https://orcid.org/0000-0003-0636-6911
Figures
Abstract
Background
Many studies have shown that Taiwan’s diabetes pay-for-performance (P4P) program can improve several patients’ clinical outcomes; however, it remains unclear whether P4P initiatives yield effects on health-related quality of life (HRQoL) for all diabetes patients as well as those with comorbidities, especially comorbid gastrointestinal conditions.
Purpose
In this study, we evaluated the effects of the diabetes P4P program on QoL among all patients as well as the effects of this program on QoL among patients with comorbid gastrointestinal conditions.
Methods
Data were collected across 6 hospitals. We used the five-level version of the EuroQol-5 Dimension (EQ-5D-5L) to obtain scores for health disutility. The scores for these measures were subsequently predicted via a generalized linear model (GLM) with a gamma distribution. We also applied propensity score weighting (PSW) to adjust for potential selection bias.
Results
After PSW, P4P program participation was found to be significantly correlated with lower EQ-5D disutility scores among all patients (parameter = −0.169, [−0.325, −0.012], P = 0.034). Additionally, P4P program participation was significantly correlated with a lower EQ-5D disutility score among patients with comorbid gastrointestinal conditions (parameter = −0.604, [−0.891, −0.317], P < 0.001).
Conclusion
Our findings suggest that participation in the diabetes P4P program is associated with improved HRQoL among all patients with type 2 diabetes, including those with gastrointestinal comorbidities. A P4P model incorporating continuing education, coordinated care, and regular follow-up may be particularly beneficial in enhancing HRQoL for patients with gastrointestinal complications.
Citation: Kuang S-H, Chen T-T, Ma W-Y, Chen S-T, Shih M-H, Su C-C, et al. (2025) Is the pay-for-performance program associated with better quality of life among type 2 diabetes patients, including those with gastrointestinal conditions, in Taiwan? A cross-sectional survey. PLoS One 20(8): e0328910. https://doi.org/10.1371/journal.pone.0328910
Editor: Alexandre Morais Nunes, University of Lisbon, Institute of Social and Political Sciences, PORTUGAL
Received: October 20, 2024; Accepted: July 8, 2025; Published: August 22, 2025
Copyright: © 2025 Kuang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared beyond individual quotations included in the paper because participants did not consent to the publication of their full responses. The data contain sensitive patient information. This restriction was approved by the ethics committee of Taipei City Hospital. For data access requests, please contact Miss Chih, the administrative coordinator at the Department of Public Health, Fu Jen Catholic University: fjuph316@gmail.com.
Funding: Ministry of Science and Technology [NSC: MOST 105-2410-H- 030-057] and the Cardinal Tien Hospital [CTH-103-1-2B05] in Taiwan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The prevalence of diabetes in Taiwan was approximately 13.1% in 2021 [1]. Patients with diabetes are more likely to have comorbidities, such as gastrointestinal conditions [2]. For example, patients with diabetes had a significantly higher prevalence of hepatitis C infection than hepatitis B infection [3]. The prevalence rate of hepatitis C infection among patients with diabetes was 6.1% in Taiwan, which is higher than the global prevalence rate of 4.2% [4]. Moreover, patients with diabetes and hepatitis C infections may have a greater risk of developing cirrhosis [5], which can lead to the development of hepatocellular carcinoma [6], increased mortality [4], and comorbidities that negatively impact patients’ quality of life (QoL) [7,8]. For example, diabetes patients with hepatitis B may exhibit poor glycemic control and poor QoL [9]. Regarding QoL, during this era of patient-centered care, the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and the American Diabetes Association (ADA) are currently highlighting the importance of patient-reported outcome (PRO) research, including QoL research, among patients with type 2 diabetes [10]. QoL is a multidimensional, subjective and dynamic measure that is used to evaluate what patients want from life, understand their perceived health status and assess how the disease affects their lives [11]. Furthermore, QoL measures can be used to detect individual psychosocial problems that may impact a patient’s therapeutic response [12]. Therefore, understanding whether interventions can improve QoL is important.
