SEARCH

SEARCH BY CITATION

Keywords:

  • Cost-effectiveness analysis;
  • decision analytic modelling;
  • embryo transfer;
  • in vitro fertilisation;
  • paediatric outcomes

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

Objective  The objective of this study was to assess the cost-effectiveness of different embryo transfer strategies for a single cycle when two embryos are available, and taking the NHS cost perspective.

Design  Cost-effectiveness model.

Setting  Five in vitro fertilisation (IVF) centres in England between 2003/04 and 2004/05.

Population  Women with two embryos available for transfer in three age groups (<30, 30–35 and 36–39 years).

Methods  A decision analytic model was constructed using observational data collected from a sample of fertility centres in England. Costs and adverse outcomes are estimated up to 5 years after the birth. Incremental cost per live birth was calculated for different embryo transfer strategies and for three separate age groups: less than 30, 30–35 and 36–39 years.

Main outcome measures  Premature birth, neonatal intensive care unit admissions and days, cerebral palsy and incremental cost-effectiveness ratios.

Results  Single fresh embryo transfer (SET) plus frozen single embryo transfer (fzSET) is the more costly in terms of IVF costs, but the lower rates of multiple births mean that in terms of total costs, it is less costly than double embryo transfer (DET). Adverse events increase when moving from SET to SET+fzSET to DET. The probability of SET+fzSET being cost-effective decreases with age. When SET is included in the analysis, SET+fzSET no longer becomes a cost-effective option at any threshold value for all age groups studied.

Conclusions  The analyses show that the choice of embryo transfer strategy is a function of four factors: the age of the mother, the relevance of the SET option, the value placed on a live birth and the relative importance placed on adverse outcomes. For each patient group, the choice of strategy is a trade-off between the value placed on a live birth and cost.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

The chances of pregnancy and live birth following in vitro fertilisation (IVF) have risen steadily over the past two decades. IVF accounts for over 1:100 births in many European countries and 1.4% of births in the UK.1 Together with the undoubted benefits of assisted reproductive technologies in helping infertile couples to have families, there are several negative consequences, the most important being the large increase in the number of multiple births seen after IVF. Although there has been a fall in triplet pregnancies since 1998, the number of twin births in UK resulting from assisted reproductive technologies continues to rise.2 The perinatal morbidity and mortality of twin pregnancy was reviewed by a Human Fertility and Embryology Authority Expert Group on Single Embryo Transfer in 2006.3 The expert group has advised that UK policymakers consider a move towards increasing use of single embryo transfer in IVF to reduce the unacceptable morbidities that follow twinning.

The Northern European countries in which ‘single embryo transfer policies’ have most successfully been introduced are also those which have a more generous amount of State support for assisted reproductive technologies than is currently seen in UK. Although the National Institute for Clinical Excellence (NICE) guidelines on management of subfertility recommended provision of three full cycles of IVF to women less than 40 years who require this treatment, little progress has been made in implementing this guidance.4 One recent survey has shown that of those Primary Care Trusts (PCTs) that offer any treatment at all, most offer one cycle and only 9% offer two cycles of IVF.5

One means by which UK Primary Care Trusts might be encouraged to provide the services offered to the nation by the Minister for Health in 2004 would be to make explicit the cost and outcomes associated with different embryo transfer strategies. A number of studies have attempted to assess the cost of multiple births to the NHS by modelling the obstetric and postnatal costs of complications of multiparity.6,7 However, economic evaluations of single embryo transfer undertaken to date, with one exception,8 have omitted maternal and paediatric adverse events.9–13

Estimating the cost-effectiveness of different strategies, including the costs of adverse events, would help commissioners plan for the consequences of meeting the NICE targets, in terms of costs, birth rates and adverse event rates. As part of a regional initiative—the Evidence-Based Commissioning Collaboration—we devised a decision analytic modelling approach to evaluate this problem in terms of antenatal, maternal, neonatal and paediatric costs, birth rates and adverse event rates. The approach also allowed us to assess how the optimal strategy changes with maternal age, the availability of embryo storage and the number of embryos available (although this final issue is not addressed in this paper). This study, therefore, assesses the cost-effectiveness of different embryo transfer strategies for a single cycle when two embryos are available and takes the NHS cost perspective.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

There are two main aspects to consider when an IVF treatment is applied: (a) the effectiveness of the embryo transfer(s) and (b) the costs/outcomes related to the resultant birth outcomes. These are shown schematically in Figure 1. The model defines the outcomes as: no transfer, no live birth, singleton live birth, twin live birth or triplet or higher multiple live birth. When summarising cost-effectiveness in an incremental cost per live birth, live birth means any type of birth (i.e. singleton, twin and triplet all receive the same weighting).

image

Figure 1. Model structure.

