Publications
scientifiques

PURPOSE: This study aimed to describe the management of antihypertensive medications in pregnancy by general practitioners in the UK and compare it with current guidelines. METHODS: We used electronic medical records from The Health Improvement Network database from 1996 to 2010 to identify completed pregnancies. The study cohort included the first pregnancy identified during the study period in women aged 13-49 years. Information on both hypertension diagnoses and prescription of specific antihypertensive medications within the 90 days before the last menstrual period (LMP) and during pregnancy was ascertained from electronic medical records. RESULTS: Among 148,544 eligible pregnancies, we identified 1995 (1.3%) during which the women had pre-existing hypertension diagnosed by the LMP date. Overall, the prevalence of antihypertensive medications during the first trimester was 1.5%; beta-blockers were the most commonly prescribed antihypertensive. Among women with pre-existing hypertension, 36% were prescribed an antihypertensive medication during the 90 days before the LMP. Among those, 9.6% and 22.2% had discontinued their medication by the first and second trimesters, respectively. For contraindicated drugs such as angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers, the corresponding discontinuation rates were around 25% and 70%. Women who switched therapy received preferably either methyldopa or an alpha/beta-blocker. CONCLUSIONS: In this population of UK pregnant women, prescription patterns of antihypertensive medications were dominated by recommended treatments, although some patients continued on contraindicated drugs throughout pregnancy or switched to preferred agents in a delayed fashion.

BACKGROUND: Injury is a significant cause of childhood death and can result in substantial long-term disability. Injuries are more common in children from socio-economically deprived families, contributing to health inequalities between the most and least affluent. However, little is known about how the relationship between injuries and deprivation has changed over time in the UK. METHODS: We conducted a cohort study of all children under 5 registered in one of 495 UK general practices that contributed medical data to The Health Improvement Network database between 1990-2009. We estimated the incidence of fractures, burns and poisonings by age, sex, socio-economic group and calendar period and adjusted incidence rate ratios (IRR) comparing the least and most socio-economically deprived areas over time. Estimates of the UK annual burden of injuries and the excess burden attributable to deprivation were derived from incidence rates. RESULTS: The cohort of 979,383 children experienced 20,804 fractures, 15,880 burns and 10,155 poisonings, equating to an incidence of 75.8/10,000 person-years (95% confidence interval 74.8-76.9) for fractures, 57.9 (57.0-58.9) for burns and 37.3 (35.6-38.0) for poisonings. Incidence rates decreased over time for burns and poisonings and increased for fractures (p<0.001 test for trend for each injury). They were significantly higher in more deprived households (IRR test for trend p<0.001 for each injury type) and these gradients persisted over time. We estimate that 865 fractures, 3,763 burns and 3,043 poisonings could be prevented each year in the UK if incidence rates could be reduced to those of the most affluent areas. CONCLUSIONS: The incidence of burns and poisonings declined between 1990 and 2009 but increased for fractures. Despite these changes, strong socio-economic inequalities persisted resulting in an estimated 9,000 additional medically-attended injuries per year in under-5s.

BACKGROUND: There is increasing evidence that impaired mucosal defence mechanisms are implicated in the pathogenesis of the functional gastrointestinal disorders (FGIDs), allowing inappropriate immune activation. AIM: To test the hypothesis that an excess of autoimmune disorders among sufferers, using a large primary care database to examine this. METHODS: Cases were diagnosed with FGIDs - irritable bowel syndrome (IBS), functional dyspepsia (FD), chronic idiopathic constipation (CIC), and multiple FGIDs. Controls were those without FGIDs. Prevalence of autoimmune disorders was compared between cases and controls using odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: We included 23,471 patients (mean age 51.4 years, 66.1% female). Prevalence of autoimmune disorders was greater among all FGIDs, compared with controls without. In those with FD (OR 1.35; 95% CI 1.12-1.63), CIC (OR 1.75; 95% CI 1.11-2.75), or multiple FGIDs (OR 1.49; 95% CI 1.25-1.77) this was statistically significant after controlling for age and gender. Rheumatological autoimmune disorders were significantly more frequent in those with FD (OR 1.44; 95% CI 1.15-1.80), CIC (OR 1.84; 95% CI 1.08-3.13), or multiple FGIDs (OR 1.53; 95% CI 1.24-1.88), after controlling for age and gender. However, endocrine autoimmune disorders were no more frequent in those with FGIDs, after controlling for age and gender. CONCLUSIONS: In a large sample of primary care patients, there was a significantly higher prevalence of autoimmune disorders among those with FD, CIC, or multiple FGIDs not explained by differences in age or gender. We were unable to control for concomitant drug use, which may partly explain this association.

