Disruption of hospital care during the first year of the COVID-19 pandemic impacted socioeconomic groups differently: population based study using routine registration data.

BACKGROUND: During the COVID-19 pandemic, provision of non-COVID healthcare was recurrently severely disrupted. The objective was to determine whether disruption of non-COVID hospital use, either due to cancelled, postponed, or forgone care, during the first pandemic year of COVID-19 impacted socioeconomic groups differently compared with pre-pandemic use. METHODS: National population registry data, individually linked with data of non-COVID hospital use in the Netherlands (2017-2020). in non-institutionalised population of 25-79 years, in standardised household income deciles (1 = low, 10 = high) as proxy for socioeconomic status. Generic outcome measures included patients who received hospital care (dichotomous): outpatient contact, day treatment, inpatient clinic, and surgery. Specific procedures were included as examples of frequently performed elective and acute procedures, e.g.: elective knee/hip replacement and cataract surgery, and acute percutaneous coronary interventions (PCI). Relative risks (RR) for hospital use were reported as outcomes from generalised linear regression models (binomial) with log-link. An interaction term was included to assess whether income differences in hospital use during the pandemic deviated from pre-pandemic use. RESULTS: Hospital use rates declined in 2020 across all income groups. With baseline (2019) higher hospital use rates among lower than higher income groups, relatively stronger declines were found for lower income groups. The lowest income groups experienced a 10% larger decline in surgery received than the highest income group (RR 0.90, 95% CI 0.87 - 0.93). Patterns were similar for inpatient clinic, elective knee/hip replacement and cataract surgery. We found small or no significant income differences for outpatient clinic, day treatment, and acute PCI. CONCLUSIONS: Disruption of non-COVID hospital use in 2020 was substantial across all income groups during the acute phases of the pandemic, but relatively stronger for lower income groups than could be expected compared with pre-pandemic hospital use. Although the pandemic's impact on the health system was unprecedented, healthcare service shortages are here to stay. It is therefore pivotal to realise that lower income groups may be at risk for underuse in times of scarcity.

Socioeconomic differences in healthcare expenditure and utilization in The Netherlands.

BACKGROUND: Worldwide, socioeconomic differences in health and use of healthcare resources have been reported, even in countries providing universal healthcare coverage. However, it is unclear how large these socioeconomic differences are for different types of care and to what extent health status plays a role. Therefore, our aim was to examine to what extent healthcare expenditure and utilization differ according to educational level and income, and whether these differences can be explained by health inequalities. METHODS: Data from 18,936 participants aged 25-79 years of the Dutch Health Interview Survey were linked at the individual level to nationwide claims data that included healthcare expenditure covered in 2017. For healthcare utilization, participants reported use of different types of healthcare in the past 12 months. The association of education/income with healthcare expenditure/utilization was studied separately for different types of healthcare such as GP and hospital care. Subsequently, analyses were adjusted for general health, physical limitations, and mental health. RESULTS: For most types of healthcare, participants with lower educational and income levels had higher healthcare expenditure and used more healthcare compared to participants with the highest educational and income levels. Total healthcare expenditure was approximately between 50 and 150 % higher (depending on age group) among people in the lowest educational and income levels. These differences generally disappeared or decreased after including health covariates in the analyses. After adjustment for health, socioeconomic differences in total healthcare expenditure were reduced by 74-91 %. CONCLUSIONS: In this study among Dutch adults, lower socioeconomic status was associated with increased healthcare expenditure and utilization. These socioeconomic differences largely disappeared after taking into account health status, which implies that, within the universal Dutch healthcare system, resources are being spent where they are most needed. Improving health among lower socioeconomic groups may contribute to decreasing health inequalities and healthcare spending.

Health expenditure of employees versus self-employed individuals; a 5 year study.

