Search

The Rare Disease and Orphan Drug Act (the Act) was enacted in 2000 in Taiwan for the facilitation of the research, development, and accessibility of orphan drugs and special nutritional foods; for the prevention and early diagnosis of rare diseases; and for providing intensive care for patients with rare diseases. The aim was to investigate the impact of the Act on the availability and use of orphan drugs in Taiwan in the hope of identifying the remaining challenges and possible solutions to assist future policy making, which may be applicable in other countries as well. The information and statistics for rare diseases and orphan drugs retrieved from the official annual reports and documents were analyzed. There were 225 diseases recognized as rare diseases, and one-third (75/225) of them were congenital metabolic disorders. Among the 110 designated orphan drugs that could apply for listing in the National Health Insurance (NHI) Pharmaceutical Benefits and Reimbursement Scheme, approximately half (62/110) of them were granted marketing authorization. While the NHI program compulsory for all citizens increased patient accessibility to orphan drugs, the rapidly increasing economic burden became an urgent issue for the government. Emerging gene therapies may be the solution to unmet medical needs and also a financial obstacle to tackle. The Act increased the availability of orphan drugs while the NHI system facilitated patient access, which benefited many patients with rare diseases in Taiwan. However, the soaring economic burden was noticed and was anticipated to aggravate. More communication and cooperation between stakeholders is critical in finding solutions for the long-term sustainability of the NHI system.

Drug lag, also known as drug innovation lag, can be categorized into marketing lag and reimbursement lag. This study investigates the trend and factors influencing drug innovation lag in Taiwan, and makes a comparison with major countries like USA, Canada, Japan, and Europe on marketing lag. The reimbursement lag in the National Health Insurance (NHI) of Taiwan is also explored. This is a retrospective study covering a total of 347 new drugs that were reimbursed by the Bureau of National Health Insurance (BNHI) from 1996 to 2002. Data collection includes time of regulatory approval, type of innovation, country of origin, and time and price of reimbursement by BNHI. The time of new drug approvals in study countries was obtained from relevant websites for comparison. The drug lag index between Taiwan and study countries was also analyzed. Data were analyzed by SPSS software for frequency distribution and multiple regressions. We found that new drugs are predominantly imported, and predominantly imported, and the average marketing lag was up to 30.5 months in Taiwan. Most of the new drugs were of me-too nature; only very few could be classified as breakthrough new drugs. In terms of marketing lag, USA was the shortest (5.6 months), followed by European countries (8.2 months), Canada (18.0 months), and Taiwan (30.5 months). The reimbursement lag was 11.7 months on average after product license granted by DOH, yet it was not affected by the NHI reimbursement price. The drug lag index from smallest to largest was: the USA (0.14), Europe (0.21), Canada (0.45), and Taiwan (0.76). Drug marketing lag is a serious issue in Taiwan. The average marketing lag of 30.5 months could be attributed to the fact that nearly all new drugs were of foreign origins. The average time after DOH's regulatory approval to NHI reimbursement was as long as 11.7 months. The government should reexamine the current function of the regulatory and reimbursement systems. More specifically, the health authorities should focus on faster regulatory process for breakthrough medicines instead of approving only ”me-too new drugs”.

Medication compliance is an important determinant in schizophrenia treatment outcomes. This study aimed to investigate medication compliance in cases of patients receiving a single antipsychotic medication among adult outpatients with schizophrenia, and examine whether their antipsychotic therapy and other factors may affect the medication compliance. Subjects with schizophrenia who had received any single outpatient first- or second-generation antipsychotic therapy from the Yuli Veterans Hospital, Taiwan, during August and November 2006 were identified from medical chart reviews. The selected subjects were surveyed for information including their medication compliance, sociodemographics, treatment-related side effects, social support, and perceived treatmentrelated benefits. Their physicians were also surveyed on their medication compliance, Clinical Global Impression (CGI) scores, and comorbid mental illness (es). Good medication compliance was defined as consumption of more than 75% of prescribed antipsychotics in the previous month. Subjects were categorized into this group when both their self-reported consumption and their physicians' estimation were higher than 75%. A logistic regression model was adopted to evaluate the association of each factor with medication compliance. Of the 76 subjects surveyed, 39 (51%) received second-generation antipsychotics (amisulpride, risperidone, or olanzapine), while 37 (49%) received first-generation antipsychotics (haloperidol or sulpiride). Thirty-eight (50%) consumed more than 75% of prescribed medications in the previous month. The average age was 41, and 62% of the sample was male. Education level and employment type were significantly different between individuals with good and poor medication compliance. The regression result indicated that education of above nine years was significantly associated with an increased likelihood of good medication compliance. Employment and perceived treatment benefits were marginally significantly correlated with good medication compliance. Second-generation antipsychotics, however, had no correlation with good medication compliance.

In order to (1) present the structure of Taiwan’s National Health Insurance (NHI) research databases, along with the comparison with automated databases in other countries, (2) estimate the strengths and weaknesses of the NHI research databases, and (3) systematically review pharmacoepidemiology studies using the NHI research databases, we compared the characteristics of existing automated databases to reveal the strengths and weaknesses of the NHI research databases. In addition, the Medline was used as a tool to search pharmacoepidemiology studies using Taiwan’s NHI research databases since 1997. The automated NHI research databases are very comparable with large research databases of other countries (US, Canada and UK). As a result, they serve as major resources for up to 11 pharmacoepidemiology studies published since 2002. However, these studies usually focused on the analysis of drug utilization pattern and drug utilization volume. As a result, there is still a huge lack of identification of potential drug safety issues using the NHI research databases. The construction of NHI research databases absolutely provides abundant research resources for scholars not only in medical fields but also in public health-related disciplines. Many studies using the NHI research databases have been published in international journals. Yet, researchers in Taiwan could make even greater progress toward thorough pharmacoepidemiology studies using the NHI research databases.

Many studies explore the effects of medication switching from second generation antipsychotic (SGA) to first generation antipsychotic (FGA), yet few exam the health effect of backward medication switching. This study investigated whether backward medication switching (from SGA to FGA) and reduced dosages on SGA increased adverse events in outpatient schizophrenia patients treated by SGA in Taiwan. This study applied a three-year historical cohort design and compared the risks of adverse events (outcome) in 2004, measured by emergency room and hospital admissions, of patients experienced backward medication switching or dosage reduction on SGA (exposure status) in 2003 with those who did not (control group). This study was based on a national fixed cohort (census) of 12,107 patients who have ever received SGA during the study period and used hospitals as regular source of care during 2002-2004. Most data were drawn from National Health Insurance claims provided by Department of Health. Cox proportional hazards model for recurrent events using Prentice, Williams and Peterson total-time approach was applied to estimate the hazard ratio (HR) of adverse events by exposure status of the patients while controlling the covariates. The results showed that backward medication switching in 2003 significantly increased the risk of emergency room or hospital admissions (HR=1.15, 1.00-1.31) in 2004. Backward medication switching and reduced dosages on SGA might increase the risk of adverse events of the schizophrenia patients treated by SGA in Taiwan.

