Publications
scientifiques

OBJECTIVE: Gout is the most common inflammatory arthritis. Erectile dysfunction (ED) is common in the general population; however, evidence regarding ED among patients with gout is limited. Our purpose was to study the association between incident gout and the risk of incident ED in the general population. METHODS: We conducted a cohort study using The Health Improvement Network, an electronic medical record database in the United Kingdom. Up to 5 individuals without gout were matched to each case of incident gout by age, enrollment time, and body mass index (BMI). Multivariate HR for ED were calculated after adjusting for smoking, alcohol consumption, comorbidities, and medication use. RESULTS: We identified 2290 new cases of ED among 38,438 patients with gout (mean age 63.6 yrs) and 8447 cases among 154,332 individuals in the comparison cohort over a 5-year median followup (11.9 vs 10.5 per 1000 person-years, respectively). Univariate (matched for age, entry time, and BMI) and multivariate HR for ED among patients with gout were 1.13 (95% CI 1.08-1.19) and 1.15 (95% CI 1.09-1.21), respectively. In our sensitivity analysis, by restricting gout cases to those receiving anti-gout treatment (n = 27,718), the magnitude of relative risk was stronger than the primary analysis (multivariate HR 1.31, 95% CI 1.23-1.39). CONCLUSION: This population-based study suggests that gout is associated with an increased risk of developing ED, supporting a possible role for hyperuricemia and inflammation as independent risk factors for ED.

The 73rd Annual Meeting of The Canadian Rheumatology Association (CRA) was held at the JW Marriott Parq Vancouver, Vancouver, British Columbia, Canada, February 21-24, 2018. The program consisted of presentations covering original research, symposia, awards, and lectures.

We sought to examine the risk of psychiatric symptoms associated with a first prescription for specific antiepileptic drugs (AEDs) used in monotherapy in a general cohort of patients with epilepsy. We used The Health Improvement Network database (comprising the years 2000-2012) to identify incident patients with epilepsy. The index date was that on which they met the case definition for epilepsy, and analyses only included patients who remained on monotherapy or received no AED therapy following diagnosis to avoid confounding by polytherapy. Psychiatric symptoms were defined using mental health clinical or treatment (medical or therapeutic) code. We analyzed the AED of interest as a time-varying covariate in multivariate Cox proportional hazard regression models controlling for confounding factors. We identified 9595 patients with incident epilepsy, 7400 of whom (77%) received a first-recorded AED prescription. Prescriptions for newer generation AEDs (lamotrigine and levetiracetam) steadily increased (constituting over 30% of all AED prescriptions by 2012) while valproate use significantly declined in females (~40% in 2002 to just over 20% by 2012). A total of 2190 patients were first exposed to carbamazepine (29.3%) and 222 to lamotrigine (3%), both of which were associated with a lower hazard of any coded psychiatric symptom or disorder in multivariate analyses (hazard ratio [HR]: 0.84, 95% confidence interval [95% CI]: 0.73-0.97; p = 0.02 and HR: 0.83, 95% CI: 0.70-0.99; p = 0.03, respectively, for carbamazepine and lamotrigine). Carbamazepine was also associated with a lower hazard for depression (HR: 0.81; 95% CI: 0.69-0.96; p = 0.013) and anxiety (HR: 0.77; 95% CI: 0.63-0.95; p = 0.013) in secondary analyses. This study provides evidence that carbamazepine and lamotrigine are associated with lower hazards for psychiatric symptoms following a diagnosis of epilepsy. These estimates can be used in clinical settings, and the precision should improve with more contemporary data that include larger proportions of newer generation AEDs.