Many studies have shown that Taiwan’s diabetes pay-for-performance (P4P) program can improve common patients’ different clinical outcomes [13–17], and P4P in diabetes management has been shown to be cost-effective [16,17,18]. However, these studies have not shown that participants in P4P programs experience significantly better quality of life. Some studies have explored the benefits of P4P programs for patients with diabetes and multiple chronic conditions, but they have not specifically examined the outcomes for diabetic patients with comorbid gastrointestinal conditions [19,20]. Therefore, there is a lack of clarity regarding whether P4P initiatives affect the QoL of patients with comorbidities, especially those with gastrointestinal comorbidities. In this study, we evaluated the effect of the diabetes P4P program on QoL among all type 2 diabetes patients as well as the effects of this program on QoL among patients with gastrointestinal comorbidities.
Methods
Study design
In this cross-sectional survey study, data were collected from patient surveys and record reviews across 6 hospitals, including two tertiary hospitals and four regional hospitals in northern Taiwan, from October to December 2016.
Setting
The National Health Insurance Administration (NHIA) integrated external incentives with a shared care network (SCN) to form the diabetes P4P program in 2001, which not only enforces the execution of suggested activities by the SCN but also suggests adherence to guidelines for physicians conducting the necessary examinations. The incentives (US $108) include three follow-up fees (total US $21), a one-time yearly evaluation fee (US $27), and physician fees that are paid four times, once for every patient visit except for visits to stand-alone clinics (total US $60) [21]. Physicians have the autonomy to decide whether to enroll treated patients with diabetes in the P4P program.
Additionally, the diabetes P4P program in Taiwan has many distinctive features of the chronic care model (CCM), which could result in better patient outcomes [21], such as coordination between diabetes-related professionals, autonomy, and the assembly of family resources. The diabetes P4P program in Taiwan has embedded this CCM-based disease management program (DMP) to lead to better patient outcomes than the hepatitis P4P program without an embedded DMP [21,22]. The P4P program-enrolled patients can be treated by teams consisting of trained professionals, including providers (specialists in endocrinology, family medicine or internal medicine), health educators, dieticians, and laboratory technicians. The teams motivate and support patients’ autonomy by encouraging healthy behavior depending on the patient’s lifestyle and preferences [21,23]. Detailed descriptions of the diabetes P4P program in Taiwan can be found in a previous article [21].
Participants
The inclusion criteria were as follows: (1) had a diagnosis of diabetes (ICD-9-CM 250); (2) were 18 years or older; and (3) had sufficient mental capacity to respond to the complete questionnaire. In this study, 1512 questionnaires were distributed after informed consent was obtained, and 1512 valid questionnaires were returned. The final analytical sample consisted of 1512 individuals. We excluded 30 subjects because of missing information regarding income and the EQ-5D index. Therefore, the final sample included 1,482 individuals. We further stratified these patients based on whether they participated in a P4P program during the year of enrollment in this study. Patients in the P4P group were also required to be enrolled for at least 1 year after the first enrollment day. The remaining patients not included in the P4P group were categorized into the control group (non-P4P group).
Variables and data sources/measurement
We accounted for common patient demographic characteristics and diabetes-related variables. The former demographic variables include age, sex, income, employment, marital status, education, and urbanization of the residence area. The communities in Taiwan are stratified into seven urbanization categories according to the Taiwanese National Health Research Institute (NHRI). Cluster analysis was used to divide these areas into urbanization groups from the highest level of urbanization to the lowest level as follows: high-level, median-level, emerging, common, aging, agricultural, and remote areas [24]. The latter diabetes-related variables included treatment, multiple shots of insulin per day (yes or no), any complications [25], and previous hospital admission within one year. The treatments included 4 categories: exercise and diet control, oral medications only, oral medications and use of insulin, and use of insulin only.
We assessed gastrointestinal conditions by using the chronic illness with complexity (CIC) index, which assesses peptic ulcers, inflammatory bowel disease, diverticulitis, chronic hepatitis, gallbladder disease and gallstones [2]. Additionally, we used the EQ-5D 5 L to obtain health utility scores; the EQ-5D 5 L is suitable for utility measurements for diabetes patients [26]. The scores of all 5 dimensions of the EQ-5D index were combined with preference weights derived from the Taiwanese value sets [27].
