Staging and clean room: Constructs designed to facilitate transparency and reduce bias in comparative analyses of real-world data.

PURPOSE: We describe constructs designed to protect the integrity of the results from comparative analyses using real-world data (RWD): staging and clean room. METHODS: Staging involves performing sequential preliminary analyses and evaluating the population size available and potential bias before conducting comparative analyses. A clean room involves restricted access to data and preliminary results, policies governing exploratory analyses and protocol deviations, and audit trail. These constructs are intended to allow decisions about protocol deviations, such as changes to design or model specification, to be made without knowledge of how they might affect subsequent analyses. We describe an example for implementing staging with a clean room. RESULTS: Stage 1 may involve selecting a data source, developing and registering a protocol, establishing a clean room, and applying inclusion/exclusion criteria. Stage 2 may involve attempting to achieve covariate balance, often through propensity score models. Stage 3 may involve evaluating the presence of residual confounding using negative control outcomes. After each stage, check points may be implemented when a team of statisticians, epidemiologists and clinicians masked to how their decisions may affect study outcomes, reviews the results. This review team may be tasked with making recommendations for protocol deviations to address study precision or bias. They may recommend proceeding to the next stage, conducting additional analyses to address bias, or terminating the study. Stage 4 may involve conducting the comparative analyses. CONCLUSIONS: The staging and clean room constructs are intended to protect the integrity and enhance confidence in the results of analyses of RWD.

Migraine and risk of hypertensive disorders of pregnancy: A population-based cohort study.

BACKGROUND: Migraine is associated with hypertensive disorders of pregnancy through common pathophysiological features. This study evaluates the association between migraine diagnosis and treatment, and risk of hypertensive disorders of pregnancy. METHODS: We conducted a prospective cohort study in the Clinical Practice Research Datalink GOLD, a large longitudinal database of patient records in the UK. We analyzed data from liveborn or stillborn singleton deliveries from 1993-2020 with at least 24 months of medical history and no history of cardiovascular disease (n = 1,049,839). We ascertained migraine through diagnosis or prescription codes before 20 weeks of gestation and hypertensive disorders of pregnancy through diagnosis codes between 20 weeks of pregnancy and delivery. We used log-binomial regression models to estimate the risk ratio and 95% confidence intervals, comparing risk of hypertensive disorders of pregnancy among individuals with migraine to those without migraine, adjusting for confounders. CONCLUSIONS: A history of migraine prior to pregnancy was associated with an increased risk of hypertensive disorders of pregnancy (RR = 1.17, 95% CI: 1.09-1.26). The greatest risk was among those with pre-pregnancy migraine that persisted into the first trimester (RR = 1.84, 95% CI: 1.35-2.50). Use of migraine medication was associated with a higher risk of hypertensive disorders of pregnancy compared to non-migraineurs (RR = 1.50, 95% CI: 1.15-1.97). Results from this study indicate that migraine is a potential risk factor for hypertensive disorders of pregnancy.

Quality of rheumatoid arthritis recording in United Kingdom Clinical Practice Research Datalink Aurum.

PURPOSE: While several studies have assessed quality and completeness of recording acute medical events in Clinical Practice Research Datalink (CPRD) Aurum, evaluation of additional chronic conditions is warranted. METHODS: We selected patients with a first diagnosis of rheumatoid arthritis (RA) coded in their CPRD Aurum record between 2005 and 2019. We assessed quality of RA diagnosis by evaluating additional information in the patient record that would corroborate the diagnosis. We report recording of diagnoses, prescriptions, labs, and referrals expected to be present based on NICE guidelines for RA management. RESULTS: There were 53 083 patients with a first recorded RA diagnosis during the study period: 43606 (82%) patients had RA drug treatments in their record, 7596 (14%) had supporting codes without drug treatment, and 1881 (4%) patients had only a RA diagnoses recorded in their medical record with no supporting codes or RA treatments. Patients with RA diagnosis only were more likely to be first diagnosed in the earliest time period of study. Labs for diagnosing and monitoring RA were most common among patients with RA treatment. Analgesic and glucocorticoid prescriptions were common in all study patients but were highest among patients with RA treatment. Among patients with RA diagnosis only, the overwhelming majority had only one RA diagnosis recorded (76%). CONCLUSIONS: Our findings suggest that codes expected for monitoring and treatment of RA are routinely recorded in CPRD Aurum. These results support previous assessments, which found data recorded in CPRD Aurum to be of good quality for use in research.

