Differential Effects of Metformin on Immune-Mediated and Androgen-Mediated Non-Cancer Skin Diseases in Diabetes Patients: A Retrospective Cohort Study.

BACKGROUND: Metformin's effects on non-cancer skin diseases are rarely investigated. OBJECTIVE: The aim of the study was to investigate immune-mediated (urticaria, allergic contact dermatitis, and psoriasis) and androgen-mediated (acanthosis nigricans, hidradenitis suppurativa, and acne) skin diseases associated with metformin use. METHODS: Metformin initiators (n = 234,585) and non-metformin initiators (n = 125,921) within the initial 12 months of antidiabetic drug prescription during 1999-2009 were followed up until December 31, 2011. Cox regression weighted for propensity score was used to estimate hazard ratios for metformin initiators versus non-metformin initiators in intention-to-treat (ITT) and per-protocol (PP) analyses. RESULTS: For immune-mediated skin diseases, hazard ratios were 0.930 (95% confidence interval: 0.920-0.940) and 0.930 (0.918-0.943) in ITT and PP analyses, respectively, and the hazard ratios for each specific outcome were all significantly below unity. For androgen-mediated skin diseases, the ITT and PP hazard ratios were 1.110 (1.060-1.162) and 0.990 (0.935-1.048), respectively, and all hazard ratios were not significant for each specific outcome except for acne in the ITT analysis (hazard ratio: 1.116, 95% confidence interval: 1.064-1.170). CONCLUSION: Metformin use is associated with a significantly lower risk of immune-mediated skin diseases but lacks a preventive effect on androgen-mediated skin diseases.

Pioglitazone and breast cancer risk in female patients with type 2 diabetes mellitus: a retrospective cohort analysis.

BACKGROUND: Whether pioglitazone may affect breast cancer risk in female diabetes patients is not conclusive and has not been investigated in the Asian populations. METHODS: The reimbursement database of Taiwan's National Health Insurance was used to enroll an unmatched cohort and a propensity score-matched cohort of ever users and never users of pioglitazone in female patients with newly diagnosed type 2 diabetes during 1999-2008. The patients were alive on January 1, 2009 and were followed up for breast cancer incidence until December 31, 2011. Cox regression was used to estimate hazard ratios for ever users and tertiles of cumulative duration of pioglitazone therapy versus never users, and for cumulative duration of pioglitazone therapy treated as a continuous variable. Three models were created for the unmatched cohort and the matched cohort, respectively: 1) without adjustment for covariates; 2) after adjustment for covariates that differed with statistical significance (P-value < 0.05) between ever users and never users; and 3) after adjustment for all covariates. RESULTS: There were 174,233 never users and 6926 ever users in the unmatched cohort; and 6926 never users and 6926 ever users in the matched cohort. After a median follow-up of 2.8 years, the numbers of incident breast cancer were 1044 in never users and 35 in ever users in the unmatched cohort and were 41 and 35, respectively, in the matched cohort. Hazard ratios suggested a null association between pioglitazone and breast cancer in all three models in either the unmatched cohort or the matched cohort. The overall hazard ratio after adjustment for all covariates was 0.758 (95% confidence interval: 0.539-1.065) in the unmatched cohort and was 0.824 (95% confidence interval: 0.524-1.296) in the matched cohort. None of the hazard ratios for the tertiles of cumulative duration of pioglitazone therapy and for the cumulative duration being treated as a continuous variable were statistically significant. CONCLUSIONS: This study suggests a null association between pioglitazone and breast cancer risk in female patients with type 2 diabetes mellitus. However, because of the small breast cancer cases and the limited follow-up time, further studies are warranted to confirm our findings.

Metformin reduces risk of varicose veins in patients with type 2 diabetes.

