|
Cost effectiveness of preventive home visits to the elderly: economic evaluation alongside randomized controlled study |
Kronborg C, Vass M, Lauridsen J, Avlund 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 study examined a programme of preventive home visits to elderly individuals in the community. The programme aimed to maintain the health and autonomy of older adults and to prevent disability and subsequent nursing home admission. Specifically, the programme involved health visitors and general practitioners (GPs). Health visitors were offered standard geriatric assessment tools and were taught to interpret unexplained tiredness in daily activities as an early sign of disability that should alert the visitor to search for the reason for such tiredness in the health, mental, or social domains. Any suspicion of a health problem would result in contact with the GP, who was urged to incorporate a short geriatric assessment in their usual clinical practice. Health visitors were offered training twice per year during a study period of 3 years, while GPs were offered a training session at the beginning of the study.
Type of intervention Primary and secondary prevention.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised 75- and 80-year-old individuals living in the community.
Setting The setting was primary care and the community. The economic study was carried out in Denmark.
Dates to which data relate The effectiveness and resource use data were gathered from 1999 to 2001. The price year was 2005.
Source of effectiveness data The effectiveness evidence was derived from a single study.
Link between effectiveness and cost data The costing was carried out retrospectively on the same sample of patients as that used in the effectiveness analysis.
Study sample Power calculations were not reported. Eligible individuals were identified in 4 of the 14 Danish counties in which GPs were involved in a preventive home visits programme. Of the 81 municipalities included in the 4 counties, 50 were eligible according to predetermined criteria, but only 34 agreed to participate. Municipalities that did not participate were similar to those that did participate. In 21 municipalities all 75- and 80-year-old home-dwelling citizens were invited to join the study. In the remaining 13 municipalities, all eligible individuals were invited from a specified district of the municipalities because of their size. Of the 5,788 persons initially contacted in the 34 municipalities, 4,060 (70%) agreed to participate. The reasons for refusing to participate were being too healthy, too ill, fear of registries, and by principle of not participating in research projects. Since 22 persons died and 4 were institutionalised before the intervention began, the final study sample comprised 4,034 individuals. In the 75-year-old cohort, there were 1,460 individuals (54% female) in the intervention group and 1,403 (55% female) in the control group. In the 80-year-old cohort, there were 632 individuals (59% female) in the intervention group and 539 (58% female) in the control group.
Study design This was a prospective, randomised controlled trial that was carried out in different municipalities in Denmark. The unit of randomisation was the municipality. The authors stated that the randomisation approach was performed independently of the investigators. Intervention and control municipalities were paired according to county, size, urban/rural status and geriatric department serving the municipality. The follow-up period was 3 years. Three individuals were lost to follow-up because of emigration. Data for these persons were censored by the date of emigration.
Analysis of effectiveness The analysis of the clinical study was conducted on an intention to treat basis. The primary clinical end point was functional ability, based on a mobility scale that measured the individual person's ability to manage six activities without help from others. The six activities considered were transferring, walking indoors, going outdoors, walking outdoors in both nice and poor weather, and climbing stairs. Functional ability was measured at baseline and at 1.5 and 3 years' follow-up. The lifetime within the study period in which the participant managed all six activities without help from others was defined as active life-years (LYs). The two groups were generally comparable at baseline in terms of the clinical and demographic characteristics. The exception was the proportion of individuals that lived alone in the 75-year-old cohort, which was significantly higher in the intervention group.
Effectiveness results In the 75-year-old cohort, the expected active LYs were 2.077 (+/- 0.0345) in the intervention group and 2.042 (+/- 0.0335) in the control group. The difference was 0.035 (95% confidence interval, CI: -0.060 to 0.129).
In the 80-year-old cohort, the expected active LYs were 1.730 (+/- 0.0722) in the intervention group and 1.526 (+/- 0.0640) in the control group. The difference was 0.204 (95% CI: 0.014 to 0.393).
Clinical conclusions The effectiveness analysis showed that no statistically significant difference was observed between the intervention and control groups in the 75-year-old cohort. However, in the 80-year-old cohort the intervention was more effective than usual care in terms of active LYs.
Measure of benefits used in the economic analysis The summary benefit measure was the number of active LYs. This was derived directly from the effectiveness analysis. Active LYs discounted at annual rates of 3% and 6% were also reported.