It is unclear to what extent self-employed choose to become self-employed. This study aimed to compare the health care expenditures-as a proxy for health-of self-employed individuals in the year before they started their business, to that of employees. Differences by sex, age, and industry were studied. In total, 5,741,457 individuals aged 25-65 years who were listed in the tax data between 2010 and 2015 with data on their health insurance claims were included. Self-employed and employees were stratified according to sex, age, household position, personal income, region, and industry for each of the years covered. Weighted linear regression was used to compare health care expenditures in the preceding (year x-1) between self-employed and employees (in year x). Compared with employees, expenditures for hospital care, pharmaceutical care and mental health care were lower among self-employed in the year before they started their business. Differences were most pronounced for men, individuals ≥40 years and those working in the industry and energy sector, construction, financial institutions, and government and care. We conclude that healthy individuals are overrepresented among the self-employed, which is more pronounced in certain subgroups. Further qualitative research is needed to investigate the reasons why these subgroups are more likely to choose to become self-employed.

Cost-effectiveness of mandatory bicycle helmet use to prevent traumatic brain injuries and death.

BACKGROUND: Traumatic brain injury (TBI) is the main cause of mortality and severe morbidity in cyclists admitted to Dutch emergency departments (EDs). Although the use of bicycle helmets is an effective way of preventing TBI, this is uncommon in the Netherlands. An option to increase its use is through a legal enforcement. However, little is known about the cost-effectiveness of such mandatory use of helmets in the Dutch context. The current study aimed to assess the cost-effectiveness of a law that enforces helmet use to reduce TBI and TBI-related mortality. METHODS: The cost-effectiveness was estimated through decision tree modelling. In this study, wearing bicycle helmets enforced by law was compared with the current situation of infrequent voluntary helmet use. The total Dutch cycling population, consisting of 13.5 million people, was included in the model. Model data and parameters were obtained from Statistics Netherlands, the National Road Traffic Database, Dutch Injury Surveillance System, and literature. Effects included were numbers of TBI, death, and disability-adjusted life years (DALY). Costs included were healthcare costs, costs of productivity losses, and helmet costs. Sensitivity analysis was performed to assess which parameter had the largest influence on the incremental cost-effectiveness ratio (ICER). RESULTS: The intervention would lead to an estimated reduction of 2942 cases of TBI and 46 deaths. Overall, the incremental costs per 1) death averted, 2) per TBI averted, and 3) per DALY averted were estimated at 1) € 2,002,766, 2) € 31,028 and 3) € 28,465, respectively. Most favorable were the incremental costs per DALY in the 65+ age group: € 17,775. CONCLUSIONS: The overall costs per DALY averted surpassed the Dutch willingness to pay threshold value of € 20,000 for cost-effectiveness of preventive interventions. However, the cost per DALY averted for the elderly was below this threshold, indicating that in this age group largest effects can be reached. If the price of a helmet would reduce by 20%, which is non-hypothetical in a situation of large-scale purchases and use of these helmets, the introduction of this regulation would result in an intervention that is almost cost-effective in all age groups.

Single transitions and persistence of unemployment are associated with poor health outcomes.

BACKGROUND: Although job loss has been associated with decline in health, the effect of long term unemployment is less clear and under-researched. Furthermore, the impact of an economic recession on this relationship is unclear. We investigated the associations of single transitions and persistence of unemployment with health. We subsequently examined whether these associations are affected by the latest recession, which began in 2008. METHODS: In total, 57,911 participants from the Dutch Health Interview Survey who belonged to the labour force between 2004 and 2014 were included. Based on longitudinal tax registration data, single employment transitions between time point 1 (t1) and time point 2 (t2) and persistent unemployment (i.e. number of years individuals were unemployed) between t1 and time point 5 (t5) were defined. General and mental health, smoking and obesity were assessed at respectively time point 3 (t3) and time point 6 (t6). Logistic regression models were performed and interactions with recession indicators (year, annual gross domestic product estimates and regional unemployment rates) were tested. RESULTS: Compared with individuals who stayed employed at t1 and t2, the likelihood of poor mental health at the subsequent year was significantly higher in those who became unemployed at t2. Persistent unemployment was associated with poor mental health, especially for those who were persistently unemployed for 5 years. Similar patterns, although less pronounced for smoking, were found for general health and obesity. Indicators of the economic recession did not modify these associations. CONCLUSIONS: Single transitions into unemployment and persistent unemployment are associated with poor mental and general health, obesity, and to a lesser extend smoking. Our study suggests that re-employment might be an important strategy to improve health of unemployed individuals. The relatively extensive Dutch social security system may explain that the economic recession did not modify these associations.