The common patient demographic characteristics, patient complexity (gastrointestinal conditions and complications) and EQ-5D 5 L scores were derived mainly from the patients’ questionnaires. To ensure consistency among the interviewers in this study, we have interviewer training sessions, which were held prior to the commencement of the interviews. During these sessions, each question and its corresponding response options were read aloud. We trained the interviewers in important details of the EQ-5D questionnaire, such as the rationale behind it and how the EQ-5D questionnaire should be administered. Patient complexity was also clarified and explained in the training sessions. To obtain complete information about gastrointestinal conditions for patients with diabetes, we provided a list of the diseases described above to allow patients to select whether they had at least one of the gastrointestinal conditions, and we asked patients to verify that the reported diseases or conditions had ever been diagnosed by physicians. After the completion of the questionnaire, the interviewers verified the gastrointestinal conditions that the patients provided by asking patients the same questions again. If the respondents could not read or were unfamiliar with the questionnaire, we asked the interviewers to read the entire questionnaire in a manner that the interviewees could understand. After completing the questionnaire, we extracted each patient’s P4P status from the medical records.
Statistical methods
The variance inflation factor (VIF) was calculated to determined whether there was multicollinearity between independent variables [28]. Regarding the skewness and ceiling effect of the EQ-5D index, in general, the utility values are expected to be skewed, as many patients report perfect health [29,30]. To address these problems, we first conducted a Kolmogorov–Smirnov test to examine normality. Then, we used a two-part model, which was estimated using the STATA twopm command, to adjust the skewness and ceiling effects of the EQ-5D index. In general, the two parts of the model were 1) a logistic regression and 2) a generalized linear model (GLM) using backward elimination. The outcome was the disutility score. The logistic regression modeled the probability of disutility (1-EQ-5D index) for transformation of the probability of disutility because too many respondents reported a perfect EQ-5D index (Pr (y > 0 | x)). The second part of the model utilized a GLM with a gamma distribution and a log link function under the condition of probability of disutility [22,29,31]. The independent variables included common patient characteristics and diabetes-related variables, as described in the variables section.
To prevent selection bias in the P4P group, we applied propensity score weighting (PSW) with inverse probability weighting (IPW) to appropriately weight patients in the P4P intervention and control groups [32], and the resulting propensity score was used as a control variable in the two-part model. This research used the aforementioned characteristics of patients to generate inverse probability weighting (propensity score) for each patient via logistic regression. Compared with the matching method, weighting the propensity score has the advantage of incorporating most of the analytical observations and can increase precision in estimating treatment effects [33]. IPW involves weighting by the inverse probability of receiving the study intervention (1/propensity score for the P4P group and 1/(1 − propensity score) for the non-P4P group) [32]. All analyses were conducted via Stata Statistical Software version 15. The study was approved by the Institutional Review Board (TCHIRB-104–12123-E) of the Taipei City Hospital.
Results
Table 1 shows that the average age of the patients was 63 years. 45% of the patients were P4P program patients; approximately 54% were male; 64% had an income less than US$1,000 per month; 28% had a full-time job; 77% were married; 62% were religious; 53% had an education level below junior high school; 69% lived in an area with high-level urbanization and took oral medication only(72%) without the use of insulin; 10% took multiple shots of insulin per day; 19% had gastrointestinal conditions; 30% had any complications; and 4% had previous hospital admissions within one year due to diabetes. Additionally, there were significant differences in employment, marital status, level of urbanization, and treatment method for diabetes between P4P patients and non-P4P program patients. The patients in the P4P group had a significantly greater average EQ-5D index (0.935) than did the patients in the control group, who had an average EQ-5D index of 0.90 (P < 0.001) (data not shown).
As shown in S1 Table of the supporting information, after applying PSW, the absolute values of the standardized mean differences (SMDs) between the P4P and non-P4P groups significantly decreased for every variable and its related categories. The largest of these differences is 0.022 in absolute value, which is much lower than the upper limit of 0.25 recommended by another study.
Table 2 shows the distribution of EQ-5D-5L responses of diabetes patients and compares the results on the basis of the status of P4P program participation. Compared with non-P4P group patients, patients in the P4P group presented higher percentages of level 1 in all five domains. The differences between the two groups were statistically significant in two domains: usual activity (P = 0.03) and anxiety/depression (P = 0.001). The distribution of EQ-5D-5L responses among diabetes patients and a comparison of the results between patients with and without gastrointestinal comorbidities are shown in Table 3. Compared with patients without gastrointestinal comorbidities, patients with gastrointestinal comorbidities presented lower percentages of level 1 in all five domains. The differences between the two groups were statistically significant in all domains.
