Scientific
publications

PURPOSE: With increased concomitant chronic diseases in type 2 diabetes mellitus (T2DM), the use of multiple drugs increases as well as the risk of drug-drug interactions (DDI) and adverse drug reactions (ADR). Nevertheless, how medication patterns vary in T2DM patients across different sex and age groups is unclear. This study aims to identify and quantify common drug combinations in first-time metformin users with polypharmacy (≥5 co-medications). METHODS: New users of metformin were identified from the IQVIA Medical Research Data incorporating data from THIN, A Cegedim Database (2016-2019). A descriptive cohort study explored prescription patterns in patients with polypharmacy. The Apriori algorithm, used to find frequent item-sets in databases, was first-time applied to identify and quantify drug combinations of up to seven drugs to investigate potential harmful polypharmacy patterns. RESULTS: The cohort included 34 169 new-users of metformin, of which 20 854 (61.0%) received polypharmacy. Atorvastatin was the most frequently co-prescribed drug with metformin overall (38.7%), in women (34.3%) and men (42.6%). In the stratified analysis, a higher proportion of women received polypharmacy (65.6%) compared to men (57.4%). Moreover, the proportion of patients receiving polypharmacy increased with age (18-39 years = 30.4%, 40-59 years = 50.5%, 60-74 years = 70.9%, and ≥75 years = 84.3%). CONCLUSION: This study is the first to identify and quantify commonly prescribed combinations of drugs compounds in patients with polypharmacy using the Apriori algorithm. The high polypharmacy prevalence at all strata indicates the need to optimize polypharmacy to minimize DDI and ADR.

BACKGROUND: Many dementia prediction models have been developed, but only few have been externally validated, which hinders clinical uptake and may pose a risk if models are applied to actual patients regardless. Externally validating an existing prediction model is a difficult task, where we mostly rely on the completeness of model reporting in a published article. In this study, we aim to externally validate existing dementia prediction models. To that end, we define model reporting criteria, review published studies, and externally validate three well reported models using routinely collected health data from administrative claims and electronic health records. METHODS: We identified dementia prediction models that were developed between 2011 and 2020 and assessed if they could be externally validated given a set of model criteria. In addition, we externally validated three of these models (Walters' Dementia Risk Score, Mehta's RxDx-Dementia Risk Index, and Nori's ADRD dementia prediction model) on a network of six observational health databases from the United States, United Kingdom, Germany and the Netherlands, including the original development databases of the models. RESULTS: We reviewed 59 dementia prediction models. All models reported the prediction method, development database, and target and outcome definitions. Less frequently reported by these 59 prediction models were predictor definitions (52 models) including the time window in which a predictor is assessed (21 models), predictor coefficients (20 models), and the time-at-risk (42 models). The validation of the model by Walters (development c-statistic: 0.84) showed moderate transportability (0.67-0.76 c-statistic). The Mehta model (development c-statistic: 0.81) transported well to some of the external databases (0.69-0.79 c-statistic). The Nori model (development AUROC: 0.69) transported well (0.62-0.68 AUROC) but performed modestly overall. Recalibration showed improvements for the Walters and Nori models, while recalibration could not be assessed for the Mehta model due to unreported baseline hazard. CONCLUSION: We observed that reporting is mostly insufficient to fully externally validate published dementia prediction models, and therefore, it is uncertain how well these models would work in other clinical settings. We emphasize the importance of following established guidelines for reporting clinical prediction models. We recommend that reporting should be more explicit and have external validation in mind if the model is meant to be applied in different settings.