OBJECTIVE: Post-stroke spasticity (PSS) is a common complication following stroke. This study describes the differences in healthcare resource utilization between patients who do and do not develop PSS in the UK. METHODS: Adults registered in The Health Improvement Network database with a recorded stroke between 2007 and 2011 were included. PSS was identified through Read codes; machine learning was used to retrospectively identify unrecorded PSS events. Patients with diagnosed or predicted PSS in the 12 months after stroke were matched to those with no PSS on age, sex, number of strokes, socioeconomic status, and comorbidities using the nearest neighbor algorithm. Utilization and costs associated with general practitioner visits, nurse visits, hospitalizations, referrals to specialists, laboratory tests, and medications in the 12 months after stroke were compared. RESULTS: Overall, 2,951 PSS cases were matched to 37,753 controls. During the first year, more PSS cases visited a physiotherapist (19% vs 7%) and occupational therapist (12% vs 5%) compared to controls. A greater proportion of cases were also referred to specialists (76% vs 64%) and hospitalized (33% vs 9%) compared to controls. Medication for spasticity was, on average, 14.68 prescriptions for cases and 5.64 for controls. Total mean costs per patient were £1,270 (standard deviation [SD] = 772) and £635 (SD = 273) for cases and controls, respectively. CONCLUSION: Costs after stroke for patients developing PSS are twice as high compared to patients who do not develop it, with the major driver being the number of hospital admissions. This highlights the need for better recording and closer management of PSS.

BACKGROUND: We do not know how primary care treatment of depression varies by age across both psychotropic medication and psychological therapies. METHODS: Cohort study including 19 710 people aged 55+ with GP recorded depression diagnoses and 26 276 people with recorded depression symptoms during the period 2009-2013, from 373 General Practices in The Health Improvement Network (THIN) database in England. Main outcomes were initiation of treatment with anti-depressants, anxiolytics, hypnotics, anti-psychotic drugs, referrals to psychological therapies within 6 months of onset. RESULTS: Treatment rates with antidepressants are high for those recorded with new depression diagnoses (87.1%) or symptoms of depression (58.7%). Treatment in those with depression diagnoses varies little by age. In those with depressive symptoms there was a J-shaped pattern with reduced antidepressant treatment in those in their 60s and 70s followed by increased treatment in the oldest age groups (85+ years), compared with those aged 55-59 years. Other psychotropic drug prescribing (hypnotics/anxiolytics, antipsychotics) all increase with increasing age. Recorded referrals for psychological therapies were low, and decreased steadily with increasing age, such that women aged 75-79 years with depression diagnoses had around six times lower odds of referral (OR 0.17, 95% CI 0.1-0.29) than those aged 55-59 years, and men aged 80-84 years had around seven times lower (OR 0.14, 95% CI 0.05-0.36). CONCLUSIONS: The oldest age groups with new depression diagnoses and symptoms have fewer recorded referrals to psychological therapies, and higher psychotropic drug treatment rates in primary care. This suggests potential inequalities in access to psychological therapies.

AIMS: To evaluate the association between use of non-insulin antidiabetics in early pregnancy and the risk of miscarriages, stillbirths and major structural malformations. MATERIALS AND METHODS: A cohort of 1511 pregnant women with pre-gestational diabetes linked to live births was identified using electronic medical records from The Health Improvement Network (THIN) for the period 1995 to 2012. Information on prescriptions, foetal outcomes and potential confounders was ascertained from both codes and free text in the THIN database. Odds ratios (OR) and 95% confidence intervals (CI) of adverse foetal outcomes in women treated with non-insulin antidiabetics during the first trimester compared to those on insulin were estimated using logistic regression to adjust for type of diabetes, glycaemic control and other maternal characteristics. RESULTS: Among 311 pregnant women on non-insulin antidiabetics, 21.9% had a miscarriage and 1.6% a stillbirth; 1.9% of live births had major malformations. The corresponding frequencies for the 883 women on insulin were 13.3%, 1.7% and 9.6%. Insulin users more often had type 1 diabetes and poor glycaemic control. Compared to women with type 1 diabetes, those with type 2 diabetes had a higher risk of miscarriages (20.5% vs 12.8%) but a lower prevalence of malformations (4.0% vs 9.2%). Compared to women with HbA1c ≤7%, those with HbA1c >7% had a higher prevalence of malformations (12.6% vs 2.7%). After adjustment for diabetes type and glycaemic control, compared to insulin, non-insulin antidiabetic patients were associated with an OR for miscarriage of 1.19 (95% CI, 0.75-1.89), for stillbirths of 0.65 (95% CI, 0.16-2.58), and for major malformations of 0.25 (95% CI, 0.08-0.84). CONCLUSION: Among women with diabetes, use of non-insulin antidiabetics early in pregnancy was not associated with greater risks of foetal losses or major malformations than was insulin.

