Economic Impact and Cost-Effectiveness of Hungary's Salmonella Control Program

Categories: Science

Essay, Pages 8 (1927 words)

Views

Introduction to Salmonellosis and Labor Productivity Loss

Salmonellosis presents a considerable public health challenge due to its impact on labor productivity. The economic evaluation requires an understanding of the disease's burden on the workforce, particularly the number of workdays lost, the average daily earnings, and the labor market's size. The analysis includes cases under 65 years of age, reflecting potential productivity loss, and considers pediatric cases by assuming a parent's absence from work.

Calculation of Productivity Loss

To estimate the opportunity cost of labour due to salmonellosis, data are needed on the number of days lost due to illness, the average daily earnings and the size of the labour market.

Don't use plagiarized sources. Get your custom paper on

“ Economic Impact and Cost-Effectiveness of Hungary’s Salmonella Control Program ”

Get high-quality paper

NEW! smart matching with writer

First, the number of cases under 65 were calculated, because in these cases can productivity loss arise. Paediatric cases were included, because it can be assumed that one patent stays at home with the sick child. For estimating the labour market participation data from KSH was used. The activity rate of the working age population varies between 63.

95% and 72.52% during the examined period. Data on the average monthly wage in Hungary were taken from the UNECE database. The average wages were given at current prices (2019.03.19.), so they had to be adjusted to their 2017 year value, by applying the Hungarian consumer price indices taken from KSH. To calculate the average daily wage the average monthly wage was divided by the average workdays of a month (appr. 21), that was calculated based on the annual number of workdays of the given year divided by 12.

For the number of days absent from work data from the literature were applied, namely from the FCC report, 2010.

This model assumes:

0.5 workday loss for mild cases, who do not seek medical care
1.6 days absence from work for those mild and moderate cases who visit their GP
4.5 workdays loss for the hospitalized cases who recovered
4.5 workdays loss for fatal cases.

These values take into account that some days of the illness may fall on the weekend. During the interview with the gastroenterologist, the doctor estimated that within 7-8 days the parameters, e.g. liver functions, of the hospitalised patients’ return to normal. Assuming that a patient does not get hospitalised on the first day when the symptoms appear, the average workday loss due to hospitalisation seemed to be in line with the estimated 4.5 workday loss.

The total cost of productivity loss for those who recovered was calculated based on the following formula:

∑ annual number of cases by health states * related work day loss by health states * average daily wage of the given year.

The estimated total costs of productivity loss on both arms are presented in Table 14. The annual costs by health states for both arms are presented in Annex 8.

When Hoffmann et al., 2012 (ERS-USDA model) estimated the economic burden of food borne pathogens in the UD, the authors did not calculate productivity loss or health care costs for fatal cases. Instead, they calculated with a total cost of premature death. In that model based on willingness to pay studies they calculated with a value of a statistical life (VSL) of 7.9 million (in 2009 US dollars), adjusted for age and sex. The authors also took into account that in some cases of neonatal deaths, stillbirths and miscarriages the loss is compensated by another siblings, thus they calculated with a fraction of the total VSL (Hoffmann et al., 2012).

For premature death cases, occurring among the patients under 65, the cost of productivity loss was calculated for the years that they would have spent working if they lived. For this estimation the difference between the age of retirement (for simplification for all cases 65 was considered as retiring age) and the age of death was calculated. The cost of one year productivity loss was calculated by multiplying the annual number of fatal cases (that occurred among the working age population) with the activity rate of the working age population and the annual average gross income of the active population (based on the 2017 average monthly income data of KSH) . The costs occurring after 2017 were discounted with a 3.7% discounting rate. This calculation was performed for each case (average cases were taken into account by age groups and sex).

For the QALY calculation, the health states and QALY loss per case values were adopted from Batz et al., 2014. As mentioned earlier, no chronic sequelae were included in the QALY calculation. For the estimation of the total number of illnesses an underreporting factor of 66.8 (95% Credibility Interval: 10.2 – 199.1)) was used that was calculated for Hungary by De Knegt et al, 2015a. In their model the authors used data from EFSA and the European Surveillance System (TESSy) for the years 2007-2009, provided by the ECDC. This result is equal to the underreporting factor calculated by Havelaar et al, 2013 for Hungary (66.7).

For the calculation of QALY loss in non-fatal cases the QALY loss per case values taken from Batz et al., 2014 were multiplied with the annual number of cases in the related health states. The sum of these values equals the total QALY loss of the recovered cases in a given year.

For the calculation of QALY loss due to mortality the Eurostat data on Hungarian life expectancies by age and sex were used. For each age group the average age by sex was calculated based on the Eurostat data on population by age and sex. For these “average” cases of each age group for both sexes the corresponding life expectancies for each calendar year were matched. These lost life years were then corrected with the corresponding life-quality index values by age groups and sex, based on EQ-5Dindex_TTO values, taken from the OLEF2000 (Hungarian national Health Interview Survey of 2000) report (Boros et al. (eds.), 2002).

