Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Patients’ willingness to pay for their drugs in primary care clinics in an urbanized setting in Malaysia: a guide on drug charges implementation

Patients’ willingness to pay for their drugs in primary care clinics in an urbanized setting in... Background: Malaysia is an upper middle income country that provides subsidized healthcare to ensure universal coverage to its citizens. The challenge of escalating health care cost occurs in most countries, including Malaysia due to increase in disease prevalence, which induced an escalation in drug expenditure. In 2009, the Ministry of Health has allocated up to Malaysian Ringgit (MYR) 1.402 billion (approximately USD 390 million) on subsidised drugs. This study was conducted to measure patients’ willingness to pay ( WTP) for treatment of chronic condition or acute illnesses, in an urbanized population. Methods: A cross-sectional study, through face-to-face interview was conducted in an urban state in 2012–2013. Systematic random sampling of 324 patients was selected from a list of patients attending ten public primary cares with Family Medicine Specialist service. Patients were asked using a bidding technique of maximum amount (in MYR) if they are WTP for chronic or acute illnesses. Results: Patients are mostly young, female, of lower education and lower income. A total of 234 respondents (72.2%) were not willing to pay for drug charges. WTP for drugs either for chronic or acute illness were at low at median of MYR10 per visit (USD 3.8). Bivariate analysis showed that lower numbers of dependent children (≤3), higher personal and household income are associated with WTP. Multivariate analysis showed only number of dependent children (≤3) as significant ( p = 0.009; 95% CI 1.27–5.44) predictor to drugs’ WTP. Conclusion: The result indicates that primary care patients have low WTP for drugs, either for chronic condition or acute illness. Citizens are comfortable in the comfort zone whereby health services are highly subsidized through universal coverage. Hence, there is a resistance to pay for drugs. Keywords: Willingness-to-pay, drugs expenditure, Acute diseases, Chronic conditions, Urban state Health delivers comprehensive medical, health, dental Background and pharmaceutical services at a subsidized rate lead- The increase in drug expenditure is a crucial challenge in ing to increase burden on the government. Since year Malaysia. As there is no established compulsory National 2000, the national drug expenditure has increased from Health Insurance yet, most of the patients obtain treat- MYR 346 million (USD 91 million) to MYR 915 million ment including prescription of drugs from government (USD 241 million) in 2005. From the year 2004–2005, funded health clinics and hospitals [1]. The Ministry of an increase of 13.3% was recorded [2] and according to the Pharmaceutical Services Division Annual Report, the *Correspondence: sh_ezat@ppukm.ukm.edu.my; sh_ezat@yahoo.com Malaysia Ministry of Health drug expenditure had esca- Faculty of Medicine, Department of Community Health, University Kebangsaan Malaysia Medical Center, Jalan Yaacob Latif, Bandar Tun lated to MYR 1.402 billion in 2009 [3]. Razak, Cheras, 56000 Kuala Lumpur, Malaysia Full list of author information is available at the end of the article © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/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://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 2 of 8 Numerous factors [2, 3] are found to contribute the future, a national health insurance system has been to an increase drug prices including growth of aging proposed to be implemented in the hope of reducing the population, higher consumer expectations, long-term cost burden of the government providers. drug treatment for chronic conditions, polypharmacy, Many primary care clinics in Selangor are equipped improvements in diagnostics/treatment of diseases, and with Family Medicine Specialist (FMS) and case-mix ser- novel expensive drugs offered due to advancement of vices (using the ICD-10 ambulatory coding and group- health technology [4]. As a result, the government has an ing). These facilities and logistics are placed due to the increasing demand on the rising health care costs [1, 3]. higher number of population seeking services, higher Provision for chronic conditions and acute diseases drugs patients’ expectancy, an increasing aging population and take place in many primary care facilities throughout the the increasing severe and complex case mix cases. state and yet, they are free of charge. A national health insurance has been proposed to be strategically imple- Methods mented in the government funded primary care centers Study population in Malaysia in the very near future [5]. Therefore, this This study is a cross-sectional study conducted in year study intends to measure patients’ WTP for their drugs 2012–2013 Selangor, Malaysia. The population represents who attend to public primary care services. both high and low income quintile level of population coming from both urban and rural areas. In each district, Willingness to pay for drugs there are two to six public primary clinics, equipped with Willingness to pay is a methodological tool to discover clinician with specialty trainings, for example the Family the hypothetical monetary value for programs and spe- Medicine Specialists (FMS). A total of 10 public primary cific medical interventions and treatments [6]. A study care with existing FMS services were chosen as those that examined the WTP for Praziquantel treatment (for clinics are visited by patients who suffer from various Schistosoma parasites) in Ogun State, Nigeria, showed types of diseases; managed with various types of medica- 92.3% respondents were willing to pay for the drug to tions including the “list A” drugs (i.e. for example drugs treat the infected household members [7]. However, in that can only be prescribed by specialists). a coronary restenosis (re-narrowing) study that assessed patients’ WTP to avoid revascularization procedure, Definition of “chronic condition” and “acute illness” found that the proportion of patients’ WTP is higher with This study aims to determine patients’ WTP for drugs, greater absolute risk reductions [8]. This is logical in the both for chronic conditions or acute illnesses. Chronic sense that higher risk would entail higher commitment condition is defined as conditions that had lasted or were and WTP. Similarly, in assessing WTP for cancer preven- expected to last 12 or more months and result in func- tion, the results revealed that income and the probability tional limitations and/or the need for on-going medi- of developing cancer were positively correlated to WTP cal care [10]. The WTP among patients were examined [9]. with three most prevalent chronic conditions in Malaysia Malaysia state healthcare system has been heavily which were hypertension, type-2 diabetes mellitus and financed through the government tax based financing, ischemic heart disease (IHD). Acute illness is defined although its health care prosper under bipartite public as rapid onset and/or short course disease less than and private providers. Under the universal health cover- 3  months duration, which is self-limiting or requiring age, Malaysia’s public out patients healthcare is provided minimal treatment [11]. almost fee exempted and nominal payment upon admis- sion as inpatient. Employer-based financing covers a lim - Sample size and sampling method ited number of companies and private insurance, that The sample size was calculated by considering the engage private health providers, mostly are concentrated assumption of two population proportion formula: the in urban areas. Primary care facilities in Malaysia are proportion (P1)  =  1.2% of low income that are willing highly subsidized by the government under the universal to pay, while proportion (P2) = 11% of high income that coverage, leading to very low state cost recovery. Malay- willing to pay [12]. The Z (standard normal) distribution sian citizens have to pay a user fee of only MYR1 (esti- value at 95% confidence level was taken at 1.96, 5% of mated to be USD 0.38) for each primary care outpatients’ absolute precision, and 20% non-response rate. Hence, payment. The course of drugs prescribed ranged between the total sample size with consideration of study design 5  days–1  week for acute illness and 1–3  months’ supply effect was n  =  324. Implementing two-stages random for chronic condition. Besides drugs; disease screening, sampling, ten public primary cares with FMS service investigations such as blood test and radiological imag- were selected through computer generated random sam- ing are highly subsidised and mostly done for free. In pling. Subsequently, 33 patients from each health clinic Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 3 of 8 were designated via systematic random sampling. A Table 1 Frequency distribution of  patients’ socio-demo- graphic total of 324 adults (aged 18 years and above) and of local residents were chosen as our respondents. Clinically or Variables Frequency % mentally unstable patients or with the case mix of sever- (n = 324) ity level III (from either acute or chronic conditions) that Age (years) were deemed too severe for the bidding interviews were Younger (18–47) 169 52.2 excluded from the study. Older (48 and above) 155 47.8 Gender Study tools Male 146 45.1 The combination of modified questionnaire and vali - Female 178 54.9 dated self-developed questionnaire were used. The Ethnicity questionnaire consisted of four parts which are socio- Malay 192 59.3 demographic