Willingness to pay for a quality adjusted life year in bavi district hanoi 2014

By using a “ceiling threshold”, any health technology producing one QALY or one year living in full health gained with its cost less than the ceiling threshold is considered cost-effecti

Background: Cost effective threshold is

essential in an economic evaluation This

study aimed to estimate the willingness to pay

(WTP) for a Quality Adjusted Life Year

(QALY) in Bavi district, Hanoi 2014 and

examine some associated factors Method:

360 respondents from Bavi district, Hanoi

were interviewed Dichotomous bidding

choice followed by open-ended question was

employed in this study Results: Mean of

willingness to pay for a Quality Adjusted life

year in Bavi, Hanoi, 2014 ranged from

13,934,010 to 20,737,620 VND (~667.3 –

993.1$ US) The WTP per QALY for worse

health states are higher than those for better

states Gender, utility of health status assessed

by respondents and monthly household

income were determined as associated factors

Conclusions: The WTP/QALY values were

slightly lower than the recommendation of WHO It is recommended to have more than one threshold for every situation based on the severity

Keywords: contingent valuation, dichotomous bidding choice, quality adjusted life year, willingness to pay

INTRODUCTION

Cost-effectiveness analysis (CEA) is essential for allocating healthcare resources more efficiently In CEA, the additional consumption of medical resources is divided

by the benefits (e.g quality-adjusted life-years) gained from healthcare interventions, in order to calculate an incremental cost-effectiveness ratio (ICER) Generally, an

Willingness to pay for a Quality Adjusted

Life Year in Bavi district, Hanoi 2014

Bui Cam Nhung 1* , Kim Bao Giang 1 , Nguyen Hoang Thanh 1 ,

Doan Thu Huyen 1 , Nguyen Hoang Long 2 , Hoang Van Minh 1

1 Hanoi Medical University, Hanoi, Vietnam

2 School of Medicine and Pharmacy, Vietnam National University, Hanoi, Vietnam

* Corresponding Author: Bui Cam Nhung, Hanoi Medical University, 1 Ton That Tung, Hanoi, Vietnam

Email: nhung305hmu@gmail.com

intervention is considered cost effective if the

ICER (e.g cost per QALY) is below a

predetermined threshold By using a “ceiling

threshold”, any health technology producing

one (QALY) or one year living in full health

gained with its cost less than the ceiling

threshold is considered cost-effective1, 2

At present, an arbitrary threshold of US$

50,000 per Quality-Adjusted Life Year as well

as the thresholds of 1-3 times of Gross

Domestic Product (GDP) per capita per

Disability-Adjusted-Life Year (DALY)

recommended by the Commission on

cited with several arguments4-6 In England, a

National Institute for Health and Clinical

Excellence (NICE) refer to an arbitrary

Nevertheless, rather than an arbitrary ceiling

threshold, a WTP/QALY value, estimated by

combining WTP and utility value measured

simultaneously, should be adopted as a ceiling

threshold

In fact, country-specific threshold is essential

because different countries have different

affordability and preference with respect to

how much health care resources would be

located In recent years, Vietnam Ministry of

Health has recognized the importance of

health technology assessment, in which CEA

is an essential component for the development

gained threshold with the context of Vietnam

will be a great support to the implementation

of health technology assessment in the future

However, there is no survey to determine the

threshold of cost-effectiveness in Vietnam

This study is the first step to examine the

threshold for cost-effectiveness in Vietnam

with the aim of estimating the WTP/QALY

values as well as examine the associated factor

to WTP associated in Bavi district, Hanoi, by using three hypothetical scenarios about improving quality of life in mild, moderate, and severe health conditions; and measuring the WTP of people by contingent valuation method

METHOD

Study design and sampling

A cross-sectional study was conducted in May

2014 in Bavi A multi-stage sampling technique was implemented to ensure the representativeness of population Firstly, all communes were classified into three regions based on their geographic locations (Mountainous, riparian and hilly area) Then, five communes in each region were randomly selected resulted in 15 communes Finally, subjects in each household of selected communes meeting the eligibility criteria were randomly chosen for interview Inclusion criteria included: 1) age between 18-60 years, and 2) able to read and write Vietnamese The exclusion criteria were: 1) be a student (who cannot make decisions on financial matters), 2) inability to answer a series of complex theoretical questions and 3) refuse to participate in the study

Study instrument The questionnaire comprised three main components: demographic characteristics, health utility measure and scenarios to measure WTP per one QALY gained There were three versions of questionnaire Each respondent was asked to answer only one version of the questionnaire

