Publications
scientifiques

BACKGROUND & AIMS: Childhood obesity is increasing and is associated with adult obesity. Antibiotics have been used to promote weight gain in livestock for several decades. Antibiotics are commonly prescribed for children, but it is not clear how exposure to antibiotics early in life affects risk for obesity. We performed a population-based cohort study to assess the association between antibiotic exposure before age 2 years and obesity at age 4 years. METHODS: We performed a retrospective cohort study of 21,714 children in The Health Improvement Network-a population-representative dataset of >10 million individuals derived from electronic medical records from 1995 through 2013 in the United Kingdom. Eligible subjects were registered within 3 months of birth with complete follow-up and height and weight were recorded within 12 months of their 4th birthday. Antibiotic exposure was assessed before age 2 years, and classified based on anti-anaerobic activity. The primary outcome was obesity at age 4 years. We performed logistic regression analyses, adjusting for maternal and sibling obesity, maternal diabetes, mode of delivery, socioeconomic status, year and country of birth, and urban dwelling. RESULTS: In the cohort, 1306 of the children (6.4%) were obese at 4 years of age. Antibiotic exposure was associated with an increased risk of obesity at 4 years (odds ratio [OR] = 1.21; 95% confidence interval [CI]: 1.07-1.38). ORs increased with repeated exposures: for 1-2 prescriptions, OR = 1.07 (95% CI, 0.91-1.23); for 3-5 prescriptions, OR = 1.41 (95% CI, 1.20-1.65); and for 6 or more prescriptions, OR = 1.47 (95% CI, 1.19-1.82). Antifungal agents were not associated with obesity (OR = 0.81; 95% CI, 0.59-1.11). CONCLUSIONS: Administration of 3 or more courses of antibiotics before children reach an age of 2 years is associated with an increased risk of early childhood obesity.

OBJECTIVE: We aimed to examine whether the current users of specific NSAIDs have an increased risk of venous thromboembolism (VTE) among knee OA patients. METHODS: We conducted a population-based case-control study using The Health Improvement Network, a database of patient records from general practices in the UK. For every VTE case, we identified five controls matched on age, sex and calendar year of study enrolment. We used conditional logistic regression to assess the association between current use of specific NSAIDs and risk of VTE relative to remote NSAID users. RESULTS: Among knee OA patients with at least one NSAID prescription, we identified 4020 incident cases of VTE and 20 059 matched controls. Adjusted odd ratios (ORs) relative to the remote users were 1.38 (95% CI: 1.32, 1.44) for recent users and 1.43 (95% CI: 1.36, 1.49) for current users. Among the current NSAID users, the risk of VTE was increased with diclofenac [OR 1.63 (95% CI: 1.53, 1.74)], ibuprofen [OR = 1.49 (95% CI: 1.38, 1.62)], meloxicam [OR = 1.29 (95% CI: 1.11, 1.50)] and coxibs [celecoxib, OR = 1.30 (95% CI: 1.11, 1.51); rofecoxib, OR = 1.44 (95% CI: 1.18, 1.76)]; naproxen did not increase VTE risk [OR = 1.00 (95% CI: 0.89, 1.12)]. CONCLUSION: Compared with the remote users of NSAIDs, the risk of VTE increased for current users of diclofenac, ibuprofen, meloxicam, and coxibs, but not for naproxen, in the knee OA population. Clinicians should consider the risk profile for specific NSAIDs when recommending their use.