Estimating disease prevalence from drug utilization data using the Random Forest algorithm.

BACKGROUND: Aggregated claims data on medication are often used as a proxy for the prevalence of diseases, especially chronic diseases. However, linkage between medication and diagnosis tend to be theory based and not very precise. Modelling disease probability at an individual level using individual level data may yield more accurate results. METHODS: Individual probabilities of having a certain chronic disease were estimated using the Random Forest (RF) algorithm. A training set was created from a general practitioners database of 276 723 cases that included diagnosis and claims data on medication. Model performance for 29 chronic diseases was evaluated using Receiver-Operator Curves, by measuring the Area Under the Curve (AUC). RESULTS: The diseases for which model performance was best were Parkinson's disease (AUC = .89, 95% CI = .77-1.00), diabetes (AUC = .87, 95% CI = .85-.90), osteoporosis (AUC = .87, 95% CI = .81-.92) and heart failure (AUC = .81, 95% CI = .74-.88). Five other diseases had an AUC >.75: asthma, chronic enteritis, COPD, epilepsy and HIV/AIDS. For 16 of 17 diseases tested, the medication categories used in theory-based algorithms were also identified by our method, however the RF models included a broader range of medications as important predictors. CONCLUSION: Data on medication use can be a useful predictor when estimating the prevalence of several chronic diseases. To improve the estimates, for a broader range of chronic diseases, research should use better training data, include more details concerning dosages and duration of prescriptions, and add related predictors like hospitalizations.

Commitment Lotteries Promote Physical Activity Among Overweight Adults-A Cluster Randomized Trial.

Background: The World Health Organization has identified physical inactivity as the fourth leading risk factor for global mortality. People often intend to engage in physical activity on a regular basis, but have trouble doing so. To realize their health goals, people can voluntarily accept deadlines with consequences that restrict undesired future behaviors (i.e., commitment devices). Purpose: We examined if lottery-based deadlines that leverage regret aversion would help overweight individuals in attaining their goal of attending their gym twice per week. At each deadline a lottery winner was drawn from all participants. The winners were only eligible for their prize if they attained their gym-attendance goals. Importantly, nonattending lottery winners were informed about their forgone prize. The promise of this counterfactual feedback was designed to evoke anticipated regret and emphasize the deadlines. Methods: Six corporate gyms with a total of 163 overweight participants were randomized to one of three arms. We compared (i) weekly short-term lotteries for 13 weeks; (ii) the same short-term lotteries in combination with an additional long-term lottery after 26 weeks; and (iii) a control arm without lotteries. Results: After 13 weeks, participants in the lottery arms attained their attendance goals more often than participants in the control arm. After 26 weeks, we observe a decline in goal attainment in the short-term lottery arm and the highest goal attainment in the long-term lottery arm. Conclusions: With novel applications, the current research adds to a growing body of research that demonstrates the effectiveness of commitment devices in closing the gap between health goals and behavior. Clinical Trial information: This trial is registered in the Dutch Trial Register. Identifier: NTR5559.

Physical activity after commitment lotteries: examining long-term results in a cluster randomized trial.

To overcome self-control difficulties, people can commit to their health goals by voluntarily accepting deadlines with consequences. In a commitment lottery, the winners are drawn from all participants, but can only claim their prize if they also attained their gym-attendance goals. In a 52-week, three-arm trial across six company gyms, we tested if commitment lotteries with behavioral economic underpinnings would promote physical activity among overweight adults. In previous work, we presented an effective 26-week intervention. In the present paper we analyzed maintenance of goal attainment at 52-week follow-up and the development of weight over time. We compared weight and goal attainment (gym attendance ≥ 2 per week) between three arms that-in the intervention period- consisted of (I) weekly short-term lotteries for 13 weeks; (II) the same short-term lotteries in combination with an additional long-term lottery after 26 weeks; and (III) a control arm without lottery-deadlines. After a successful 26-week intervention, goal attainment declined between weeks 27 and 52 in the long-term lottery arm, but remained higher than in the control group. Goal attainment did not differ between the short-term lottery arm and control arm. Weight declined slightly in all arms in the first 13 weeks of the trial and remained stable from there on. Commitment lotteries can support regular gym attendance up to 52 weeks, but more research is needed to achieve higher levels of maintenance and weight loss.