AIM: This population-based retrospective cohort study compared the incidence of varicose veins in an unmatched cohort and a cohort of 1:1 propensity score (PS)-matched pairs of ever and never users of metformin in type 2 diabetes patients. METHODS: Patients with new-onset type 2 diabetes during 1999 to 2005 were enrolled from Taiwan's National Health Insurance and followed until December 31, 2011. Analyses were conducted in an unmatched cohort of 123 710 ever users and 15 095 never users and in 15 088 PS-matched pairs of ever users and never users. Hazard ratios were estimated by Cox proportional hazards model incorporated with the inverse probability of treatment weighting using the PS. RESULTS: New-onset varicose veins were diagnosed in 126 never users and 633 ever users in the unmatched cohort and in 126 never users and 80 ever users in the matched cohort. The respective incidences were 191.36 and 110.04 per 100 000 person-years in the unmatched cohort and 191.41 and 115.81 per 100 000 person-years in the matched cohort. The hazard ratio for ever versus never users in the unmatched cohort was 0.57 (95% confidence interval, 0.47-0.69) and was 0.60 (0.45-0.80) for the matched cohort. In the unmatched cohort, the hazard ratios for the first, second, and third tertiles of cumulative duration were 1.03 (0.83-1.28), 0.55 (0.44-0.69), and 0.29 (0.23-0.37), respectively. The respective hazard ratios in the matched cohort were 0.97 (0.65-1.43), 0.79 (0.55-1.15), and 0.24 (0.13-0.42). CONCLUSION: Metformin use is associated with a lower risk of varicose veins in patients with type 2 diabetes.

Metformin and risk of hepatocellular carcinoma in patients with type 2 diabetes.

BACKGROUND: Whether metformin may reduce hepatocellular carcinoma (HCC) risk requires confirmation. METHODS: Type 2 diabetes patients newly diagnosed during 1999-2005 and with 2 or more prescriptions of antidiabetic drugs were enrolled from the Taiwan's National Health Insurance database. A total of 173 917 ever-users and 21 900 never-users of metformin were identified (unmatched cohort). A 1:1 matched-pair cohort of 21 900 ever-users and 21 900 never-users based on a propensity score (PS) was created. Hazard ratios were estimated by Cox regression incorporated with the inverse probability of treatment weighting using the PS. In addition, interactions with aspirin and statin were evaluated. RESULTS: In the unmatched cohort, 619 never-users and 2642 ever-users developed HCC, with a respective incidence of 668.0 and 330.7 per 100 000 person-years and an overall hazard ratio of 0.49 (95% confidence interval: 0.45-0.54). The hazard ratios for the first (<25.7 months), second (25.7-56.9 months) and third (>56.9 months) tertile of cumulative duration of metformin therapy were 0.89 (0.81-0.98), 0.50 (0.46-0.56) and 0.23 (0.21-0.26) respectively. Analyses of the matched cohort showed an overall hazard ratio of 0.76 (0.67-0.85), and the hazard ratios for the respective tertiles were 1.39 (1.19-1.62), 0.77 (0.65-0.91) and 0.37 (0.30-0.45). Aspirin and statin were observed to have a significant interaction with metformin. CONCLUSIONS: Metformin was associated with a reduced risk of HCC in a dose-response pattern. Users of both metformin and aspirin or metformin and statin had the lowest risk.

Metformin is associated with decreased skin cancer risk in Taiwanese patients with type 2 diabetes.

BACKGROUND: Metformin, an antidiabetic drug, is associated with decreased cancer risk, but its effect on skin cancer is unknown. OBJECTIVE: To evaluate skin cancer risk associated with metformin use. METHODS: In total, 16,237 matched pairs of ever and never metformin users with new-onset type 2 diabetes diagnosed during 1999-2005 were retrospectively enrolled and followed until December 31, 2011, using Taiwan's National Health Insurance database. Hazard ratios (HRs) were estimated using Cox regression weighted for propensity scores. RESULTS: Skin cancer incidence was 45.59 and 83.90 per 100,000 person-years among ever and never users, respectively (HR 0.540, 95% confidence interval [CI] 0.357-0.819). Among ever users, the HRs (95% CIs) for the first (<21.00 months), second (21.00-45.83 months), and third (>45.83 months) cumulative duration tertiles were 0.817 (0.448-1.489), 0.844 (0.504-1.412), and 0.114 (0.036-0.364), respectively, and the HRs (95% CIs) for the first, second, and third cumulative dose tertiles were 1.006 (0.579-1.748), 0.578 (0.317-1.051), and 0.229 (0.099-0.530), respectively. HRs (95% CIs) were 0.523 (0.175-1.562) for melanoma and 0.496 (0.319-0.772) for nonmelanoma skin cancer. LIMITATIONS: Few patients had skin cancer and information on ultraviolet light exposure and tumor histopathology was lacking. CONCLUSION: Metformin use is associated with a decreased skin cancer risk.

Sitagliptin and pancreatic cancer risk in patients with type 2 diabetes.