Changes in Medication Use During Pregnancy for Women with Chronic Conditions: An Analysis of Claims Data.

PURPOSE: Evaluation of drug safety during pregnancy is dependent on the number of exposed women during routine clinical practice with data available for analysis. We examined medication fills in pregnant and nonpregnant women within select disease cohorts: general population, migraine, diabetes, and hyperlipidemia to explore the potential use of claims data to assess medication use and safety during pregnancy. METHODS: This cohort study, using IBM MarketScan® Research Databases claims data, included women 10-54 years of age with pregnancy resulting in a liveborn infant between January 2010 and September 2015 and matched nonpregnant women. Medication use (antidepressants, antihypertensives, sedatives, glucose-lowering medications, antiepileptics, antipsychotics, lipid-lowering medications) was abstracted from pharmacy claims 180 days before last menstrual period through 180 days postdelivery. RESULTS: Among 753,760 women in the general pregnancy population (including 73,268 migraine, 50,155 hyperlipidemia, and 8361 diabetes; non-exclusive cohorts), antidepressants, antihypertensives, and sedatives were the most commonly used medications during pregnancy. Medications of interest were less commonly used in the pregnancy cohort than in the matched nonpregnant cohort within each time period (e.g., 3.7% vs 13.1% antidepressant use in 1st trimester). Most prescription fills were less common during pregnancy then pre-pregnancy. Post-pregnancy, prescription fills increased to or exceeded pre-pregnancy levels, except antihypertensive and glucose-lowering medications, which increased during pregnancy. CONCLUSIONS: Medication use among pregnant women was low and different from that among matched nonpregnant women. The underlying size of large commercial claims databases offer opportunities for efficient evaluation of potential safety concerns, particularly for rare drug exposures, compared to traditional pregnancy registries.

Pre-pregnancy migraine diagnosis, medication use, and spontaneous abortion: a prospective cohort study.

BACKGROUND: Migraine is common among females of reproductive age (estimated prevalence:17-24%) and may be associated with reproductive health through underlying central nervous system excitability, autoimmune conditions, and autonomic dysfunction. We evaluated the extent to which pre-pregnancy migraine diagnosis and medication use are associated with risk of spontaneous abortion (SAB). METHODS: We analyzed data from a preconception study of pregnancy planners (2013-2021). Eligible participants self-identified as female, were aged 21-45 years, resided in the USA or Canada, and conceived during follow-up (n = 7890). Participants completed baseline and bimonthly follow-up questionnaires for up to 12 months or until a reported pregnancy, whichever occurred first. Pregnant participants then completed questionnaires during early (~ 8-9 weeks) and late (~ 32 weeks) gestation. We defined migraineurs as participants who reported a migraine diagnosis or use of a medication to treat migraine. Preconception questionnaires elicited migraine medication use during the past 4 weeks, and SAB on follow-up and pregnancy questionnaires. We used Cox regression models with gestational weeks as the time scale to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for associations among preconception migraine, migraine medication use, and SAB, controlling for potential demographic, medical, and lifestyle confounders. RESULTS: Nineteen percent of study pregnancies ended in SAB. History of migraine before conception was not appreciably associated with SAB risk (HR = 1.03, 95% CI: 0.91-1.06). Use of any migraine medication was associated with a modest increase in SAB risk overall (HR = 1.14, 95% CI: 0.96-1.36). We observed the greatest increase in risk among those taking migraine medications daily (HR = 1.38, 95% CI: 0.81-2.35) and those taking prescription migraine prophylaxis (HR = 1.43, 95% CI: 0.72-2.84) or combination analgesic and caffeine medications (HR = 1.42, 95% CI: 0.99-2.04). CONCLUSIONS: Migraine medication use patterns suggesting greater underlying migraine severity were associated with increased risk of SAB. This research adds to the limited information available on the reproductive effects of migraine.