data, patient treatment expenses, patients’ Chinese 30 9.3 WTP and the maximum monetary amount that each Indian 95 29.3 patient is willing to pay for drugs procurement. Others 7 2.2 Marital status Data collection Married 256 79.0 Primary care clinics are perceived as to be more close- Single 52 16.1 knitted with the local community, as compared to Widow 16 4.9 patients from hospitals that may receive referral from Level of education other states. Pretested and structured questionnaires Lower level of education 246 75.9 through face-to-face interviews with 324 respondents Higher level of education 78 24.1 were successfully carried out. Personal income Lower income (RM0-950) 163 50.3 Data analysis Higher income (RM951 and above) 161 49.7 Analysis was generated using SPSS version 20 software, Household income and the p value of less than 0.05 was accepted as statisti- Lower income (RM0-1900) 165 50.9 cally significant. The relationships, proportion and con - Higher income (RM1901 and above) 159 49.1 trol of confounders, respectively were conducted using Disease status the Mann–Whitney U test, Pearson Chi square test and Chronic conditions 162 50.0 patients’ WTP as the categorical dependent variable. The Acute illness 162 50.0 multivariate model was adjusted for demographic char- Insurance status acteristics. Model fitting was assessed by the change in Yes 103 31.8 2 log-likelihood value and the significance of random No 221 68.2 parameter variance estimate was assessed using the Wald NHI implementation test. The values will be presented in Malaysian Ringgit Yes 181 55.9 (MYR) currency which stands at an estimated conversion No 130 40.1 of MYR 3.8 to USD 1 (as of July 2015). Not sure 13 4.0 Results Socio‑demographic and characteristics of health condition of study participants payment of MYR110 per month. Only 28 respond- The total respondents of 324 showed the distribution ents (8.6%) had ever accessed private hospital treat- of the patients’ age from the younger age group. A total ment. All respondents visit the subsidised public health of n  =  169 (52.2%) respondents were from the younger care extensively, while only 54.9% of patients had ever age (18–47  years), while the rest were from older age been admitted to public hospitals. 50% of patients also group (≥48  years) (Table  1). The majority of respond - regularly visit private health clinics for their healthcare ents at n = 116 respondents (35.8%) were working with services. the private sector. Half of the respondents were from the Almost half of respondents (n  =  162, 52.2%) had con- lower personal income level (MYR 0–950/month) and tracted some form of chronic diseases and the other half low household income (MYR 0–1900/month) group. came to the clinics for acute illnesses. Among patients Only 103 (31.8%) of respondents had any form of private with chronic conditions, 123 patients suffered hyperten - voluntary health insurance with an average premium sion (75.9%), diabetes n  =  79 (48.8%) and IHD n  =  15 Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 4 of 8 (9.3%) respectively. A patient may have multiple co- Factors associated with patients’ WTP for their drugs morbidities simultaneously. A total of 43 (26.5%) had Table 2 shows the bivariate analysis elucidating the asso- two co-morbidities and 6 (3.7%) respondents had three ciation between socio-demographic factors and patients’ chronic conditions. Among those with acute illness, 61 WTP for drugs. Amongst all socio-demographic fac- (37.7%) respondents had upper respiratory tract infec- tors, personal income and household income were tions (URTI), 111 (68.5%) had some form of viral fever, found to have significant relationship with patients’ and 15 (9.3%) had experienced acute gastroenteritis WTP (p = 0.028 and p = 0.022 respectively). One of the (AGE) within the last 6 months. Out of 324 respondents, inclusion criteria for respondents, was patients must be the majority of them (72.2%) had never practiced Tradi- 18  years old and above. Age was initially stratified into tional and Complementary Medicine (TCM) and the rest 4 categories. They were less than 30  years old (i.e. 18 admitted practicing some form of TCM. to  ≤30  years old), 30–40  years old, 40–50  years old and more than 50 years old. Then we tested for normality and Patients’ willingness to pay for their drugs calculated the median age of patients. The relationship A total of 234 out of 324 respondents (72.2%) were not between the above 4 age categories and WTP (2 catego- willing to pay for any drug charges and 34 (10.5%) of ries of Yes and No) was then analysed. The Chi square them strongly disagreed on any sort of drugs payment. analysis was not significant. Then we tested between Only 46 (14.2%) respondents were willing to pay for chronic condition or acute illness drugs, and another 6 (1.9%) were uncertain. Table 2 The association between  socio-demographic fac- Respondents’ health condition was not significantly tors and willingness to pay for drugs (n = 324) associated with their WTP for drugs (p  =  0.356). Only Variables WTP for drugs χ p value 28 (17.3%) out of 162 respondents with chronic condition Yes (%) No (%) were found to be willing to pay for their drugs, while for respondents with acute illness, only 22 (13.6%) were will- Age (years) ing to pay. Among respondents with hypertension, 19.5% 18–47 21 (12.4) 148 (87.6) 2.446 0.118 were willing to pay for their drugs followed by respond- 48 and above 29 (18.7) 126 (81.3) ents with diabetes (13.9%) and with IHD (6.7%). Gender Among respondents who have two co-morbidities, 14% Male 25 (17.1) 121 (82.9) 0.582 0.445 were willing to pay for their drugs, while 16.7% with three Female 25 (14.0) 153 (86.0) co-morbidities were willing to pay for their drugs. How- Ethnicity ever, co-morbidities were also not significantly associated Malay 29 (15.1) 163 (84.9) 0.039 0.844 with patients’ WTP. Non-Malay 21 (15.9) 111 (84.1) The overall level of WTP for drugs are presented using Marital status the median value and its’ interquartile range (IQR). Married 43 (16.8) 213 (83.2) 1.741 0.187 Patients were willing to pay for chronic diseases treat- Single 7 (10.3) 61 (89.7) ment at MYR10 (IQR 1, 30) and slightly higher range was Level of education seen at MYR10 (IQR 5, 30) for acute illnesses. However, Lower level of education 36 (14.6) 210 (85.4) 0.499 0.480 the Mann–Whitney-U Test showed that there was no Higher level of education 14 (17.9) 64 (82.1) significant difference between patients’ WTP for either Personal income chronic or acute illness (p = 0.588). Lower level (RM0-950) 18 (11.0) 145 (89.0) 4.842 0.028* Among diabetic patient, the median value (MYR) and Higher level (RM951 and above) 32 (19.9) 129 (80.1) IQR to pay for diabetic drugs was at MYR 10 (1, 25); Household income hypertension at MYR10 (1, 30) and IHD at MYR1 (0, 40). Lower level (RM0-1900) 18 (10.9) 147 (89.1) 5.271 0.022* Among patients with acute illnesses i.e. URTI, viral fever Higher level (RM1901 and 32 (20.1) 127 (79.9) and AGE, levels patients were willing to pay for drugs at above) the median value of MYR10 (IQR 5, 25), MYR10 (IQR 5, Number of children 30) and MYR10 (IQR 1, 45) respectively. Less than 3 37 (19.1) 157 (80.9) 4.909 0.027* A total of 181 respondents (55.9%) agreed to the sug- 4 and above 13 (10.0) 117 (90.0) 1.051 gestion that the government is right to implement a Status of prior insurance national health insurance scheme in the near future. 98 Have insurance 19 (18.4) 84 (81.6) 0.305 respondents of the 181 respondents (54.1%) chosen to No insurance 31 (14.0) 190 (86.0) pay the premium at only 1% from their monthly basic WTP willingness to pay salary. * p < 0.05; χ Pearson Chi Square Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 5 of 8 actual value of age and amount WTP using Spearman chronic condition, and hypertensive disease predomi- correlation. This was also not significant. The median nates. As in many upper developing countries, unhealthy value was 47  years old. Lastly using the Chi square diet, sedentary lifestyle and chronic diseases such as analysis, age is classified into two categories which were hypertension are a one of the major issue in this coun- ‘younger and older’ age. The group 18–47 was classified try as well. Among the acute illnesses, the majority of as “younger” and 48 and above is classified as “older” age patients contracted acute fever within the last 6 months. group. The two categories were then tested against WTP. As high as 72.2% of respondents were not willing to pay This was also not significant. The final test through multi - for drugs. However, 55.9% of respondents supported the variate analysis also did not show any significant findings. future implementation of the proposed future National Perhaps even though the age here varies but the pur- Health Insurance to be implemented by the government chasing power of the patients are almost similar. Hence [5]. According to data published [15] by the Depart- did not show any relationship between age and WTP, ment of Statistics Malaysia in 2010; a household average unlike income of patients. Higher level income group was monthly expenditure, consisting of an average of four more willing to pay for their drugs compared to the low members per household was approximately MYR 2200. income group, at 19.9% compared to 11.0%. Respond- This government-funded survey revealed that the aver - ents with higher income level were more willing to pay age Malaysians spend 23% of their household income on for drugs (20.1%), compared to the lower level income housing and utilities, 20% for foods and beverages and group (10.9%). Among respondents with chronic condi- only 1% is