BACKGROUND: Women prescribed antipsychotics face the dilemma on whether to continue medication in pregnancy in terms of balancing risks and benefits. Previous research on other psychotropic medications suggests that many women discontinue treatment in early pregnancy. However, very limited evidence exists on discontinuation of antipsychotic medication. METHODS: We identified 495,953 pregnant women from THIN primary care database. Kaplan-Meier plots were used to examine time to last antipsychotic prescription. Poisson regression was used to examine characteristics of those who stopped treatment during pregnancy. RESULTS: There has been an overall increase in prevalence of antipsychotic prescribing since 2007. However, antipsychotics were more likely to be stopped in pregnant than non-pregnant women. Only 107/279 (38%) of women on atypical antipsychotics and 39/207 (19%) of women on typical antipsychotics before pregnancy still received treatment at the start of third trimester. Older women were more likely to continue typical antipsychotic treatment in pregnancy (35+ versus <25 years risk ratio: 3.09 [95% CI 1.76, 5.44]). Likewise, those who received typical antipsychotics for longer periods before were most likely to continue treatment in pregnancy (12+ versus <6 months: RR: 3.12 [95% CI 1.97, 4.95]). For atypical antipsychotics length and dose of prior prescribing were also associated with continuation in pregnancy. CONCLUSIONS: Pregnancy was a major determinant of cessation of antipsychotics. Only 38% of women on atypical and 19% on typical antipsychotics were still prescribed the drug in the third trimester. Duration of prior treatment, maternal age as well as dose was significantly associated with continued treatment of antipsychotics in pregnancy.

PURPOSE: The aims of this study are to examine the validity of diagnostic codes for psoriatic arthritis in The Health Improvement Network (THIN) and to examine the agreement between General Practitioner (GP) report and prescription records for disease modifying antirheumatic drugs (DMARDs). METHODS: Questionnaires were sent to the GPs of 100 randomly selected patients with at least one medical record code for psoriatic arthritis. The positive predictive value (PPV) for a GP confirmed diagnosis was calculated, and alternative algorithms were examined to determine which method resulted in the highest PPV. RESULTS: The PPV for a single code for psoriatic arthritis was 85% (95%CI: 75.8-91.7%). Adding a prescription for a DMARD increased the PPV to 91% but with a substantial loss in sensitivity. Agreement between GPs and prescription data for use of an oral DMARD was 69%. CONCLUSIONS: The diagnosis codes for psoriatic arthritis used in THIN are valid. All prescriptions for DMARDs may not be accounted for in THIN.

BACKGROUND: The UK incidence of pancreatic ductal adenocarcinoma (PDAC) is approximately 9/100,000 population compared with 1-2/100,000 for biliary tract cancer (BTC). This study explores the incidence of these cancers over time and the influence of socio-demographic and geographic factors in a UK primary care cohort. METHODS: This study uses data from a large UK primary care database, The Health Improvement Network (THIN). All adult patients contributing data to THIN between January 2000 and December 2010 were included. Annual incidence rates were calculated, adjusted for age, gender, time period, deprivation score (Townsend quintile) and strategic health authority. RESULTS: From 2000-2010, the annual incidence of PDAC increased by an average of 3% per year (95% CI 1.00-4.00%) and BTC by 4% (95% CI 2.00-6.00%). Incidence of both cancers increased steeply with age and was higher in men. BTC was associated with increasing deprivation (most deprived versus least deprived quintile (OR: 1.45 [95% CI: 1.17, 1.79.]). CONCLUSIONS: The overall incidence of both cancers is low but increasing. Variations in incidence may reflect changes in coding practice or increased exposure to associated risk factors.

BACKGROUND: Licensing arrangements for nicotine replacement therapy (NRT) in the UK were broadened in 2005 to allow prescribing to pregnant smokers. However, estimates of NRT prescribing in pregnant females in the UK are currently lacking. AIM: To assess trends in NRT prescribing around pregnancy, and variation in prescribing by maternal characteristics. DESIGN AND SETTING: Population-based descriptive study using pregnancy data from The Health Improvement Network primary care database, 2001-2012. METHOD: NRT prescriptions were identified during pregnancy and in the 9 months before and after. Annual prescribing prevalence was calculated. Logistic regression was used to assess females' likelihood of receiving prescriptions by maternal characteristics. RESULTS: Of 388 142 pregnancies studied, NRT was prescribed in 7551 for an average duration of 2 weeks. The prescribing prevalence of NRT increased from 0.03% (0.7% in smokers) in 2001 to 2.6% (11.4% in smokers) in 2005, after which it remained stable. Prescribing prevalence of NRT before and after pregnancy was half the prevalence during pregnancy. The odds of prescribing NRT during pregnancy in smokers increased with socioeconomic deprivation (OR = 1.29, 95% CI = 1.15 to 1.45 in the most compared with the least deprived group). Prescribing was 33% higher in pregnant smokers with asthma (OR = 1.33, 95% CI = 1.22 to 1.45) and mental illness (OR = 1.33, 95% CI = 1.23 to 1.44) compared with smokers without these diagnoses. CONCLUSION: NRT prescribing is higher during pregnancy compared with before and after, and is higher in smokers from more socioeconomically deprived groups, those with asthma or those diagnosed mental illness.

