Explore the vast range of titles available.

Cardiovascular disease (CVD) continues to be a significant health threat to humans globally, and a significant burden on healthcare systems. Cardiovascular risk prediction utilizing machine learning (ML) models in patients with asthma remains vastly underexplored. In this cohort study consisting of 641,042 participants, we used routinely collected electronic healthcare record data to explore various ML algorithms including logistic regression, penalized logistic regression, decision trees, random forest and gradient boost to develop a model with high specificity. The penalized logistic regression model was identified to be the best and simplest classification model in terms of discriminatory power (AUC = 0.85). The gradient boost model was found to be the best predictive model in terms of calibration where the predicted and observed probabilities at risk of CVD match or are closely aligned. In all models, the number of previous cardiovascular events was the most influential predictor, followed by age and prescriptions related to cardiovascular medications. The top predictor alone produced a reasonable level of predictive power (AUC = 0.66). We have created a novel prediction model for predicting CVD within a year of asthma diagnosis for patients with asthma at least 50 years old. Using penalized logistic regression, we achieved a high level of accuracy. By implementing this model, it would be possible to screen out patients with low risk of CVD with high specificity and acceptable sensitivity. Penalized logistic regression and gradient boost models have similar accuracy in screening out individuals at low risk of CVD. For this objective, penalized logistic regression may be more suitable than gradient boost models for implementation as it is simpler to use and more transparent. At the probability threshold of 8% (outcome prevalence), both models' effectiveness in reducing unnecessary treatments was by approximately 52%. These ML models performed better compared with traditional statistical-based risk prediction models. The unique contribution of the study is the construction of prediction models for CVD disease within 12 months from asthma diagnosis based on regression and machine learning models and the comparison of their accuracy to identify the best model based on suitable statistical measures such as AUC and calibration. Further prospective studies using different populations and external validation are required to assess and validate the ML risk prediction models.

Polypharmacy, prescription of multiple medications to a patient, is a major challenge for health systems. There have been no peer-reviewed studies of polypharmacy prevalence and medication cost at a population level in England. To determine prevalence and medication cost of polypharmacy, by patient characteristics. Design and setting: Retrospective cohort study of North West London electronic health records. We quantified prevalence and direct cost of polypharmacy (five or more regular medications), stratified by demographics and frailty. We fitted a mixed-effects logistic regression for polypharmacy. Of 1.7 million adults, 167,665 (9.4%) were on polypharmacy. Age and socio-economic deprivation were associated with polypharmacy (OR 9.24 95% CI 8.99 to 9.50, age 65-74 compared with 18-44; OR 0.68 95% CI 0.65 to 0.71, least deprived compared with most). Polypharmacy prevalence increased with frailty (OR 1.53 95% CI 1.53 to 1.54 per frailty component, for White women). Men had higher odds of polypharmacy than women at average frailty (OR 1.26 95% CI 1.24 to 1.28) and with additional frailty components (OR 1.10 95% CI 1.09 to 1.10). Black people had lower odds of polypharmacy at average frailty (OR 0.82 95% CI 0.79 to 0.85, compared with White), but along with other ethnicities, saw greater odds increases with increasing frailty (OR 1.02 95% CI 1.01 to 1.03). Annual medication cost 8.2 times more for those on polypharmacy compared with not (£370.89 and £45.31). Demographic characteristics are associated with polypharmacy, after adjusting for frailty. Further research should explore why, to reduce health inequities and optimise cost associated with polypharmacy.

Summary Objectives To identify the number and characteristics of patients with polypharmacy receiving structured medication reviews (SMRs) and medication reviews in primary care in 2022, and to evaluate whether the provision of these services is equitable across different demographic and socio-economic groups. Design Cross-sectional study. Setting Primary care networks in North West London, UK. Participants Adults registered with a general practitioner (GP) and regularly prescribed at least five medicines or more. Main outcome measures Receipt of at least one SMR and any kind of medication review during the study period (2022). Results Among 515,042 adults regularly prescribed with medication, 167,482 were regularly prescribed at least five medicines, defined as polypharmacy. 53.3% (89,220) of these patients received at least one kind of medication review and 17.2% (11,954) of them received SMRs. Patients who were males, black, more affluent, and frailer, were more likely to receive medication reviews, while those who were males, less affluent, and frailer, were more likely to receive SMRs. Conclusions Although polypharmacy was common in North West London, only about half of eligible patients received medication reviews, and only 17.2% received SMRs. Different distributions of medication reviews and SMRs by demographic and socio-economic characteristics may indicate inequities in the provision of these services. Policy makers should consider effective ways to incentivise the equitable provision of SMRs.

To determine demographic and clinical characteristics associated with uptake of COVID-19 vaccines among pregnant women, and quantify the relationship between vaccine uptake and admission to hospital for COVID-19. Pregnant women are at increased risk of severe adverse outcomes from COVID-19. Since April 2021, COVID-19 vaccines were recommended for pregnant women in the UK. Despite this, evidence shows vaccine uptake is low. However, this evidence has been based only on women admitted to hospital, or on qualitative or survey-based studies. Retrospective cohort study including all pregnancies ending between 18 June 2021 and 22 August 2022, among adult women registered with a Northwest London general practice. Statistical analyses were mixed-effects multiple logistic regression models. We conducted a nested case-control analysis to quantify the relationship between vaccine uptake by end of pregnancy and hospitalisation for COVID-19 during pregnancy. Our study included 47,046 pregnancies among 39,213 women. In 26,724 (57%) pregnancies, women had at least one dose of vaccine by the end of pregnancy. Uptake was lowest in pregnant women aged 18–24 (33%; reference group), Black women compared with White (37%; OR 0.55, 95% CI: 0.51 to 0.60), and women in more deprived areas (50%; reference group). Women with chronic conditions were more likely to receive the vaccine than women without (Asthma OR 1.21, 95% CI: 1.13 to 1.29). Patterns were similar for the second dose. Women admitted to hospital were much less likely to be vaccinated (22%) than those not admitted (57%, OR 0.22, 95% CI: 0.15 to 0.31). Women who received the COVID-19 vaccine were less likely to be hospitalised for COVID-19 during pregnancy. COVID-19 vaccine uptake among pregnant women is suboptimal, particularly in younger women, Black women, and women in more deprived areas. Interventions should focus on increasing uptake in these groups to improve health outcomes and reduce health inequalities.