Copayment Adjustments Under Bundled and Two-Tiered Pricing: Welfare Implications for Outpatient Care in Taiwanese Hospitals

Abstract

This study investigates the welfare implications of copayment adjustments on non-prescription drug services and prescription drugs provided through outpatient care in Taiwan’s hospitals under the bundled and 2-tiered pricing scheme. The novelty of this study lies in its analysis of the welfare effects of outpatient copayments under this pricing structure. Structural change models were employed to estimate the welfare effects of copayment adjustments. The analysis reveals that higher copayments prior to physician consultations may lead to a pricing-out effect for low-income individuals in medical centers, while the impact is negligible in regional hospitals. The estimated willingness-to-pay (WTP) for an outpatient visit involving regular and refillable prescription drugs in medical centers is approximately NT$2024 (95% CI: [1978, 2072]) and NT$2312 (95% CI: [2267, 2357]), respectively. These values are 1.71 to 1.87 times higher than the WTP for comparable outpatient visits in regional hospitals, with estimates for regular (NT$1078; 95% CI: [1045, 1111]) and refillable prescription drugs (NT$1353; 95% CI: [1304, 1400]). The WTP for prescription drugs bundled within an outpatient visit ranges from NT$939 (95% CI: [908, 971]) to NT$1210 (95% CI: [1174, 1244]) in medical centers and from NT$308 (95% CI: [293, 322]) to NT$544 (95% CI: [514, 574]) in regional hospitals. Additionally, the share of deadweight loss (DWL) per outpatient visit attributable to regular prescription drugs (with a cost-sharing scheme) ranges from 14.58% to 31.15%, which is significantly lower than that attributable to refillable prescription drugs (without a cost-sharing scheme), ranging from 17.32% to 42.80%. Our findings highlight the underestimation of welfare effects from copayment policy changes and emphasize the 2-tiered scheme as an effective approach to mitigating DWL in outpatient care utilization in Taiwan’s hospitals.

Keywords

copayment adjustments outpatient prescription drugs welfare analysis deadweight loss bundle pricing two-tiered pricing willingness-to-pay compensated variation Taiwan National Health Insurance

Highlights

● Bundled two-tiered pricing better reflects welfare gains from copayment changes.

● Charge higher copayments for refillable prescriptions to cut waste.

● Apply stronger refillable copayments in regional hospitals than in medical centers.

● Evidence backs raising the cap to NT$300 and extending cost-sharing to refillables.

● Guard against regressive effects—higher fees may push low-income patients away from care.

Introduction

Background and Policy Context

Taiwan’s healthcare system operates under a government-run, single-payer National Health Insurance (NHI) program, providing universal health coverage to all residents. This program categorizes healthcare providers into 4 tiers—local clinics (primary care), district hospitals (secondary care), regional hospitals (tertiary care), and medical centers (advanced care, teaching, and research), and offers low-cost outpatient care services in hospitals and allows beneficiaries to visit any physician without referral restrictions. Consequently, since its inception in 1995, a significant portion of Taiwan’s NHI expenditures has been allocated to outpatient care services.^1,2 For instance, in 2021, approximately 70% of total medical expenditures were directed toward outpatient care, with 61.8% of these costs incurred by medical centers (24.5%), regional hospitals (23.7%), and district hospitals (13.6%).³ In addition, a recent study reported that the average number of hospital outpatient visits per resident in Taiwan is approximately 18 times per year, more than double the OECD average.⁴ This high frequency of hospital outpatient visits reflects the prevalence of doctor and hospital shopping in Taiwan’s healthcare system as documented in the healthcare literature.^5
-7 Consequently, the potential economic welfare loss associated with outpatient care utilization in hospitals, as measured by deadweight loss (DWL) due to moral hazard behavior, poses a significant concern regarding the sustainability of Taiwan’s NHI program.^2,8

In order to address the excessive patient volume and medical expenditure associated with outpatient care in hospitals, Taiwan’s National Health Insurance Administration (NHIA) implemented a copayment policy featuring a price-differentiating scheme. Under this policy, patients who visited hospitals without a referral from a local clinic faced a higher copayment fee, while those referred from a local clinic paid a lower copayment fee for their hospital outpatient visits. During our study period from January 2000 to December 2021, copayment adjustments made in 2005 and 2017 were part of this price-differentiating policy.^1,2 Nevertheless, previous studies found that this policy had minimal impact on altering patients’ behavior when seeking outpatient care.^1,2 In response, the NHIA introduced a new copayment policy on July 1, 2023, implementing a bundled and 2-tiered pricing scheme^9,10 aimed at reducing both healthcare expenditures and potential economic welfare losses associated with outpatient care utilization in hospitals. Specifically, outpatient care services provided by Taiwan’s hospitals typically consist of 2 components: non-prescription drug services (such as therapy or treatment) and prescription drugs. Under the new policy, patients pay a user fee for their outpatient visit before seeing a physician, followed by an additional cost-sharing rate (namely, 20% of the reimbursement amount capped at NT$300) for prescription drugs after receiving medical orders, creating a 2-tiered pricing system for the bundled outpatient care services.

It is important to note that the new copayment policy retains the fixed copayment fee for outpatient visits (eg, NT$420 for medical centers and NT$240 for regional hospitals as of April 2017) before patients consult with a physician. Nevertheless, it introduces a cost-sharing scheme for outpatient prescription drugs after medical orders are issued. Specifically, the upper limit on user fees for regular outpatient prescription drugs in both medical centers and regional hospitals has been increased from NT$200 to NT$300. For refillable prescriptions for patients with chronic illnesses, a user fee equivalent to 20% of the reimbursement amount, capped at NT$300, is charged when the prescription is filled in hospitals for the first time; subsequent refills are exempt from additional charges. It is worth noting that the user fee for outpatient prescription drugs, depending on the physician’s orders, ranges from 0 to a maximum of NT$300 in both medical centers and regional hospitals.⁹ Patients were not informed whether an additional user fee for outpatient prescription drugs would apply until after their consultation with physicians. Consequently, the effective price per outpatient-visit for patients consisted solely of the fixed copayment for outpatient care prior to consulting with physicians. This fixed copayment is designed to constrain outpatient care utilization in hospitals. Since non-prescription services (eg, therapies or treatments) and prescription drugs are typically delivered jointly in Taiwan’s hospitals—as complementary goods—the imposition of additional cost-sharing on prescription drugs is unlikely to directly influence their utilization. Instead, it increases the overall cost of hospital outpatient care for patients and serves as a policy tool to reduce potential economic welfare losses due to moral hazard behavior. Further discussions regarding this copayment reform can be found in the NHI Important Policy Forum organized by the NHIA.¹⁰

The research question is whether the bundled and 2-tiered pricing scheme effectively mitigates DWL in outpatient care utilization. we employed a time-series structural change model, as proposed by Bai and Perron,¹¹ to estimate the time-varying price and income elasticities of outpatient care demand in these settings.

Following this, we evaluated the welfare effects of increased outpatient user fees by simulating WTP values for hospital outpatient care services and assessing the DWL in outpatient care utilization due to the moral hazard behavior.⁸ The main contributions of this study are as follows:

Bundled Service Framework: This study evaluates the welfare effects of user fees under a bundled and 2-tiered pricing scheme within Taiwan’s NHI program.

Dynamic Elasticity Estimation: By employing a time-series structural change model, this research advances beyond traditional regression approaches, enabling the estimation of time-varying price and income elasticities rather than assuming constant effects.