Psoriasis increases the incidence of hypertension and cardiovascular disease. However, its effect on the course of cardiovascular disease remains unknown. To investigate whether patients with psoriasis and hypertension have a higher requirement for cardiovascular procedure and surgery than patients with hypertension but without psoriasis, we used the Taiwan National Health Insurance Research Database to identify patients with new-onset hypertension during 2005-2006. Among these patients, those with psoriasis (n = 4039) were matched in a 1:1 ratio by age and sex with patients without psoriasis. The association between psoriasis and cardiovascular interventions was examined using time-varying Cox proportional hazards models. The mean follow-up period was 5.62 years. Psoriasis was associated with an increased risk for cardiovascular procedure and surgery in patients with hypertension (adjusted hazard ratio [aHR], 1.28; 95% confidence interval [CI], 1.07-1.53). When no psoriasis served as a reference group, the aHRs were higher for women than for men, and for patients aged 50-64 years than for younger and older patients. Patients with severe psoriasis or psoriatic arthritis tended to have higher risks of cardiovascular procedure and surgery than patients with mild psoriasis (aHR, 1.22; 95% CI, 0.98-1.51) or patients without psoriatic arthritis (aHR, 1.15; 95% CI, 0.84-1.58), respectively, did, although not reaching statistical significance. In conclusion, patients with hypertension and psoriasis had a greater requirement for cardiovascular interventions than hypertensive patients without psoriasis. More intense assessments for cardiovascular interventions may be necessary in patients with concurrent hypertension and psoriasis than general hypertension patients.

Background: Incidence of actinic keratosis (AKs) and nonmelanoma skin cancer appears to be increasing worldwide due to increasing levels of ultraviolet radiation, lifestyle changes, and an aging population. However, there is no population-based study focusing on AKs among Asian population. We aimed to investigate the incidence, the treatment pattern, the medical utilization and the risk of malignant neoplasm of the AKs patients in Taiwan. Methods: We conducted a retrospective cohort study to investigate patients with AKs from 2003 to 2011 by using the National Health Insurance Research Database (NHIRD). The inclusion criteria were patients who had at least two outpatient visits or one hospital admission of AK identified by dermatologist from 2004 to 2011. There were 35,933 patients with AKs (elderly [aged ≥65 years]: n = 17,004; non-elderly [aged

Background Phototherapy might increase bone mineral density. However, it is unknown whether phototherapy can reduce the risk of fractures in patients with vitiligo. Objectives To investigate the effect of phototherapy on fracture risks in vitiligo patients aged 40 or older. Methods This population-based cohort study used the 2000–2010 Taiwan National Health Insurance Research Database (NHIRD) to identify 3863 patients newly diagnosed with vitiligo between 2003 and 2009 at age ≥40 years. Study subjects were classified into three cohorts: (1) frequent phototherapy; (2) infrequent phototherapy; and (3) no phototherapy. Patients were followed until the first hip or vertebral fracture or 31 December 2010. Data were analysed using Cox regression models and also stratified by age and gender. Results Frequent phototherapy decreased the fracture risks (adjusted hazard ratio (aHR) = 0.32, p = 0.009) in vitiligo patients. Stratification by age and gender confirmed the fracture prevention effect of frequent phototherapy in patients aged 40–64 years (aHR = 0.14, p = 0.016) and in female patients (aHR = 0.31, p = 0.024). Conclusions This study provides the first evidence that frequent phototherapy can reduce the risk of fractures among middle-aged and among female vitiligo patients.

Reduced male fertility potential is not uncommon in some rheumatic diseases, i.e. systemic lupus erythematosus, dermatomyositis, and rheumatoid arthritis (RA) (1). Varicocele – abnormal venous dila...

Background Deferasirox (DFX) is an effective and well-tolerated oral iron chelator elevating the adherence to iron chelating therapy among patients with iron overload. However, the US Food and Drug Administration issued a warning about the potential adverse events associated with DFX in 2010. Methods To examine the risks of gastrointestinal (GI) bleeding, acute liver necrosis, and acute renal failure among DFX users compared with desferrioxamine (DFO) users in a real-world setting, first-time users of DFX or DFO between 2005 and 2008 in Taiwan's National Health Insurance database were observed in this population-based retrospective cohort study. The risks of different adverse events were individually analyzed by Cox proportional hazards models and adjusted by age, sex, concomitant medications, and prior medical conditions. Results Deferasirox users had the highest incidence rates of GI bleeding (2.03 per 10 000 patient-days), acute liver necrosis (0.26 per 10 000 patient-days) and acute renal failure (1.45 per 10 000 patient-days) compared with other iron chelator users. Compared with DFO users, DFX users were not associated with the risk of GI bleeding (adjusted HR 1.03, 95% CI 0.61–1.74, p = 0.90) and the risk of acute liver necrosis (adjusted HR 2.13, 95% CI 0.49–9.33, p = 0.32). The association between DFX use and acute renal failure was found to be statistically significant (HR 2.18, 95% CI 1.18–4.02, p = 0.01; adjusted HR 2.41, 95% CI 1.27–4.58, p = 0.01). Conclusion In this study, we found statistically significant higher risk of acute renal failure and non-statistically significant higher risk of GI bleeding and acute liver necrosis associated with DFX use. More researches are warranted to evaluate the association between DFX use and potential adverse events. Copyright © 2014 John Wiley & Sons, Ltd.

Evidence has emerged that pioglitazone may increase the risk of bladder cancer, but the association has not been confirmed. This potential risk also has not been evaluated in users of rosiglitazone. Using Taiwan’s National Health Insurance Research Database (NHIRD), this large population-based nested case–control study was conducted to explore the relationship between the use of rosiglitazone or pioglitazone and risk of bladder cancer in diabetic patients. We identified 3,412 cases of newly diagnosed bladder cancer and 17,060 controls (1:5 matched by age and sex) among a diabetic patient cohort from the NHIRD. We defined an index date for each case as the date of first hospitalization for bladder cancer. Each control was assigned the index date of their corresponding case. Multivariable conditional logistic regressions were used to estimate the association between exposure (timing and duration) to rosiglitazone or pioglitazone and bladder cancer. We defined rosiglitazone or pioglitazone exposure as “current” if the prescription duration covered the index date or ended at 90 days before, as “recent” if it ended 91–180 days before the index date, or as “past” if the last prescription ended more than 180 days before. Duration of rosiglitazone or pioglitazone use was defined based on the cumulative days of exposure prior to the index date

Background Population aging has been a critical issue around the world and people will have to face living problems when they get old. In Western countries, older people are more used to live alone or in institutions. In Eastern countries, due to filial piety of Chinese culture, the elderly prefer to live with their children or their relatives. There was no empirical study to investigate the relationship between health and living arrangement among older Taiwanese. Objective This study was designed to explore the association between living arrangement and health characteristics among the elderly in Taiwan. Method This study used national representative data from the Taiwan Longitudinal Study on Aging surveyed in 2007. We identified 2621 elders aged older than 65 in 2007 and categorized them into 3 types of living arrangement by the questionnaire. Linear regressions were used to analyze the relationship between living arrangement and health status (activities of daily living [ADLs], instrumental activities of daily living [IADLs], and Center of Epidemiological Studies–depression [CES-D]) among the elderly. Results Elderly individuals who indicated they rotationally lived with family members had poorer health conditions, including IADLs (Coeff = 0.23; 95% confidence interval [CI]: –0.06–0.52) and CES-D (Coeff = 0.41; 95% CI: –0.59–1.40), than those who steadily lived with family. In contrast, elderly individuals who lived alone had better health conditions in IADLs (Coefficient = –0.38; 95% CI: –0.53 to –0.22) than those who indicated they lived steadily with family. Conclusions These findings reveal that this type of rotational living is not a good living arrangement for the elderly.

