Interrogating disease clusters, multimorbidity and adverse outcomes in chronic kidney disease

Background:

People with kidney disease often have lots of health problems. They usually have more health problems than people whose kidneys work normally. They are more likely to have heart problems or die at a younger age. We do not know enough about why this is, or what can be done to stop these things from happening.

Looking after patients with kidney disease and multiple health problems is complicated. It is not always possible to use tests and treatments that are used in people whose kidneys work normally. We need to understand better the effects of having multiple health problems together with kidney disease on people's lives.

Aims:

1) Find out the differences between people whose kidneys work normally and people with kidney disease.
2) Among people with kidney disease and multiple health problems, find out the rates of the following health events: heart attacks, strokes, admissions to hospital and deaths.
3) Among people with kidney disease and multiple health problems, find clusters of people who are most likely to have health events.
4) Find the most effective way to predict which people with kidney disease and multiple health problems will have health events.

Information Sources:

This project will use information that has already been collected by doctors and researchers. The information about patients is held on computer systems. One source is a study called the UK Biobank, which involved 500,000 volunteers from across the UK. The people who took part had blood and urine tests and filled out questionnaires about their lives and medical problems. The second source is called the SAIL databank. This is where hospitals and GPs in Wales keep the information about 2.4 million people. The SCREAM cohort, based on 1.2 million people in Sweden, is the third source. The patients in all three sources cannot be identified because the information is anonymous.
Benefits from the Planned Programme of Research:

We will find people with kidney disease and multiple health problems who are more likely to have health events than others. If we can identify them early, they could be given better treatment. We might be able to keep them healthy for longer or find certain patients in whom more research needs to be done.

Patients have been involved in our research from the start, helping us plan the project. However, we will not be collecting new information from patients. Instead, we will be using the data from four million people whose data has already been collected. Patients therefore will not need to spend extra time with researchers for the main part of our project. The information they have already given will make the research possible and they will be involved later, helping us to use the results of our research to help patients.

Aim:

To improve our understanding of the risks of adverse clinical outcomes associated with chronic kidney disease (CKD) and multimorbidity.

Context:

I have performed a systematic review on the area and identified an association between multimorbidity and mortality in patients with CKD. However, little is known about the risks in patients with mild to moderate CKD and about the risk of cardiovascular (CV) outcomes.

Objectives:

1) To compare the characteristics of patients with CKD to those with normal kidney function.
2) To determine the associations between multimorbidity count (number of of Long Term Conditions (LTCs)) and adverse outcomes (mortality, hospitalisation and CV events) in patients with CKD.
3) To identify clusters of LTCs and patient factors in patients with CKD who are highest risk for adverse clinical outcomes.
4) To determine the most effective predictor of adverse outcomes in patients with CKD and multimorbidity: multimorbidity count or clusters of LTCs and patient factors.

Methods:

This Fellowship will use a data-driven approach in three large and high-quality datasets. UK Biobank has comprehensive data on 500,000 volunteers. Data are partly self-reported and are validated by linked health care records. The SAIL databank provides data from the routine care of 2.4 million patients in Wales. The SCREAM cohort is from the routine care of 1.2 million Swedish patients. Data from all three datasets will be analysed using survival analyses and clustering statistical techniques.

Opportunities:

It is well known that patients with CKD are at high risk of early mortality and CV events, but little is known about how multimorbidity influences different adverse clinical outcomes, especially in mild to moderate CKD. By targeting this evidence gap, this Fellowship will contribute to the development of clinical guidelines and direct future health policy and clinical trials.

Patients with chronic kidney disease (CKD) and multimorbidity are at high risk of cardiovascular events and early mortality. Our recent research has shown that little is known about the population with mild to moderate CKD and how specific risk factors lead to different outcomes. This Fellowship will identify risk factors associated with accelerated risks and inform the development of clinical guidelines with focus on the management of patients with multimorbidity. Potential beneficiaries from increased understanding of these problems are patients, the public, carers, clinicians, policy-makers and academics.

Patients, Carers and the Public:

In the long term, the main beneficiaries of this research will be patients with CKD. As patients live longer with chronic diseases, the number of patients with CKD is rising gloablly. We must understand how to improve the care of these patients and to help patients and their carers to understand what risks they face and how to manage these risks. This could be by finding individuals who are high-risk and are likely to benefit from lifestyle changes or interventions to reduce risk. Low-risk individuals may not benefit from intense risk factor modification and therefore would be helped by deprescribing. The views of patients and the public have been integral to this study's design and, in order to maximise the impact for the public, the Fellowship has a key component of public engagement throughout its lifespan.

Clinicians:

The study findings will help clinicians to identify those patients with CKD and multimorbidity that are highest risk of adverse clinical outcomes. Such knowledge will help clinicians to understand the risks faced by individuals with CKD and multimorbidity. They will have access to better information on this subject and therefore be able to communicate risk to patients better and explain the risks and benefits of treatments in greater detail.

Policy-makers:

Identifying new combinations of risk factors for adverse outcomes will inform new policies and strategies that aim to reduce harm. Current clinical guidelines on CKD such as KDIGO and NICE have little evidence regarding how to account for multimorbidity in patient care. This Fellowship will contribute to the development of future guidelines, moving from a single disease-centred approach to a patient-centred one.

Academics:

Dissemination of the methods used in this Fellowship will benefit researchers who can apply these methods to similar problems, especially epidemiological areas where there are numerous variables to consider. The computer code developed as part of the methods will be made available via github.com, which will help researchers working in similar fields.

New information from this Fellowship will facilitate nephrologists in generating new ideas in improving our understanding of multimorbidity and CKD. Results from the Fellowship will inform future clinical trial design, leading to more research in these complex groups of patients. Qualitative researchers may also be interested in exploring patient perspectives of living with the different patterns of multimorbibidy and CKD identified through this work so this research may open up new opportunities for future sociological research too.