Clinical Guidelines, No. 11.)

Cover of Fertility

Fertility: Assessment and Treatment for People with Fertility Problems.

Show details

Appendix BEconomic models

B.1. Aim of the economic models

The purpose of the economic modelling was to synthesise the estimates of the costs and clinical effectiveness of assisted reproduction for couples seeking treatment for fertility problems after initial investigation. The assisted reproduction techniques for which sufficient data were available to construct models were IVF alone and IVF with ICSI. The economic analysis focused on the effect of age on the cost-effectiveness of IVF and ICSI and the cost effectiveness of these treatments according to the number of previous unsuccessful cycles. Different scenarios were explored using sensitivity analysis since published evidence reported a range of estimates for several important parameters.

B.2. Structure of the economic models

In vitro fertilisation treatment

Two separate models were constructed in order to estimate age-specific and cycle-specific costs per live birth. The models had to be structured differently because different forms of data were available in relation to age and number of cycles.

Age-specific model

The model based upon age was structured so that couples were offered up to six fresh cycles of IVF treatment. This model was based on age-specific success rates obtained from the HFEA (see Chapter 11, Tables 11.1 to 11.5). The lowest age used in the economic mode was 24 years because below this age there were fewer than 100 treatment cycles (see Table 11.1). For each unsuccessful fresh cycle, couples would be offered up to two attempts at frozen embryo transfer. It is assumed that, on average, one-third of couples whose fresh IVF treatment cycles are unsuccessful will have enough viable embryos for two attempts at frozen embryo transfer. This model also assumed that live birth rates were constant for each treatment cycle. The structure of the model is presented in Figure B.1, which, for the purposes of illustration, shows only one of the six potential fresh cycles of IVF treatment. The outcomes of each IVF cycle with fresh or frozen embryo transfer are:

Figure B.1. Structure of the in vitro fertilisation treatment model for deriving age-specific cost per live birth; the figure shows one of the six potential fresh cycles of IVF treatment for the model and up to two frozen embryo transfers (ET) and uses constant live birth rates for different cycles.

Figure B.1

Structure of the in vitro fertilisation treatment model for deriving age-specific cost per live birth; the figure shows one of the six potential fresh cycles of IVF treatment for the model and up to two frozen embryo transfers (ET) and uses constant live (more...)

  • a live birth (in which case treatment ceases)
  • an ectopic pregnancy
  • a miscarriage
  • no pregnancy.

The options for couples without a live birth are:

  • to discontinue treatment
  • to attempt a frozen embryo transfer
  • to proceed straight to the next fresh cycle of IVF treatment if there are no embryos suitable for frozen embryo transfer.

The model assumed that no couples would choose to discontinue treatment until they has used up all embryos suitable for frozen embryo transfer. The purpose of estimating costs per live birth from up to to six fresh cycles of IVF treatment was to explore the impact of offering treatment beyond the three fresh cycles that have been shown to be of similar clinical effectiveness (see Chapter 11).

The model also allowed for the possibility of OHSS but it was assumed that having OHSS would not affect the outcome of IVF treatment. A detailed description of the clinical effectiveness data used in this model is presented in Table B.1. The discontinuation rates used in the model were estimated in studies based on experience in the independent sector, which may be higher than those that would occur if couples were not paying for treatment themselves.

Table B.1. Clinical effectiveness data used in the in vitro fertilisation treatment model for deriving age-specific cost per live birth.

Table B.1

Clinical effectiveness data used in the in vitro fertilisation treatment model for deriving age-specific cost per live birth.

There is little robust clinical evidence to determine whether any long-term adverse outcome for the couple are associated with IVF treatment and so long-term consequences of treatment were not included in our models. Such consequences would include the potential costs to people with fertility problems in terms of psychological ill health relating to waiting for treatment and the stress associated with assisted reproduction, irrespective of the outcome of treatment.

Cycle-specific models

The models based on the number of cycles were structured so that couples were offered three or four fresh cycles of IVF treatment and no frozen embryo transfers. Two models were used because two data sets with different structures were available.