Background Avascular necrosis (AVN) and psoriasis have some pathogenic mechanisms and associated conditions in common. Objective To examine the association between psoriasis and AVN. Methods This study used data from the Taiwan National Health Insurance Research Database for the period 2004–2006 and identified 28,268 patients with psoriasis, who were then matched for age and sex with 113,072 controls without psoriasis from the Taiwan Longitudinal Health Insurance Database 2000. Multivariate Cox proportional hazards models were used for the analysis. Results The unadjusted risk of AVN was significantly higher for patients with psoriasis than for controls (hazard ratio [HR] 2.29) and remained significant after adjustment for other risk factors (adjusted HR 1.96; 95% confidence interval 1.62-2.38). The risk for AVN increased in relation to psoriasis severity and was higher for patients with psoriasis and arthritis than for patients without arthritis. The adjusted HRs were higher for male patients than for female patients and for patients younger than 30 years compared with older patients. Limitations We lacked information on daily tobacco use, alcohol consumption, and physical activity. Conclusion The risk for AVN increased with the disease severity of psoriasis.

Background Inflammation of systemic and vascular tissues besides the skin in psoriasis is associated with cardiovascular morbidity and mortality. Objective We sought to investigate whether or not patients with psoriasis have an increased risk of aortic aneurysm (AA). Methods This population-based cohort study identified 34,301 patients with psoriasis in the Taiwan National Health Insurance Research Database during 2004 to 2006, who were matched for age and sex with 137,204 control subjects without psoriasis from the Taiwan Longitudinal Health Insurance Database 2000. Each individual was individually followed up for 5 years to identify those who subsequently developed AA. Results After adjusting for medical history and medication use, patients with psoriasis were at increased overall risk of AA (adjusted hazard ratio [HR] 1.80; 95% confidence interval 1.25-2.61). The risk for AA increased with the severity of psoriasis. The adjusted HRs were higher for male than female patients (adjusted HR 1.84 vs 1.56), and for patients younger than 50 years versus older patients (adjusted HR 2.81 vs 1.64). Limitations There is a lack of information regarding patients' Psoriasis Area and Severity Index score, daily tobacco use, or alcohol consumption. Conclusion Patients with psoriasis are predisposed to developing AA: this risk increases with psoriasis severity and is independent of established cardiovascular risk factors.

BACKGROUND:The increased rates of cardiovascular morbidity and mortality in patients with psoriasis are not adequately explained by traditional risk factors. Whether concomitant sleep disorders (SDs) modify the risk of cardiovascular disease (CVD) in patients with psoriasis remains unknown. METHODS:Using the Taiwan National Health Insurance Research Database (NHIRD), we conducted a cohort study to investigate the association between concomitant SDs and CVD risk in patients with psoriasis. Data from 99,628 adults who received a psoriasis diagnosis during the period from 2004 to 2010 were analyzed. Cox proportional hazards regression analysis models were used to compare the risks of ischemic heart disease (IHD) and stroke between patients with and without SDs. RESULTS:Psoriasis patients with a concomitant SD had significantly higher risks of IHD (adjusted hazard ratio [aHR], 1.25; 95% confidence interval [CI], 1.22-1.28) and stroke (aHR, 1.24; 95% CI, 1.16-1.33) as compared with psoriasis patients without SDs. All psoriasis patient subgroups, including those with mild and severe psoriasis and those with and without arthritis, had increased HRs for IHD and stroke. The increases in IHD and stroke risks conferred by SDs were proportional to the dose of hypnotics used. The effect of SDs on the risks of IHD and stroke was greater in young adults than in middle-aged and older adults. CONCLUSIONS:The risks of IHD and stroke were higher for psoriasis patients with SDs than for those without SDs. Clinicians should carefully evaluate CVD risk, particularly in young patients with psoriasis.

Background Thiazolidinediones (TZDs) may reduce in-stent restenosis and improve clinical outcomes in type 2 diabetic patients after bare-metal stent implantation. However, it is still unknown whether diabetic patients with drug-eluting stents (DESs) could benefit from treatment with TZDs. Objective The objective was to evaluate the clinical outcomes of TZDs in type 2 diabetic patients within 1 year of receiving DESs. Methods This retrospective cohort study was performed in 1743 Taiwanese type 2 diabetic patients (1137 men; 606 women) who received DESs between December 1, 2006 and December 31, 2007. Patients were classified into TZD (n = 268) or non-TZD groups (n = 1,475) using medication records within 3 months of the index hospitalization. Follow-up data were available through December 31, 2008. Clinical outcome measurements included death, myocardial infarction (MI), and repeat revascularization within 1 year after the index date of hospitalization. Cox proportional hazards model and other analyses were performed for the study. Results For the TZD and non-TZD groups, the mean ages were 65.07 and 66.09 years, respectively, for those with limus-eluting stents (LESs) and 65.61 and 65.81 years, respectively, for those with paclitaxel-eluting stents (PESs). With or without TZD medication, there were no significant differences in the adjusted hazard ratios of death, MI, or repeat revascularization for diabetic patients who received LESs or PESs. TZD treatment in patients who received LESs and had a history of MI was associated with a higher risk of MI (hazard ratio = 5.292; 95% CI, 1.028–27.232). Conclusions TZDs did not improve the clinical outcomes in Taiwanese type 2 diabetes patients who received DESs. TZDs might have been a contributor to higher risk of MI in patients with LESs and a history of MI. Larger clinical trials are still needed to study this issue further.

