Addressing Limitations in Observational Studies of the Association Between Glucose-Lowering Medications and All-Cause Mortality: A Review

Importance of accounting for timing of time-varying exposures in association studies: Hydrochlorothiazide and non-melanoma skin cancer

PURPOSE

Hydrochlorothiazide (HCTZ), a widely prescribed antihypertensive drug with photosensitising properties, has been linked with non-melanoma skin cancer (NMSC) risk. However, previous analyses did not fully explore if and how the impact of past HCTZ exposures accumulates with prolonged use and/or depends on time elapsed since exposures. Therefore, we used different models to more comprehensively assess how NMSC risk vary with HCTZ exposure, and explore how the results may depend on modeling strategies.

METHODS

We used different parametric models with alternative time-varying exposure metrics, and the flexible weighted cumulative exposure model (WCE) to estimate associations between HCTZ exposures and NMSC risk in a population-based cohort of HCTZ users over 65 years old, in the province of Ontario, Canada.

RESULTS

Among 3844 HCTZ users, 273 developed NMSC during up to 8 years of follow-up. In parametric models, based on all exposures, increased duration of past HCTZ use was associated with an increase of NMSC risk but cumulative dose showed no systematic association. Yet, WCE results suggested that only exposures taken 2.5-4 years in the past were associated with the current NMSC hazard. This finding led us to re-define the parametric models, which also confirmed that any HCTZ dose taken outside this time-window were not systematically associated with NMSC incidence.

CONCLUSIONS

Our analyses illustrate how flexible modeling may yield new insights into complex temporal relationships between a time-varying drug exposure and risks of adverse events. Duration and recency of antihypertensive agents exposures must be taken into account in evaluating risk and benefits.

A Road Map for Peer Review of Real-World Evidence Studies on Safety and Effectiveness of Treatments

The growing acceptance of real-world evidence (RWE) in clinical and regulatory decision-making, coupled with increasing availability of health care data and advances in automated analytic approaches, has contributed to a marked expansion of RWE studies of diabetes and other diseases. However, a recent spate of high-profile retractions highlights the need for improvements in the conduct of RWE research as well as in the associated peer review and editorial processes. We review best pharmacoepidemiologic practices and common pitfalls regarding design, measurement, analysis, data validity, appropriateness, and generalizability of RWE studies. To enhance RWE study assessments, we propose that journal editors require 1) study authors to complete RECORD-PE, a reporting guideline for pharmacoepidemiological studies on routinely collected data, 2) availability of predetermined study protocols and analysis plans, 3) inclusion of pharmacoepidemiologists on the peer review team, and 4) provision of detail on data provenance, characterization, and custodianship to facilitate assessment of the data source. We recognize that none of these steps guarantees a high-quality research study. Collectively, however, they permit an informed assessment of whether the study was adequately designed and conducted and whether the data source used was fit for purpose.

Reproducibility of real-world evidence studies using clinical practice data to inform regulatory and coverage decisions

Studies that generate real-world evidence on the effects of medical products through analysis of digital data collected in clinical practice provide key insights for regulators, payers, and other healthcare decision-makers. Ensuring reproducibility of such findings is fundamental to effective evidence-based decision-making. We reproduce results for 150 studies published in peer-reviewed journals using the same healthcare databases as original investigators and evaluate the completeness of reporting for 250. Original and reproduction effect sizes were positively correlated (Pearson's correlation = 0.85), a strong relationship with some room for improvement. The median and interquartile range for the relative magnitude of effect (e.g., hazard ratiooriginal/hazard ratioreproduction) is 1.0 [0.9, 1.1], range [0.3, 2.1]. While the majority of results are closely reproduced, a subset are not. The latter can be explained by incomplete reporting and updated data. Greater methodological transparency aligned with new guidance may further improve reproducibility and validity assessment, thus facilitating evidence-based decision-making. Study registration number: EUPAS19636.

Evolving channeling in prescribing SGLT-2 inhibitors as first-line treatment for type 2 diabetes

PURPOSE

Sodium-glucose cotransporter-2 inhibitors (SGLT-2i) are increasingly being considered as first-line treatment for type 2 diabetes (T2D). The benefits of SGLT-2i from cardiovascular outcome trials may lead to preferential prescribing of SGLT-2i to patients at high cardiovascular risk, possibly causing confounding in non-randomized studies of SGLT-2i as first-line treatment. We assessed evolving imbalances in characteristics of patients starting SGLT-2i versus metformin as first-line monotherapy.

