Personalised care for older patients with myocardial infarction (MiRisk)

People are living longer. As a result, an increasing proportion of our society are living with two or more long-term health conditions, a state called multimorbidity, or suffer a degree of frailty. Heart disease is common and a heart attack is the leading cause of death or disability in older patients. Increased age, multimorbidity and frailty all influence how a patient recovers from a heart attack as well as the risk and benefit of potential treatments. Despite accounting for a large proportion of patients seen in clinical practice, older patients are underrepresented in large clinical trials and we therefore do not know the best way of managing these complex patients.

In our research so far, we have shown that older patients receive fewer recommended treatments for a heart attack compared to younger patients, despite being at higher risk of future heart problems and therefore potentially having the most to gain from therapy. This likely reflects the complex nature of these patients and the difficulty clinicians have in identifying older patients in whom the benefits of available treatments outweigh the risks which may be increased due to age, frailty and other health problems.

We want to learn the reasons behind treatment differences between younger and older patients and use this information to develop tools, personalised for each patient, which can aid decision making in this challenging and increasingly common patient group. To do this, we can take advantage of the large amounts of anonymised electronic patient data that is routinely collected and securely stored in national healthcare databases. This allows us to study patients who have suffered a heart attack at some point in the past. This data has already been collected and means patients will not need to spend any additional time with researchers or have changes made to the care they would normally receive. We can see what impact age, multimorbidity and frailty had on the management of these patients and assess how these factors influenced their recovery. This information will allow us to design a decision tool specifically for these complex patients, called MiRisk, which will provide information on the risk of key outcomes, such as death due to a heart problem, following a heart attack. This will help patients make informed decisions about their care and doctors to identify patients who will benefit from treatments, regardless of age.

This work will be performed by a team of specialist heart researchers based at the Centre for Cardiovascular Sciences, University of Edinburgh, who also jointly work within the NHS to look after patients with heart disease. Patient data will be accessed using the newly established DataLoch, a secure repository of all health and social care data for the South East Region of Scotland. DataLoch was created to help find solutions to the major health and social care challenges we all face. It brings together patient data that is currently routinely collected as part of people's day to day interactions with health and social care services. This includes visits to hospitals or GPs, treatments and medicines received, and outcomes and test results. All data is stored securely in line with national data protection law and the project has been developed with public involvement. Data used in this study will be anonymised meaning individual patients cannot be identified.

We believe knowing how multimorbidity and frailty impact a patient's recovery following a heart attack will help patients and their families make informed decision about their care and allow doctors to target treatments to those who have most to benefit.

Guidelines recommend the use of risk stratification tools to guide the management of patients with myocardial infarction. Frailty and multimorbidity are strong predictors of adverse outcomes following myocardial infarction and are increasingly common in older patients yet neither are objectively measured or used to guide patient management. There is evidence to suggest incorporating these measures into current risk tools improves discrimination in older patients.

I aim to develop a personalised risk prediction model for use in older patients with myocardial infarction. I hypothesise that a risk prediction model which includes important non-cardiovascular factors such as frailty and multimorbidity will lead to improved risk stratification. I hypothesis that a risk prediction model capable of estimating the separate risks of myocardial infarction, cardiovascular death and non-cardiovascular death would greatly aid decision making in older patients. These hypotheses will be tested in three phases.
Study 1: Analysis of the impact of patient age, frailty and multimorbidity on the management and outcomes of 48,282 consecutive patients with suspected myocardial infarction who participated in a randomised control trial. We anticipate frailty and multimorbidity are important factors in decision making and have an incremental impact on cardiovascular and non-cardiovascular outcomes.
Study 2: Derivation, internal validation and independent external validation of a multiple outcome clinical risk prediction model (MiRisk). MiRisk will be derived and validated using a population database containing detailed primary, secondary and social care data on patients suffering a myocardial infarction between 2012-2019 in the South East Region of Scotland.
Study 3: Comparison of MiRisk and the Global Registry of Acute Cardiac Events (GRACE) score in the 2,949 patients with myocardial infarction present in the HighSTEACS trial database.