Background: Health-care utilities differ considerably from country to country. Our objective was to examine the association of cultural values based on Hofstede’s cultural dimensions’ theory with utility values that were identified using the time trade off method. Methods: We performed a literature search to determine preference-based value algorithms in the general popula- tion of a given country. We then fitted a second-order quadratic function to assess the utility function curve that links health status with health-care utilities. We ranked the countries according to the concavity and convexity proper- ties of their utility functions and compared this ranking with that of the Hofstede index to check if there were any similarities. Results: We identified 10 countries with an EQ-5D-5L-based value set and 7 countries with an EQ-5D-3L-based value set. Japan’s degree of concavity was highest, while Germany’s was lowest, based on the EQ-5D-3L and EQ-5D-5L value sets. Japan also ranked first in the Hofstede long-term orientation index, and rankings related to the degree of concav- ity, indicating a low time preference rate. Conclusions: This is the first evaluation to identify and report an association between different cultural beliefs and utility values. These findings underline the necessity to take local values into consideration when designing health technology assessment systems. higher). One possible explanation is that the quality of Background health status in Thai people is simply not as high as that Utility is a preference weight, where preference can be of the Japanese. But even based on comparable health sta measured in terms of value or desirability . Accord- - ingly, health-care utilities need to be built on prefer- tuses, higher utility values are reported for the Japanese ences for the different health statuses. The more desirable population. Our hypothesis is that this is probably due to health statuses generally receive a greater utility value. the way utilities are identified, which ultimately reflects Utility is measured on an interval scale of 0–1, where cultural values and beliefs. Take the EQ-5D, for exam- 0 indicates death and 1 indicates full health status with ple, where utilities for a small number of health statuses negative values assigned to states worse than death . are obtained by means of a time trade-off (TTO) exer - Utility value sets are country-specific and there are cise and later on generalized with regression techniques huge differences between countries. The reasons for these to the remaining health states. Using the TTO concept, differences are mostly unclear. A recent study [ 3] found respondents are asked to choose either to live 10 years in that the utility value as measured by EuroQOL-5D-3L in a specified current health status or to give up some life the Thai population at age 60 was nearly 0.65, while that years to live for a shorter period in full health. The num - of the Japanese population was found to be 0.91 (40% ber of years in full health that are deemed of equal value to 10 years in the current health state describes the utility value . If the respondent is not willing to give up life *Correspondence: Joerg.firstname.lastname@example.org years against full health at all, then the utility value of the Health Economics, Janssen Pharmaceutical KK, 5-2, Nishi-kanda current health state is 1. 3-chome Chiyoda-ku, Tokyo 101-0065, Japan Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 2 of 8 It is evident that the willingness to trade lengths of life which means that they would assign high utility to any for quality of life reflects a person’s time preference which given health state. Recently, it was shown that female in turn is a cultural value. The Dutch sociologist Geert students reported higher quality of life values when they Hofstede classified cultures according to their short-term were elicited by SG . This outcome was found to be or long-term planning horizon. In his “Confucian dimen- associated with lower risk preference by females, a well- sion”, the Asian cultures of China, Hong Kong, Taiwan established finding in behavioral economics [11–19]. One and Japan have the longest-term orientations while many plausible explanation for this was proposed as the display Western countries such as the US and UK only achieve of overconfidence exhibited by men about their deci - ranks 17 and 18, respectively [5, 6]. If time preference is sions, especially when forecasting potential outcomes determined by cultural values, then utilities obtained by more precisely than women . Not only do women and TTO have a