Welfare Simulations: The study conducts numerical simulations of WTP and DWL for bundled outpatient services, offering empirical support for refining copayment structures and cost-sharing policies.

Policy Implications: The results indicate that the 2-tiered pricing scheme may help reduce welfare loss arising from moral hazard in the NHI system.

Literature Reviews

It is essential to recognize that the effectiveness of any user fee policy in reducing both healthcare utilization and potential economic welfare losses is highly dependent on how patients’ behavior in seeking outpatient care responds to changes in copayment fees. Previous studies on the effectiveness of copayment policies in reducing healthcare expenditures indicate a significant negative relationship between copayments and healthcare utilization. For instance, Guindon et al conducted a systematic review that included 43 eligible studies investigating the impact of cost-sharing on drug and healthcare utilization.¹² This review examined the effects of prescription drug insurance and cost-sharing schemes on drug use, healthcare utilization, and health outcomes. The evidence suggested that drug use is negatively correlated with cost-sharing and positively correlated with drug insurance, though the magnitude of these associations with healthcare utilization is relatively small. Similarly, Kiil and Houlberg conducted a systematic review on the effects of copayments on healthcare demand, identifying 47 eligible studies, most of which found that copayments reduce outpatient care utilization.¹³

Beyond these systematic reviews, Xu and Pei used the difference-in-differences (DID) approach to explore the impact of coinsurance reduction on outpatient care utilization in China, finding no significant impact.¹⁴ Nevertheless, applying the same DID approach in Germany, Xu and Bittschi concluded that the abolition of copayments led to a short-term increase in outpatient care utilization.¹⁵ In addition, Kato et al found a significant influence of cost-sharing on outpatient care utilization in Japan through a regression discontinuity analysis,¹⁶ while Park and Choi observed a similar effect in Korea using the DID approach.¹⁷ Similarly, Jo et al used a linear probability model with a DID approach to examine the impact of differential copayments on patient healthcare choices, finding significant effects on the choice between secondary/tertiary and primary care.¹⁸ A negative relationship between various cost-sharing schemes—such as deductibles,^19,20 capitation,²¹ and tiered cost-sharing^22,23—and healthcare utilization has also been documented in the literature. Furthermore, the significant influence of user fees on various types of healthcare utilization has also been documented in studies focusing on dental care,²⁴ drug use,^20,23,25,26 emergency care,^8,27,28 mental healthcare,²⁹ prenatal care,³⁰ primary care,^31
-34 rehabilitation care,³⁵ and long-term care.³⁶ The literature review mentioned above incorporates cross-country evidence, offering a broader understanding of the impacts of cost-sharing mechanisms across different healthcare systems.

It is important to note that Taiwan’s Ministry of Health and Welfare (MOHW) has revised copayment policies for outpatient care services multiple times since the inception of Taiwan’s NHI program in 1995. The impact of these copayment changes on outpatient care utilization has been a significant focus of health policy research. For instance, Chen et al employed a segmented time-series model to assess the effect of the price-differentiated copayment policy introduced on July 15, 2005, on outpatient care utilization,³⁷ while another study by Chen et al. utilized a sample selection model for count data to analyze the same policy.² Both studies concluded that copayment adjustments negatively influenced outpatient care utilization. Expanding on these studies, Juan et al applied a difference-in-differences (DID) approach to examine the 2005 copayment adjustment effect on effective care for patients with persistent asthma, similarly identifying a negative relationship between copayments and healthcare utilization.³⁸ Additionally, Lin et al.¹ and Chen et al.²⁷ simulated the effectiveness of copayment policies, revealing adverse effects on outpatient and emergency care utilization, respectively. Chen further conducted a welfare analysis of copayment adjustments for emergency department visits, demonstrating that while higher copayments did not produce a pricing-out effect or significant welfare gains, a negative long-term association between copayments and emergency department utilization was evident.⁸ Moreover, Yang et al investigated the influence of persistent behavior and cost-sharing policies on outpatient care utilization among Taiwan’s elderly population using a dynamic panel count data model.³⁹ Their findings suggest that elderly individuals exhibit greater price sensitivity over time, indicating that copayment interventions may be more effective in the long term. Furthermore, Liu et al employed a conventional time-series model to identify the determinants of medical spending, concluding that the 1999 upward copayment adjustments significantly reduced medical spending.⁴⁰ This negative relationship was further corroborated by Huang and Tung, who confirmed the reduction in outpatient care utilization following the 1999 adjustments.⁴¹ This body of literature underscores the consistent evidence of a negative relationship between copayment adjustments and healthcare utilization across various settings and methodologies.

Motivations and Objectives

Although previous studies have indicated that copayment policies may effectively reduce healthcare utilization, they typically do not consider outpatient care services as a bundle of non-prescription drug services (such as therapy or treatment) and prescription drugs under Taiwan’s NHI program. Moreover, these studies often overlook the reasons behind the NHIA’s transition from a price-differentiating scheme (with different copayment fees for non-referral and referral patients) to a bundled and 2-tiered pricing scheme (with a fixed copayment fee before consultation with a physician and an additional user fee for prescription drugs after receiving medical orders). To address this research gap, the objective of this study is to simulate the welfare effects of the new bundled and 2-tiered pricing cost-sharing policy on outpatient care utilization in medical centers and regional hospitals under Taiwan’s NHI program.

Data and Methods

Research Design & Welfare Effect Analysis

This study adopts an observational time series design, employing multiple time series models to analyze secondary data and address its primary research objective. It is important to clarify that this study adopts a macroeconomic simulation of the welfare effects of the bundled and 2-tiered pricing scheme prior to the enforcement of the 2023 copayment policy reform, rather than as a microeconomic evaluation of the policy’s effectiveness after its implementation. This approach is necessary because the 2023 claims data from the NHI program may not be available for research purposes until as late as the end of 2025. Furthermore, under a new data collection regulation introduced in 2025, NHI beneficiaries now have the right to withhold their data from being used for research purposes, which may further restrict data accessibility. The empirical procedure used to investigate the welfare effect of copayment adjustment on outpatient care utilization in Taiwan’s hospitals 2 steps: First, we applied the time-series structural change model to reveal dynamic change of patient behavior when seeking outpatient care. Second, the net welfare effect due to copayment adjustment was examined. The implementation of the user fee policy with a bundled and 2-tiered pricing scheme for outpatient care utilization in hospitals by the NHIA involves 2 main stages (see Figure 1). In the first stage, patients are required to pay a fixed copayment (P₀) for their outpatient visits (Q₀) before consulting with a physician. The full price for outpatient visits (Q₀), which includes non-prescription drug services (such as treatment or therapy), is represented as P₁, where P₁ equals P₀ plus the reimbursement for these non-prescription drug services (P₁ − P₀). In the second stage, the full price for outpatient visits (Q₀), including both non-prescription and prescription drugs, is denoted as P_f, where P_f equals P₁ plus the reimbursement for prescription drugs (P_f − P₁).

Figure 1.

Illustration for welfare effects.