Non-steroidal anti-inflammatory drugs and risk of myocardial infarction adjusting for use of proton pump-inhibitors in patients with no major risk factors: a nested case-control study in the UK Clinical Practice Research Datalink.

AIMS: Studies have found an increased risk of myocardial infarction (MI) in association with some non-steroidal anti-inflammatory drugs (NSAIDs). We evaluated this association in patients without major cardiovascular risk factors and assessed potential reverse causality bias. METHODS AND RESULTS: In this nested case-control study of patients aged 40-79 in Clinical Practice Research Datalink GOLD who received at least one NSAID prescription between 2006 and 2019, we found 8639 MI cases and 34 556 matched controls. We calculated odds ratios (ORs) and 95% confidence intervals (CIs) for MI comparing NSAID users to non-exposed according to the number and timing of NSAID prescriptions and proton-pump inhibitor (PPI) use. Current diclofenac use was associated with a two-fold increased risk of MI regardless of duration of use (adjusted OR, 2.08; 95% CI, 1.82-2.38). ORs ranged from 3 to 5 among current and recent diclofenac users newly exposed to PPIs. There was no spike in risk in new current diclofenac users not exposed to PPIs, but ORs rose with increasing prescriptions. The risk of MI in ibuprofen users was concentrated in new PPI users. There was no material increased risk in naproxen users, nor in past users of most NSAIDs in the absence of PPIs. CONCLUSION: The risk of MI was elevated in current diclofenac users, particularly in new concomitant PPI users. ORs increased in new users of ibuprofen and PPIs but declined with extended use and were lower in non-PPI users. This suggests that some of the findings may be explained by reverse causality bias.

Presence of Codes for Indication for Use in Clinical Practice Research Datalink Aurum: An Assessment of Benign Prostatic Hyperplasia Treatments.

Background: Assessments of strengths and limitations of new data sources are critical for making decisions about suitability for specific research questions. For some studies, it is necessary to capture a drug's indication for use. Objective: To assess the presence of indications for prescription use in Clinical Practice Research Datalink (CPRD) Aurum (January 1988-June 2021) by describing the proportion of men in CPRD Aurum who had a recorded indication for use of prescriptions for 5-alpha reductase inhibitors (5-ARI), alpha blockers (AB), or tadalafil, which have multiple indications. Methods: From a random sample of 154 practices of CPRD Aurum data, we selected 85,597 male patients with a prescription for a 5-ARI, an AB, or tadalafil. Among these patients, we described presence of codes indicating whether the patient had benign prostatic hyperplasia, hypertension, erectile dysfunction, or alopecia using three indication definitions: narrow (specific diagnoses recorded within one year before and up to 90 days after the prescription), broad (specific diagnoses or supporting clinical codes in the time period described above), and widest (diagnoses or supporting codes recorded at any time before the prescription and up to 90 days after the prescription). Results: Using the narrow indication definition limited to diagnoses only, 39,861 (46.6%) patients' records contained an indication for use. The broad definitions, which additionally included supporting codes, captured indications for 62,912 (73.5%) patients and the widest definition, which additionally included supporting codes and all available data before the first prescription date, captured indications for 71,478 (83.5%) patients. Indications were present more often for prescriptions in 2005 and later (85.9%). Conclusion: The findings of this assessment suggest that CPRD Aurum can be used for studies that require information on treatment indications for BPH and potentially for treatments of other chronic diseases managed in the primary care setting.

Characterization of the deterioration of diabetes control in patients with a subsequent diagnosis of pancreatic cancer: A descriptive study.