IMPORTANCE: Suboptimal adherence to endocrine therapy (ET) among patients with hormone-receptor-positive breast cancer significantly affects survival outcomes and is associated with higher hospitalization rates and health care costs. Weak adherence to long-term treatments has multiple determinants, including disease characteristics, treatment adverse effects, and patients' attributes, such as age and comorbidities. OBJECTIVE: To examine whether potential drug-drug interactions (PDDI) with tamoxifen or aromatase inhibitor were associated with adherence to ET in patients with early and advanced breast cancer. DESIGN, SETTING, AND PARTICIPANTS: This cohort study used anonymized health record data of women with breast cancer who received ET in a private observational primary care database. Patients eligible for analysis included women aged 18 years or older who had a reported diagnosis of breast cancer and received ET with tamoxifen or aromatase inhibitor between 1994 and 2021. Data were analyzed 2021. EXPOSURES: Adherence to ET during a given year was defined by a medication possession ratio of 80% or greater over 1-year prescription periods. PDDI were categorized into absent, minor (a combination to take into account), moderate (combination requiring precautions for use), major (combination not recommended), and contraindicated according to guidelines in the Claude Bernard Drug Database. MAIN OUTCOMES AND MEASURES: We used regression models to estimate odds ratios (ORs) and 95% CIs for the associations between adherence and age, baseline comorbidities, PDDI, and adherence to ET during the previous year. RESULTS: A total of 10 863 patients who were prescribed ET for breast cancer were eligible for the analysis (age 70 years or older, 3509 patients [32.3%]). In the tamoxifen cohort (3564 patients), PDDI were reported in 497 of 3670 patients (13.5%) at baseline (moderate, 254 patients [51.1%]; major, 227 patients [45.7%]), 2047 of 4831 patients (42.4%) at year 1, 1127 of 2751 patients (41.0%) at year 2, 761 of 1861 patients (40.9%) at year 3, 376 of 1058 patients (35.5%) at year 4, and 201 of 593 patients (33.9%) at year 5. In the aromatase inhibitor cohort (7299 patients), PDDI were reported in 592 of 7437 patients (8.0%) at baseline (moderate in 588 of 592 patients [99.3%]), which reached 2875 of 9031 patients (31.8%) at year 1 and ranged between 31.4% (1802 of 5730 patients in year 2) and 32.8% (791 of 2411 in year 4) throughout the study period. No association between adherence and PDDI was found in the tamoxifen (OR, 0.99; 95% CI, 0.91-1.08) or aromatase inhibitor (OR, 1.05; 95% CI, 0.95-1.15) cohort. CONCLUSIONS AND RELEVANCE: In this cohort of patients with hormone-receptor-positive breast cancer, PDDI with tamoxifen and aromatase inhibitors were not associated with adherence to ET.

PURPOSE: To compare the risk of acute kidney injury (AKI) among users of rivaroxaban vs warfarin. PATIENTS AND METHODS: We identified two cohorts of patients with non-valvular atrial fibrillation (NVAF) who initiated rivaroxaban (15/20 mg/day, N = 6436) or warfarin (N = 7129) excluding those without estimated glomerular filtration rate values recorded in the year before oral anticoagulant (OAC) initiation and those with a history of end-stage renal disease or AKI. We used two methods to define AKI during follow-up (mean 2.5 years): coded entries (method A) and the Aberdeen AKI phenotyping algorithm (method B) using recorded renal function laboratory values during the study period to identify a sudden renal deterioration event. Cox regression was used to calculate hazard ratios (HRs) for AKI with rivaroxaban vs warfarin use, adjusted for confounders. RESULTS: The number of identified incident AKI cases was 249 (method A) and 723 (method B). Of the latter, 104 (14.4%) were also identified by method A. After adjusting for age, sex, baseline renal function and comorbidity, HRs (95% CIs) for AKI were 1.19 (0.92-1.54; =0.18) using method A and 0.80 (0.68-0.93; <0.01) using method B. Estimates stratified by baseline level of chronic kidney disease were largely consistent with the main estimates. CONCLUSION: Our results support a beneficial effect of rivaroxaban over warfarin in terms of AKI occurrence in patients with NVAF. More research into how best to define AKI using primary care records would be valuable for future studies.

BACKGROUND: Low-dose aspirin and metformin have been individually associated with a reduced risk of cancer. Whether their concurrent use in adults with type 2 diabetes mellitus (T2DM) is associated with a reduced risk of colorectal cancer (CRC) is unclear. OBJECTIVE: Among individuals with T2DM taking metformin, we sought to evaluate the association between low-dose aspirin versus no aspirin and the risk of CRC. METHODS: A multiple-database new-user cohort study of patients with T2DM taking metformin was conducted between 2007 and 2010 (Clinical Data Analysis and Reporting System [CDARS], Hong Kong) and 2007-2016 (The Health Improvement Network [THIN], UK). The primary outcome was incident CRC. Patients were followed from index date of prescription until the earliest occurrence of an outcome of interest, an incident diagnosis of any cancer, death, or until 31 December 2019. Cox proportional hazards regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CI). Estimates were pooled using an inverse variance random effects model, and heterogeneity was assessed using I . RESULTS: After one-to-one propensity-score matching, 57,534 patients were included (CDARS = 16,276; THIN = 41,258). The median (IQR) follow-up was 9.3 (6.5-10.7) years in CDARS and 3.2 (1.1-5.8) years in THIN. The concurrent use of low-dose aspirin and metformin was not associated with a lower risk of CRC compared to metformin only (HR = 0.89, 95% CI 0.75-1.05, I  = 0%). CONCLUSION: Low-dose aspirin was not associated with a lower risk of CRC in patients with T2DM taking metformin. Our study does not support the routine use of low-dose aspirin in this population.