IMPORTANCE: Self-harm is a prominent cause of morbidity in patients with bipolar disorder and is strongly associated with suicide. There is evolving evidence that lithium use may reduce suicidal behavior, in addition to concerns that the use of anticonvulsants may increase self-harm. Information is limited about the effects of antipsychotics when used as mood stabilizer treatment. Rates of unintentional injury are poorly defined in bipolar disorder, and understanding drug associations with this outcome may shed light on mechanisms for lithium's potential antisuicidal properties through reduction in impulsive aggression. OBJECTIVE: To compare rates of self-harm, unintentional injury, and suicide in patients with bipolar disorder who were prescribed lithium, valproate sodium, olanzapine, or quetiapine fumarate. DESIGN, SETTING, AND PARTICIPANTS: This investigation was a propensity score (PS)-adjusted and PS-matched longitudinal cohort study in a nationally representative UK sample using electronic health records data collected between January 1, 1995, and December 31, 2013. Participants included all patients diagnosed as having bipolar disorder who were prescribed lithium, valproate, olanzapine, or quetiapine as maintenance mood stabilizer treatment. MAIN OUTCOMES AND MEASURES: The primary outcome was any form of self-harm. Secondary outcomes were unintentional injury and suicide. RESULTS: Of the 14 396 individuals with a diagnosis of BPD, 6671 were included in the cohort, with 2148 prescribed lithium, 1670 prescribed valproate, 1477 prescribed olanzapine, and 1376 prescribed quetiapine as maintenance mood stabilizer treatment. Self-harm rates were lower in patients prescribed lithium (205; 95% CI, 175-241 per 10 000 person-years at risk [PYAR]) compared with those prescribed valproate (392; 95% CI, 334-460 per 10 000 PYAR), olanzapine (409; 95% CI, 345-483 per 10 000 PYAR), or quetiapine (582; 95% CI, 489-692 per 10 000 PYAR). This association was maintained after PS adjustment (hazard ratio [HR], 1.40; 95% CI, 1.12-1.74 for valproate, olanzapine, or quetiapine vs lithium) and PS matching (HR, 1.51; 95% CI, 1.21-1.88). After PS adjustment, unintentional injury rates were lower for lithium compared with valproate (HR, 1.32; 95% CI, 1.10-1.58) and quetiapine (HR, 1.34; 95% CI, 1.07-1.69) but not olanzapine. The suicide rate in the cohort was 14 (95% CI, 9-21) per 10 000 PYAR. Although this rate was lower in the lithium group than for other treatments, there were too few events to allow accurate estimates. CONCLUSIONS AND RELEVANCE: Patients taking lithium had reduced self-harm and unintentional injury rates. This finding augments limited trial and smaller observational study results. It supports the hypothesis that lithium use reduces impulsive aggression in addition to stabilizing mood.

A number of epidemiologic studies have demonstrated associations between obesity and diabetes and the risk of monoclonal gammopathy of undetermined significance (MGUS). However, since MGUS is an asymptomatic condition we evaluated whether these are true associations or the result of detection-bias. We conducted a nested case-control study using a large primary-care database. Cases were defined as those with incident diagnosis of MGUS. For every case, four eligible controls matched on age, sex, practice site, and duration of follow-up were selected. Exposure variables included obesity and diabetes (including antidiabetic therapies) as well as other metabolic risk factors. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using conditional logistic regression. The study included 2363 MGUS patients and 9193 matched controls. In the primary analysis, obesity and diabetes were associated with higher MGUS risk with an adjusted ORs of 1.15 (95% CI 1.02-1.29) and 1.30 (95% CI 1.13-1.50), respectively. However, after adjustment to the number of laboratory tests prior to the MGUS diagnosis, there was no association between obesity and diabetes and MGUS risk (ORs of 1.08, 95% CI 0.96-1.22 and 1.08, 95% CI 0.93-1.25, respectively). In an additional analysis of antidiabetic therapies and MGUS risk, there was a nonsignificant decrease in MGUS risk among diabetes patients treated with metformin alone compared to subjects without diabetes (OR 0.77, 95% CI 0.56-1.05). In summary, while previously described risk factors for MGUS might be the result of detection bias, metformin should be further evaluated as a possible chemoprevention modality. Am. J. Hematol. 91:581-584, 2016. © 2016 Wiley Periodicals, Inc.

BACKGROUND: Prescription drugs are a central component of healthcare worldwide. We investigated changes in drug-prescribing patterns over time in the general population. METHODS: Secular trends were analyzed using 1999-2012 prescription data from The Health Improvement Network. Prevalence of receipt of medication prescriptions was computed by age, sex, and therapeutic category for each calendar year. Spearman correlations were computed to assess change over time. RESULTS: Between 1999 and 2012, the percentage of the population that received at least one medication prescription increased from 64.5% to 69.2% (rho = 0.96, p < 0.001). The percentage of patients receiving prescriptions for one to four unique agents declined from 45.6% to 42.1% (Spearman's rho = -0.98, p < 0.001). Meanwhile, the percentage receiving five to nine and 10 or more unique agents increased from 14.1% to 17.5% (rho = 0.996, p < 0.001) and 4.7% to 9.6% (rho = 1.000, p < 0.001) respectively. Largest increases were seen in use of drugs for gastrointestinal disease among women and cardiovascular disease among men. In 2012, the most commonly used agents were for infection or nervous system drugs, with 32.0% and 28.9% of patients receiving at least one prescription, respectively. CONCLUSIONS: Nearly 70% of the United Kingdom population has received prescriptions for one or more medication with increasing proportions receiving prescriptions for five or more. The high rates of medication use increase the complexity and cost of healthcare. These data can be used for public health planning and to design pharmacoepidemiology and comparative effectiveness studies. Copyright © 2015 John Wiley & Sons, Ltd.

