A Guide to Handling Missing Data in Cost-Effectiveness Analysis Conducted Within Randomised Controlled Trials

Faria, Rita; Gomes, Manuel; Epstein, David; White, Ian R.

doi:10.1007/s40273-014-0193-3

Cited by 490 publications

(623 citation statements)

References 40 publications

Supporting

Mentioning

616

Contrasting

Unclassified

Order By: Relevance

“…Missing data on costs and clinical effects for the various follow‐up periods were imputed using multiple imputation by chained equations (MICE) as implemented in STATA 14 (Van Buuren, Boshuizen, & Knook, 1999). For this, we made the assumption that data were “missing at random” (Faria, Gomes, Epstein, & White, 2014). Imputations were performed separately for the intervention and the control group within each study.…”

Section: Methodsmentioning

confidence: 99%

Cost effectiveness of guided Internet-based interventions for depression in comparison with control conditions: An individual-participant data meta-analysis

Kolovos

Dongen²,

Riper

et al. 2018

Depress Anxiety

View full text Add to dashboard Cite

BackgroundThere is limited evidence on the cost effectiveness of Internet‐based treatments for depression. The aim was to evaluate the cost effectiveness of guided Internet‐based interventions for depression compared to controls.MethodsIndividual–participant data from five randomized controlled trials (RCT), including 1,426 participants, were combined. Cost‐effectiveness analyses were conducted at 8 weeks, 6 months, and 12 months follow‐up.ResultsThe guided Internet‐based interventions were more costly than the controls, but not statistically significant (12 months mean difference = €406, 95% CI: − 611 to 1,444). The mean differences in clinical effects were not statistically significant (12 months mean difference = 1.75, 95% CI: − .09 to 3.60 in Center for Epidemiologic Studies Depression Scale [CES‐D] score, .06, 95% CI: − .02 to .13 in response rate, and .00, 95% CI: − .03 to .03 in quality‐adjusted life‐years [QALYs]). Cost‐effectiveness acceptability curves indicated that high investments are needed to reach an acceptable probability that the intervention is cost effective compared to control for CES‐D and response to treatment (e.g., at 12‐month follow‐up the probability of being cost effective was .95 at a ceiling ratio of 2,000 €/point of improvement in CES‐D score). For QALYs, the intervention's probability of being cost effective compared to control was low at the commonly accepted willingness‐to‐pay threshold (e.g., at 12‐month follow‐up the probability was .29 and. 31 at a ceiling ratio of 24,000 and 35,000 €/QALY, respectively).ConclusionsBased on the present findings, guided Internet‐based interventions for depression are not considered cost effective compared to controls. However, only a minority of RCTs investigating the clinical effectiveness of guided Internet‐based interventions also assessed cost effectiveness and were included in this individual–participant data meta‐analysis.

show abstract

Section: Methodsmentioning

confidence: 99%

Cost effectiveness of guided Internet-based interventions for depression in comparison with control conditions: An individual-participant data meta-analysis

Kolovos

Dongen²,

Riper

et al. 2018

Depress Anxiety

View full text Add to dashboard Cite

show abstract

“…MI of both costs and QALYs is increasingly recognised as an appropriate approach to deal with missing observation and missing follow-up data (29). All missing cost and utility data were treated as missing at random.…”

Section: Missing Datamentioning

confidence: 99%

“…The imputations were conducted in STATA version 13, using predictive mean matching and sequential chained equations. The variables included in the models were selected on the basis of potential predictive ability identified from descriptive and regression analyses of the pooled baseline and follow up data (29).…”

Section: Missing Datamentioning

confidence: 99%

Cost-effectiveness of structured group psychoeducation versus unstructured group support for bipolar disorder: Results from a multi-centre pragmatic randomised controlled trial

Camacho

Ntais

Jones

et al. 2017

Journal of Affective Disorders

View full text Add to dashboard Cite

Background Bipolar disorder (BD) costs the English economy an estimated £5.2billion/year, largely through incomplete recovery. This analysis estimated the cost-effectiveness of group psychoeducation (PEd), versus group peer support (PS), for treating BD. Methods A 96-week pragmatic randomised controlled trial (RCT), conducted in NHS primary care. The primary analysis compared PEd with PS, using multiple imputed datasets for missing values. An economic model was used to compare PEd with treatment as usual (TAU). The perspective was Health and Personal Social Services. Results Participants receiving PEd (n=153) used more (costly) health-related resources than PS (n=151) (net cost per person £1098 (95% CI, £252-£1943)), with a quality-adjusted life year (QALY) gain of 0.023 (95% CI, 0.001-0.056). The cost per QALY gained was £47,739. PEd may be costeffective (versus PS) if decision makers are willing to pay at least £37,500 per QALY gained. PEd costs £10,765 more than PS to avoid one relapse. The economic model indicates that PEd may be costeffective versus TAU if it reduces the probability of relapse (by 15%) or reduces the probability of and increases time to relapse (by 10%). Limitations Participants were generally inconsistent in attending treatment sessions and low numbers had complete cost/QALY data. Factors contributing to pervasive uncertainty of the results are discussed. Conclusions This is the first economic evaluation of PEd versus PS in a pragmatic trial. PEd is associated with a modest improvement in health status and higher costs than PS. There is a high level of uncertainty in the data and results.