BACKGROUND: The risk of pancreatic cancer associated with incretin-based therapies is controversial. METHODS: This study retrospectively analysed the National Health Insurance database including patients with newly diagnosed type 2 diabetes mellitus at an age ≥ 25 years between 1999 and 2010. A total of 71 137 ever users of sitagliptin and 933 046 never users were followed for pancreatic cancer until 31 December 2011. A time-dependent approach was used to calculate incidence and estimate hazard ratios adjusted for propensity score using Cox regression. RESULTS: During follow-up, 83 ever users and 3658 never users developed pancreatic cancer, representing an incidence of 73·6 and 55·0 per 100 000 person-years, respectively. The adjusted hazard ratio (95% confidence intervals) for ever versus never users was 1·40 (1·13-1·75). The respective adjusted hazard ratio for the first, second and third tertile of cumulative dose < 14 700, 14 700-33 700 and > 33 700 mg was 1·83 (1·28-2·62), 1·97 (1·41-2·76) and 0·72 (0·45-1·15). For average daily dose of < 50, 50-99·9 and ≥ 100 mg, the respective hazard ratio was 3·10 (1·17-8·26), 1·01 (0·63-1·61) and 1·53 (1·18-1·97). CONCLUSIONS: Sitagliptin is significantly associated with a higher risk of pancreatic cancer, especially when the cumulative dose is < 33 700 mg. The risk diminished in users with a higher cumulative dose. The daily dose of sitagliptin should better be kept < 100 mg, and its use should be reconsidered in patients who suffer from severe renal impairment and thus a daily dose of < 50 mg is always recommended. Future studies are required to confirm the findings with more appropriate adjustment for smoking.

Metformin reduces ovarian cancer risk in Taiwanese women with type 2 diabetes mellitus.

BACKGROUND: Whether metformin therapy affects ovarian cancer risk in Asian patients with type 2 diabetes mellitus has not been investigated. METHODS: Data analysis was performed in 2014. The reimbursement databases of Taiwanese female patients with a new diagnosis of type 2 diabetes mellitus between 1998 and 2002 (n = 479,475) were retrieved from the National Health Insurance for follow-up of ovarian cancer until the end of 2009. Metformin was treated as a time-dependent variable; and of these patients, 286,106 were never-users, and 193,369 were ever-users. A time-dependent approach was used to calculate ovarian cancer incidence and estimate hazard ratios by Cox regression for never-users (as referent group), ever-users and subgroups of metformin exposure (tertiles of cumulative duration and cumulative dose). RESULTS: During follow-up, 601 metformin ever-users and 2600 never-users developed ovarian cancer, representing an incidence of 49.4 and 146.4 per 100,000 person-years, respectively. The overall fully adjusted hazard ratio (95% confidence intervals) for ever-users versus never-users was 0.658 (0.593-0.730). The fully adjusted hazard ratios for the first, second and third tertiles of cumulative duration of metformin therapy were 1.169 (1.019-1.341), 0.761 (0.644-0.898) and 0.276 (0.225-0.340), respectively (p trend < 0.01) and 1.220 (1.067-1.395), 0.610 (0.513-0.725) and 0.305 (0.248-0.374), respectively (p trend < 0.01), for a cumulative dose of metformin. In additional analyses, sulfonylureas but not the other antidiabetic drugs were associated with a reduced risk of ovarian cancer. CONCLUSIONS: Metformin use is associated with a decreased risk of ovarian cancer.

Prolonged use of human insulin increases breast cancer risk in Taiwanese women with type 2 diabetes.