OBJECTIVE: Observational studies analyzing multiple exposures simultaneously have been limited by difficulty distinguishing relevant results from chance associations due to poor specificity. Set-based methods have been successfully used in genomics to improve signal-to-noise ratio. We present and demonstrate medication class enrichment analysis (MCEA), a signal-to-noise enhancement algorithm for observational data inspired by set-based methods. MATERIALS AND METHODS: We used The Health Improvement Network database to study medications associated with Clostridium difficile infection (CDI). We performed case-control studies for each medication in The Health Improvement Network to obtain odds ratios (ORs) for association with CDI. We then calculated the association of each pharmacologic class with CDI using logistic regression and MCEA. We also performed simulation studies in which we assessed the sensitivity and specificity of logistic regression compared to MCEA for ORs 0.1-2.0. RESULTS: When analyzing pharmacologic classes using logistic regression, 47 of 110 pharmacologic classes were identified as associated with CDI. When analyzing pharmacologic classes using MCEA, only fluoroquinolones, a class of antibiotics with biologically confirmed causation, and heparin products were associated with CDI. In simulation, MCEA had superior specificity compared to logistic regression across all tested effect sizes and equal or better sensitivity for all effect sizes besides those close to null. DISCUSSION: Although these results demonstrate the promise of MCEA, additional studies that include inpatient administered medications are necessary for validation of the algorithm. CONCLUSIONS: In clinical and simulation studies, MCEA demonstrated superior sensitivity and specificity for identifying pharmacologic classes associated with CDI compared to logistic regression.

BACKGROUND: Long-term, low-level exposure to toxic elements in soil may be harmful to human health but large longitudinal cohort studies with sufficient follow-up time to study these effects are cost-prohibitive and impractical. Linkage of routinely collected medical outcome data to systematic surveys of soil quality may offer a viable alternative. METHODS: We used the Geochemical Baseline Survey of the Environment (G-BASE), a systematic X-ray fluorescence survey of soil inorganic chemistry throughout England and Wales to obtain estimates of the concentrations of 15 elements in the soil contained within each English and Welsh postcode area. We linked these data to the residential postcodes of individuals enrolled in The Health Improvement Network (THIN), a large database of UK primary care medical records, to provide estimates of exposure. Observed exposure levels among the THIN population were compared with expectations based on UK population estimates to assess representativeness. RESULTS: Three hundred seventy-seven of three hundred ninety-five English and Welsh THIN practices agreed to participate in the linkage, providing complete residential soil metal estimates for 6,243,363 individuals (92% of all current and former patients) with a mean period of prospective computerised medical data collection (follow-up) of 6.75 years. Overall agreement between the THIN population and expectations was excellent; however, the number of participating practices in the Yorkshire & Humber strategic health authority was low, leading to restricted ranges of measurements for some elements relative to the known variations in geochemical concentrations in this area. CONCLUSIONS: The linked database provides unprecedented population size and statistical power to study the effects of elements in soil on human health. With appropriate adjustment, results should be generalizable to and representative of the wider English and Welsh population.

The benefits of using electronic health records (EHRs) for disease risk screening and personalized health-care decisions are being increasingly recognized. Here we present a computationally feasible statistical approach with which to address the methodological challenges involved in utilizing historical repeat measures of multiple risk factors recorded in EHRs to systematically identify patients at high risk of future disease. The approach is principally based on a 2-stage dynamic landmark model. The first stage estimates current risk factor values from all available historical repeat risk factor measurements via landmark-age-specific multivariate linear mixed-effects models with correlated random intercepts, which account for sporadically recorded repeat measures, unobserved data, and measurement errors. The second stage predicts future disease risk from a sex-stratified Cox proportional hazards model, with estimated current risk factor values from the first stage. We exemplify these methods by developing and validating a dynamic 10-year cardiovascular disease risk prediction model using primary-care EHRs for age, diabetes status, hypertension treatment, smoking status, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol in 41,373 persons from 10 primary-care practices in England and Wales contributing to The Health Improvement Network (1997-2016). Using cross-validation, the model was well-calibrated (Brier score = 0.041, 95% confidence interval: 0.039, 0.042) and had good discrimination (C-index = 0.768, 95% confidence interval: 0.759, 0.777).