cultural dimension as well. A person with men have different attitudes towards risk, but societies as a low preference for the present, which is equivalent to a whole. In terms of Hofstede’s risk avoidance ranking, a long-term orientation, would be less willing to trade countries such as Greece, Belgium, Italy, and Korea rank lengths of life for quality of life and would therefore have high while the US und the United Kingdom have very low a higher utility value than a person from a culture with a scores on the uncertainty avoidance index. strong preference for the present seeking gratification in To the best of our knowledge, there is no published the here and now. Of note, such differences would occur study on the association of culture and health utility to in the same health states due to differences in cultural date. Therefore, we aim to determine this association by values with far reaching implications for health tech- assessing utility functions of different countries in terms nology assessment. If the higher utility values of Asians of their curvature properties and profiles. simply result from their long-term orientation, a medical intervention likely to restore perfect health in an Asian Methods context will not create as much value (regarding addi- To test our proposition that culture shapes utility values, tional utilities) as the same intervention in a culture with our objective was to investigate whether utility func- a short-term orientation. This is because a culture with tions that are based on the TTO method show different short-term orientation would value given health states profiles that are in line with the time preference of each lower and the potential utility increase of an interven- country. We also wanted to test our hypothesis that cul- tion is larger. On the other hand, societies with long-term turally determined risk attitudes shape utility functions orientation would probably be more willing to pay for an when utilities are elicited by the SG method. additional year of life. However, this would be reflected Regarding TTO, a country with a low time preference in a higher Incremental cost-effectiveness ratio (ICER) would have a very high marginal rate of substitution of threshold and should not be confused with the additional life span and quality of life. The two indifference curves utilities gained from an intervention. that depict a combination of two “goods”, namely lengths Similar arguments can be put forward for utilities of life and quality of life that provide the same level of sat- derived from standard gamble (SG) known as the gold isfaction to the consumer are presented in Fig. 1. I is an standard  in utility identification since it is derived from Neumann–Morgenstern’s rigorous axiom-based economic decision theory under uncertainty . Here, a utility value equals the probability value that makes a person indifferent to a certain outcome (health state) and a gamble in which-probability (1 − p)—means the per- son dies immediately and—probability p-means perfect health is restored. Culture here also plays a critical role in terms of risk aversion. Again, one of Hofstede’s cul- tural dimensions—apart from long-term orientation—is the avoidance of uncertainty. This dimension focuses on how cultures adapt to changes and manage uncertainty. Emphasis is on the extent to which a culture feels threat- ened or is concerned about ambiguity. The US and UK are considered as countries with a culture of high risk- taking, while Japan or Italy are characterized by a high uncertainty avoidance score . Followers of a risk averse Fig. 1 Indifference curves for low (I ) and high (I ) time preference society would hesitate to engage in the gambling process, 1 2 Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 3 of 8 indifference curve for a country with a low time prefer - were cross-walk value sets derived by means of a map- ence aka long-term orientation. ping approach to the three-level version of the EQ-5D. For people to trade off lengths of life against quality of The algorithms can be used to generate utility values of life, one would need to offer higher values of quality of various sets of responses. We used country specific util - life for forgoing one unit of life span than in countries ity sets to fit a second-order quadratic function so that with a higher time preference (such as I ). The utility the utility function connecting health states and utilities function that links different health states to utility val - can be estimated. To assess the curvature of this util- ues have different profiles as well. We would