Table 4 shows the results of the analysis of the disutility EQ-5D index via a two-part model among all patients and those with gastrointestinal comorbidities after PSW adjustment. P4P program participation was significantly correlated with lower EQ-5D disutility scores among all patients (parameter = −0.169, [−0.325, −0.012], P = 0.034). With respect to patient demographic variables, patients with higher education had lower disutility scores for all patients (university, parameter = −0.202, [−0.378, −0.025], P = 0.025; reference: below elementary school). Patients with jobs had lower disutility scores in the total cohort (full-time without shift, parameter = −0.396, [−0.548, −0.244], P < 0.001; full-time with shift, parameter = −0.399, [−0.700, −0.098], P = 0.009; self-employed, parameter = −0.296, [−0.533, −0.060], P = 0.014; reference: others [loss of job or housekeeper]). With respect to diabetes-related variables, patients’ treatment with exercise and diet control had lower disutility scores (parameter = −0.310, [−0.563, −0.058], P = 0.016; reference: oral medications only).
Similarly, P4P program participation was also significantly correlated with lower EQ-5D disutility scores among patients with gastrointestinal comorbidities (parameter = −0.604, [−0.891, −0.317], P < 0.001). With respect to patient demographic variables, higher-income patients had lower EQ-5D disutility scores in the gastrointestinal comorbidity cohort (income U.S.$2,001-$3,333, parameter = −0.777, [−1.055, −0.498], P < 0.001; reference: income < U.S.$1,000). With respect to diabetes-related variables, patients with multiple shots of insulin per day had higher disutility scores (parameter = 0.810, [0.256, 1.364], P = 0.00; reference: no or a single shot of insulin).
Discussion
Previous studies have demonstrated that diabetes P4P programs could be associated with better clinical outcomes. However, no studies have proven that diabetes P4P programs could also be associated with better QoL among patients with gastrointestinal comorbidities. We not only detected positive effects of the diabetes P4P program on QoL among all patients with diabetes but also observed significant positive effects of the P4P program on QoL among patients with gastrointestinal comorbidities.
Previous studies on the quality of life (QoL) of patients with diabetes have shown that factors such as employment [34], higher income [35], and higher educational attainment [36] were positively associated with EQ-5D index values. Consistent with these findings, we observed similar patterns in our analysis. In addition, patients who managed diabetes through exercise and diet control, rather than oral medications, and those receiving only a single daily insulin injection, as opposed to multiple injections, reported better QoL. These differences likely reflect variations in disease severity, suggesting that individuals with milder form of diabetes tend to experience higher quality of life [34,37–39]. Additionally, previous research using the EQ-5D index has shown that the average EQ-5D-5L index value ranges from 0.77 to 0.95 for patients with diabetes [34,40–42]. In our study, the average EQ-5D-5L score was 0.92, which is within the range reported in a previous local study [17]. Finally, the diabetes P4P program in Taiwan was associated with better QoL among all patients and those with gastrointestinal comorbidities. The results were robust because we controlled for the confounders above, and PSW was applied to prevent selection bias.
The probable reasons for the P4P program having spillover effects on a better EQ-5D index among patients with gastrointestinal comorbidities. The diabetes P4P program in Taiwan not only has an incentive mechanism for adherence to physician guidelines but also integrates a shared care network (SCN) [21] that has distinctive characteristics, such as continuing education, coordinated care and regular follow-up, thereby improving clinical outcomes in diabetic patients with gastrointestinal comorbidities. First, in Taiwan, participation in the diabetes P4P program requires that health care professionals undergo regular continuing education. Although hepatitis HCV-specific training has become more prominent in recent years, particularly after the Ministry of Health and Welfare (MoHW) introduced the National Hepatitis C Policy Guideline (2018–2025) [43], it is likely that even before 2017, general education on managing comorbidities may have increased health care professionals’ awareness of liver-related complications in patients with diabetes. Second, the P4P program was developed on the basis of the principles of the chronic care model (CCM) [21], in which one of the core components is effective coordination among medical specialties. In addition to common interdisciplinary collaborations such as those with ophthalmology or cardiology [44], some health care institutions have initiated integrative care pathways between endocrinology and gastroenterology for patients with diabetes [45]. Systematic reviews have shown that such integrative care models can significantly enhance patients’ quality of life [46]. Third, more health care institutions may be unable to implement integrated care with gastroenterology departments due to various institutional constraints [47]. However, patients enrolled in the P4P program typically attend regular follow-up visits [48], which enables physicians involved in diabetes management—such as endocrinologists or family medicine specialists—to become more aware of gastrointestinal issues. Consequently, these patients are more likely to be proactively referred to gastroenterology for further evaluation [2,49]. Alternatively, patients may be referred by diabetes care physicians to health educators for counseling on lifestyle modifications, such as interventions addressing alcohol cessation (i.e., alcohol consumption leading to severe hepatitis [6,35]), which is associated with better health-related QoL among patients with diabetes [50].