Over a period of 5 years, the Innovative Medicines Initiative PROTECT (Pharmacoepidemiological Research on Outcomes of Therapeutics by a European ConsorTium) project has addressed key research questions relevant to the science of safety signal detection. The results of studies conducted into quantitative signal detection in spontaneous reporting, clinical trial and electronic health records databases are summarised and 39 recommendations have been formulated, many based on comparative analyses across a range of databases (e.g. regulatory, pharmaceutical company). The recommendations point to pragmatic steps that those working in the pharmacovigilance community can take to improve signal detection practices, whether in a national or international agency or in a pharmaceutical company setting. PROTECT has also pointed to areas of potentially fruitful future research and some areas where further effort is likely to yield less.

BACKGROUND: Statins are widely prescribed for the primary prevention of cardiovascular disease. Guidelines exist for statin prescriptions, but there is little recent analysis concerning prescription trends over time and how these vary with respect to demographic variables. METHODS AND RESULTS: Using The Health Improvement Network primary care database, statin therapy initiation and statin prescription prevalence rates were calculated using data from 7,027,711 individuals across the UK for the years 1995 to 2013, overall and stratified by sex, age group, and socioeconomic deprivation level (Townsend score). Statin therapy initiation rates rose sharply from 1995 (0.51 per 1,000 person-years) up to 2006 (19.83 per 1,000 person-years) and thereafter declined (10.76 per 1,000 person-years in 2013). Males had higher initiation rates than females and individuals aged 60-85 years had higher initiation rates than younger or more elderly age groups. Initiation rates were slightly higher as social deprivation level increased, after accounting for age and sex. Prescription prevalence increased sharply from 1995 (2.36 per 1,000 person-years) to 2013 (128.03 per 1,000 person-years) with males generally having a higher prevalence rate, over time, than females. Prevalence rates over time were generally higher for older age groups but were similar with respect to social deprivation level. CONCLUSION: The uptake of statins within UK primary care has increased greatly over time with statins being more commonly prescribed to older patients in general and, in recent years, males appear to have been prescribed statins at higher rates than females. After accounting for age and sex, the statin therapy initiation rate increases with the level of social deprivation.

BACKGROUND: Despite the lack of scientific data, many cancer patients hold the belief that glucose 'feeds' cancer and might affect disease outcome. We aimed to evaluate associations between glucose, hemoglobin A1C (HbA1C), and survival among individuals with diabetes and diabetes associated cancers. METHODS: Five retrospective cohort studies were conducted in a large population-representative database. The study population included all patients with diabetes and an incident diagnosis of colorectal, breast, bladder, pancreatic and prostate cancers. Exposure of interest was serum glucose or HbA1C levels within 6 months prior to cancer diagnosis. Cox regression model was used to calculate hazard-ratio (HR) and 95% confidence-interval (CI) for overall survival. Analyses were adjusted for cancer-specific confounders. A subgroup analysis was performed among insulin-treated patients. RESULTS: Study cohorts included 7916 individuals with incident cancers and concurrent diabetes. There was no association between HbA1C levels and overall survival in colorectal (HR 1.00, 95% CI 0.95-1.06), breast (HR 1.03, 95% CI 0.95-1.11), bladder (HR 0.94, 95% CI 0.86-1.01), pancreatic (HR 0.98, 95% CI 0.94-1.02), or prostate (HR 1.02, 95% CI 0.96-1.08) cancers. Among diabetes patients treated with insulin, there was increased survival with increasing serum glucose, most prominent for bladder cancer (HR 0.91, 95% CI 0.84-0.99, per 1 mmol/l increase). CONCLUSIONS: Higher glucose and HbA1C levels in diabetes patients with incident cancer are not associated with worse overall survival following cancer diagnosis. Among insulin-treated patients, higher glucose levels may be associated with improved survival.