Direct costs The analysis of the costs was carried out from a societal perspective. It included the costs associated with preventive home visits, hospital stay, outpatient visits at the hospital, visits to the emergency department, visits to other health care providers (such as GPs and other medical specialists, dentists, physiotherapists, chiropractors, chiropodists, psychological counsellors), drugs, rehabilitation, day care, respite care, nursing home, aids and appliances for disabled persons, house modifications because of disability, and meals on wheels. The unit costs were presented separately from the quantities of resources used for most items. Resource use was derived from multiple sources, including the National Patient Registry for all hospital data, the National Health Insurance Service Registry for other health care providers, the Registry of Medicinal Product Statistics for drugs, and the local health authorities for other items. The costs were obtained from diagnosis-related group charges, the National Health Insurance fee schedules, retail prices, national statistics on wage rates, the National Board of Social Services, and the Danish municipalities. Details of the cost calculation were extensively reported. Discounting was relevant as the costs were incurred during more than 2 years, and two annual discount rates were used (3% and 6%). The authors reported both the undiscounted and discounted costs. The price year was 2005.
Statistical analysis of costs The costs were presented as mean values with standard deviations.
Indirect Costs Indirect costs were not included given the advanced age of the individuals considered in the analysis.
Currency Danish kroner (DKK). These were converted into euros (EUR) using the exchange rate EUR 1.00 = DKK 7.46.
Sensitivity analysis A univariate sensitivity analysis was carried out to assess the robustness of the costs and cost-effectiveness ratios to variations in some quantities of resources used and costs. The sources of the alternative values were not explicitly reported. The costs and benefits were treated stochastically in order to generate cost-effectiveness acceptability curves (based on 10,000 bootstrap resamples).
Estimated benefits used in the economic analysis When discounting benefits at the 3% rate, the expected active LYs in the 75-year-old cohort were 2.019 (+/- 0.0335) in the intervention group and 1.985 (+/- 0.0325) in the control group (difference 0.034, 95% CI: -0.058 to 0.125). The corresponding values in the 80-year-old cohort were 1.683 (+/- 0.07) in the intervention group and 1.486 (+/- 0.062) in the control group (difference 0.197, 95% CI: 0.013 to 0.38).
When discounting benefits at the 6% rate, the expected active LYs in the 75-year-old cohort were 1.965 (+/- 0.0325) in the intervention group and 1.933 (+/- 0.0315) in the control group (difference 0.034, 95% CI: -0.056 to 0.121). The corresponding values in the 80-year-old cohort were 1.639 (+/- 0.0679) in the intervention group and 1.448 (+/- 0.0602) in the control group (difference: 0.191, 95% CI: 0.013 to 0.369).
Cost results In the 75-year-old cohort, the undiscounted costs (discounted at 3%; discounted at 6%) were EUR 12,899 +/- 605.36 (EUR 12,492 +/- 585.25; EUR 12,177 +/- 566.62) in the intervention group and EUR 13,778 +/- 587.94 (EUR 13,348 +/- 568.36; EUR 12,951 +/- 550.27) in the control group.
The undiscounted cost-difference was -EUR 879 (95% CI: -2,534 to 776). The cost-difference was -EUR 856 (95% CI: -2,455 to 744) when discounted at 3% and -EUR 834 (95% CI: -2,383 to 715) when discounted at 6%.
In the 80-year-old cohort, the undiscounted costs (discounted at 3%; discounted at 6%) were EUR 17,773 +/- 1,332.17 (EUR 17,231 +/- 1,288.07; EUR 16,730 +/- 1,247.59) in the intervention group and EUR 17,059 +/- 1,180.97 (EUR 16,537 +/- 1,141.82; EUR 16,055 +/- 1,105.90) in the control group.
The undiscounted cost-difference was EUR 714 (95% CI: -2,779 to 4,207). The cost-difference was EUR 694 (95% CI: -2,684 to 4,071) when discounted at 3% and EUR 675 (95% CI: -2,596 to 3,946) when discounted at 6%.
Synthesis of costs and benefits Incremental cost-effectiveness ratios (ICERs; i.e. the incremental cost per active LY gained) were calculated in order to combine the costs and benefits of the intervention over usual care.
In the cohort of 75-year-old individuals, the costs and benefits were not combined as they were similar in both groups.
In the cohort of 80-year-old individuals, the ICER was EUR 3,522.
The univariate sensitivity analysis showed that the total costs were sensitive only to variations in the costs of nursing home care. The ICER ranged from EUR 2,906 to EUR 6,294 in the cohort of 80-year-old individuals.
The cost-effectiveness acceptability curve showed that the probability that the intervention was cost-saving in comparison with usual care was 82% in 75-year-olds and 36% in 80-year-olds when a 3% discount rate was applied. When considering alternative thresholds for the societal willingness-to-pay for an active LY gained, the probability that the intervention was more cost-effective than usual care did not exceed 86% for thresholds up to EUR 60,000 per active LY gained in 75-year-old individuals. Among 80-year-olds, the probability of the intervention being more cost-effective than usual care increased to 93% if a decision-maker was willing to pay at least EUR 20,000 per active LY gained, and to 98% if they were willing to pay at least EUR 27,000 per active LY gained.