spent for health [15]. This shows that on aver - tions, household income was significantly associated with age, a household is willing to spend approximately MYR patients’ WTP. Among the respondents suffering from 22 per month on health. This amount reflected the ten - hypertension, 32.5% of the higher income group was dency of population to be heavily dependent on the long willing to pay, compared to respondents with diabetes standing universal health coverage provided and mini- (23.8%) and IHD (30.0%). This relationship was signifi - mal cross subsidisation. The mean income per capita for cant at p = 0.012. Malaysian was at MYR 5000 per month in the year 2012. The number of dependents’ children have a significant Inequality exists with urbanites progressing further than association with patients’ WTP (p = 0.027). Respondents ruralises. The urban household monthly income had having between 0 and 3 children were more willing to increased at a rate of 6.6% per year. This was from MYR pay for drugs (19.1%) compared to those who had 4 and 4705 monthly in 2009, to MYR 5742 last 2015 while the above dependents (10.0%). However, stratification analy - rural household monthly income increased 6.4% annu- sis based on chronic conditions showed that there was no ally from MYR 2545 to MYR 3080. This leads to the rural significant association between numbers of dependent population at higher risk due to higher financial com - children with patients’ WTP. Other factors such as age, mitment of household dependents, children’s education, gender, ethnicity, marital status, having prior insurance housing rent, transport (cars, motorbike) instalment and status and level of education were found to have no sig- other responsibilities [1, 5]. From Table 2, females, Malay nificant association with patients’ WTP for drugs. and singles, were less willing to pay for drugs. However Further analysis via multivariate logistic regression these associations were not significant. (Table  3), the Wald’s estimates gave the upmost impor- Almost all clinics attendees pay a nominal fee of MYR tance to number of dependents towards patients WTP 1 (or USD 0.38), including for all investigations and (p = 0.009). Respondents with lower number of depend- drugs prescribed per visit [1–3]. Respondents also rec- ents (0 to 3 children) were nearly three times more likely ommended that the government should continuously to pay for drugs, compared to those with higher depend- subsidies the health care services especially for lower ents (≥4 children) at adjusted OR  =  2.63 (95% CI 1.27, income group earners. Patients with chronic conditions 5.44). have to undergo a considerable amount of suffering com - pared to those with acute or common illnesses [7–10]. Discussion In addition, it is obligatory for them to go to the clinic The patients’ profile showed that the majority of patients more often for follow up and continuation of treatment who attended the facilities were from the younger age [1, 2]. If drugs charges are to be implemented abruptly group 18–47  years, females, of Malay ethnicity, com- in health clinics without proper targeting for fee subsidi- pleted lower education, lower income group and suffered sation, this situation may decrease visits to hospitals or from chronic condition. This fits with the scenario of clinics in the future [4, 5, 7, 10]. However, our statistical urbanised population in Malaysia that attends the subsi- test proved that health status has no significant associa - dised free primary care. Almost 30% of the 162 patients tion with patients’ WTP for drugs. Regardless of a pri- with chronic condition attendees had more than one mary care patient’s health condition of having chronic or Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 6 of 8 Table 3 Logistic regression of influencing factors towards willingness to pay Variable β S.E Wald p value Exp (β) 95% CI for exp (β) Lower Upper Age Older 0.428 0.441 0.941 0.332 1.534 0.646 3.642 Younger 1.000 Gender Male 0.037 0.342 0.012 0.913 1.038 0.531 2.031 Female 1.000 Ethnicity Non-Malay 0.047 0.337 0.019 0.889 1.048 0.542 2.027 Malay 1.000 Marital status Married 0.576 0.483 1.419 0.234 1.778 0.690 4.585 Single 1.000 Level of education Higher level 0.128 0.425 0.091 0.763 1.137 0.494 2.616 Lower level 1.000 Personal income Higher level 0.449 0.477 0.887 0.346 1.567 0.615 3.988 Lower level 1.000 Household income Higher level 0.510 0.403 1.600 0.206 1.666 0.755 3.673 Lower level 1.000 Type of occupation Employed 0.092 0.454 0.041 0.840 1.096 0.450 2.671 Unemployed 1.000 Number of dependents 0 to 3 0.965 0.372 6.732 0.009** 2.625 1.266 5.442 4 and above 1.000 Type of health condition Chronic 0.033 0.406 0.007 0.935 1.034 0.467 2.290 Acute 1.000 TCM practice Practice TCM 0.045 0.375 0.014 0.905 1.046 0.502 2.179 Not practice TCM 1.000 Health insurance No health insurance −0.127 0.382 0.110 0.740 0.881 0.416 1.864 With health insurance 1.000 β standardized coefficient, S.E standard error, Exp (β) odds ratio ** p < 0.05 acute illness, this did not influence the WTP for drugs [8, have the highest percentage of WTP as compared to 16]. Most patients were not enthusiastic to pay for their diabetes and IHD. This is in tandem with Bradford et al. drugs and proposed that the government drug subsidy finding, which state that hypertensive patients are more to be sustained [1, 5, 14]. The economic effect of chronic responsive to pay compared to other patients’ with other disease has adversely and disproportionately affected chronic conditions [16, 17]. poor and vulnerable populations in the developing world Our findings illustrated that personal and household [13, 16] including Malaysia. The results indicate that only incomes were significantly associated with WTP for 17.3% of respondents with chronic condition are willing drugs, for both chronic and acute illness. Those who to pay for their drugs. Respondents with hypertension earned higher income were more inclined to pay more for Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 7 of 8 drugs. This result supports our hypothesis and was con - tablet per month which costs about MYR 15 to MYR sistent with the previous studies, which indicated that the 30 [22, 23]. This rough estimation indicates that most higher income level with more disposable income is asso- Malaysians are not prepared to endure [3, 5] the entire ciated with WTP [18, 19, 21]. This relevancy of income cost of their chronic treatment. and WTP, indicate that if drug charge is to be imple- Patients who were not practicing TCM were more will- mented in public dispensing, this may potentially become ing to pay for drugs as compared with those currently a burden to the countries of lower income groups. Inter- practicing TCM. However, the association of WTP and estingly even though patients with higher income were TCM practice were not significant. Previous studies more enthusiastic to pay more for drugs, the percentage revealed that patients who contracted chronic condition was very low. The current price inflation and exuberant are more likely to utilize and purchase complementary cost of living have led to the drastic increase in the cost and alternative medicine [24–26]. They may perceive less of drugs and healthcare as a whole [18, 20, 21], albeit a need on modern drugs [26], or need to purchase them; higher personal or household income [10, 23]. hence the less willing to pay. Respondents with less number of dependents were more willing to pay for drugs (19.1%). This finding is con - Limitation of study sistent with previous study by Okyere which found that Since this study did not represent the private sector pri- the chances of an individual to pay for a ¢5000 health mary care patients, the result would be very centered insurance premium, would be reduced by 10.8%, if there towards public primary care patients. Thus, it is sug - is an increase of 1% in the dependency ratio in the fam- gested that future study should include private primary ily [18, 19]. This elucidated that an increase in number care. As this state is also one of the richest urbanized of dependents was associated with reduced WTP. The states in Malaysia, population of WTP would be expected concept of opportunity cost is seen here [1, 4]; if patients to greatly contradict from other less developed states are required to pay more for expensive chronic disease with a larger number of low income groups. drugs, this would affect their other daily expenses and basic needs [23]. Hence, this substantiates our hypothesis Recommendation that patients with a high number of dependents have low The majority of the respondents (72.2%) were not willing WTP for drugs as compared with those with low number to pay for their drugs. Utilisations on other mechanisms of dependents. Therefore, the increase in the number of need to be addressed tin inspiring its citizens to share the dependents will contribute to less willingness to pay. burden of drug cost. It is proposed that public facilities start The median value of WTP for drugs among patients to implement a staggered payment mechanism whereby who have two co-morbidities was MYR 5 per visit. step-by-step fee charges are implemented within the next Whereas for patients who have three co-morbidities were few years period in order to heighten citizens’ awareness in willing to pay up to a higher level at MYR 10.50 but this assisting the government to control health care costs. Phas- was not statistically significant. Valderas [20] stated as ing these implementations would be much palatable to edu- the number of chronic disease increases, the costs of the cate the public and finally remove the norm of citizens who treatment will