expect that ity function, we then calculated its second derivative to utility functions from long-term orientation countries determine whether the function is concave (i.e. nega- are more convex than those of short-term orientation tive sign of the second derivative) or convex (i.e. positive countries. To illustrate this, imagine an individual whose sign of the second derivative). This can be interpreted long-term preference is so high that he would by no as a quasi-Arrow–Pratt measure, which uses the curva- means trade a unit of his life span against quality. Conse- ture of the utility functions to determine risk aversion quently, his utility values will be one for all health states . Following that, we ranked the countries according except death. The same argument can be developed for to the concavity and convexity properties of their utility SG derived utilities. An extremely risk averse individual functions and compared their ranking with that of the would never engage in the gamble and would attach the Hofstede index to check if there were any similarities as value of 1 for all health states except death, which is an hypothesized. extreme example of a convex utility function. A graphical In the final step, we ran a scenario analysis to assess the representation of the utility functions according to risk practical implications of our findings. We constructed preference is presented in Fig. 2. two scenarios for four countries that were character- We performed a literature search using PubMed to ized by a different culture based on Hofstede’s taxonomy, determine preference-based value algorithms derived namely Japan (JP), Germany (GE), United Kingdom (UK), from either standard gamble or time trade-off in the gen - and the US. eral population. The search terms were “utility” OR “pref - The first scenario was a breakthrough innovation. In erence” OR “standard gamble” OR “time trade off ” OR this scenario, we assumed that the hypothetical prod- “EQ-5D” OR “health related quality of life” OR “health uct could increase any EQ-5D-5L-health status from 5 utility index” OR “SF-36” OR “SF-6D” AND “healthy vol- (worst) to 1 (best). For example: before receiving treat- unteer” OR “healthy population” OR “healthy”. The Euro - ment, the health status of a patient was 55,523 which qol and 36-item short form websites were searched as corresponds to a utility value of 0.068 in JP, 0.085 in GE, well in January 2017 [21, 22]. Most of the value sets for − 0.142 in the UK, and 0.127 in the US . We assumed EQ-5D-3L were from Szende et al. . The EQ-5D-3L that breakthrough treatment changes the health status value set from Thailand is from Tongsiri and Cairns , to 11,123, thus changing the utility values to 0.721 in JP, and the value set for the EQ-5D-5L can be found in Van 0.909 in GE, 0.75 in the UK, and 0.809 in the US. We Hout et al. . Some of the value sets included in here then calculated the incremental utility for each change Fig. 2 Utility functions according to risk preference Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 4 of 8 (JP 0.653, GE 0.824, UK 0.893 and US 0.682). We per- The estimates of the 2nd order quadratic utility func - formed the same calculations for all possible health states tion are reported in Table 1. Depending on the second and averaged the incremental change for this hypotheti- derivative of this function, the estimates were categorized cal product. Then, we calculated the ICER value based as concave or convex. on the following assumptions. First, we assumed a drug The lowest second derivative was ranked as number 1 price of 1,000,000 Japanese Yen (JPY), without any other on a scale in descending order. As previously mentioned, incremental costs or cost offsets. Second, we assumed a the maximum rank indicates the highest degree of con- time horizon of 1 year. cavity. The more concave a function, the higher the long- The second scenario we implemented was incremen - term preference for the respective country. In addition, tal innovation. Here, we assumed that the hypotheti- Figs. 3, 4 are graphical representations of the utility func- cal product can improve the EQ-5D-5L-health status by tions of the countries selected based on EQ-5D-3L and only one unit in all five dimensions (e.g. from 55,523 to EQ-5D-5L value sets. 44,412). The calculations were the same as in the break - When using the EQ-5D-3L value set, it is evident that through scenario. the concave relation between health status and utility was notable for Japan, Denmark, and Zimbabwe (Table 1 and Fig. 3), with Japan ranked as the country with the highest Results order of