Utility measure The EQ-5D-3L version was used, which consisted of two parts: the EQ-5D-3L

descriptive system and the EQ visual

Analogue scale (VAS) The former comprises

the 5 following dimensions: mobility,

self-care, usual activities, pain/discomfort and

anxiety/depression with 3 response levels: 1 as

no problems, 2 as some problems, 3 as extreme

problems9 Each of health state was assigned a

preference weight by using tariff of general

population based on time trade-off, standard

single index reflects full health (as 1) and

death (as 0) In some health states, the index

has negative value, suggesting the health states

are considered to be worse than death9 Among

tariffs of many countries in the world,

Thailand is a country in the same region and

has similar socioeconomic and cultural

characteristics to Vietnam comparing to other

countries Therefore, Thailand’s tariff, with a

range from -0.454 to 1, was applied to convert

the index10 For VAS, the respondents were

asked to look at the scale of 20 cm, 0-100

thermometer scale where 100 is labeled “The

best health state or perfect health“, and 0 is

labeled “the worst health state or dead”9

To measure health utility, the respondents were

firstly asked to indicate his/her health state

according to five questions They were also

asked to indicate their current health state by

VAS Then each respondent was assigned to

imagine being in 1 hypothetical health state

based on his/her version of questionnaire

Each hypothetical health state was also

described by 5 dimensions of EQ-5D-3L

instrument Finally, they were asked to rate the

hypothetical health state by VAS

The population-based values for EQ-5D health

states derived from Thailand population’s

study were used to establish five hypothetical

scenarios for measuring WTP 10 Three health

states were selected for the scenarios: 11212

represented for mild health states (utility

>0.7); 22222 represented for moderate health states (utility =0.36 – 0.7); and 22332 represented for severe health state (utility < 0.36) In which, 11212 indicates a health status

of having no problem in mobility, self-care and

no pain/discomfort but having some problems

in usual activities and anxiety/depression A health state of 22222 shows some problems in all dimensions including mobility, self-care, usual activities, pain/discomfort, and anxiety/depression 22332 represents for a health state with some problems in mobility, self-care, anxiety or depression but having extreme problems in usual activities and extreme pain/discomfort

In order to avoiding ceiling effect, each questionnaire version contained two scenarios for 0.2 or 0.4 QALY gained Time spent for treatment in each health states was calculated based on the formula bellowed:

WTP measure Double-Bounded dichotomous bidding technique followed by open-ended question was performed to examine respondents’ WTP per one QALY gained Based on the pilot survey and the information about GDP per capita of Vietnam in 2012, four different starting prices were selected for the study The list of prices was described in Table 1

Table 1 Bid values in double-bound

dichotomous choice

A specified period of time being in that

hypothetical health state followed by complete

recovery was assumed Respondents were

asked to indicate his/her WTP for the treatment

that can make him/her immediately recover to

perfect health (EQ-5D state: 11111) He/she

had to pay out-of-pocket one time within the

next 6 months To avoid starting point bias,

each respondent was randomly assigned on a

certain starting price The yes/no answer to the

first price offered to the respondent determine

the next price offered If the answer is “yes”,

the bid amount increased in the second bid If

the initial answer is “no”, the bid amount

would be reduced The open-ended question

was asked after the second bidding to examine

the maximum WTP amount

Statistical analysis

STATA version 12.0 was used to analyze the

data Student-t, ANOVA, Kruskal-Wallis and

χ2test were used to determine the differences

in demographic characteristics among three

levels of hypothetical health status From

open-ended response, WTP/QALY value was

calculated using disaggregated approach

(Mean of ratios) based on the following

formula:

Multivariate analysis (logistic regression and

linear regression) were conducted to examine

the related factors to the proportion and WTP

for a QALY value after adjusted for possible

confounders Results with p<0.05 is

statistically significant

RESULTS

The socio-demographic information of 360

respondents classified by each version is

shown in Table 2 Respondents were predominantly female (51.9%), and mean age was 42.6 years old with the standard deviation

of 10.1 years Most of people were married, farmers, in secondary or high school level of education, head of the household and in good health (EQ-5D 0.74; EQ-VAS 0.75) The utility of given health status assessed by respondents was lower than their health status and decreases from mild to severe health states No significant differences across questionnaire versions were found in health status of respondents

Table 3 displays the general information of the households Average income of the households was 8,268,234VND while the mean of households spending was 5,844,103 VND The proportion of living area of household was equivalent No significant differences across questionnaire versions were found of any variables of general information of households

Table 2 Socio-demographic characteristics of

the study respondents

*Testing across 3 versions using Chi-square test for gender, education, occupation, marriage, status of respondents variables; and ANOVA test for age variable