Download figure to PowerPoint

This model was used to investigate costs and outcomes associated with different embryo transfer strategies for a varying number of embryos. A strategy was defined as a combination of single/double, fresh/frozen embryo transfer within a cycle.14 The strategy was based on the number of available embryos at the beginning of the cycle. In this paper, we evaluate the situation where two embryos are available. The main comparison is between the transfer of two fresh embryos, which is referred to as double embryo transfer (DET), and single fresh embryo transfer (SET), which if unsuccessful leads to the transfer of a previously frozen single embryo transfer (fzSET).

Antenatal, maternal, neonatal and paediatric costs together with adverse outcomes are all based on the type of live birth, defined as being singleton, twin or triplet and higher. The individual components of costs were not modelled here as previous work was used to define a quantum of cost associated with each birth outcome (see ‘Data’). These costs were assessed up to 5 years after delivery. The adverse outcomes were specified through consultation with commissioners and included premature delivery, cerebral palsy, neonatal intensive care unit (NICU) admissions and length of stay. These were calculated by rates being applied to the different birth outcomes (see ‘Data’).

Data

Birth probabilities

Observational data were used in preference to that produced in randomised controlled trials (RCTs). The reasons for this are that most trials reflect a selected patient group (e.g. only those who have two high-quality embryos available), do not allow separate age-based analyses and do not reflect UK practice.12,15,16 Seven IVF centres in the region were approached for embryo transfer data relating to 2003/04 and 2004/05. In summary, the data were single and double, fresh and frozen transfers, for three age groups (<30, 30–35 and 36–39 years), together with birth outcomes. Data were available for five centres, which in total covered 5508 fresh cycles and 1295 frozen cycles.

Due to the very low numbers of elective single embryo transfers, that is where a single embryo is transferred regardless of the number of embryos available, the single transfer data provided by the centres represent a poor prognosis sample. In essence, they describe transfers where only one viable embryo is available. This is reflected by the low birth probabilities in Table 1. Consequently, SET and fzSET birth rates were imputed by fitting the collaborative model of Matorras et al.17 to the DET and fzDET regional data (Table 1).

Table 1.  Summary of birth probabilities
 Regional data* (%)Regional data with imputed values for SET and fzSET17 (%)
  • *

    Calculated from regional data of women less than the age of 40 years; n = 5508 fresh cycles and 1295 frozen cycles.

Fresh DET26.426.4
Elective SET10.014.5
Frozen double embryo transfer18.618.6
Elective fzSET5.78.4

Individual birth outcome probabilities for regional data are set out in Table 2. The imputed SET and fzSET birth rates are split into different birth outcomes by multiplying the overall live birth rate by the ratio of live birth outcomes (for SET and fzSET) taken from published trials.12,15,16

Table 2.  Live birth outcomes
Age group (years)Transfer typeCycles started*Drop-out rate (%)No live birthSingletonTwinsTriplets or above
Probability given transfer occurs
  • fzDET, frozen double embryo transfer.

  • *

    The number of SET cycles is 48% of the number of DET cycles. This was based on the elicitation exercise described under ‘Data’.

<30DET83314.50.6930.2280.0770.002
SET40014.50.8290.1680.0030.000
fzDET1498.80.7920.1610.0470.000
fzSET728.80.9000.1000.0000.000
30–35DET260714.70.7160.2160.0680.000
SET125114.70.8440.1540.0020.000
fzDET55614.60.8110.1560.0320.000
fzSET26714.60.9150.0850.0000.000
36–39DET168816.70.7870.1740.0390.001
SET81016.70.8850.1130.0020.000
fzDET41214.70.8250.1550.0190.000
fzSET19814.70.9220.0780.0000.000

Within the model, the birth outcomes were described by a Dirichlet distribution using counts of events rather than the associated probabilities. For DET and fzDET, the raw regional data were used. For SET and fzSET, the underlying data do not exist as the birth probabilities are constructed from other data sources (as described in Table 1). Consequently, an estimate of the uncertainty surrounding the SET and fzSET live birth rate was elicited from a panel of six fertility experts (see Acknowledgements) in terms of a 95% confidence interval (CI). The Dirichlet distribution was then defined such that the event numbers produced a 95% CI around the birth rate equal to that elicited from expert opinion. From this exercise, the CI elicited was 0.16–0.23 (central probability of 0.19 for women less than the age of 30 years), which implied a sample size for SET that was 48% of that for DET. A continuity correction summing to unity was added to the event counts for each type of embryo transfer.

