|
An economic evaluation of increased uptake in Q fever vaccination among meat and agricultural industry workers following implementation of the National Q Fever Management Program |
Kermode M, Yong K, Hurley S, Marmion B |
|
|
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 implementation of a National Q Fever Management Program (NQFMP) in order to immunise meat and agricultural industry workers at risk of infection. The NQFMP screens all people at risk of infection, and vaccination is administered if both the skin and serology tests are negative. The vaccine used was Q-VAX (CSL Limited).
Economic study type Cost-effectiveness analysis and cost-utility analysis.
Study population The study population comprised a hypothetical cohort of 30-year-old meat industry (MI) and agricultural industry (AI) workers who were not vaccinated prior to the introduction of the vaccine programme (i.e. 35% of MI workers and all AI workers).
Setting The setting was the community. The economic study was carried out in Australia.
Dates to which data relate The transition probabilities used in the model were gathered from studies published between 1980 and 2001. One further source was referenced to an Internet site accessed in 2002. The cost data were obtained from published studies, a pharmaceutical company and government department sources from 1989 to 2001. The costs were adjusted to 2001 prices.
Source of effectiveness data The estimates for the final outcomes were derived from a synthesis of published studies and several authors' assumptions.
Modelling A Markov model was used to estimate the costs and benefits of each strategy. The time horizon for the model was 20 years with each cycle lasting for one year. Eleven health states were used in the model. According to these the workers could be:
susceptible (not vaccinated and without pre-existing immunity),
not susceptible (vaccinated or with pre-existing immunity),
in an asymptomatic infection state,
in a temporary state of acute Q fever,
recovered,
Q fever fatigue syndrome (QFS),
post-QFS,
pre-endocarditis,
endocarditis with or without surgery, and
death.
Outcomes assessed in the review The following effectiveness outcomes were obtained from the review:
the risk of infection with Coxiella burnetti;
the probability of symptomatic infection;
the probability of hospitalisation with acute Q fever;
the probability of death from acute Q fever;
the probability of developing QFS;
the probability of developing endocarditis;
the probability of having valve replacement surgery following endocarditis; and
the probability of death from endocarditis.
These effectiveness outcomes were included in the model as input parameters.
Study designs and other criteria for inclusion in the review 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 At least 18 studies appear to have been reviewed.
Methods of combining primary studies A narrative explanation of how the study results were combined was provided.
Investigation of differences between primary studies Results of the review The results of the review were as follows:
the risk of infection with Coxiella burnetti was 0.03 per year for the first 5 years and 0.02 for the next 5 years in the primary analysis;
the probability of symptomatic infection was 0.23 (range: 0.10 - 0.40);
the probability of hospitalisation with acute Q fever was 0.05 (range: 0.02 - 0.10);
the probability of death from acute Q fever was 0.01 (range: 0.005 - 0.024);
the probability of developing QFS was 0.20 (range: 0.10 - 0.30);
the probability of developing endocarditis was 0.02 (range: 0.01 - 0.16);
the probability of having valve replacement surgery following endocarditis was 0.44; and
the probability of death from endocarditis was 0.34.
Methods used to derive estimates of effectiveness The authors made assumptions to obtain some of the estimates of effectiveness.
Estimates of effectiveness and key assumptions The model assumed the following:
the efficacy of the vaccine was 98% and it provided lifelong protection;
compliance with screening and vaccination was 100% among MI workers not vaccinated in the old scenario, and 20% (range: 10 - 40) among AI workers;
pre-existing immunity was 17% (range: 10 - 50) among MI workers and 28% (range: 10 - 50) among AI workers;
time in the industry was 10 years for both groups (range: 3 - 20); and
all-cause mortality was based on age-specific all-cause rates.
Measure of benefits used in the economic analysis The summary measures of health benefit used were the life-years gained (LYG), quality-adjusted life-years (QALYs) gained and Q fever cases avoided. The study also reported the number of deaths avoided. These outcomes were obtained from the model.