BACKGROUND: Deterioration of diabetes control can be the first harbinger of pancreatic cancer. However, little is known about how to distinguish patients with pancreatic cancer-related diabetes deterioration from those with type 2 diabetes progression. We aimed to characterize the glycated hemoglobin (HbA1c) and body weight profile of pancreatic cancer patients with deteriorating diabetes before the cancer diagnosis. METHODS: Using data from the UK-based Clinical Practice Research Datalink (CPRD) GOLD, we established a study population including pancreatic cancer patients with diabetes deterioration in the >0.5-3 years before the cancer diagnosis and non-cancer patients with deterioration of type 2 diabetes (comparison group). Patients were considered to have diabetes deterioration if their glucose-lowering treatment was intensified. We characterized the longitudinal trajectories of HbA1c and body weight in pancreatic cancer patients compared with non-cancer patients before and after treatment intensification. RESULTS: The mean absolute increase in HbA1c from the pre-deterioration period, i.e. the time >1-2 years before treatment intensification, to the time of treatment intensification, was 1.5% ± 1.6% in pancreatic cancer patients vs. 0.9% ± 1.4% in non-cancer patients. After treatment intensification, mean HbA1c remained elevated in pancreatic cancer patients, while it returned to the pre-deterioration level in non-cancer patients. Body weight decreased by 1.9% ± 6.4% in cancer patients and increased by 0.3% ± 5.2% in non-cancer patients between the pre-deterioration period and treatment intensification, on average. CONCLUSIONS: Pancreatic cancer-related diabetes deterioration may frequently be characterized by pronounced increases in HbA1c, persistent elevation of HbA1c after treatment intensification, and concomitant weight loss.

ß-Blockers and the Risk of Depression: A Matched Case-Control Study.

INTRODUCTION: Depression is a commonly cited adverse effect of ß-blockers but the evidence for a causal relationship is limited. OBJECTIVE: We aimed to explore whether ß-blockers are associated with an increased risk of new-onset depression. METHODS: We conducted a case-control study using the UK population-based Clinical Practice Research Datalink (CPRD) GOLD. We identified patients aged 18-80 years with an incident depression diagnosis between 2000 and 2016, and matched controls, and estimated the risk (odds ratio [OR]) of depression in association with use of ß-blockers. We also conducted analyses of exposure, categorised by number and timing of prescriptions and by indication for ß-blocker use. RESULTS: The study encompassed 118,705 patients with incident depression and the same number of matched controls. The odds of developing depression were increased for current short-term use of any ß-blocker (adjusted OR [aOR] 1.91, 95% confidence interval [CI] 1.72-2.12), whereas current long-term use was not associated with the risk of depression compared with never use. The elevated risk of depression among short-term users was mostly confined to propranolol users with a neuropsychiatric disorder (aOR 6.33, 95% CI 5.16-7.76), while propranolol users with a cardiovascular indication were only at marginally increased risk of depression (aOR 1.44, 95% CI 1.14-1.82). CONCLUSIONS: This study suggests that the association between use of ß-blockers and depression may not be causal but rather a result of protopathic bias. Propranolol is often prescribed to treat neuropsychiatric symptoms, suggesting that the onset of depression may be related to the underlying indication rather than to an effect of a ß-blocker therapy.

Association between glycemic control and risk of venous thromboembolism in diabetic patients: a nested case-control study.

