J. Bradley Layton

Propensity Score Methods for Confounding Control in Nonexperimental Research

Nonexperimental studies are increasingly used to investigate the safety and effectiveness of medical products as they are used in routine care. One of the primary challenges of such studies is confounding, systematic differences in prognosis between patients exposed to an intervention of interest and the selected comparator group. In the presence of uncontrolled confounding, any observed difference in outcome risk between the groups cannot be attributed solely to a causal effect of the exposure on the outcome. Confounding in studies of medical products can arise from a variety of different sociomedical processes.1 The most common form of confounding arises from good medical practice, physicians prescribing medications and performing procedures on patients who are most likely to benefit from them. This leads to a bias known as confounding by indication, which can cause medical interventions to appear to cause events that they prevent.2,3 Conversely, patients who are perceived by a physician to be near the end of life may be less likely to receive preventive medications, leading to confounding by frailty or comorbidity.4–6 Additional sources of confounding bias can result from patients’ health-related behaviors. For example, patients who initiate a preventive medication may be more likely than other patients to engage in other healthy, prevention-oriented behaviors leading to bias known as the healthy user/adherer effect.7–9 Many statistical approaches can be used to remove the confounding effects of such factors if they are captured in the data. The most common statistical approaches for confounding control are based on multivariable regression models of the outcome. To yield unbiased estimates of treatment effects, these approaches require that the researcher correctly models the effect of the treatment and covariates on the outcome. However, correct specification of an outcome model can be challenging, particularly in studies …

Testosterone lab testing and initiation in the United Kingdom and the United States, 2000 to 2011:

CONTEXT New formulations, increased marketing, and wider recognition of declining testosterone levels in older age may have contributed to wider testosterone testing and supplementation in many countries. OBJECTIVE Our objective was to describe testosterone testing and testosterone treatment in men in the United Kingdom and United States. DESIGN This was a retrospective incident user cohort. SETTING We evaluated commercial and Medicare insurance claims from the United States and general practitioner healthcare records from the United Kingdom for the years 2000 through 2011. PARTICIPANTS We identified 410,019 US men and 6858 UK men who initiated a testosterone formulation as well as 1,114,329 US men and 66,140 UK men with a new testosterone laboratory measurement. MAIN OUTCOME MEASURES Outcome measures included initiation of any injected testosterone, implanted testosterone pellets, or prescribed transdermal or oral testosterone formulation. RESULTS Testosterone testing and supplementation have increased pronouncedly in the United States. Increased testing in the United Kingdom has identified more men with low levels, yet US testing has increased among men with normal levels. Men in the United States tend to initiate at normal levels more often than in the United Kingdom, and many men initiate testosterone without recent testing. Gels have become the most common initial treatment in both countries. CONCLUSIONS Testosterone testing and use has increased over the past decade, particularly in the United States, with dramatic shifts from injections to gels. Substantial use is seen in men without recent testing and in US men with normal levels. Given widening use despite safety and efficacy questions, prescribers must consider the medical necessity of testosterone before initiation.

A clinical tool to measure the components of health-care transition from pediatric care to adult care: the UNC TR(x)ANSITION scale.

Objective: To describe the development of the University of North Carolina (UNC) TRxANSITION Scale that measures the health-care transition and self-management skills by youth with chronic health conditions. Methods: Item and scale development of the UNC TRxANSITION Scale was informed by two theoretical models, available literature, and expert opinion interviews and feedback from youth with chronic conditions, their parents, and interdisciplinary collaboration. Through an iterative process, three versions of the scale were piloted on a total of 185 adolescents and emerging adults with different chronic illnesses. This clinically administered scale relies on a semi-structured interview format of the patient and does not rely solely on patient report, but is verified with information from the medical record to validate responses. Results: Following the item development and the three iterations of the scale, version 3 was examined in a more intensive fashion. The current version of the UNC TRxANSITION Scale comprises 33 items scattered across the following 10 domains: Type of illness, Rx=medications, Adherence, Nutrition, Self-management, Informed-reproduction, Trade/school, Insurance, Ongoing support, and New health providers. It requires approximately 7–8 min to administer. With a sample of 128 adolescents and young adults, ranging in age from 12 to 20, inter-rater reliability was strong (r = 0.71) and item-total correlation scores were moderate to high. Content and construct validity were satisfactory, and the overall score was sensitive to advancing age. The univariate linear regression yielded a beta coefficient of 1.08 (p < 0.0001), indicating that the total score increased with advancing age. Specifically, there was about a one point increase in the total score for each year of age. Conclusion: The UNC TRxANSITION Scale is a disease-neutral tool that can be used in the clinical setting. Initial findings suggest that it is a reliable and valid tool that has the potential to measure health-care transition skill mastery and knowledge in a multidimensional fashion.