concavity. On the other hand, a convex relation We found four utility sets for countries where utilities was found for the Netherlands, Germany, UK, Spain, and were elicited by means of SG [27–29]. Since this number Thailand, with Germany ranking as the country with the was too low for a meaningful analysis, only results for lowest order of concavity. studies based on TTO were reported. We identified TTO Based on EQ-5D-5L value set, the concave relation of based value sets for the following countries: The EQ- health status and utility were found for Japan, Denmark, 5D-5L based value sets consisted of 10 countries, namely, Zimbabwe, and Thailand (Fig. 4). Similar to the results Japan, Denmark, France, Germany, the Netherlands, from the EQ-5D-3L value set, Japan was ranked as the Spain, Thailand, UK, US, and Zimbabwe. Seven countries country with the highest order of concavity, indicating were identified from EQ-5D-3L, including Japan, Den - the highest degree of long-term preference, or, in other mark, Germany, the Netherlands, Spain, Thailand, UK, words, the lowest time preference rate. On the other and Zimbabwe. hand, a convex relationship was detected for France, Table 1 Second order quadratic utility functions Country 2-nd order quadratic line Tool Interpretation Rank Rank (equation 5D-5D-5L) (EQ-5D-3L) Japan Y = − 0.0000000550x + 0.0003563210x − 0.0742209093 EQ-5D-5L Concave 1 – Japan y = − 0.0000062x + 0.0045459x − 0.0090418 EQ-5D-3L Concave – 1 Denmark Y = − 0.0000000446x + 0.0004197915x − 0.1655666889 EQ-5D-5L Concave 2 – Denmark y = − 0.0000054x + 0.0057193x − 0.2949838 EQ-5D-3L Concave – 2 France Y = 0.0000000237x + 0.0001943942x − 0.2303118192 EQ-5D-5L Convex 6 – Germany Y = 0.0000000323x + 0.0001660183x + 0.0082592409 EQ-5D-5L Convex 10 – Germany y = 0.0000094x + 0.0014504x – 0.0378014 EQ-5D-3L Convex – 8 Netherlands Y = 0.0000000268x + 0.0001734792x − 0.0647394984 EQ-5D-5L Convex 8 – Netherlands y = 0.0000055x + 0.0021384x − 0.0380753 EQ-5D-3L Convex – 5 Spain Y = 0.0000000251x + 0.0002372594x − 0.2745223392 EQ-5D-5L Convex 7 – Spain y = 0.0000083x + 0.0025337x − 0.3504985 EQ-5D-3L Convex – 7 Thailand Y = − 0.0000000134x + 0.0002816214x − 0.1901163980 EQ-5D-5L Concave 4 – Thailand y = 0.0000015x + 0.0033337x − 0.2652283 EQ-5D-3L Convex – 4 UK Y = 0.0000000299x + 0.0002053902x − 0.2343214473 EQ-5D-5L Convex 9 – UK y = 0.0000079x + 0.0023519x − 0.3067348 EQ-5D-3L Convex – 6 USA Y = 0.0000000092x + 0.0001832790x + 0.0991735591 EQ-5D-5L Convex 5 – Zimbabwe Y = − 0.0000000189x + 0.0002390259x + 0.1619338051 EQ-5D-5L Concave 3 – Zimbabwe y = − 0.0000025x + 0.0034765x + 0.0751729 EQ-5D-3L Concave – 3 The ranking of concavity was based on the second derivative of each country. Lowest second derivative was ranked as the number 1 of the ranking Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 5 of 8 Fig. 3 2nd-order quadratic predictions for longevity derived by EQ-5D-3L Germany, Netherlands, Spain, UK, and US. Once again, the UK that has the lowest ICER in both scenarios, the Germany was ranked as the country with the lowest value for Japan is twice as high. degree of concavity. The ranks in the Hofstede index [5, 30] as well as per Discussion utility function curvature are reported in Table 2. Our findings indicate that different utility values are The results for Japan are fairly consistent in that they associated with various cultural beliefs which may impact rank first in the Hofstede long-term orientation index as cost-utility analysis. We also found a low level of evidence well as in both rankings representing the degree of con- that countries with a long-term horizon (according to cavity. Conversely, we found conflicting results for both Hofstede) show a more concave profile of utility function. Germany and Denmark. Although Germany has a high Paradoxically, a social planner with a restricted budget degree of long-term orientation, similar to Japan, this is who aims at maximizing the global welfare (sum of utili- not reflected in the shape of the EQ-5D utility functions. ties) would potentially distribute less budget to countries For Denmark, the opposite holds true: long term orienta- with a low time preference rate. The reason is that those tion (LTO) according to Hofstede is lower than the utility countries already have higher utility values, resulting in functions would suggest. a smaller scope for improvement, because, by definition, The results of the scenario analysis are reported in utility values are capped at 1. Only in health states that Table 3. We found that the ICER of both the break- close to death are the slopes of the utility