It is important to note that the user fee for prescription drugs is a portion of the full price of the prescription drugs (P_f − P₁) since the theoretical maximum user fee charged for prescription drugs corresponds to the full price of the prescription drugs (P_f − P₁). Nevertheless, the imposition of a user fee for prescription drugs effectively reduces the DWL associated with their outpatient care utilization, as patients are more likely to adhere to physicians’ orders and pay the associated user fee for prescription drugs. Consequently, the DWL for outpatient visits (Q₀) is less than the area Δafh, while the WTP per outpatient visit remains within the area Δacf, divided by the number of outpatient visits (Q₀). Since we specified our demand function for outpatient care as the Cobb-Douglas functional form, the change in consumer surplus(ΔCS) can be approximated using the compensating variation (CV) derived by Hausman⁴² for a change in copayment per outpatient visit from P_i to P_j. The technical details of the empirical procedure for the welfare analyses are provided in the Supplemental Materials.

Variables Definitions & Sources of Data

The data for this research were obtained from Taiwan’s National Health Insurance Research Database (NHIRD), the Demographic Statistics Database (DSD), and the Macroeconomics Statistics Database (MSD) administered by the Taiwanese government. The data from the DSD and MSD databases are publicly available; however, access to data from the NHIRD is restricted and requires a formal application and approval. Links to these databases are provided in the Supplemental Materials. This study utilized only aggregate data from all residents in Taiwan and did not involve any individual-level data. As such, the use of de-identified data does not pose any ethical concerns for this study. The monthly outpatient care demand per patient in medical centers and regional hospitals was calculated as the average number of outpatient visits per patient in these hospitals. Copayment fee for physician consultations and user fee for both regular and refillable prescription drugs were determined based on the NHIA regulations for outpatient care visits. Note that refillable prescriptions are issued exclusively to patients with chronic illnesses under NHI regulations. Thus, monthly reimbursements for physician consultations and regular prescription drugs per outpatient visit in medical centers and regional hospitals were sourced from the entire pool of Taiwan’s NHI beneficiaries, while data on reimbursements for first-time refillable prescriptions were specifically gathered from patients with chronic illnesses.

The monthly wage income level was specified as the regular wage income in the industrial sector. To ensure consistency and comparability across time, all monetary values, including copayments, user fees, reimbursements, and income variables, were adjusted to 2021 price levels using relevant price indices, such as the medical price index and consumer price index. Control variables in our empirical analysis included the total wage income index for the healthcare industry (which reflects the income level of healthcare workers), the old-age healthcare-worker dependency ratio (measured as the number of individuals aged 65 and above per 100 healthcare workers, indicating the burden on healthcare workers due to population aging), monthly dummy variables, indicators for copayment policy changes, and the COVID-19 pandemic period. The breakpoint unit root test⁴³ was employed to assess stationarity in the time series, while the Hodrick-Prescott filter⁴⁴ was applied to extract the cyclical component of the old-age healthcare-worker dependency ratio, ensuring stationarity.

Data Collection

Given that the revised copayment policy applies exclusively to regional hospitals and medical centers in Taiwan, this study focuses on aggregate outpatient care utilization by all NHI beneficiaries in these 2 hospital types from January 2000 to December 2021, yielding a total of 264 monthly observations. The study adheres to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines,⁴⁵ and the completed EQUATOR checklist is presented in Supplemental Table A3 of the Supplemental Materials.

Results

Figure 2 presents time series plots of all variables utilized in this study. To maintain conciseness, the descriptive statistics and the results of unit root tests and structural break identification in Table 1 are not discussed here; interested readers are referred to the Supplemental materials for detailed information. Table 2 presents the estimation results of the time-series structural change model for outpatient care demand in medical centers and regional hospitals. As indicated at the bottom of Table 2, the time-series structural change model for outpatient care demand in medical centers satisfies the assumptions of goodness-of-fit tests for normality, serial correlation, heteroskedasticity, and stability in the residuals. Although the time-series structural change model for outpatient care demand in regional hospitals meets all goodness-of-fit assumptions, it fails to reject the null hypothesis of no serial correlation in the residuals (F-statistic = 5.195, p-value < .01). Consequently, T-statistics for regional hospitals were computed by dividing the estimated coefficients by the Newey and West standard errors to correct for potential bias due to serial correlation in the residuals.⁴⁶

Figure 2.

Time plots for all variables used in this study. (a) Outpatient care visits. (b) Copayment per visits. (c) Regular wage income. (d) Total wage income index for healthcare industry. (e) Old-age healthcare-worker dependence ratio. (f) Reimbursement payment for MC outpatient care. (g) Reimbursement payment for RH outpatient care.

Table 1.

Descriptive Statistics, Unit Root Tests, and Structural Breaks Identification.

Panel A: Variables used for the structure change model
Variables	Description	Mean	SD	Med	Min	Max
Q^MC	Monthly medical center outpatient visits per patient.	1.768	0.089	1.773	1.528	1.971
Q^RH	Monthly regional hospital outpatient visits per patient.	1.768	0.085	1.772	1.532	1.944
COP^MC	Real copayment per medical center outpatient visit at 2021 price level	374.191	55.787	389.400	262.467	443.514
COP^RH	Real copayment per regional hospital outpatient visit at 2021 price level	240.731	32.742	251.243	174.978	295.676
INC	Monthly regular wage income in the industrial sector at 2021 price level (NT$1000;1 US$ = 30 NT$)	39.142	1.016	39.161	36.433	41.411
HWI	Monthly total wage income index for the healthcare industry, with 2021 as the base year.	94.167	19.965	88.770	72.350	188.360
OHDR	Old-age healthcare-worker dependency ratio as the number of individuals aged 65 and above per 100 healthcare workers	24.997	2.111	25.127	20.783	27.793
ln(OHDR)	Logarithmic transformation of old-age health-care-worker dependency ratio	3.215	0.087	3.224	3.034	3.325
C_ln(OHDR)	Cyclical component of logarithmic transformation of old-age healthcare-worker dependency ratio	0.000	0.008	0.000	−0.023	0.018
CPC	Copayment policy adjustment dummy variable, (CPC = 1 for July and August of 2005 and April and May of 2017).	0.015	0.122	0.000	0.000	1.000
COV-19	Pandemic surge from Dec, 2019 to the end of the study period	0.095	0.293	0.000	0.000	1.000

Panel B: Variables used for the willingness-to-pay for outpatient care services
Variables	Description	Mean	SD	Med	Min	Max
OPCV^MC	Reimbursement payment for non-prescription drug services per medical center outpatient visit at 2021 price level (NT$)	1148.189	104.884	1159.347	891.615	1341.422
OPCV^RH	Reimbursement payment for non-prescription drug services per regional hospital outpatient visit at 2021 price level (NT$)	1118.266	95.730	1122.445	868.187	1293.820
REPD^MC	Reimbursement payment for regular prescription drugs per medical center outpatient visit at 2021 price level (NT$)	1229.982	148.291	1204.630	949.033	2060.667
REPD^RH	Reimbursement payment for regular prescription drugs per regional hospital outpatient visit at 2021 price level (NT$)	717.447	68.058	707.325	573.256	1174.843
RFPD^MC	Reimbursement payment for the first refill of prescription drugs at medical centers, expressed in 2021 prices (NT$)	1622.104	140.220	1589.595	1390.976	2029.996
RFPD^RH	Reimbursement payment for the first refill of prescription drugs at regional hospitals, expressed in 2021 prices (NT$)	1276.027	192.359	1204.950	1042.398	1809.269