METHODS

Using claims data from two US commercial health insurance and Medicare, we identified patients with T2D aged ≥18 years (>65 years in Medicare) initiating first-line SGLT-2i or metformin from 2013 through 2019. Standardized differences (SDs) for patient characteristics were assessed during four consecutive calendar time blocks (T1:4/2013-12/2014; T2:1/2015-6/2016; T3:7/2016-12/2017; and T4:1/2018-12/2019). We also estimated the propensity score of receiving SGLT-2i versus metformin within each time block and evaluated time trends in model discrimination with c-statistics.

RESULTS

We identified 9113 initiators of first-line SGLT-2i and 810 348 initiators of first-line metformin. During T1, SGLT-2i initiators were younger (SD = -0.24) and less likely to have seen cardiologists (-0.07) with a similar prevalence of CVD (0.04) compared with metformin. During T4, patients were more balanced for age (-0.01). Cardiologist visits (0.08) and CVD (0.25) became more prevalent among SGLT-2i initiators.

CONCLUSIONS

When comparing initiators of first-line SGLT-2i versus metformin, imbalances in patient characteristics evolved from 2013 through 2019, particularly channeling SGLT-2i to individuals at high cardiovascular risk. Evolving channeling in prescribing first-line SGLT-2i should be expected and accounted for in non-randomized comparative effectiveness research.

Role of Metformin in Morbidity and Mortality Associated with Urinary Tract Infections in Patients with Type 2 Diabetes

We conducted this study to compare the morbidity and mortality associated with UTI and sepsis, between metformin users and nonusers in patients with diabetes. As such, 40,774 propensity score-matched metformin users and nonusers were identified from Taiwan's National Health Insurance Research Database, between 1 January 2000 and 31 December 2017. We adopted the Cox proportional hazards model with robust standard error estimates for comparing the risks of UTI, sepsis, and death due to UTI or sepsis, in patients with T2DM. Compared with the nonuse of metformin, the aHRs (95% CI) for metformin use in UTI, recurrent UTI, sepsis, and death due to UTI or sepsis were 1.06 (0.98, 1.15), 1.08 (0.97, 1.2), 1.01 (0.97, 1.06), and 0.58 (0.42, 0.8), respectively. The cumulative incidence of death due to UTI or sepsis was significantly lower in metformin users than in nonusers (p = 0.002). A longer cumulative duration of metformin use had a lower aHR in the risk of death due to UTI or sepsis than metformin nonuse. In patients with T2DM, metformin use showed no significant differences in the risks of UTI, recurrent UTI, or sepsis. However, it was associated with a lower risk of death due to UTI or sepsis than metformin nonuse.

Cardiovascular outcomes, heart failure and mortality in type 2 diabetic patients treated with glucagon-like peptide 1 receptor agonists (GLP-1 RAs): A systematic review and meta-analysis of observational cohort studies

BACKGROUND

Cardiovascular outcomes trials (CVOTs) have assessed the effects of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) on major adverse cardiovascular events (MACE) and mortality in high cardiovascular (CV) risk populations. Observational research can provide complementary evidence about these effects in unselected populations.

AIM

To systematically review retrospective observational cohort studies conducted in electronic healthcare databases (EHDs) assessing GLP-1 RAs´ effects on MACE and/or hospitalisation for heart failure (HHF) and/or all-cause mortality in Type 2 diabetes mellitus (T2DM) patients.

METHODS

We systematically searched studies meeting inclusion criteria, compared design, methods and population characteristics, assessed risk for bias and did a meta-analysis (MA) using a random-effects model to calculate overall hazard ratios (HRs) and 95% CI (confidence intervals).