There were several limitations in our study. First, we could not control for all of the factors related to QoL scores; however, we controlled for possible and important socioeconomic and clinical factors and applied those factors in the PSW analysis to prevent selection bias. Second, there were some concerns about the low rate of reports of gastrointestinal conditions. To address this problem, we asked the interviewers to pay attention to the results reported by the patients. However, we cannot completely rule out the possibility of underreporting gastrointestinal conditions. In our study, the prevalence of gastrointestinal conditions was 19% (284/1477), which was slightly higher than that reported in other studies conducted in Asia if we considered only HBV infection in individuals with type 2 diabetes mellitus [51]. We tried to use the NHIA’s national claims data to verify the results, and we first reported that in 2013, the average age of patients with diabetes from the national database was 62.51 years (data not shown), which is within the range of our sample. The sex distribution was also identical (data not shown). Additionally, the prevalence rate of gastrointestinal conditions from the CIC index from the national database was approximately 24% (data not shown), which indicates that the lower prevalence rate (19%) of gastrointestinal conditions in our study was not severe, thus supporting the findings that the P4P program has more benefits for patients with gastrointestinal comorbidities. In the future, a link to the database of the NHIA should be established to verify patients with gastrointestinal comorbidities. Third, QoL was assessed in a cross-sectional manner, thus decreasing the generalizability of the results. In the future, patients should be followed up to evaluate their subsequent quality of life, and a quasi-experimental design incorporating preceding QoL outcomes should be used. Fourth, we cannot rule out the possibility that patients referred by physicians or health care personnel to conduct questionnaires may have certain characteristics that incur bias. However, after adjustment through PSW, the P4P and non-P4P groups categorized from all patients were comparable. Therefore, the bias introduced by referrals can be mitigated. Fifth, our data cover the year 2016, and although this may raise concerns about data currency, we believe the findings remain relevant. The definitions and implementation of the diabetes P4P program may have evolved since its initiation in 2001, which could affect the interpretation of our results. Prior to 2012, the diabetes P4P proposal underwent nine revisions, characterized by significant reforms such as the 1) inclusion of outcome-based incentives, 2) the implementation of pay-for-improvement mechanisms, and 3) initially, introduced as a pilot program, the diabetes P4P was formally integrated into the official payment system. From 2012 to the end of our data period in 2016, as well as after 2016, there were no significant changes in the diabetes P4P program. In recent years, the NHIA has merged with the early chronic kidney disease (CKD) P4P [52]. Consequently, this new initiative may benefit diabetic patients with CKD complications but potentially has a limited effect on gastrointestinal conditions. In summary, although the data in this study may be old, the conclusions remain potentially applicable. Sixth, this study employs a two-part model. Under the conditions of α = 0.05, power = 0.8, and an effect size of 0.1 (considering both the binomial and normal parts), the minimum sample size required is 96 [53]. In this study, there were 249 diabetic patients with gastrointestinal conditions. Therefore, the sample size in this study may be sufficient. However, since the actual sample size is not large, caution should be exercised when interpreting the results of this study.
Conclusions
Our findings suggest that participation in the diabetes P4P program is associated with improved health-related quality of life (QoL) among all patients with type 2 diabetes, including those with gastrointestinal comorbidities. A P4P model incorporating continuing education, coordinated care, and regular follow-up may be particularly beneficial in enhancing QoL for patients with gastrointestinal complications.
Supporting information
S1 Table. Standardized mean difference and percent reduction before/after propensity score weighting for every variable and its categories.
https://doi.org/10.1371/journal.pone.0328910.s001
(DOCX)
Acknowledgments
The authors would like to thank Miss Hsin-Yu Pan of New Taipei City Hospital for her valuable support in participant recruitment and data collection throughout the study.
References
- 1.
International Diabetes Federation. Taiwan information. Available from: https://idf.org/our-network/regions-and-members/western-pacific/members/taiwan/.