Most implementations of multiple imputation (MI) of missing data are designed for simple rectangular data structures ignoring temporal ordering of data. Therefore, when applying MI to longitudinal data with intermittent patterns of missing data, some alternative strategies must be considered. One approach is to divide data into time blocks and implement MI independently at each block. An alternative approach is to include all time blocks in the same MI model. With increasing numbers of time blocks, this approach is likely to break down because of co-linearity and over-fitting. The new two-fold fully conditional specification (FCS) MI algorithm addresses these issues, by only conditioning on measurements, which are local in time. We describe and report the results of a novel simulation study to critically evaluate the two-fold FCS algorithm and its suitability for imputation of longitudinal electronic health records. After generating a full data set, approximately 70% of selected continuous and categorical variables were made missing completely at random in each of ten time blocks. Subsequently, we applied a simple time-to-event model. We compared efficiency of estimated coefficients from a complete records analysis, MI of data in the baseline time block and the two-fold FCS algorithm. The results show that the two-fold FCS algorithm maximises the use of data available, with the gain relative to baseline MI depending on the strength of correlations within and between variables. Using this approach also increases plausibility of the missing at random assumption by using repeated measures over time of variables whose baseline values may be missing.

STUDY OBJECTIVE: To investigate whether endometrial ablation is associated with increased risk or delayed diagnosis of endometrial cancer compared with medical management of abnormal uterine bleeding. DESIGN: Multi-centered retrospective cohort study (Canadian Task Force classification II-2). SETTING: The study was performed using data from The Health Improvement Network, a representative population-based cohort of patients in 495 outpatient general practitioner practices in the United Kingdom. PATIENTS: Women aged >25 years with abnormal uterine bleeding diagnosed between June 1994 and September 2010. INTERVENTIONS: Endometrial ablation, medical management, or both. MEASUREMENTS AND MAIN RESULTS: A total of 234 721 women met study inclusion and exclusion criteria, 4776 of whom underwent endometrial ablation and the remaining 229 945 received medical management. Cox models compared endometrial cancer rates between ablation and medical management groups using hazard ratios. To investigate a possible diagnostic delay, the median time from bleeding diagnosis to endometrial cancer diagnosis in women in whom endometrial cancer developed was compared using the Mann-Whitney U test. All statistical tests were 2-tailed, with α = .05. During a median observation period of 4.07 years (interquartile range [IQR], 1.88-7.17), endometrial cancer developed in 3 women in the ablation group and 601 women in the medical management group (ablation hazard ratio, 0.45; 95% confidence interval, 0.15-1.40; p = .17). Median time to diagnosis was 237 in the ablation group, and 299 days in the medical management group (ablation IQR, 155-1350; medical management IQR, 144-1133.5; p = .99). Adjusted and sensitivity analyses did not change the results. CONCLUSIONS: No difference was observed in endometrial cancer rates, and there was no delay in diagnosis when comparing endometrial ablation vs medical management. Further studies are needed to investigate the effect of previous ablation exposure on histology or cancer stage at manifestation of endometrial cancer.

BACKGROUND: Rotator cuff disease is a common condition in the general population, but relatively little is known about its associated risk factors. MATERIALS AND METHODS: We have undertaken a large case-control study using The Health Improvement Network database to assess and to quantify the relative contributions of some constitutional and environmental risk factors for rotator cuff disease in the community. Our data set included 5000 patients with rotator cuff disease who were individually matched with a single control by age, sex, and general practice (primary care practice). RESULTS: The median age at diagnosis was 55 years (interquartile range, 44-65 years). Multivariate analysis showed that the risk factors associated with rotator cuff disease were Achilles tendinitis (odds ratio [OR] = 1.78), trigger finger (OR = 1.99), lateral epicondylitis (OR = 1.71), and carpal tunnel syndrome (OR = 1.55). Oral corticosteroid therapy (OR = 2.03), oral antidiabetic use (OR = 1.66), insulin use (OR = 1.77), and "overweight" body mass index of 25.1 to 30 (OR = 1.15) were also significantly associated. Current or previous smoking history, body mass index of greater than 30, any alcohol intake, medial epicondylitis, de Quervain syndrome, cubital tunnel syndrome, and rheumatoid arthritis were not found to be associated with rotator cuff disease. CONCLUSIONS: We have identified a number of comorbidities and risk factors for rotator cuff disease. These include lateral epicondylitis, carpal tunnel syndrome, trigger finger, Achilles tendinitis, oral corticosteroid use, and diabetes mellitus. The findings should alert the clinician to comorbid pathologic processes and guide future research into the etiology of this condition.