Panel C: Break point unit tests
Variables	Constant	Break date	Variables	Constant + trend	Break date
ln(Q^MC)	−8.466^***	2001:M02	ln(Q^MC)	−18.968^***	2000:M08
ln(Q^RH)	−8.518^***	2001:M04	ln(Q^RH)	−18.510^***	2001:M02
ln(COP^MC)	−9.735^***	2005:M06	ln(COP^MC)	−10.264^***	2005:M06
ln(COP^RH)	−8.105^***	2005:M06	ln(COP^RH)	−53.533^***	2005:M06
ln(INC)	−5.029^***	2000:M05	ln(INC)	−7.287^***	2008:M05
ln(HWI)	−11.550^***	2001:M01	ln(HWI)	−13.817^***	2001:M01
ln(OHDR)	−4.245^*	2001:M12	ln(OHDR)	−4.257	2009:M12
C_ln(OHDR)	−5.911^***	2003:M08	C_ln(OHDR)	−5.407^**	2002:M05

Panel D: Tests for structural breaks of the outpatient care demand function
	Medical centers			Regional hospitals
	Breaks	Sup F	Date	Breaks	Sup F	Date
	0 vs 1	388.010^***	2005:M07	0 vs 1	460.242^***	2005:M07
Sequential sup F tests	1 vs 2	47.201^***	2008:M10	1 vs 2	41.628^***	2008:M10
	2 vs 3	19.932^**	2012:M02	2 vs 3	35.635^***	2012:M02
	3 vs 4	13.397		3 vs 4	28.587^***	2018:M09
				4 vs 5	2.732

Note. SD and Med represent standard deviation and median, respectively, and the sample period ran from January 2001 to December 2021, resulting in a total of 264 monthly observations. The upper bound of breaks (M) and the trimming percentage (τ) of samples were set to 5% and 15%, respectively. The number of breaks was chosen if the sequential Sup F statistic exceeded the critical values generated by Bai and Perron.¹¹

“***”, “**”, and “*” represent 1%, 5%, and 10% significance levels, respectively.

Table 2.

Estimated Results for the Structure Change Model.

Medical centers				Regional hospitals
Regime	Parameter	Coeff	T-stat	Regime	Parameter	Coeff	T-stat
First regime 2000:M01\|2005:M06	ln(COP ^MC )	−0.112	−1.836 ^*	First regime 2000:M01\|2005:M06	ln(COP ^RH )	−0.282	−4.745 ^***
	ln(INC)	0.610	6.927 ^***		ln(INC)	0.418	3.403 ^***
	ln(HWI)	−0.034	−1.910 ^*		ln(HWI)	−0.050	−2.495 ^**
	C_ln(OHDR)	−0.082	−0.367		C_ln(OHDR)	0.175	0.645
Second regime 2005:M07\|2008:M09	ln(COP ^MC )	−0.344	−2.634 ^***	Second Regime 2005:M07\|2008:M09	ln(COP ^RH )	−0.288	−2.713 ^***
	ln(INC)	0.737	5.795 ^***		ln(INC)	0.422	3.116 ^***
	ln(HWI)	−0.010	−0.487		ln(HWI)	−0.026	−1.324
	C_ln(OHDR)	−0.166	−0.240		C_ln(OHDR)	−0.443	−0.801
Third regime 2008:M10\|2012:M01	ln(COP ^MC )	−0.492	−2.700 ^***	Third regime 2008:M10\|2012:M01	ln(COP ^RH )	−0.599	−5.286 ^***
	ln(INC)	0.837	6.805 ^***		ln(INC)	0.607	5.313 ^***
	ln(HWI)	−0.035	−1.687 ^*		ln(HWI)	−0.065	−2.610 ^***
	C_ln(OHDR)	−0.268	−0.425		C_ln(OHDR)	−0.478	−1.284
Fourth regime 2012:M02\|2021:M12	ln(COP ^MC )	−0.080	−1.751 ^*	Fourth regime 2012:M02\|2018:M08	ln(COP ^RH )	−0.375	−3.692 ^***
	ln(INC)	0.601	5.284 ^***		ln(INC)	0.483	4.846 ^***
	ln(HWI)	−0.025	−1.692 ^*		ln(HWI)	−0.047	−3.541 ^***
	C_ln(OHDR)	−0.777	−2.372 ^**		C_ln(OHDR)	−0.439	−1.774 ^*
				Fifth regime 2018:M09\|2021:M12	ln(COP ^RH )	−1.185	−2.111 ^**
					ln(INC)	0.890	3.095 ^***
					ln(HWI)	−0.024	−0.714
					C_ln(OHDR)	−0.159	−0.410
Non-breaking control variables	Constant	−5.148	−4.953 ^***	Non-breaking control variables	Constant	−2.187	−1.519
	D2(February)	−0.054	−7.518 ^***		D2(February)	−0.059	−7.931 ^***
	D3(March)	0.005	0.613		D3(March)	0.000	0.014
	D4(April)	−0.014	−1.587		D4(April)	−0.021	−2.313^**
	D5(May)	−0.001	−0.097		D5(May)	−0.009	−0.981
	D6(June)	−0.014	−1.564		D6(June)	−0.022	−2.310 ^**
	D7(July)	0.006	0.730		D7(July)	−0.003	−0.346
	D8(August)	0.009	1.031		D8(August)	−0.002	−0.292
	D9(September)	−0.006	−0.693		D9(September)	−0.014	−1.607
	D10(October)	0.012	1.414		D10(October)	0.004	0.440
	D11(November)	−0.003	−0.346		D11(November)	−0.012	−1.411
	D12(December)	0.006	0.746		D12(December)	−0.003	−0.352
	CPC	−0.001	−0.147		CPC	0.001	0.077
	Cov-19	−0.005	−0.895		Cov-19	−0.014	−1.164
Goodness-of-fit	Null hypotheses	Stat	p value	Goodness-of-fit	Null hypotheses	Stat	p value
	Normality	BJ = 1.047	0.593		Normality	BJ = 3.294	0.193
	No-serial-corr	F = 1.236	0.293		No-serial-corr	F = 5.195 ^***	<0.01
	Homoskedasticity	F = 0.894	0.626		Homoskedasticity	F = 1.090	0.347
	Stability	F = 1.586	0.209		Stability	F = 0.224	0.952
	R²	88.99%			R²	90.39%
	Adjusted-R²	87.63%			Adjusted-R²	89.02%

Note. T-statistics for regional hospitals were computed as the estimated coefficients divided by the Newey-West standard errors. Bold fonts represent 10% or rigorous significance levels.

“***”, “**”, and “*” represent 1%, 5%, and 10% significance levels, respectively.

Importantly, the results from Table 2 indicate that the price elasticities of outpatient care demand are significantly negative, and income elasticities are significantly positive across various regimes in both medical centers and regional hospitals. These results suggest that an increase in the copayment fee per outpatient visit prior to physician consultations leads to a decrease in outpatient care demand, and confirm that outpatient care is a normal good for patients. Additionally, the total wage income index for the healthcare industry is significantly and negatively associated with outpatient visits across most regimes in both medical centers and regional hospitals. These findings imply that lower income among healthcare workers correlates with higher outpatient care volume, consistent with the implementation of the global budgeting method for hospital services. Moreover, the cyclic component of the old-age healthcare-worker dependency ratio is significantly and negatively correlated with outpatient visits after February 2012 in both medical centers and regional hospitals. These findings indicate that an increasing elder care burden reduces outpatient care volume after this period. Furthermore, the study found that indicators for copayment policy changes and the COVID-19 pandemic period did not produce significant effects on outpatient visits in either medical centers or regional hospitals. These results suggest that the time-series structural change model effectively accommodated these significant events. Finally, the R² values for the models range between 88.99% and 90.39%, indicating that the explanatory variables included in the model can explain approximately 90% of the variation in outpatient visits in medical centers and regional hospitals.

