|
Cost-effectiveness of referrals to high-volume hospitals: an analysis based on a probabilistic Markov model for hip fracture surgeries |
Gandjour A, Weyler E J |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The authors assessed high-volume hospitals (HVHs) in comparison with low-volume hospitals (LVHs) for hip fracture surgery. HVHs were defined as those hospitals performing more than 45 hip fracture surgeries per year. LVHs were defined as those hospitals carrying out less than 15 procedures per year.
Study population The study population comprised hypothetical hip fracture patients who were hospitalised for surgery at an average age of 74 years and who could potentially be transported either to an LVH or an HVH (i.e. those who lived within 20 km of an existing HVH).
Setting The setting was secondary care. The economic study was carried out in Germany.
Dates to which data relate The effectiveness data came from studies published between 1988 and 2005. The cost data came from studies published between 2001 and 2005.
Source of effectiveness data The following data were collected during the review:
the mortality rates,
the surgical complication rates,
the hip replacement rates,
the transition probabilities,
the probabilities that long-term care patients were cared for in different environments, and
the revision rate for survivors.
Modelling The authors used a decision tree and Markov model to evaluate the clinical and economic outcomes of the two alternatives over 6-month and 20-year time horizons. The Markov model had a cycle length of 6 months and incorporated four health states. The health states were home, long-term care, revision (with and without rehabilitation) and death. Transition probabilities were used to define a patient's journey between heath states. A half-cycle correction was used to allow for transition events occurring part-way through a given cycle. Full details of the modelling exercise were reported.
Sources searched to identify primary studies Although references were provided for all data, information about the nature of the source (e.g. randomised controlled trial) was not provided. The majority of the data were derived from published literature and national data when required. The mortality data were taken from one of two large international studies (Wenning et al. 2000, see 'Other Publications of Related Interest' below for bibliographic details), which the authors stated were the only studies to meet the inclusion criteria of a recent review and were reported to be of "the highest quality for analysing the volume outcome relationship".
Methods used to judge relevance and validity, and for extracting data It was unclear whether a review of the literature was conducted. The authors seem to have selected studies that met their requirements. No details of the search methods, inclusion criteria or estimate selection methods were provided.
Measure of benefits used in the economic analysis The summary measures of health benefit were the life expectancy and quality-adjusted life-years (QALYs). Health-related quality of life data were taken from a published source that used a preference-based questionnaire (EQ-5D) (Tidermark et al. 2002, see 'Other Publications of Related Interest' below for bibliographic details). The benefits were discounted at a rate of 3%.
Direct costs The costing was carried out from both the societal and the statutory health insurance perspectives. The analysis aimed to estimate the total annual costs of patients referred to HVHs by considering length of stay, labour and material costs for both clinical care and ancillary services (such as radiology, catering and cleaning), long-term care, rehabilitation and informal carer travel costs. The societal analysis also incorporated the cost of hospital infrastructure. A budgetary impact analysis was also carried out. The costs were adjusted to 2004 prices and were discounted at a rate of 3%.
Statistical analysis of costs The costs were treated deterministically in the base-case.
Indirect Costs Although a societal perspective was adopted, productivity costs were explicitly excluded because of the average age of the patients involved.
Sensitivity analysis A univariate sensitivity analysis was conducted around variables, using 95% confidence intervals to define the sensitivity limits where possible and plausible limits where confidence intervals were not available. A Monte Carlo simulation was used to estimate the impact of altering parameter values simultaneously. Best- and worst-case scenarios were also considered.
Estimated benefits used in the economic analysis The results of the model showed that:
6 months after surgery, LVHs gave a 0.41 gain in life-years and a 0.2 gain in QALYs;
6 months after surgery, HVHs gave a 0.42 gain in life-years and a 0.21 gain in QALYs;
after 20 years, LVHs gave a 12.45 gain in undiscounted life-years, a 10.36 gain in discounted life-years and a 4.83 gain in discounted QALYs;
after 20 years, HVHs gave a 12.55 gain in undiscounted life-years, a 10.44 gain in discounted life-years and a 4.93 gain in discounted QALYs.