Background Bisphosphonates are the class of medication used most widely to treat osteoporosis. Since an article reported that patients who used zoledronic acid, a bisphosphonate, had a higher proportion of atrial fibrillation (AF) in 2007, the issue of bisphosphonates and AF has become a growing concern. Due to the widespread use of bisphosphonates, it is necessary to explore the relationship between bisphosphonates and AF and other cardiovascular diseases. Objective We aimed to investigate the risk of AF, stroke, or acute myocardial infarction (AMI) associated with the use of the bisphosphonates alendronate and raloxifene in patients with osteoporosis. We also focused our analysis on the impact of different dosing regimens of alendronate. Methods The National Health Insurance Research Database was used to conduct an 8-year, population-based, retrospective cohort study. The study population comprised women who first took alendronate or raloxifene between 2002 and 2006 and who had a history of osteoporosis and vertebral or spinal fracture. Follow-up was conducted for every patient until the first diagnosis of AF, stroke, or AMI or until the end of the 1-year follow-up period. The Cox proportional hazards model was used to evaluate the association between the risk of cardiovascular disease and the prescription of alendronate or raloxifene. Results We identified 9609 women who had been prescribed either alendronate (n = 6949) or raloxifene (n = 2660). The patients treated with alendronate were at a lower risk of AF, stroke, or AMI compared with the raloxifene group (AF: hazard ratio [HR] = 0.60 [95% CI, 0.42–0.85]; stroke: HR = 0.47 [95% CI, 0.39–0.57]; AMI: HR = 0.51 [95% CI, 0.36–0.72]). However, when analyzing the groups by different alendronate dosing regimens, those patients who received alendronate 10 mg had a significantly higher risk of AF and stroke compared with patients who received raloxifene (AF: HR = 1.66 [95% CI, 1.12–2.46]; stroke: HR = 1.56 [95% CI, 1.23–1.98]). The alendronate 70-mg group demonstrated a lower risk of cardiovascular disease, be it AF, stroke, or AMI (AF: HR = 0.28 [95% CI, 0.18–0.43]; stroke: HR = 0.23 [95% CI, 0.18–0.30]; AMI: HR = 0.28 [95% CI, 0.18–0.41]). When we assigned alendronate 10 mg as the reference group, the alendronate 70 mg group had a lower risk of 3 cardiovascular diseases (AF: HR = 0.17 [95% CI, 0.10–0.27]; stroke: HR = 0.16 [95% CI, 0.12–0.22]; AMI: HR = 0.21 [95% CI, 0.13–0.35]). Conclusions Alendronate 10 mg was associated with a higher risk of cardiovascular disease than alendronate 70 mg. Further studies are required to investigate this relationship.

Background There is conflicting evidence regarding the potential interaction between clopidogrel and proton pump inhibitors (PPIs), with observational studies suggesting an increased risk of adverse cardiovascular (CV) outcomes and clinical trials suggesting there is no such risk. Methods We conducted a retrospective cohort study to assess CV outcomes of 9753 patients taking dual antiplatelet therapy of aspirin plus clopidogrel with or without a PPI after hospitalization for acute coronary syndrome (ACS). Cox proportional hazards models were used to assess our primary endpoint of re-hospitalization for ACS in overall sample and a propensity score matching subsample. Results Among patients taking clopidogrel plus aspirin, concomitant use of PPI was not associated with the risk of re-hospitalization for ACS (adjusted hazard ratio [HR] 1.12 [95%CI 0.72–1.73]). The findings were consistent in the propensity score matching cohort (adjusted HR 0.82 [95%CI 0.43–1.54]). Compared with PPI nonusers, there is no significant association between each specific PPI users and the risk of re-hospitalization for ACS (adjusted HR; omeprazole 0.96 [95%CI 0.35–2.66], pantoprazole 1.05 [95%CI 0.38–2.92], rabeprazole 0.60 [95%CI 0.17–2.17], esomeprazole 0.31 [95%CI 0.10–0.99], and lansoprazole 0.82 [95%CI 0.32–2.07]). Conclusion In conclusion, this population-based cohort study found that concomitant use of clopidogrel and PPI in patients who received dual antiplatelet therapy after ACS was not associated with risk of ACS re-hospitalization. Together, our study and findings of recently published clinical trials suggest that there was no apparent CV interaction between clopidogrel and PPI in patients who received dual antiplatelet therapy. Copyright © 2011 John Wiley & Sons, Ltd.

Background Carbamazepine has been associated with severe cutaneous adverse drug reactions (ADR). It is important for patients with these ADRs under off-label prescriptions are not eligible for drug injury relief in Taiwan. We conducted a study to depict the demography and possible factors related to the off-label carbamazepine use in Taiwan. We also explored the policy influence of carbamazepine use. Methods We used the dataset of one million randomly-sampled insured persons in the Taiwan National Health Insurance Research Database for 2002 and 2006 to conduct a cross-sectional study. With the use of a computerized clinical information system, carbamazepine prescriptions were categorized into five groups (A, B, C, D and E) according to these indications: Group A, B were defined as on-label use; group C, D and E were defined as various levels of off-label use, depending on the strength of support from the literature. A logistic regression model was conducted to find the factors related to off-label use. Results Based on the one million representative samples, 6305 and 5703 patients received 31 146 and 27 579 carbamazepine prescriptions in 2002 and 2006 respectively. In both years, nearly 43% of total prescriptions were related to on-label uses. Prescriptions from primary clinics and departments of internal medicine and psychiatry, the physician's age, and the patient's age were factors associated with higher risk of off-label carbamazepine use. Conclusions Our study echoes the highly prevalent off-label use of carbamazepine in Taiwan and adds to the rare research on this subject in the East Asian population. The carbamazepine relabelling in 2004 did not change either prescription patterns or factors related to off-label use.

To compare Helicobacter pylori eradication therapy with antisecretory therapy alone on the risk of hospitalization for a major ulcer event.Retrospective, population-based cohort study.The 2000-2006 National Health Insurance database in Taiwan.A total of 838,176 patients diagnosed with a gastrointestinal ulcer and who filled at least one prescription for antiulcer therapy, either H. pylori eradication therapy (331,364 patients [39.53%]) or antisecretory therapy alone (506,812 patients [60.47%]), between January 1, 2001, and December 31, 2006.The primary outcome was hospitalization for a major ulcer event, defined as a gastrointestinal ulcer with hemorrhage and/or perforation. Cox proportional hazards models, adjusted for demographic and clinical characteristics, were used to compare the risk of hospitalization for a major ulcer event between the group receiving H. pylori eradication therapy (triple or quadruple combination therapy that includes an antisecretory agent) and the group receiving antisecretory therapy alone (histamine₂-receptor blocker or proton pump inhibitor). The H. pylori eradication therapy group was divided into initial users (combination therapy received immediately after gastrointestinal ulcer diagnosis) and late users (combination therapy received after antisecretory therapy with time lag ≤ 180 days, 181-365 days, or365 days from ulcer diagnosis). A secondary analysis was conducted in the three late H. pylori eradication therapy subgroups to determine if risk of hospitalization for major ulcer events differed by timing of receipt of therapy. Compared with the antisecretory therapy alone group, the H. pylori therapy group (initial users) had a significantly decreased risk of hospitalization for major ulcer events (adjusted hazard ratio [AHR] 0.57, 95% confidence interval [CI] 0.54-0.59, p0.001). However, later use of H. pylori therapy was associated with a higher risk of hospitalization for major ulcer events (time lag 181-365 days, AHR 1.68, 95% CI 1.51-1.86, p0.001;365 days, AHR 1.74, 95% CI 1.67-1.80, p0.001) compared with those who received H. pylori therapy within 6 months (≤ 180 days) after gastrointestinal ulcers were diagnosed.Helicobacter pylori therapy given within 6 months of a diagnosis of gastrointestinal ulcer was associated with a reduced risk of hospitalization for major ulcer events. Our findings extend the evidence from clinical trials that report the value of H. pylori eradication therapy in reducing ulcer recurrence by documenting the real-world benefit of reducing the risk of hospitalization for major gastrointestinal ulcer events.