The first cycle-specific model was based on live birth rates by number of previous unsuccessful cycles obtained from the HFEA (see Table 11.6) This dataset included estimates for up to four fresh (not frozen) cycles of treatment (see Section 11.4). The dataset did not include miscarriage or ectopic pregnancy rates by number of previous unsuccessful cycles. However, overall miscarriage rates and ectopic pregnancy rates (irrespective of the number of previous treatment cycles) were available from the HFEA and these were used in this model (see Tables 11.4 and 11.5, respectively). The structure of the model is presented in Figure B.2, which shows all four potential fresh cycles of IVF treatment. The potential outcomes of each IVF cycle are:

Figure B.2. Structure of the in vitro fertilisation treatment model for deriving cycle-specific cost per live birth; the figure shows all four potential fresh cycles of IVF treatment for the model, and uses different live birth rates for different cycles.

Figure B.2

Structure of the in vitro fertilisation treatment model for deriving cycle-specific cost per live birth; the figure shows all four potential fresh cycles of IVF treatment for the model, and uses different live birth rates for different cycles.

  • a live birth (in which case treatment ceases)
  • an ectopic pregnancy
  • a miscarriage
  • no pregnancy.

The options for couples without a live birth are:

  • to discontinue treatment
  • to proceed straight to the next fresh cycle of IVF treatment.

A detailed description of the clinical effectiveness data used in this model is presented in Table B.2.

Table B.2. Clinical effectiveness data used in the first in vitro fertilisation treatment model for deriving cycle-specific cost per live birth.

Table B.2

Clinical effectiveness data used in the first in vitro fertilisation treatment model for deriving cycle-specific cost per live birth.

The second cycle-specific model was based on live birth rates by number of previous unsuccessful cycles obtained from the Oxford Fertility Unit (see Table 11.7) This dataset included estimates for up to three fresh (not frozen) cycles of treatment for two different age group (under 39 years versus 39 years and over; see Chapter 11, Section 11.2). The dataset also included miscarriage rates but not ectopic pregnancy rates. The possibility of ectopic pregnancy was, therefore, not included in this model. The structure of the model is similar to that presented in Figure B.2, except that only three fresh cycles of IVF treatment are modelled, and the possibility of ectopic pregnancy is not considered. A detailed description of the clinical effectiveness data used in this model is presented in Table B.3.

Table B.3. Clinical effectiveness data used in the second in vitro fertilisation treatment model for deriving cycle-specific cost per live birth.

Table B.3

Clinical effectiveness data used in the second in vitro fertilisation treatment model for deriving cycle-specific cost per live birth.

Intracytoplasmic sperm injection

We used one model to estimate the cost per live birth of IVF plus ICSI. This model had the same basic structure as the age-specific model for IVF treatment (that is, it included fresh and frozen treatment cycles; see Figure B.1). However, no data were available on the clinical effectiveness of ICSI and so this model was based on overall (not age-specific) success rates for IVF treatment obtained from the HFEA (see Tables 11.1 to 11.5). A detailed description of the clinical effectiveness data used in this model is presented in Table B.4.

Table B.4. Clinical effectiveness data used in the intracytoplasmic sperm injection model for deriving overall cost per live birth.

Table B.4

Clinical effectiveness data used in the intracytoplasmic sperm injection model for deriving overall cost per live birth.

B.3. Costs used in the economic models

Treatment costs were estimated using a variety of published and unpublished sources of data. Table B.5 summarises the cost data used in the model. NHS reference costs were used where no published research papers reporting specific costs could be identified. NHS reference costs are second-best cost estimates since they show wide variation and are not derived from detailed bottom-up calculation of the true inputs into a service. The best cost data are derived from UK-based economic evaluation studies that report resource use and unit costs as well as a cumulative mean cost estimate. Such data were not available for many of the estimates used in the model.

Table B.5. Cost data used in the in vitro fertilisation treatment models.

Table B.5

Cost data used in the in vitro fertilisation treatment models.

A range of estimates for the cost of an IVF cycle was obtained from different sources. A web-published review by the voluntary organisation, Fertility Confidential, reported in 2002 that the average charge for IVF treatment in the UK at the 71 fee-paying clinics was £1,737 per treatment cycle, with the lowest reported charge around £1,000 and the highest around £2,500. The HFEA reported on its website that the cost of an IVF cycle is around £1,771 excluding drug costs. The HFEA also reported on its website that the cost of an ICSI cycle is £1,936 (without drugs).