Other outcomes

Rates of adverse outcomes were derived from previous work. The outcomes and sources used were premature birth,18 NICU admissions and days and cerebral palsy.7,19 Distributions around these point estimates were not specified in the model.

Costs

IVF costs were taken as the mean of six local providers. Antenatal, neonatal and maternal costs associated with IVF by multiplicity are derived from Ledger et al.7 Adjustments were made to the original estimates by using updated Reference Costs and Hospital Episode Statistics and by using obstetric per diem unit costs instead of generic per diem rates for any hospitalisations.20,21 Paediatric hospital costs (defined as those following the initial neonatal admission) up to 5 years are derived from Henderson et al.,6 by excluding neonatal admission from their estimates using Ledger’s figures and updating to 2003/04 prices using the Hospital and Community Health Services Index.22 Unit costs are summarised in Table 3. Costs were discounted at 3.5% per annum where appropriate.23

Table 3.  Costs associated with IVF treatment (2003/04)
ResourceMean costSourceDistributionStandard error
  • EBCC, Evidence-based Commissioning Collaboration; HDU, high dependency unit; SCBU, special care baby unit.

  • *

    The cost of a fresh embryo transfer is based on a women starting treatment, regardless of outcome. Contracts with providers vary in the amount that is ‘refunded’ in the event of a transfer not occurring. However, this cost is the same across all strategies and therefore will not impact on the incremental results.

Fresh transfer cycle2242*EBCC centresNormal161
Frozen transfer cycle (first)1064EBCC centresNormal145
Frozen transfer cycle (subsequent)790EBCC centresNormal113
Maternal care singleton4297Adapted from Ledger et al.7Lognormal1418
Maternal care twins7915Adapted from Ledger et al.7Lognormal2612
Maternal care triplets15,130Adapted from Ledger et al.7Lognormal4993
NICU/HDU/SCBU singleton207Adapted from Ledger et al.7Lognormal68
NICU/HDU/SCBU twin4026Adapted from Ledger et al.7Lognormal1329
NICU/HDU/SCBU triplets27,825Adapted from Ledger et al.7Lognormal9182
Paediatric care singleton2071Henderson et al.6Lognormal683
Paediatric care twin7898Henderson et al.6Lognormal2606
Paediatric care triplet20,100Henderson et al.6Lognormal6633

Distributions for the IVF costs were estimated directly from the provider estimates (Table 3). Distributions for the other unit costs were thought to be skew based on the summary statistics given in NHS Reference Costs and the Henderson study.6,20 A log normal distribution was assumed and SDs identified such that the resultant distributions approximately fit the Reference Cost and Henderson summary statistics (Table 3).

Analysis

Analyses were undertaken using the regional data for SET+fzSET versus DET. Additionally, we assessed the impact of not freezing the additional embryo following SET in an additional set of analyses. This approach is adopted by many providers in the UK and results in the destruction of any additional embryos.3

Mean costs and outcomes were estimated, and results ranked so as to allow the estimation of incremental cost-effectiveness ratios (ICERs). ICERs represent the additional cost associated with an additional unit of outcome—in this case, the additional cost per additional live birth—and are used as a summary measure of value for money. A probabilistic sensitivity analysis was undertaken by specifying distributions around the model inputs. Four thousand samples were drawn to construct cost-effectiveness acceptability curves.24 These curves show the probability that the chosen strategy is cost-effective at different valuations placed on the outcome measure, which in this case are live births.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

For all age groups, the ranking of the strategies in terms of effectiveness is the same; SET is least effective, followed by SET+fzSET and DET being the most effective (Table 4). SET+fzSET is the most costly in terms of IVF costs, but the lower costs associated with multiple births mean that in terms of total costs, it is less costly than DET.