Comparative safety of testosterone dosage forms

IMPORTANCE Increases in testosterone use and mixed reports of adverse events have raised concerns about the cardiovascular safety of testosterone. Testosterone is available in several delivery mechanisms with varying pharmacokinetics; injections cause spikes in testosterone levels, and transdermal patches and gels cause more subtle but sustained increases. The comparative cardiovascular safety of gels, injections, and patches has not been studied. OBJECTIVE To determine the comparative cardiovascular safety of testosterone injections, patches, and gels. DESIGN, SETTING, AND PARTICIPANTS A retrospective cohort study was conducted using administrative claims from a commercially insured (January 1, 2000, to December 31, 2012) and Medicare (January 1, 2007, to December 31, 2010) population in the United States and general practitioner records from the United Kingdom (January 1, 2000, to June 30, 2012). Participants included men (aged ≥18 years) who initiated use of testosterone patches, gels, or injections following 180 days with no testosterone use. Our analysis was conducted from December 11, 2013, to November 12, 2014. EXPOSURES New initiation of a testosterone dosage form, with use monitored for up to 1 year. MAIN OUTCOMES AND MEASURES Inpatient or outpatient medical records, diagnoses, or claims for cardiovascular and cerebrovascular events including myocardial infarction (MI), unstable angina, stroke, and composite acute event (MI, unstable angina, or stroke); venous thromboembolism (VTE); mortality; and all-cause hospitalization. RESULTS We identified 544,115 testosterone initiators between the 3 data sets: 37.4% injection, 6.9% patch, and 55.8% gel. The majority of men in the Medicare cohort were injection initiators (51.2%), most in the US commercially insured population were gel initiators (56.5%), and the UK database included equal proportions of injections and gel users (approximately 41%). With analysis conducted using hazard ratios and 95% CIs, compared with men using gels, injection initiators had higher hazards of cardiovascular events (ie, MI, unstable angina, and stroke) (1.26; 1.18-1.35), hospitalization (1.16; 1.13-1.19), and death (1.34; 1.15-1.56) but not VTE (0.92; 0.76-1.11). Compared with gels, patches did not confer increased hazards of cardiovascular events (1.10; 0.94-1.29), hospitalization (1.04; 1.00-1.08), death (1.02; 0.77-1.33), or VTE (1.08; 0.79-1.47). CONCLUSIONS AND RELEVANCE Testosterone injections were associated with a greater risk of cardiovascular events, hospitalizations, and deaths compared with gels. Patches and gels had similar risk profiles. However, this study did not assess whether patients met criteria for use of testosterone and did not assess the safety of testosterone among users compared with nonusers of the drug.

Effect of Statin Use on Acute Kidney Injury Risk Following Coronary Artery Bypass Grafting

