Data source
Taiwan’s National Health Insurance (NHI) program was established in 1995 and covers > 99% of the Taiwanese population (> 23 million beneficiaries). Taiwan’s National Health Insurance Research Database (NHIRD) contains the following encrypted data: patient identification number, birthday, sex, date of admission and discharge, International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic and procedure codes (up to five each), and outcomes. The Longitudinal Health Insurance Database (LHID) 2005 is a subset of the NHIRD used in this study and contains data on the medical service utilization of approximately 1 million randomly selected beneficiaries, who represented approximately 5% of Taiwan’s population in 2005. The NHIRD was used to extract data from 2000 to 2013. The NHI administration regularly conducts random reviews of medical records to ensure diagnostic accuracy. This study was conducted in accordance with the World Medical Association’s Code of Ethics (Declaration of Helsinki). The Institutional Review Board of Tri-Service General Hospital at the National Defense Medical Center in Taipei, Taiwan approved this study, and the need for individual consent was waived because all identifying data were encrypted (TSGHIRB No. E202216004). The NHIRD is a freely accessible database that contains de-identified patient information to protect patient anonymity.
Participants
This study included a cohort of patients from the LHID 2005 database who were newly diagnosed with VI (ICD-9-CM 369.3 and 369.4). We excluded patients who had a VI before 2000, poor prognosis before VI, unknown sex, and incomplete tracking data. Patients with baseline comorbidities of diabetes mellitus (ICD-9-CM 250), hypertension (HTN) (ICD-9-CM 401–405), renal disease (ICD-9-CM 580–589), hyperlipidemia (ICD-9-CM 272), thyrotoxicosis (ICD-9-CM 242), septicemia (ICD-9-CM 003.1, 036.1, 038), pneumonia (ICD-9-CM 480–486), chronic liver disease (ICD-9-CM 571), injury (ICD-9-CM 800–999), and tumor (ICD-9-CM 140–208). The inclusion and exclusion criteria are shown in Fig. 1. The date of the diagnosis of VI was used as the index date. Similarly, participants in the control group were selected from the LHID 2005 cohort. The study and control cohorts were matched 1:4 according to sex, age, and the index date.
Outcome Measurement And Covariates
All participants were followed up from the index date until the first diagnosis of VI, poor prognosis, death, withdrawal from the NHI program, or December 31, 2018. The covariates included sex, age group, geographical area of residence (north, center, south, and east of Taiwan), urbanization level of residence (levels 1–4), and monthly income (in New Taiwan Dollars: <18,000, 18,000–34,999, and ≥ 35,000). The urbanization level of the residence was defined according to the population and various indicators of development. Level 1 was defined as a population of > 1,250,000 with specific designation of political, economic, cultural, and metropolitan development. Level 2 was defined as a population between 500,000 and 1,249,999, with an important role in politics, the economy, and culture. Urbanization levels 3 and 4 were defined as populations between 149,999 and 499,999 and < 149,999, respectively.
Risk factors evaluated for poor prognosis included mental disorders, ≥ 3 outpatient or inpatient visits, anxiety disorders (ICD-9-CM 300), depression (ICD-9-CM 296.2, 296.3, 300.4, and 311), bipolar disorder (ICD-9-CM 296.0, 296.4–296.8), sleep disorders (ICD-9-CM 307.4 and 780.5), post-traumatic stress disorder/acute stress disorder (ICD-9-CM 308 and 309.81), dementia (ICD-9-CM 290.0–290.4, 290.8, 290.9, and 331.0), eating disorders (ICD-CM 307.1 and 307.5), substance-related disorders (SRD; ICD-CM 291–292, 303.3, 303.9, and 304–305), psychotic disorders (ICD-CM 295, and 297–298), autism (ICD-CM 299.00), other mental disorders (ICD-9-CM 290–319, not listed above), suicide (ICD-9-CM E950–E959), and death from all causes (ICD-9-CM 800–999).
Statistical Analyses
The clinical characteristics of the participants are expressed numerically. We compared the distribution of categorical characteristics and baseline comorbidities between the case and control groups using Fisher’s exact test and chi-square test. Continuous variables were presented as means and standard deviations and were compared using the t-test. As the primary goal of this study was to determine whether the clinical characteristics of patients were associated with the poor prognosis, Fine and Gray’s survival analysis and regression analysis were used to determine the risk of poor prognosis (competing with mortality), and the results were presented as hazard ratios (HRs) with the associated 95% CIs. Associations between time-to-event outcomes and clinical characteristics were examined using the Kaplan–Meier method and multivariate Cox regression analysis with stepwise selection. The results are presented as adjusted HRs with the corresponding 95% CIs. All statistical analyses were performed using IBM SPSS Statistics for Windows version 22.0. (released 2013, IBM Corp., Armonk, NY, USA). A two-tailed p < 0.05 was considered statistically significant.