To evaluate whether adjustments in outpatient care copayments may have regressive effects on outpatient care utilization across different income groups, Figure 3 presents scatter plots of negative price elasticities of outpatient care demand. The scatter plots are divided into 4 quadrants based on the median or mean values of negative price elasticities and monthly wage income. Observations in the first and third quadrants indicate concordance between negative price elasticities and monthly wage income, while those in the second and fourth quadrants represent discordance. As illustrated in Figure 3, the odds ratios of concordance versus discordance between negative price elasticities and monthly wage income are significant for medical centers, but not for regional hospitals, regardless of whether the median or mean is used to define concordant and discordant areas. Specifically, the odds ratios for medical centers are 0.428 (95% CI: [0.247, 0.739]) and 0.477 (95% CI: [0.277, 0.820]) when using the median and mean as the cut-off points, respectively. These findings suggest that the low (high) income group exhibits higher (lower) price elasticity for outpatient care utilization in medical centers, whereas no significant differences in price elasticities are observed for outpatient care utilization in regional hospitals across different income groups.

Figure 3.

Price elasticities versus monthly regular wage. (a) NegativePE_MC (median = 0.113) versus income (median = 39.161). (b) NegativePE_RH (median = 0.375) versus income (median = 39.161). (c) NegativePE_MC (mean = 0.189) versus income (median = 39.142). (d) NegativePE_RH (mean = 0.496) versus income (mean = 39.142).

Table 3 presents the simulated results for WTP and DWL values per outpatient visit in medical centers and regional hospitals under Taiwan’s NHI program. The BJ statistics for testing the null hypothesis of normality in WTP and DWL values yielded p-values less than 0.1, indicating a violation of the normality assumption. As a result, the 95% confidence intervals for WTP and DWL for an outpatient visit (with bundled non-prescription drug services and prescription drugs) were computed using the bias-corrected and accelerated (BCa) bootstrap method with 10 000 repetitions. Specifically, the WTP for an outpatient visit with regular prescription drugs in medical centers is approximately NT$2024 (95% CI: [1978, 2071]). Of this amount, 46.39% (NT$939; 95% CI: [908, 971]) is attributable to regular prescription drugs, while the remaining 53.61% (NT$1085; 95% CI: [1069, 1102]) is associated with non-prescription drug services. Similarly, the WTP for an outpatient visit with refillable prescription drugs in medical centers is approximately NT$2312 (95% CI: [2267, 2357]), with 52.33% (NT$1210; 95% CI: [1174, 1244]) attributable to refillable prescription drugs, paid during the first refill, and 47.67% (NT$1102; 95% CI: [1087, 1118]) allocated to non-prescription drug services. In contrast, the WTPs for outpatient visits with regular and refillable prescription drugs in regional hospitals are notably lower than those in medical centers. Specifically, the WTP for an outpatient visit with regular prescription drugs in regional hospitals is approximately NT$1078 (95% CI: [1045, 1111]). Of this amount, 28.54% (NT$308; 95% CI: [293, 322]) can be attributed to regular prescription drugs, while 71.46% (NT$771; 95% CI: [751, 790]) pertains to non-prescription drug services. Additionally, the WTP for an outpatient visit with refillable prescription drugs in regional hospitals is approximately NT$1353 (95% CI: [1304, 1400]). Of this amount, 40.22% (NT$544; 95% CI: [514, 574]) is related to refillable prescription drugs, paid during the first refill, while 59.78% (NT$809; 95% CI: [788, 829]) corresponds to non-prescription drug services.

Table 3.

Estimated Results for Willingness-To-Pay (WTP) and Deadweight Los (DWL) Per Outpatient Care Visit in Hospitals.

Variables	Descriptions	Medical centers					Regional hospitals
Variables	Descriptions	Mean	95% LB	95% UB	Share (%)	BJ Stat	Mean	95% LB	95% UB	Share (%)	BJ stat
WREPD (1)	WTP for regular prescription drugs per outpatient care visit	939.148	907.507	970.748	46.39	4.753^*	307.776	293.145	322.077	28.54	36.948
WNREPD (2)	WTP for non-prescription drugs services per outpatient care visit	1085.271	1068.728	1102.250	53.61	23.071	770.674	750.826	789.744	71.46	70.012
WTPRE (1) + (2)	WTP for an outpatient care visit (considered as a bundle of outpatient care services).	2024.419	1977.920	2071.164	100.00	15.615	1078.450	1044.892	1111.186	100.00	67.220
WRFPD (3)	WTP for refillable prescription drugs per outpatient visit	1210.138	1174.474	1244.195	52.33	31.166	544.067	513.502	573.875	40.22	15.610
WNRFPD (4)	WTP for non-prescription drugs services per outpatient care visit	1102.277	1086.829	1118.163	47.67	20.590	808.766	787.752	828.886	59.78	81.168
WTPRF (3) + (4)	WTP for an outpatient care visit (considered as a bundle of outpatient care services).	2312.415	2266.792	2357.184	100.00	24.109	1352.832	1303.987	1400.412	100.00	46.104
DREPD (1)	DWL for an outpatient care visit attributable to regular prescription drugs	74.610	52.380	97.624	14.58	50.699	266.202	249.247	283.587	31.15	97.976
DNREPD (2)	DWL for an outpatient care visit attributable to non-prescription drug services	437.108	425.296	449.136	85.42	17.901	588.323	562.268	615.171	68.85	73.324
DWLRE (1) + (2)	DWL for an outpatient care visit (considered as a bundle of outpatient care services).	511.719	478.537	546.211	100.00	47.804	854.525	812.459	898.322	100.00	82.392
DRFPD (3)	DWL for an outpatient care visit attributable to refillable prescription drugs	88.029	55.943	121.128	17.32	45.059	401.773	381.869	421.980	42.20	42.957
DNRFPD (4)	DWL for an outpatient care visit attributable to non-prescription drugs services	420.103	408.816	431.673	82.68	11.070	550.231	522.837	578.341	57.80	74.600
DWLRF (3) + (4)	DWL for an outpatient care visit (considered as a bundle of outpatient care services).	508.132	466.817	550.924	100.00	49.293	952.004	906.960	997.953	100.00	77.134

Note. “WTP,” and” DWL” mean “willingness-to-pay” and “deadweight loss,” respectively. The 95% confidence intervals were computed using the bias-corrected and accelerated (BCa) bootstrap method with 10 000 repetitions. Bold fonts (“*”) denote 5% (10%) or rigorous significance levels.