Table 4.  Summary cost-effectiveness results for different embryo transfer strategies
Age (years)StrategyResults per cycleResults per 1000 cycles
Live birth rateIVF costs (£2003/04)Other costs (£2003/04)Total cost (£2003/04)ICER* (£2003/04)Cumulative multiple birthsCumulative cerebral palsyCumulative premature birthsCumulative NICU** admissionsCumulative NICU days
  • *

    ICER: incremental cost per additional live birth.

  • **

    NICU: not including high dependency unit and special care baby unit.

<30SET0.171224211933435 3021834
SET+fzSET0.24329701801477118,46371311378
DET0.30822423193543510,3398026637250
30–35SET0.156224210703312 3019726
SET+fzSET0.21829831510449319,24040261042
DET0.28422422794503681846815931191
36–39SET0.11522427923034 2014520
SET+fzSET0.17230061213421921,0293021840
DET0.213224219644206−3174014020123

In terms of the ICER, three issues are worth considering (Table 4). First, the ICER for SET+fzSET over SET increases with age, from around £18,000 (ages <30 years) to around £21,000 (ages 36–39 years). Second, for less than 30 and 30–35 years of age groups, DET exhibits weak dominance over SET+fzSET; in other words, if the ICER for SET+fzSET is considered worthwhile, then the (lower) ICER for DET is more worthwhile. Third, for the age group of 36–39 years, DET exhibits strong dominance over SET+fzSET; it is more effective and less costly. Calculating an ICER against SET for this group produces a cost per live birth of £11,959.

The results also show how adverse events increase when moving from SET to SET+fzSET to DET. The outcomes associated with these events are not included in the denominator of the ICER used.

When the probabilistic results are considered, we can see how the probability of SET+fzSET being cost-effective varies with age (Figure 2). In general, the probability of SET+fzSET being cost-effective decreases with age in all age groups, if SET is not considered. When SET is included in the analysis (Figure 3), SET+fzSET no longer becomes a cost-effective option at any threshold value for any age group. Figure 3 only shows the results for less than 30 years; however, SET is less prominent in the other age groups.

image

Figure 2. Cost-effectiveness acceptability curves of SET+fzSET versus DET.

Download figure to PowerPoint

image

Figure 3. Cost-effectiveness acceptability curve for two embryo strategies including SET (age less than 30 years).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

The analyses show that the choice of embryo transfer strategy is a function of four factors: the age of the mother, the relevance of the SET option, the value placed on a live birth and the relative importance placed on adverse outcomes. Increasing age increases the probability of DET being cost-effective, as too does an increasing value placed on a live birth. Inclusion of SET as a viable treatment option effectively removes SET in cost-effectiveness terms; low valuations of live births see SET as having the highest probability of being cost-effective, while higher values favour DET. All of these conclusions from the cost-effectiveness analysis are clear, but when considering policy options, we also need to consider what weight should be attached to the adverse outcomes.

Other studies

Previous studies have looked at this issue;8,10–13 however, these have been criticised in a recent review for not using UK data on unselected women, including longer term costs, or assessing patient subgroups.25 These criticisms can also be applied to a review of single versus double embryo transfer,26 which were based around the same studies critiqued by Scotland et al. Only one has included neonatal costs within their calculations.8 The study by Kjellberg et al. is based on a previous RCT,16 which was undertaken in relatively young women (aged less than 36 years) who had produced two high-quality embryos. The high birth rates in the SET+fzSET group within the RCT are largely responsible for the high ICER relating to DET reported in the economic study (€71940, price level not reported).

While Lukassen et al.12 concluded that SET+fzSET and DET were both equally costly and effective, most other studies identified a trade-off between high birth rates and higher costs associated with DET.8,10,13 De Sutter et al.11 stand alone as using cost per child born as their summary cost-effectiveness measure, which makes DET relatively more cost-effective compared with using live births.

The other main difference between studies is the number of cycles evaluated, with only one study evaluating more than one cycle.11 The impact of subsequent cycles was also considered by the economic model underpinning the 2004 NICE Fertility Guideline,4 however, that work does not address the specific issue of type of embryo transfer.

A study by Little et al.27 is also worth consideration, although it has not been identified as an economic evaluation by previous reviews.25,26 While Little et al. do not produce cost-effectiveness ratios, they model costs and outcomes in a similar way to this study. They also improve on previous studies by looking at different age bands and including some longer term costs for cerebral palsy. Like us, they also use IVF centre data rather than trial-based data; however, their adjustment to produce realistic live birth rates for SET is based on an assumption that the rate is 50% of that for DET. Also, in common with the other studies, they also do not include a full stochastic analysis of cost-effectiveness.

