Patient satisfaction versus political support: Korea's drug-dispensing law revisited

Article information

Abstract

Introduction

In 2000, Korea enacted a law prohibiting doctors from dispensing drugs mainly to prevent the misuse or overuse of drugs, control drug expenditures through appropriate drug use, and improve the quality of healthcare services. Before the law, doctors prescribed and dispensed drugs through their clinics, and pharmacists could give prescriptions for minor conditions such as colds and migraines. Pharmacists could dispense dugs without a prescription, including powerful drugs such as antibiotics and steroids. Under the law, only doctors are allowed to prescribe drugs, and prescription drugs are dispensed only through pharmacies. Patients' complaints about increased medical care costs (i.e., paying for both prescriptions and drugs) and the inconvenience of visiting both clinics and pharmacies have focused the public's attention on the effectiveness of the law. Despite the justifiable objectives of the law, there has been strong opposition from various interest groups, particularly from doctors. Doctors went on strike for months, and many healthcare researchers have questioned the law's intentions [1,2]. These interest groups have paid close attention to various issues surrounding the law, such as increases in healthcare expenditures and patients' inconvenience. They have argued that the law allows large pharmaceutical firms to enter the market, which has traditionally favored small pharmacies [3].

Some political scientists have insisted that the Korean government enacted the law to obtain additional support from the public, infringing on vested rights of doctors, who have been recognized as a politically powerful group. They have argued that the law was enacted based on political considerations, not on healthcare implications, although the law has serious healthcare implications. Hwang [4] raised the question of why the law received support from the Dae Jung Kim administration, although it failed to secure sufficient support from previous administrations, and focused on the role of the government and the president in the policymaking process, whereas most studies have focused on the role of civic groups such as non-governmental organizations [5-11]. This argument provides a new perspective on healthcare policymaking by suggesting that healthcare policies can be implemented by governments driven by political interests, not by professional decisions.

This study addresses three important policy questions from the patients' perspective: whether patients (consumers) are satisfied with the law, whether they support the law, and whether their political alignment influences their satisfaction with and support for the law. For this, the study considers Korea's regional characteristics to evaluate the effects of political alignment on the policy decision because presidential candidates from specific regions have historically been elected as presidents. In addition, the study assesses patients' rational decision on the law by considering their satisfaction with the use of healthcare services.

1. Political background

Korea's political parties or factions can be geographically divided into Gyeongsang (hereafter "the southeastern region"), Jeolla (hereafter "the southwestern region"), and others. Today's political differences stem mainly from the regional background of political parties back in the 1970s. However, these regional differences have a long history, reflecting Korea's 5,000-year history. For instance, there were two countries -"Baekje" in the southwestern region and "Silla" in the southeastern region -in ancient times, and a long - standing feud between them continued until they were united to become "Unified Silla." The political party based in the southeastern region maintained its power by achieving the unification, whereas the southwestern region became politically alienated because it was used as a place of exile. This propensity continues to this day. From 1962 (the start of the industrialization era), three presidents from the southeastern region were consecutively elected, and they intensified the regional divide by prioritizing certain regions and implementing policies that benefited residents in the southeastern region [12,13]. This continued until the election of president Dae Jung Kim, who was supported mainly by the southwestern region. This reflects Koreans' tendency to choose presidential candidates based on their regional identity, not on their long-term political views [14].

Although the Dae Jung Kim administration assumed office with 95% of the votes from the southwestern region in the 1997 presidential election, its political base was weak in other regions[15]. In that election, 40.3% of the voters chose Kim, but his opponent received a similar number of votes (38.7%), indicating weak political support for the Kim administration (Table 1). For this reason, the administration pushed various policies in conjunction with progressive civic groups. This was mainly to curb the establishment through participatory democracy, inducing the public and laborers to become more involved in the decision-making process [17]. The drug-dispensing law and the unification of national health insurance were part of Kim's electoral platform and became the main policies to be implemented.