A UK study1143 reported the cost of a stimulated cycle of IVF to be around £4,250 and a natural cycle to be around £898. An earlier study reported the cost per couple of IVF to range from £1,786 to £5,749 and a single cycle to cost £1,100.13 Another UK study undertaken earlier in the 1990s reported a cost of IVF to be £1,005 for stimulated IVF.707

In our models, we have explored the cost per live birth of IVF at the lower and higher ranges of cost estimates. We have also estimated the cost per live birth with and without the costs of IVF drugs since gonadotrophins can increase the cost per cycle by around £500 to 1,000, depending on the drugs used. We used three costs in our models. The baseline cost was £2,771 (£1,771, which includes the costs associated with health services use and counselling, plus £1,000 for drugs); a lower value of £1,771 (the cost without drugs); and a higher value of £3,500 (£2,500, which was the highest value reported in the Fertility Confidential survey, plus £1,000 for drugs). We also explored the impact of assuming an even higher cost of £5,000 per IVF cycle (£4,000 plus £1,000 for drugs). The cost for an ICSI cycle in our model was £2,936 (£1,936, plus £1,000 for drugs).

The costs of miscarriage and ectopic pregnancy after IVF treatment could not be estimated from the published literature and so we used NHS reference costs relating to miscarriage and upper genital tract (intermediate procedures) for ectopic pregnancy.

A detailed description of the cost data used in this model is presented in Table B.5.

B.4. Sensitivity analysis

Sensitivity analyses were undertaken to explore the effects on the total cost and cost per live birth of changing the following parameters in the models:

  • the cost (without drugs) per cycle of IVF/ICSI
  • the number of couples who would choose to discontinue treatment rather than starting a new fresh cycle
  • the rate and cost of OHSS per fresh cycle of IVF
  • the source of clinical effectiveness data (HFEA or Oxford Fertility Unit).

B.5. Results

In vitro fertilisation treatment

Age-specific model

Age-specific costs per live birth using the three cost estimates (baseline, lower and upper) for IVF treatment and an OHSS incidence rate of 0.2% are shown in Figure B.3. The figure shows that the costs per live birth are very similar for ages 24 years to 33 years, after which they rise steeply with increasing age. Detailed tables of costs for three specific ages (24 years, 35 years and 39 years) using the baseline cost of IVF treatment (£2,771) are presented in Tables B.6, B.7 and B.8, respectively. The tables show that the costs per live birth were £11,917 at 24 years, £12,931 at 35 years and £20,056 at 39 years. The total costs after three cycles of treatment based on 1000 couples at the start of treatment and using the baseline cost of IVF treatment and a discontinuation rate of 17.7% were £6.2 million in women aged 24 years, £6.3 million in women aged 35 years and £6.9 million in women aged 39 years. The percentage of couples who achieved a live birth after three cycles of treatment were 52% at 24 years, 49% at 35 years and 34% at 39 years.

Figure B.3. Age-specific cost per live birth using three cost estimates for a cycle of in vitro fertilisation treatment.

Figure B.3

Age-specific cost per live birth using three cost estimates for a cycle of in vitro fertilisation treatment.

Table B.6. Cost per live birth for women aged 24 years using baseline cost for a cycle of in vitro fertilisation treatment (1000 couples).

Table B.6

Cost per live birth for women aged 24 years using baseline cost for a cycle of in vitro fertilisation treatment (1000 couples).

Table B.7. Cost per live birth for women aged 35 years using baseline cost for a cycle of in vitro fertilisation treatment.

Table B.7

Cost per live birth for women aged 35 years using baseline cost for a cycle of in vitro fertilisation treatment.

Table B.8. Cost per live birth for women aged 39 years using baseline cost for a cycle of in vitro fertilisation treatment (1000 couples).

Table B.8

Cost per live birth for women aged 39 years using baseline cost for a cycle of in vitro fertilisation treatment (1000 couples).

The sensitivity analyses using lower and higher costs for IVF treatment (£1,771 and £3,500, respectively) resulted in costs per live birth of £8,103 and £14,697 at 24 years, £8,800 and £15,943 at 35 years and £13,723 and £24,673 at 39 years. The sensitivity analyses using the baseline cost of IVF treatment and the higher discontinuation rate (50%) resulted in total costs after three cycles of IVF treatment based on 1000 couples at the start of treatment of £4.7 million for women aged 24 years, £4.8 million for women aged 35 years and £5.0 million for women aged 39 years.

Cycle-specific models

Cycle-specific costs per live birth using the baseline cost estimate for IVF treatment and HFEA clinical effectiveness data are shown in Table B.9. The cost per live birth in the first cycle of IVF treatment was £15,281. The corresponding costs for the second, third and fourth cycles of IVF treatment were £16,169, £14,793 and £14,336, respectively. These costs reflect the varying live birth rates by number of previous unsuccessful IVF cycles (see Table 11.6). The total cost at the end of three cycles based on 1000 couples at the start of treatment was £5.9 million, with 38% of couples achieving a live birth.