Cost results Six months after surgery, the discounted costs were EUR 15,229in LVHs and EUR 14,074 in HVHs. Twenty years after surgery, the discounted costs were EUR 46,202 in LVHs and EUR 47,755 in HVHs.
Synthesis of costs and benefits The model showed that at 6 months after surgery, LVHs were dominated (greater costs and poorer outcomes). At 20 years after surgery, the incremental cost-effectiveness ratio (incremental cost per QALY gained) of HVHs over LVHs was EUR 15,530.
The discount rate was reported to have had the greatest impact on the incremental cost-effectiveness ratio 20 years post hip surgery. A cost-effectiveness curve, derived from the Monte Carlo simulation, was presented. This showed that at a willingness-to-pay of EUR 50,000 per QALY, the probability that transfer to HVHs was cost-effective was 99%.
Authors' conclusions "Transferring hip fracture surgery to HVHs (high-volume hospitals) saves costs at 6 months after surgery, but increases costs at a 20-year horizon". The authors noted that this additional cost at 20 years was "acceptable"'.
CRD COMMENTARY - Selection of comparators The authors compared HVHs with LVHs. However, the respective definitions of high and low volume do not appear to have covered hospitals performing between 15 and 45 surgeries per year. The reader must assess whether these definitions are applicable within their own setting.
Validity of estimate of measure of effectiveness The authors clearly reported the references for data that they used to populate their model. However, they omitted to provide information about the methods used to identify, select and include studies. In addition, there was no information on the sources used (e.g. randomised controlled trials). Given the limited reporting it was not possible to ascertain the validity of the estimates used. However, the extensive sensitivity analysis conducted helps reduce uncertainty in the parameter estimates. Validity of estimate of measure of benefit Life expectancy and QALYs were used as the summary measures of health benefit. These are generic measure that can be compared with any health-related technology. The source of the utility values was clearly reported and the reader is referred to this study to assess the quality of these inputs.
Validity of estimate of costs The costing analysis appears to have included all the cost categories relevant to the stated perspectives. The sources were clearly identified, and the authors reported the discount rates. It was unclear whether any costs were reflated to ensure 2004 prices were used and, if so, what adjustment was used for reflation. Given the relative comparability of the cost estimates, changes due to omissions or different perspectives could easily influence the principal results and conclusions of the study.
Other issues Comparisons with other studies were not possible as this was reported to be the first study to analyse the long-term cost-effectiveness of referral to HVHs. The authors acknowledged difficulties in generalising the results because of differences in referral strategies and resource consumption in other settings. The results were presented clearly and the conclusions were an accurate reflection of these results. Limitations relating to the availability of information for input into the model were noted. However, where possible, the authors tried to input data that estimated a conservative case for HVHs and were clear about any assumptions that were made. The authors provided a thorough discussion of the relative merits and limitations of their study.
Implications of the study The authors noted that implementing a strategy of referral to HVHs would have several implications, such as creating a weaker financial position for LVHs. Strategies to prevent the need for hip fracture surgery are noted as an area for further work.
Bibliographic details Gandjour A, Weyler E J. Cost-effectiveness of referrals to high-volume hospitals: an analysis based on a probabilistic Markov model for hip fracture surgeries. Health Care Management Science 2006; 9(4): 359-369 Other publications of related interest 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.
Wenning M, Hupe K, Scheuer I, et al. Does quantity mean quality? An analysis of 116,000 patients regarding the connection between the number of cases and the quality of the results. Chirurg 2000;71:717-22.
Tidermark J, Zethraeus N, Svensson O, Tornkvist H, Ponzer S. Femoral neck fractures in the elderly: functional outcome and quality of life according to EuroQol. Qual Life Res 2002;11:473-81.
Indexing Status Subject indexing assigned by NLM MeSH Aged; Cost-Benefit Analysis; Germany; Hip Fractures /economics /surgery; Hospitals /classification; Humans; Markov Chains; Monte Carlo Method; Referral and Consultation AccessionNumber 22007006002 Date bibliographic record published 30/09/2007 Date abstract record published 30/09/2007 |
|
|
|