Acute kidney injury (AKI) is a serious complication of cardiovascular surgery. Although some nonexperimental studies suggest that statin use may reduce postsurgical AKI, methodologic differences in study designs leave uncertainty regarding the reality or magnitude of the effect. The aim of this study was to estimate the effect of preoperative statin initiation on AKI after coronary artery bypass grafting (CABG) using an epidemiologic approach more closely simulating a randomized controlled trial in a large CABG patient population. Health care claims from large, employer-based and Medicare insurance databases for 2000 to 2010 were used. To minimize healthy user bias, patients were identified who underwent nonemergent CABG who either newly initiated a statin <20 days before surgery or were unexposed for ≥200 days before CABG. AKI was identified <15 days after CABG. Multivariate-adjusted risk ratios (RRs) and 95% confidence intervals (CIs) were calculated using Poisson regression. Analyses were repeated using propensity score methods adjusted for clinical and health care utilization variables. A total of 17,077 CABG patients were identified. Post-CABG AKI developed in 3.4% of statin initiators and 6.2% of noninitiators. After adjustment, a protective effect of statin initiation on AKI was observed (RR 0.78, 95% CI 0.63 to 0.96). This effect differed by age, with an RR of 0.91 (95% CI 0.68 to 1.20) for patients aged ≥65 years and an RR of 0.62 (95% CI 0.45 to 0.86) for those aged <65 years, although AKI was more common in the older group (7.7% vs 4.0%). In conclusion, statin initiation immediately before CABG may modestly reduce the risk for postoperative AKI, particularly in younger CABG patients.

Association Between Direct-to-Consumer Advertising and Testosterone Testing and Initiation in the United States, 2009-2013

Importance Testosterone initiation increased substantially in the United States from 2000 to 2013, especially among men without clear indications. Direct-to-consumer advertising (DTCA) also increased during this time. Objective To investigate associations between televised DTCA and testosterone testing and initiation in the United States. Design, Setting, and Population Ecologic study conducted in designated market areas (DMAs) in the United States. Monthly testosterone advertising ratings were linked to DMA-level testosterone use data from 2009-2013 derived from commercial insurance claims. Associations between DTCA and testosterone testing, initiation, and initiation without recent baseline tests were estimated using Poisson generalized estimating equations. Exposures Monthly Nielsen ratings for testosterone DTCA in the 75 largest DMAs. Main Outcomes and Measures (1) Rates of new serum testosterone testing; (2) rates of testosterone initiation (in-office injection, surgical implant, or pharmacy dispensing) for all testosterone products combined and for specific brands; and (3) rates of testosterone initiation without recent serum testosterone testing. Results Of 17 228 599 commercially insured men in the 75 DMAs, 1 007 990 (mean age, 49.6 [SD, 11.5] years) had new serum testosterone tests and 283 317 (mean age, 51.8 [SD, 11.3] years) initiated testosterone treatment. Advertising intensity varied by geographic region and time, with the highest intensity seen in the southeastern United States and with months ranging from no ad exposures to a mean of 13.6 exposures per household. Nonbranded advertisements were common prior to 2012, with branded advertisements becoming more common during and after 2012. Each household advertisement exposure was associated with a monthly increase in rates of new testosterone testing (rate ratio [RR], 1.006; 95% CI, 1.004-1.008), initiation (RR, 1.007; 95% CI, 1.004-1.010), and initiation without a recent test (RR, 1.008; 95% CI, 1.002-1.013). Mean absolute rate increases were 0.14 tests (95% CI, 0.09-0.19), 0.05 new initiations (95% CI, 0.03-0.08), and 0.02 initiations without a recent test (95% CI, 0.01-0.03) per 10 000 men for each monthly ad exposure over the entire period. Conclusions and Relevance Among US men residing in the 75 designated market areas, regional exposure to televised direct-to-consumer advertising was associated with greater testosterone testing, new initiation, and initiation without recent testing.

Acute Kidney Injury and Long-term Risk of Cardiovascular Events After Cardiac Surgery: A Population-Based Cohort Study