functions of low through innovation as well as the incremental innovation time preference rate countries like Japan higher, as are is highest in Japan, indicating that the value it brings in marginal utilities when health is improving. Such coun- terms of additional utilities is the lowest. Compared with tries would therefore benefit more from interventions Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 6 of 8 Fig. 4 2nd-order quadratic predictions for longevity derived by EQ-5D-5L providing only marginal improvement of a patient’s now introducing an HTA system that is modeled on the health state instead of restoring perfect health. Other UK system, using cost-utility analysis as a basis . If countries with a high time preference would consider- the equivalent threshold is to be applied, the same drugs ably benefit from interventions providing restoration of would then sell at lower prices in Japan than in the UK, a perfect health state (utility level 1). The results of the due to differences in utility values that in turn differ due scenario analyses further support these assessments. For to different time preference rates. the two hypothetical drugs, the calculated ICER for Japan While according to Hofstede time preference is cul- was twice that of the UK, or 37% higher than that of the turally determined, there might be other influenc - US. The implications of these findings for health tech - ing factors involved. In the economic literature, time nology assessment (HTA) are quite significant. Japan is preference is linked to a state’s level of economic Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 7 of 8 Table 2 Long term orientation index of included income. The authors observed that impatient children countries. Source: Hofstede et al. (2010)  perform worse in school and consequently earn less income, are more often unemployed, and more often Country Long term EQ-5D-3L EQ-5D-3L 5L orientation rank concavity rank concavity rank depend on welfare benefits. Moreover, the study even reported that impatient children are more likely to die Japan 1 1 1 young, become obese, or become pregnant while still in Germany 2 8 10 their teens . Subsequent research further demon- Netherlands 3 5 8 strated that impatient people are more likely to become France 4 – 6 involved in crimes later in life [40, 41]. UK 5 6 9 While it is difficult to disentangle cultural and indi - Spain 6 7 7 vidual drivers of time preference, we believe that Denmark 7 2 2 cultural factors shape preferences and should be Thailand 8 4 4 accounted for in health economics. One implication of USA 9 – 5 this debate is that the HTA concept of cost-utility anal- Zimbabwe Not applicable 3 3 ysis depends on cultural beliefs and values and cannot be easily transferred across cultures without significant adaptations. While we have looked at only two dimen- Table 3 Results of scenario analysis sions of Hofstede’s concept of culture, we acknowledge Germany Japan UK US other cultural values. Hofstede himself defined four other pillars, namely power distance, individualism, Break through innovation masculinity, and indulgence. Religion is another impor- Incremental cost (JPY ) 1,000,000 1,000,000 1,000,000 1,000,000 tant part of cultural values that either influence Hof - Incremental QALY 0.506 0.289 0.575 0.400 stede’s dimensions or even constitute an independent ICER (JPY per QALY ) 1,976,121 3,463,208 1,738,434 2,499,378 pillar of culture impacting on choices of patients . Incremental innovation Future research could apply a more holistic approach to Incremental cost (JPY ) 1,000,000 1,000,000 1,000,000 1,000,000 culture and relate it to health economics concepts. Incremental QALY 0.340 0.194 0.387 0.269 Another limitation is that some of the EQ-5D-5L ICER (JPY per QALY ) 2,939,257 5,151,130 2,585,724 3,717,543 value sets were mapped from the EQ-5D-3L and those UK United Kingdom, US United States; QALY quality-adjusted life year, ICER values are therefore not directly obtained preferences. incremental cost-effectiveness ratio, JPY Japanese Yen We cannot rule out the possibility that the use of cross- walk value sets influences the results. development . Time series data analysis for instance can show that for Taiwan and Japan, the time prefer- Conclusion ence rate decreased up to a certain point during eco- We maintain that cultural beliefs determine utility val- nomic catch-up, with a further decline afterwards as ues that are beneficial in cost-utility analysis and health the populations became more hedonistic . Fur- technology assessments. Countries such as Japan are thermore, individual