Although the WTP for outpatient visits with regular and refillable prescription drugs in regional hospitals is significantly lower than that in medical centers, the DWL for these visits shows the opposite trend. Specifically, the DWL for an outpatient visit with regular prescription drugs in regional hospitals is approximately NT$855 (95% CI: [812, 898]), which is considerably higher than in medical centers (NT$512; 95% CI: [479, 546]). The DWL attributable to regular prescription drug services for an outpatient visit in regional hospitals is approximately NT$266 (95% CI: [249, 284]), also higher than in medical centers (NT$75; 95% CI: [52, 98]). Similarly, the DWL associated with non-prescription drug services for an outpatient visit with refillable prescription drugs in regional hospitals is approximately NT$588 (95% CI: [562, NT$615]), which again surpasses that in medical centers (NT$437; 95% CI: [425, 449]). In terms of the distribution of total DWL for an outpatient visit with regular prescription drugs, 14.58% is attributable to prescription drug services and 85.42% to non-prescription drug services in medical centers, compared to 31.15% and 68.85%, respectively, in regional hospitals. A similar pattern is observed for outpatient visits with refillable prescription drugs. The DWL in regional hospitals is approximately NT$952 (95% CI: [907, 998]), which is significantly higher than in medical centers (NT$508; 95% CI: [467, 551]). The DWL attributable to refillable prescription drug services in regional hospitals is approximately NT$402 (95% CI: [382, 422]), again higher than in medical centers (NT$88; 95% CI: [56, 121]). In addition, the DWL related to non-prescription drug services in regional hospitals is approximately NT$550 (95% CI: [523, 578]), compared to NT$420 (95% CI: [409, 432]) in medical centers. The shares of DWL for an outpatient visit with refillable prescription drugs attributable to prescription drug services and non-prescription drug services are approximately 17.32% and 82.68% in medical centers, versus 42.20% and 57.80% in regional hospitals, respectively. All results reported in Table 3 have been validated through robustness checks using the pre-COVID-19 baseline data (see Supplemental Materials for details).

Discussion

Policy Implications

Several policy implications can be derived from our empirical findings, which are outlined as follows: First, the design of the copayment policy for outpatient care utilization in hospitals under Taiwan’s NHI program is divided into 2 schemes: the price-differentiating scheme^1,2 (ie, varying copayments for referral vs non-referral outpatient visits) and the bundled and 2-tiered pricing scheme^9,10 (ie, a fixed copayment for non-prescription drug services and a cost-sharing rate with a cap for prescription drugs for an outpatient visit). It is important to note that non-prescription and prescription drug services are complementary and bundled in hospital outpatient visits, and patients are often unaware of the prescription drug costs until after their consultation with a physician. The fixed copayment for non-prescription drug services represents the effective price patients face, while the cost-sharing rate for prescription drugs is unlikely to influence patients’ WTP for outpatient visits, as demonstrated in Figure 1. Nevertheless, the cost-sharing rate for prescription drugs was intended as a tool to reduce DWL associated with outpatient care utilization in hospitals.

Second, prior research on copayment adjustments under the price-differentiating scheme has indicated negligible effects on medical care utilization and DWL associated with outpatient care utilization.^1,2,8,27 Evidence from Table 3 shows that the DWL per outpatient visit resulting from regular prescription drugs (with a 20% cost-sharing rate capped at NT$200) is significantly lower than that associated with refillable prescription drugs (which have no cost-sharing rate) in both medical centers and regional hospitals. These findings suggest that neglecting to account for the bundled and 2-tiered pricing scheme may lead to an underestimation of the welfare gains resulting from copayment policy changes under Taiwan’s NHI program.

Third, the results from Table 3 indicate that the share of DWL attributed to refillable prescription drugs bundled within an outpatient care visit, ranging from 17.32% in medical centers to 42.20% in regional hospitals, is substantially higher than that of regular prescription drugs, which ranges from 14.58% in medical centers to 31.15% in regional hospitals. Therefore, in order to reduce the DWL associated with outpatient care utilization, a higher cost-sharing rate should be applied to refillable prescription drugs compared to regular prescription drugs within an outpatient visit. Additionally, the higher cost-sharing rate should be more prominently applied to refillable prescription drugs obtained from outpatient care visits to regional hospitals compared to medical centers.

Fourth, the results from Table 3 indicate that the share of WTP for non-prescription drug services bundled within an outpatient care visit ranges from 47.67% to 53.61% of the total WTP for an outpatient visit in medical centers, and from 59.78% to 71.46% in regional hospitals. These findings may validate the NHIA’s practice of using copayment adjustments for non-prescription drugs services as a means to manage outpatient care utilization in hospitals, despite the documented ineffectiveness of such policies in the literature.^1,2,8,27,47 Furthermore, the WTP for prescription drugs bundled within an outpatient visit ranges from NT$939 to NT$1210 in medical centers and from NT$308 to NT$544 in regional hospitals, all of which exceed the newly implemented cost-sharing cap of NT$300, effective as of July 2023. These findings provide strong evidence in support of the new copayment policy, which advocates for an upward adjustment of the cost-sharing cap from NT$200 to NT$300 for outpatient regular prescription drugs, as well as the imposition of a 20% cost-sharing rate with a cap of NT$300 for outpatient refillable prescription drugs. Consistent results from the expanded simulation scenarios presented in the Supplemental Materials further reinforce this policy direction.

Fifth, although the effective price faced by patients is the fixed physician consultation copayment fee, the cost-sharing scheme for prescription drugs within an outpatient visit may increase the overall cost for patients seeking follow-up outpatient care in hospitals. This could potentially price out the poor and the sick, who are most likely to require hospital outpatient care, from Taiwan’s NHI program. While previous studies on the pricing-out effects due to increased user fees for healthcare services have found that such increases generally produce only slight or insignificant pricing-out effects on healthcare utilization under Taiwan’s NHI program,^8,47 evidence from Figure 3 suggests that adjustments in outpatient care copayments may have regressive income effects on outpatient care utilization in medical centers. Therefore, for low-income individuals, even modest copayment increases are perceived as losses due to limited resources, consistent with prospect theory’s emphasis on loss aversion.⁴⁸ This perception heightens the risk of care avoidance or delay, despite positive health outcomes.

Policy Recommendations

Lower Cost-Sharing Rates in Medical Centers

Given the regressive effects of copayment observed in medical centers, policymakers should carefully consider the potential exclusion of vulnerable populations when implementing higher cost-sharing rates on prescription drugs bundled within outpatient care visits in these settings.

Bundled and 2-Tiered Pricing Scheme for Medical Tests

As our analysis supports the effectiveness of the 2-tiered pricing scheme in mitigating DWL in outpatient care utilization, policymakers may consider extending this approach to medical tests provided through outpatient care. Specifically, copayment adjustments for such tests in Taiwan’s hospitals could be explored under the bundled and 2-tiered pricing scheme to enhance efficiency and equity.

Research Limitations

This study has several limitations. First, we lack access to the out-of-pocket payment per outpatient visit, preventing us from accurately measuring the true user fee and full price of outpatient care utilization. Nevertheless, any simultaneous underestimation of these 2 prices is unlikely to introduce significant bias in the net welfare gain or loss, as the net effect on our welfare analysis remains minimal. Second, as Chen argues, the price elasticity of outpatient care demand driven by real copayment changes (primarily due to increases in the medical price index in this study) may differ from that driven by nominal copayment adjustments.⁸ While nominal copayment adjustments may lead to short-term Money Illusion, Lucas’s critique suggests that rational individuals are not susceptible to such illusions.⁴⁹ Nominal copayments are measured in monetary terms, whereas real copayments account for the value of healthcare services. Differences in real copayments reflect variations in the quantities of outpatient care services that an individual’s income can purchase. From the perspective of rational expectations in resource allocation, price elasticity based on real copayment changes is more suitable for calculating the welfare effects of copayment adjustments in the long run. Third, numerical simulations of WTP and DWL based on time-series aggregate data rest on key economic assumptions: stable preferences within a relatively homogeneous population over the study period, the absence of supply constraints—as is typical under Taiwan’s NHI program—and the use of linear approximation methods (eg, the triangle rule illustrated in Figure 2). These assumptions limit the generalizability of our findings to other healthcare systems with differing institutional or market conditions. Fourth, the implementation of a bundled and tiered pricing scheme may heighten patients’ awareness of total costs associated with follow-up outpatient visits. Nevertheless, as the empirical analysis is based on aggregate time-series data, the resulting inferences pertain only to initial outpatient visits and do not capture behavior related to subsequent follow-up care. Therefore, future research should employ individual-level data to better capture the impact of the bundled and tiered pricing scheme on care-seeking behavior across the continuum of visits—from initial to follow-up. Microeconomic evaluation methods, such as difference-in-differences methodology, should be applied to identify the causal effects of copayment changes on healthcare utilization decisions and behavioral responses.