Emotional responses to behavioral economic incentives for health behavior change.

Many people aim to change their lifestyle, but have trouble acting on their intentions. Behavioral economic incentives and related emotions can support commitment to personal health goals, but the related emotions remain unexplored. In a regret lottery, winners who do not attain their health goals do not get their prize but receive feedback on what their forgone earnings would have been. This counterfactual feedback should provoke anticipated regret and increase commitment to health goals. We explored which emotions were actually expected upon missing out on a prize due to unsuccessful weight loss and which incentive-characteristics influence their likelihood and intensity. Participants reported their expected emotional response after missing out on a prize in one of 12 randomly presented incentive-scenarios, which varied in incentive type, incentive size and deadline distance. Participants primarily reported feeling disappointment, followed by regret. Regret was expected most when losing a lottery prize (vs. a fixed incentive) and intensified with prize size. Multiple features of the participant and the lottery incentive increase the occurrence and intensity of regret. As such, our findings can be helpful in designing behavioral economic incentives that leverage emotions to support health behavior change.

Determinants of first-time utilization of long-term care services in the Netherlands: an observational record linkage study.

BACKGROUND: Since in an ageing society more long-term care (LTC) facilities are needed, it is important to understand the main determinants of first-time utilization of (LTC) services. METHODS: The Andersen service model, which distinguishes predisposing, enabling and need factors, was used to develop a model for first-time utilization of LTC services among the general population of the Netherlands. We used data on 214,821 persons registered in a database of general practitioners (NIVEL Primary Care Database). For each person the medical history was known, as well as characteristics such as ethnicity, income, home-ownership, and marital status. Utilization data from the national register on long-term care was linked at a personal level. Generalized Linear Models were used to determine the relative importance of factors of incident LTC-service utilization. RESULTS: Top 5 determinants of LTC are need, measured as the presence of chronic diseases, age, household size, household income and homeownership. When controlling for all other determinants, the presence of an additional chronic disease increases the probability of utilizing any LTC service by 45% among the 20+ population (OR = 1.45, 95% CI: 1.41-1.49), and 31% among the 65+ population (OR = 1.31, 95% CI: 1.27-1.36). With respect to the 20+ population, living in social rent (OR = 2.45, 95% CI = 2.25-2.67, ref. = home-owner) had a large impact on utilizing any LTC service. In a lesser degree this was the case for living alone (OR = 1.63, 95% CI = 1.52-1.75, ref. = not living alone). A higher household income was linked with a lower utilization of any LTC service. CONCLUSIONS: All three factors of the Anderson model, predisposing, enabling, and need determinants influence the likelihood of future LTC service utilization. This implies that none of these factors can be left out of the analysis of what determines this use. New in our analysis is the focus on incident utilization. This provides a better estimate of the effects of predictors than a prevalence based analysis, as there is less confounding by changes in determinants occurring after LTC initiation. Especially the need of care is a strong factor. A policy implication of this relative importance of health status is therefore that LTC reforms should take health aspects into account.

The effect of trends in health and longevity on health services use by older adults.