BACKGROUND AND OBJECTIVES: Physician prescribing habits for opiates and headache therapies have not been previously evaluated in a large, matched cohort study in idiopathic intracranial hypertension (IIH). Our objective was to evaluate opiate and headache medication prescribing habits in women with IIH compared with matched women with migraine and population controls. We also investigated the occurrence of new onset headache in IIH compared with population controls. METHODS: We performed a population-based matched, retrospective cohort study to explore headache outcomes. Cross-sectional analyses were used to describe medication prescribing patterns. We used data from IQVIA Medical Research Data, an anonymized, nationally representative primary care electronic medical record database in the United Kingdom, from January 1, 1995, to September 25, 2019. Women aged 16 years and older were eligible for inclusion. Women with IIH (exposure) were matched by age and body mass index with up to 10 control women without IIH but with migraine (migraine controls), and without IIH or migraine (population controls). RESULTS: A total of 3,411 women with IIH, 13,966 migraine controls, and 33,495 population controls were included. The adjusted hazard ratio for new onset headache in IIH compared with population controls was 3.09 (95% CI 2.78-3.43). In the first year after diagnosis, 58% of women with IIH were prescribed acetazolamide and 20% topiramate. In total, 20% of women with IIH were prescribed opiates within the first year of their diagnosis, reducing to 17% after 6 years, compared with 8% and 11% among those with migraine, respectively. Twice as many women with IIH were prescribed opiates compared with migraine controls, and 3 times as many women with IIH were prescribed opiates compared with population controls. Women with IIH were also prescribed more headache preventative medications compared with migraine controls. DISCUSSION: Women with IIH were more likely to be prescribed opiate and simple analgesics compared with both migraine and population controls. Women with IIH trialed more preventative medications over their disease course suggesting that headaches in IIH may be more refractory to treatment.

Type 2 diabetes mellitus (T2DM) is associated with the development of chronic comorbidities, which can lead to high drug utilization and adverse events. We aimed to identify common comorbidity clusters and explore the progression over time in newly treated T2DM patients. The IQVIA Medical Research Data incorporating data from THIN, a Cegedim database of anonymized electronic health records, was used to identify all patients with a first-ever prescription for a non-insulin antidiabetic drug (NIAD) between January 2006 and December 2019. We selected 58 chronic comorbidities of interest and used Bayesian nonparametric models to identify disease clusters and model their progression over time. Among the 175,383 eligible T2DM patients, we identified the 20 most frequent comorbidity clusters, which were comprised of 14 latent features (LFs). Each LF was associated with a primary disease (e.g., 98% of patients in cluster 2, characterized by LF2, had congestive heart failure [CHF]). The presence of certain LFs increased the probability of having another LF active. For example, LF2 (CHF) frequently appeared with LFs related to chronic kidney disease (CKD). Over time, the clusters associated with cardiovascular diseases, such as CHF, progressed rapidly. Moreover, the onset of certain diseases led to further complications. Our models identified established T2DM complications and previously unknown connections, thus, highlighting the potential for Bayesian nonparametric models to characterize complex comorbidity patterns.

AIMS: People with pre-diabetes are at high risk of progressing to type 2 diabetes. This progression is not well characterised by ethnicity, deprivation and age, which we describe in a large cohort of individuals with pre-diabetes. METHODS: A retrospective cohort study with The Health Improvement Network (THIN) database was conducted. Patients aged 18 years and over and diagnosed with pre-diabetes [HbA1c 42 mmol/mol (6.0%) to 48 mmol/mol (6.5%) were included]. Cox proportional hazards regression was used to calculate adjusted hazard rate ratios (aHR) for the risk of progression from pre-diabetes to type 2 diabetes for each of the exposure categories [ethnicity, deprivation (Townsend), age and body mass index (BMI)] separately. RESULTS: Of the baseline population with pre-diabetes (n = 397,853), South Asian (aHR 1.31; 95% CI 1.26-1.37) or Mixed-Race individuals (aHR 1.22; 95% CI 1.11-1.33) had an increased risk of progression to type 2 diabetes compared with those of white European ethnicity. Likewise, deprivation (aHR 1.17; 95% CI 1.14-1.20; most vs. least deprived) was associated with an increased risk of progression. Both younger (aHR 0.63; 95% CI 0.58-0.69; 18 to <30 years) and older individuals (aHR 0.85; 95% CI 0.84-0.87; ≥65 years) had a slower risk of progression from pre-diabetes to type 2 diabetes, than middle-aged (40 to <65 years) individuals. CONCLUSIONS: South Asian or Mixed-Race individuals and people with social deprivation had an increased risk of progression from pre-diabetes to type 2 diabetes. Clinicians need to recognise the differing risk across their patient populations to implement appropriate prevention strategies.

