|
Tai chi for the prevention of fractures in a nursing home population: an economic analysis |
Wilson C J, Datta S K |
|
|
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 use of tai chi, a martial arts-based exercise, to prevent falls and subsequent hip fractures among institutionalised elderly. The tai chi intervention consisted of two one-hour classes per week, which were monitored by a certified tai chi instructor and an assistant.
Economic study type Cost-effectiveness analysis and cost-benefit analysis.
Study population The study population comprised a hypothetical cohort of institutionalised elderly at an average risk for falls. Details of the cohort, such as age and clinical conditions, were not described.
Setting The setting was a nursing home. The economic study was conducted in the USA.
Dates to which data relate The effectiveness and resource use data were derived from studies published mainly between 1972 and 1996. The price year was 2000.
Source of effectiveness data The effectiveness evidence was derived from published studies.
Modelling A decision tree model was constructed to estimate the relative costs and benefits of the tai chi programme and standard care in a hypothetical cohort of 100 nursing home residents over one year. The possible outcomes considered in the decision tree were hip fracture and compliance. Only falls related to hip fractures were considered.
Outcomes assessed in the review The outcomes estimated in the review were the annual incidence of falls among nursing home residents;
the proportion of frequent fallers;
the proportion of falls that resulted in hip fractures;
the relative risk of falls in tai chi recipients in comparison with controls;
the case-fatality rate for hip fracture patients within one year following the injury; and
the proportion of falls resulting in fractures.
The drop-out rate was also estimated on the basis of published evidence.
Study designs and other criteria for inclusion in the review A review of the literature was performed. However, the inclusion criteria and details of the primary studies were not provided. The authors stated that one study, which was used to derive some probability values and the characteristics of the study intervention and the population of elderly, was a single-blinded randomised trial (Wolf et al., see Other Publications of Related Interest).
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included The effectiveness evidence came from 15 primary studies.
Methods of combining primary studies The primary studies appear to have been combined using narrative methods.
Investigation of differences between primary studies Results of the review The annual incidence of falls among nursing home residents was 1.5% falls per bed per year (range: 1.45 - 2.08).
The proportion of frequent fallers was 25% and the annual average was 5.8 falls.
The proportion of falls that resulted in hip fractures was 1% (range: <1 to approximately 2%).
The relative risk of fall in tai chi recipients, compared with controls, was 0.525.
The case-fatality rate for hip fractures patients within one year following the injury was 24%.
The proportion of falls resulting in fractures was 5%.
The drop-out rate was 30%, thus the compliance rate with the tai chi programme was 70%.
Measure of benefits used in the economic analysis The summary benefit measures used in the economic analysis were the avoided mortality and morbidity. These were calculated using the estimates of Viscusi (see Other Publications of Related Interest). The method used to elicit the value of the benefit measures was the willingness to pay (WTP) approach, but further details were not reported. The monetary values of avoided mortality and morbidity were updated to 2000 prices using the full consumer price index. The expected number of falls averted with the tai chi intervention relative to usual care, and the number of deaths averted, were also reported since they represented outputs of the decision model and were used in the cost-effectiveness analysis.
Direct costs Discounting was not applied because the time horizon of the model was one year. The unit costs were reported separately from the quantities of resources used. All of the cost computations were explicitly reported. The cost categories included in the economic evaluation were instructor, assistant, patient screening (to identify those who benefited from the intervention) and the treatment of hip fractures (e.g. hospitalisation and nurse time). Overhead costs and replacement/maintenance costs were not included in the analysis.
The cost/resource boundary reflected a societal perspective. The costs came from discussions with tai chi instructors across the USA, an Internet search, official wages and charges (also from Medicare reimbursement rates), and published studies. When required, a charge-to-cost ratio of 0.58 was used to estimate the true hospital costs. Resource use was based on probability data derived from the literature review and from authors' assumptions. All the costs were inflated to 2000 prices using the medical care component of the consumer price index.
Statistical analysis of costs The costs were treated deterministically in the base case.
Indirect Costs The indirect costs were not included in the economic evaluation.
Sensitivity analysis One-way sensitivity analyses were conducted to deal with uncertainty in the input estimates. For example, the annual cost of the tai chi intervention, the direct cost of the hip fracture, the number of falls per patient, the relative risk of fall with tai chi, and the percentage of falls resulting in hip fracture. Also, the probability of hip fracture with and without tai chi, the probability of non-compliance, and one-year mortality from hip fracture. The scenario in which there were no assistant training costs in subsequent years was also considered. The ranges used in the sensitivity analysis appear to have been obtained from the literature and regional variations. A threshold analysis (determination of a cut-off point beyond which tai chi was no longer cost-saving) was also conducted.
Estimated benefits used in the economic analysis The estimated values (2000 values) of avoided mortality and morbidity were $715,432.50 and $16,466.80, respectively, according to the published studies.
When such values were applied to the cohort of patients under study, the value of mortality avoided was $85,637.27 (or $856.37 per person per year). The value of morbidity avoided was $41,002.33 (or $410.02 per person per year).