BACKGROUND: The effect of population aging on future health services use depends on the relationship between longevity gains and health. Whether further gains in life expectancy will be paired by improvements in health is uncertain. We therefore analyze the effect of population ageing on health services use under different health scenarios. We focus on the possibly diverging trends between different dimensions of health and their effect on health services use. METHODS: Using longitudinal data on health and health services use, a latent Markov model has been estimated that includes different dimensions of health. We use this model to perform a simulation study and analyze the health dynamics that drive the effect of population aging. We simulate three health scenarios on the relationship between longevity and health (expansion of morbidity, compression of morbidity, and the dynamic equilibrium scenario). We use the scenarios to predict costs of health services use in the Netherlands between 2010 and 2050. RESULTS: Hospital use is predicted to decline after 2040, whereas long-term care will continue to rise up to 2050. Considerable differences in expenditure growth rates between scenarios with the same life expectancy but different trends in health are found. Compression of morbidity generally leads to the lowest growth. The effect of additional life expectancy gains within the same health scenario is relatively small for hospital care, but considerable for long-term care. CONCLUSIONS: By comparing different health scenarios resulting in the same life expectancy, we show that health improvements do contain costs when they decrease morbidity but not mortality. This suggests that investing in healthy aging can contribute to containing health expenditure growth.

Systematic review of individual-level, community-level, and healthcare system-level factors contributing to socioeconomic differences in healthcare utilisation in OECD countries with universal health coverage.

OBJECTIVES: Countries with universal health coverage (UHC) strive for equal access for equal needs without users getting into financial distress. However, differences in healthcare utilisation (HCU) between socioeconomic groups have been reported in countries with UHC. This systematic review provides an overview individual-level, community-level, and system-level factors contributing to socioeconomic status-related differences in HCU (SES differences in HCU). DESIGN: Systematic review following the Preferred Reporting Items for Systematic review and Meta-Analysis (PRISMA) guidelines. The review protocol was published in advance. DATA SOURCES: Embase, PubMed, Web of Science, Scopus, Econlit, and PsycInfo were searched on 9 March 2021 and 9 November 2022. ELIGIBILITY CRITERIA: Studies that quantified the contribution of one or more factors to SES difference in HCU in OECD countries with UHC. DATA EXTRACTION AND SYNTHESIS: Studies were screened for eligibility by two independent reviewers. Data were extracted using a predeveloped data-extraction form. Risk of bias (ROB) was assessed using a tailored version of Hoy's ROB-tool. Findings were categorised according to level and a framework describing the pathway of HCU. RESULTS: Of the 7172 articles screened, 314 were included in the review. 64% of the studies adjusted for differences in health needs between socioeconomic groups. The contribution of sex (53%), age (48%), financial situation (25%), and education (22%) to SES differences in HCU were studied most frequently. For most factors, mixed results were found regarding the direction of the contribution to SES differences in HCU. CONCLUSIONS: SES differences in HCU extensively correlated to factors besides health needs, suggesting that equal access for equal needs is not consistently accomplished. The contribution of factors seemed highly context dependent as no unequivocal patterns were found of how they contributed to SES differences in HCU. Most studies examined the contribution of individual-level factors to SES differences in HCU, leaving the influence of healthcare system-level characteristics relatively unexplored.

From test to rest: evaluating socioeconomic differences along the COVID-19 care pathway in the Netherlands.

INTRODUCTION: The COVID-19 pandemic exacerbated healthcare needs and caused excess mortality, especially among lower socioeconomic groups. This study describes the emergence of socioeconomic differences along the COVID-19 pathway of testing, healthcare use and mortality in the Netherlands. METHODOLOGY: This retrospective observational Dutch population-based study combined individual-level registry data from June 2020 to December 2020 on personal socioeconomic characteristics, COVID-19 administered tests, test results, general practitioner (GP) consultations, hospital admissions, Intensive Care Unit (ICU) admissions and mortality. For each outcome measure, relative differences between income groups were estimated using log-link binomial regression models. Furthermore, regression models explained socioeconomic differences in COVID-19 mortality by differences in ICU/hospital admissions, test administration and test results. RESULTS: Among the Dutch population, the lowest income group had a lower test probability (RR = 0.61) and lower risk of testing positive (RR = 0.77) compared to the highest income group. However, among individuals with at least one administered COVID-19 test, the lowest income group had a higher risk of testing positive (RR = 1.40). The likelihood of hospital admissions and ICU admissions were higher for low income groups (RR = 2.11 and RR = 2.46, respectively). The lowest income group had an almost four times higher risk of dying from COVID-19 (RR = 3.85), which could partly be explained by a higher risk of hospitalization and ICU admission, rather than differences in test administration or result. DISCUSSION: Our findings indicated that socioeconomic differences became more pronounced at each step of the care pathway, culminating to a large gap in mortality. This underlines the need for enhancing social security and well-being policies and incorporation of health equity in pandemic preparedness plans.