Table B.9. Cost per live birth by cycle of in vitro fertilisation treatment using baseline cost estimate and Human Fertilisation and Embryology Authority clinical effectiveness rates.

Table B.9

Cost per live birth by cycle of in vitro fertilisation treatment using baseline cost estimate and Human Fertilisation and Embryology Authority clinical effectiveness rates.

The sensitivity analyses using the lower costs for IVF treatment (£1,771) resulted in costs per live birth of £9,787 for the first cycle, £10,356 for the second cycle, £9,474 for the third cycle, and £9,181 for the fourth cycle. The corresponding costs per live birth using the higher cost for IVF treatment (£3,500) were £19,287, £20,408, £18, 671 and £18,094. Using an even higher cost of £5,000 for IVF treatment resulted in costs per live birth of £27,528, £29,129, £26,650 and £25,826).

The sensitivity analyses using the baseline cost of IVF treatment and the higher discontinuation rate (50%) resulted in a total cost after three cycles of IVF treatment based on 1000 couples at the start of treatment of £4.4 million, with 28% of couples achieving a live birth.

Cycle-specific costs per live birth using the baseline cost estimate for IVF treatment and Oxford Fertility Unit clinical effectiveness data are shown in Tables B.10 (women aged less than 39 years) and B.11 (women aged 39 years and over). For women aged less than 39 years, the cost per live birth in the first cycle of IVF treatment was £11,694. The corresponding costs for the second and third cycles of IVF treatment were £11,548 and £12,758. For women aged 39 years and over, the costs per live birth were £27,611 for the first cycle of treatment, £28,938 for the second cycle of treatment and £12,835 for the third cycle of IVF treatment. These costs reflect the varying live birth rates by number of previous unsuccessful IVF cycles (see Table 11.7); the cost per live birth for the third cycle of treatment is not reliable because of the small number of cycles on which the live birth rate was based. For women aged less than 39 years, the total cost at the end of three cycles based on 1000 couples at the start of treatment was £5.6 million, with 48% of couples achieving a live birth. For women aged 39 years and over, the total cost at the end of three cycles, based on 1000 couples at the start of treatment, was £6.4 million, with 29% of couples achieving a live birth. These costs are consistent with those obtained using the HFEA clinical effectiveness data, reflecting the differences in live birth rates according to the woman’s age, rather than variations in live birth rates between clinics.

Table B.10. Cost per live birth by cycle of in vitro fertilisation treatment for women aged less than 39 years using baseline cost estimate and Oxford Fertility Unit clinical effectiveness rates.

Table B.10

Cost per live birth by cycle of in vitro fertilisation treatment for women aged less than 39 years using baseline cost estimate and Oxford Fertility Unit clinical effectiveness rates.

Table B.11. Cost per live birth by cycle of in vitro fertilisation treatment for women aged 39 years and over using baseline cost estimate and Oxford Fertility Unit clinical effectiveness rates.

Table B.11

Cost per live birth by cycle of in vitro fertilisation treatment for women aged 39 years and over using baseline cost estimate and Oxford Fertility Unit clinical effectiveness rates.

Changing the incidence rate and cost of treating ovarian hyperstimulation syndrome

Using the baseline IVF model (live birth rates of 17.7% for fresh embryo transfers and 11.5% for frozen embryo transfers, ectopic pregnancy rate of 0.5%, miscarriage rate of 2.7%, discontinuation rate of 17.6%, cost per fresh IVF cycle of £2,771, cost per frozen embryo transfer £666, OHSS incidence rate of 0.2%, and OHSS treatment cost of £800), the overall cost per live birth was £13,301. Increasing the OHSS incidence rate to 1% led to an overall cost per live birth of £13,328, while decreasing the cost of treating OHSS to £350 (a UK private sector estimate) led to an overall cost per live birth of £13,309. These results indicate that the rate and cost of OHSS are not important factors in the overall cost-effectiveness of IVF. The lack of robust cost data for the treatment of OHSS is less important than the lack of robust data for the cost of an IVF cycle.