also escalate [3, 22]. However from Chang currently enjoy the highly subsidized drugs and health care. states those co-morbidities had no significant increase Prior implementing the new health care financing towards patients’ WTP [21]. WTP has its inert weakness policy such as the proposed National Health Insurance [1, 6] i.e. the population with more disposable income and its benefit packages, it is plausible to start charging would generally be more willing to pay for drugs [9, 12] MYR 10 as the minimum drug charges for each outpa- compared with population who have lower purchas- tient visit that entails subsidised treatment for acute and ing power and income [5]. Bivariate analysis statistically chronic diseases. In countries that already implement showed that patients who earned less than MYR950 per the casemix costing charges, this can be used to start month (considered to be lower income) are less likely to charging patients according to the cost calculated in pay (89%) for drugs compared to those who earned more the casemix system. Government facilities should start than MYR950 per month (80.1%). charging the more affluent population on health services The median value patients are willing to pay for drugs and drugs consumed. The minimum value of MYR10 is was only MYR 10. Taking the highly prevalent diabetes considering the overall median value of patients’ WTP as an example [15, 16, 22] the first line medication pre - for drugs (MYR 10 per visit). However proper targeting is scribed to patient after adequate practice on healthy diet needs for the lower income group. This is whilst exclud - and regular exercise is bi-guanides (Metformin) [22]. ing the vulnerable groups such as the low income earn- For a patient who consumes the maximum dose of Met- ers, pensioners or the elderly. This can be carried out formin (1000 mg two times daily), he or she requires 120 through the implementation of means tested financing. Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 8 of 8 References Conclusions 1. Babar ZD, Izham MIM. Eec ff t of privatization of the drug distribution The results indicate that patients have low WTP for drugs system on drug prices in Malaysia. Public Health. 2009;123:523–33. for both chronic and acute illness even in an urban state of 2. Annual report: Malaysian Ministry of Health. Kuala Lumpur: Pharmaceuti- cal Services Division; 2005. Malaysia. The maximum amount that public patients are 3. Annual report: Malaysia Ministry of Health drug expenditure. Kuala Lum- willing to pay for acute or chronic diseases drugs are consid- pur: Pharmaceutical Services Division; 2010. ered low, the median value was only MYR 10 (USD 3.8) per 4. Hussain HS. Drug control and formulary management in Malaysia. Phar- macoecon Outcomes Res Asia. 2008;11:S158–9. visit. Only three factors were found to be associated with 5. Bridel A. Explaining 1Care: its proposals, rationale and feasibility. Philip- WTP; which are the level personal and household income, pines: Center for Public Policy Studies (CPPS); 2012. and number of dependents. Most of our citizens are in a 6. Quevedo JFM, Hernández IC, Espinosa JG, Escudero GS. The willingness- to-pay concept in question. Rev Saúde Pública. 2009;43:2. comfort zone of receiving highly subsidized health services 7. Adeneye AK, Mafe MA, Appelt B, Idowu ET, Akande DO. Willingness to thus, and may not be prepared to transform from the norm pay for praziquantel treatment in a hyper-endemic community of Ogun of having free services and to pay for his/her own drugs. State, Nigeria. Res Soc Adm Pharm. 2006;2:83–95. 8. Greenberg D, Bakhai A, Neumann PJ, Cohen DJ. Willingness to pay for Authors’ contributions avoiding coronary restenosis and repeat revascularization: results from a SEWP contributed to the design of the study, choosing the co-authors, revi- contingent valuation study. Health Policy. 2004;70:207–16. sions of the manuscripts, formatting, and revision of references used in publica- 9. Michael AM, Bohara AK, Pagán JA. Assessing willingness to pay for cancer tion. SNAA prepared the study logistics, carried out the study and interviewing prevention. Int J Health Care Finance Econ. 2010;10:301–14. the patients. ANA contributed comments on the WTP tool and critiques on the 10. Hwang W, Weller W, Ireys H, Anderson G. Out-of-pocket medical spending WTP concept. RZ and RI helped in the arrangement of local facilities and final for care of chronic conditions. Health Aff (Millwood). 2001;20(6):267–78. approval of draft. All authors read and approved the final manuscript. 11. Jones R, White P, Armstrong D, Ashworth M, Peters M. Managing acute illness. The King’s Fund. 2010. Author details 12. Chen Y T, Ying YH, Chang K, Hsieh YH. Study of patients’ willingness to Faculty of Medicine, Department of Community Health, University Kebang- pay for a cure of chronic obstructive pulmonary disease in Taiwan. Int J saan Malaysia Medical Center, Jalan Yaacob Latif, Bandar Tun Razak, Cheras, Environ Res Public Health. 2016;13(3):273–87. 56000 Kuala Lumpur, Malaysia. Health Administration, Faculty of Business 13. Adams R, Chou YJ, Pu C. Willingness to participate and pay for a proposed Management, Universiti Teknologi MARA, Puncak Alam, Selangor, Malaysia. national health insurance in St. Vincent and the grenadines: a cross-sectional Institute for Health Systems Research, Ministry of Health, Kuala Lumpur, contingent valuation approach. BMC Health Serv Res. 2015;15:148–58. Malaysia. Health District Office, Sepang, Malaysia. 14. Gustafsson-Wright E, Asfaw A, Van der Gaag J. Willingness to pay for health insurance: an analysis of the potential market for new low-cost Acknowledgements health insurance products in Namibia. Soc Sci Med. 2009;69:1351–9. The authors would like to thank the Dean and Hospital Director UKM Medical 15. Population distribution and basic demographic characteristic report Center and the Director of Ministry of Health, Malaysia for the support and 2010. Department of Statistics Malaysia. 2010. http://www.statistics.gov. permission to publish this paper. We would also like to thanks the Director my. Accessed 11 Jan 2013. of Selangor State Health Department and all Health District Officers for their 16. Engelgau M, Rosenhouse S, El-Saharty S, Mahal A. The economic effect fullest cooperation and commitment in this research. Last but not least, we of non-communicable diseases on households and nations: a review of would like to convey our utmost appreciation to the respondents who had existing evidence. J Commun Healthc. 2011;16(Suppl. 2):75–81. been willing to participate in our study. 17. Bradford WD, Kleit A, Krousel-Wood MA, Re RM. Comparing willingness to pay for telemedicine across a chronic heart failure and hypertension Competing interests population. Telemed J E Health. 2005;11(4):430–8. The authors declare that they have no competing interests. 18. Pinto SL, Pharm BS, Goodman MH, Black CD, Lesch D. Identifying factors that affect patients’ willingness to pay for inhaled insulin. Adm Pharm. Availability of data and supporting materials 2009;5:253–61. All data generated or analysed during this study are included in this published 19. Okyere WKA, Akoto IO, Anum A, Appiah EN. Willingness to pay for health article. insurance in a developing economy: a pilot study of the informal sector of Ghana using contingent valuation. Health Policy. 1997;42:223–37. Consent for publication 20. Valderas JM, Starfield B, Sibbald B, Salisbury C, Roland M. Defining comor - All authors approved the manuscript and submission for publication. The bidity: implications for understanding health and health services. Ann permission to publish was also obtained from the Ministry of Health Malaysia Fam Med. 2009;7(4):357–63. and UKM Medical Center. 21. Chang K. Comorbidities, quality of life and patients’ willingness to pay for a cure.for type 2 diabetes in Taiwan. Public Health. 2010;124:284–94. Ethics approval and consent to participate 22. Clinical practice guidelines. Management of type 2 diabetes mellitus. 4th Informed written consent of all respondents had been undertaken. Ethical ed. Kuala Lumpur: Persatuan Diabetes Malaysia, Academy of Medicine clearance was obtained from the Research and Ethics Committee, Medical Malaysia, Ministry of Health Malaysia; 2009. Faculty, Universiti Kebangsaan Malaysia (FF-163-2012) as well as National 23. Consumer price guide. Pharmaceutical Services Division, Ministry of Medical Research Registry, Ministry of Health (NMRR-12-393-12001). Health Malaysia. 2014. http://www.pharmacy.gov.my/v2/ms/apps/drug- price. Accessed 11 Jan 2013. Funding 24. Millar WJ. Patterns of use–alternative health care practitioners. Health We gratefully acknowledge funding support from UKM Medical Center Funda- Rep. 2001;13(1):9–21. mental Research Fund (FF-162-2012) for this study. 25. Quan H, Lai D, Johnson D, Verhoef M, Musto R. Complementary and alternative medicine use among Chinese and white Canadians. Can Fam Physician. 2008;54(11):1563–9. Publisher’s Note 26. Metcalfe A, Williams J, McChesney J, Patten SB, Jetté N. Use of comple- Springer Nature remains neutral with regard to jurisdictional claims in pub- mentary and alternative medicine by those with a chronic disease and lished maps and institutional affiliations. the general population—results of a national population based survey. BMC Complement Altern Med. 2010;10:58. Received: 12 April 2016 Accepted: 22 March 2017 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Asia Pacific Family Medicine Springer Journals

Patients’ willingness to pay for their drugs in primary care clinics in an urbanized setting in Malaysia: a guide on drug charges implementation

Loading next page...