RESULTS

Sixteen studies included 285,436 T2DM patients exposed to GLP-1 RAs (exenatide bid, liraglutide, lixisenatide, long-acting exenatide), n ranged from 219 to 160,803 patients. Comparators included: no exposure, other antidiabetic medications (OADs), combined OADs, canagliflozin or multiple comparators. Ten studies estimated all-cause mortality, hazard ratios (HRs) ranged from 0.17 (95% CI 0.02-1.22) to 1.29 (95% CI 0.54-3.13). Thirteen studies assessed cardiovascular events and/or MACE; HRs ranged from 0.27 (95% CI 0.14-0.53) to 1.11 (95% CI 0.99-1.24). Eight studies assessed HHF, HRs ranged from 0.12 (95% CI 0.02-0.66) to 1.64 (95% CI 1.28-2.13). Excluding two studies because of temporal bias, we obtained pooled estimates for all-cause mortality: HR 0.63 (0.44-0.89), CV outcomes HR 0.84 (0.75-0.94) and HHF; HR 0.94 (0.78-1.14), (high between-study variability: I²  = 83.35%; I²  = 70.3%; and I²  = 90.1%, respectively).

CONCLUSION

Pooled results of EHDs' studies assessing GLP-1 RAs effects favoured GLP-1 RAs for all-cause mortality and MACE while were neutral for HHF. Results should be interpreted cautiously because of studies' substantial heterogeneity and limitations of observational research.

The association between use of proton-pump inhibitors and excess mortality after kidney transplantation: A cohort study

BACKGROUND

Chronic use of proton-pump inhibitors (PPIs) is common in kidney transplant recipients (KTRs). However, concerns are emerging about the potential long-term complications of PPI therapy. We aimed to investigate whether PPI use is associated with excess mortality risk in KTRs.

METHODS AND FINDINGS

We investigated the association of PPI use with mortality risk using multivariable Cox proportional hazard regression analyses in a single-center prospective cohort of 703 stable outpatient KTRs, who visited the outpatient clinic of the University Medical Center Groningen (UMCG) between November 2008 and March 2011 (ClinicalTrials.gov Identifier NCT02811835). Independent replication of the results was performed in a prospective cohort of 656 KTRs from the University Hospitals Leuven (NCT01331668). Mean age was 53 ± 13 years, 57% were male, and 56.6% used PPIs. During median follow-up of 8.2 (4.7-9.0) years, 194 KTRs died. In univariable Cox regression analyses, PPI use was associated with an almost 2 times higher mortality risk (hazard ratio [HR] 1.86, 95% CI 1.38-2.52, P < 0.001) compared with no use. After adjustment for potential confounders, PPI use remained independently associated with mortality (HR 1.68, 95% CI 1.21-2.33, P = 0.002). Moreover, the HR for mortality risk in KTRs taking a high PPI dose (>20 mg omeprazole equivalents/day) compared with patients taking no PPIs (HR 2.14, 95% CI 1.48-3.09, P < 0.001) was higher than in KTRs taking a low PPI dose (HR 1.72, 95% CI 1.23-2.39, P = 0.001). These findings were replicated in the Leuven Renal Transplant Cohort. The main limitation of this study is its observational design, which precludes conclusions about causation.

CONCLUSIONS

We demonstrated that PPI use is associated with an increased mortality risk in KTRs, independent of potential confounders. Moreover, our data suggest that this risk is highest among KTRs taking high PPI dosages. Because of the observational nature of our data, our results require further corroboration before it can be recommended to avoid the long-term use of PPIs in KTRs.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT02811835, NCT01331668.

Transparency in real-world evidence (RWE) studies to build confidence for decision-making: Reporting RWE research in diabetes

Transparency of real-world evidence (RWE) studies is critical to understanding how findings of a specific study were derived and is a necessary foundation to assessing validity and determination of whether decisions should be informed by the findings. In the present paper, we lay out strategies to improve clarity in the reporting of comparative effectiveness studies using real-world data that were generated by the routine operation of a healthcare system. This may include claims data, electronic health records, wearable devices, patient-reported outcomes or patient registries. These recommendations were discussed with multiple stakeholders, including regulators, payers, academics and journal editors, and endorsed by two professional societies that focus on RWE. We remind readers interested in diabetes research of the utility of conceptualizing a target trial that is then emulated by a RWE study when planning and communicating about RWE study implementation. We recommend the use of a graphical representation showcasing temporality of key longitudinal study design choices. We highlight study elements that should be reported to provide the clarity necessary to make a study reproducible. Finally, we suggest registering study protocols to increase process transparency. With these tools the readership of diabetes RWE studies will be able to more efficiently understand each study and be more able to assess a study's validity with reasonably high confidence before making decisions based on its findings.

