General obstetrics
DOI: 10.1111/j.1471-0528.2009.02236.x www.bjog.org
The cost-effectiveness of outpatient (at home) cervical ripening with isosorbide mononitrate prior to induction of labour O Eddama,a S Petrou,a L Schroeder,a SS Bollapragada,b F Mackenzie,c J Norrie,d M Reid,e JE Normanf a National Perinatal Epidemiology Unit, University of Oxford, Oxford, UK b Obstetrics & Gynaecology, Division of Developmental Medicine, University of Glasgow, Faculty of Medicine, Glasgow Royal Infirmary, Glasgow, UK c Department of Obstetrics, Princess Royal Maternity, Glasgow, UK d Centre for Healthcare Randomised Trials (CHaRT), Health Services Research Unit, Aberdeen University, Aberdeen, UK e Public Health and Health Policy, Division of Community Based Sciences, University of Glasgow, Glasgow, UK f Division of Reproductive and Developmental Sciences, The University of Edinburgh, Edinburgh, UK Correspondence: O Eddama, National Perinatal Epidemiology Unit, University of Oxford, Oxford OX3 7LF, UK. Email
[email protected]
Accepted 16 April 2009. Published Online 28 May 2009.
Objectives To assess the cost-effectiveness of outpatient (at home) cervical ripening with isosorbide mononitrate (IMN) prior to induction of labour. Design Economic evaluation was conducted alongside a
randomised placebo controlled trial (the IMOP trial). Setting Large UK maternity hospital. Population A total of 350 nulliparous women with a singleton
pregnancy, cephalic presentation ‡37 weeks gestation, requiring cervical ripening prior to induction of labour. Interventions Isosorbide mononitrate (n = 177) or placebo
(n = 173) self-administered vaginally at home at 48, 32 and 16 hours prior to the scheduled time of admission for induction. Results Mean health service costs between the period of
randomisation and discharge for mother and infant were £1254.86 in the IMN group and £1242.88 in the placebo group,
generating a mean cost difference of £11.98 (bootstrap mean cost difference £12.86; 95%CI: )£106.79, £129.39) that was not statistically significant (P = 0.842). The incremental cost per hour prevented from hospital admission to delivery was £7.53. At the notional willingness to pay threshold of £100 per hour prevented from hospital admission to delivery, the probability that IMN is cost-effective was estimated at 0.67. This translated into a mean net monetary benefit of £98.13 for each woman given IMN. Conclusions Although the probability that IMN is cost-effective approaches 0.7 at seemingly low willingness to pay thresholds for an hour prevented from hospital admission to delivery, our results should be viewed in the light of the clinical findings from the IMOP trial. Keywords Economic evaluation, induction of labour, isosorbide
mononitrate, obstetrics.
Please cite this paper as: Eddama O, Petrou S, Schroeder L, Bollapragada S, Mackenzie F, Norrie J, Reid M, Norman J. The cost-effectiveness of outpatient (at home) cervical ripening with isosorbide mononitrate prior to induction of labour. BJOG 2009;116:1196–1203.
Introduction Post-term pregnancy is associated with increased rates of fetal complications, including meconium and meconium aspiration syndrome, oligohydramnios, macrosomia, fetal birth injury, non-reassuring fetal heart rate or fetal distress in labour and caesarean delivery.1 Labour induction, carried out to reduce the risk of complications in post-term pregnancies, is successful depending on the degree of cervical ripening.2 Isosorbide mononitrate (IMN) is a nitric oxide (NO) donor that can be used for cervical ripening without
1196
causing uterine contractions and the need for fetal monitoring, and therefore be safely administered at home.2,3 Randomised controlled trials (RCTs) have been published on the use of prostaglandins (PG), another cervical ripening agent, but the agent has since been declared unsafe for the outpatient (at home) context by many authorities because of adverse fetal and maternal effects.4,5 There is only one randomised controlled study, conducted in Sweden and published in early 2007, on the use of IMN for cervical ripening on an outpatient basis.6 That study found that outpatient administered IMN is clinically effective, safe and
ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology
The cost-effectiveness of outpatient cervical ripening with isosorbide mononitrate
well tolerated, although it provided no evidence of the costeffectiveness of the agent. In this paper, we present the results of an economic evaluation of outpatient (at home) cervical ripening using IMN based on the Isosorbide MOnonitrate against Placebo study (IMOP) trial. This economic evaluation has been written in conjunction with the paper reporting clinical outcomes from the IMOP trial.7
Methods Clinical study The methods for the economic evaluation build on those of the clinical study also reported in this issue of the journal.7 In brief, eligible nulliparous women with a singleton fetus were enrolled into a RCT at the Princess Royal Maternity Hospital, Glasgow, between March 2005 and December 2006. The objectives were to test the hypothesis that outpatient pre-induction cervical ripening with IMN (40 mg) reduces the elapsed time from hospital admission to delivery as compared with placebo. The primary outcome of the trial was elapsed time interval from hospital admission to delivery. A finding of a reduction in duration from hospital admission to delivery in excess of 4 hours was considered a priori to be clinically important. Ethical approval was obtained.
