Publications
scientifiques

BACKGROUND: The aim of this study was to use a longitudinal computerised health care dataset (The Health Improvement Network) to provide information on the overall incidence of cancer, and on the incidence of organ-specific cancers, in people with idiopathic pulmonary fibrosis (IPF) and sarcoidosis in comparison to the general population. METHODS: Incident cases of IPF and sarcoidosis were identified with up to four controls matched by age, gender and general practice. Cancer incidence rates were compared between cohorts using Cox regression and adjusting for age, gender and smoking habit. RESULTS: One thousand and sixty-four incident cases of IPF (mean age at diagnosis 71.5 years; 62.4% male) were identified. Overall, the incidence of cancer was increased in people with IPF compared to the general population (rate ratio 1.51; 95% CI 1.20-1.90), but this was largely due to a marked increase in the incidence of lung cancer (rate ratio 4.96; 95% CI 3.00-8.18). One thousand one hundred and fifty-three incident cases of sarcoidosis (mean age at diagnosis 47.0 years; 47.2% male) were identified. There was an overall increased incidence of cancer in sarcoidosis (rate ratio 1.65; 95% CI 1.22-2.24) and this was largely explained by an increase in the incidence of skin cancers (rate ratio 1.86; 95% CI 1.11-3.11). CONCLUSIONS: This study provides further evidence of a marked increase in the incidence of lung cancer in people with IPF, but we found no increase in the risk of other cancers. People with sarcoidosis did have an increase risk of skin cancers, but not cancers at other sites.

AIMS: The aim of this retrospective cohort study was to estimate the time to insulin initiation in patients with Type 2 diabetes inadequately controlled on oral glucose-lowering agents (OGLAs). METHODS: Insulin-naïve patients failing on OGLAs were identified from The Health Improvement Network database, which collects records from general practices throughout the UK. Patients were included if they were aged > or = 40 years, had concomitant prescriptions for > or = 2 OGLAs, and > or = 1 year of available records prior to the first occurrence of HbA(1c) > or = 8.0% after > or = 90 days of OGLA polytherapy at > or = 50% of maximum recommended dosages. RESULTS: A total of 2501 eligible patients with Type 2 diabetes who had an HbA(1c) above the OGLA failure threshold of > or = 8.0% were identified (54.0% male; 30.9% aged 60-69 years). It was estimated that if all the eligible patients were followed for 5 years, 25% would initiate insulin within 1.8 years of OGLA failure (95% CI 1.6-2.0), and 50% within 4.9 years (95% CI 4.6-5.8). The presence of diabetes-related complications had no substantial impact on the time to insulin initiation. CONCLUSIONS: This study found that 25% of patients with Type 2 diabetes had insulin initiation delayed for at least 1.8 years, and 50% of patients delayed starting insulin for almost 5 years after failure of glycaemic control with OGLA polytherapy, even in the presence of diabetes-related complications. Interventions that reduce this delay to insulin initiation are required to help achieve and maintain recommended glycaemic targets in patients with Type 2 diabetes.

BACKGROUND: Use of a long-acting inhaled bronchodilator, either an anticholinergic or a beta-adrenergic receptor agonist (beta-agonist), is recommended for maintenance treatment of chronic obstructive pulmonary disease (COPD). In COPD, the organ system most frequently requiring medical care, other than the respiratory system, is the cardiac system. OBJECTIVES: To compare the risk of total mortality and certain respiratory and cardiac adverse events among users of the two types of recommended long-acting bronchodilators, we conducted a cohort study. Specifically, the study compared the safety of the only approved long-acting anticholinergic, tiotropium bromide, with the single-ingredient long-acting beta-agonists (LABAs) salmeterol or formoterol in a broad population of users. METHODS: We used automated general practitioner data from the UK THIN (The Health Information Network) database as the data source for this study. We used Cox proportional hazards models to compute hazard ratio (HR) estimates and 95% CI controlling for propensity scores comprising various baseline demographic variables, medical therapies and illnesses. RESULTS: The 1061 tiotropium users and 1801 LABA users were similar with regard to risk of total mortality (HR 0.93; 95% CI 0.59, 1.44) and most cardiac events, including angina (HR 0.77; 95% CI 0.37, 1.59), atrial fibrillation or flutter (HR 0.60; 95% CI 0.25, 1.42), myocardial infarction (HR 1.29; 95% CI 0.45, 3.66) and tachycardia (HR 0.66; 95% CI 0.29, 1.51). Though imprecise, there was evidence of a decreased risk of heart failure (HR 0.65; 95% CI 0.37, 1.12) in tiotropium users. As regards respiratory endpoints, the risk of COPD exacerbation (HR 1.15; 95% CI 0.79, 1.67) and pneumonia (HR 1.11; 95% CI 0.38, 3.26) were similar among users of each type of drug, although there was a decreased risk of asthma exacerbation (HR 0.41; 95% CI 0.26, 0.64) in tiotropium users compared with LABA users. CONCLUSIONS: Users of tiotropium and single-ingredient LABA had similar risk of total mortality and cardiovascular endpoints. The decreased risk of asthma exacerbations with tiotropium may be due to residual confounding by indication. Confidence limits for most events include reduced risks for tiotropium and also small increases in risk. Nevertheless, the point estimates suggest that tiotropium was associated with a lower risk of each cardiac event except myocardial infarction. However, the small number of cases means that further studies are needed to confirm these results.