STUDY QUESTION: What is the incidence/prevalence of type 2 diabetes in women with polycystic ovary syndrome (PCOS) and the economic burden associated with PCOS in the UK? SUMMARY ANSWER: The incidence and prevalence of type 2 diabetes in women with PCOS are 3-33 per 1000 person years and 26.5%, respectively, with an associated annual healthcare burden of at least £237 million in the UK. WHAT IS KNOWN ALREADY: Although observational studies have been designed to assess the incidence of diabetes in women with PCOS, these have been open to criticism because of short periods of follow-up, small sample sizes or invalidated diagnosis of PCOS. Only one study has estimated the healthcare-related economic burden of PCOS, reporting a cost of $4.36 billion per year in the USA. STUDY DESIGN, SIZE, DURATION: This was a modelling study using individual patient data from a UK primary care database between 2004 and 2014 and aggregate data from the literature to obtain conversion rates through disease progression of PCOS. A simulation approach was applied to model the population dynamics of PCOS over a follow-up period of 25 years. PARTICIPANTS/MATERIALS, SETTING, METHODS: A total of 14 135 women with PCOS or symptoms indicative of PCOS were selected from the primary care database to estimate the incidence of confirmed diagnosis of PCOS and diagnosis of type 2 diabetes. A 'virtual' cohort including the entire PCOS population (size estimated from the UK census data) was simulated to model the population dynamics of PCOS. The economic and utility analyses were further conducted from a healthcare perspective. MAIN RESULTS AND THE ROLE OF CHANCE: The peak conversion rate from possible to diagnosed PCOS was 121 per 1000 person-year (PY). The maximal incidence of type 2 diabetes was 33 per 1000 PY. The estimated prevalence of diabetes in the PCOS population was 26.5% (95% interval: 25.4-27.8%) during a 25-year follow-up. The annual healthcare burden of PCOS based on our conservative estimate is at least £237 million for the follow-up period examined. LIMITATIONS, REASONS FOR CAUTION: Due to lack of data, a full economic evaluation including healthcare costs of all the comorbidities associated with PCOS was not possible. Simplification of the real-world situation represented by the model may be a concern. WIDER IMPLICATIONS OF THE FINDINGS: This study suggests that a large number of women with symptoms indicative of PCOS never receive a definitive diagnosis yet can suffer from a rapid conversion to diabetes. This significantly reduces the quality of life for individual patients and incurs high costs for healthcare providers. As the risk of diabetes in women with PCOS is similar to that seen in populations at high risks of diabetes, it is possible that including them in national screening programmes may be cost effective. STUDY FUNDING/COMPETING INTEREST(S): There was no funding for the current study. There are no conflicts of interest. TRIAL REGISTRATION NUMBER: Not applicable.