Table 1

Votes for presidential candidate Dae Jung Kim by region

The drug-dispensing law was a key policy enabled by support from progressive civic groups, which made efforts to fortify their organizational competency and increase their political leverage by participating in the policymaking process [8-11,18]. In other words, the government, the ruling party, and civic groups were united in favor of the law, whereas healthcare professionals, particularly doctors, were against of it. Progressive civic groups supported the law to expand their political power, and doctors opposed it, claiming that it would worsen patients' access to medicine [9-11]. However, the government pushed ahead, despite three strikes in June 2000, which involved more than 70% of all hospitals and clinics.

The law provoked intense conflicts between the two stakeholders and has been constantly criticized as a failure since its implementation. A policy assessment executed by a National Assembly member a year after the enactment of the law suggested that the misuse and overuse of drugs actually increased under the law [19]. In addition, a public survey showed that most Koreans complained about the inconvenience of the law [20,21]. In 2010, a decade since the enactment of the law, scholars, National Assembly members, and even the government called for a reevaluation of the law. Recently, it has been contended that the separation of prescription and dispensation should be terminated because the law has failed to achieve its original goals [22].

Some, particularly doctors, have believed that the law was enacted mainly for pursuing political interests, not for improving the healthcare system. Therefore, this study examines the effects of political interests on the policymaking process for the law by using survey data on drug use and political preferences after the law was enacted.

Data and Descriptive Statistics

By using random digit dialing, we obtained a random sample of 540 Korean adults (age>19) who used healthcare services and prescription drugs after the enactment of the law and interviewed them by telephone in March 2002 to obtain data.

The data focused on the respondents' selection and use of healthcare services (including prescription drugs), their level of satisfaction with healthcare services and drug dispensation under the law, and their willingness to support the drug dispensing law. We evaluated the respondents' satisfaction with the law by using a five-point Likert-type scale ranging from "very dissatisfied" (1) to "very satisfied" (5). We determined their support for the law by asking whether they supported it. In addition, we obtained data on their sociodemographic characteristics, including their age, gender, education level, occupation, monthly income, and place of residence.

Table 2 shows the characteristics of the respondents. Among the 540 respondents, 44% were male; 46% were between the ages of 20 and 39; 53% had a monthly income between USD 1,000 and USD 3,000; and 42% and 30% lived in Seoul and the southeastern region, respectively. Their characteristics generally reflect those of the general population in terms of their age, gender, and place of residence.

Table 2

Characteristics of respondents (n=540)

Methods

We analyzed the relationship between the respondents' support for/satisfaction with the law and their sociodemographic characteristics by using the χ² statistic. To examine the causal relationship between their political alignment and support for the law, we employed a logit model by controlling for their sociodemographic characteristics and satisfaction with the law. We used the place of residence as a proxy for political supports to capture the effects of the respondents' political alignment on their support for the law.

We used a bivariate probit model to check the robustness of the results. Even if patients' satisfaction with and support for a healthcare law may be correlated, it is often difficult to find theoretical or empirical evidence of the causal relationship between these two variables. In this regard, the bivariate probit model has several useful features, including the consideration of the correlation between patients' satisfaction with and support for the law through a regression. In this study, we considered the following bivariate probit model:

y₁^* = χ'₁β₁+ ε₁, y₁= 1 if y₁^* > 0 and 0 otherwise, y₂^* = χ'₂β₂+ ε₂, y₂= 1 if y₂^* > 0 and 0 otherwise, E [ε₁ | χ₁, χ₂] = Var[ε₂ | χ₁, χ₂] = 0, Var [ε₁ | χ₁, χ₂] = Var[ε₂|χ₁, χ₂] = 1, Cov [ε₁, ε₂ | χ₁, χ₂] = ρ,

where y₁^* denotes a latent variable for the patient's support for the law and is coded as 1 if the patient supports the law and 0 otherwise; y₂^* is a latent variable for the patient's satisfaction with the law and is coded as 1 if the patient is satisfied with the law and 0 otherwise; the error terms in the two equations are assumed to be correlated; and ρ represents the correlation between the patient's support for and satisfaction with the law.

Results

1. Respondents' satisfaction with the law

Table 3 shows the descriptive statistics for the respondents' satisfaction with the law. Approximately half (248, 46%) were not satisfied with the law. Their satisfaction was significantly related to their gender and occupation but not to their education level, monthly income, and place of residence.