Modeling of cumulative effects of time-varying drug exposures on within-subject changes in a continuous outcome

An accurate assessment of the safety or effectiveness of drugs in pharmaco-epidemiological studies requires defining an etiologically correct time-varying exposure model, which specifies how previous drug use affects the outcome of interest. To address this issue, we develop, and validate in simulations, a new approach for flexible modeling of the cumulative effects of time-varying exposures on repeated measures of a continuous response variable, such as a quantitative surrogate outcome, or a biomarker. Specifically, we extend the linear mixed effects modeling to estimate how past and recent drug exposure affects the way individual values of the outcome change throughout the follow-up. To account for the dosage, duration and timing of past exposures, we rely on a flexible weighted cumulative exposure methodology to model the cumulative effects of past drug use, as the weighted sum of past doses. Weights, modeled with unpenalized cubic regression B-splines, reflect the relative importance of doses taken at different times in the past. In simulations, we evaluate the performance of the model under different assumptions concerning (i) the shape of the weight function, (ii) the sample size, (iii) the number of the longitudinal observations and (iv) the intra-individual variance. Results demonstrate the accuracy of our estimates of the weight function and of the between- and within-patients variances, and good correlation between the observed and predicted longitudinal changes in the outcome. We then apply the proposed method to re-assess the association between time-varying glucocorticoid exposure and weight gain in people living with rheumatoid arthritis.

The EMPagliflozin compaRative effectIveness and SafEty (EMPRISE) study programme: Design and exposure accrual for an evaluation of empagliflozin in routine clinical care

BACKGROUND

The EMPA-REG OUTCOME trial showed that empagliflozin reduced the risk of cardiovascular death and hospitalization for heart failure (HHF) in diabetic patients with cardiovascular disease. EMPRISE is a study programme on the effectiveness, safety and healthcare utilization of empagliflozin in routine care, leveraging real-world data from two commercial and one federal US data sources from 2014 to 2019.

OBJECTIVES

To describe rationale and design of EMPRISE, assess ability to minimize confounding and evaluate the time to reach sufficient statistical power for a key study outcome, HHF, using baseline information from the first year of EMPRISE.

METHODS

In 3 claims data sets, we identified a 1:1 propensity score (PS)-matched cohort of diabetic patients ≥18 years initiating empagliflozin or a dipeptidyl peptidase-4 inhibitor (DPP4i), resulting in 6643 total pairs. The PS model included >140 baseline covariates. We measured covariate balance via standardized differences (SD) and postmatching c-statistic. We computed the incidence rate (IR) of HHF, predicted exposure accrual over time and calculated expected power.

RESULTS

After PS matching, patient characteristics were balanced with SD <0.1 and c-statistic between 0.54 and 0.59. The population IR of HHF was 4.4 per 1000 person-years using a specific HHF definition and 14.8 using a broader HHF definition. In our projection, 80%-powered analyses would require a minimum of 169 HHF events, expected to accumulate by year 3 (specific definition) or year 2 (broader definition).

CONCLUSION

Baseline information from EMPRISE provided evidence of solid confounding control and adequate exposure accrual with expected powered analyses for the primary outcomes.

Flexible Modeling of the Association Between Cumulative Exposure to Low-Dose Ionizing Radiation From Cardiac Procedures and Risk of Cancer in Adults With Congenital Heart Disease

Adults with congenital heart disease are increasingly being exposed to low-dose ionizing radiation (LDIR) from cardiac procedures. In a recent study, Cohen et al. (Circulation. 2018;137(13):1334-1345) reported an association between increased LDIR exposure and cancer incidence but did not explore temporal relationships. Yet, the impact of past exposures probably accumulates over years, and its strength may depend on the amount of time elapsed since exposure. Furthermore, LDIR procedures performed shortly before a cancer diagnosis may have been ordered because of early symptoms of cancer, raising concerns about reversal causality bias. To address these challenges, we combined flexible modeling of cumulative exposures with competing-risks methodology to estimate separate associations of time-varying LDIR exposure with cancer incidence and all-cause mortality. Among 24,833 patients from the Quebec Congenital Heart Disease Database, 602 had incident cancer and 500 died during a follow-up period of up to 15 years (1995-2010). Initial results suggested a strong association of cancer incidence with very recent LDIR exposures, likely reflecting reverse causality bias. When exposure was lagged by 2 years, an increased cumulative LDIR dose from the previous 2-6 years was associated with increased cancer incidence, with a stronger association for women. These results illustrate the importance of accurate modeling of temporal relationships between time-varying exposures and health outcomes.

