Study of Inappropriate Thiazolidinediones Prescribing among Patient with Type Two Diabetes and Heart Failure

• Main Authors: Yin-Yun Huang, 黃尹韻
• Other Authors: Weng-Foung Huang
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/54491366220266502700

碩士 === 國立陽明大學 === 醫務管理研究所 === 97 === Background Medication safety is a very important part of patient safety and medical quality at all times. For the patients with chronic diseases and need long-term medication treatment, potentially inapplicable prescriptions not only bring adverse drug reaction, increase the use of medical care, but also result in hospitalization or even death in the worst scenario. Thiazolidinediones is a kind of oral medicine category for anti-hyperglycemic indication and with certain controversies. At present, most available literature showed that the fluid retention problem occurred after taking thiazolidinediones will increase the risk of heart failure. Health authorities also issued warnings of use on type II diabetes mellitus patients who experienced heart attack or had latent heart disease and heart failure. The package insert of thiazolidinediones indicates it is not recommended for patients with serious heart failure with heart disease classification III and IV (NYHA Functional Classification). However, some foreign literature pointed out that it is still applicable to patients with serious heart failure; and the rate of such situation is gradually increasing. Therefore, it is worthy to explore whether there are similar situations in Taiwan. Objectives The objectives of this study are two folded: (A) To find out the use of TZD in outpatient services of type II diabetes mellitus patients who were hospitalized for serious heart failure after being discharged from the hospital. (B) To analyze the differences between patient characteristics, disease history, the previous TZD medication, as well as the frequency and the volume of TZD dose within one year of hospital discharge. Methods This study is a retrospective cohort study by using National Health Insurance claims data. The study subjects were patients who were diagnosed with type II diabetes mellitus, yet without heart failure in the previous year, and was first hospitalized for serious heart failure in 2005 and discharged in the same year. The risk ratio of study subjects getting TZD prescription from outpatient services after being discharged from the hospital within one year were analyzed for different patients’ characteristics, disease history, and the previous medication experiences by Cox proportional hazard model. Further analysis with multiple-regression-analysis was used to explore whether the frequency and the volume of dose are affected by patients’ characteristics, disease history, and the previous medication experience. Results The main findings of this study are summarized as the followings: 1. 77 of the 849 study subjects had doctors issued TZD type prescription after hospitalized for heart failure, discharged, and follow-up for one year of outpatient visit. The rate of potentially inapplicable prescription is 8.48%. Comparing to patients who have never taken TZD, the patients who had TZD taking history have higher risk of getting the TZD prescription during follow-up outpatient services (HR=6.554, 95% CI=3.903-11.004); Furthermore, patients who were hospitalized for 14 days or more than 14 days have lower risk of getting TZD prescription in comparison to patients who were hospitalized less than 7 days (HR=0.336, 95%CI=0.132-0.854). 2. For patients whom TZD prescription were issued to during one year follow-up after discharged form heart failure hospitalization, the total counts of doctor TZD prescription writing is 559, i.e. an average of 7.76 TZD prescriptions per year for each patient. When compare to all oral medicine for hypoglycemic medicines from outpatient services, the potentially inappropriate prescription rate is 45.56%. There was an average of 136.75 DDD TZD in total per year for each patient, i.e. the dose per day for each patient is 0.79 DDD. The daily dose of the TZD prescription for patients who had previous TZD taking history will be decreased for 10.2% (β=-0.102，t=-2.002，p=0.049) when compared to TZD non-taking subjects; however, the TZD dose for patients with Charlson comorbidity index (CCI) 3 or more will be decreased for 13.1% per day when compared to patients with CCI <3. Conclusions In this study, we found that TZD prescribing for type II diabetes mellitus patients who also complicated by serious heart failure are still in existence. Furthermore, patients who had previous TZD taking history have higher risk of receiving TZD prescription after being discharged form heart failure hospitalization. It may reflect a phenomenon that physicians’ choice of medication still related to patients’ previous medication. This research also found that the yearly frequency among high risk ratio TZD taking patients is not proportional to the amount of doses; it may show that frequency and the amount of doses should be taken into consideration when exploring potentially inappropriate prescriptions. Keywords: thiazolidinediones, potentially inappropriate prescriptions, diabetes mellitus, heart failure