No previous studies has therefore modelled the impact of age, longer term costs or the option of SET. This latter issue, is however, perhaps peculiar to the current funding situation within the UK.3 By also incorporating uncertainty into our analysis using a probabilistic sensitivity analysis, we have also addressed this further criticism of previous work.25

Weaknesses

The model developed is quite simple. Costs and adverse events are purely based on multiplicity, yet even this formulation required the birth outcomes data for SET and fzSET to be imputed due to the lack of UK data. We have factored in an element of uncertainty into the imputed figures, but more definitive results will not be possible until reliable outcomes for elective SET from routine UK practice are available. While the rates of elective SET are increasing in the UK, it is unlikely that reliable data will be available in the foreseeable future. Data from routine practice in other European countries point to our estimates being pessimistic;3 however, there are clear dangers in trying to use these figures. Higher live birth rates for SET and fzSET would have the effect of improving the cost-effectiveness of their associated strategies.

It is also worth considering that even our broad perspective with respect to costs does not capture the full impact of adverse events. Other work investigating the costs up to the age of 9 years associated with low birthweight babies who have disability highlights the potential magnitude of other costs to society. This work shows that healthcare costs are matched by those for special educational needs.28 Within their study, costs were broken down; thus, 35% neonatal costs, 6% inpatient costs, 6% outpatient costs, 1% GP costs and 51% special education needs costs.

One other element of cost is also simplified in the model. The costs of freezing embryos are included for the first year of treatment only. However, any spare embryos that are available following a successful pregnancy may be kept for several years afterwards, at a cost of around £200 per annum. In practice, this matter is a complex mix of funding issues and patient choice.

As part of our analysis, we have chosen not to use data from RCTs due to selection biases within the data that produce very high live birth rates (e.g. 43.4% for DET).16 However, by using observational data to gain greater external validity, we are potentially introducing other biases and therefore compromising internal validity. Our solution has been to use observational data for DET (where there is little selection bias) and then use these to impute birth rates for SET (where there is substantial bias). This approach produces an 11.9% point difference in live birth rates between DET and SET (Table 1), which is much closer to the differences seen in the trials than those seen in the crude regional data (Table 1). Consequently, we feel that this produces estimates that have good internal and external validity. Furthermore, uncertainty around this approach is incorporated into the results through the probabilistic sensitivity analysis.

For outcomes, we have focused on a small set of important measures. Consequently, other outcomes that may be considered important to some are excluded. For example, DET not only produces a greater number of pregnancies but also reduces the mean time to pregnancy and number of failed attempts. As the process of IVF can be stressful, reducing this time would be beneficial.29 Likewise, parental morbidity associated with caring for twins or triplets or other postnatal consequences of multiplicity are not included as summary measures.30

Other issues

The results presented here relate to the situation where two embryos are available, but more than two viable embryos are commonly produced by the stimulation cycle. These situations are slightly more complex to evaluate as more permutations of transfers are possible. Evaluations of strategies relating to three embryos have been undertaken using this approach, but the results are not presented here.

Our approach only looks at a single cycle of treatment. Pregnancy rates are fairly constant for the first three cycles of treatment for women less than the age of 40 years, and therefore, the results within each cycle will remain the same.31 However, this does not take into account the fact that differences in live birth rates will impact on the costs and outcomes of future cycles if they are made available to the woman.

Other uncertainties associated with the impact of any policy formulated are possible behavioural changes that existing providers may make in response to the new policy. Central to the embryo transfer process are several decisions that are implicit, and are assumed constant, within the model. All of these may change in response to a new funding policy.

Of most importance is the interplay between cryopreservation and quality of embryos transferred. If a greater role for frozen embryo transfer is encouraged, the number of transfers per cycle could be increased by freezing more of the low-quality embryos that would otherwise be discarded. The consequence of this would be a reduction in frozen pregnancy rates from what has been predicted. Alternatively, if reliable methods for assessing embryo transfer were available, a selective SET policy could be considered, whereby SET is only given to women with good prognosis.3

Also, the results do not give a definitive conclusion, thus allowing different commissioners to opt for different treatment strategies. A commissioner that places a low valuation on live births will opt for single embryo transfer, while another with a higher valuation may opt for DET. This will perpetuate the existing postcode lottery. Likewise, the dual market of public and private provision may cause different types of treatment to be provided within the different sectors.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

The choice of embryo transfer strategy has a profound impact on the costs, effectiveness and adverse outcomes associated with treatment. Cost-effectiveness is dependent on age and the chosen threshold value. In addition, policymakers must also consider which treatment options should be considered as viable, the relative importance of adverse outcomes and potential behavioural consequences of any policy formulation. It is important to revisit these analyses when sufficient data are available on elective SET in the UK.