factors might play a role as there characterized by a long-term time horizon, or a low is a potentially large variation in preferences within time preference rate. Cultural practices and beliefs are the same country. Low educational status for instance reflected in a concave shape of a country’s utility func - was identified as a key correlate of a high time prefer - tion. Such a curvature profile implies that incremental ence which in turn contributes to unhealthy behavior health gains are valued less. Accordingly, health-care such as smoking [34, 35]. Conversely, individuals with interventions create different value in different cultural a long-term horizon are more likely to participate in settings that should be accounted for in HTA. Conse- higher education and adopt a healthier lifestyle . quently, our study further supports the accumulating The relationship between long-term orientation, cogni - literature that argues against a one-size-fits-all cost- tive function and health together with other outcomes effectiveness approach . Instead, we maintain that has been documented in several studies [37, 38]. Gol- a context-sensitive, multiple-criteria decision-making steyn et al. for instance asked 13,606 Swedish children approach is warranted that should include values, cul- aged 13 whether they would prefer to receive $140 now tural and country specific goals and goes beyond pure or $1400 in 5 years. The study traced the children’s cost-effectiveness. Or, as it was expressed by Chalkidou long-term achievements regarding education, fertility et al. ‘[take] into account local values is the holy grail decisions, health indicators, labor market success, and for country empowerment’ . Mahlich et al. Cost Eff Resour Alloc (2018) 16:19 Page 8 of 8 Authors’ contributions 15. Harbaugh WT, Krause K, Vesterlund L. Risk attitudes of children and JM designed the study and drafted the manuscript, PD and RS performed the adults: choices over small and large probability gains and losses. Exp analysis. NC verified the analytical methods and helped to interpret the results. Econ. 2002;5(1):53–84. All authors read and approved the final manuscript. 16. Croson R, Gneezy U. Gender differences in preferences. J Econ Lit. 2009;47(2):448–74. Author details 17. Golsteyn B, Heckman J, Meijers H. Gender differences in risk aversion and Health Economics, Janssen Pharmaceutical KK, 5-2, Nishi-kanda 3-chome ambiguity aversion. J Eur Econ Assoc. 2009;7(2–3):649–58. Chiyoda-ku, Tokyo 101-0065, Japan. Düsseldorf Institute for Competition Eco- 18. Khodarahimi S. Sensation-seeking and risk-taking behaviors: a study on nomics (DICE), University of Düsseldorf, Düsseldorf, Germany. Center of Phar- young Iranian adults.Appl Res Qual. Life. 2015;10(4):721–34. maceutical Outcomes Research, Naresuan University, Phitsanulok, Thailand. 19. Bayyurt N, Karışık V, Coşkun A. Gender differences in investment prefer - School of Pharmacy, Monash University Malaysia, Subang Jaya, Malaysia. ences. Eur J Econ Political Stud. 2013;6(1):71–83. 20. Barber BM, Odean T. Boys will be boys: gender, overconfidence, and com- Competing interests mon stock investment. Q J Econ. 2001;116(1):261–92. JM and RS were employed at Janssen KK at the time the study was performed. 21. Euroqol Foundation. Instruments. http://www.euroq ol.org. Accessed 10 Jan 2017. Availability of data and materials 22. RAND corporation. 36-Item short form survey. https ://www.rand.org/ All data are publicly available and are available from the authors upon reason-healt h/surve ys_tools /mos/36-item-short -form.html. Accessed 10 Jan able request. 2017. 23. Szende A, Oppe M, Devlin N. EQ-5D value sets: inventory comparative Consent for publication review and user guide. Dordrecht: Springer; 2007. Not applicable. 24. Tongsiri S, Cairns J. Estimating population-based values for EQ-5D health states in Thailand. Value Health. 2011;14:1142–5. Ethics approval and consent to participate 25. Van Hout B, Janssen MF, et al. Interim scoring for the EQ-5D-5L: mapping Not applicable. the EQ-5D-5L to EQ-5D-3L value sets. Value Health. 2012;15(5):708–15. 26. Pratt J. Risk aversion in the small and in the large. Econometrica. Funding 1966;32(1/2):122–36. The study was funded by Janssen KK. 27. Brazier JE, Fukuhara S, Roberts J, et al. Estimating a preference-based index from the Japanese SF-36. J Clin Epidemiol. 2009;62:1323–31. 28. Cruz JN, Camey SA, Hoffmann JF, et al. Estimating the SF-6D value set for Publisher’s Note a population based sample of Brazilians. Value Health. 