BACKGROUND: Human insulin is commonly used to treat hyperglycemia in patients with diabetes, but its potential link with female breast cancer is under debate. This study investigated whether human insulin use might be associated with breast cancer risk in Taiwanese women with type 2 diabetes. METHODS: The reimbursement databases of all Taiwanese diabetic patients from 1996 to 2009 were retrieved from the National Health Insurance. An entry date was set at 1 January 2004 and a total of 482,033 women with type 2 diabetes were followed up for breast cancer incidence until the end of 2009. Incidences for ever-users, never-users and subgroups of human insulin exposure (using tertile cutoffs of time since starting insulin, cumulative dose and cumulative duration of insulin) were calculated and the adjusted hazard ratios were estimated by Cox regression. The potential risk modification by concomitant treatment with metformin, statin and angiotensin converting enzyme inhibitor/angiotensin receptor blocker (ACEI/ARB) was also evaluated. RESULTS: There were 59,798 ever-users and 422,235 never-users of human insulin, with respective numbers of incident breast cancer of 559 (0.93 %) and 4,711 (1.12 %), and respective incidence of 207.9 and 215.1 per 100,000 person-years. The overall adjusted hazard ratio (95 % confidence interval) did not show a significant association with insulin [1.033 (0.936-1.139)]. However, patients in the third tertiles of dose-response parameters might show a significantly higher risk of breast cancer while compared to never-users: 1.185 (1.026-1.368), 1.260 (1.096-1.450) and 1.257 (1.094-1.446) for ≥67 months for time since starting insulin, ≥39,000 units for cumulative dose of insulin, and ≥21.8 months for cumulative duration of insulin, respectively. Additional analyses suggested that the breast cancer risk associated with human insulin use might be beneficially modified by concomitant use of metformin, statin and ACEI/ARB. CONCLUSIONS: This study discloses a significantly higher risk of breast cancer associated with prolonged use of human insulin. The increased risk of breast cancer associated with human insulin use may be modified by medications such as metformin, statin and ACEI/ARB.

Rosiglitazone may reduce non-melanoma skin cancer risk in Taiwanese.

BACKGROUND: Whether rosiglitazone may affect the risk of non-melanoma skin cancer (NMSC) has not been investigated. METHODS: The reimbursement databases of all Taiwanese diabetic patients from 1996 to 2009 were retrieved from the National Health Insurance. An entry date was set at 1 January 2006 and a total of 886418 patients with type 2 diabetes were followed up for NMSC incidence until the end of 2009. Incidences for ever-users, never-users and subgroups of rosiglitazone exposure (using tertile cutoffs of duration of therapy and cumulative dose) were calculated and hazard ratios estimated by Cox regression. Additional models were created as sensitivity analyses. RESULTS: There were 103097 ever-users and 783321 never-users, respective numbers of incident NMSC 250 (0.24%) and 2084 (0.27%), and respective incidence 68.90 and 76.77 per 100000 person-years. Although the overall hazard ratio was not significant in the unadjusted, age-sex-adjusted or fully adjusted model, the risk was significantly lower in the third tertile of duration of therapy and cumulative dose, with significant P for trends. The fully adjusted hazard ratio (95% confidence interval) for a duration of therapy >13.77 months and a cumulative dose of >1752 mg was 0.723 (0.566, 0.923) and 0.783 (0.618, 0.993), respectively. The findings were supported by various sensitivity analyses. CONCLUSIONS: Rosiglitazone may reduce the risk of NMSC, but further confirmation is required.

An updated review on cancer risk associated with incretin mimetics and enhancers.

Incretin-based therapies, including the use of incretin mimetics of glucagon-like peptide-1 receptor (GLP-1R) agonists and incretin enhancers of dipeptidyl-peptidase 4 (DPP-4) inhibitors, are widely used by clinicians for glucose lowering in patients with type 2 diabetes mellitus. These agents have benefits of a lower risk of hypoglycemia, being neutral for body weight for DPP-4 inhibitors and having a potential for weight reduction with GLP-1R agonists. They may also have a neutral or beneficial cardiovascular effect. Despite these benefits, an increased risk of cancer (especially pancreatic cancer and thyroid cancer) associated with incretin-based therapies has been reported. In this article, we reviewed related literature of experimental animal and observational human studies, clinical trials, and meta-analyses published until December 15, 2014. Current studies suggested a probable role of GLP-1R activation on the development of pancreatic cancer and thyroid cancer in rodents, but such an effect in humans is not remarkable due to the lower or lack of expression of GLP-1R on human pancreatic ductal cells and thyroid tissues. Findings in human studies are controversial and inconclusive. In the analyses of the US Food and Drug Administration adverse events reporting system, a significantly higher risk of pancreatic cancer was observed for GLP-1R agonists and DPP-4 inhibitors, but a significantly higher risk of thyroid cancer was only observed for GLP-1R agonists. Such a higher risk of pancreatic cancer or thyroid cancer could not be similarly demonstrated in other human observational studies or analyses of data from clinical trials. With regards to cancers other than pancreatic cancer and thyroid cancer, available studies supported a neutral association in humans. Some preliminary studies even suggested a potentially beneficial effect on the development of other cancers with the use of incretins. Based on current evidence, continuous monitoring of the cancer issues related to incretin-based therapies is required, even though the benefits may outweigh the potential cancer risk in the general patients with type 2 diabetes mellitus.