Purpose To examine the potentially inappropriate prescription of thiazolidinediones (TZD). Methods Data on TZD prescriptions were collected from Taiwan's National Health Insurance dataset from 2001 to 2006. TZDs were considered inappropriately prescribed when they were prescribed to patients who were (1) under 18 years old, (2) pregnant, who had (3) type 1 diabetes, (4) severe heart failure, (5) hepatic insufficiency, or (6) renal insufficiency and taking TZD + metformin in combination. We aggregated potentially inappropriate prescriptions of TZD for each health-care institution in each month starting from March 2001, when TZD was introduced to Taiwan's market. Results The potentially inappropriate prescription of TZD increased from 9.41% in 2001 to 12.50% in 2006. Prior inappropriate prescription led to a 0.06% (95%CI: 0.04–0.08) further increase in its later inappropriate prescription. Accumulated months of experience prescribing TZD was found associated with higher proportion of inappropriate prescription of TZD (0.03%, 95%CI: 0.01–0.05). However, it was negatively associated with new incidence of inappropriate prescription of TZD (−0.20, 95%CI: −0.22 to −0.18). The greater the volume of prior TZD prescription (−0.87%, 95%CI: −0.93 to −0.81) and the greater the number of accumulated months since adoption (−0.14%, 95%CI: −0.16 to −0.12), the greater the decrease in rates of new inappropriate prescriptions. Conclusions Along with the quick penetration of the new DM drug came an increased possibility that it would be prescribed inappropriately, a trend that persisted over time. Copyright © 2010 John Wiley & Sons, Ltd.

Concomitant use of antiplatelet agents and proton-pump inhibitors (PPIs) has been recommended in patients with a history of gastrointestinal (GI) hemorrhage. However, recent studies have reported that PPIs may alter clopidogrel’s pharmacokinetics and potentially lead to an increased risk of recurrent adverse cardiovascular (CV) events. Using Taiwan’s 2000–2006 National Health Insurance database, this population-based retrospective cohort study assessed CV and GI events in patients who had a prior history of GI bleeding and had been prescribed ongoing antiplatelet therapy after acute coronary syndrome (ACS) discharge. We identified 3,580 ACS patients and categorized them into (1) those taking clopidogrel alone, (2) those taking clopidogrel plus PPIs, and (3) those taking aspirin plus PPIs. Cox proportional hazards models were used to assess the association between the use of antiplatelet therapies and CV/GI events. The clopidogrel only group and the clopidogrel plus PPI group were found to be at lower risk for GI events than the aspirin plus PPI group [adjusted hazard ratio (HR) 0.23 (95% confidence interval; CI 0.14–0.36) and HR 0.70 (0.52–0.96), respectively]. However, while the clopidogrel only group had a lower risk of CV events than the aspirin plus PPI group [HR 0.57 (0.38–0.84)], the clopidogrel plus PPI group had a significantly higher CV risk than the aspirin plus PPI group [HR 1.59 (1.18–2.13)]. Our findings suggest that although the use of clopidogrel plus PPIs provides GI benefits, with this treatment, there is an increased CV risk among patients with a history of GI bleeding.

Background Thiazolidinediones (TZDs) are widely used antidiabetic agents, but there is great concern and conflicting reports over their possible effect on cardiovascular morbidity, especially in patients with heart failure (HF). Methods Using 2000–2005 Taiwan's National Health Insurance (NHI) claims data, this population-based, retrospective cohort study investigated if there was an association between the cumulative TZD dose and clinical outcomes in type 2 diabetic patients recently hospitalized for HF. Study outcomes were death, first all-cause readmission, and first readmission for HF. Cox proportional hazard models were used to analyze the association between TZD versus sulfonylurea (SU) use and these outcomes. Results Out of a total of 8139 eligible patients, 224 were taking TZD (65.63% female; mean [SD] age, 68.30[10.60] years) and 7915 were taking SU (55.10% female; 70.30[10.50] years). Patients taking TZD were at higher risk for readmission for HF (59 cases; HR 1.58 (95% confidence interval, 95%CI 1.44–1.72)), all-cause readmission (147 cases; 1.40 (1.30–1.70)), and death (103 cases; 2.23 (1.58–3.14)). The higher the cumulative exposure to TZD, the greater the risk of HF readmission, all-cause readmission, and death. Conclusion Among diabetic patients who had been hospitalized for HF, TZD users were at significantly greater risk for readmission for HF, all-cause readmission, and death than SU users. The risk of all adverse clinical outcomes appeared to highly relate to cumulative exposure to TZD. These findings provide empirical evidence supporting the latest black box warnings issued by the United States Food and Drug Administration in August, 2007 advising that TZD not be prescribed for diabetic patients with symptomatic heart failure. Copyright © 2010 John Wiley & Sons, Ltd.

Background: Safety concerns regarding severe cardiovascular events associated with the use of selective cyclooxygenase-2 (COX-2) inhibitors resulted in the market withdrawal of rofecoxib in September 2004. Objective: Using Taiwan's National Health Insurance 2003—;2004 claims database, this population-based retrospective cohort study assessed changes in physicians' practice of prescribing a COX-2 inhibitor after market withdrawal of rofecoxib. Methods: Patients with rheumatoid arthritis (RA) or osteoarthritis (OA) who were chronic users of COX-2 inhibitors before market withdrawal of rofecoxib were identified. Eligible chronic users of COX-2 inhibitors were patients who received ≥3 consecutive prescriptions for celecoxib or rofecoxib to treat RA or OA between April 1 and September 30, 2004. The main outcome evaluated in this study was the volume of celecoxib prescribing for each patient by his or her prescribing physician at the first postwithdrawal outpatient visit related to RA or OA. For each matched physician, we calculated the prescribing practice indexes before and after the withdrawal of rofecoxib to assess changes in prescribing practice. A higher value of the index represented less potential that patients with a history of cardiovascular events who were seen by a physician received a prescription for a COX-2 inhibitor. A 2-stage model analysis was used to assess changes, in physicians' prescribing practice after rofecoxib's withdrawal, in the volume of prescribing a COX-2 inhibitor for each matched patient by his or her prescribing physician. Results: Of the chronic users of COX-2 inhibitors identified, 13,101 were taking celecoxib and 8763 were taking rofecoxib before rofecoxib was withdrawn from the market. Concerns about safety after the mar- ket withdrawal of rofecoxib reduced physicians' volume of prescribing a COX-2 inhibitor, particularly in patients who had previously received rofecoxib. After rofecoxib's withdrawal, 72.50% of previous rofecoxib users (n = 6353) and 49.50% of previous celecoxib users (n = 6485) stopped taking COX-2 inhibitors. Only 27.50% of the rofecoxib users (n = 2410) were switched to celecoxib after rofecoxib's withdrawal. Overall, physicians' prescribing practice for celecoxib increased, from baseline to 3 months after rofecoxib's withdrawal, from 0.66 to 0.79 for academic medical center physicians, 0.71 to 0.82 for metropolitan hospital physicians, 0.77 to 0.88 for local community hospital physicians, and 0.86 to 0.92 for primary care clinic physicians. After controlling for patients' demographics, the linear regression analysis revealed that changes in physicians' prescribing practice after the withdrawal of rofecoxib affected the volume of prescribing a COX-2 inhibitor ( P Conclusions: The volume of celecoxib prescribing was greatly reduced after rofecoxib was withdrawn from the market. Physicians' prescribing practice for COX-2 inhibitors significantly changed after the withdrawal, and it had a significant impact on the postwithdrawal volume of celecoxib prescribing.