Health state valuations were obtained from the literature. EQ-5D scores for patients with chronic fatigue syndrome were used to proxy post-QFS state. Utility values for endocarditis (a possible sequela of Q fever) could not be found in the literature, so utility values for congestive cardiac failure were used instead. The time horizon considered for the estimation of health benefits was 20 years. The LYG and QALYs were discounted at a discount rate of 5%.
Direct costs The direct costs to the health service were considered. The published costs included in the analysis were for pre-vaccination screening, vaccination (medical consultation and Q-VAX), acute Q fever treatment, the direct medical costs of QFS, and valve replacement surgery for endocarditis. The pre-vaccination screening costs covered a medical consultation, the skin test antigen for MI and AI workers, and serology. The treatment costs for acute Q fever covered diagnostic serology, medical consultations and hospitalisation.
The costs were obtained from the Medicare Benefits Schedule, CSL Ltd. (a pharmaceutical company), the Australian Institute of Health and Welfare, and the Department of Human Services. Discounting was appropriately performed at a rate of 5%, as recommended by the Pharmaceutical Benefits Advisory Committee. The price data were either from 2001 or were adjusted to that year using the current Australian health inflation rate of 5.9% per annum.
Statistical analysis of costs The cost data were treated deterministically.
Indirect Costs No indirect costs were considered in the base-case analysis, although they were included in the sensitivity analyses. The rationale for inclusion was that the indirect costs to the government and the community were considerable. The average cost of "Workcover" scheme compensation claims was reported. This was taken from a study published in 1999 that used data from two Australian states. The indirect costs were discounted at 5% and were adjusted to a 2001 price year allowing an inflation rate of 5.9% per annum, as described already.
Currency Australian dollars (Aus$).
Sensitivity analysis A sensitivity analysis was conducted to investigate variability in the data. The methods used were not specifically described, although it appears that one-way sensitivity analyses have been performed. Transition probabilities and key model assumptions were investigated in the sensitivity analysis, and the indirect costs were included in one of these analyses, although the unit cost data were not subject to analysis. The ranges used for the sensitivity analyses were either taken from published studies or based on authors' assumptions.
Estimated benefits used in the economic analysis The number of incremental discounted LYG by instituting the programme was 43 among MI workers and 87 among AI workers.
The number of incremental discounted QALYs gained was 138 among MI workers and 273 among AI workers.
The number of incremental Q fever cases avoided was 400 among MI workers and 785 among AI workers.
The number of Q-fever related deaths was 4 with the vaccination programme, and 8 for the period prior to the implementation of the programme.
Cost results The discounted total medical costs of the intervention programme were Aus$1,247,785 for MI workers and Aus$5,923,060 for AI workers.
The discounted total medical costs of the comparator were Aus$381,439 for MI workers and Aus$3,747,420 for AI workers.
The incremental cost of the programme was Aus$866,346 for MI workers and Aus$2,175,640 for AI workers.
The costs associated with the side effects of vaccination were not considered in the analysis. Although side effects can be severe if vaccination is given to a person with pre-existing immunity, the authors reported that reactions to the vaccine are extremely rare.
Synthesis of costs and benefits The incremental costs and benefits were combined by calculating cost-effectiveness ratios (cost per LYG, cost per QALY and cost per Q-fever case avoided) and incremental cost-effectiveness ratios (incremental cost per additional LYG, QALY or Q-fever case avoided with the vaccination programme, compared with the period before the implementation of the programme).
The discounted cost per LYG by the intervention programme was Aus$20,002 for MI workers and Aus$24,950 for AI workers.
The discounted cost per QALY gained was Aus$6,294 for MI workers and Aus$7,984 for AI workers.
The discounted cost per Q fever case avoided was Aus$2,164 for MI workers and Aus$2,772 for AI workers.
The undiscounted cost per LYG was Aus$10,406 for MI workers and Aus$13,164 for AI workers.
The sensitivity analysis showed that, for both groups of workers, the outcomes were most sensitive to duration in the industry, the proportion developing acute Q fever, and the time horizon of the model. The AI model was also sensitive to the proportion dying from acute Q fever and the discount rate. When the indirect costs were included (compensation claims), cost-savings were observed for both industry groups.
Authors' conclusions Increasing the uptake of Q fever vaccine among meat and agricultural industry workers is a cost-effective public health strategy.
