Healthcare during Covid-19

Einleitung / Introduction

Willingness to pay (WTP) is an established measure for valuing non-marketed goods, such as health care services. We analyze if unexpected events, i.e. information shocks, can cause a shift in the willingness to pay. We study the effect of health-associated information shocks (HIS) on WTP by analyzing how SARS-CoV-2 infections in the proximity of a subject alter the WTP for booster vaccinations.

Methode / Method

We elicited the WTP for booster vaccinations against Covid-19 in representative samples in Germany, Denmark, Italy, France, Portugal, Spain, the United Kingdom and the Netherlands. The data of 8,454 individuals was collected as part of the 9th wave of the European Covid Survey (ECOS) between 23 December 2021 and 11 January 2022. The WTP for booster shots was elicited in a three-stepped approach, asking for (1) the minimum WTP, (2) the maximum WTP, (3) the final WTP considering previous answers. We used dynamic validation to ensure the consistency and quality of answers. The setting specified that a booster vaccination is recommended but must be paid out of pocket and individual’s health permits vaccination. For HIS measurement, participants were asked for the existence, proximity and severity of close Covid-19 cases in the weeks before the survey. Descriptive statistics on the sample, the WTP values and HIS were calculated. The difference in WTP depending on the occurrence of HIS was estimated using a two-part model. Average marginal effects (AME) were calculated to facilitate interpretation.

Ergebnisse / Results

Of 8,454 collected observations, we excluded 137 as careless responders and 2,508 respondents indicating protest answers. Of the final sample comprising 5,809 observations, 76.8% stated a WTP greater than €0. At least one close COVID-19 case, i.e. a HIS, was reported by 61.9% of participants. The occurrence of a HIS was associated with an increase in WTP by €14.53 (p<0.0001) compared to no HIS according to a model without any covariates. Analyzing the temporal proximity, HIS was associated with an increase in WTP, which was stronger if the HIS occurred in the weeks before the survey (AME weeks 0-2: €20.05, p=0.0002, weeks 3-4: €29.20, p=0.005), while after more than four weeks the HIS association is still positive but no longer becomes significant, compared to no HIS. If experience of HIS was milder than expected, HIS was associated with a decrease in WTP by €-19.81 (p=0.0001). Controlling for socio-demographic and COVID-19-related covariates decreases the significance and effect sizes for temporal proximity and severity of HIS.

Zusammenfassung / Conclusion

The analysis seems to confirm HIS shifting the WTP of individuals. Timing for WTP measurement matters, and researchers should be aware of potential effects of HIS when informing decision-makers using WTP values.

AutorInnen

Carolin Brinkmann, Hamburg Center for Health Economics, Universität Hamburg

Sebastian Neumann-Böhme, Hamburg Center for Health Economics, Universität Hamburg, Erasmus School of Health Policy & Management, Erasmus University Rotterdam

Werner B F Brouwer, Erasmus School of Health Policy & Management, Erasmus University Rotterdam

Tom Stargardt, Hamburg Center for Health Economics, Universität Hamburg

Einleitung / Introduction

Many countries including Germany utilized government-imposed restrictions such as public and private gathering restrictions, school closures, and cancellation or postponement of elective health care services during the COVID-19 pandemic to keep the COVID-19 related mortality and morbidity rate low. However, underutilization of health care services, especially lower well-child visits and lower screening rates for cancer can lead to unintended long-term public health problems. While outpatient utilization reduced considerably during the pandemic, government-imposed restrictions are found to have a limited impact on the underutilization of health care services in the US. In this work, we examined the effect of government-imposed restrictions on health care utilization during the first three waves of COVID-19 pandemic in Germany.

Methode / Method

We used monthly health care utilization data obtained from Zentralinstitut für die kassenärztliche Versorgung by state from January 2019- June 2021. We considered cancer screenings (colonoscopy, mammogram, and melanoma screenings), obstetrics and well-child visits, non-COVID-19 vaccinations, chronic disease management (dialysis, disease management programs, and cancer treatments), specialist and total outpatient visits. We employed an instrumental variables approach, using the time to state elections as an instrument for state-imposed gathering restrictions to examine the causal effect of gathering restrictions on health care utilization. We included state fixed effects and weather controls. Since policy measures are expected to be imposed depending on the progression of the pandemic, we controlled for the state-level exposure to the COVID-19 pandemic to avoid the bias introduced by non-random assignment of policy measures or time-varying treatment effects.

Ergebnisse / Results

We find that one month of gathering restriction caused 36% decline in specialist visits and 18% decline in total outpatient visits. Colonoscopies, obstetrics and well-child visits, non-COVID-19 vaccinations, dialysis and cancer treatments were not significantly adversely affected by the gathering restrictions. However, disease management programs utilization decreased by 59% and mammograms and melanoma screenings came to a stall as a result of gathering restrictions.