This study used Taiwan's National Health Insurance claim database (years 2000-2005) to examine how thiazolidinediones (TZD), a new class of drugs for diabetes, penetrated into Taiwan's hospitals, and its association with the concentration of all diabetes drugs at the hospital level. We collected 72 monthly summaries of diabetes prescriptions from all hospitals in Taiwan. Hospital-level pharmaceutical concentration was measured by penetration of TZD, defined as monthly market share of TZD in each hospital. Concentration of diabetes drugs was measured by Herfindahl-Hirschman indices. We found a negative association (coefficient = -0.3610) between TZD penetration and concentration of diabetes drug but a positive association between penetration of TZD and the volume of prescribed diabetes drugs (coefficient = 0.4088). In conclusion, hospital characteristics and volume of services determined the concentration of pharmaceuticals at the institution level, reflecting the heterogeneous competition between pharmaceutical companies within each hospital. Institution-level pharmaceutical concentration influences the adoption and penetration of new drugs.

Background Psoriasis is associated with cardiovascular morbidity and mortality. However, the association between psoriasis and arrhythmia has not been adequately studied. Objective We sought to investigate whether patients with psoriasis have an increased risk of arrhythmia. Methods This population-based cohort study identified 40,637 patients with psoriasis and 162,548 subjects without psoriasis matched by age, sex, history of coronary artery disease, hypertension, and diabetes in the Taiwan National Health Insurance Research Database during 2004 through 2006. Results After adjusting for medical history and medication use, patients with psoriasis were at increased risk of overall arrhythmia (adjusted hazard ratio [aHR] 1.34; 95% confidence interval [CI] 1.29-1.39). The risks of arrhythmia were higher in all subgroups, including patients with severe (aHR 1.25; 95% CI 1.12-1.39) and mild (aHR 1.35; 95% CI 1.30-1.41) psoriasis, and in patients with (aHR 1.46; 95% CI 1.22-1.74) and without (aHR 1.33; 95% CI 1.28-1.39) psoriatic arthritis. Limitations The National Health Insurance Research Database did not contain information regarding Psoriasis Area and Severity Index, cigarette smoking, or alcohol consumption. Conclusion Patients with psoriasis were at higher risk of developing arrhythmia, particularly for those with psoriatic arthritis, independent of traditional cardiovascular risk factors.

Objective: This study was based on Taiwan's National Health Insurance (NHI) claim records with the aim of identifying specific types of potentially inappropriate sedative-hypnotic prescribing in elderly outpatients with insomnia. The potentially inappropriate prescribing included duplicate treatment, excessive dosage and duration or treatment and prescribing of hypnotics that are too long-acting. Material and methods: This cross-sectional study was based on annual outpatient claim data for 2001 released by Taiwan's Bureau of National Health Insurance (BNHI). A subset was created for patients aged 65 years or older and coded as having insomnia. Physician consultation claim data were extracted and merged in 1 claim file consisting of ICD-9-CM codes, patient demographic data, specialty of physicians, medical institution code and pharmaceutical prescription content. Results: Elderly patients with insomnia constituted 216,994 of the 1,000,193 files surveyed on outpatient claims in 2001. The mean age was 74.33 years and gender distribution was nearly equal. Of the prescribed sleep medications, 41.26% were hypnotic benzodiazepines, 29.36% were hypnotic non-benzodiazepines and 29.38% were sedative-anxiolytics. Approximately 1 in 25 patients (4.12%) of all the patients prescribed hypnotics received duplicate treatment, 1 in 8 an inappropriately (12.27%) high dosage (daily dose > 1.5 DDD, Defined Daily Dose), 1 in 3 (32.25%) more than 28 DDD per prescription and 1 in 6 (17.52%) a drug with an effect which was too prolonged. Physicians tended to consider patient gender but not age when prescribing. Clear trends were found between the specialty of the physician and the type of inappropriate prescribing. The type of medical institution was significantly related only to the excessive quantity of medication prescribed. Conclusions: Elderly people in Taiwan with insomnia receive potentially inappropriate prescriptions for sleep medications. Similar data could possibly be extracted from similar databases in other countries throughout the world. Some of these potentially inappropriate prescriptions are avoidable in terms of restricting the length of outpatient sedative-hypnotic treatment, introducing hypnotics in small dosage forms and continuously educating clinicians on the safety of geriatric medication.

To examine the relationship between different anti-diabetic therapies (dipeptidyl peptidase-4 (DPP-4), metformin and sulfonylureas) and risk of acute pancreatitis among type 2 diabetic patients in Taiwan, and explore each drug's dose-response relationship.We derived a nationwide retrospective cohort of patients with type 2 diabetes in Taiwan. The inclusion criteria are adult diabetic patients with continuous baseline enrollment, new users of the studied drugs, and without missing demographics. There were 4113/101 498/44 772 DPP-4/Metformin/Sulfonylurea users. Adjusted hazards ratios for pancreatitis associated with DPP-4, derived from Cox proportional hazard models with propensity score weighting, were estimated; dose-response analyses were also conducted.Dipeptidyl peptidase-4 was statistically significantly associated with a decreased risk of acute pancreatitis compared with sulfonylureas (adjusted HR: 0.36, 95%CI [0.17, 0.75]) but not metformin (adjusted HR: 0.67, 95%CI [0.32, 1.41]); metformin was statistically significantly associated with a lower risk of pancreatitis than sulfonylurea (adjusted HR: 0. 53; 95%CI [0.37, 0.76]). In addition, low-dose metformin was statistically significantly associated with a lower risk of pancreatitis compared with high-dose metformin (HR: 0.65; 95%CI [0.44, 0.97]).Our findings suggest that sulfonylureas may potentially be associated with an increased risk of pancreatitis compared with DPP-4 or metformin. Studies with longer follow up, larger sample sizes, and more precise capture of confounders may be needed to determine the risk of pancreatitis associated with incretin based therapies.