Table 3

Descriptive statistics for respondents' satisfaction (n=540)

2. Respondents' support for the law

Table 4 shows the descriptive statistics for the respondents' support for the law. Slightly less than half (227, 42.1%) did not support the law. There were significant relationships between their support for the law and all their sociodemographic characteristics except for gender. As in the case of their satisfaction with the law, those between the ages of 40 and 59 were less likely to support the law. College graduates and office workers/professionals were more likely to support the law than high school graduates and blue-collar workers, respectively. Those from the southwestern region were most likely to support the law, and those with a monthly income less than USD 1,000 were more likely to support the law than those making more.

Table 4

Descriptive statistics for respondents' support for the law (n=540)

3. Effects of respondents' political alignment on their support for the law

Table 5 shows the results of the logit regression, which indicate that the respondents who were satisfied with the law were more likely to support it. In addition, those over the age of 40 were more likely to support the law. Blue-collar workers were less likely to support the law than office workers/professionals (odds ratio [OR]=0.23). In addition, those respondents living in Seoul (the national capital), the eastern region, and the southwestern region were much more likely to support the law than those living in the southeastern region (OR=1.55, 3.04, and 3.99, respectively). In particular, the coefficient of region was much higher for the southwestern region than for other regions.

Table 5

Logit result for respondents' support for the law

4. Robustness

Table 6 shows the results for the bivariate probit model (with the correlation between the respondents' support for and satisfaction with the law controlled for). These results verify that the respondents' place of residence had a significant effect on their support for the law but not on their satisfaction with it. Those from the southwestern region were significantly more likely to support the law than those from Seoul, but this was not the case for their satisfaction with the law. Those from the southeastern region were less likely to support the law, but their place of residence had no effect on their satisfaction with the law.

Table 6

Results for the bivariate probit model for respondents' support for the law and satisfaction

Discussion

The study employs cross-sectional data obtained through telephone interviews to investigate how patients' political alignment influences their satisfaction with and support for the law in the context of the policymaking process. The study uses a logit model and a bivariate probit model to capture the effects of patients' political alignment on their satisfaction with and support for the law. To the authors' knowledge, no study has provided an empirical analysis of the effects of regional characteristics on healthcare policymaking. In this regard, this study contributes to the literature by being the first to provide such an analysis, revealing new insights into the relationship between political alignment and healthcare policymaking in Korea.

In order to capture the effect of the political alignment on the law, we used the place of residence as a proxy for the respondents' political propensity because presidential candidates from specific regions have been elected with unquestioning support of the people from the same area historically. For example, Dae Jung Kim, who implemented the law, won the presidency with approximately 95% of the votes from the southwestern region but only about 14% from the southeastern region.

This implies that the Kim administration was established through strong regional support. In other words, if the variable for the region is regarded as a proxy for political alignment, then our results implies that the respondents' political alignment had considerable influence on their support for the law. In short, our results suggest that the respondents' political alignment was a significant determinant of their support for the law but not of their satisfaction with it.

Conclusion

The results provide empirical evidence that regional characteristics have considerable influence on political support and the policymaking process. The level of satisfaction with the law was low across the country. The respondents from the southwestern region, the political base of the ruling party when the law was enacted, were most likely to support the law, whereas those from the southeastern region were less likely to support it than those from other regions. That is, the respondents' place of residence was not significantly related to their satisfaction with the law, but it was to their support for the law, particularly for those respondents from the same region as the ruling party.

These results suggest that patients' political preferences are much more likely than their need for convenience to influence the policymaking process and that political power may induce undesirable outcomes for healthcare policies. This indicates an urgent need for a systematic reform that can prevent a repeat of such trial-and-error policymaking in the design and implementation of healthcare policies. Such a reform can be based on education programs for the public and politicians, the wide dissemination of information, and the facilitation of decision-making processes that make active use of expert opinions on healthcare policies.

This study is limited in that the telephone survey overlooked those individuals without a telephone at home or those who did not answer the call. Nonetheless, the study provides a better understanding of the role of political power in the policymaking process.

Article information Continued