OBJECTIVE To examine the impact of postoperative acute kidney injury (AKI) on the long-term risk of myocardial infarction, heart failure, stroke, and all-cause mortality after elective cardiac surgery. The authors investigated whether time of onset of AKI altered the association between AKI and the adverse events. DESIGN Population-based cohort study in 2006-2011. SETTING Two university hospitals. PARTICIPANTS Adult elective cardiac surgical patients. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS AKI was defined as an increase in baseline creatinine according to the Kidney Disease Improving Global Outcomes criteria. AKI was defined within 30 days of surgery, and also analyzed as early- or late-onset AKI. The authors followed patients from postoperative day 30 until hospitalization with myocardial infarction, heart failure, stroke, or death. Adjustment for confounding factors was done using propensity scores and standardized-mortality-ratio weights. A total of 1,457 (30.7%) of 4,742 patients developed AKI within 30 days of surgery and 470 (9.9%) patients experienced a composite cardiovascular endpoint. Comparing patients with and without postoperative AKI, weighted hazard ratio (HR) and 95% confidence intervals (CI) of 5-year risk of the composite cardiovascular endpoint was 1.41 (95% CI: 1.11-1.80). For each endpoint separately the weighted HR was similarly increased. Ninety-one days to 5-year weighted HR of all-cause mortality was 1.37 (95% CI: 1.05-1.80). The effect of AKI was similar for early- and late-onset AKI. CONCLUSIONS Early- and late-onset AKI within 30 days of elective cardiac surgery was associated with a similarly increased 5-year risk of myocardial infarction, heart failure, stroke, and increased all-cause mortality.

Cognitive Pharmacy Services at a Pediatric Nephrology and Hypertension Clinic

Purpose: Pediatric patients require special attention from pediatric pharmacists. This is particularly true for pediatric patients with chronic kidney disease (CKD) as the number of their medications and the complexity of their treatment increase with disease progression. However, there is paucity of information describing pediatric cognitive pharmacy services in this setting. The objective of this study is to identify the potential roles of a clinical pharmacist as a provider in a pediatric nephrology and hypertension clinic. Methods: Pediatric patients (≤18 years of age) who chronically took at least one medication were consecutively enrolled at the University of North Carolina (UNC) Pediatric Nephrology and Hypertension Clinic from 1 August 2007 to 15 April 2008. Demographic information and the interventions performed during the clinic visit by a clinical pharmacist were examined. Results: Three hundred and seventy-four visits made in 283 participants were evaluated. The mean (SD) number of cognitive pharmacy interventions per patient was 2.3 (1.0) on the first visit, with medication counseling and verification of current medications comprising the most common activity (85%). The mean (SD) number of medications per patient was 5.7 (4.8) and of medications counseled per visit was 4.0 (3.4). Medication adherence was investigated in 141 (38%) visits. Pretransplant education on medications was performed in 3% of the patients. Discrepancies of medications were discovered in 12 of the 374 visits. Conclusion: Pediatric cognitive pharmacy services to patients at the UNC pediatric nephrology clinic were feasible, which improved the quality of services and promoted better outcomes for these complex patients.

Effectiveness of chemoradiation for head and neck cancer in an older patient population

PURPOSE The purpose of this study was to compare chemoradiation therapy (CRT) with radiation therapy (RT) only in an older patient population with head and neck squamous cell carcinoma (HNSCC). METHODS AND MATERIALS Using the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database (1992-2007), we identified a retrospective cohort of nonmetastatic HNSCC patients and divided them into treatment groups. Comparisons were made between CRT and RT cohorts. Propensity scores for CRT were estimated from covariates associated with receipt of treatment using multivariable logistic regression. Standardized mortality ratio weights (SMRW) were created from the propensity scores and used to balance groups on measured confounders. Multivariable and SMR-weighted Cox proportional hazard models were used to estimate the hazard ratio (HR) of death for receipt of CRT versus RT among the whole group and for separate patient and tumor categories. RESULTS The final cohort of 10,599 patients was 68% male and 89% white. Median age was 74 years. Seventy-four percent were treated with RT, 26% were treated with CRT. Median follow-up points for CRT and RT survivors were 4.6 and 6.3 years, respectively. On multivariable analysis, HR for death with CRT was 1.13 (95% confidence interval [CI]: 1.07-1.20; P<.01). Using the SMRW model, the HR for death with CRT was 1.08 (95% CI: 1.02-1.15; P=.01). CONCLUSIONS Although the addition of chemotherapy to radiation has proven efficacious in many randomized controlled trials, it may be less effective in an older patient population treated outside of a controlled trial setting.

Acute kidney injury in statin initiators

Statins are widely used for preventing cardiovascular disease, yet recent reports suggest an increased risk of acute kidney injury (AKI). We estimated the one‐year risk of AKI associated with statin initiation and determined the comparative safety of individual statin formulations.