To explore prescription patterns and determinants of sleep-related medications prescribed to elderly outpatients with insomnia in Taiwan. This cross-sectional study was based on 2001 annual outpatient claims data released by the Bureau of National Health Insurance in Taiwan. The claims data of each physician consultation were extracted and merged in one claim file. International Classification of Diseases (9th Edition)-Clinical Modification codes, patient’s demographics, physician’s specialty, the medical institution code and the content of pharmaceutical prescription constituted a file. Patients were included if they were: (i) ≥65 years of age; and (ii) coded as having ‘insomnia’. Elderly insomniacs made up 216 994 of the 1 000 193 outpatient claim files we surveyed. Patients had a mean age of 74.33 years, and the sex distribution was nearly equal. Based on the data above, 11.14% of the elderly had been diagnosed as having ‘insomnia’ for the year 2001. This population preferred primary-care clinics over hospital-ambulatory departments; patients most frequently sought medical help from internal medicine specialists. The most popular sleep medication was lorazepam, followed by zolpidem. The first-choice off-label drug used to treat insomnia was trazodone. Hypnotics, sedatives and anxiolytics were prescribed 12.6 times more frequently than off-label used drugs. When treating insomnia with an off-label drug, physicians usually prescribed a therapeutic dosage much lower than that recommended in the package insert. Choice of sleep medication and off-label drug were most often influenced by physician specialty. Off-label prescriptions were common but not prevalent. Choice of hypnotic or sedative-anxiolytic was related to how long the drug acted and how much it cost; choice of off-label drug was related to physicians’ familiarity with specific drugs and patients’ characteristics. Concomitant anxiety or depression was significantly associated with higher consumption of hypnotics. Benzodiazepines and newer non-benzodiazepine hypnotics are still the most frequently used drugs for treating insomnia in the elderly in Taiwan. Elderly patients with concomitant anxiety or depression consumed more hypnotics. Further studies conducted over several years are needed to identify trends in the pharmacological treatment of insomnia.

After the implementation of good manufacturing practice in the early 1980s, Taiwan established a relatively effective regulatory process for pharmaceutical registration. The regulatory process for new drug applications (NDAs) in Taiwan, however, relied on foreign-derived documentation, especially the clinical data. The evolution of clinical trial requirements for product registration can be broken down into four phases over the past 15 years. It started with the initiation of postmarketing surveillance (PMS) in the early 1980s to restrict the entry of generic imitations and to provide adequate protection of pharmaceutical innovations. This was followed by a series of legislation or amendments on relevant legal requirements, such as the Patent Law, the Medical Care Act, the Bioavailability and Bioequivalence Guidelines, and the Pharmaceutical Affairs Law. The emerging pipeline protection in the early 1990s resulted in the 1993 Amendment of 1983 PMS that requires domestic clinical data for new drug applications.

Deferasirox (DFX) is an effective and well-tolerated oral iron chelator elevating the adherence to iron chelating therapy among patients with iron overload. However, the US Food and Drug Administration issued a warning about the potential adverse events associated with DFX in 2010.To examine the risks of gastrointestinal (GI) bleeding, acute liver necrosis, and acute renal failure among DFX users compared with desferrioxamine (DFO) users in a real-world setting, first-time users of DFX or DFO between 2005 and 2008 in Taiwan's National Health Insurance database were observed in this population-based retrospective cohort study. The risks of different adverse events were individually analyzed by Cox proportional hazards models and adjusted by age, sex, concomitant medications, and prior medical conditions.Deferasirox users had the highest incidence rates of GI bleeding (2.03 per 10 000 patient-days), acute liver necrosis (0.26 per 10 000 patient-days) and acute renal failure (1.45 per 10 000 patient-days) compared with other iron chelator users. Compared with DFO users, DFX users were not associated with the risk of GI bleeding (adjusted HR 1.03, 95% CI 0.61-1.74, p = 0.90) and the risk of acute liver necrosis (adjusted HR 2.13, 95% CI 0.49-9.33, p = 0.32). The association between DFX use and acute renal failure was found to be statistically significant (HR 2.18, 95% CI 1.18-4.02, p = 0.01; adjusted HR 2.41, 95% CI 1.27-4.58, p = 0.01).In this study, we found statistically significant higher risk of acute renal failure and non-statistically significant higher risk of GI bleeding and acute liver necrosis associated with DFX use. More researches are warranted to evaluate the association between DFX use and potential adverse events.

Objectives. To determine if access to medical services differed by regions and to demonstrate the extent of the differences of adopting a claims-based risk-adjustment system versus a demographic model for regional resource allocation. Methods. The claims of a 1% random sample of Taiwan’s National Health Insurance enrollees (N = 173 175) in 2002 was used. The number of visits and morbidity-adjusted resource consumption were calculated individually then collapsed regionally. Regional expected resource allocation was compared with actual consumption. Results. After controlling for diagnosis-based health measures, the average numbers of visits were stable across regions. Two models were consistent in showing over- or underutilization; the overall difference between two models in resource allocation was 5.8% at the district level. We observed strong urban overutilization and rural underutilization. Conclusions. Access to medical services is similar across regions. The adoption of a diagnosis-based model over a demographic-adjusted budgeting method would affect resource allocation considerably.

Objective To evaluate the outcome of vertebroplasty or kyphoplasty (VK), in comparison with non-VK treatment, among patients hospitalized for first-ever vertebral compression fractures (VCFs). Design A population-based retrospective cohort study. Setting Taiwan’ s National Health Insurance claims data. Participants All individuals aged ≥60 years who were newly discharged after hospitalization for a primary VCF diagnosis. Intervention Percutaneous vertebroplasty or kyphoplasty. Measurements Study outcomes were discharge outcome (re-hospitalization within 1 week, 1 month or 6 months, categorized by diagnosis) and the prescription of anti-osteoporosis medication for secondary fracture prevention. Potential selection bias was adjusted by using propensity score matching to select one conservatively treated patient (e.g. lumbar brace, analgesics, or physical therapy) matched to one patient receiving VK. Results The study cohort consisted of 9238 patients who had been discharged after hospitalization for a first-ever VCF between 2004 and 2007. During the index hospitalization, 1018 patients received VK, compared with 8,220 patients who did not receive VK. Patients receiving percutaneous procedure group had a consistently lower incidence of 7-day re-hospitalization for any of the three outcomes (OR = 0.48; 95% CI: 0.32–0.72). Considering the cause of re-hospitalization separately, the vertebroplasty/kyphoplasty group had a significantly lower risk of 7-day re-hospitalization for fracture-related diagnosis (OR = 0.28, 95% CI: 0.12–0.68) and musculoskeletal diagnosis (OR = 0.08, 95% CI: 0.01–0.88) as well as a significantly lower risk of 1-month re-hospitalization (OR = 0.74; 95% CI: 0.59–0.93). Conclusions VK may protect patients with VCFs from short-term re-hospitalization and a greater need exists for anti-osteoporosis medication as secondary prevention for this at-risk patient group.

Background: A link between the use of alendronate and atypical diaphyseal femoral fracture has been suggested. Objective: The goal of this study was to evaluate the benefits of alendronate in preventing rehospitalization due to hip fractures and whether its use increases risk of hospitalization for atypical diaphyseal femoral fractures in Taiwan. Methods: Using Taiwan’s National Health Insurance database, we identified women with osteoporosis with a first-ever hospitalization for vertebral or hip fractures between 2001 and 2007, which consisted of all patients receiving alendronate, raloxifene, calcitonin salmon, or teriparatide after the index fracture hospitalization. Data of untreated women were obtained as the untreated cohort. Study outcomes were defined as a rehospitalization due to hip fracture or a new hospitalization for subtrochanteric or diaphyseal femoral fracture. Results: Among 11,278 women identified (mean age, 77 years), 2425 (21.5%) received alendronate, 2694 (23.9%) received other antiosteoporosis drugs, and 6159 (54.6%) were untreated. Patients in each group were comparable in fracture history and major comorbidities; untreated patients were more likely to have stroke (11.2%; P 0.01) and those treated with alendronate were more likely to have a history of hyperlipidemia (16.2%; P 0.03). Compared with the untreated patient cohort, our analysis suggested that patients prescribed alendronate were associated with decreased risk of rehospitalization due to hip fracture (hazard ratio 0.67 [95% CI, 0.54‐0.82]). Neither patients prescribed alendronate, nor those prescribed other antiosteoporosis drugs, differed significantly from the untreated patient cohort in terms of risk of hospitalization for atypical femoral fracture (adjusted hazard ratios 0.77 and 0.49 [95% CI, 0.40‐1.47 and 0.22‐1.12], respectively). Consistent with these data, short- or long-term alendronate use was not found to be significantly associated with higher risk of atypical femoral fractures. Conclusions: This study in Taiwanese patients suggests that alendronate use was associated with a reduction in risk of rehospitalization due to hip fracture. We did not find a significant association between alendronateuseandriskofhospitalizationforatypicalfemoral fracture. (Clin Ther. 2011;33:1659‐1667) © 2011

