The study examined newborn screening for medium chain acyl-CoA dehydrogenase deficiency (MCADD) using tandem mass spectrometry (MS/MS).

Screening.

The objective of the study was to assess the cost-effectiveness of MS/MS screening for MCADD in comparison with no screening (clinical diagnosis) in newborns. Full details of the study were reported in a health technology assessment (HTA) published on the website of the Canadian Agency for Drugs and Technologies in Health (CADTH). In brief, a review of clinical and economic data was carried out along with a primary economic evaluation based on a decision model to determine whether Canadian jurisdictional programmes should be expanded to include MCADD on the grounds of their value for money. The current abstract will report primarily on the economic modelling analysis. The perspective of the public health care payer was adopted in the study.

Cost-utility analysis.

The study population comprised a hypothetical cohort of newborns.

A decision tree analytical model was constructed to assess the long-term costs and benefits of screening versus no screening in a hypothetical cohort of newborns. The time horizon of the model was lifetime. In the model, there were 16 pathways for screening and 6 for no screening. The pathways under screening were distinguished from each other by the results (positive or negative), MCADD status (true or false positive), the time of acute episode (early or late) and the long-term health consequences (asymptomatic, acute complications only, mild or severe neurological impairment, death). The pathways under no screening differed from each other only in terms of MCADD status (with or without) and long-term health consequences. A graphical representation of the model was provided in the primary CADTH report, but not in this publication.

The setting was a hospital. The economic study was carried out in Canada.

The effectiveness data were derived from studies published between 1995 and 2004. No dates for the costs or resource use were provided. The price year was not reported.

The clinical data used in the model were:

the incidence of MCADD;

screening accuracy (sensitivity, specificity, positive predictive value and negative predictive value);

the outcomes of MCADD patients detected by MS/MS-based screening versus newborns being clinically diagnosed (rate of homozygosity for A985G mutation, rate of heterozygosity for A985G mutation, percentage of asymptomatic cases, percentage of symptomatic cases, and percentage of fatal cases); and

health outcomes (probabilities of patient experiencing an acute episode, recovering from one, having neurological impairment and dying).

Details of the data sources (including at least study design, sample sizes and diagnostic results) were provided in the primary Canadian HTA. Twenty-one studies were included in this analysis (6 based on abstracts alone), most of which were prospective cohort studies.

A systematic review of the literature was undertaken to identify the primary studies. Clinical and economic databases such as MEDLINE, BIOSIS Previews, Pascal, Social SciSearch, PsycINFO, ERIC and EMBASE were searched, beginning at 1995. Grey literature was identified by searching websites of regulatory agencies, HTA and related agencies. The search terms were reported in detail. The authors also provided extensive details of the inclusion and exclusion criteria, and the methods used to assess the validity of the primary studies. In particular, studies were eligible if they specified newborn screening for MCADD using MS/MS, if they compared outcomes of MS/MS-based screening with clinical diagnosis, or if they reported outcomes of patients identified only by clinical diagnosis. The validity of the studies was assessed using the QUADAS, an instrument used to evaluate the quality of studies of diagnostic accuracy.

The summary benefit measure used was the expected number of quality-adjusted life-years (QALYs). The QALYs were derived using a modelling approach. The utility weights used to adjust expected survival appear to have been obtained from three published US studies, but no further details of these analyses were provided. The benefits were discounted at an annual rate of 3%.

The analysis of the costs was performed from the perspective of the public health payer, thus only the direct medical costs were included. As reported in the primary HTA, the main cost categories were screening visits, acute episode and severe neurological impairment. The authors stated that the bulk of the information on costs and resource use was derived from the Nova Scotia screening programme. Discounting was relevant, as the long-term costs were evaluated, and an annual rate of 3% was applied. The price year was not explicitly reported.

The productivity costs were not considered, which was appropriate given the perspective adopted in the study.

Canadian dollars (CAD).

The costs were treated deterministically in the base-case.

The robustness of the base-case cost-effectiveness analysis was tested by running one- and two-way sensitivity analyses, as well as investigating best- and worst-case scenarios. The authors reported all the assumptions that were made in the different scenarios considered. Variability was mainly tested for in incidence rates, diagnostic accuracy, screening costs and costs associated with the treatment of acute episodes.

The lifetime additional QALYs gained with screening over no screening in a hypothetical cohort of 330,803 newborns were 181. These ranged from 52 (worst-case scenario) to 378 (best-case scenario).

The total cost of screening a hypothetical cohort of 330,803 newborns over a 77-year horizon was CAD 934,923, compared with CAD 450,521 for no screening (cost-difference CAD 484,402).