BACKGROUND: Previous studies suggested an elevated risk of venous thromboembolism (VTE) among patients with type 2 diabetes mellitus (T2DM), with a possible sex difference. The impact of glycemic control on the risk of VTE is unclear. Our objective was to analyze the association between glycemic control and the risk of unprovoked (idiopathic) VTE in men and women with T2DM. METHODS: We conducted a nested case-control analysis (1:4 matching) within a cohort of patients with incident T2DM between 1995 and 2019 using data from the CPRD GOLD. We excluded patients with known risk factors for VTE prior to onset of DM. Cases were T2DM patients with an unprovoked treated VTE. The exposure of interest was glycemic control measured as HbA1c levels. We conducted conditional logistic regression analyses adjusted for several confounders. RESULTS: We identified 2'653 VTE cases and 10'612 controls (53.1% females). We found no association between the HbA1c level and the risk of VTE in our analyses. However, when the most recent HbA1c value was recorded within 90 days before the index date, women with HbA1c levels > 7.0% had a 36-55% increased relative risk of VTE when compared to women with HbA1c > 6.5-7.0%. CONCLUSIONS: Our study raises the possibility that female T2DM patients with HbA1c levels > 7% may have a slightly higher risk for unprovoked VTE compared to women with HbA1c levels > 6.5-7.0%. This increase may not be causal and may reflect differences in life style or other characteristics. We observed no effect of glycemic control on the risk of VTE in men.

Glucocorticoids and Risk of Venous Thromboembolism in Asthma Patients Aged 20-59 Years in the United Kingdom's CPRD 1995-2015.

BACKGROUND: Glucocorticoids, the class of steroids used in management of asthma, have been observed to be associated with adverse events such as increased coagulation and inhibition of fibrinolysis. This study evaluated the risk of VTE in relation to the use of glucocorticoids in patients with asthma. METHODS: We conducted a nested case-control study among patients aged 20-59 years with asthma who received at least one glucocorticoid prescription during 1995-2015 in the UK-based Clinical Practice Research Datalink GOLD. We used descriptive analyses and conditional logistic regression to evaluate the risk of VTE associated with glucocorticoid use. RESULTS: The adjusted ORs (aORs) (95% CI) for VTE in patients exposed to glucocorticoids were 1.9 (1.6-2.3), 1.4 (1.1-1.8), and 1.2 (0.9-1.5) for current, recent, and past glucocorticoid users, respectively, compared to the unexposed. The aORs (95% CI) for VTE in patients exposed to systemic glucocorticoid and inhaled glucocorticoids, compared to the unexposed, were 3.5 (2.7-4.5) and 1.5 (1.3-1.8), respectively. CONCLUSION: Current and systemic glucocorticoid use was associated with a dose-response increased risk of incident idiopathic VTE.

Quality and Completeness of Myocardial Infarction Recording in Clinical Practice Research Datalink Aurum.

BACKGROUND: Validation studies of the Clinical Practice Research Datalink (CPRD) Aurum database in the UK are critical for making decisions about its suitability and validity for research purposes. OBJECTIVE: To examine data source agreement of myocardial infarction (MI) diagnoses recorded in CPRD Aurum compared with linked Hospital Episode Statistics (HES) data. This comparison provides information on CPRD Aurum data correctness (accuracy, validity) and completeness (presence, missingness). METHODS: Patients with MI diagnoses recorded in either data source were selected from a random sample of 50,000 patients in CPRD Aurum with HES linkage (1997-2017). Correctness was defined as the proportion of MI cases in CPRD Aurum with a concordant MI diagnosis recorded in HES or with strong supporting evidence in either data source. Completeness was defined as the proportion of patients with primary HES-coded MIs with strong supporting evidence that were also present in CPRD Aurum. RESULTS: There were 1260 patients with MI recorded in the CPRD Aurum sample. The overall correctness of the recorded MI diagnoses was 94%: 986 patients (78%) had concordant diagnoses in HES within 90 days; 123 (10%) were concordant with HES, but with an inconclusive date and another 71 (6%) had strong supporting evidence for being a true MI case. There were 1125 patients with MI recorded in HES primary diagnosis fields with strong supporting evidence in either data source. Of these, 880 (78%) were present in CPRD Aurum, with completeness somewhat higher in more recent years. CONCLUSION: MI diagnoses recorded in CPRD Aurum were highly likely to be correct, supporting its use in clinical research studies. Completeness was lower, indicating the need for data linkage for some studies.