 
/lp/springer-journals/patients-willingness-to-pay-for-their-drugs-in-primary-care-clinics-in-Y4YeIpQRxl
Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Medicine & Public Health; General Practice / Family Medicine; Primary Care Medicine
eISSN
1447-056X
DOI
10.1186/s12930-017-0035-5
Publisher site
See Article on Publisher Site

Abstract

Background: Malaysia is an upper middle income country that provides subsidized healthcare to ensure universal coverage to its citizens. The challenge of escalating health care cost occurs in most countries, including Malaysia due to increase in disease prevalence, which induced an escalation in drug expenditure. In 2009, the Ministry of Health has allocated up to Malaysian Ringgit (MYR) 1.402 billion (approximately USD 390 million) on subsidised drugs. This study was conducted to measure patients’ willingness to pay ( WTP) for treatment of chronic condition or acute illnesses, in an urbanized population. Methods: A cross-sectional study, through face-to-face interview was conducted in an urban state in 2012–2013. Systematic random sampling of 324 patients was selected from a list of patients attending ten public primary cares with Family Medicine Specialist service. Patients were asked using a bidding technique of maximum amount (in MYR) if they are WTP for chronic or acute illnesses. Results: Patients are mostly young, female, of lower education and lower income. A total of 234 respondents (72.2%) were not willing to pay for drug charges. WTP for drugs either for chronic or acute illness were at low at median of MYR10 per visit (USD 3.8). Bivariate analysis showed that lower numbers of dependent children (≤3), higher personal and household income are associated with WTP. Multivariate analysis showed only number of dependent children (≤3) as significant ( p = 0.009; 95% CI 1.27–5.44) predictor to drugs’ WTP. Conclusion: The result indicates that primary care patients have low WTP for drugs, either for chronic condition or acute illness. Citizens are comfortable in the comfort zone whereby health services are highly subsidized through universal coverage. Hence, there is a resistance to pay for drugs. Keywords: Willingness-to-pay, drugs expenditure, Acute diseases, Chronic conditions, Urban state Health delivers comprehensive medical, health, dental Background and pharmaceutical services at a subsidized rate lead- The increase in drug expenditure is a crucial challenge in ing to increase burden on the government. Since year Malaysia. As there is no established compulsory National 2000, the national drug expenditure has increased from Health Insurance yet, most of the patients obtain treat- MYR 346 million (USD 91 million) to MYR 915 million ment including prescription of drugs from government (USD 241 million) in 2005. From the year 2004–2005, funded health clinics and hospitals [1]. The Ministry of an increase of 13.3% was recorded [2] and according to the Pharmaceutical Services Division Annual Report, the *Correspondence: sh_ezat@ppukm.ukm.edu.my; sh_ezat@yahoo.com Malaysia Ministry of Health drug expenditure had esca- Faculty of Medicine, Department of Community Health, University Kebangsaan Malaysia Medical Center, Jalan Yaacob Latif, Bandar Tun lated to MYR 1.402 billion in 2009 [3]. Razak, Cheras, 56000 Kuala Lumpur, Malaysia Full list of author information is available at the end of the article © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/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://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 2 of 8 Numerous factors [2, 3] are found to contribute the future, a national health insurance system has been to an increase drug prices including growth of aging proposed to be implemented in the hope of reducing the population, higher consumer expectations, long-term cost burden of the government providers. drug treatment for chronic conditions, polypharmacy, Many primary care clinics in Selangor are equipped improvements in diagnostics/treatment of diseases, and with Family Medicine Specialist (FMS) and case-mix ser- novel expensive drugs offered due to advancement of vices (using the ICD-10 ambulatory coding and group- health technology [4]. As a result, the government has an ing). These facilities and logistics are placed due to the increasing demand on the rising health care costs [1, 3]. higher number of population seeking services, higher Provision for chronic conditions and acute diseases drugs patients’ expectancy, an increasing aging population and take place in many primary care facilities throughout the the increasing severe and complex case mix cases. state and yet, they are free of charge. A national health insurance has been proposed to be strategically imple- Methods mented in the government funded primary care centers Study population in Malaysia in the very near future [5]. Therefore, this This study is a cross-sectional study conducted in year study intends to measure patients’ WTP for their drugs 2012–2013 Selangor, Malaysia. The population represents who attend to public primary care services. both high and low income quintile level of population coming from both urban and rural areas. In each district, Willingness to pay for drugs there are two to six public primary clinics, equipped with Willingness to pay is a methodological tool to discover clinician with specialty trainings, for example the Family the hypothetical monetary value for programs and spe- Medicine Specialists (FMS). A total of 10 public primary cific medical interventions and treatments [6]. A study care with existing FMS services were chosen as those that examined the WTP for Praziquantel treatment (for clinics are visited by patients who suffer from various Schistosoma parasites) in Ogun State, Nigeria, showed types of diseases; managed with various types of medica- 92.3% respondents were willing to pay for the drug to tions including the “list A” drugs (i.e. for example drugs treat the infected household members [7]. However, in that can only be prescribed by specialists). a coronary restenosis (re-narrowing) study that assessed patients’ WTP to avoid revascularization procedure, Definition of “chronic condition” and “acute illness” found that the proportion of patients’ WTP is higher with This study aims to determine patients’ WTP for drugs, greater absolute risk reductions [8]. This is logical in the both for chronic conditions or acute illnesses. Chronic sense that higher risk would entail higher commitment condition is defined as conditions that had lasted or were and WTP. Similarly, in assessing WTP for cancer preven- expected to last 12 or more months and result in func- tion, the results revealed that income and the probability tional limitations and/or the need for on-going medi- of developing cancer were positively correlated to WTP cal care [10]. The WTP among patients were examined [9]. with three most prevalent chronic conditions in Malaysia Malaysia state healthcare system has been heavily which were hypertension, type-2 diabetes mellitus and financed through the government tax based financing, ischemic heart disease (IHD). Acute illness is defined although its health care prosper under bipartite public as rapid onset and/or short course disease less than and private providers. Under the universal health cover- 3  months duration, which is self-limiting or requiring age, Malaysia’s public out patients healthcare is provided minimal treatment [11]. almost fee exempted and nominal payment upon admis- sion as inpatient. Employer-based financing covers a lim - Sample size and sampling method ited number of companies and private insurance, that The sample size was calculated by considering the engage private health providers, mostly are concentrated assumption of two population proportion formula: the in urban areas. Primary care facilities in Malaysia are proportion (P1)  =  1.2% of low income that are willing highly subsidized by the government under the universal to pay, while proportion (P2) = 11% of high income that coverage, leading to very low state cost recovery. Malay- willing to pay [12]. The Z (standard normal) distribution sian citizens have to pay a user fee of only MYR1 (esti- value at 95% confidence level was taken at 1.96, 5% of mated to be USD 0.38) for each primary care outpatients’ absolute precision, and 20% non-response rate. Hence, payment. The course of drugs prescribed ranged between the total sample size with consideration of study design 5  days–1  week for acute illness and 1–3  months’ supply effect was n  =  324. Implementing two-stages random for chronic condition. Besides drugs; disease screening, sampling, ten public primary cares with FMS service investigations such as blood test and radiological imag- were selected through computer generated random sam- ing are highly subsidised and mostly done for free. In pling. Subsequently, 33 patients from each health clinic Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 3 of 8 were designated via systematic random sampling. A Table 1 Frequency distribution of  patients’ socio-demo- graphic total of 324 adults (aged 18 years and above) and of local residents were chosen as our respondents. Clinically or Variables Frequency % mentally unstable patients or with the case mix of sever- (n = 324) ity level III (from either acute or chronic conditions) that Age (years) were deemed too severe for the bidding interviews were Younger (18–47) 169 52.2 excluded from the study. Older (48 and above) 155 47.8 Gender Study tools Male 146 45.1 The combination of modified questionnaire and vali - Female 178 54.9 dated self-developed questionnaire were used. The Ethnicity