We determined the trends in incidence of skin basal cell carcinoma (BCC) using a primary care population-based cohort study in the UK. 11,113 adults with a BCC diagnosis were identified from a total of 7.22 million person-years of data between 1996 and 2003 from the Health Improvement Network database. From a random subsample of BCC cases identified from the database, 93% were confirmed by hospital letter and/or pathology report. The incidence of BCC was 153.9 per 100,000 person-years (95% CI 151.1, 156.8) and was slightly higher in men as compared to women (Incidence Rate Ratio 1.10, 95% CI 1.06, 1.14). There was a 3% increase year on year across the study period (IRR 1.03, 95% CI 1.01, 1.04), with the largest increase in incidence seen in the 30-39 year age groups, although this did not reach statistical significance. Our study indicates 53,000 new cases of BCC are estimated every year in the UK and figures are continuing to rise on a yearly basis. Incidence rates are highest for men and in particular in the older age categories. These findings are consistent with those reported for various other populations. We have also found an increase in incidence in ages 30-39, which may suggest a cohort effect of increasing ultraviolet exposure in successive younger generations. This may have a huge public and service impact in future years in countries such as the UK, with predominantly fair-skinned population, with high leisure exposure to ultraviolet light. Our findings underline the need for more elaborate preventive measures.

OBJECTIVES: Cost-effectiveness analyses are routinely based on data from group averages, restricting its generalizibility to those with below- or above-average risk. A pharmaco-economic model that used individualized risks for fractures was developed in order to take into account patient heterogeneity. METHODS: Data were obtained from The Health Improvement Network research database of general practitioners, comprising a UK general population of women aged more than 50 years (N = 330,000). Mortality and hip, vertebral, and other osteoporotic fracture risks for each individual were estimated by age, body mass index (BMI), smoking, and other clinical risk factors. Estimates on costs, EuroQol (EQ-5D) utilities, and treatment efficacy were obtained from a UK national report (the National Institute for Clinical Excellence) and outcomes were simulated over a 10-year period. RESULTS: It was found that the cost per quality-adjusted life-year (QALY) gained was lower in elderly women and in women with fracture history. There was a large variability in the cost-effectiveness with baseline fracture risk and with clinical risk factors. Patients with low BMI (<20) had considerable better cost-effectiveness than patients with high BMI (>or=26). Using a cost-acceptability ratio of 30k pounds per QALY gained, bisphosphonate treatment became cost-effective for patients with a 5-year risk of 9.3% (95% confidence interval [CI] 8.0-10.5%) for osteoporotic fractures and of 2.1% (95% CI 1.5-2.7%) for hip fractures. Including bone mineral density in the risk assessment, the cost per QALY gained was 35k pounds in women at age 60 with a fracture history and a T-score of -2.5 (at age 80, this was 3k pounds). CONCLUSION: A pharmacoeconomic model based on individual long-term risks of fracture improves the selection of postmenopausal women for cost-effective treatment with bisphosphonates.

AIMS: To assess how trends in the incidence of coronary heart disease (CHD) and mortality rates among people with CHD have affected the prevalence of CHD in the UK. METHODS AND RESULTS: A time trend analysis using computerized general practice clinical records of people aged 35 years and over was performed. From 1996 to 2005, age-standardized incidence of CHD decreased by 2.2% in men and 2.3% in women per year (average percentage change). Age-standardized all-cause mortality among those with CHD decreased by 4.5% in men and 3.4% in women per year (average percentage change). Age-standardized prevalence increased by 1.3% in men and 1.7% in women per year (average percentage change). Although the decline in incidence had some impact on limiting the increase in prevalence, its effect was offset by the increase in prevalence occurring as a result of improved survival among people with CHD. CONCLUSION: The results suggest that increasing prevalence is largely due to decreasing mortality among people with CHD. Further increases in prevalence are likely even if the incidence of CHD continues to fall.

BACKGROUND: Previous studies have associated tetracyclines and, perhaps more specifically, minocycline use for the treatment of acne with onset of drug-induced lupus erythematosus (LE). OBJECTIVES: To determine the frequency of LE among those with acne who used antibiotics from the tetracycline class of antibiotics. METHODS: A retrospective cohort study of individuals aged 15-35 years with acne within the practices of the general practice physicians in the U.K. who participate in The Health Information Network (THIN). Our outcome measure was physician reports of LE. RESULTS: We identified 97 694 subjects with acne who were followed for about 520 000 person-years. They were on average about 22 years old and 57.5% were female. Minocycline exposure was noted in 24.8% of our subjects, doxycycline exposure in 15.6%, other tetracyclines in 42.3%, and 17.3% had not received a tetracycline antibiotic. The overall hazard ratio for the association of minocycline to LE was 2.64 (95% confidence interval 1.51-4.66) and when adjusted for age and gender was 3.11 (1.77-5.48). Those affected were often treated for LE. No association was noted for doxycycline and the other tetracyclines. CONCLUSIONS: The use of minocycline and not the other tetracyclines is associated with LE. LE as reported in THIN often required systemic therapy. Overall, the event is uncommon but the risk and benefit of minocycline therapy must be carefully considered.