OBJECTIVE: Gout patients often have multiple comorbidities, making them susceptible to SARS-CoV-2 infection and poor outcomes. This study was undertaken to examine the association between gout and the risk of SARS-CoV-2 infection and severe outcomes, especially in patients who have received a SARS-CoV-2 vaccine. METHODS: We conducted 2 cohort studies using The Health Improvement Network in the UK. Individuals with gout and those without gout from the general population were followed up from December 8, 2020 to October 31, 2021. We estimated the rate difference (RD) and hazard ratio (HR) of SARS-CoV-2 infection and severe outcomes (i.e., hospitalization and death within 30 days after SARS-CoV-2 infection) for individuals with gout versus those without gout using a Cox proportional hazards model according to SARS-CoV-2 vaccination status. We adjusted for potential confounders by using overlap weighting of exposure scores. RESULTS: Among the vaccinated cohort, 1,955 cases of breakthrough COVID-19 infection occurred in 54,576 individuals with gout (4.68 cases per 1,000 person-months), and 52,468 cases occurred in 1,336,377 individuals without gout (3.76 cases per 1,000 person-months). The partially adjusted RD of breakthrough infection was 0.91 cases per 1,000 person-months (95% confidence interval [95% CI] 0.62-1.20 cases per 1,000 person-months), and the partially adjusted HR was 1.24 (95% CI 1.19-1.30). Gout was also associated with an increased risk of hospitalization (adjusted HR 1.30 [95% CI 1.10-1.53]) and death (adjusted HR 1.36 [95% CI 0.87-2.13]). Women with gout had an increased risk of hospitalization (adjusted HR 1.55 [95% CI 1.15-2.10]) and death (adjusted HR 2.46 [95% CI 1.12-5.41]). Similar associations with gout were observed in the unvaccinated cohort. CONCLUSION: These general population data suggest that individuals with gout, especially women, have higher risks of SARS-CoV-2 infection and severe outcomes, even when vaccinated.

INTRODUCTION: Although essential tremor (ET) is considered a common adult movement disorder, evidence on its incidence is still scant. This study aims at estimating ET incidence in two European countries, namely, the UK and France. METHODS: Incident cases of ET were identified within the Health Improvement Network (THIN®) database between 1st January 2014 and 31 December 2019. Yearly crude and standardized incidence rates (IR) were estimated across the study period for both countries. Poisson regression models were built to assess temporal trends in IRs and differences between sexes and age classes. RESULTS: In total, 4,970 and 4,905 incident cases of ET were identified in the UK and France, respectively. The yearly average crude IR (per 100,000 person-years) was 18.20 (95%CI: 15.09-21.32) in UK and 21.42 (17.83-25.00) in France, whereas standardized ones were 19.51 (18.97-20.01) and 19.50 (18.97-20.05). Regression analyses showed slightly increasing trends in both countries, higher incidence among males, and a significant increase with age. Yearly average IR increased from 3.96 (0.95-6.97) and 5.28 (1.12-9.44) in subjects aged <20 years to 49.27 (26.29-72.24) and 51.52 (30.19-72.86) in those aged >80 year in UK and France. CONCLUSIONS: Standardized ET incidence was comparable in the UK and France, showing a slight increase in both countries, reporting a higher value among people aged 60 years and older. This study outlines the need to conduct future studies to estimate the burden of ET in terms of disease control and healthcare resource utilization.