questionnaire consisted of four parts which are socio- Malay 192 59.3 demographic data, patient treatment expenses, patients’ Chinese 30 9.3 WTP and the maximum monetary amount that each Indian 95 29.3 patient is willing to pay for drugs procurement. Others 7 2.2 Marital status Data collection Married 256 79.0 Primary care clinics are perceived as to be more close- Single 52 16.1 knitted with the local community, as compared to Widow 16 4.9 patients from hospitals that may receive referral from Level of education other states. Pretested and structured questionnaires Lower level of education 246 75.9 through face-to-face interviews with 324 respondents Higher level of education 78 24.1 were successfully carried out. Personal income Lower income (RM0-950) 163 50.3 Data analysis Higher income (RM951 and above) 161 49.7 Analysis was generated using SPSS version 20 software, Household income and the p value of less than 0.05 was accepted as statisti- Lower income (RM0-1900) 165 50.9 cally significant. The relationships, proportion and con - Higher income (RM1901 and above) 159 49.1 trol of confounders, respectively were conducted using Disease status the Mann–Whitney U test, Pearson Chi square test and Chronic conditions 162 50.0 patients’ WTP as the categorical dependent variable. The Acute illness 162 50.0 multivariate model was adjusted for demographic char- Insurance status acteristics. Model fitting was assessed by the change in Yes 103 31.8 2 log-likelihood value and the significance of random No 221 68.2 parameter variance estimate was assessed using the Wald NHI implementation test. The values will be presented in Malaysian Ringgit Yes 181 55.9 (MYR) currency which stands at an estimated conversion No 130 40.1 of MYR 3.8 to USD 1 (as of July 2015). Not sure 13 4.0 Results Socio‑demographic and characteristics of health condition of study participants payment of MYR110 per month. Only 28 respond- The total respondents of 324 showed the distribution ents (8.6%) had ever accessed private hospital treat- of the patients’ age from the younger age group. A total ment. All respondents visit the subsidised public health of n  =  169 (52.2%) respondents were from the younger care extensively, while only 54.9% of patients had ever age (18–47  years), while the rest were from older age been admitted to public hospitals. 50% of patients also group (≥48  years) (Table  1). The majority of respond - regularly visit private health clinics for their healthcare ents at n = 116 respondents (35.8%) were working with services. the private sector. Half of the respondents were from the Almost half of respondents (n  =  162, 52.2%) had con- lower personal income level (MYR 0–950/month) and tracted some form of chronic diseases and the other half low household income (MYR 0–1900/month) group. came to the clinics for acute illnesses. Among patients Only 103 (31.8%) of respondents had any form of private with chronic conditions, 123 patients suffered hyperten - voluntary health insurance with an average premium sion (75.9%), diabetes n  =  79 (48.8%) and IHD n  =  15 Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 4 of 8 (9.3%) respectively. A patient may have multiple co- Factors associated with patients’ WTP for their drugs morbidities simultaneously. A total of 43 (26.5%) had Table 2 shows the bivariate analysis elucidating the asso- two co-morbidities and 6 (3.7%) respondents had three ciation between socio-demographic factors and patients’ chronic conditions. Among those with acute illness, 61 WTP for drugs. Amongst all socio-demographic fac- (37.7%) respondents had upper respiratory tract infec- tors, personal income and household income were tions (URTI), 111 (68.5%) had some form of viral fever, found to have significant relationship with patients’ and 15 (9.3%) had experienced acute gastroenteritis WTP (p = 0.028 and p = 0.022 respectively). One of the (AGE) within the last 6 months. Out of 324 respondents, inclusion criteria for respondents, was patients must be the majority of them (72.2%) had never practiced Tradi- 18  years old and above. Age was initially stratified into tional and Complementary Medicine (TCM) and the rest 4 categories. They were less than 30  years old (i.e. 18 admitted practicing some form of TCM. to  ≤30  years old), 30–40  years old, 40–50  years old and more than 50 years old. Then we tested for normality and Patients’ willingness to pay for their drugs calculated the median age of patients. The relationship A total of 234 out of 324 respondents (72.2%) were not between the above 4 age categories and WTP (2 catego- willing to pay for any drug charges and 34 (10.5%) of ries of Yes and No) was then analysed. The Chi square them strongly disagreed on any sort of drugs payment. analysis was not significant. Then we tested between Only 46 (14.2%) respondents were willing to pay for chronic condition or acute illness drugs, and another 6 (1.9%) were uncertain. Table 2 The association between  socio-demographic fac- Respondents’ health condition was not significantly tors and willingness to pay for drugs (n = 324) associated with their WTP for drugs (p  =  0.356). Only Variables WTP for drugs χ p value 28 (17.3%) out of 162 respondents with chronic condition Yes (%) No (%) were found to be willing to pay for their drugs, while for respondents with acute illness, only 22 (13.6%) were will- Age (years) ing to pay. Among respondents with hypertension, 19.5% 18–47 21 (12.4) 148 (87.6) 2.446 0.118 were willing to pay for their drugs followed by respond- 48 and above 29 (18.7) 126 (81.3) ents with diabetes (13.9%) and with IHD (6.7%). Gender Among respondents who have two co-morbidities, 14% Male 25 (17.1) 121 (82.9) 0.582 0.445 were willing to pay for their drugs, while 16.7% with three Female 25 (14.0) 153 (86.0) co-morbidities were willing to pay for their drugs. How- Ethnicity ever, co-morbidities were also not significantly associated Malay 29 (15.1) 163 (84.9) 0.039 0.844 with patients’ WTP. Non-Malay 21 (15.9) 111 (84.1) The overall level of WTP for drugs are presented using Marital status the median value and its’ interquartile range (IQR). Married 43 (16.8) 213 (83.2) 1.741 0.187 Patients were willing to pay for chronic diseases treat- Single 7 (10.3) 61 (89.7) ment at MYR10 (IQR 1, 30) and slightly higher range was Level of education seen at MYR10 (IQR 5, 30) for acute illnesses. However, Lower level of education 36 (14.6) 210 (85.4) 0.499 0.480 the Mann–Whitney-U Test showed that there was no Higher level of education 14 (17.9) 64 (82.1) significant difference between patients’ WTP for either Personal income chronic or acute illness (p = 0.588). Lower level (RM0-950) 18 (11.0) 145 (89.0) 4.842 0.028* Among diabetic patient, the median value (MYR) and Higher level (RM951 and above) 32 (19.9) 129 (80.1) IQR to pay for diabetic drugs was at MYR 10 (1, 25); Household income hypertension at MYR10 (1, 30) and IHD at MYR1 (0, 40). Lower level (RM0-1900) 18 (10.9) 147 (89.1) 5.271 0.022* Among patients with acute illnesses i.e. URTI, viral fever Higher level (RM1901 and 32 (20.1) 127 (79.9) and AGE, levels patients were willing to pay for drugs at above) the median value of MYR10 (IQR 5, 25), MYR10 (IQR 5, Number of children 30) and MYR10 (IQR 1, 45) respectively. Less than 3 37 (19.1) 157 (80.9) 4.909 0.027* A total of 181 respondents (55.9%) agreed to the sug- 4 and above 13 (10.0) 117 (90.0) 1.051 gestion that the government is right to implement a Status of prior insurance national health insurance scheme in the near future. 98 Have insurance 19 (18.4) 84 (81.6) 0.305 respondents of the 181 respondents (54.1%) chosen to No insurance 31 (14.0) 190 (86.0) pay the premium at only 1% from their monthly basic WTP willingness to pay salary. * p < 0.05; χ Pearson Chi Square Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 5 of 8 actual value of age and amount WTP using Spearman chronic condition, and hypertensive disease predomi- correlation. This was also not significant. The median nates. As in many upper developing countries, unhealthy value was 47  years old. Lastly using the Chi square diet, sedentary lifestyle and chronic diseases such as analysis, age is classified into two categories which were hypertension are a one of the major issue in this coun- ‘younger and older’ age. The group 18–47 was classified try as well. Among the acute illnesses, the majority of as “younger” and 48 and above is classified as “older” age patients contracted acute fever within the last 6 months. group. The two categories were then tested against WTP. As high as 72.2% of respondents were not willing to pay This was also not significant. The final test through multi - for drugs. However, 55.9% of respondents supported the variate analysis also did not show any significant findings. future implementation of the proposed future National Perhaps even though the age here varies but the pur- Health Insurance to be implemented by the government chasing power of the patients are almost similar. Hence [5]. According to data published [15] by the Depart- did not show any relationship between age and WTP, ment of Statistics Malaysia in 2010; a household average unlike income of patients. Higher level income group was monthly expenditure, consisting of an average of four more willing to pay for their drugs compared to the low members per household was approximately MYR 2200. income group, at 19.9% compared to 11.0%. Respond- This government-funded survey revealed that the aver - ents with higher income level were more willing to pay age Malaysians spend 23% of their household income on for drugs (20.1%), compared to the lower level income