Metformin and endometrial cancer risk in Chinese women with type 2 diabetes mellitus in Taiwan.

OBJECTIVE: To evaluate metformin effects on endometrial cancer risk in Chinese female patients with type 2 diabetes mellitus (T2DM) in Taiwan. METHODS: This is a retrospective cohort analysis using the National Health Insurance database of Taiwan. Female patients with newly diagnosed T2DM and without endometrial cancer in 1998-2002 were followed to end of 2009 (n=478,921). Among them, 285,916 were never-users and 193,005 were ever-users of metformin. A time-dependent approach was used to calculate endometrial cancer incidence and estimate hazard ratios by Cox regression for ever-users, never-users, and subgroups of metformin exposure (tertiles of cumulative duration and cumulative dose). Sensitivity analyses were conducted in various subgroups. RESULTS: During follow-up, 728 metformin ever-users and 2157 never-users developed endometrial cancer, representing an incidence of 60.00 and 121.69 per 100,000 person-years, respectively. The overall hazard ratio (95% confidence intervals) for ever- versus never-users after adjustment for propensity score (PS) was 0.675 (0.614-0.742). The PS-adjusted hazard ratios for the first, second, and third tertiles of cumulative duration of metformin therapy were 1.089 (0.966-1.228), 0.707 (0.616-0.812) and 0.313 (0.262-0.374), respectively (P-trend<0.0001); and 1.062 (0.942-1.197), 0.620 (0.538-0.715) and 0.376 (0.317-0.447), respectively (P-trend<0.0001), for cumulative dose of metformin. The dose-response relationship was demonstrated in various models and an overall reduced risk was consistently supported by sensitivity analyses. CONCLUSIONS: The use of metformin in women with T2DM was associated with an overall significantly lower risk of endometrial cancer with dose-response relationship.

Metformin may reduce breast cancer risk in Taiwanese women with type 2 diabetes.

Whether metformin therapy affects breast cancer risk in Asian patients with type 2 diabetes mellitus (T2DM) has not been investigated. The reimbursement databases of Taiwanese female patients with a new diagnosis of T2DM between 1998 and 2002 (n = 476,282) were retrieved from the National Health Insurance for follow-up of breast cancer until the end of 2009. Metformin was treated as a time-dependent variable; and of these patients, 285,087 were never-users and 191,195 were ever-users. A time-dependent approach was used to calculate breast cancer incidence and estimate hazard ratios by Cox regression for ever-users, never-users, and subgroups of metformin exposure (tertiles of cumulative duration and cumulative dose). During follow-up, 2,412 (1.26 %) metformin ever-users and 9,322 (2.10 %) never-users developed breast cancer, representing an incidence of 201.08 and 535.88 per 100,000 person-years, respectively. The overall multivariable-adjusted hazard ratio (95 % confidence intervals) for ever- versus never-users was 0.630 (0.597-0.665). The multivariable-adjusted hazard ratios for the first, second, and third tertiles of cumulative duration of metformin therapy were 1.122 (1.043-1.207), 0.754 (0.692-0.820), and 0.280 (0.253-0.310), respectively, (P-trend <0.0001); and 1.099 (1.021-1.182), 0.664 (0.611-0.723), and 0.311 (0.281-0.344), respectively, (P-trend <0.0001), for cumulative dose of metformin. Metformin use is associated with a decreased risk of breast cancer.

Diabetes and breast cancer in Taiwanese women: a detection bias?