Conclusion

The primary objective of this study is to investigate the welfare implications of copayment adjustments on non-prescription drug services and prescription drugs offered through outpatient care in Taiwan’s hospitals under the bundled and 2-tiered pricing scheme. Our empirical results indicate that the WTPs for prescription drugs bundled within an outpatient visit in both medical centers and regional hospitals exceed the new cost-sharing cap of NT$300 for prescription drugs, which came into effect in July 2023. Additionally, the share of DWL per outpatient visit attributed to regular prescription drugs (with a cost-sharing scheme) is significantly lower than the DWL share for refillable prescription drugs (which have no cost-sharing scheme). These results carry substantive policy implications, endorsing the bundled and 2-tiered pricing scheme as an effective mechanism for reducing DWL in hospital outpatient care utilization. Furthermore, the findings suggest the potential extension of this pricing approach to medical tests provided in outpatient settings under Taiwan’s NHI program.

Supplemental Material

sj-docx-1-inq-10.1177_00469580251366876 – Supplemental material for Copayment Adjustments Under Bundled and Two-Tiered Pricing: Welfare Implications for Outpatient Care in Taiwanese Hospitals

Supplemental material, sj-docx-1-inq-10.1177_00469580251366876 for Copayment Adjustments Under Bundled and Two-Tiered Pricing: Welfare Implications for Outpatient Care in Taiwanese Hospitals by Wen-Yi Chen in INQUIRY: The Journal of Health Care Organization, Provision, and Financing

Footnotes

Acknowledgements

I sincerely thank three anonymous reviewers for their insightful comments and constructive suggestions, which have significantly enhanced the quality of this research. I am deeply grateful to Tr. W.W. Liu for her unwavering support throughout the study period. I also wish to acknowledge Dr. Ching-Yuan Chen (Taichung Tzu Chi Hospital, Taiwan) and Lisa Brutcher (Washington State University, USA) for their valuable assistance in addressing the reviewers’ comments and for the final proofreading of the manuscript. All remaining errors are my own.

ORCID iD

Wen-Yi Chen

Ethical Considerations

I used secondary data that does not require taking consent. The data collection process was approved by the Research Ethics Committee of Taichung Tzu Chi Hospital with the Certificate of Exempt Review ID: REC112-16.

Author Contributions

WY Chen contributed the conceptualization, analysis, and writing of this work.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Taiwan National Science and Technology Council under the following projects: “An ex-ante evaluation of the effectiveness of the user-pricing design of the new copayment policy under Taiwan’s National Health Insurance program: Insights from the bundled pricing design of two complementary healthcare services” (Grant No. 112-2410-H-025-011), and “Determinants of employment behavior network connectedness among nursing professionals and the impact of this network connectedness on healthcare quality in Taiwan” (Grant No. 113-2410-H-025-006).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data will be made available on request.

Supplemental Material

Supplemental material for this article is available online.

References

Lin

Chen

Shieh

SH.

Age structural transitions and copayment policy effectiveness: evidence from Taiwan’s national health insurance system. Int J Environ Res Public Health. 2020;17(12):4183. doi:10.3390/ijerph17124183

Chen

Bernell

McMullen

BS.

The new co-payment policy under Taiwan’s National Health Insurance: welfare gain or welfare loss?

Expert Rev Pharmacoecon Outcomes Res. 2008;8(2):141-149. doi:10.1586/14737167.8.2.141

Ministry of Health and Welfare. National Health Insurance Statistics 2021. Taipei: National Health Insurance Administration, Ministry of Health and Welfare; 2022. Accessed September 24, 2024. https://dep.mohw.gov.tw/dos/lp-5103-113-xCat-y110.html

Peng

Chou

YC.

How can Taiwan’s National Health Insurance program be saved?

Common Wealth Magazine. 2024;(794). https://english.cw.com.tw/article/article.action?id=3648

Wong

Guide to Taiwan’s Health Industries. PricewaterhouseCoopers (PwC); 2024. Accessed September 24, 2024. https://www.pwc.tw/en/publications/assets/taiwan-health-industries-guide.pdf

Biernikiewicz

Taieb

Toumi

Characteristics of doctor-shoppers: a systematic literature review. J Mark Access Health Policy. 2019;7(1):1595953. doi:10.1080/20016689.2019.1595953

Lin

Chang

Chen

Hwang

SJ.

Doctor-shopping behaviors among traditional Chinese medicine users in Taiwan. Int J Environ Res Public Health. 2015;12(8):9237-9247. doi:10.3390/ijerph120809237

Chen

WY.

The welfare effect of co-payment adjustments on emergency department visits in medical centers: evidence from Taiwan. Health Policy. 2020;124:1192-1199. doi:10.1016/j.healthpol.2020.03.013

National Health Insurance Administration. New NHI copayment system has been implemented on July 1; NHIA offices around Taiwan mobilized for policy promotion. NHI Policy Announcement. 2024. Accessed September 24, 2024. https://www.nhi.gov.tw/en/cp-1270-b8ac1-8-2.html

10.

National Health Insurance Administration. 2023 Copayment Adjustment Q&A, 2023. Accessed May 2, 2025. https://www.nhi.gov.tw/ch/dl-7540-a99c47012bb5466392ba0b8e97b8c4dc-1.pdf

11.

Bai

Perron

Multiple structural change models: a simulation analysis. In: Corbea

Durlauf

Hansen

, eds. Econometric Theory and Practice: Frontiers of Analysis and Applied Research. Cambridge University Press; 2006;212-237.

12.

Guindon

Stone

Trivedi

Garasia

Khoee

Olaizola

The associations of prescription drug insurance and cost-sharing with drug use, health services use, and health: a systematic review of Canadian studies. Value Health. 2023;26(7):1107-1129. doi:10.1016/j.jval.2023.02.010

13.

Kiil

Houlberg

How does copayment for health care services affect demand, health and redistribution? A systematic review of the empirical evidence from 1990 to 2011. Eur J Health Econ. 2014;15(8):813-828. doi:10.1007/s10198-013-0526-8

14.

Pei

Does coinsurance reduction influence informer-sector workers' and farmers' utilization of outpatient care? A quasi-experimental study in China. BMC Health Serv Res. 2022;22(1):914. doi:10.1186/s12913-022-08301-x

15.

Bittschi

Does the abolition of copayment increase ambulatory care utilization? A quasi-experimental study in Germany. Eur J Health Econ. 2022;23(8):1319-1328. doi:10.1007/s10198-022-01430-4

16.