housing and utilities, 20% for foods and beverages and group (10.9%). Among respondents with chronic condi- only 1% is spent for health [15]. This shows that on aver - tions, household income was significantly associated with age, a household is willing to spend approximately MYR patients’ WTP. Among the respondents suffering from 22 per month on health. This amount reflected the ten - hypertension, 32.5% of the higher income group was dency of population to be heavily dependent on the long willing to pay, compared to respondents with diabetes standing universal health coverage provided and mini- (23.8%) and IHD (30.0%). This relationship was signifi - mal cross subsidisation. The mean income per capita for cant at p = 0.012. Malaysian was at MYR 5000 per month in the year 2012. The number of dependents’ children have a significant Inequality exists with urbanites progressing further than association with patients’ WTP (p = 0.027). Respondents ruralises. The urban household monthly income had having between 0 and 3 children were more willing to increased at a rate of 6.6% per year. This was from MYR pay for drugs (19.1%) compared to those who had 4 and 4705 monthly in 2009, to MYR 5742 last 2015 while the above dependents (10.0%). However, stratification analy - rural household monthly income increased 6.4% annu- sis based on chronic conditions showed that there was no ally from MYR 2545 to MYR 3080. This leads to the rural significant association between numbers of dependent population at higher risk due to higher financial com - children with patients’ WTP. Other factors such as age, mitment of household dependents, children’s education, gender, ethnicity, marital status, having prior insurance housing rent, transport (cars, motorbike) instalment and status and level of education were found to have no sig- other responsibilities [1, 5]. From Table 2, females, Malay nificant association with patients’ WTP for drugs. and singles, were less willing to pay for drugs. However Further analysis via multivariate logistic regression these associations were not significant. (Table  3), the Wald’s estimates gave the upmost impor- Almost all clinics attendees pay a nominal fee of MYR tance to number of dependents towards patients WTP 1 (or USD 0.38), including for all investigations and (p = 0.009). Respondents with lower number of depend- drugs prescribed per visit [1–3]. Respondents also rec- ents (0 to 3 children) were nearly three times more likely ommended that the government should continuously to pay for drugs, compared to those with higher depend- subsidies the health care services especially for lower ents (≥4 children) at adjusted OR  =  2.63 (95% CI 1.27, income group earners. Patients with chronic conditions 5.44). have to undergo a considerable amount of suffering com - pared to those with acute or common illnesses [7–10]. Discussion In addition, it is obligatory for them to go to the clinic The patients’ profile showed that the majority of patients more often for follow up and continuation of treatment who attended the facilities were from the younger age [1, 2]. If drugs charges are to be implemented abruptly group 18–47  years, females, of Malay ethnicity, com- in health clinics without proper targeting for fee subsidi- pleted lower education, lower income group and suffered sation, this situation may decrease visits to hospitals or from chronic condition. This fits with the scenario of clinics in the future [4, 5, 7, 10]. However, our statistical urbanised population in Malaysia that attends the subsi- test proved that health status has no significant associa - dised free primary care. Almost 30% of the 162 patients tion with patients’ WTP for drugs. Regardless of a pri- with chronic condition attendees had more than one mary care patient’s health condition of having chronic or Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 6 of 8 Table 3 Logistic regression of influencing factors towards willingness to pay Variable β S.E Wald p value Exp (β) 95% CI for exp (β) Lower Upper Age Older 0.428 0.441 0.941 0.332 1.534 0.646 3.642 Younger 1.000 Gender Male 0.037 0.342 0.012 0.913 1.038 0.531 2.031 Female 1.000 Ethnicity Non-Malay 0.047 0.337 0.019 0.889 1.048 0.542 2.027 Malay 1.000 Marital status Married 0.576 0.483 1.419 0.234 1.778 0.690 4.585 Single 1.000 Level of education Higher level 0.128 0.425 0.091 0.763 1.137 0.494 2.616 Lower level 1.000 Personal income Higher level 0.449 0.477 0.887 0.346 1.567 0.615 3.988 Lower level 1.000 Household income Higher level 0.510 0.403 1.600 0.206 1.666 0.755 3.673 Lower level 1.000 Type of occupation Employed 0.092 0.454 0.041 0.840 1.096 0.450 2.671 Unemployed 1.000 Number of dependents 0 to 3 0.965 0.372 6.732 0.009** 2.625 1.266 5.442 4 and above 1.000 Type of health condition Chronic 0.033 0.406 0.007 0.935 1.034 0.467 2.290 Acute 1.000 TCM practice Practice TCM 0.045 0.375 0.014 0.905 1.046 0.502 2.179 Not practice TCM 1.000 Health insurance No health insurance −0.127 0.382 0.110 0.740 0.881 0.416 1.864 With health insurance 1.000 β standardized coefficient, S.E standard error, Exp (β) odds ratio ** p < 0.05 acute illness, this did not influence the WTP for drugs [8, have the highest percentage of WTP as compared to 16]. Most patients were not enthusiastic to pay for their diabetes and IHD. This is in tandem with Bradford et al. drugs and proposed that the government drug subsidy finding, which state that hypertensive patients are more to be sustained [1, 5, 14]. The economic effect of chronic responsive to pay compared to other patients’ with other disease has adversely and disproportionately affected chronic conditions [16, 17]. poor and vulnerable populations in the developing world Our findings illustrated that personal and household [13, 16] including Malaysia. The results indicate that only incomes were significantly associated with WTP for 17.3% of respondents with chronic condition are willing drugs, for both chronic and acute illness. Those who to pay for their drugs. Respondents with hypertension earned higher income were more inclined to pay more for Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 7 of 8 drugs. This result supports our hypothesis and was con - tablet per month which costs about MYR 15 to MYR sistent with the previous studies, which indicated that the 30 [22, 23]. This rough estimation indicates that most higher income level with more disposable income is asso- Malaysians are not prepared to endure [3, 5] the entire ciated with WTP [18, 19, 21]. This relevancy of income cost of their chronic treatment. and WTP, indicate that if drug charge is to be imple- Patients who were not practicing TCM were more will- mented in public dispensing, this may potentially become ing to pay for drugs as compared with those currently a burden to the countries of lower income groups. Inter- practicing TCM. However, the association of WTP and estingly even though patients with higher income were TCM practice were not significant. Previous studies more enthusiastic to pay more for drugs, the percentage revealed that patients who contracted chronic condition was very low. The current price inflation and exuberant are more likely to utilize and purchase complementary cost of living have led to the drastic increase in the cost and alternative medicine [24–26]. They may perceive less of drugs and healthcare as a whole [18, 20, 21], albeit a need on modern drugs [26], or need to purchase them; higher personal or household income [10, 23]. hence the less willing to pay. Respondents with less number of dependents were more willing to pay for drugs (19.1%). This finding is con - Limitation of study sistent with previous study by Okyere which found that Since this study did not represent the private sector pri- the chances of an individual to pay for a ¢5000 health mary care patients, the result would be very centered insurance premium, would be reduced by 10.8%, if there towards public primary care patients. Thus, it is sug - is an increase of 1% in the dependency ratio in the fam- gested that future study should include private primary ily [18, 19]. This elucidated that an increase in number care. As this state is also one of the richest urbanized of dependents was associated with reduced WTP. The states in Malaysia, population of WTP would be expected concept of opportunity cost is seen here [1, 4]; if patients to greatly contradict from other less developed states are required to pay more for expensive chronic disease with a larger number of low income groups. drugs, this would affect their other daily expenses and basic needs [23]. Hence, this substantiates our hypothesis Recommendation that patients with a high number of dependents have low The majority of the respondents (72.2%) were not willing WTP for drugs as compared with those with low number to pay for their drugs. Utilisations on other mechanisms of dependents. Therefore, the increase in the number of need to be addressed tin inspiring its citizens to share the dependents will contribute to less willingness to pay. burden of drug cost. It is proposed that public facilities start The median value of WTP for drugs among patients to implement a staggered payment mechanism whereby who have two co-morbidities was MYR 5 per visit. step-by-step fee charges are implemented within the next Whereas for patients who have three co-morbidities were few years period in order to heighten citizens’ awareness in willing to pay up to a higher level at MYR 10.50 but this assisting the government to control health care costs. Phas- was not statistically significant. Valderas [20] stated as ing these implementations would be much