Disclosure of interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

S.D., F.F.N., J.B.C., A.D., P.S. have no conflicts of interest. W.L.L. has supervised and undertaken research funded by Organon Laboratories Ltd and Serono Pharmaceuticals Ltd and have been reimbursed for conference attendance and lectures by these organisations and Ferring Pharmaceuticals Ltd. E.A.L. was the Director and Person Responsible of one of the IVF clinics that contributed data (CARE Sheffield) until December 2005.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

This work was funded by the Evidence-Based Commissioning Collaboration. Four commissioning consortia funded the Collaboration at the time of the work. These were: North Trent Commisioners (NORCom), Trent Commisioners (TRENTCom), West Yorkshire Primary Care Organisations (WYPCO) and North and East Yorkshire and North Lincolnshire Commissioners (NEYNL).

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

S.D., F.F.N., J.B.C. participated in the conception and design, acquisition of data, analysis/interpretation and drafting of the article. P.S. participated in the acquisition of data and analysis/interpretation and drafting of the article. A.D. participated in the analysis/interpretation and drafting of the article. W.L.L. participated in the conception and design, acquisition of data and analysis/interpretation and drafting of the article. E.A.L. participated in the conception and design, acquisition of data and analysis/interpretation and drafting of the article. All authors have seen and approved the final version.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References