CRD COMMENTARY - Selection of comparators Although no explicit justification was provided for the comparator used, it represented usual practice before the introduction of the Q fever vaccination programme. You should decide if the comparator represents current practice in your own setting.
Validity of estimate of measure of effectiveness The authors stated that a systematic search of the literature had been undertaken, although it was not possible to assess how systematic this was as no information was given about the methods used (sources searched, criteria used to select and include studies and to assess their relevance and validity). Therefore, it cannot be assessed whether the review was conducted in such a way as to identify relevant research and minimise bias. However, the authors stated that it is very likely that all relevant studies were identified, therefore reducing these potential biases, since the literature related to Q-fever is not extensive.
The estimates for the parameters appear to have been chosen using a narrative synthesis. A sensitivity analysis was conducted on many of the effectiveness model input parameters. The ranges used were mostly derived from the literature and appear to have been appropriate, but several were based on authors' assumptions. The authors commented that the assumptions used in the model were conservative. They did not consider the impact of differences between the studies when estimating effectiveness.
Validity of estimate of measure of benefit The measures of benefit used were LYG, QALY gains and Q fever cases avoided. These were derived directly from the model. QALY gains were proxied from quality of life studies, among patients with different illnesses, which had used the EQ-5D and the time trade-off approach to estimate utility. The validity of the QALY gains relies on the unproven assumptions that the quality of life with QFS is the same as that of chronic fatigue syndrome, and similarly for Q fever endocarditis and congestive cardiac failure.
Validity of estimate of costs The cost analysis was performed from the perspective of the health service (i.e. the government funding the vaccination programme). It appears that most of the relevant categories of costs have been included in the analysis. The authors noted that some relevant costs were omitted (i.e. the costs of complications due to acute Q fever other than endocarditis and endocarditis-associated stroke). The authors anticipated that the inclusion of these would have increased the cost-effectiveness of the intervention. The costs and resource use related to the vaccination programme were reported disaggregated, although the medical costs associated with some of the health states considered in the model were reported in an aggregated manner. This may limit to some extent the reproducibility of the study results in other settings. The costs were treated deterministically and no sensitivity analysis of the prices was undertaken. Discounting was applied, which was appropriate given the model time horizon. Costs, rather than charges, were reported, thus reflecting the true opportunity costs of the intervention. The cost data were taken from sources published between 1989 and 2001, and were adjusted to a single price year using an appropriate inflation rate.
Other issues The authors did not compare their findings with those from other studies, so it is not known how far their results agree with other published results. They also did not directly address the issue of the generalisability of the results to other settings. The authors do not appear to have presented their results selectively. The authors acknowledged the uncertainty in several estimates used in the model. Specifically, there was uncertainty in the level of compliance among AI workers, the rate at which AI workers leave the industry, the risk of infection over time, the proportion developing QFS, and the utility values for QFS and endocarditis. In terms of the cost results, it would have been useful to have investigated the sensitivity of the outcomes to medical costs.
Implications of the study The authors noted that Australia does not use an explicit cost-effectiveness or cost-utility threshold to guide funding decisions for health care interventions, but the NQFMP seems to provide excellent value for money in terms of health benefit gains.
Bibliographic details Kermode M, Yong K, Hurley S, Marmion B. An economic evaluation of increased uptake in Q fever vaccination among meat and agricultural industry workers following implementation of the National Q Fever Management Program. Australian and New Zealand Journal of Public Health 2003; 27(4): 390-398 Indexing Status Subject indexing assigned by NLM MeSH Adult; Agriculture; Australia /epidemiology; Bacterial Vaccines /administration & Cost-Benefit Analysis /statistics & Coxiella burnetii /immunology; Female; Humans; Immunization Programs /economics /utilization; Male; Markov Chains; Meat-Packing Industry; Middle Aged; Occupational Diseases /economics /prevention & Q Fever /epidemiology /prevention & Quality-Adjusted Life Years; control; control; dosage /economics; numerical data AccessionNumber 22003009829 Date bibliographic record published 31/07/2005 Date abstract record published 31/07/2005 |
|
|
|