Epidemiology of skin event rates among users of pumps for the subcutaneous administration of drugs for chronic conditions.

OBJECTIVE: To provide the epidemiology of skin events occurring during long-term administration of medications delivered by continuous subcutaneous infusion pump (CSIP) systems as background rates for the development of novel CSIP treatments to use in community-based settings. METHODS: Using a United Kingdom general practice database, we conducted a study to assess the rates of skin events among new users of apomorphine and insulin delivered by CSIP in patients with Parkinson's disease or diabetes, respectively. Skin events included skin infections, skin nodules/localized swelling, dermatitis/eczema, urticaria/erythema, and rash/other non-specific skin eruptions. RESULTS: Five hundred and fifty-seven adults (age 30+) were included in this descriptive cohort. The median duration of CSIP use was 17 months among 255 apomorphine users and 41 months among 302 insulin users. By 60 months, ∼40% of both cohorts experienced skin events. Repeated skin events occurred in 11% of the apomorphine cohort and 14% of the insulin cohort at any time during follow-up. The overall skin event rate in the apomorphine cohort was 17 per 1000 person-months (PM) and 13 per 1000 PM in the insulin cohort. The most common skin events in both cohorts were infection and rash/unspecified skin eruptions. The highest rates of skin events occurred soon after apomorphine CSIP initiation (36 per 1000 PM in weeks 1-2 and 50 per 1000 PM in weeks 3-4), with lower rates after 4 weeks. Insulin CSIP users' skin event rates were consistent over the treatment duration. CONCLUSIONS: Clinically important skin events are common during long-term administration of medications by CSIP.

Validity of bariatric surgery codes in the UK Clinical Practice Research Datalink (CPRD) GOLD compared with Hospital Episodes Statistics.

OBJECTIVES: To assess completeness and validity of bariatric surgery codes in the UK Clinical Practice Research Datalink (CPRD) GOLD compared with Hospital Episodes Statistics (HES). METHODS: We conducted a validation study among patients in the UK-based CPRD GOLD with linkage to HES (1998 to 2017). Since the same surgery codes are used for bariatric and other gastrointestinal surgery we assessed code distribution patterns used in patients with bariatric versus other gastrointestinal surgery by presence of other conditions such as obesity and gastrointestinal cancer. We developed algorithms to identify bariatric surgery and calculated validity measures (ie, positive/negative predictive value [PPV/NPV], sensitivity, and specificity) of each in CPRD GOLD compared with HES (gold standard). RESULTS: Among 7 357 007 available patients we identified 10 190 patients who had a total of 14 046 potential bariatric surgery codes in CPRD GOLD and/or HES. Surgery code patterns differed between bariatric surgery and assumed other gastrointestinal surgery. The sensitivity of CPRD GOLD bariatric surgery coding improved from an overall of 56% to 69-71% when applying stricter algorithms (ie, in obese patients or obese, gastrointestinal disease/complication free patients) but PPVs remained at 53%-55%. NPVs and specificities of CPRD GOLD bariatric surgery coding achieved ≥99.8% for all algorithms. CONCLUSION: Our results suggest that using CPRD GOLD and HES data and a wide selection of surgery codes will result in the most complete and accurate capture of bariatric surgery events. Validity measures of CPRD GOLD bariatric surgery codes were identical in obese patients and more restrictive populations.

The Risk of Muscular Events Among New Users of Hydrophilic and Lipophilic Statins: an Observational Cohort Study.