PURPOSE: To evaluate whether diabetes is a risk factor for breast cancer considering confounders and potential detection examinations. METHODS: National Health Insurance data on 501,747 women without breast cancer were retrieved. Three-year cumulative incidence (2003-2005) and risk ratios (RRs) between diabetic and nondiabetic women were calculated. Potential detection examinations were compared between diabetic and nondiabetic women by chi-square test. Odds ratios (ORs) were estimated by logistic regression for diabetes status/duration with and without adjustment for potential detection examinations and confounders. RESULTS: The crude RR (95% confidence interval [CI]) for all ages, and age groups < 50, 50-64 and ≥ 65 years, was 2·62 (2·31-2·91), 2·69 (2·11-3·44), 1·39 (1·15-1·68) and 1·37 (1·03-1·84), respectively. Patients with diabetes more frequently received potential detection examinations than nondiabetes (17·5% vs. 7·4%, P-value < 0·001). The unadjusted OR (95% CI) for breast cancer for diabetes status (yes vs. no) was 2·63 (2·31-2·98) and was significant for any diabetes duration. The OR for diabetes status was 1·81 (95% CI: 1·59-2·06) after adjustment for potential detection examinations. In models adjusted for potential detection examinations, age, living region, occupation, comorbidities and used medications, OR for diabetes status attenuated to 1·13 (95% CI 0·96-1·32, P-value = 0·14) and none was significant for any diabetes duration. Potential detection examinations were associated with a fivefold to sevenfold higher risk in various models, indicating a strong impact of detection bias. CONCLUSIONS: An association between diabetes and breast cancer is observed, but this can be due to potential detection bias and confounders.

Treatment with human insulin does not increase thyroid cancer risk in patients with type 2 diabetes.

BACKGROUND: Whether human insulin therapy may increase thyroid cancer risk in patients with type 2 diabetes mellitus (T2DM) has not been investigated. MATERIALS AND METHODS: The reimbursement databases of all Taiwanese diabetic patients from 1996 to 2009 were retrieved from the Bureau of National Health Insurance. The entry date was set at 1 January 2004, and 968,384 patients with T2DM were followed up for thyroid cancer incidence until the end of 2009. Ever-users, never-users and subgroups of human insulin exposure (using tertile cut-offs of time since starting insulin, duration of therapy and cumulative dose) at entry date were calculated for thyroid cancer incidence. Insulin glargine was not marketed until after the entry date. Therefore, to exclude the potential contamination of insulin glargine, patients who happened to use insulin glargine were censored at the time of its initiation when calculating the period of follow-up. Hazard ratios were estimated by Cox regression. RESULTS: There were 111,121 ever-users and 857,263 never-users of human insulin, with respective numbers of incident thyroid cancer of 118 (0·11%) and 1047 (0·12%), and respective incidences of 23·9 and 23·8 per 100,000 person-years. The overall hazard ratios (95% confidence intervals) did not show a significant association with human insulin in either the age-sex-adjusted or the fully adjusted model: 0·942 (0·778-1·141) and 1·096 (0·888-1·353), respectively. When categorized into tertiles of the dose-response parameters, none of the hazard ratios was significant. CONCLUSIONS: This study does not support the role of human insulin therapy in increasing the risk of thyroid cancer in patients with T2DM.

A review on thiazolidinediones and bladder cancer in human studies.

There is a concern of an increased risk of bladder cancer associated with the use of thiazolidinediones, a class of oral glucose-lowering drugs commonly used in patients with type 2 diabetes with a mechanism of improving insulin resistance. Human studies on related issues are reviewed, followed by a discussion on potential concerns on the causal inference in current studies. Pioglitazone and rosiglitazone are discussed separately, and findings from different geographical regions are presented. Randomized controlled trials designed for primarily answering such a cancer link are lacking, and evidence from clinical trials with available data for evaluating the association may not be informative. Observational studies have been reported with the use of population-based administrative databases, single-hospital records, drug adverse event reporting system, and case series collection. Meta-analysis has also been performed by six different groups of investigators. These studies showed a signal of higher risk of bladder cancer associated with pioglitazone, especially at a higher cumulative dose or after prolonged exposure; however, a weaker signal or null association is observed with rosiglitazone. In addition, there are some concerns on the causal inference, which may be related to the use of secondary databases, biases in sampling, differential detection, and confounding by indications. Lack of full control of smoking and potential biases related to study designs and statistical approaches such as prevalent user bias and immortal time bias may be major limitations in some studies. Overlapping populations and opposing conclusions in studies using the same databases may be of concern and weaken the reported conclusions of the studies. Because randomized controlled trials are expensive and unethical in providing an answer to this cancer issue, observational studies are expected to be the main source in providing an answer in the future. Furthermore, international comparison studies using well-designed and uniform methodology to clarify the risk in specific sexes, ethnicities, and other subgroups and to evaluate the interaction with other environmental risk factors or medications will be helpful to identify patients at risk.

Continuity of care, medication adherence, and health care outcomes among patients with newly diagnosed type 2 diabetes: a longitudinal analysis.

