The study investigated to compare health care utilization and expenditures between diabetic patients with and without depression in Taiwan.
Health care utilization and expenditure among diabetic patients with and without depression disorder during 2000 and 2004 were examined using Taiwan's population-based National Health Insurance claims database. Health care utilization included outpatient visits and the use of inpatient services, and health expenditures were outpatient, inpatient, and total medical expenditures. Moreover, general estimation equation models were used for analyzing the factors associated with outpatient visits and expenditures. Multiple logistic regression analysis was applied for identifying the factors associated with hospitalization.
The average annual outpatient visits and annual total medical expenditures in the study period were 44.23–52.20; NT$87,496–133,077 and 30.75–32.92; NT$64,411–80,955 for diabetic patients with and without depression. After adjustment for covariates, our results revealed that gender and complication were associated with out-patient visits. Moreover, the time factor was associated with the total medical expenditure, and residential urbanization and complication factors were associated with hospitalization.
Health care utilization and expenditures for diabetic patients with depression were significantly higher than those without depression. Sex, complications, time, and urbanization are the factors associated with health care utilization and expenditures.
Diabetes mellitus, a chronic and disabling disease, majorly contributes to disability-adjusted life years.
Comorbid mental and physical illness, often categorized as “chronic illness with complexity (CIC),” is a new and emerging research area. CIC is defined as multiple chronic conditions occurring simultaneously regardless of the causal pathways and associations.
Many studies have discussed the health care utilization and expenditures of patients with diabetes comorbid with physical complications.
Taiwan's National Health Insurance (NHI) program is a mandatory, single-payer system established in 1995 under the NHI Administration (NHIA); approximately 98% of all Taiwanese residents are enrolled in the program, and almost all medical care providers in Taiwan, including those employed at medical and primary care centers, are contracted by the NHIA to provide outpatient and inpatient services. Through a fee-for-service payment system, all health care providers fill monthly service claims to the NHI to get reimbursement for their medical fees. These claims include inpatient, ambulatory, and home care, and information such as patient demographics (age and sex), clinical details [disease and procedure codes based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)], and health care utilization and expenditure (hospital days, drug use, and costs).
Our study was conducted retrospectively using a study cohort drawing from a random sample in the National Health Research Institutes (NHRI) database from 2000 to 2004. The cohort was randomly selected from the population that enrolled in the NHI in 2000 (n=23,735,407). Excluding those from the uninsured population, a sample group of 200,432 patients represented 1% of the total NHI database. No significant differences were observed in age, sex, or the average insured payroll-related amount between the random sample and whole population. Although the administrative claims database has limitations regarding the accuracy of diagnostic coding,
Patients with at least two service claims for diabetes (ICD-9-CM 250) for ambulatory care or with one service claim for inpatient care between 2000 and 2004 were defined as having diabetes.
Health care utilization included outpatient visits and hospital admission per person per year. Outpatients included those who visited physician clinics and hospital outpatient departments. Hospital admission was defined as admission to a general or psychiatric hospital.
Healthcare expenditure included outpatient, inpatient, and total medical expenditures per person per year. Outpatient expenditure included payments for office-based provider visits and hospital outpatient visits. Outpatient health expenditures consisted of physician fee, medication, laboratory exams, and others. Inpatient expenditures included 14 items, including medication, room, procedure, imaging, and others.
Patient demographics characteristics included age, sex, urbanization, and income, whereas disease characteristics included types of complications and diabetes. Patient ages were calculated as the year difference of mid-year (June 30) and patient birthdays. Age was categorized as <45, 45–64, and ≥65 years, and urbanization was categorized as urban, suburban, and rural.
Student's t test was used for comparing the differences of health care utilization and expenditures between patients with diabetes comorbid with and without depression each year. After adjustment for time, we primarily identified the NHI 2000 cohort and followed the patients with depression for determining the factors possibly associated with health care utilization and expenditures. General estimation equations (GEE) were used for examining the factors associated with outpatient visits and the total medical expenditure of patients with diabetes comorbid with depression after adjusting for covariates. Multiple logistic regression analysis was used for exploring the associations between the factors of interest and hospitalization of patients with diabetes comorbid with depression. All analyses were performed using SPSS version 12.0 (SPSS Inc., Chicago, IL, USA).
The demographic data of patients with diabetes comorbid with and without depression disorder are shown in
The average annual outpatient, inpatient, and total expenditures increased for the two cohorts across the five-year period. The outpatient visits, hospital admission, outpatient expenditures, inpatient expenditures, and total expenditures were significantly higher for patients with diabetes comorbid with depression than for those without depression.
The associations between longitudinal outpatient visits and associated factors analyzed using the GEE model are presented in
The factors associated with the total medical expenditures for patients with diabetes comorbid with depression were analyzed using the GEE model (
The risk of hospitalization in the observation period was analyzed using logistic regression models (
This study has several key findings. First, compared with diabetic patients without depression, those with depression availed more health care services and had higher expenditures during the study period. These results were consistent with those of previous studies.
Second, the study findings indicated that men used less health care services than women do. The finding is consistent with that of a previous study that reported women as being more capable of identifying depressive symptoms and more willing to seek help.
In addition, residing in rural areas and ophthalmic complications were highly associated with hospitalization. This finding was consistent with that of a previous study stating that depressed rural patients had high odds of being hospitalized for physical and mental illnesses.
Investigating health care utilization and expenditure for patients with diabetes comorbid with depression by using an insurance database has many advantages, including a large sample size, less resources required for psychiatric assessments, and the availability of longitudinal data for detecting incidence and risk factors.
However, this study has several limitations. First, the primary data source in this study was from health insurance claim data. There may exist dual diagnoses, over- and underdiagnoses for billing purposes.
Different study designs and instruments may be considered when reviewing health care utilization and expenditures in future studies. ICD-9-CM 296.2x, 296.3x, 296.82, 300.4, 309.0, 309.1, and 311 were included in the definition of depression,
In conclusion, in this large-scale, population-based study, we revealed the health care utilization and expenditure of patients with diabetes comorbid with depression. Thus far, little research has been conducted for clinical practice with respect to health care utilization and expenditures of such patients in Asia, particularly in China. Our findings indicated that health care utilization and expenditures for patients with diabetes comorbid with depression are significantly higher than those for patients with diabetes without depression. Sex, complications, time, and urbanization are factors associated with health care utilization and expenditures. This study, conducted on patients from ethnic groups in Taiwan, provide valuable epidemiological information.
This study was supported by grants from the Kaohsiung Medical University Hospital (KMUH99-9M26) and Kaohsiung Municipal Ta-Tung Hospital (kmtth-100-010 and kmtth-101-013). This study is in part based on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health and managed by National Health Research Institutes (NHIRD-100-100 and NHIRD-102-135). The interpretation and conclusions contained herein do not represent those of the aforementioned agencies.
The online-only Data Supplement is available with this article at
ICD9-CM codes for complications used in this study
Dependents: including family dependents, students and the unemployed. Exchange rate between US$:NT$=1:30
Exchange rate between US$:NT$=1:30. *p<0.05, **p<0.01, ***p<0.001. SD: standard deviation
Exchange rate between US$:NT$=1:30. *dependents: including family dependents, students and the unemployed. GEE: general estimation equations, CI: confidence interval
Dependents: including family dependents, students and the unemployed. Exchange rate between US$:NT$=1:30. CI: confidence interval, OR: odds ratio