- 2. Meduru P, Helmer D, Rajan M, Tseng C-L, Pogach L, Sambamoorthi U. Chronic illness with complexity: implications for performance measurement of optimal glycemic control. J Gen Intern Med. 2007;22 Suppl 3(Suppl 3):408–18. pmid:18026810
- 3. White DL, Ratziu V, El-Serag HB. Hepatitis C infection and risk of diabetes: a systematic review and meta-analysis. J Hepatol. 2008;49(5):831–44. pmid:18814931
- 4. Yu M-L, Wang C-Y, Lee M-H, Ou H-Y, Cheng P-N, Tu S-T, et al. TASL, TADE, and DAROC consensus for the screening and management of hepatitis C in patients with diabetes. J Formos Med Assoc. 2023;122(3):202–20. pmid:36750398
- 5. Memon MS, Arain ZI, Naz F, Zaki M, Kumar S, Burney AA. Prevalence of type 2 diabetes mellitus in hepatitis C virus infected population: a Southeast Asian study. J Diabetes Res. 2013;2013:539361. pmid:23984431
- 6. Yu M-L, Chen P-J, Dai C-Y, Hu T-H, Huang C-F, Huang Y-H, et al. 2020 Taiwan consensus statement on the management of hepatitis C: part (I) general population. J Formos Med Assoc. 2020;119(6):1019–40. pmid:32359879
- 7. Adriaanse MC, Drewes HW, van der Heide I, Struijs JN, Baan CA. The impact of comorbid chronic conditions on quality of life in type 2 diabetes patients. Qual Life Res. 2016;25(1):175–82. pmid:26267523
- 8. Nguyen HV, Tran TT, Nguyen CT, Tran TH, Tran BX, Latkin CA, et al. Impact of Comorbid Chronic Conditions to Quality of Life among Elderly Patients with Diabetes Mellitus in Vietnam. Int J Environ Res Public Health. 2019;16(4):531. pmid:30781767
- 9. Testa MA, Simonson DC, Turner RR. Valuing quality of life and improvements in glycemic control in people with type 2 diabetes. Diabetes Care. 1998;21 Suppl 3:C44–52. pmid:9850489
- 10. Marrero DG, Hilliard ME, Maahs DM, McAuliffe-Fogarty AH, Hunter CM. Using patient reported outcomes in diabetes research and practice: Recommendations from a national workshop. Diabetes Res Clin Pract. 2019;153:23–9. pmid:31128133
- 11. Speight J, Holmes-Truscott E, Hendrieckx C, Skovlund S, Cooke D. Assessing the impact of diabetes on quality of life: what have the past 25 years taught us? Diabet Med. 2020;37(3):483–92. pmid:31797443
- 12. Ligda G, Ploubidis D, Foteli S, Kontou PI, Nikolaou C, Tentolouris N. Quality of life in subjects with type 2 diabetes mellitus with diabetic retinopathy: A case-control study. Diabetes Metab Syndr. 2019;13(2):947–52. pmid:31336549
- 13. Kornelius E, Chiou J-Y, Yang Y-S, Lu Y-L, Peng C-H, Huang C-N. The Diabetes Shared Care Program and Risks of Cardiovascular Events in Type 2 Diabetes. Am J Med. 2015;128(9):977-85.e3. pmid:25908394
- 14. Lin T-Y, Chen C-Y, Huang YT, Ting M-K, Huang J-C, Hsu K-H. The effectiveness of a pay for performance program on diabetes care in Taiwan: A nationwide population-based longitudinal study. Health Policy. 2016;120(11):1313–21. pmid:27780591
- 15. Hsieh H-M, Lin T-H, Lee I-C, Huang C-J, Shin S-J, Chiu H-C. The association between participation in a pay-for-performance program and macrovascular complications in patients with type 2 diabetes in Taiwan: A nationwide population-based cohort study. Prev Med. 2016;85:53–9. pmid:26740347
- 16. Tan EC-H, Pwu R-F, Chen D-R, Yang M-C. Is a diabetes pay-for-performance program cost-effective under the National Health Insurance in Taiwan?. Qual Life Res. 2014;23(2):687–96. pmid:23975377