Bias in pharmacoepidemiologic studies using secondary health care databases: a scoping review

BACKGROUND

The availability of clinical and therapeutic data drawn from medical records and administrative databases has entailed new opportunities for clinical and epidemiologic research. However, these databases present inherent limitations which may render them prone to new biases. We aimed to conduct a structured review of biases specific to observational clinical studies based on secondary databases, and to propose strategies for the mitigation of those biases.

METHODS

Scoping review of the scientific literature published during the period 2000-2018 through an automated search of MEDLINE, EMBASE and Web of Science, supplemented with manually cross-checking of reference lists. We included opinion essays, methodological reviews, analyses or simulation studies, as well as letters to the editor or retractions, the principal objective of which was to highlight the existence of some type of bias in pharmacoepidemiologic studies using secondary databases.

RESULTS

A total of 117 articles were included. An increasing trend in the number of publications concerning the potential limitations of secondary databases was observed over time and across medical research disciplines. Confounding was the most reported category of bias (63.2% of articles), followed by selection and measurement biases (47.0% and 46.2% respectively). Confounding by indication (32.5%), unmeasured/residual confounding (28.2%), outcome misclassification (28.2%) and "immortal time" bias (25.6%) were the subcategories most frequently mentioned.

CONCLUSIONS

Suboptimal use of secondary databases in pharmacoepidemiologic studies has introduced biases in the studies, which may have led to erroneous conclusions. Methods to mitigate biases are available and must be considered in the design, analysis and interpretation phases of studies using these data sources.

Observational research on sodium glucose co-transporter-2 inhibitors: A real breakthrough?

Sodium glucose co-transporter-2 inhibitors have attracted the interest of the scientific community following the results from dedicated cardiovascular outcome trials, which demonstrated remarkable reduction in all-cause mortality and other cardiovascular (CV) endpoints with empagliflozin and canagliflozin. These impressive results raised further expectations on real world data from large observational cohort studies. They were designed to address the possible existence of a class effect, and the uncertainty on whether this benefit can be extended from secondary to primary CV prevention of patients with type 2 diabetes. In this review, we collated data from existing observational studies (including the celebrated CVD-REAL cohorts) and critically appraised results and methodological issues with the aim of providing clinical insight, including unsettled aspects, and proposing a research agenda for future investigations.

Acute vs cumulative benefits of metformin use in patients with type 2 diabetes and heart failure

AIMS

To evaluate the association between metformin use and heart failure (HF) exacerbation in people with type 2 diabetes (T2D) and pre-existing HF using alternative exposure models.

MATERIALS AND METHODS

We analysed data for patients with T2D and incident HF from a national US insurance claims database. We compared the results of several multivariable Cox models where time-varying use of metformin was modelled as: (1) current use; (2) total duration of past use; and (3) use within the past 30 days or 10 days. The outcome was defined as time to HF-related hospitalization. We then re-analysed the data using flexible weighted cumulative exposure (WCE) models.

RESULTS

A total of 7620 patients with diabetes and incident HF were analysed. The mean (SD) patient age was 54 (8) years, and 58% (n = 4440) were men. In all, 3799 individuals (50%) were exposed to metformin, and 837 HF hospitalizations (11%) occurred (mean follow-up 1.7 years). Results of conventional models suggested potential acute benefits in reducing HF exacerbation with metformin use in the past 10 days (adjusted hazard ratio [aHR] 0.76, 95% confidence interval [CI] 0.60-0.97), while WCE models, which provided a better fit for the data, suggested lack of a systematic effect (aHR 0.91, 95% CI 0.69-1.20).

CONCLUSIONS

Our results suggest that cumulative metformin exposure does not decrease the risk of HF-related exacerbation. Use of other anti-hyperglycaemic agents with proven efficacy in patients with HF should also be considered as treatment options in this population.

Claims-based studies of oral glucose-lowering medications can achieve balance in critical clinical variables only observed in electronic health records

AIM

To evaluate the extent to which balance in unmeasured characteristics of patients with type 2 diabetes (T2DM) was achieved in claims data, by comparing against more detailed information from linked electronic health records (EHR) data.