BACKGROUND: Statins are effective lipid-lowering drugs for the prevention of cardiovascular disease, but muscular adverse events can limit their use. Hydrophilic statins (pravastatin, rosuvastatin) may cause less muscular events than lipophilic statins (e.g. simvastatin, atorvastatin) due to lower passive diffusion into muscle cells. OBJECTIVE: To compare the risk of muscular events between statins at comparable lipid-lowering doses and to evaluate if hydrophilic statins are associated with a lower muscular risk than lipophilic statins. DESIGN/SETTING: Propensity score-matched cohort study using data from the United Kingdom-based Clinical Practice Research Datalink (CPRD) GOLD. PATIENTS: New statin users. Cohort 1: pravastatin 20-40 mg (hydrophilic) vs simvastatin 10-20 mg (lipophilic), cohort 2: rosuvastatin 5-40 mg (hydrophilic) vs atorvastatin 10-80 mg (lipophilic), and cohort 3: simvastatin 40-80 mg vs atorvastatin 10-20 mg. MAIN MEASURES: The outcome was a first record of a muscular event (myopathy, myalgia, myositis, rhabdomyolysis) during a maximum follow-up of 1 year. KEY RESULTS: The propensity score-matched cohorts consisted of 1) 9,703, 2) 7,032, and 3) 37,743 pairs of statin users. Comparing the risk of muscular events between low-intensity pravastatin vs low-intensity simvastatin yielded a HR of 0.86 (95% CI 0.64-1.16). In the comparison of moderate- to high-intensity rosuvastatin vs equivalent doses of atorvastatin, we observed a HR of 1.17 (95% CI 0.88-1.56). Moderate- to high-intensity simvastatin was associated with a HR of 1.33 (95% CI 1.16-1.53), when compared with atorvastatin at equivalent doses. LIMITATIONS: We could not conduct other pairwise comparisons of statins due to small sample size. In the absence of a uniform definition on the comparability of statin doses, the applied dose ratios may not fully match with all literature sources. CONCLUSIONS: Our results do not suggest a systematically lower risk of muscular events for hydrophilic statins when compared to lipophilic statins at comparable lipid-lowering doses.

Antiseizure drugs and risk of developing smoking-related chronic obstructive pulmonary disease or lung cancer: A population-based case-control study.

AIMS: To determine whether enzyme-inducing antiseizure drugs (ASDs) affect the risk of developing chronic obstructive pulmonary disease (COPD) or lung cancer in smokers. METHODS: Cases of COPD and lung cancer and matched controls without these conditions were identified from a population of smokers with ≥1 prescription for any type of ASD in the Clinical Practice Research Datalink UK database of patients managed in primary care (1995-2016). A matched case-control study was performed utilising multivariate logistic regression analyses of exposure to enzyme-inducing ASDs compared to non-enzyme-inducing ASDs. The duration of ASD exposure and level of tobacco exposure were also assessed. RESULTS: We identified 5952 incident COPD and 1373 incident lung cancer cases, and 59 328 and 13 681 matched controls, respectively. Compared with never use, ever use of enzyme-inducing ASDs was associated with slightly decreased risk estimates of COPD (adjusted odds ratio: 0.85, 95% confidence interval: 0.81-0.89) and lung cancer (adjusted odds ratio: 0.82, 95% confidence interval: 0.73-0.92). These risk estimates were attenuated in heavy smokers. CONCLUSION: We found slightly decreased risk estimates of COPD and lung cancer among smokers taking enzyme-inducing ASDs and hypothesise that this may be related to induction of detoxification of tobacco-specific lung toxins.

Antidiabetic Treatment, Level of Glycemic Control, and Risk of Fracture in Type 2 Diabetes: a Nested, Case-Control Study.