Measurement of resource use Data were collected on all significant health service resource inputs used in the treatment of each woman and infant during the period between randomisation and hospital discharge. The trial data collection forms recorded the duration and intensity of maternal and neonatal care, based on standard criteria for level of care, and profiled maternal and neonatal complications. Observational work was undertaken to estimate the disaggregated resource inputs associated with complications experienced by women and infants, as well as the disaggregated resource inputs associated with alternative modes of delivery. Specifically, this involved face-to-face interviews with clinicians and midwives at the Princess Royal Maternity Hospital in order to establish staff times associated with complications and modes of delivery, as well as any equipment, drugs and disposables used. This was externally validated against clinical protocols for various aspects of maternity and neonatal care within the Princess Royal Maternity Hospital and within other hospital settings across the UK. Electronic data collection systems at the Princess Royal Maternity Hospital were accessed to ascertain occupancy rates for each maternity and neonatal ward for the recruitment period of the RCT.
Valuation of resource use Unit costs for each resource item were obtained from a variety of sources. All unit costs employed followed recent
guidelines on costing healthcare services as part of economic evaluation.8 An average (per diem) cost per day for different forms of maternity care was calculated by sending a detailed questionnaire to the hospital finance department, requesting cost data for the main resource categories of medical and nursing staff, equipment, consumables and revenue and capital overheads, and then apportioning these to different wards (prenatal assessment ward, labour ward and postnatal ward) using a ‘top–down’ methodology.9 Each per dime cost was then adjusted to reflect occupancy rates with each ward over the recruitment period of the trial. These data were validated by feeding back the costs to the hospital finance department for confirmation of accuracy and comparison with other national data. An average cost per day for each level of neonatal care, as well as intensive care for the mother, and an average cost of ambulance transfer, was derived from national Department of Health reference costs.10 Department of Health reference costs are pre-calculated on a full absorption costing basis so that costs for each level of care include staff salary, on-costs, equipment, consumables and revenue and capital overheads, bringing this is line with the costing methods we applied in our top–down methodology. The costs of drugs prescribed by the hospital were obtained from hospital pharmacy bulk pricing data. Unit costs were combined with resource volumes to obtain a net cost per mother and infant during the trial period. All costs are expressed in UK pounds sterling and valued at 2007 prices.
Measurement and representation of cost effectiveness We performed an incremental cost-effectiveness analysis in which we calculated the incremental costs (DC) and incremental effectiveness (DE) of IMN compared with placebo and expressed these as an incremental cost-effectiveness ratio (ICER; DC/DE). The cost-effectiveness of IMN was expressed in terms of the incremental cost per in patient hour prevented in the interval from hospital admission to delivery.
Data analysis The economic evaluation was conducted from a health service perspective on the basis of intention to treat. Discounting of costs and health effects was unnecessary because of the restricted time horizon for the economic evaluation. All results are reported as mean values with standard deviations and as mean differences in costs and effects with 95% confidence intervals (CIs) where applicable. We tested for differences in resource use and costs between the comparator groups using the independentsamples t-test procedure. The differences in resource use, costs and effects were considered significant if two-tailed P values were 0.05 or less. As the data for costs were
ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology
1197
Eddama et al.
skewed, we used non-parametric bootstrap estimation to derive 95% CIs for mean cost differences between the comparator groups.11 The bootstrap method does not rely on parametric assumptions concerning the underlying distribution of the data, hence its usefulness for generating CIs for skewed data.11 Using a large number of simulations, and based on sampling with replacement from the original data, the bootstrap method estimates the distribution of a sampling statistic.11 Each of these CIs was calculated using 1000 bias-corrected bootstrap replications. Non-parametric bootstrap simulation of the cost-effect pairs was also performed to generate 1000 replications of the incremental cost-effectiveness ratio, which were represented graphically on a four quadrant cost-effectiveness plane.12 Mean net benefits, defined as Rc.DE ) DC,13 were estimated for alternative values of Rc, the willingness to pay threshold for the primary outcome, namely each hour of care between hospital admission and delivery that is prevented. We undertook sensitivity analyses to illustrate the impact of the principal aspects of uncertainty on the estimates of cost-effectiveness. The values of the following variables were varied as part of the sensitivity analyses: (i) maternal hospital occupancy rates (10% increase and decrease to reflect variations in longer-term measures of occupancy over a 3 year instead of 22-month period, obtained from the hospital’s finance database); (ii) the per diem costs of each level of hospital maternity care (set at the mean and the upper and lower inter-quartile limits of national Department of Health reference costs10 in order to reflect variations in the relative price structures of resource inputs across alternative maternity settings) and (iii) the per diem cost of each level of neonatal care (set at the upper and lower inter-quartile limits of national Department of Health reference costs). All analyses were performed with a microcomputer running Excel version 2003.
Results A total of 350 women were randomised in the trial; 177 to the active group and 173 to the placebo group. The groups were similar in terms of mean maternal age, mean gestational age at delivery, number of previous pregnancies, indication for labour induction and modified bishop score.7 Isosorbide mononitrate reduced the hospital admission to delivery interval as compared with placebo by an average of 1.6 hours (95% CI: )1.9, 5.1; P = 0.37) (Table 1). There were no significant differences in mode of delivery, length of labour, oxytocin requirements during labour and the use of epidural analgesia in labour between the trial groups. The proportion of women with an unripe cervix (defined as Bishop score of