The expected number of falls averted with the tai chi intervention relative to usual care was 49.875 over one year in the cohort of 100 nursing home residents. Thus, the hip fracture incidence per 100 participants per year was 1.00125 with the tai chi programme and 1.5 with usual care.
The number of deaths averted with the tai chi programme was 0.1197.
Cost results The estimated total costs were $27,517.10 in the tai chi group and $28,321.50 in the usual care group. The difference in costs was $804.40 in favour of the tai chi programme. Thus, the expected direct net cost-savings was $8.04 per patient, which rose to $14.04 if there was no assistant training cost in subsequent years.
Synthesis of costs and benefits An average cost-effectiveness ratio was calculated to combine the costs and benefits of the two study strategies within a cost-effectiveness analysis approach.
The cost per fall averted with tai chi was $172.68.
The cost per hip fracture averted was $17,268.17.
The cost per death averted ranged from $64,294 to $66,694 depending on whether 2 or 4 new assistants were trained during the study period.
The total net benefits associated with the tai chi programme were calculated (sum of the direct cost-savings and the values of mortality and morbidity avoided). These amounted to $1,274.43 per person per year.
The threshold analysis revealed that tai chi was no longer cost-saving beyond a price of $95 per person (it was $86 in the base-case), or when the costs of treating hip fractures were below $17,268.17.
The results were also sensitive to variations in the probability of hip fractures in the absence of tai chi or with non-compliance.
Authors' conclusions The findings supported the implementation of a tai chi programme among institutionalised elderly because the economic benefits exceeded its costs, independently of considering direct or indirect benefits.
CRD COMMENTARY - Selection of comparators The choice of the comparator appears to have been appropriate because no tai chi represented the usual approach used for institutionalised elderly individuals. You should decide whether it represents usual care in your own setting.
Validity of estimate of measure of effectiveness The effectiveness evidence came from published studies. The authors stated that a review of the literature was conducted to identify the relevant primary studies. However, the method and conduct of the review were not reported and only the design of one primary study was given. The authors did not state whether differences between the studies (e.g. sample characteristics, comparators and setting) were taken into account and the estimates were combined using narrative methods. The authors also acknowledged that the main characteristics and probabilities of the tai chi programme came from a single study in which strict inclusion criteria were adopted. Therefore, it appears difficult to evaluate the internal validity of the effectiveness study.
Validity of estimate of measure of benefit The summary benefit measures used in the economic analysis appear to have been appropriate for the two types of economic evaluations cost-effectiveness and cost-benefit analyses) conducted. In particular, in the cost-benefit analysis, the WTP approach was used to quantify the economic value of avoided mortality and morbidity. Such values came from a published study whose results were updated to present prices. The authors discussed the implications of using WTP rather than an approach in which quality-adjusted life-years were estimated.
Validity of estimate of costs The economic analysis appears to have been conducted from a societal perspective, although the indirect costs were not included. However, lost productivity would be irrelevant among institutionalised elderly. Therefore, it appears likely that all the relevant costs were considered. The authors stated that some categories of costs were not included, but provided justifications for such exclusions. The unit costs and the source of each cost component were provided. The price year was reported and all the cost computations were explicitly described, thus it would be possible to replicate the study in other settings. Cost inflation was considered and a charge-to-cost ratio was applied. The costs were treated deterministically in the base-case, but wide variations were investigated in the sensitivity analysis. These issues tend to enhance the internal validity of the cost analysis.
Other issues The authors made few comparisons of their findings with those from other studies. On the issue of the generalisability of the study results, the authors stated that sensitivity analyses were conducted to increase the external validity of their conclusions. However, the evidence on tai chi came largely from a single study, thus limiting the generalisability of the study results. The conclusions of the analysis should therefore be limited to institutionalised elderly patients at an average risk of fall, although the concept of "average risk" was not explicitly defined. The authors discussed some advantages and disadvantages of implementing the tai chi programme.
Implications of the study The study results suggested that tai chi is an economically feasible intervention in the setting of nursing homes, not only from a societal perspective but also from the point of view of patient insurers (such as Medicare or Medicaid), who might see the financial benefits of the intervention. Future research should evaluate the benefits of a tai chi programme on patients at a high risk of falls.
Bibliographic details Wilson C J, Datta S K. Tai chi for the prevention of fractures in a nursing home population: an economic analysis. Journal of Clinical Outcomes Management 2001; 8(3): 19-27 Other publications of related interest Wolf SL, Barnhart HX, Kutner NG, et al. Reducing frailty and falls in older persons: an investigation of the Tai Chi and computerized balance training. Atlanta FICSIT group. Frailty and Injuries: Cooperative Studies of Intervention Techniques. Journal of the American Geriatrics Society 1996;44:489-497.
Viscusi WK. The value of risks to life and health. Journal of Economic Literature 1993;31:1912-46.
Indexing Status Subject indexing assigned by CRD MeSH Accidental Falls /prevention & Aged; Cost-Benefit Analysis; Female; Fractures, Bone /prevention & Humans; Male; Residential Facilities; Tai Ji; control; control AccessionNumber 22001007675 Date bibliographic record published 31/03/2004 Date abstract record published 31/03/2004 |
|
|
|