Background Using non-steroidal anti-inflammatory drugs (NSAIDs) as a case, we used Taiwan's National Health Insurance (NHI) database, to empirically explore the association between policy interventions (price regulation, new drug entry, and an information shock) and drug expenditures, utilization, and market structure between 2001 and 2004. Methods All NSAIDs prescribed in ambulatory visits in the NHI system during our study period were included and aggregated quarterly. Segmented regression analysis for interrupted time series was used to examine the associations between two price regulations, two new drug entries (cyclooxygennase-2 inhibitors) and the rofecoxib safety signal and expenditures and utilization of all NSAIDs. Herfindahl index (HHI) was applied to further examine the association between these interventions and market structure of NSAIDs. Results New entry was the only variable that was significantly correlated with changes of expenditures (positive change, p = 0.02) and market structure of the NSAIDs market in the NHI system. The correlation between price regulation (first price regulation, p = 0.62; second price regulation, p = 0.26) and information shock (p = 0.31) and drug expenditure were not statistically significant. There was no significant change in the prescribing volume of NSAIDs per rheumatoid arthritis (RA) or osteoarthritis (OA) ambulatory visit during the observational period. The market share of NSAIDs had also been largely substituted by these new drugs up to 50%, in a three-year period and resulted in a more concentrated market structure (HHI 0.17). Conclusions Our empirical study found that new drug entry was the main driving force behind escalating drug spending, especially by altering the market share.

Thiazolidinediones (TZDs) are widely used antidiabetic agents, but there is great concern and conflicting reports over their possible effect on cardiovascular morbidity, especially in patients with heart failure (HF).Using 2000-2005 Taiwan's National Health Insurance (NHI) claims data, this population-based, retrospective cohort study investigated if there was an association between the cumulative TZD dose and clinical outcomes in type 2 diabetic patients recently hospitalized for HF. Study outcomes were death, first all-cause readmission, and first readmission for HF. Cox proportional hazard models were used to analyze the association between TZD versus sulfonylurea (SU) use and these outcomes.Out of a total of 8139 eligible patients, 224 were taking TZD (65.63% female; mean [SD] age, 68.30[10.60] years) and 7915 were taking SU (55.10% female; 70.30[10.50] years). Patients taking TZD were at higher risk for readmission for HF (59 cases; HR 1.58 (95% confidence interval, 95%CI 1.44-1.72)), all-cause readmission (147 cases; 1.40 (1.30-1.70)), and death (103 cases; 2.23 (1.58-3.14)). The higher the cumulative exposure to TZD, the greater the risk of HF readmission, all-cause readmission, and death.Among diabetic patients who had been hospitalized for HF, TZD users were at significantly greater risk for readmission for HF, all-cause readmission, and death than SU users. The risk of all adverse clinical outcomes appeared to highly relate to cumulative exposure to TZD. These findings provide empirical evidence supporting the latest black box warnings issued by the United States Food and Drug Administration in August, 2007 advising that TZD not be prescribed for diabetic patients with symptomatic heart failure.

To examine the potentially inappropriate prescription of thiazolidinediones (TZD).Data on TZD prescriptions were collected from Taiwan's National Health Insurance dataset from 2001 to 2006. TZDs were considered inappropriately prescribed when they were prescribed to patients who were (1) under 18 years old, (2) pregnant, who had (3) type 1 diabetes, (4) severe heart failure, (5) hepatic insufficiency, or (6) renal insufficiency and taking TZD + metformin in combination. We aggregated potentially inappropriate prescriptions of TZD for each health-care institution in each month starting from March 2001, when TZD was introduced to Taiwan's market.The potentially inappropriate prescription of TZD increased from 9.41% in 2001 to 12.50% in 2006. Prior inappropriate prescription led to a 0.06% (95%CI: 0.04-0.08) further increase in its later inappropriate prescription. Accumulated months of experience prescribing TZD was found associated with higher proportion of inappropriate prescription of TZD (0.03%, 95%CI: 0.01-0.05). However, it was negatively associated with new incidence of inappropriate prescription of TZD (-0.20, 95%CI: -0.22 to -0.18). The greater the volume of prior TZD prescription (-0.87%, 95%CI: -0.93 to -0.81) and the greater the number of accumulated months since adoption (-0.14%, 95%CI: -0.16 to -0.12), the greater the decrease in rates of new inappropriate prescriptions.Along with the quick penetration of the new DM drug came an increased possibility that it would be prescribed inappropriately, a trend that persisted over time.

Objectives Concerns have been raised about bisphosphonate use and risk of atrial fibrillation (AF) in women with osteoporosis. This study compares the risk of AF and of flutter or acute myocardial infarction (AMI) in women with osteoporosis taking alendronate or raloxifene. Methods Using Taiwan's National Health Insurance database to conduct a population-based retrospective cohort study, we reviewed the medical and prescription histories of 27,257 women with osteoporosis (21,037 receiving alendronate and 6,220 receiving raloxifene) between 2001 and 2007. Mean (SD) follow-up was 303.62 (422.87) days. For the main outcome measures, we calculated the adjusted relative risk of AF and AMI using the Cox proportional hazards model, adjusting for various confounders. Results Incidence rates (per patient-year) of AF in the alendronate group (1.00%) and the raloxifene group (1.02%) were similar. Alendronate use was not associated with risk of AF (hazard ratio [HR], 1.06; 95% CI, 0.85-1.32) and AMI (HR, 1.02; 95% CI, 0.86-1.19) compared with raloxifene use. However, alendronate users who had previous cardiovascular events and had taken their medications for more than 1 year were at significantly greater risk of AMI than were the group taking raloxifene (HR, 2.24; 95% CI, 1.07-4.71). Users who received 70 mg of alendronate once a week were at significantly lower risk of AF than were those taking 10 mg daily (HR, 0.56; 95% CI, 0.47-0.68). Conclusions Compared with raloxifene, alendronate did not increase the risk of AF and flutter in women with osteoporosis. Medical history contributed most to the development of AF or AMI in the women who received either raloxifene or alendronate. Long-term treatment with alendronate is not suggested for women with a history of cardiovascular events because they are at increased risk of AMI.