CONTEXT: Patients with type 2 diabetes mellitus (T2DM) have an increased risk of low-trauma fractures. However, the effect of antidiabetic medication in relation to glycemic control on the risk of fracture is poorly understood. OBJECTIVE: This work aimed to evaluate the association between the level of glycemic control, use of antidiabetic medication, and risk of low-trauma fractures in patients with newly diagnosed T2DM. METHODS: We conducted a nested case-control analysis among individuals registered in the Clinical Practice Research Datalink. The base population consisted of patients with newly diagnosed T2DM from 1995 to 2017. Cases were patients with a low-trauma fracture after T2DM diagnosis. We matched 4 controls to each case. Exposures of interest were glycemic control (last glycated hemoglobin [HbA1c] level before fracture) and type of diabetes treatment. We conducted conditional logistic regression analyses adjusted for several confounders. RESULTS: We identified 8809 cases and 35 219 controls. Patients with current metformin use and HbA1c levels of less than 7.0% and between 7.0-8.0% had a reduced risk of fractures (adjusted odds ratio 0.89; 95% CI, 0.83-0.96 and 0.81; 95% CI, 0.73-0.90, respectively) compared with untreated patients. However, in patients receiving metformin plus 1 or 2 other antidiabetic drugs, or insulin (alone or in addition to other antidiabetic medication), the level of glycemic control was not associated with the risk of fracture compared with untreated patients. CONCLUSIONS: While patients with good or medium glycemic control receiving current metformin monotherapy had a lower risk of fracture compared with untreated patients, glycemic control in patients receiving treatment other than metformin was not associated with risk of fracture.

CPRD Aurum database: Assessment of data quality and completeness of three important comorbidities.

PURPOSE: The Clinical Practice Research Datalink (CPRD) now provides a new medical record database, CPRD Aurum. This is the second of several studies being undertaken to assess the quality of CPRD Aurum data for research. METHODS: We included patients aged 20+, with at least one lab test result of any type from a random sample of 50 000 patients in CPRD Aurum. We assessed whether diagnosis codes for type 2 diabetes, hyperlipidemia, and iron deficiency or unspecified anemia were accompanied by supporting codes including lab results and treatments (correctness) and whether lab results, treatments, or other codes indicate a missing diagnosis record (completeness). RESULTS: Among 37 502 patients in CPRD Aurum, correctness of type 2 diabetes, hyperlipidemia, and anemia diagnoses was high (99%, 93%, and 97%, respectively). Completeness was only high for type 2 diabetes (94%-98%); completeness for hypercholesterolemia and anemia diagnoses was modest even when the presence of treatments and lab results indicated the conditions were likely present (51%-59% and 58%-70%, respectively). CONCLUSIONS: Our findings indicate that for studies of type 2 diabetes, hyperlipidemia, and iron deficiency or unspecified anemia, the diagnosis code is likely to be correct where present. However, a significant proportion of cases of hyperlipidemia or anemia will be missed if only diagnosis codes are used to select patients with these conditions. Researchers should consider using treatments, supporting codes, and, when available, lab data to supplement diagnosis codes and enhance case capture when including these conditions in studies using CPRD Aurum.

Quality and completeness of diagnoses recorded in the new CPRD Aurum Database: evaluation of pulmonary embolism.

PURPOSE: The Clinical Practice Research Datalink (CPRD) now provides a new medical record database, CPRD Aurum. This is the first of several studies being undertaken to assess the quality and completeness of CPRD Aurum data for research endeavors. METHODS: We identified patients with a pulmonary embolism (PE) diagnosis from a random sample of 50 000 patients in CPRD Aurum and compared the diagnoses using data from Hospital Episode Statistics (HES). We calculated the proportion of PE cases recorded in CPRD Aurum who also had a PE diagnosis recorded in HES. We also evaluated completeness by identifying all PE diagnoses in HES and calculating the proportion also present in CPRD Aurum. RESULTS: The study included 781 PE patients: 580 had a PE in CPRD Aurum, 632 had a PE in HES, and 431 had a PE in both. The proportion of patients with anticoagulated PE in CPRD Aurum confirmed by HES was 76.8%. The completeness of primary hospitalized PE HES events compared to CPRD Aurum was 79.1%. In most instances, there was a plausible explanation for the presence of a PE in only one of the two data sources. CONCLUSIONS: The results of this study are reassuring and suggest that the correctness (eg, quality, accuracy) and completeness of diagnosis information in CPRD Aurum are promising with respect to serious acute conditions that require medical attention. Evaluation of other data elements will provide additional insight into this new data resource and its utility for medical research.