BACKGROUND: Hydroxychloroquine (HCQ) retinopathy may be more common than previously recognized; recent ophthalmology guidelines have revised recommendations from ideal body weight (IBW)-based dosing to actual body weight (ABW)-based dosing. However, contemporary HCQ prescribing trends in the UK remain unknown. METHODS: We examined a UK general population database to investigate HCQ dosing between 2007 and 2016. We studied trends of excess HCQ dosing per ophthalmology guidelines (defined by exceeding 6.5 mg/kg of IBW and 5.0 mg/kg of ABW) and determined their independent predictors using multivariable logistic regression analyses. RESULTS: Among 20,933 new HCQ users (78% female), the proportions of initial HCQ excess dosing declined from 40% to 36% using IBW and 38% to 30% using ABW, between 2007 and 2016. Among these, 47% of women were excess-dosed (multivariable OR 12.52; 95% CI 10.99-14.26) using IBW and 38% (multivariable OR 1.98; 95% CI,1.81-2.15) using ABW. Applying IBW, 37% of normal and 44% of obese patients were excess-dosed; however, applying ABW, 53% of normal and 10% of obese patients were excess-dosed (multivariable ORs = 1.61 and 0.1 (reference = normal); both p < 0.01). Long-term HCQ users showed similar excess dosing. CONCLUSION: A substantial proportion of HCQ users in the UK, particularly women, may have excess HCQ dosing per the previous or recent weight-based guidelines despite a modest decline in recent years. Over half of normal-BMI individuals were excess-dosed per the latest guidelines. This implies the potential need to reduce dosing for many patients but also calls for further research to establish unifying evidence-based safe and effective dosing strategies.

OBJECTIVES: Major depressive disorder (MDD) is associated with increased levels of systemic proinflammatory cytokines, including tumour necrosis factor alpha. As these cytokines are pathogenic in autoimmune diseases such as rheumatoid arthritis (RA), our aim was to explore on a population-level whether MDD increases the risk of developing RA. METHODS: A retrospective cohort study was conducted using The Health Improvement Network (THIN) database (from 1986 to 2012). Observation time was recorded for both the MDD and referent cohorts until patients developed RA or were censored. Cox proportional hazards models were used to determine the risk of developing RA among patients with MDD, accounting for age, sex, medical comorbidities, smoking, body mass index and antidepressant use. RESULTS: A cohort of 403 932 patients with MDD and a referent cohort of 5 339 399 patients without MDD were identified in THIN. Cox proportional hazards models revealed a 31% increased risk of developing RA among those with MDD in an unadjusted model (HR=1.31, 95% CI 1.25 to 1.36, p<0.0001). When adjusting for all covariates, the risk remained significantly increased among those with MDD (HR=1.38, 95% CI 1.31 to 1.46, p<0.0001). Antidepressant use demonstrated a confounding effect that was protective on the association between MDD and RA. CONCLUSION: MDD increased the risk of developing RA by 38%, and antidepressants may decrease this risk in these patients. Future research is necessary to confirm the underlying mechanism of MDD on the pathogenesis of RA.

OBJECTIVES: The aim of this study was to estimate the patterns of care and annual levels of healthcare resource use attributable to managing pressure ulcers (PUs) in clinical practice in the community by the UK's National Health Service (NHS), and the associated costs of patient management. METHODS: This was a retrospective cohort analysis of the records of 209 patients identified within a randomly selected population of 6000 patients with any type of wound obtained from The Health Improvement Network (THIN) Database, who developed a PU in the community and excluded hospital-acquired PUs. Patients' characteristics, wound-related health outcomes and healthcare resource use were quantified over 12 months from initial presentation, and the corresponding total NHS cost of patient management was estimated at 2015/2016 prices. RESULTS: 50% of all the PUs healed within 12 months from initial presentation, but this varied between 100% for category 1 ulcers and 21% for category 4 ulcers. The mean time to healing ranged from 1.0 month for a category 1 ulcer to 8 months for a category 3/4 ulcer and 10 months for an unstageable ulcer. Patients were predominantly managed in the community by nurses with minimal clinical involvement of specialist clinicians. Up to 53% of all the ulcers may have been clinically infected at the time of presentation, and 35% of patients subsequently developed a putative wound infection a mean 4.7 months after initial presentation. The mean NHS cost of wound care over 12 months ranged from £1400 for a category 1 ulcer to >£8500 for the other categories of ulcer. Additionally, the cost of managing an unhealed ulcer was 2.4 times more than that of managing a healed ulcer (mean of £5140 vs £12 300 per ulcer). CONCLUSION: This study provides important insights into a number of aspects of PU management in clinical practice in the community that have been difficult to ascertain from other studies, and provides the best estimate available of NHS resource use and costs with which to inform policy and budgetary decisions.