BACKGROUND: Inhaled corticosteroids are used increasingly to treat people with COPD, but the extent to which these drugs increase the risk of fracture is unclear. AIM: To quantify the dose-response relationship between fracture risk and inhaled corticosteroids in people with COPD, independent of the effects of percent predicted FEV(1) and oral corticosteroids. DESIGN: Nested case-control study. METHODS: Cases and controls were COPD patients aged > or =40 years or more at diagnosis, with a FEV(1) measurement recorded in The Health Improvement Network database, up to 5 July 2005. Cases (people with a fracture event after 1 January 1998, n = 1235) were assigned up to four controls (n = 4598), matched by gender and general practice. RESULTS: Mean FEV(1) was 57.5% in cases, and 58.5% in controls. Inhaled corticosteroids had been prescribed in 69% of cases (median dose 269 mcg/day) and 66% (226 mcg/day) of controls. Oral corticosteroids had been prescribed in 60% of cases (median annual prescription rate 0.6) and 56% of controls (also 0.6 per year). Risk of fracture increased with increasing mean daily doses of inhaled corticosteroid (p for trend 0.007), and was most marked in those whose daily dose was > or =1600 mcg (OR 1.80, 95% CI 1.04-3.11). This effect was virtually unchanged by adjustment for mean percent predicted FEV(1) and annual prescription rate for oral corticosteroids (OR for highest dose exposure 1.74, 95% CI 1.00-3.01). DISCUSSION: Our findings add to the evidence that the use of inhaled corticosteroids is associated with a small increase in fracture risk, particularly at higher doses.

OBJECTIVE: Strategies for initiating statin use among adult patients with diabetes for primary cardiovascular disease (CVD) prevention include treating all patients (assuming diabetes is a coronary risk equivalent) or treating patients who are at risk of developing CVD. The aim of the study was to combine both strategies to derive an appropriate age cutoff for prescribing statins. By considering different strategies, we also aim to assess the effectiveness and efficiency of different strategies to reduce CVD events. RESEARCH DESIGN AND METHODS: This was a cross-sectional primary care population study using electronic patient files from 304 general practitioner practices in England and Wales. Of 60,258 patients with diabetes, 11,005 men and women aged 30-74 years fullfilled criteria for primary CVD prevention. Model outcomes were extrapolated to an estimated national diabetes prevalence of 3.6%. RESULTS: The age transition from a low-risk to a moderate-risk category for diabetic men and women occurred at ages 40.6 and 44.2 years, respectively, and sensitivity and specificity for fulfilling moderate CVD risk criteria were 97.9 and 61.8% for men and 92.0 and 77.0% for women. When applied to the national population, the age cutoff strategies were an effective and efficient strategy, potentially avoiding 11,094 events with a number needed to treat of 25.1. CONCLUSIONS: A strategy to treat all men and women with diabetes aged >40 and 45 years, respectively, with statins showed good compromise between high effectiveness and high efficiency for reducing CVD events. Strategy to intervene if cholesterol was >5 mmol/l was the least effective and efficient in preventing CVD events.

PURPOSE: Standard formulation alpha-blocker use has been associated with an increased risk of hypotension-related adverse events, particularly with cardiovascular indications and on treatment initiation. We investigated the association between those fractures that are commonly due to falls and modified-release doxasozin exposure. METHODS: A case-control study using UK primary care records (the THIN database) showed that 6540 cases aged 50 years or older had a record of first fracture of the hip/femur, humerus or wrist after May 2001; 26 495 controls had no fracture and were matched on practice, year of birth and sex. Exclusions for the primary analysis were previous benign prostatic hypertrophy (BPH), lower urinary tract symptoms (LUTS) or prostate cancer, major trauma at the time of the case's fracture, or a history of falls. Conditional logistic regression estimated the association between modified-release doxasozin treatment and the occurrence of fractures. RESULTS: No increased risk of fractures was associated with current use of modified-release doxasozin, adjusted odds ratios (ORs) 0.90 (95%CI 0.68, 1.19); previous use, adjusted OR 0.92 (95%CI 0.73, 1.16) or the start of a treatment episode (</=28 days), crude OR 0.57 (0.17, 1.92). Including those with a history of BPH, LUTS and prostatic carcinoma had little effect, adjusted OR for current use was 0.95 (95%CI 0.74, 1.21). No significant increased risk of fractures was associated with the start of standard preparation alpha-blockers, crude OR 1.42 (95%CI 0.65, 3.07). CONCLUSIONS: We found no association between modified-release doxasozin use and fractured hip/femur, humerus or wrist in those without a recent history of falls, or trauma.