Kato

Goto

Tsuji

Kondo

The effects of patient cost-sharing on health expenditure and health among older people: heterogeneity across income groups. Eur J Health Econ. 2022;23(5):847-861. doi:10.1007/s10198-021-01399-6

17.

Park

Choi

Who benefits from the fixed copayment of medical and pharmaceutical expenditure among the Korean elderly?

Int J Environ Res Public Health. 2020;17(21):8118. doi:10.3390/ijerph17218118

18.

Jun

Park

Impact of differential copayment on patient healthcare choice: evidence from South Korean National Cohort Study. BMJ Open. 2021;11(6):e044549. doi:10.1136/bmjopen-2020-044549

19.

Gidwani

Yank

Burgette

Kofner

Asch

Wagner

The impact of telehealth cost-sharing on healthcare utilization: evidence from high-deductible health plans. Health Serv Res. 2024;59(6):e14343. doi:10.1111/1475-6773.14343

20.

Law

Cheng

Worthington

, et al Impact of a household-level deductible on prescription drug use among lower-income adults: a quasi-experimental study. CMAJ Open. 2019;7(1):E167-E173. doi:10.9778/cmajo.20180198

21.

Bharadwaj

Gill

Dyjack

, et al Impact of outpatient palliative care services on resource utilization and cost management in a capitated Medicare population. J Palliat Med. Published online March 10, 2025. doi:10.1089/jpm.2024.0325

22.

Ackley

CA.

Tiered cost sharing and health care demand. J Health Econ. 2022;85:102663. doi:10.1016/j.jhealeco.2022.102663

23.

Herr

Suppliet

Tiered co-payments, pricing, and demand in reference price markets for pharmaceuticals. J Health Econ. 2017;56:19-29. doi:10.1016/j.jhealeco.2017.08.008

24.

Cooray

Aida

Watt

, et al Effect of copayment on dental visits: a regression discontinuity analysis. J Dent Res. 2020;99(12):1356-1362. doi:10.1177/0022034520946022

25.

Hamina

Tanskanen

Tiihonen

Taipale

Medication use and health care utilization after a cost-sharing increase in schizophrenia: a nationwide analysis. Med Care. 2020;58(9):763-769. doi:10.1097/MLR.0000000000001369

26.

Ito

Hara

Yoo

Tomio

Kobayashi

Can income-based co-payment rates improve disparity? The case of the choice between brand-name and generic drugs. BMC Health Serv Res. 2019;19(1):780. doi:10.1186/s12913-019-4598-8

27.

Chen

Lin

Yang

CH.

Effect of demographic structural change on copayment policy effectiveness for emergency department visits. Taiwan J Public Health. 2021;40(5):525-544. doi:10.6288/TJPH.202110_40(5).110051

28.

Petrou

Ingleby

Co-payments for emergency department visits: a quasi-experimental study. Public Health. 2019;169:50-58. doi:10.1016/j.puhe.2018.12.014

29.

Lambregts

van Vliet

RCJA

. The impact of copayments on mental healthcare utilization: a natural experiment. Eur J Health Econ. 2018;19(6):775-784. doi:10.1007/s10198-017-0921-7

30.

Di Giacomo

Piacenza

Siciliani

Turati

. The effect of co-payments on the take-up of prenatal tests. J Health Econ. 2022;81:102553. doi:10.1016/j.jhealeco.2021.102553

31.

Haaga

Böckerman

Kortelainen

Tukiainen

Effects of nurse visit copayment on primary care use: do low-income households pay the price?

J Health Econ. 2024;94:102866. doi:10.1016/j.jhealeco.2024.102866

32.

Johansson

Jakobsson

Svensson

Effects of primary care cost-sharing among young adults: varying impact across income groups and gender. Eur J Health Econ. 2019;20(8):1271-1280. doi:10.1007/s10198-019-01095-6

33.

Landsem

Magnussen

The effect of copayments on the utilization of the GP service in Norway. Soc Sci Med. 2018;205:99-106. doi:10.1016/j.socscimed.2018.03.034

34.

Olsen

Melberg

HO.

Did adolescents in Norway respond to the elimination of copayments for general practitioner services?

Health Econ. 2018;27(7):1120-1130. doi:10.1002/hec.3660

35.

Ziebarth

NR.

Assessing the effectiveness of health care cost containment measures: evidence from the market for rehabilitation care. Int J Health Care Finance Econ. 2014;14(1):41-67. doi:10.1007/s10754-013-9138-1

36.

Kim

Yoo

Effects of copayment reduction in long-term care insurance on medical utilization in South Korea. J Aging Soc Policy. 2025;37(3):411-431. doi:10.1080/08959420.2025.2461942

37.

Chen

Schafheutle

Noyce

PR.

The impact of nonreferral outpatient co-payment on medical care utilization and expenditures in Taiwan. Res Soc Adm Pharm. 2009;5(3):211-224. doi:10.1016/j.sapharm.2008.08.003

38.

Juan

Chang

Lee

YC.

Copayment policy reforms and effective care utilization by patients with persistent asthma in Taiwan. Health Policy. 2022;126(2):143-150. doi:10.1016/j.healthpol.2021.11.010

39.

Yang

Tsai

Tien

JJ.

The impacts of persistent behaviour and cost-sharing policy on demand for outpatient visits by the elderly: evidence from Taiwan’s National Health Insurance. Geneva Pap Risk Insur Issues Pract. 2017;42(1):31-52. doi:10.1057/s41288-016-0022-3

40.

Liu

Hsu

Huang

YK.

The determinants of health expenditures in Taiwan: modeling and forecasting using time series analysis. J Stat Manag Syst. 2010;13(3):515-534. doi:10.1080/09720510.2010.10701484

41.

Huang

Tung

CM.

The effects of outpatient co-payment policy on healthcare usage by the elderly in Taiwan. Arch Gerontol Geriatr. 2006;43:101-116. doi:10.1016/j.archger.2005.09.007

42.

Hausman

Exact consumer’s surplus and deadweight loss. Am Econ Rev. 1981;71(4):662-676. https://www.jstor.org/stable/1806188

43.

Perron

Further evidence on breaking trend functions in macroeconomic variables. J Econom. 1997;80:355-385. doi:10.1016/S0304-4076(97)00049-3

44.

Hodrick

Prescott

Postwar

U.S.

Business cycles: an empirical investigation. J Money Credit Bank. 1997;29(1):1-16. doi:10.2307/2953682

45.

Vandenbroucke

von Elm

Altman

, et al Strengthening the reporting of observational studies in Epidemiology (STROBE): explanation and elaboration. Epidemiology. 2007;18(6):805-835. doi:10.1097/EDE.0b013e3181577511

46.

Newey

West

KD.

A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica. 1987;55(3):703-708. doi:10.2307/1913610

47.

Chen

Liang

Lin

YH.

Pricing-out effect from possible user-fee policies under Taiwan's national health insurance. Expert Rev Pharmacoecon Outcomes Res. 2010;10(4):453-464. doi:10.1586/erp.10.43

48.

Kahneman

Tversky

On the interpretation of intuitive probability: a reply to Jonathan Cohen. Cognition. 1979;7(4):409-411. doi:10.1016/0010-0277(79)90024-6.1979.

49.

Lucas

. Econometric policy evaluation: a critique. In: Brunner

Meltzer

, eds. The Phillips Curve and Labor Markets, Carnegie-Rochester Conference Series on Public Policy. Vol 1. Amsterdam: North-Holland; 1976:19-46.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

10.86 MB