Current approaches to predict cardiovascular risk fail to identify many people who would benefit from preventive treatment, while others receive unnecessary intervention. Machine-learning offers opportunity to improve accuracy by exploiting complex interactions between risk factors. We assessed whether machine-learning can improve cardiovascular risk prediction. Prospective cohort study using routine clinical data of 378,256 patients from UK family practices, free from cardiovascular disease at outset. Four machine-learning algorithms (random forest, logistic regression, gradient boosting machines, neural networks) were compared to an established algorithm (American College of Cardiology guidelines) to predict first cardiovascular event over 10-years. Predictive accuracy was assessed by area under the ‘receiver operating curve’ (AUC); and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) to predict 7.5% cardiovascular risk (threshold for initiating statins). 24,970 incident cardiovascular events (6.6%) occurred. Compared to the established risk prediction algorithm (AUC 0.728, 95% CI 0.723–0.735), machine-learning algorithms improved prediction: random forest +1.7% (AUC 0.745, 95% CI 0.739–0.750), logistic regression +3.2% (AUC 0.760, 95% CI 0.755–0.766), gradient boosting +3.3% (AUC 0.761, 95% CI 0.755–0.766), neural networks +3.6% (AUC 0.764, 95% CI 0.759–0.769). The highest achieving (neural networks) algorithm predicted 4,998/7,404 cases (sensitivity 67.5%, PPV 18.4%) and 53,458/75,585 non-cases (specificity 70.7%, NPV 95.7%), correctly predicting 355 (+7.6%) more patients who developed cardiovascular disease compared to the established algorithm. Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others.

This study is to identify, summarise and synthesise literature on the causes of the evidence to practice gap for complex interventions in primary care. This study is a systematic review of reviews. MEDLINE, EMBASE, CINAHL, Cochrane Library and PsychINFO were searched, from inception to December 2013. Eligible reviews addressed causes of the evidence to practice gap in primary care in developed countries. Data from included reviews were extracted and synthesised using guidelines for meta-synthesis. Seventy reviews fulfilled the inclusion criteria and encompassed a wide range of topics, e.g. guideline implementation, integration of new roles, technology implementation, public health and preventative medicine. None of the included papers used the term “cause” or stated an intention to investigate causes at all. A descriptive approach was often used, and the included papers expressed “causes” in terms of “barriers and facilitators” to implementation. We developed a four-level framework covering external context, organisation, professionals and intervention. External contextual factors included policies, incentivisation structures, dominant paradigms, stakeholders’ buy-in, infrastructure and advances in technology. Organisation-related factors included culture, available resources, integration with existing processes, relationships, skill mix and staff involvement. At the level of individual professionals, professional role, underlying philosophy of care and competencies were important. Characteristics of the intervention that impacted on implementation included evidence of benefit, ease of use and adaptability to local circumstances. We postulate that the “fit” between the intervention and the context is critical in determining the success of implementation. This comprehensive review of reviews summarises current knowledge on the barriers and facilitators to implementation of diverse complex interventions in primary care. To maximise the uptake of complex interventions in primary care, health care professionals and commissioning organisations should consider the range of contextual factors, remaining aware of the dynamic nature of context. Future studies should place an emphasis on describing context and articulating the relationships between the factors identified here. PROSPERO CRD42014009410

The presence of additional chronic conditions has a significant impact on the treatment and management of type 2 diabetes (T2DM). Little is known about the patterns of comorbidities in this population. The aims of this study are to quantify comorbidity patterns in people with T2DM, to estimate the prevalence of six chronic conditions in 2027 and to identify clusters of similar conditions. We used the Clinical Practice Research Datalink (CPRD) linked with the Index of Multiple Deprivation (IMD) data to identify patients diagnosed with T2DM between 2007 and 2017. 102,394 people met the study inclusion criteria. We calculated the crude and age-standardised prevalence of 18 chronic conditions present at and after the T2DM diagnosis. We analysed longitudinally the 6 most common conditions and forecasted their prevalence in 2027 using linear regression. We used agglomerative hierarchical clustering to identify comorbidity clusters. These analyses were repeated on subgroups stratified by gender and deprivation. More people living in the most deprived areas had ≥ 1 comorbidities present at the time of diagnosis (72% of females; 64% of males) compared to the most affluent areas (67% of females; 59% of males). Depression prevalence increased in all strata and was more common in the most deprived areas. Depression was predicted to affect 33% of females and 15% of males diagnosed with T2DM in 2027. Moderate clustering tendencies were observed, with concordant conditions grouped together and some variations between groups of different demographics. Comorbidities are common in this population, and high between-patient variability in comorbidity patterns emphasises the need for patient-centred healthcare. Mental health is a growing concern, and there is a need for interventions that target both physical and mental health in this population.