- 17. Hsieh H-M, Tsai S-L, Shin S-J, Mau L-W, Chiu H-C. Cost-effectiveness of diabetes pay-for-performance incentive designs. Med Care. 2015;53(2):106–15. pmid:25397966
- 18. Hsieh H-M, Gu S-M, Shin S-J, Kao H-Y, Lin Y-C, Chiu H-C. Cost-Effectiveness of a Diabetes Pay-For-Performance Program in Diabetes Patients with Multiple Chronic Conditions. PLoS One. 2015;10(7):e0133163. pmid:26173086
- 19. Chen C-C, Cheng S-H. Does pay-for-performance benefit patients with multiple chronic conditions? Evidence from a universal coverage health care system. Health Policy Plan. 2016;31(1):83–90. pmid:25944704
- 20. Huang Y-C, Lee M-C, Chou Y-J, Huang N. Disease-specific Pay-for-Performance Programs: Do the P4P Effects Differ Between Diabetic Patients With and Without Multiple Chronic Conditions?. Med Care. 2016;54(11):977–83. pmid:27547944
- 21. Chen T-T, Oldenburg B, Hsueh Y-S. Chronic care model in the diabetes pay-for-performance program in Taiwan: Benefits, challenges and future directions. World J Diabetes. 2021;12(5):578–89. pmid:33995846
- 22. Su W-C, Chen T-T, Yang S-S, Shih L-N, Liu C-K, Wang C-C, et al. The effect of a pay-for-performance program on health-related quality of life for patients with hepatitis in Taiwan. Health Qual Life Outcomes. 2022;20(1):130. pmid:36064530
- 23. Tien K-J, Hung H-C, Hsiao J-Y, Hsu S-C, Hsin S-C, Shin S-J, et al. Effectiveness of comprehensive diabetes care program in Taiwanese with type 2 diabetes. Diabetes Res Clin Pract. 2008;79(2):276–83. pmid:17904243
- 24. Chang W-P, Wu S-J, Chang W-C, Kuo H-C. Population-based study of the association between urbanization and Kawasaki disease in Taiwan. Sci World J. 2013;2013:169365. pmid:23864819
- 25. Young BA, Lin E, Von Korff M, Simon G, Ciechanowski P, Ludman EJ, et al. Diabetes complications severity index and risk of mortality, hospitalization, and healthcare utilization. Am J Manag Care. 2008;14(1):15–23. pmid:18197741
- 26. Clarke PM, Hayes AJ, Glasziou PG, Scott R, Simes J, Keech AC. Using the EQ-5D index score as a predictor of outcomes in patients with type 2 diabetes. Med Care. 2009;47(1):61–8. pmid:19106732
- 27. Lin H-W, Li C-I, Lin F-J, Chang J-Y, Gau C-S, Luo N, et al. Valuation of the EQ-5D-5L in Taiwan. PLoS One. 2018;13(12):e0209344. pmid:30586400
- 28. Kim JH. Multicollinearity and misleading statistical results. Korean J Anesthesiol. 2019;72(6):558–69. pmid:31304696
- 29. Jalkanen K, Aarnio E, Lavikainen P, Jauhonen H-M, Enlund H, Martikainen J. Impact of type 2 diabetes treated with non-insulin medication and number of diabetes-coexisting diseases on EQ-5D-5 L index scores in the Finnish population. Health Qual Life Outcomes. 2019;17(1):117. pmid:31286982
- 30. Ong SC, Mak B, Aung MO, Li S-C, Lim S-G. Health-related quality of life in chronic hepatitis B patients. Hepatology. 2008;47(4):1108–17. pmid:18318043
- 31. Shafie AA, Chhabra IK, Wong JHY, Mohammed NS. EQ-5D-3L health state utility values in transfusion-dependent thalassemia patients in Malaysia: a cross-sectional assessment. Health Qual Life Outcomes. 2021;19(1):10. pmid:33413416
- 32.
Guo S, Fraser MW. Propensity score analysis: statistical methods and applications. Thousand Oaks, Calif.: Sage Publications. 2010.