Figure 1 shows mean/median WTP/QALY

values for each scenario The overall average

of WTP/QALY for 0.2 QALY gained scenario

was 20,737,620 VND compared with average

13,934,010 VND WTP/QALY for 0.4 QALY

gained scenario The lower utility score of the

health states was, the higher WTP/QALY

value was estimated The WTP/QALY for 0.2

scenarios was higher than the WTP/QALY for

0.4 scenarios in every health state

According to Table 4, the proportion of willing

to pay respondents for first bidding choice in

0.2 QALY gained scenario ranged from 55% to 63% with average 58%, while slightly higher proportion was presented in 0.4 QALY gained scenario For the second bidding choice, lower proportion than first bidding choice was found The proportion of agreement for second bidding choice range was from 42% to 54% in 0.2 QALY gained scenario and from 48% to 58% in 0.4 QALY gained scenario In both bidding choice and scenarios, moderate health state had the lowest proportion of respondents who are willing to pay in all scenarios

Table 5 shows the association between some related factors with the proportion of willingness to pay after adjusted for other factors shown in the table using multivariable logistic regression Gender, utility of health status assessed by respondents and monthly household income were found to be significant predictors of whether the respondents would pay or not

DISCUSSION

Currently, there has been a numerous of evidences concerning the WTP for one additional QALY11-15 Our study was the first study examining the WTP of one QALY gained in the context of Vietnam In this study, mean of WTP/QALY derived from open-ended question varied approximately from 13,934,000 VND (0.4 QALY gained) to 20,738,000 VND (0.2 QALY gained) It was

Table 3 General information of the study

households

* Testing across 3 versions using Chi-square test for living area

variable; Kruskal-Wallis test for income, household spending,

number of people in family variables; and ANOVA test for

number of under-6-year-old children variable.

Figure 1 Mean of WTP for each scenario

derived from open-ended question

Table 4 Proportion of willingness to pay across

bidding choices

calculated equal to 0.38 – 0.56 GDP per capita

of Vietnam in 2012, which was two times less

than the WTP/QALY estimated from Thai

and markedly lower than the ranged of 1 - 3

times of GDP per capita/QALY, recommended

by the Commission on Macroeconomics and

interpreted for this results were that this study

was conducted in rural area (approximately

60% of respondents was agriculture) and

35.7% of respondents living in difficult

economic area (mountainous area), therefore

the affordability can be lower than urban area

The founded mean of WTP/QALY from this

study was approximately 667.3 – 993.1$ US

(Average exchange rate in 2012: 1USD =

20,882.6 VND17) Similarly, compared to other

WTP/QALY studies conducted in other Asian

countries as well as Europe countries6, 7, 18,19,this

result was considerably lower many times

However, these countries are developed

countries with higher living standard of

people, by dint of which the comparison is

unsatisfactory

Consistent with the previous studies, the severity of the health states influenced the WTP for a QALY15, and it should be more than one ceiling threshold for all situations15, 20-22 In this study, the mean of WTP/QALY was higher showing the forward trend from mild to severe health states Additionally, the proportion of unwillingness to pay in mild health states is higher than moderate and severe health states Such these behavior can be interpreted through the Health Belief Model23 When the issue is not considered as serious (perceived seriousness), individuals are less likely to take that action (willing to pay) Therefore, when making interventions plan in community, the seriousness of health event and attention of community should be considered carefully The proportion of unwillingness to pay in 0.2 QALY gained scenario (11.4%) was lower than in 0.4 QALY gained scenario (10.0%) For mild health state, most of unwilling to pay respondents claimed that this health state was not too bad, so they can live with it For severer health states, limited financial was the most reason reported by unwilling to pay respondents That unwilling to pay anything at all to avoid some duration of that health state does not accord with economic theory according to which goods are only valued if individuals are willing to pay a positive amount for them24 The implication of these findings needs further consideration

The number of QALY gained chosen in this study was tested through pilot test to ensure the avoidance of ceiling effect The ceiling effect happens when subjects are faced with a large QALY gained leading to less amount of WTP than expected due to the limited finance The chosen 0.2 and 0.4 QALY gained are neither too small nor too large If the chosen QALY gained is too small, the high proportion