Estimating healthcare expenditures after becoming divorced or widowed using propensity score matching.

Becoming divorced or widowed are stressful life events experienced by a substantial part of the population. While marital status is a significant predictor in many studies on healthcare expenditures, effects of a change in marital status, specifically becoming divorced or widowed, are less investigated. This study combines individual health claims data and registered sociodemographic characteristics from all Dutch inhabitants (about 17 million) to estimate the differences in healthcare expenditure for individuals whose marital status changed (n = 469,901) compared to individuals who remained married, using propensity score matching and generalized linear models. We found that individuals who were (long-term) divorced or widowed had 12-27% higher healthcare expenditures (RR = 1.12, 95% CI 1.11-1.14; RR = 1.27, 95% CI 1.26-1.29) than individuals who remained married. Foremost, this could be attributed to higher spending on mental healthcare and home care. Higher healthcare expenditures are observed for both divorced and widowed individuals, both recently and long-term divorced/widowed individuals, and across all age groups, income levels and educational levels.

Self-reported sick leave and long-term health symptoms of Q-fever patients.

BACKGROUND: In The Netherlands, 1168 Q-fever patients were notified in 2007 and 2008. Patients and general practitioners (GPs) regularly reported persisting symptoms after acute Q-fever, especially fatigue and long periods of sick leave, to the public health authorities. International studies on smaller Q-fever outbreaks demonstrate that symptoms may persist years after acute illness. Data for the Dutch outbreaks were unavailable. The aim of this study is to quantify sick leave after acute Q-fever and long-term symptoms. METHODS: Our study targeted 898 acute Q-fever patients, notified in 2007 and 2008 residing in the Province Noord-Brabant. Patients from the 2008 cohort were mailed a questionnaire at 12 months and those of the 2007 cohort at 12-26 months after onset of illness. Patients reported underlying illness, Q-fever-related symptoms and sick leave. RESULTS: The response rate was 64%. Forty percent of the working patients reported long-term (>1 month) sick leave. Pre-existent heart disease odds ratio (OR) 4.50; confidence interval (CI) 1.27-16.09), hospitalization in the acute phase (OR 3.99; 95% CI 2.15-7.43) and smoking (OR 1.69; 95% CI 1.01-2.84) were significant predictors for long-term absence. Of the patients who resumed work, 9% were-at the time of completing the questionnaire-still unable to function at pre-infection levels due to fatigue or concentration problems. Of the respondents, 40% reported persisting physical symptoms at the time of follow-up. Fatigue (20%) was most frequently reported. Daily activities were affected in 30% of cases. CONCLUSIONS: Q-fever poses a serious persisting long-term burden on patients and society.

Exploring Health Trends Prior to State Pension Age for The Netherlands up to 2040.

BACKGROUND: In many Western countries, the state pension age is being raised to stimulate the extension of working lives. It is not yet well understood whether the health of older adults supports this increase. In this study, future health of Dutch adults aged 60 to 68 (i.e., the expected state pension age) is explored up to 2040. METHODS: Data are from the Dutch Health Interview Survey 1990-2017 (N ≈ 10,000 yearly) and the Dutch Public Health Monitor 2016 (N = 205,151). Health is operationalized using combined scores of self-reported health and limitations in mobility, hearing or seeing. Categories are: good, moderate and poor health. Based on historical health trends, two scenarios are explored: a stable health trend (neither improving nor declining) and an improving health trend. RESULTS: In 2040, the health distribution among men aged 60-68 is estimated to be 63-71% in good, 17-28% in moderate and 9-12% in poor health. Among women, this is estimated to be 64-69%, 17-24% and 12-14%, respectively. CONCLUSIONS: This study's explorations suggest that a substantial share of people will be in moderate or poor health and, thus, may have difficulty continuing working. Policy aiming at sustainable employability will, therefore, remain important, even in the case of the most favorable scenario.