We would like to thank the following clinicians who in addition to W.L.L. supplied us with their estimates of the uncertainty surrounding single embryo transfer: Adrian Lower, Adam Balen, Geraldine Hartshorne, Masoud Afnan and Mark Hamilton.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of interests
  9. Funding
  10. Contribution to authorship
  11. Acknowledgements
  12. References
  • 1
    Nyboe Andersen A, Erb K. Register data on Assisted Reproductive Technology (ART) in Europe including a detailed description of ART in Denmark. Int J Androl 2006;29:1216.
  • 2
    Simmons R, Doyle P, Maconochie N. Dramatic reduction in triplet and higher order births in England and Wales. BJOG 2004;111:8568.
  • 3
    Braude P. One child at a time: reducing multiple births after IVF. Report of the Expert Group on Multiple Births after IVF. London, UK: Human Fertilisation and Embryology Authority, 2006.
  • 4
    National Institute for Clinical Excellence (NICE). Fertility: Assessment and Treatment for People with Fertility Problems. Clinical guideline No. 11. London: NICE, 2004.
  • 5
    Kennedy R, Kingsland C, Rutherford A, Hamilton M, Ledger W. Implementation of the NICE Guideline—Recommendations from the British Fertility Society for National Criteria for NHS Funding of Assisted Conception. Bristol, UK: British Fertility Society, 2006.
  • 6
    Henderson J, Hockley C, Petrou S, Goldacre M, Davidson L. Economic implications of multiple births: inpatient hospital costs in the first 5 years of life. Arch Dis Child Fetal Neonatal Ed 2004;89:F5425.
  • 7
    Ledger WL, Anumba D, Marlow N, Thomas CM, Wilson EC.; Cost of Multiple Births Study Group (COMBS Group). The costs to the NHS of multiple births after IVF treatment in the UK. BJOG 2006;113:215.
  • 8
    Kjellberg AT, Carlsson P, Bergh C. Randomized single versus double embryo transfer: obstetric and paediatric outcome and a cost-effectiveness analysis. Hum Reprod 2006;21:21016.
  • 9
    Wolner-Hanssen P, Rydhstroem H. Cost-effectiveness analysis of in-vitro fertilization: estimated costs per successful pregnancy after transfer of one or two embryos. Hum Reprod 1998;13:8894.
  • 10
    Gerris J, De Sutter P, De Neubourg D, Van Royen E, Vander Elst J, Mangelschots K, et al. A real-life prospective health economic study of elective single embryo transfer versus two-embryo transfer in first IVF/ICSI cycles. Hum Reprod 2004;19:91723.
  • 11
    De Sutter P, Gerris J, Dhont M. A health-economic decision-analytic model comparing double with single embryo transfer in IVF/ICSI. Hum Reprod 2002;17:28916.
  • 12
    Lukassen HG, Braat DD, Wetzels AM, Zielhuis GA, Adang EM, Scheenjes E, et al. Two cycles with single embryo transfer versus one cycle with double embryo transfer: a randomized controlled trial. Hum Reprod 2005;20:7028.
  • 13
    Fiddelers AA, Van Montfoort AP, Dirksen CD, Dumoulin JC, Land JA, Dunselman GA, et al. Single versus double embryo transfer: cost-effectiveness analysis alongside a randomized clinical trial. Hum Reprod 2006;21:20907.
  • 14
    Department of Health. Health Secretary welcomes new fertility guidance. Press Release 2004/0069, 2004. www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4074060&chk=2mc5cA]. Accessed 11 December 2006.
  • 15
    Martikainen H, Tiitinen A, Tomas C, Tapanainen J, Orava M, Tuomivaara L, et al. One versus two embryo transfer after IVF and ICSI: a randomized study. Hum Reprod 2001;16:19003.
  • 16
    Thurin A, Hausken J, Hillensjo T, Jablonowska B, Pinborg A, Strandell A, et al. Elective single-embryo transfer versus double-embryo transfer in in vitro fertilization. N Engl J Med 2004;351:2392402.
  • 17
    Matorras R, Matorras F, Mendoza R, Rodríguez M, Remohí J, Rodríguez-Escudero FJ, et al. The implantation of every embryo facilitates the chances of the remaining embryos to implant in an IVF programme: a mathematical model to predict pregnancy and multiple pregnancy rates. Hum Reprod 2005;20:292331.
  • 18
    Helmerhorst FM, Perquin DA, Donker D, Keirse MJ. Perinatal outcome of singletons and twins after assisted conception: a systematic review of controlled studies. BMJ 2004;328:261.
  • 19
    Petterson B, Nelson KB, Watson L, Stanley F. Twins, triplets, and cerebral palsy in births in Western Australia in the 1980s. BMJ 1993;307:123943.
  • 20
    Department of Health. NHS Reference Costs. 2004. [www.dh.gov.uk/PolicyAndGuidance/OrganisationPolicy/FinanceAndPlanning/NHSReferenceCosts/fs/en]. Accessed 13 March 2006.
  • 21
    The Information Centre. Hospital episode statistics. 2004. [www.hesonline.nhs.uk/Ease/servlet/ContentServer?siteID=1937&categoryID=192]. Accessed 13 March 2006.
  • 22
    Curtis L, Netten A. Unit Costs of Health and Social Care 2005. Canterbury, UK: Personal and Social Services Research Unit, 2005.
  • 23
    Her Majesty’s Treasury. The Green Book. Appraisal and Evaluation in Central Government. London: The Stationery Office, 2003.
  • 24
    Fenwick E, Byford S. A guide to cost-effectiveness acceptability curves. Br J Psychiatry 2005;187:1068.
  • 25
    Scotland GS, McNamee P, Bhattacharya S. Is elective single embryo transfer a cost-effective alternative to double embryo transfer? BJOG 2007;114:57.
  • 26
    Fiddelers AA, Severens JL, Dirksen CD, Dumoulin JC, Land JA, Evers JL. Economic evaluations of single- versus double-embryo transfer in IVF. Hum Reprod 2007;13:513.
  • 27
    Little SE, Ratcliffe J, Caughey AB. Cost of transferring one through five embryos per in vitro fertilization cycle from various payor perspectives. Obstet Gynecol 2006;108:593601.
  • 28
    Stevenson RC, Pharoah P, Stevenson CJ, McCabe CJ, Cooke RWI. Cost of care for a geographically determined population of low birthweight infants to age 8-9 years. II. Children with disability. Arch Dis Child Fetal Neonatal Ed 1996;74:F11821.
  • 29
    Weaver SM, Clifford E, Hay DM, Robinson J. Psychosocial adjustment to unsuccessful IVF and GIFT treatment. Patient Educ Couns 1997;31:718.
  • 30
    Klock SC. Psychosocial adjustment to twins after infertility. Best Pract Res Clin Obstet Gynaecol 2004;18:64556.
  • 31
    Croucher C, Lass A, Margara R, Winston RM. Predictive value of the results of a first in-vitro fertilization cycle on the outcome of subsequent cycles. Hum Reprod 1998;13:4038.