Table 5 Related factors to willingness to pay for

a QALY using logistic regression

* p<0.5; **p<0.01

of unwillingness to pay leads to inaccurately

estimated value of a QALY If it is too large,

the ceiling effect will occur In this study, the

respondents are allowed to pay money within

6 months and by multiple sources of money

Therefore, ceiling effect can be controlled in

this study

The non-linear relationship between WTP and

QALY gained can be observed clearly in this

study Albeit the increase from 0.2 to 0.4

QALY gains is 2 times, the actual amount of

WTP for 0.2 QALY gained and 0.4 QALY

gained observed in this study ranged only from

1.3 to 1.5 times Consistent with the previous

knowledge, the relationship between WTP and

QALY is curving inward possibly due to

declining utility and ceiling effect25, 26

Using multivariable logistic regression,

several related factors to proportion of WTP

were found in our study Male respondents

were 2.08 to 2.32 times more likely to pay for

treatment than female respondents This was

probably explained that the majority of male

respondents were the head of the household

and they were more independent when making

decisions Even though male respondents have

more high risk behavior but males are more

permissive than females who manage finance

of whole family Willingness to pay higher in

men than in women was also observed in other

willingness to pay studies27, 28

Utility assessment of giving health status was

an associated factor to the proportion of WTP

for a QALY For each second increase in utility

assessment of given health status, the odd of

willingness to pay increases from 0.96 to 0.98

times This means the further low utility

assessment, the more proportion of WTP for

that health condition treatment In other words,

if the respondents are possible to understand

and evaluate the hypothetical condition properly, they probably willing to pay higher Therefore, the role of interviewers in this study was extremely crucial

Albeit monthly household income was statistically significant, it was accounted for modestly change in the WTP proportion For a one-unit of monthly household income changes, the odd of willingness to pay raised

or fell 1.0001 times However, household income was expected to affect the willingness

to pay of respondents stronger than observed result in this study This may due to small sample size Thus, further investigation should

be conducted to clarify this associated factor Gender, utility of given health status, education level and monthly household income were concluded to be the most related factor to both the proportion and amount of willingness to pay This result was consistent with the related factors founded in Thailand study16and other study18, 20

There are also some limitations in this study Initially, this study was conducted in a rural district of Vietnam and the sampling might not represent Vietnam population For more evidences for decision-making in Vietnam, the further studies advisedly employed larger sample size and represented for Vietnam population Regarding to starting price of 0.4 QALY gained scenario, it was selected to be equal to starting price of 0.2 QALY gained scenario, on account of which the respondents may not recognize the difference between two scenarios leading to unchanged amount of willingness to pay in the second situation In addition to this, low level of education of respondents caused difficulties in understanding the hypothetical condition, and this would be a limitation of this study

Health technology assessment and

cost-effectiveness analysis are progressively

crucial in making decisions in Vietnam This

study was conducted with the aim to estimate

the willingness to pay for an additional QALY

in Vietnam and associated factors Despite the

limitations, this study is the first step in

estimating the social ceiling threshold in

Vietnam The mean of willingness to pay for

three health states ranged from 13,934,010 to

20,737,620 VND (~0.38 – 0.56 GDP per capita

of Vietnam in 2012) According to our

findings, the WTP per QALY for worse health

states are higher than those for better states

Therefore, it is recommended to have more

than one threshold for every situation based on

the severity

ACKNOWLEDGEMENT

We would like to thank Center of Health System Research – Hanoi Medical University for financial support for this study We would also like to acknowledge Dr Montarat Thavorncharoensap for supporting the research methods, the investigators from FILABAVI as well as all the respondents contributed in this study

DECLARATION OF CONFLICTING INTERESTS

We have read and understood VJPH policy on declaration of interests and declare the following interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article

REFERENCES

1 Drumond MF, Stoddart GL Principles of economic

evaluation of health programmes World health

statistics quarterly Rapport trimestriel de

statistiques sanitaires mondiales

1985;38(4):355-367.

2 Mills A Economic evaluation of health

programmes: application of the principles in

developing countries World health statistics

quarterly Rapport trimestriel de statistiques

sanitaires mondiales 1985;38(4):368-382.

3 WHO Commission on Macroeconomics and

Health Macroeconomics and health: investing in

health for economic development Geneva:

Report of the Commission on Macroeconomics

and Health: World Healh Organization;2001.

4 Ubel P, Hirth R, Chernew M, A F What is the price

of life and why doesn't it increase at the rate of

inflation? Arc Int Med 2002;163:1637-1641.

5 Grosse SD Assessing cost-effectiveness in

healthcare: history of the $50,000 per QALY

threshold Expert review of pharmacoeconomics

& outcomes research Apr 2008;8(2):165-178.

6 Hirth R A, Chernew M E, Miller E, Fendrick M, Weissert W G Willingness to pay for a quality-adjusted life year: in search of a standard Medical decision making : an international journal of the Society for Medical Decision Making 2000(20):332-342.

