health prior to having sufficient hyperglycaemia to be diagnosed as having diabetes. Along with treatment inequalities, these physiological differences may also contribute to the observed differences in relative risk of cardiovascular disease between men and women with diabetes. It may also be that the relative effect of diabetes (or any other single cardiovascular risk factor) is greater in low-risk groups than high-risk groups (as noted above), and that this simply reflects that many risk factors have overlapping mechanistic pathways. Thus, adding the first or second risk factor to low-risk groups will have more of an effect than adding the fourth or fifth risk factor to high-risk groups.
Predicting Risk in Individuals
Increasingly, international guidelines are recommending treatment decisions be based on an individual patient’s overall cardiovascular risk rather than solely targeting particular risk factor levels [27, 28]. There is also a growing focus on tailoring treatment targets, including glycaemic targets, for individuals according to a personalised risk-benefit assessment. Therefore, clinicians need to be equipped with strategies to interpret and apply epidemiological evidence in their provision of care to individual patients. Such strategies should be rigorously investigated in interventional trials. Determining an individual’s absolute risk of cardiovascular disease and knowing at what threshold of risk to intervene and with what therapy are not straightforward. Hence, existing guidelines have varying suggested approaches.
Intensive pharmacological risk factor modification for patients with established cardiovascular disease (i.e., “secondary prevention”) is an accepted standard of care with a robust evidence base. Clinical risk assessment and management for patients with diabetes but without known cardiovascular disease is more complicated, with guidelines suggesting various potential approaches. Essentially, the goal is to translate available evidence into day-to-day clinical care, to maximise benefits to individuals, while minimising treatment-related harm and economic costs. Individual patients and healthcare settings will have unique factors that influence treatment decisions that may not always align with published guidelines. Nevertheless, guidelines provide a valuable framework for standardisation of care.
Figure 6 provides a conceptual overview of different approaches to considering cardiovascular risk in patients with diabetes. One potential approach in primary prevention is to treat specific risk factors in all patients when they are outside of the normal range. For example, to commence an antihypertensive agent if the systolic blood pressure remains above 140 mm Hg or a statin if the LDL level is 3.5 mmol/L (135 mg/dL) or higher.
Fig. 6. Various approaches used to assess the need for pharmacological therapy to modify cardiovascular risk factors in patients with diabetes.
An alternate approach is to determine initiation of pharmacotherapy on the basis of an individual’s overall cardiovascular risk. Some guidelines essentially recommend treating all patients with diabetes as being at high risk. For example, the 2013 European Society of Cardiology and European Association for the Study of Diabetes guidelines consider all patients with type 2 diabetes at “high risk” and recommend statin therapy with a target LDL of <2.5 mmol/L (97 mg/dL) [27]. Patients with type 2 diabetes and an additional risk factor are recommended statin therapy with a target LDL of less than 1.8 mmol/L (80 mg/dL). Similarly, the 2018 American Diabetes Association guidelines recommend moderate intensity statin therapy to all patients with diabetes aged 40 years and above [28].
A more nuanced risk assessment option is to calculate an estimated absolute cardiovascular risk for a patient using a validated risk score and then initiate treatment in patients who score above a specified threshold. For example, the 2014 National Institute for Health and Care Excellence guidelines recommend commencement of statin therapy in patients with type 2 diabetes and a 10% or greater 10-year risk of developing cardiovascular disease on the QRISK2 assessment tool [29].
Cardiovascular Disease Risk Scores
As early as the late 1970s, the first widely used cardiovascular risk score was developed, which is the Framingham Risk Score [30]. Over recent decades, more than 45 additional scores applicable to patients with diabetes have been published [31]. Some have been developed in populations with diabetes and others in general populations but include diabetes status as a variable. Most cardiovascular risk scores that have been assessed in diabetes populations include similar parameters such as age, sex, smoking status, systolic blood pressure, lipids and, for those scores that are not diabetes-specific, diabetes status. Some scores include additional factors such as HbA1c; weight or body mass index; chronic kidney disease stage, estimated glomerular filtration rate (eGFR), or presence of albuminuria; presence of retinopathy; ethnicity; family history; and markers of social deprivation. Risk scores are now being incorporated into diabetes guidelines and increasingly used in clinical practice, mostly being available as charts or as web-based calculators. The statistical models underlying these scores use a patient’s personal risk factor values and the average cardiovascular disease risk of the population to predict the individual’s risk of a cardiovascular event within a specified time frame, often 10-years. Establishing a clinical risk score should involve 3 key stages [31]:
1. development and internal validation
2. external validation
3. model impact studies
Risk Score Development and Internal Validation
Risk scores are generally developed using data from prospective observational cohort studies. The predictive models calculate a score from a series of weighted risk factors. The weights are regression coefficients or hazard ratios that account for how important each risk factor is in predicting cardiovascular disease. There are a number of stages in selecting the variables to be incorporated into a model, with various statistical methodologies employed. Potentially relevant risk factor variables are selected for investigation on the basis of clinical knowledge and existing epidemiological evidence. The associations between these variables and cardiovascular outcomes are modelled, initially in univariate analyses. Various models are then built, adding and removing variables to see whether they benefit the model in terms of goodness of fit (i.e., whether they improve the correlation between risk predicted by the model and risk observed in the study cohort). Variables that improve the model’s predictive capabilities are generally retained. Other factors may also contribute to the decision to retain or exclude variables from the final predictive model, including cost and ease of measurement and potential impact on patient motivation for lifestyle modification for example [32].
External Validation
External validation of risk scores is essential because prediction models generally perform better within the cohort in which they were created than when applied to another population. A systematic review of cardiovascular risk prediction models for patients with type 2 diabetes found that just under a third of published risk scores had been validated in external study cohorts [31]. Where there are substantial differences between populations, the predictive model may need to be recalibrated. Large studies of pooled international observational cohorts have shown that