2011;14:S108–14. Springer Nature remains neutral with regard to jurisdictional claims in pub- 29. Craig BM, Pickard AS, Stolk E, Brazier JE. US valuation of the SF-6D. Med lished maps and institutional affiliations. Decis Making. 2013;33:793–803. 30. Hofstede G, Hofstede GJ, Minkov M. Cultures and organizations: software Received: 3 October 2017 Accepted: 24 May 2018 of the mind. revised and expanded. 3rd ed. New York: McGraw-Hill; 2010. 31. Mahlich J, Kamae I, Rossi B. A new health technology assessment system for Japan? Simulating the potential impact on the price of Simeprevir. Int J Technol Assess Health Care. 2017;33(1):121–7. 32. Uzawa H. An endogenous rate of time preference, the Penrose effect, References and dynamic optimality of environmental quality. Proc Natl Acad Sci USA. 1. Weinstein MC, Torrance G, McGuire A. QALYs: the basics. Value Health. 1996;93(12):5770–6. 2009;12:S5–9. 33. Ogawa K. Economic development and time preference schedule: the 2. Lenert L, Kaplan RM. Validity and interpretation of preference-based Case of Japan and East Asian NICs. J Dev Econ. 1993;42:175–95. measures of health-related quality of life. Med Care. 2000;38(9):138–50. 34. Peretti-Watel P, L’Haridon O, Seror V. Time preferences, socioeconomic 3. Shiroiwa T, Fukuda T, Ikeda S, Igarashi A, Noto S, Saito S, et al. Japanese status and smokers’ behaviour, attitudes and risk awareness. Eur J Public population norms for preference-based measures: EQ-5D-3L, EQ-5D-5L, Health. 2013;23:783e–8e. and SF-6D. Qual Life Res. 2016;25(3):707–19. 35. Jusot F, Khlat M. The role of time and risk preferences in smoking inequali- 4. Burstrom K, Johannesson M, Diderichsen F. A comparison of individual ties: a population-based study. Addict Behav. 2013;38:2167–73. and social time-trade-off values for health states in the general popula- 36. Farrell P, Fuchs V. Schooling and health: the cigarette connection. J Health tion. Health Policy. 2006;76(3):359–70. Econ. 1982;1(3):217–30. 5. Hofstede G. Culture’s consequences: comparing values, behaviors, institu- 37. Dohmen T, Falk A, Huffman D, Sunde U. Are risk aversion and impatience tions, and organizations across nations. Thousand Oaks: Sage; 2001. related to cognitive ability? Am Econ Rev. 2010;100(3):1238–60. 6. Hofstede G, Bond M. The confucius connection: from cultural roots to 38. Benjamin D, Brown S, Shapiro J. Who is ‘behavioural’? cognitive ability and economic growth. Organ Dyn. 1988;16(4):4–21. anomalous preferences. J Eur Econ Assoc. 2013;11(6):1231–55. 7. Gafni A. The standard gamble method: what is being measured and how 39. Golsteyn B, Grönqvist H, Lindahl L. Adolescent time preferences predict it is interpreted. Health Serv Res. 1994;29(2):207–24. lifetime outcomes. Econ J. 2014;124:F739–61. 8. von Neumann J, Morgenstern O. Theory of games and economic behav- 40. Åkerlund D, Golsteyn B, Grönqvist H, Lindahl L. Time discounting and ior. London: Wiley; 1944. criminal behavior. Proc Natl Acad Sci USA. 2016;113(22):6160–5. 9. Hofstede G. Culture’s consequences: international differences in work- 41. Nagin D, Pogarsky G. Time and punishment: delayed consequences and related values. Beverly Hills: Sage; 1980. criminal behavior. J Quant Criminol. 2004;20:295–317. 10. Al Obaidi L, Mahlich J. A potential gender bias in assessing quality of 42. Ohr S, Jeong S, Saul P. Cultural and religious beliefs and values, and their life—a standard gamble experiment among university students. Clinico- impact on preferences for end-of-life care among four ethnic groups of Economics Outcomes Res. 2015;7:227–33. community-dwelling older persons. J Clin Nurs. 2017;26(11–12):1681–9. 11. Dohmen T, Falk A, Huffman D, Sunde U, Schupp J, Wagner GG. Individual 43. Balthussen R, Jansen MP, Mikkelsen E, Tromp N, Hontelez J, Bijlmakers L, risk attitudes: measurement, determinants, and behavioral conse- Van der Wilt G. Priority setting for universal health coverage: we need quences. J Eur Econ Assoc. 2011;9(3):522–50. evidence-informed deliberative processes, not just more evidence on 12. Eckel CC, Grossman PJ. Forecasting risk attitudes: an experimental costeffectiveness. Int J Health Policy Manag. 2016;5(11):615–8. study using actual and forecast gamble choices. J Econ Behav Organ. 44. Chalkidou K, Glassman A, Marten R, et al. Priority-setting for achieving 2008;68(1):1–17. universal health coverage. Bull World Health Organ. 2016;94(6):462–7. 13. Hartog J, Ferrer-i-Carbonell A, Jonker N. Linking measured risk aversion to individual characteristics. Kyklos. 2002;55(1):3–26. 14. Holt CA, Laury SK. Risk aversion and incentive effects. Am Econ Rev. 2002;92(5):1644–55.
Cost Effectiveness and Resource Allocation
– Springer Journals
Published: Jun 1, 2018