show abstract

“…54 Current guidance on handling missing data in cost-effectiveness analysis was followed to inform such analysis. 55 Only missing utility scores (for both EQ-5D-3L index and SF-6D utilities) and individual cost items were imputed and the distribution of responses for both instruments was reported for data available.…”

Section: Discussionmentioning

confidence: 99%

A multicentre, randomised controlled trial of position during the late stages of labour in nulliparous women with an epidural: clinical effectiveness and an economic evaluation (BUMPES)

Bick

Briley

Brocklehurst

et al. 2017

Health Technol Assess

View full text Add to dashboard Cite

BackgroundEpidural analgesia leads to increased risk of instrumental vaginal delivery (IVD). There is debate about whether or not posture in second-stage labour influences the incidence of spontaneous vaginal birth (SVB).ObjectivesIn nulliparous women with epidural analgesia, does a policy of adopting an ‘upright position’ throughout second-stage labour increase the incidence of SVB compared with a policy of adopting a ‘lying-down’ position?DesignTwo-arm randomised controlled trial.SettingMaternity units in England and Wales.ParticipantsNulliparous women aged ≥ 16 years, at ≥ 37 weeks’ gestation with singleton cephalic presentation and intended SVB, in second-stage labour with an epidural providing effective pain relief.Interventions(1) Upright position to maintain the pelvis in as vertical a plane as possible; and (2) lying-down position to maintain the pelvis in as horizontal a plane as possible.Main outcome measuresThe primary outcome measure was incidence of SVB. Secondary outcomes included augmentation, interventions to maintain blood pressure, duration of labour, episiotomy, genital tract trauma, post-partum haemorrhage, maternal satisfaction, neonatal metabolic acidosis, 5-minute Apgar score of < 4, resuscitation at birth and admission to neonatal unit. At 1 year for (1) women: urinary or faecal incontinence, dyspareunia and health-related quality of life; (2) for infants: major morbidity. A cost–consequences analysis with a time horizon of 1 year after the birth from a NHS perspective.ResultsBetween October 2010 and January 2014, 3236 women were randomised from 41 centres in England and Wales. There was a statistically significant difference in the incidence of SVB between groups, with 35.2% of women achieving a SVB in the upright group, compared with 41.1% in the lying-down group (adjusted risk ratio 0.86, 95% confidence interval 0.78 to 0.94). There was no evidence of differences in most of the secondary maternal or neonatal outcomes, or in long-term outcomes at the 12-month follow-up. No significant overall cost differences were observed between upright and lying-down positions for mothers or their babies.LimitationsMeasurement of adherence was challenging in this unmasked trial, and adherence could be influenced by midwives’ beliefs about the allocated positions. If adherence was poor, this would have diluted the difference between the two groups.ConclusionsThere is clear evidence of the benefit of adopting a lying-down position in second-stage labour in nulliparous women with epidural analgesia, with no apparent disadvantages in either short- or long-term outcomes for mother or baby, and this is cost neutral for the NHS.Future workQuestions remain about whether or not other positions could increase the incidence of SVB further in this group of women. The results also raise questions about the role of maternal position in second-stage labour in women without an epidural.Trial registrationCurrent Controlled Trials ISRCTN35706297.FundingThis project was funded by the National Institute for Health Research Health Technology Assessment programme and will be published inHealth Technology Assessment, Vol 21, No. 65. See the NIHR Journals Library website for further project information.

show abstract

A Guide to Handling Missing Data in Cost-Effectiveness Analysis Conducted Within Randomised Controlled Trials

Cited by 490 publications

References 40 publications

Cost effectiveness of guided Internet-based interventions for depression in comparison with control conditions: An individual-participant data meta-analysis

Cost effectiveness of guided Internet-based interventions for depression in comparison with control conditions: An individual-participant data meta-analysis

Cost-effectiveness of structured group psychoeducation versus unstructured group support for bipolar disorder: Results from a multi-centre pragmatic randomised controlled trial

A multicentre, randomised controlled trial of position during the late stages of labour in nulliparous women with an epidural: clinical effectiveness and an economic evaluation (BUMPES)

Contact Info

Product

Resources

About