METHODS

Within a large US commercial insurance database and using a cohort design, we identified patients with T2DM initiating linagliptin or a comparator agent within class (ie, another dipeptidyl peptidase-4 inhibitor) or outside class (ie, pioglitazone or a sulphonylurea) between May 2011 and December 2012. We focused on comparators used at a similar stage of diabetes to linagliptin. For each comparison, 1:1 propensity score (PS) matching was used to balance >100 baseline claims-based characteristics, including proxies of diabetes severity and duration. Additional clinical data from EHR were available for a subset of patients. We assessed representativeness of the claims-EHR-linked subset, evaluated the balance of claims- and EHR-based covariates before and after PS-matching via standardized differences (SDs), and quantified the potential bias associated with observed imbalances.

RESULTS

From a claims-based study population of 166 613 patients with T2DM, 7219 (4.3%) patients were linked to their EHR data. Claims-based characteristics in the EHR-linked and EHR-unlinked patients were similar (SD < 0.1), confirming the representativeness of the EHR-linked subset. The balance of claims-based and EHR-based patient characteristics appeared to be reasonable before PS-matching and generally improved in the PS-matched population, to be SD < 0.1 for most patient characteristics and SD < 0.2 for select laboratory results and body mass index categories, which was not large enough to cause meaningful confounding.

CONCLUSION

In the context of pharmacoepidemiological research on diabetes therapy, choosing appropriate comparison groups paired with a new-user design and 1:1 PS matching on many proxies of diabetes severity and duration improves balance in covariates typically unmeasured in administrative claims datasets, to the extent that residual confounding is unlikely.

Competing risks modeling of cumulative effects of time-varying drug exposures

An accurate assessment of drug safety or effectiveness in pharmaco-epidemiology requires defining an etiologically correct time-varying exposure model, which specifies how previous drug use affects the hazard of the event of interest. An additional challenge is to account for the multitude of mutually exclusive events that may be associated with the use of a given drug. To simultaneously address both challenges, we develop, and validate in simulations, a new approach that combines flexible modeling of the cumulative effects of time-varying exposures with competing risks methodology to separate the effects of the same drug exposure on different outcomes. To account for the dosage, duration and timing of past exposures, we rely on a spline-based weighted cumulative exposure modeling. We also propose likelihood ratio tests to test if the cumulative effects of past exposure on the hazards of the competing events are the same or different. Simulation results indicate that the estimated event-specific weight functions are reasonably accurate, and that the proposed tests have acceptable type I error rate and power. In real-life application, the proposed method indicated that recent use of antihypertensive drugs may reduce the risk of stroke but has no effect on the hazard of coronary heart disease events.

Adjustment for time-dependent unmeasured confounders in marginal structural Cox models using validation sample data

Large databases used in observational studies of drug safety often lack information on important confounders. The resulting unmeasured confounding bias may be avoided by using additional confounder information, frequently available in smaller clinical “validation samples”. Yet, no existing method that uses such validation samples is able to deal with unmeasured time-varying variables acting as both confounders and possible mediators of the treatment effect. We propose and compare alternative methods which control for confounders measured only in a validation sample within marginal structural Cox models. Each method corrects the time-varying inverse probability of treatment weights for all subject-by-time observations using either regression calibration of the propensity score, or multiple imputation of unmeasured confounders. Two proposed methods rely on martingale residuals from a Cox model that includes only confounders fully measured in the large database, to correct inverse probability of treatment weight for imputed values of unmeasured confounders. Simulation demonstrates that martingale residual-based methods systematically reduce confounding bias over naïve methods, with multiple imputation including the martingale residual yielding, on average, the best overall accuracy. We apply martingale residual-based imputation to re-assess the potential risk of drug-induced hypoglycemia in diabetic patients, where an important laboratory test is repeatedly measured only in a small sub-cohort.

Association of insulin dosage with mortality or major adverse cardiovascular events: a retrospective cohort study

BACKGROUND

Existing studies have shown conflicting evidence regarding the safety of exogenous insulin therapy in patients with type 2 diabetes. In particular, observational studies have reported an increased risk of death and cardiovascular disease among users of higher versus lower doses of insulin. We aimed to quantify the association between increasing dosage of insulin exposure and death and cardiovascular events, while taking into account time-dependent confounding and mediation that might have biased previous studies.