BACKGROUND: The effects of continuity of care (COC) on health care outcomes are well established. However, the mechanism of this association is not fully understood. OBJECTIVE: The objective of this study was to examine the relationship between COC and medication adherence, as well as to investigate the mediating effect of medication adherence on the association between COC and health care outcomes, in patients with newly diagnosed type 2 diabetes. RESEARCH DESIGN AND SUBJECTS: This study utilized a longitudinal design and included a 7-year follow-up period from 2002 to 2009 under a universal health insurance program in Taiwan. Patients aged 18 years or older who were first diagnosed with type 2 diabetes in 2002 were included in the study. Random intercept models were conducted to assess the temporal relationship between COC, medication adherence, and health care outcomes. RESULTS: Patients with high or intermediate COC scores were more likely to be adherent to medications than those with low COC scores [odds ratio (OR), 3.37; 95% confidence interval (CI), 3.15-3.60 and OR, 1.84; 95% CI, 1.74-1.94, respectively]. In addition, the association between COC and health care outcomes was partly mediated by better medication adherence in patients with newly diagnosed type 2 diabetes. CONCLUSIONS: Improving the COC for patients with type 2 diabetes may result in higher medication adherence and better health care outcomes.

Diabetes and thyroid cancer mortality: a 12-year prospective follow-up of Taiwanese.

BACKGROUND: The association between diabetes and thyroid cancer is rarely studied. This study evaluated thyroid cancer mortality trend in Taiwanese population, mortality rate ratios between diabetic patients and general population, and risk factors in diabetic patients. METHODS: In general population, age-standardized trends were evaluated from 1995 to 2006. A total of 113,347 diabetic men and 131,573 diabetic women aged ≥ 25 years recruited during 1995-1998 were followed to 2006. Age- and sex-specific mortality rate ratios were calculated and Cox's regression evaluated the risk factors. RESULTS: A steady trend of thyroid cancer mortality was observed in the general population. A total of 20 diabetic men and 45 diabetic women died of thyroid cancer, with overall mortality rate 2.32 and 4.26 per 100,000 person-years, respectively. Mortality rate ratios showed positive association with magnitude increased with decreasing age: 1.85 (0.77, 4.43), 1.21 (0.54, 2.73), 2.53 (1.14, 5.59) and 5.80 (2.10, 16.01) for ≥ 75, 65-74, 55-64 and 25-54 years old, respectively, for men; and 0.78 (0.35, 1.74), 2.03 (1.31, 3.13), 2.99 (1.77, 5.04) and 5.34 (2.20, 13.00), respectively, for women. After adjustment, only age was significantly associated with thyroid cancer mortality. Sex, diabetes duration, diabetes type, body mass index, smoking, insulin use and area of residence were not significantly predictive for thyroid cancer mortality. CONCLUSIONS: The annual thyroid cancer mortality during 1995-2006 in the Taiwanese general population has been steady. Our data suggest a higher risk in diabetic patients, with especially higher mortality rate ratios in younger age. Obesity, smoking and insulin use are not modifiable risk factor.

Diabetes is not an independent risk factor for hepatocellular carcinoma.

BACKGROUND: It is controversial whether diabetes is a risk factor for hepatocellular carcinoma (HCC) in countries with high prevalence of viral hepatitis such as Taiwan. METHODS: From the Taiwanese National Health Insurance, 1 000 000 insurants were randomly selected for the study. Among them, 494 080 men and 502 841 women without HCC were followed from 2003 until 2005. Relative risks were estimated taking into account alcohol-related diagnoses, chronic liver disease, potential HCC detection examinations and other covariates. RESULTS: The diabetes patients were more likely to receive examinations of abdominal sonography, HBsAg, anti-HCV and alpha-fetoprotein. Multivariable-adjusted relative risk (95% confidence interval) was not significant for diabetes as a risk factor for HCC in either men or women: 0.932 (0.788-1.101) for men and 1.158 (0.968-1.386) for women. While comparing diabetes duration <1, 1-3 and ≥ 3 years to non-diabetes individuals, the respective relative risks were 1.043 (0.751-1.448), 0.945 (0.738-1.209) and 0.901 (0.744-1.091) for men; and were 0.950 (0.610-1.480), 1.169 (0.884-1.546) and 1.189 (0.974-1.452) for women. For both sexes, age, alcohol-related diagnoses, chronic liver disease and potential detection bias were significantly associated with higher risk, but dyslipidemia and stroke a lower risk. In addition, chronic obstructive pulmonary disease was associated with a lower risk and sulfonylurea a higher risk in men. Models that did not consider alcohol-related diagnoses, chronic liver disease and potential detection bias would result in spurious association between diabetes and HCC. CONCLUSIONS: Diabetes is not an independent risk factor for HCC after considering the effects of alcohol-related diagnoses, chronic liver disease and potential detection bias.