7 Littlejohns P Does NICE have a threshold? A response In: Towse A PC, Devlin N, ed Cost-effectivenss thresholds Economic and ethical issues London: King's Fund and Office of Health Economics; 2002.

8 Health VIetnam Mo, Group HP Joint Annual Health Review 2013: Towards Universal Health Coverage Hanoi: Ministry of Health;2013.

9 Group E EQ-5D-3L User Guide: Basic information

on how to use the EQ-5D-3L instrument 2013.

10 Tongsiri S, Cairns J Estimating population-based values for EQ-5D health states in Thailand Value in health : the journal of the International Society for

Pharmacoeconomics and Outcomes Research Dec

2011;14(8):1142-1145.

11 Bobinac A, Van Exel NJ, Rutten FF, Brouwer WB.

Willingness to pay for a quality-adjusted life-year:

the individual perspective Value in health : the

journal of the International Society for

Pharmacoeconomics and Outcomes Research Dec

2010;13(8):1046-1055.

12 Byrne MM, O'Malley K, Suarez-Almazor ME.

Willingness to pay per quality-adjusted life year in

a study of knee osteoarthritis Medical decision

making : an international journal of the Society for

Medical Decision Making Nov-Dec

2005;25(6):655-666.

13 King JT, Jr., Tsevat J, Lave JR, Roberts MS.

Willingness to pay for a quality-adjusted life year:

implications for societal health care resource

allocation Medical decision making : an

international journal of the Society for Medical

Decision Making Nov-Dec 2005;25(6):667-677.

14 Thavorncharoensap M, Teerawattananon Y,

Natanant S, Kulpeng W, Yothasamut J,

Werayingyong P Estimating the willingness to pay

for a quality-adjusted life year in Thailand: does

the context of health gain matter?

ClinicoEconomics and outcomes research : CEOR.

2013;5:29-36.

15 Shiroiwa T, Igarashi A, Fukuda T, Ikeda S WTP for a

QALY and health states: More money for severer

health states? Cost effectiveness and resource

allocation : C/E 2013;11:22.

16 Montarat Thavorncharoensap, Pattara

Leerahavarong, al e Asian study on value for a

Asian study on value for a QALY: Thailand Health

Intervention and Technology Assessment

Program; 2012.

17 Limited FM USD/VND - US Dollar Vietnamese

Dong Historical Data 2014;

http://www.investing.com/currencies/usd-vnd-historical-data.

18 Shiroiwa T, Igarashi A, Fukuda T, Ikeda S WTP for a

QALY and health states: More money for severer

health states? Cost effectiveness and resource

allocation : C/E Sep 1 2013;11(1):22.

19 Shiroiwa T, Sung YK, Fukuda T, Lang HC, Bae SC, Tsutani K International survey on willingness-to-pay (WTP) for one additional QALY gained: what is the threshold of cost effectiveness? Health economics 2010;19(4):422-437.

20 Gyrd-Hansen D Willingness to pay for a QALY: theoretical and methodological issues PharmacoEconomics 2005;23(5):423-432.

21 Lieu TA, Ray GT, Ortega-Sanchez IR, Kleinman K, Rusinak D, Prosser LA Willingness to pay for a QALY based on community member and patient preferences for temporary health states associated with herpes zoster PharmacoEconomics 2009;27(12):1005-1016.

22 Zhao FL, Yue M, Yang H, Wang T, Wu JH, Li SC Willingness to pay per quality-adjusted life year: is one threshold enough for decision-making?: results from a study in patients with chronic prostatitis Medical care Mar 2011;49(3):267-272.

23 Nancy K Janz R, MS The Health Belief Model: A Decade Later Health Education Behavior 1984;11(1):1-47.

24 Mason H, Jones-Lee M, Donaldson C Modelling the monetary value of a QALY: a new approach based on UK data Health economics Aug 2009;18(8):933-950.

25 Bobinac A, van Exel NJ, Rutten FF, Brouwer WB GET MORE, PAY MORE? An elaborate test of construct validity of willingness to pay per QALY estimates obtained through contingent valuation Journal of health economics Jan 2012;31(1):158-168.

26 Smith RD, Richardson J Can we estimate the 'social' value of a QALY? Four core issues to resolve Health policy Sep 28 2005;74(1):77-84.

27 Van Minh H, Nguyen-Viet H, Thanh NH, Yang JC Assessing willingness to pay for improved sanitation in rural Vietnam Environmental health and preventive medicine Jul 2013;18(4):275-284.

28 Van Hoi L, Thi Kim Tien N, Van Tien N, et al Willingness to use and pay for options of care for community-dwelling older people in rural Vietnam BMC health services research 2012;12:36.