METHODS

We did a cohort study using primary care records from the UK-based Clinical Practice Research Datalink (CPRD). New users of metformin monotherapy were identified in the period between Jan 1, 2001, and Dec 31, 2012. We then identified those in this group with a new prescription for insulin. Insulin exposure was categorised into groups according to the mean dose (units) per day within 180-day time segments throughout each patient's follow-up. Relative differences in mortality and major adverse cardiovascular events (non-fatal myocardial infarction, non-fatal stroke, cardiovascular-related mortality) were assessed using conventional multivariable Cox proportional hazards models. Marginal structural models were then applied to reduce bias introduced by the time-dependent confounders affected by previous treatment.

FINDINGS

We identified 165 308 adults with type 2 diabetes in the CPRD database. After applying our exclusion criteria, 6072 (mean age 60 years [SD 12·5], 3281 [54%] men, mean HbA_1c 8·5% [SD 1·75], and median follow-up 3·1 years [IQR 1·7-5·3) were new add-on insulin users and were included in the study cohort; 3599 were new add-on insulin users and were included in the subcohort linked to hospital records and death certificate information. Crude mortality rates were comparable between insulin dose groups; <25 units per day (46 per 1000 person-years), 25 to <50 units per day (39 per 1000 person-years), 50 to <75 units per day (27 per 1000 person-years), 75 to <100 units per day (34 per 1000 person-years), and at least 100 units per day (32 per 1000 person-years; p>0·05 for all; mean rate of 31 deaths per 1000 person-years [95% CI 29-33]). With adjustment for baseline covariates, mortality rates were higher for increasing insulin doses: less than 25 units per day [reference group]; 25 to <50 units per day, hazard ratio (HR) 1·41 [95% CI 1·12-1·78]; 50 to <75 units per day, 1·37 [1·04-1·80]; 75 to <100 units per day, 1·85 [1·35-2·53]; and at least 100 units per day, 2·16 [1·58-2·93]. After applying marginal structural models, insulin dose was not associated with mortality in any group (p>0·1 for all).

INTERPRETATION

In conventional multivariable regression analysis, higher insulin doses are associated with increased mortality after adjustment for baseline covariates. However, this effect seems to be confounded by time-dependent factors such as insulin exposure, glycaemic control, bodyweight gain, and the occurrence of cardiovascular and hypoglycaemic events. This study provides reassurance of the overall safety of insulin use in the treatment of type 2 diabetes and contributes to our understanding of the contrasting conclusions from non-randomised and randomised studies regarding dose-dependent effects of insulin on cardiovascular events and mortality.

FUNDING

Canadian Institutes of Health Research, Heart and Stroke Foundation of Canada, and the Newfoundland and Labrador Research and Development Corporation.

Common medications and prostate cancer mortality: a review

PURPOSE

Most prostate cancer patients also have comorbidities that are treated with both prescription and nonprescription medications; furthermore, many use dietary supplements. We assess their association with prognosis after prostate cancer diagnosis, and we discuss methodological challenges and clinical implications.

METHODS

We reviewed high-quality observational studies investigating the association of commonly used medications and supplements with prostate cancer-specific mortality.

RESULTS

There is preliminary evidence that statins and metformin use may be associated with lower risk of cancer-specific mortality after prostate cancer diagnosis; conversely, high calcium and multivitamin supplementation may be associated with increased risk. Evidence is inconclusive for nonsteroidal anti-inflammatory drugs, acetylsalicylic acid (aspirin), insulin, antihypertensives such as angiotensin-converting enzyme inhibitors and beta-blockers, digoxin, and warfarin. Common limitations of the internal validity of studies examined include unmeasured confounding and confounding by indication, competing risks, and time-related biases such as immortal time bias. The majority of studies focused on Caucasian men with specific comorbidities, while heterogeneity among patients and tumors was mostly not assessed.

CONCLUSIONS

Commonly prescribed medications and over-the-counter supplements may influence prognosis among prostate cancer patients. Further well-designed pharmacoepidemiologic studies and randomized controlled trials of selected medications in appropriate patient groups are necessary before these drugs can bear new indications for prostate cancer treatment. We discuss considerations when deciding about use of these drugs in clinical practice at the present time.