- 33. Desai RJ, Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: a primer for practitioners. BMJ. 2019;367:l5657. pmid:31645336
- 34. Abedini MR, Bijari B, Miri Z, Shakhs Emampour F, Abbasi A. The quality of life of the patients with diabetes type 2 using EQ-5D-5 L in Birjand. Health Qual Life Outcomes. 2020;18(1):18. pmid:32000785
- 35. Tamornpark R, Utsaha S, Apidechkul T, Panklang D, Yeemard F, Srichan P. Quality of life and factors associated with a good quality of life among diabetes mellitus patients in northern Thailand. Health Qual Life Outcomes. 2022;20(1):81. pmid:35596219
- 36. Prabowo MH, Febrinasari RP, Pamungkasari EP, Mahendradhata Y, Pulkki-Brännström A-M, Probandari A. Health-related Quality of Life of Patients With Diabetes Mellitus Measured With the Bahasa Indonesia Version of EQ-5D in Primary Care Settings in Indonesia. J Prev Med Public Health. 2023;56(5):467–74. pmid:37828874
- 37. Meraya AM, Alwhaibi M. Health related quality of life and healthcare utilization among adults with diabetes and kidney and eye complications in the United States. Health Qual Life Outcomes. 2020;18(1):85. pmid:32228619
- 38. Trikkalinou A, Papazafiropoulou AK, Melidonis A. Type 2 diabetes and quality of life. World J Diabetes. 2017;8(4):120–9. pmid:28465788
- 39. Shim YT, Lee J, Toh MPHS, Tang WE, Ko Y. Health-related quality of life and glycaemic control in patients with Type 2 diabetes mellitus in Singapore. Diabet Med. 2012;29(8):e241-8. pmid:22507291
- 40. Arifin B, Idrus LR, van Asselt ADI, Purba FD, Perwitasari DA, Thobari JA, et al. Health-related quality of life in Indonesian type 2 diabetes mellitus outpatients measured with the Bahasa version of EQ-5D. Qual Life Res. 2019;28(5):1179–90. pmid:30649698
- 41. Wong ELY, Xu RH, Cheung AWL. Measurement of health-related quality of life in patients with diabetes mellitus using EQ-5D-5L in Hong Kong, China. Qual Life Res. 2020;29(7):1913–21. pmid:32140920
- 42. Gebremariam GT, Biratu S, Alemayehu M, Welie AG, Beyene K, Sander B, et al. Health-related quality of life of patients with type 2 diabetes mellitus at a tertiary care hospital in Ethiopia. PLoS One. 2022;17(2):e0264199. pmid:35180266
- 43.
Ministry of Health and Welfare T. National Hepatitis C Policy Guideline 2018–2025. Taipei: Ministry of Health and Welfare. 2019. https://gpi.culture.tw/books/1010800763
- 44. Chiou ST, Lin HD, Yu NC, Hseuh HK, Lin LH, Lin LT, et al. An initial assessment of the feasibility and effectiveness of implementing diabetes shared care system in Taiwan--some experiences from I-Lan County. Diabetes Res Clin Pract. 2001;54 Suppl 1:S67–73. pmid:11580971
- 45.
Health Promotion Administration M of H and W. Guidelines for the Diabetes Shared Care Program. Taipei: Health Promotion Administration, Ministry of Health and Welfare. 2017.
- 46. Wang Y, Hu M, Zhu D, Ding R, He P. Effectiveness of Collaborative Care for Depression and HbA1c in Patients with Depression and Diabetes: A Systematic Review and Meta-Analysis. Int J Integr Care. 2022;22(3):12. pmid:36117873
- 47. Chi C, Liang L-L, Lee J-L, Núñez A, Tuepker A. Healthcare for all - A critical review of Taiwan’s national health insurance system through a social institution lens. J Formos Med Assoc. 2024;123 Suppl 3:S228–33. pmid:38972766
- 48. Yen S-M, Kung P-T, Sheen Y-J, Chiu L-T, Xu X-C, Tsai W-C. Factors related to continuing care and interruption of P4P program participation in patients with diabetes. Am J Manag Care. 2016;22(1):e18–30. pmid:26799201
- 49. Bae S, Rosenthal MB. Patients with multiple chronic conditions do not receive lower quality of preventive care. J Gen Intern Med. 2008;23(12):1933–9. pmid:18810557
- 50. Hsieh P-L, Yang F-C, Hu Y-F, Chiu Y-W, Chao S-Y, Pai H-C, et al. Continuity of Care and the Quality of Life among Patients with Type 2 Diabetes Mellitus: A Cross-Sectional Study in Taiwan. Healthcare (Basel). 2020;8(4):486. pmid:33202699
- 51. Lu J, Hou X, Tu H, Tang Z, Xiang Y, Bao Y, et al. Chronic hepatitis B virus infection status is more prevalent in patients with type 2 diabetes. J Diabetes Investig. 2017;8(4):619–25. pmid:27930871
- 52.
National Health Insurance Administration NH. Integration program of diabetes and early chronic kidney disease. Available from: https://www.nhi.gov.tw/ch/cp-14177-adf81-3576-1.html.
- 53. Lachenbruch PA. Power and sample size requirements for two-part models. Stat Med. 2001;20(8):1235–8. pmid:11304738