Authors' conclusions The results of the analysis did not provide conclusive evidence on the cost-effectiveness of preventive home visits for the elderly in Denmark.
CRD COMMENTARY - Selection of comparators The authors justified the choice of the comparators, which were appropriately selected since the new intervention was compared with usual care. Some details of the new programme were given but the readers were referred to the original trial for more information. You should decide whether they are valid comparators in your own setting.
Validity of estimate of measure of effectiveness The effectiveness data were derived from a clinical trial. This was appropriate for the study question since the two interventions were compared simultaneously in a randomised fashion. This should have limited the potential impact of selection bias. The study groups were comparable at baseline in both cohorts for almost all clinical and demographic factors that were measured. This represents a further strength of the analysis. The method of sample selection was described and the reasons for refusal were reported. The authors pointed out that almost 30% of the patients refused to participate, which could limit how representative the study sample was. Non-participants were found to have poorer functional ability, which might have affected the conclusions of the analysis. The authors also highlighted that the length of follow-up might have been too short. The basis of the analysis was intention to treat, which strengthens the robustness of the study. Other details, such as the use of power calculations, were not reported since the main clinical analysis had already been published in a separate paper. The evidence came from multiple centres, although the sampling of study participants for outcome measurement could have caused some bias, as the authors noted. These issues should be considered when assessing the validity of the clinical analysis.
Validity of estimate of measure of benefit The summary benefit measure was appropriate as survival was combined with functional ability. The authors noted that active LYs were evaluated at predetermined time points. This might represent a limitation of the analysis, although this might be typical for data derived from intervention studies.
Validity of estimate of costs The perspective chosen for the analysis was appropriate and all the relevant categories of costs appear to have been included. Indirect costs, such as productivity losses, were not included given the advanced age of the individuals considered in the analysis. However, the authors stated that the costs of informal care, which represent a relevant aspect for the care of the elderly, were not included. Extensive information on the unit costs and quantities of resources consumed was provided. A detailed breakdown of the cost items was reported, which will help in replicating the analysis in other settings. All sources of data were reported for each item. Discounting was relevant but the authors also presented undiscounted results. In addition, two alternative discount rates were used. The cost estimates were specific to the study setting but alternative costs were used in the sensitivity analysis. The price year was reported, which will facilitate reflation exercises in other time periods.
Other issues The authors reported the results from some other studies, which were not directly comparable with the current analysis. The issue of the generalisability of the study results to other settings was not explicitly addressed, although most cost estimates were varied in the sensitivity analysis. The authors noted some limitations of their analysis. For example, the high refusal rate might have affected both costs and benefits because of the poorer health conditions of non-participants: active LYs might have been overestimated, whilst health care costs might have been underestimated.
Implications of the study The study results did not enable conclusions about the cost-effectiveness of preventive home visits to be drawn. Further research on the cost-effectiveness of preventive programmes in the long run is needed.
Source of funding Supported by grants from the Danish Medical Research Council, Research Foundation for General Practice and Primary Care, Eastern Danish Research Forum, County Value-Added Tax Foundation, and Ministry of Social Affairs.
Bibliographic details Kronborg C, Vass M, Lauridsen J, Avlund K. Cost effectiveness of preventive home visits to the elderly: economic evaluation alongside randomized controlled study. European Journal of Health Economics 2006; 7: 238-246 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.
Vass M, Avlund K, Hendriksen C, et al. Preventive home visits to older people in Denmark: methodology of a randomized controlled study. Aging Clin Exp Res 2002;14:509-15.
Vass M, Avlund K, Kvist K, et al. Structured home visits to older people. Are they only of benefit for women? A randomised controlled trial. Scand J Prim Health Care 2004;22:106-11.
Robertson MC, Gardner MM, Devlin N, et al. Effectiveness and economic evaluation of a nurse delivered home exercise programme to prevent falls. II. Controlled trial in multiple centres. BMJ 2001;22:701-4.
Smith RD, Widiatmoko D. The cost-effectiveness of home assessment and modification to reduce falls in the elderly. Aust N Z J Public Health 1998;22:436-40.
Indexing Status Subject indexing assigned by NLM MeSH Activities of Daily Living; Aged; Aged, 80 and over; Community Health Services /economics /organization & Cost-Benefit Analysis; Female; Health Services for the Aged /economics /organization & House Calls /economics; Humans; Male; Outcome Assessment (Health Care) /economics /organization & administration; administration; administration AccessionNumber 22007008008 Date bibliographic record published 30/04/2007 Date abstract record published 30/04/2007 |
|
|
|