As a final step, QALY loss by age category was multiplied by the number of fatal cases that occurred in that age category. Future forgone lifeyears after 2017 were discounted with the 3.7% discounting rate. The sum of these values equals the total QALY loss due to premature death in a given year. The total QALY loss of salmonellosis in a given year is the sum of total QALY loss due to premature death and the total QALY loss of the recovered cases in a given year. The annual QALY losses by health states for both arms are presented in Annex 10.

For the no-intervention arm with unchanged case numbers and rates based on data from 2007 and for the programme arm 2007-2017 both costs (programme costs, productivity loss, health care costs) and health outcomes were calculated.

For each costs and health outcomes that occurred before 2017 a correction were made according to the consumer price indices. In case of productivity loss and QALY loss occurring after 2017 discounting were performed by using the 3.7% discount rate.

To decide on the cost-effectiveness of the programme as threshold the 3x GDP/capita per QALY gain was used. According to KSH data the GDP per capita in Hungary was 3 919 000 HUF in 2017, thus the threshold value is 11 757 000 HUF per QALY gain. According to my calculation, the ICER is 5 943 912.7 HUF, that is below the national HTA threshold. This result means that the Salmonella control programme was cost-effective.

Evaluation and Sensitivity Analyses

Salmonella is still one of the most common food borne diseases in the EU, hence the control of the disease is a high priority within the EU and its MSs. Based on the results of the presented analysis, the Hungarian Salmonella Control Programme was cost-effective so far. The costs spent on the interventions are in line with the health gain due to the decreasing number of human cases, including hospitalizations and fatal cases and the declining health care costs and productivity loss.

However, this result should be treated with caution. The applied methodology has some weaknesses and limitations that cannot be overlooked when evaluating the results. The uncertainties and weaknesses should be addressed by sensitivity analyses.

Sensitivity Analyses

In case of such a complex analysis, there are several sources of uncertainty in the estimates of cost of illness and QALY losses. These can be model uncertainties due to assumptions about structural relations or measurement uncertainties in relation to the estimation of parameter values.

The most important methodological weaknesses are:

the rate and number of fatal cases have been recalculated retrospectively;
the total number of cases were calculated with a 66.8 underreporting factor, which is a strong assumption
the utilisation of health services was adopted from the literature (Frenzen et al., 1999), however we cannot know what the real situation is in Hungary
the cost of productivity loss was added in the analysis, although this might not occur at the state but at the employers
the estimated productivity loss of fatal cases (premature death).

Other factors may also had an effect on the results such as the estimated costs of health care services, or the number of work day losses due to the illness that was based on the data from the FCC report, 2010. The estimated cost of human salmonellosis was also conservative, since it did not take into account the costs of possible sequelae.

Out of these factors, most likely the number of fatal cases had the greatest effect on the results, since according to the Batz et al., 2014 QALY loss of salmonellosis is dominated by acute mortality (>99%). Therefore, uncertainties of undiagnosed mild cases or hospitalization rates should have a smaller impact on the results.

As it was expected, by calculating with the initial number of fatal cases extracted from the ECDC database, the result of the cost-effectiveness analysis of the Salmonella control programme changed drastically. While the costs have not decreased so notably, the QALY losses on both arms decreased. When calculating the delta cost and delta QALY, it turned out, that on the programme arm a greater QALY loss was estimated than on the comparator arm. As a result, the ICER value of this scenario is – 131 297 049.5 HUF per QALY. Therefore, in this case the Salmonella control programme is all but not cost-effective.

One might argue that the cost of productivity loss should not be taken into account, because in each year a certain amount of days of absence from work are paid by the employer instead of the state (In Hungary 15 days of sick leave is paid by the employer in a calendar year). Another deterministic sensitivity analysis was performed by not including the costs of productivity loss in the calculation, since this cost might not pose a burden to the state but the employers. If the cost-effectiveness analysis is rerun according to this scenario, the ICER increases to 10 392 341.1 HUF / QALY gain. This value is still below the 3xGDP/per capita threshold; hence the Salmonella control programme is still cost-effective.

A deterministic sensitivity analysis was conducted where the cost of the Salmonella control programme was changed. In the original analysis the total expenditures spent on the programme was included. However, the EU co-finances these interventions, thus until 2015 50%, since 2015 75% of the eligible costs are transferred back to the state. If the amount of the EU-financing is deducted from the total programme cost, the estimated cost that the Hungarian state spent on the programme is 6 396.5 million HUF. If we recalculate the cost-effectiveness analysis with this programme cost, the ICER significantly decreases, its value is 28 589 HUF / QALY gain, that is below the selected threshold. IN this scenario, the Salmonella control programme is very cost-effective.

Conclusion

The economic evaluation of the Hungarian Salmonella Control Programme underscores the importance of rigorous cost-effectiveness analyses in public health decision-making. By quantifying the impact of salmonellosis on labor productivity and assessing the intervention program's economic viability, the study provides critical insights into managing foodborne diseases and optimizing resource allocation for public health interventions.