OBJECTIVE: Transient ischaemic attacks (TIA) serve as warning signs for future stroke, and the impact of TIA on long term survival is uncertain. We assessed the long-term hazards of all-cause mortality following a first episode of a transient ischaemic attack (TIA). DESIGN: Retrospective matched cohort study. METHODS: Cohort study using electronic primary health care records from The Health Improvement Network (THIN) database in the United Kingdom. Cases born in or before 1960, resident in England, with a first diagnosis of TIA between January 1986 and January 2017 were matched to three controls on age, sex and general practice. The primary outcome was all-cause mortality. The hazards of all-cause mortality were estimated using a time-varying Double-Cox Weibull survival model with a random frailty effect of general practice, while adjusting for different socio-demographic factors, medical therapies, and comorbidities. RESULTS: 20,633 cases and 58,634 controls were included. During the study period, 24,176 participants died comprising of 7,745 (37.5%) cases and 16,431(28.0%) controls. In terms of hazards of mortality, cases aged 39 to 60 years at the first TIA event had the highest hazard ratio (HR) of mortality compared to their 39-60 years matched controls (HR = 3.04 (2.91 - 3.18)). The HR for cases aged 61-70 years, 71-76 years and 77+ years were 1.98 (1.55 - 2.30), 1.79 (1.20 - 2.07) and 1.52 (1.15 - 1.97) compared to their same-aged matched controls. Cases aged 39-60 at TIA onset who were prescribed aspirin were associated with reduced HR of 0.93 (0.84 - 1.01), 0.90 (0.82 - 0.98) and 0.88 (0.80 - 0.96) at 5, 10 and 15 years respectively, compared to the same aged cases who were not prescribed any antiplatelet. Statistically significant reductions in hazard ratios were observed with aspirin at 10 and 15 years in all age groups. Hazard ratio point estimates for other antiplatelets (dipyridamole or clopidogrel) and dual antiplatelet therapy were very similar to aspirin at 5, 10 and 15 years but with wider confidence intervals that included 1. There was no survival benefit associated with antiplatelet prescription in controls. CONCLUSIONS: The overall risk of death was considerably elevated in all age groups after a first-ever TIA event. Aspirin prescription was associated with a reduced risk. These findings support the use of aspirin in secondary prevention for people with a TIA. The results do not support the use of antiplatelet medication in people without TIA.

BACKGROUND: Cow's milk allergy (CMA) is one of the most common food allergies among children. Whilst avoidance of cow's milk protein is the cornerstone of management, further treatment of symptoms including those affecting the gastrointestinal, skin and respiratory systems plus other allergic comorbidities, maybe required. This study aimed to quantify the wider economic impact of CMA and its management in the United Kingdom (UK). METHODS: We conducted a retrospective matched cohort study on children with CMA (diagnosis read code and/or hypoallergenic formula prescription for ≥3 months) examining healthcare data (medication prescriptions and healthcare professional contacts) from case records within The Health Improvement Network (A Cegedim Proprietary Database) in the UK. A comparative cost analysis was calculated based on healthcare tariff and unit costs in the UK. RESULTS: 6998 children (54% male; mean observation period 4.2 years) were included (n = 3499 with CMA, mean age at diagnosis 4.04 months; n = 3499 matched controls without CMA). Compared to those without CMA, medications were prescribed to significantly more children with CMA (p < 0.001) at a higher rate (p < 0.001). Children with CMA also required significantly more healthcare contacts (p < 0.001) at higher rate (p < 0.001) compared to those without CMA. CMA was associated with additional potential healthcare costs of £1381.53 per person per year. CONCLUSION: The findings of this large cohort study suggest that CMA and its associated co-morbidities presents a significant additional healthcare burden with economic impact due to higher prescribing of additional medications. Further research into management approaches that may impact these clinical and economic outcomes of CMA is warranted.

AIMS: There is concern that cardiovascular disease (CVD) in young adults is rising. However, current trends in the UK are unknown. We investigated sex-specific trends in the incidence and prevalence of CVD in young UK adults. METHODS AND RESULTS: A series of annual (1998-2017) cohort and cross-sectional studies were conducted to estimate incidence rates and prevalence in men and women aged 16-50. Joinpoint regression models were fitted to evaluate changes in trends. From 1998 to 2017, incidence and prevalence had an overall downward trend for ischaemic heart disease (IHD) and angina, while coronary revascularization, stroke/transient ischaemic attack (TIA), and heart failure (HF) had an upward trend in both sexes. Myocardial infarction (MI) trends were stable in men and increased in women. For incidence, the average annual percentage change (AAPC) for men vs. women, respectively, was IHD -2.6% vs. -3.4%; angina -7.0% vs. -7.3%; MI 0.01% vs. 2.3%; revascularization 1.1% vs. 3.9%; stroke/TIA 1.9% vs. 0.6%; HF 5.6% vs. 5.0% (P for trend &lt;0.05 for all except MI and revascularization in men and stroke/TIA in women). For prevalence, AAPCs for men vs. women, respectively, were IHD -2.8% vs. -4.9%; angina -7.2% vs. -7.8%; MI -0.2% vs. 2.0; revascularization 3.2% vs. 4.1%; stroke/TIA 3.1% vs. 3.6%; HF 5.0% vs. 3.0% (P for trend &lt;0.05 for all except MI in men). In recent years, IHD and revascularization trends levelled off, while stroke/TIA and HF trends increased in both sexes. CONCLUSION: Overall trends in incidence and prevalence of CVD are worsening in young adults. Factors behind unfavourable trends warrant investigation and public health intervention.