International comparison

The cost-effectiveness ratios (cost per live birth) presented here can be compared with cost-effectiveness ratios reported for other countries using RCT clinical effectiveness evidence. A review of this evidence shows far higher cost-effectiveness ratios (cost of IVF per delivery) in the USA (as might be expected), but similar results in Scandinavian countries.1145 The data reported below are for the year 1994:

  • Sweden £10,295
  • Denmark £11,858
  • Norway £13,413
  • Finland £11,211
  • Iceland £7,400

Intracytoplasmic sperm injection

The cost per live birth for couples undergoing ICSI using the baseline cost of ICSI treatment (£2,936) and an OHSS incidence rate of 0.2% is presented in Table B.12. The table show that the cost per live birth was £14,002. The total cost after three cycles of ICSI treatment was £6.5 million, with 48% of couples achieving a live birth. At a lower cost per ICSI treatment (£1936, which excludes drugs) the cost per live birth was £9,056.

Table B.12. Cost per live birth using baseline cost for intracytoplasmic sperm injection.

Table B.12

Cost per live birth using baseline cost for intracytoplasmic sperm injection.

The cost of the PESA/TESA procedure can add to the average cost of a live birth for when sperm retrieval is necessary before ICSI. We did not identify any studies that reported the cost per live birth of ICSI with PESA/TESA in the UK, nor any that reported the cost of the procedure alone. A German study published in 2000 reported a hospital fee of 369 euros (around £250) at one institution,1146 but no further data were provided to indicate which health care resources were included in the cost. In the UK, private fertility clinics charge a range of prices for PESA/TESA. The lower end of the range of prices published on the internet in November 2003 was £625 and the higher end was around £1,400. A voluntary organisation reported a guide price £1,100 for couples seeking treatment with PESA/TESA (www.oneinsix.com, accessed 4 November 2003).

Table B.13Typical treatment schedule and cost of antagonists and agonists in ovulation induction

Generic/proprietary nameDuration, route and treatment scheduleDose/dayDays of treatment per cycle (n)Price and unit advertised in BNF181 and total cost of antagonist or agonistMean dose gonadotrophinGonadotrophin days (n)Price and unit advertised in BNF181 and total cost of gonadotrophinTotal price (agonist/antagonist plus gonadotrophin)
Antagonists e.g. cetrorelix (brand name: Cetrotide®)Short Injection

5 days of gonadotrophin and then 5 days of gonadotrophin plus antagonist

(total 10 days)
250 micrograms5 daysNet price 250-microgram vial = £24.00

£24 x 5 days = £120
1500 iu (3000 iu for poorly responding patients)

150–300 iu/day, depending on patient characteristics
10

starting on 1st day of cycle
150-unit ampoule of gonal-f = £52.50

150-unit vial of puregon = £54.43

At 150 iu/day:

£52.50 x 10 days = £525 for gonal-f

£54.43 x 10 days = £544.30 for puregon

At 300 iu/day
(high dose):

2 x 150 units per day x 10 days = £1050 (rounded to nearest £10)
With low dose Gonal-f: £120 + £525 (Gonal-f) = £645

With high dose Gonal-f (or for 300 iu/day): £120 + £1050 = £1170

With low dose Puregon: £120 + £544.30 (Puregon) = £664.30

With high dose Puregon: £120 + £1090 = 1210

Range: £645—1170
Agonists e.g. nafarelin (brand name Synarel®)Long Nasal

Start agonist on 21st day of cycle for 2–3 weeks, then 10 days of gonadotrophin plus agonist

(total 24–31 days)
200 micrograms in each nostril, twice a day

= 800 micrograms per day
14 days (minimum duration) to 21 days (maximum duration)

plus 10 days while on gonadotrophin

= 96 doses (minimum duration)

= 124 doses (maximum duration)
200 micrograms per metered spray. Net price 30-dose unit = £32.28; 60-dose unit = £55.66

= £87.94 for 90 doses

= £111.34 for 120 doses
1500–3000 iu total

or

150–300 iu/day
10Minimum duration of agonist, high dose of gonadotrophin: £88 + 1050 = £1138

Maximum duration of agonist, high dose of gonadotrophin: £111 + £1050 = £1162

Minimum duration, low dose gonadotrophin: £88 + £535 (mean cost) = £623

Maximum duration agonist, high dose of gonadotrophin: £111 + £535 = £646

Range: £623–1138
Copyright © 2004, National Collaborating Centre for Women’s and Children’s Health.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

Bookshelf ID: NBK45930
PubReader format: click here to try

Views

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...