palatable to edu- the number of chronic disease increases, the costs of the cate the public and finally remove the norm of citizens who treatment will also escalate [3, 22]. However from Chang currently enjoy the highly subsidized drugs and health care. states those co-morbidities had no significant increase Prior implementing the new health care financing towards patients’ WTP [21]. WTP has its inert weakness policy such as the proposed National Health Insurance [1, 6] i.e. the population with more disposable income and its benefit packages, it is plausible to start charging would generally be more willing to pay for drugs [9, 12] MYR 10 as the minimum drug charges for each outpa- compared with population who have lower purchas- tient visit that entails subsidised treatment for acute and ing power and income [5]. Bivariate analysis statistically chronic diseases. In countries that already implement showed that patients who earned less than MYR950 per the casemix costing charges, this can be used to start month (considered to be lower income) are less likely to charging patients according to the cost calculated in pay (89%) for drugs compared to those who earned more the casemix system. Government facilities should start than MYR950 per month (80.1%). charging the more affluent population on health services The median value patients are willing to pay for drugs and drugs consumed. The minimum value of MYR10 is was only MYR 10. Taking the highly prevalent diabetes considering the overall median value of patients’ WTP as an example [15, 16, 22] the first line medication pre - for drugs (MYR 10 per visit). However proper targeting is scribed to patient after adequate practice on healthy diet needs for the lower income group. This is whilst exclud - and regular exercise is bi-guanides (Metformin) [22]. ing the vulnerable groups such as the low income earn- For a patient who consumes the maximum dose of Met- ers, pensioners or the elderly. This can be carried out formin (1000 mg two times daily), he or she requires 120 through the implementation of means tested financing. Puteh et al. Asia Pac Fam Med (2017) 16:5 Page 8 of 8 References Conclusions 1. Babar ZD, Izham MIM. Eec ff t of privatization of the drug distribution The results indicate that patients have low WTP for drugs system on drug prices in Malaysia. Public Health. 2009;123:523–33. for both chronic and acute illness even in an urban state of 2. Annual report: Malaysian Ministry of Health. Kuala Lumpur: Pharmaceuti- cal Services Division; 2005. Malaysia. The maximum amount that public patients are 3. Annual report: Malaysia Ministry of Health drug expenditure. Kuala Lum- willing to pay for acute or chronic diseases drugs are consid- pur: Pharmaceutical Services Division; 2010. ered low, the median value was only MYR 10 (USD 3.8) per 4. Hussain HS. Drug control and formulary management in Malaysia. Phar- macoecon Outcomes Res Asia. 2008;11:S158–9. visit. Only three factors were found to be associated with 5. Bridel A. Explaining 1Care: its proposals, rationale and feasibility. Philip- WTP; which are the level personal and household income, pines: Center for Public Policy Studies (CPPS); 2012. and number of dependents. Most of our citizens are in a 6. Quevedo JFM, Hernández IC, Espinosa JG, Escudero GS. The willingness- to-pay concept in question. Rev Saúde Pública. 2009;43:2. comfort zone of receiving highly subsidized health services 7. Adeneye AK, Mafe MA, Appelt B, Idowu ET, Akande DO. Willingness to thus, and may not be prepared to transform from the norm pay for praziquantel treatment in a hyper-endemic community of Ogun of having free services and to pay for his/her own drugs. State, Nigeria. Res Soc Adm Pharm. 2006;2:83–95. 8. Greenberg D, Bakhai A, Neumann PJ, Cohen DJ. Willingness to pay for Authors’ contributions avoiding coronary restenosis and repeat revascularization: results from a SEWP contributed to the design of the study, choosing the co-authors, revi- contingent valuation study. Health Policy. 2004;70:207–16. sions of the manuscripts, formatting, and revision of references used in publica- 9. Michael AM, Bohara AK, Pagán JA. Assessing willingness to pay for cancer tion. SNAA prepared the study logistics, carried out the study and interviewing prevention. Int J Health Care Finance Econ. 2010;10:301–14. the patients. ANA contributed comments on the WTP tool and critiques on the 10. Hwang W, Weller W, Ireys H, Anderson G. Out-of-pocket medical spending WTP concept. RZ and RI helped in the arrangement of local facilities and final for care of chronic conditions. Health Aff (Millwood). 2001;20(6):267–78. approval of draft. All authors read and approved the final manuscript. 11. Jones R, White P, Armstrong D, Ashworth M, Peters M. Managing acute illness. The King’s Fund. 2010. Author details 12. Chen Y T, Ying YH, Chang K, Hsieh YH. Study of patients’ willingness to Faculty of Medicine, Department of Community Health, University Kebang- pay for a cure of chronic obstructive pulmonary disease in Taiwan. Int J saan Malaysia Medical Center, Jalan Yaacob Latif, Bandar Tun Razak, Cheras, Environ Res Public Health. 2016;13(3):273–87. 56000 Kuala Lumpur, Malaysia. Health Administration, Faculty of Business 13. Adams R, Chou YJ, Pu C. Willingness to participate and pay for a proposed Management, Universiti Teknologi MARA, Puncak Alam, Selangor, Malaysia. national health insurance in St. Vincent and the grenadines: a cross-sectional Institute for Health Systems Research, Ministry of Health, Kuala Lumpur, contingent valuation approach. BMC Health Serv Res. 2015;15:148–58. Malaysia. Health District Office, Sepang, Malaysia. 14. Gustafsson-Wright E, Asfaw A, Van der Gaag J. Willingness to pay for health insurance: an analysis of the potential market for new low-cost Acknowledgements health insurance products in Namibia. Soc Sci Med. 2009;69:1351–9. The authors would like to thank the Dean and Hospital Director UKM Medical 15. Population distribution and basic demographic characteristic report Center and the Director of Ministry of Health, Malaysia for the support and 2010. Department of Statistics Malaysia. 2010. http://www.statistics.gov. permission to publish this paper. We would also like to thanks the Director my. Accessed 11 Jan 2013. of Selangor State Health Department and all Health District Officers for their 16. Engelgau M, Rosenhouse S, El-Saharty S, Mahal A. The economic effect fullest cooperation and commitment in this research. Last but not least, we of non-communicable diseases on households and nations: a review of would like to convey our utmost appreciation to the respondents who had existing evidence. J Commun Healthc. 2011;16(Suppl. 2):75–81. been willing to participate in our study. 17. Bradford WD, Kleit A, Krousel-Wood MA, Re RM. Comparing willingness to pay for telemedicine across a chronic heart failure and hypertension Competing interests population. Telemed J E Health. 2005;11(4):430–8. The authors declare that they have no competing interests. 18. Pinto SL, Pharm BS, Goodman MH, Black CD, Lesch D. Identifying factors that affect patients’ willingness to pay for inhaled insulin. Adm Pharm. Availability of data and supporting materials 2009;5:253–61. All data generated or analysed during this study are included in this published 19. Okyere WKA, Akoto IO, Anum A, Appiah EN. Willingness to pay for health article. insurance in a developing economy: a pilot study of the informal sector of Ghana using contingent valuation. Health Policy. 1997;42:223–37. Consent for publication 20. Valderas JM, Starfield B, Sibbald B, Salisbury C, Roland M. Defining comor - All authors approved the manuscript and submission for publication. The bidity: implications for understanding health and health services. Ann permission to publish was also obtained from the Ministry of Health Malaysia Fam Med. 2009;7(4):357–63. and UKM Medical Center. 21. Chang K. Comorbidities, quality of life and patients’ willingness to pay for a cure.for type 2 diabetes in Taiwan. Public Health. 2010;124:284–94. Ethics approval and consent to participate 22. Clinical practice guidelines. Management of type 2 diabetes mellitus. 4th Informed written consent of all respondents had been undertaken. Ethical ed. Kuala Lumpur: Persatuan Diabetes Malaysia, Academy of Medicine clearance was obtained from the Research and Ethics Committee, Medical Malaysia, Ministry of Health Malaysia; 2009. Faculty, Universiti Kebangsaan Malaysia (FF-163-2012) as well as National 23. Consumer price guide. Pharmaceutical Services Division, Ministry of Medical Research Registry, Ministry of Health (NMRR-12-393-12001). Health Malaysia. 2014. http://www.pharmacy.gov.my/v2/ms/apps/drug- price. Accessed 11 Jan 2013. Funding 24. Millar WJ. Patterns of use–alternative health care practitioners. Health We gratefully acknowledge funding support from UKM Medical Center Funda- Rep. 2001;13(1):9–21. mental Research Fund (FF-162-2012) for this study. 25. Quan H, Lai D, Johnson D, Verhoef M, Musto R. Complementary and alternative medicine use among Chinese and white Canadians. Can Fam Physician. 2008;54(11):1563–9. Publisher’s Note 26. Metcalfe A, Williams J, McChesney J, Patten SB, Jetté N. Use of comple- Springer Nature remains neutral with regard to jurisdictional claims in pub- mentary and alternative medicine by those with a chronic disease and lished maps and institutional affiliations. the general population—results of a national population based survey. BMC Complement Altern Med. 2010;10:58. Received: 12 April 2016 Accepted: 22 March 2017

Journal

Asia Pacific Family MedicineSpringer Journals

Published: Apr 4, 2017

References