Benign prostatic hyperplasia is a significant risk factor for bladder cancer in diabetic patients: a population-based cohort study using the National Health Insurance in Taiwan.

BACKGROUND: Diabetic patients have a higher risk of bladder cancer and benign prostatic hyperplasia (BPH). Theoretically, BPH patients may have an increased risk of bladder cancer because residual urine in the bladder surely increases the contact time between urinary excreted carcinogens and the urothelium. However, whether BPH increases bladder cancer risk in patients with type 2 diabetes has not been studied. METHODS: The reimbursement databases of all Taiwanese diabetic patients under oral anti-diabetic agents or insulin from 1996 to 2009 were retrieved from the National Health Insurance. An entry date was set at 1 January 2006 and a total of 547584 men with type 2 diabetes were followed up for bladder cancer incidence until the end of 2009. Incidences of bladder cancer for BPH by status and by duration were calculated and adjusted hazard ratios (95% confidence intervals) were estimated by Cox regression. The effects of diabetes duration and medications used for diabetic control in relation with bladder cancer risk were also evaluated by Cox regression in BPH men. RESULTS: The incidences were 258.77 and 69.34 per 100,000 person-years for patients with and without BPH, respectively, adjusted hazard ratio 1.794 (1.572, 2.047). For BPH patients, those who underwent surgical procedures for BPH had a higher incidence than those who did not (355.45 vs. 250.09 per 100,000 person-years), respective adjusted hazard ratios: 2.459 (1.946, 3.109) and 1.709 (1.492, 1.958). The significantly higher risk could be demonstrated for BPH of any duration: respective adjusted hazard ratios 1.750 (1.430, 1.605), 1.844 (1.543, 2.203), 2.011 (1.680, 2.406) and 1.605 (1.341, 1.921) for BPH <1, 1-3, 3-5 and ≥ 5 years versus patients without BPH. Sensitivity analyses for patients aged ≥ 60 years and after excluding BPH patients with surgical procedures or without surgical procedures, respectively, yielded similar results. In BPH men, diabetes duration was not significantly related with bladder cancer; but metformin was consistently associated with a significantly lower risk, with adjusted hazard ratio of 0.719 (0.590, 0.875) for all ages and 0.742 (0.604, 0.912) for age ≥ 60 years. CONCLUSIONS: BPH is a significant risk factor for bladder cancer in men with type 2 diabetes. Metformin may protect against bladder cancer in BPH men.

Pioglitazone does not affect the risk of ovarian cancer: analysis of a nationwide reimbursement database in Taiwan.

BACKGROUND: The association between pioglitazone and ovarian cancer has not been studied. METHODS: The reimbursement databases of all Taiwanese patients with a diagnosis of diabetes and under oral anti-diabetic agents or insulin from 1996 to 2009 were retrieved from the National Health Insurance. An entry date was set at 1 January 2006 and a total of 546,632 female patients with type 2 diabetes were followed up for ovarian cancer incidence until the end of 2009. Incidences for ever-users, never-users and subgroups of pioglitazone exposure [using cutoffs of the Kaiser Permanente Northern California study and tertile cutoffs derived from the databases] were calculated and the hazard ratios were estimated by Cox regression in unadjusted, age-adjusted and fully adjusted models. RESULTS: There were 30,783 ever-users and 515,849 never-users, with respective numbers of incident ovarian cancer of 49 (0.16%) and 946 (0.18%), and respective incidence of 43.08 and 51.47 per 100,000 person-years. The overall hazard ratios (95% confidence intervals) in unadjusted, age-adjusted and fully adjusted models were 0.822 (0.616-1.095), 0.823 (0.617-1.097) and 0.968 (0.718-1.305), respectively. In the dose-response analyses, none of the categories showed a significant hazard ratio, and all P-trends were >0.05 without statistical significance. CONCLUSIONS: This study does not support a positive or negative association between pioglitazone use and ovarian cancer in female patients with type 2 diabetes.