Instrumental variable (IV) methods are widely used to adjust for the bias in estimating treatment effects caused by unmeasured confounders in observational studies. It is common that a comparison between two treatments is focused on and that only subjects receiving one of these two treatments are considered in the analysis even though more than two treatments are available. In this paper, we provide empirical and theoretical evidence that the IV methods may result in biased treatment effects if applied on a data set in which subjects are preselected based on their received treatments. We frame this as a selection bias problem and propose a procedure that identifies the treatment effect of interest as a function of a vector of sensitivity parameters. We also list assumptions under which analyzing the preselected data does not lead to a biased treatment effect estimate. The performance of the proposed method is examined using simulation studies. We applied our method on The Health Improvement Network (THIN) database to estimate the comparative effect of metformin and sulfonylureas on weight gain among diabetic patients.

Analyses of healthcare data from 30 million individuals in three countries showed that current use of bisphosphonates may be associated with a small increased risk of cardiac valvulopathy (vs. those not exposed within the previous year), although confounding cannot be entirely ruled out. The observed tendency for decreased valvulopathy risk with cumulative duration of bisphosphonate use >6 months may even indicate a protective effect with prolonged use. Further studies are still needed to evaluate whether bisphosphonates increase or decrease the risk of valvulopathy. INTRODUCTION: A signal of cardiac valve disorders with use of bisphosphonates was identified in the literature and EudraVigilance database, which contains reports of suspected adverse drug reactions from worldwide sources. The aim of this study was to evaluate the association using population-based healthcare data. METHODS: This was a case-control study among users of bisphosphonates and other drugs for osteoporosis in six healthcare databases covering over 30 million individuals in Italy, Netherlands and the UK from 1996 to 2012. Prescriptions/dispensations were used to assess drug exposure. Newly diagnosed cases of cardiac valvulopathy were identified via disease codes/free-text search. Controls were matched to each case by age, sex, database and index date. Adjusted odds ratios (ORs) were estimated using conditional logistic regression for the pooled data and meta-analysis of individual database risk estimates. RESULTS: A small but statistically significant association was found between exposure to bisphosphonates as a class and risk of valvulopathy. Overall risk was 18 % higher (95 % CI 12-23 %) in those currently exposed to any bisphosphonate (mainly alendronate and risedronate) vs. those not exposed within the previous year. Risk of valve regurgitation was 14 % higher (95 % CI 7-22 %). Decreased valvulopathy risk was observed with longer cumulative duration of bisphosphonate use, compared to use of less than 6 months. Meta-analyses of database-specific estimates confirmed results from pooled analyses. CONCLUSIONS: The observed increased risks of cardiac valvulopathy with bisphosphonate use, although statistically significant, were quite small and unlikely to be clinically significant. Further studies are still needed to evaluate whether bisphosphonates increase or decrease the risk of valvulopathy and to investigate possible mechanisms for the association.

AIM: In Europe, pregabalin is approved for treatment of neuropathic pain, general anxiety disorder (GAD) and as adjunctive therapy for epilepsy. The purpose of this study was to assess utilisation of pregabalin in the UK, including patients with a recorded history of substance abuse, from a large general practice database. METHODS: This observational drug utilisation study (DUS) analysed pregabalin prescription data from the UK Health Improvement Network primary care database between September 2004 and July 2009. Patient demographics, diagnoses (by READ codes) and pregabalin dosing data were collected. Diagnosis codes were used as proxy for approved indication for pregabalin. RESULT: A cohort of 18,951 patients was prescribed pregabalin; dosing information was available for 13,480 (71.1%). Median age of patients was 58 years, and majority were female (60.1%). Median (interquartile range) prescribed average daily dose (ADD) of pregabalin for all patients was 150.0 (162.5) mg/day; this was highest in patients with epilepsy (191.9 mg/day), followed by neuropathic pain (158.0 mg/day) and GAD (150.0 mg/day). Only 1.0% (136/13,480) of patients were prescribed an ADD of pregabalin over the maximum approved dose of 600 mg/day. Of these, 18.4% (25/136) of patients had a history of substance abuse compared with 14.0% (1884/13,480) in the full population. CONCLUSION: Data from this DUS indicated that the majority of pregabalin prescribing in the UK was consistent with product labelling. The proportion of patients with prescribed ADD > 600 mg/day was small and with a similar proportion with a history of substance abuse as in the full population.