Family history is a risk factor for many chronic diseases, including cancer, cardiovascular disease, and diabetes. Professional guidelines usually include family history to assess health risk, initiate interventions, and motivate behavioral changes. The advantages of family history over other genomic tools include a lower cost, greater acceptability, and a reflection of shared genetic and environmental factors. However, the utility of family history in public health has been poorly explored. To establish family history as a public health tool, it needs to be evaluated within the ACCE framework (analytical validity; clinical validity; clinical utility; and ethical, legal, and social issues). Currently, private and public organizations are developing tools to collect standardized family histories of many diseases. Their goal is to create family history tools that have decision support capabilities and are compatible with electronic health records. These advances will help realize the potential of family history as a public health tool.

Lipoprotein(a), Lp(a), is a modified atherogenic low-density lipoprotein particle that contains apolipoprotein(a). Its levels are highly heritable and variable in the population. This consensus statement by HEART UK is based on the evidence that Lp(a) is an independent cardiovascular disease (CVD) risk factor, provides recommendations for its measurement in clinical practice and reviews current and emerging therapeutic strategies to reduce CVD risk. Ten statements summarise the most salient points for practitioners and patients with high Lp(a). HEART UK recommends that Lp(a) is measured in adults as follows: 1) those with a personal or family history of premature atherosclerotic CVD; 2) those with first-degree relatives who have Lp(a) levels >200 nmol/l; 3) patients with familial hypercholesterolemia; 4) patients with calcific aortic valve stenosis and 5) those with borderline (but <15%) 10-year risk of a cardiovascular event. The management of patients with raised Lp(a) levels should include: 1) reducing overall atherosclerotic risk; 2) controlling dyslipidemia with a desirable non-HDL-cholesterol level of <100 mg/dl (2.5 mmol/l) and 3) consideration of lipoprotein apheresis.

Objective To identify, summarise and synthesise available literature on the effectiveness of implementation strategies for optimising implementation of complex interventions in primary care. Design Systematic review of reviews. Data sources MEDLINE, EMBASE, CINAHL, Cochrane Library and PsychINFO were searched, from first publication until December 2013; the bibliographies of relevant articles were screened for additional reports. Eligibility criteria for selecting studies Eligible reviews had to (1) examine effectiveness of single or multifaceted implementation strategies, (2) measure health professional practice or process outcomes and (3) include studies from predominantly primary care in developed countries. Two reviewers independently screened titles/abstracts and full-text articles of potentially eligible reviews for inclusion. Data synthesis Extracted data were synthesised using a narrative approach. Results 91 reviews were included. The most commonly evaluated strategies were those targeted at the level of individual professionals, rather than those targeting organisations or context. These strategies (eg, audit and feedback, educational meetings, educational outreach, reminders) on their own demonstrated a small to modest improvement (2–9%) in professional practice or behaviour with considerable variability in the observed effects. The effects of multifaceted strategies targeted at professionals were mixed and not necessarily more effective than single strategies alone. There was relatively little review evidence on implementation strategies at the levels of organisation and wider context. Evidence on cost-effectiveness was limited and data on costs of different strategies were scarce and/or of low quality. Conclusions There is a substantial literature on implementation strategies aimed at changing professional practices or behaviour. It remains unclear which implementation strategies are more likely to be effective than others and under what conditions. Future research should focus on identifying and assessing the effectiveness of strategies targeted at the wider context and organisational levels and examining the costs and cost-effectiveness of implementation strategies. PROSPERO registration number CRD42014009410.