Sharp upturn of life expectancy in the Netherlands: effect of more health care for the elderly?

During the 1980s and 1990s life expectancy at birth has risen only slowly in the Netherlands. In 2002, however, the rise in life expectancy suddenly accelerated. We studied the possible causes of this remarkable development. Mortality data by age, gender and cause of death were analyzed using life table methods and age-period-cohort modeling. Trends in determinants of mortality (including health care delivery) were compared with trends in mortality. Two-thirds of the increase in life expectancy at birth since 2002 were due to declines in mortality among those aged 65 and over. Declines in mortality reflected a period rather than a cohort effect, and were seen for a wide range of causes of death. Favorable changes in mortality determinants coinciding with the acceleration of mortality decline were mainly seen within the health care system. Health care expenditure rose rapidly after 2001, and was accompanied by a sharp rise of specialist visits, drug prescriptions, hospital admissions and surgical procedures among the elderly. A decline of deaths following non-treatment decisions suggests a change towards more active treatment of elderly patients. Our findings are consistent with the idea that the sharp upturn of life expectancy in the Netherlands was at least partly due to a sharp increase in health care for the elderly, and has been facilitated by a relaxation of budgetary constraints in the health care system.

The relationship between baseline health and longitudinal costs of hospital use.

In this paper, we investigate the relationship between baseline health and costs of hospital use over a period of eight years. We combine cross-sectional survey data with information from the Dutch national hospital register. Four different indicators of health (self-perceived health, long-term impairments, ADL limitations and comorbidity) are considered. We find that for ages 50 to 70, differences in hospital costs between good health and bad health are substantial and persist during the whole time period. However, for higher ages expected hospital costs for individuals in bad health decline rapidly and become lower than those for people in good health after about six to seven years. The higher mortality rate among people in bad health is the primary cause here. Our results are confirmed for all four health indicators. We conclude that relying on better health to contain healthcare expenditures is too optimistic, and the interaction between health and mortality should be taken into account when projecting healthcare costs. Healthy ageing is important, but more for health gains than for cost savings.

Exploring the influence of proximity to death on disease-specific hospital expenditures: a carpaccio of red herrings.

It has been demonstrated repeatedly that time to death is a much better predictor of health care expenditures than age. This is known as the 'red herring' hypothesis. In this article, we investigate whether this is also the case regarding disease-specific hospital expenditures. Longitudinal data samples from the Dutch hospital register (n=11 253 455) were used to estimate 94 disease-specific two-part models. Based on these models, Monte Carlo simulations were used to assess the predictive value of proximity to death and age on disease-specific expenditures. Results revealed that there was a clear effect of proximity of death on health care expenditures. This effect was present for most diseases and was strongest for most cancers. However, even for some less fatal diseases, proximity to death was found to be an important predictor of expenditures. Controlling for proximity to death, age was found to be a significant predictor of expenditures for most diseases. However, its impact is modest when compared to proximity to death. Considering the large variation in the degree to which proximity to death and age matter for each specific disease, we may speak not only of age as a 'red herring' but also of a 'carpaccio of red herrings'.

Economic evaluation and the postponement of health care costs.

The inclusion of medical costs in life years gained in economic evaluations of health care technologies has long been controversial. Arguments in favour of the inclusion of such costs are gaining support, which shifts the question from whether to how to include these costs. This paper elaborates on the issue how to include cost in life years gained in cost effectiveness analysis given the current practice of economic evaluations in which costs of related diseases are included. We combine insights from the theoretical literature on the inclusion of unrelated medical costs in life years gained with insights from the so-called 'red herring' literature. It is argued that for most interventions it would be incorrect to simply add all medical costs in life years gained to an ICER, even when these are corrected for postponement of the expensive last year of life. This is the case since some of the postponement mechanism is already captured in the unadjusted ICER by modelling the costs of related diseases. Using the example of smoking cessation, we illustrate the differences and similarities between different approaches. The paper concludes with a discussion about the proper way to account for medical costs in life years gained in economic evaluations.