Under the screening scenario, the cost of screening was the major cost driver (85% of the total cost). In the no screening scenario, the major costs were shifted to providing care for patients with acute episodes (57%) and providing long-term care for patients with severe neurological impairment (43%).

An incremental cost-utility ratio was calculated in order to combine the costs and QALYs.

The incremental cost per QALY gained with screening in comparison with no screening was CAD 2,676.

The incremental cost per QALY was CAD 1,029 in the best-case scenario and CAD 11,436 in the worst-case scenario.

The univariate sensitivity analysis showed that the results were sensitive to changes in screening cost and specificity. When the cost of screening increased to CAD 5.60 (CAD 2.40 in the base-case), the incremental cost per QALY rose to CAD 6,963.40. Increasing the specificity from 99.95% to 99.995% decreased the cost per QALY from CAD 3,095 to CAD 2,078.

The two-way sensitivity analysis indicated that when screening cost and specificity were varied simultaneously, the cost per QALY ranged from -CAD 756 to CAD 7,875. When screening cost and cost associated with an acute episode were varied together, the cost per QALY ranged from -CAD 860 to CAD 7,363. In all cases analysed the incremental cost per QALY of screening with respect to no screening was below a threshold of CAD 20,000.

Screening for medium chain acyl-CoA dehydrogenase deficiency (MCADD) using tandem mass spectrometry (MS/MS) was a cost-effective strategy in the Canadian health care system.

The rationale for the choice of the comparator was clear. No screening was an appropriate strategy as it reflected the current pattern of care in the authors' setting. You should decide whether this is a valid comparator in your own setting.

Extensive information on the decision model was presented in the primary  HTA, where a justification was provided for the time horizon and the  choice of the health states.  The model parameters and their sources  were reported in full in the HTA.  Fewer details were available in this  publication, although key parameters values were provided.

The clinical evidence used in the decision model came from a review of the literature, details of which (e.g. inclusion criteria and reasons for exclusion) were provided. Limited information on the design and other characteristics of the studies was given in this publication, but it could be found in the HTA. The authors evaluated the validity of the clinical data using a scoring system, which suggested that most of the published studies were of a poor quality. In particular, the authors stated that clinical studies scored poorly in items covering bias, variability and reporting. The impact of changing clinical assumptions was extensively addressed in the sensitivity analysis.

The estimation of health benefits (QALYs) was modelled using a decision tree model. The methods used to estimate the utility weights were not described as they were taken from published sources. The authors acknowledged that utility weights were derived from studies that had no MCADD patients. Discounting was appropriately carried out.

The analysis of the costs appears to have been consistent with the perspective of the health payer, although information on the categories of costs included and their sources was limited. The majority of the information was reported in the primary HTA. The costs were treated deterministically but alternative cost estimates were used in the sensitivity analysis. The price year was not reported, thus limiting the possibility of performing reflation exercises in other time periods.

The authors reported the results of other economic evaluations that were systematically searched in the published literature. However, no explicit comparisons with current findings were made. The issue of the generalisability of the study results to other settings was addressed in the sensitivity analysis, in which alternative scenarios were taken into account. The authors acknowledged some limitations of their study, mainly arising from the poor quality of the clinical evidence available. Other issues, such as the use of country-specific data sources or the exclusion of repetition of acute episodes, were also highlighted. However, the study results were robust to variation in the clinical and economic parameters.

The study results appear to support the use of MCADD screening using MS/MS. The authors pointed out the key issue of parental distress associated with screening.

Funded by the Canadian Agency for Drugs and Technologies in Health.

Because readers are likely to encounter and assess individual publications, NHS EED abstracts reflect the original publication as it is written, as a stand-alone paper. Where NHS EED abstractors are able to identify positively that a publication is significantly linked to or informed by other publications, these will be referenced in the text of the abstract and their bibliographic details recorded here for information.

Tran K, Banerjee S, Li H, et al. Newborn screening for medium chain acyl-CoA dehydrogenase deficiency using tandem mass spectrometry: clinical and cost-effectiveness. Technology Report. Ottawa: Canadian Coordinating Office for Health Technology Assessment; 2006. Report No.: 62.

Insinga RP, Laessig RH, Hoffman GL. Newborn screening with tandem mass spectrometry: examining its cost-effectiveness in the Wisconsin Newborn Screening Panel. J Pediatr 2002;141:524-31.

Venditti LN, Venditti CP, Berry GT, et al. Newborn screening by tandem mass spectrometry for medium-chain acyl-CoA dehydrogenase deficiency: a cost-effectiveness analysis. Pediatrics 2003;112:1005-15.

This record was compiled by CRD commissioned reviewers according to a set of guidelines developed in collaboration with a group of leading health economists.