Zusammenfassung / Conclusion

Policy measures taken to fight the COVID-19 pandemic had a significant impact on some preventive, chronic and acute care in Germany, while the most critical health care services were not significantly affected by the government-imposed restrictions. Promoting the utilization of disease management programs, mammograms, and melanoma screenings can help avoid unintended consequences of government-imposed restrictions during the COVID-19 pandemic.

AutorInnen

Esra Eren Bayindir, Hamburg Center for Health Economics, Universität Hamburg

Jonas Schreyögg, Hamburg Center for Health Economics, Universität Hamburg

Einleitung / Introduction

Implementing lockdown measures, as non-pharmaceutical containment policies, during pandemics like COVID-19, is crucial to contain the spread of the disease early on. One key determinant for the effectiveness of such containment policies is that the public supports the measures and complies with the policies. Therefore, it is important to assess the public’s perception of measures through means like the perceived strictness of lockdown policies. Therefore, the aim of this study is to investigate the effect of objective measures as well as other factors on subjective perception of lockdown stringency.

Methode / Method

Using a panel ordered logistic regression model with random effects, we employed a representative sample of 21,348 observations from seven European countries collected in three waves of the European Covid Survey (ECOS) between January 19th and July 5th 2021. The subjective stringency was raised in ECOS while the objective stringency was obtained from the Oxford COVID-19 Government Response Tracker.

Ergebnisse / Results

This study finds that subjective and objective stringency are positively associated. Most notably, we find clear differences in gender, where the effect of objective on subjective stringency is smaller for males than for females. Furthermore, being female, trusting government information as well as increasing age, income, and perceived risk from COVID-19 significantly decrease the likelihood that respondents perceive measures as stricter. In contrast, trusting information from social media has the opposite effect.

Zusammenfassung / Conclusion

Our results suggest that the perception of containment measures is not only driven by the objective strictness, but also by adjustable factors like perceived risk and trust in information sources. This offers important insights for policy makers who aim to increase the acceptability of measures and potentially also compliance to them.

AutorInnen

Lasse Falk, Universität Hamburg

Sebastian Neumann-Böhme, Universität Hamburg

Iryna Sabat, Nova School of Business and Economics

Jonas Schreyögg, Universität Hamburg

Einleitung / Introduction

The benefits of lockdown measures lie in reducing the potential health loss in society as a result of a pandemic. One of the main costs relate to reductions in population well-being. Trading off these costs and benefits is important, especially because they do not necessarily fall into the same groups within society. Currently, health benefits of measures like lockdowns are concentrated in older age groups, while well-being reductions appear larger in younger age groups. Knowledge about how citizens think about trade-offs between health and well-being within and across age groups is lacking. The objective of this study is to elicit these preferences, which may inform future policy decisions involving such trade-offs.

Methode / Method

A discrete choice experiment is set up using a pandemic setting and the Dutch context. Choice tasks include two policy options that differ with respect to four attributes: Health change, well-being change, age group experiencing the health change, and age group experiencing the well-being change. Health is expressed as prevented deaths, while well-being change is expressed on a well-being scale from 0 to 100. Changes occur in one year in a specified number of individuals. A heterogenous Bayesian D-efficient design was used to create seven blocks of choice tasks with 14 questions each. Sampling and survey administration was conducted online. A mixed logit model with main effects and linearly specified health and well-being changes builds the base model. Calculations are conducted in willingness-to-pay space, using prevented deaths as numerical denominator.

Ergebnisse / Results

Responses from 1123 individuals from the Netherlands were included in the analysis. All levels of the four attributes were significant in the mixed logit model, with coefficients of prevented deaths and well-being reduction being postive and negative, respectively. A health gain in terms of prevented deaths is most preferred in the mid age group (25-64) and least preffered in the oldest age group (65+). Well-being loss receives highest disutility in the youngest age group (<25). The numerical results in willingness-to-pay space indicate that the number of deaths that need to be prevented to warrant a 10% well-being reduction for 1 year in 1,000 people was 2.0 if both effects occure in the mid-age group, 1.3 if the well-being loss occurs among 65+ and prevented deaths among 25-64, and 8.2 if the well-being loss occurs among <25 and prevented deaths among 65+.

Zusammenfassung / Conclusion

This study provides information on people’s relative preferences for health and well-being, in a pandemic setting, within and between age groups. This information may be useful for guiding (pandemic response) policies involving health and well-being trade-offs, and increasing their acceptability in society.

AutorInnen

Sebastian Himmler, Technische Universität München