Search

To derive a simple score for estimating the long-term risk of osteoporotic and hip fracture in individual patients with MS.

To describe the patterns of risks of acute myocardial infarction (MI) during exposure to long-acting beta2-agonists (LABA).

The extent of the heterogeneity of drug utilization among NSAID users has not been extensively studied. We studied the longitudinal prescribing and switching patterns of NSAID users in a 1‐year follow‐up study in four German pharmacies. The study population consisted of 526 persons with an average age of 57 years. We observed that the legend duration of prescription increased with age; 14.3 days for patients aged 44 or younger to 25.1 days for persons 75 years or older, and was dependent on disease chronicity; 16.0 days for acutely ill persons compared to 23.9 days for chronic patients. The average legend duration also varied between different types of NSAIDs, from 18.0 days for ibuprofen to 29.1 days for tiaprofen. Switching from one type of NSAID to another proved to be related to the legend duration of prescription and patient characteristics such as compliance with NSAID therapy, duration of the disease and the effectiveness (poor tolerability or insufficient effect) of the NSAID therapy reported by the patients 4 weeks after recruitment. We conclude that NSAID users cannot be viewed as an homogeneous group of patients with respect to exposure time to the drug. This heterogeneity should be considered in the exposure definition, the ‘time‐window’ design of observational studies dealing with risk comparisons in particular. © 1997 John Wiley & Sons, Ltd.

ObjectiveTo assess the consistency of machine learning and statistical techniques in predicting individual level and population level risks of cardiovascular disease and the effects of censoring on risk predictions.DesignLongitudinal cohort study from 1 January 1998 to 31 December 2018.Setting and participants3.6 million patients from the Clinical Practice Research Datalink registered at 391 general practices in England with linked hospital admission and mortality records.Main outcome measuresModel performance including discrimination, calibration, and consistency of individual risk prediction for the same patients among models with comparable model performance. 19 different prediction techniques were applied, including 12 families of machine learning models (grid searched for best models), three Cox proportional hazards models (local fitted, QRISK3, and Framingham), three parametric survival models, and one logistic model.ResultsThe various models had similar population level performance (C statistics of about 0.87 and similar calibration). However, the predictions for individual risks of cardiovascular disease varied widely between and within different types of machine learning and statistical models, especially in patients with higher risks. A patient with a risk of 9.5-10.5% predicted by QRISK3 had a risk of 2.9-9.2% in a random forest and 2.4-7.2% in a neural network. The differences in predicted risks between QRISK3 and a neural network ranged between –23.2% and 0.1% (95% range). Models that ignored censoring (that is, assumed censored patients to be event free) substantially underestimated risk of cardiovascular disease. Of the 223 815 patients with a cardiovascular disease risk above 7.5% with QRISK3, 57.8% would be reclassified below 7.5% when using another model.ConclusionsA variety of models predicted risks for the same patients very differently despite similar model performances. The logistic models and commonly used machine learning models should not be directly applied to the prediction of long term risks without considering censoring. Survival models that consider censoring and that are explainable, such as QRISK3, are preferable. The level of consistency within and between models should be routinely assessed before they are used for clinical decision making.

ObjectivesTo estimate fracture risk in patients receiving bariatric surgery versus matched controls.DesignPopulation based, retrospective cohort study. SettingUse of records from the United Kingdom General Practice Research Database, now known as the Clinical Practice Research Datalink (from January 1987 to December 2010). ParticipantsPatients with a body mass index of at least 30, with a record of bariatric surgery (n=2079), and matched controls without a record (n=10 442). Each bariatric surgery patient was matched to up to six controls by age, sex, practice, year, and body mass index. Patients were followed from the date of bariatric surgery for the occurrence of any fracture. We used time dependent Cox regression to calculate relative rates of fracture, adjusted for disease and previous drug treatment, and time-interaction terms to evaluate fracture timing patterns.Main outcome measureRelative rates of any, osteoporotic, and non-osteoporotic fractures.ResultsMean follow-up time was 2.2 years. Overall, there was no significantly increased risk of fracture in patients who underwent bariatric surgery, compared with controls (8.8 v8.2 per 1000 person years; adjusted relative risk 0.89, 95% confidence interval 0.60 to 1.33). Bariatric surgery also did not affect risk of osteoporotic and non-osteoporotic fractures. However, we saw a trend towards an increased fracture risk after three to five years following surgery, as well as in patients who had a greater decrease in body mass index after surgery, but this was not significant.ConclusionBariatric surgery does not have a significant effect on the risk of fracture. For the first few years after surgery, these results are reassuring for patients undergoing such operations, but do not exclude a more protracted adverse influence on skeletal health in the longer term.

Background Two recent case–control studies by Meier et al. and van Staa et al. used the UK General Practice Research Database (GPRD) to examine the association between the use of statins and the risk of fractures, with different results. The objective of the present study was to examine methodological explanations for the discrepant results. Methods We created two datasets, which mimicked the previous study designs: a ‘selected population’ (SP) case–control dataset, with fracture cases matched to controls nested within a selected cohort (Meier et al.), and an ‘entire population’ (EP) case–control dataset, with both cases and controls sampled from the total GPRD population (van Staa et al.). Cases and controls were matched by gender, age (year of birth or 5 year age bands), and general practice. Results The study included 131 855 fracture cases. The crude odds ratio (OR) for hip fracture in statin users was 0.37 (95% CI 0.27–0.52) in the SP and 0.54 (95% CI 0.39–0.74) in the EP dataset. This difference was reduced when matching by year of birth, rather than by 5 year age bands: crude ORs were 0.58 (95% CI 0.43–0.79) and 0.61 (95% CI 0.44–0.88), respectively. In the SP dataset, 37% of the cases could be matched by year of birth, while this was achieved for 99% in the ‘EP’ dataset. The exposure time-window, the selection of confounders, and exclusion of high-risk patients also influenced results. Conclusion Residual confounding by a matching variable and different definitions of the exposure time window explained differences in results. In case–control studies of drug use and fracture risk, broad matching criteria for age should be avoided and the selection of the time-window for exposure should be carefully considered.