Exercise-based cardiac rehabilitation may protect people with coronary heart disease from future hospitalisation through several biological and behavioural pathways, according to a new individual participant data meta-analysis that attempts to explain not only whether rehabilitation works, but how it works. The CaReMATCH analysis brings together patient-level information from multiple clinical trials to examine the mechanisms connecting structured exercise programmes with reduced hospitalisation risk. Its findings add a more detailed layer to the established evidence that cardiac rehabilitation can improve recovery after a heart attack, coronary revascularisation or other manifestations of coronary disease. Rather than treating rehabilitation as a single intervention with a single outcome, the study follows the chain of events between exercise training and subsequent health, identifying the changes that may carry much of its protective effect.
Coronary heart disease develops when fatty deposits and inflammatory processes narrow or obstruct the arteries supplying the heart. Even after treatment, patients remain vulnerable to recurrent symptoms, myocardial infarction, heart failure and complications that require hospital care. Exercise-based cardiac rehabilitation is designed to reduce that risk through supervised or prescribed physical activity, education, lifestyle support and clinical monitoring. Patients may train on treadmills, bicycles or through individually tailored walking programmes, with intensity adjusted according to symptoms, fitness and cardiovascular risk. The intervention can improve the ability of the heart, lungs, blood vessels and muscles to work together, but its effects are not limited to aerobic performance. It can also influence blood pressure, body composition, psychological wellbeing, medication adherence and everyday physical activity.
The CaReMATCH investigators used individual participant data rather than relying solely on summary results published by each trial. This approach allows researchers to analyse comparable variables across studies at the level of the individual patient, while accounting for differences in age, sex, baseline disease, treatment and starting fitness. It also makes it possible to investigate mediation. In statistical terms, mediation analysis asks whether an intervention affects an outcome partly because it changes an intermediate factor. For example, rehabilitation may reduce hospitalisation directly through broad physiological adaptation, but it may also reduce admissions indirectly by increasing exercise capacity, improving risk-factor control or helping patients remain physically active after the programme ends. The analysis therefore seeks to map the route from rehabilitation to hospitalisation rather than simply calculating an overall average effect.
The most important candidate mediator is exercise capacity, a measure of how much physical work a person can perform before fatigue, breathlessness or other limiting symptoms appear. It is commonly assessed through a cardiopulmonary exercise test, a symptom-limited exercise test or a functional walking assessment. Physiologically, improved capacity can reflect better cardiac output, more efficient oxygen delivery, improved endothelial function and greater ability of skeletal muscles to extract and use oxygen. These adaptations may allow patients to perform daily activities with less cardiovascular strain. Better fitness is also associated with improved reserve during illness, meaning that a patient may be more capable of tolerating an acute stress before complications become severe enough to require hospital admission. In the CaReMATCH framework, changes in functional capacity are examined as a potential bridge between participation in rehabilitation and later hospitalisation.
The analysis also considers a network of modifiable cardiovascular risk factors. Regular exercise can reduce resting and exertional blood pressure, improve insulin sensitivity and support healthier body composition. It may improve lipid metabolism and help patients adopt behaviours that strengthen the effects of prescribed cardiovascular medicines. Each of these pathways is biologically plausible. Lower blood pressure reduces mechanical stress on arterial walls; better glucose regulation can limit vascular damage; and improved lipid control may slow the progression of atherosclerotic plaque. However, mediation analysis is more demanding than simply observing that two variables improve at the same time. The investigators must assess whether a change in a risk factor occurs after rehabilitation, whether it is associated with hospitalisation, and how much of the intervention’s apparent effect remains after that factor is statistically taken into account.
Physical activity outside formal training is another potential link. A rehabilitation course may provide more than a temporary period of supervised exercise: it can teach patients how to judge intensity, manage symptoms and incorporate movement into ordinary life. This distinction matters because a few weeks of clinic-based exercise may not be enough to produce sustained protection if participants return to prolonged sitting and inactivity. Increased daily activity can reinforce improvements in fitness, weight, blood pressure and metabolic health. It may also reduce the loss of conditioning that often follows a cardiac event. By examining participant-level changes, the CaReMATCH study seeks to determine whether the benefits associated with rehabilitation are connected to what patients do after the structured programme as well as what happens during it.
Psychological and behavioural factors may contribute to the same chain. Anxiety, depression and fear of exertion are common after a heart attack or cardiac procedure, and they can lead patients to avoid activity even when exercise is medically safe. Rehabilitation programmes typically combine exercise with education, reassurance and contact with health professionals, helping patients distinguish normal exertional sensations from warning symptoms. Greater confidence may improve attendance, treatment adherence and willingness to remain active. These changes could influence hospitalisation indirectly by reducing symptom-related emergency visits or supporting long-term disease management. At the same time, the researchers treat such pathways cautiously because psychological measures are not always collected in the same way across trials, and mediation estimates can be sensitive to missing data and differences in follow-up.
A central message of the study is that no single measurement is likely to explain the entire association between exercise-based rehabilitation and hospitalisation. Cardiovascular adaptation, muscle conditioning, risk-factor modification, behaviour and psychological recovery overlap and reinforce one another. A patient who becomes fitter may also walk more, lose weight, sleep better and feel less anxious; separating these effects statistically is difficult because the mediators are not independent switches. The individual participant data design improves the analysis, but it does not turn observational pathways into definitive proof of causation. Mediation models rely on assumptions about timing, confounding and the accuracy of measurements. A factor may appear to mediate an outcome because it is linked to another unmeasured change, such as medication use, social support or access to continuing care.
Even with those limitations, the CaReMATCH findings have practical implications for the future of cardiac rehabilitation. If improvements in exercise capacity account for a substantial part of reduced hospitalisation risk, programmes may need to prioritise reliable fitness assessment and progression of training rather than simply recording attendance. If daily activity, psychological recovery or control of specific risk factors carries additional influence, rehabilitation could be made more personalised, combining exercise prescription with targeted support for inactivity, depression, blood pressure or metabolic health. The broader lesson is that rehabilitation should not be viewed as an optional add-on after acute cardiac treatment. It is a complex secondary-prevention strategy whose effects may emerge through several mutually reinforcing pathways. By identifying those pathways, the new analysis offers clinicians a clearer basis for designing programmes and gives patients a more vivid explanation of why regular, appropriately prescribed exercise can help keep them out of hospital.
Subject of Research: The mechanisms by which exercise-based cardiac rehabilitation influences hospitalisation risk in people with coronary heart disease.
Article Title: Mediators of the effect of exercise-based cardiac rehabilitation on hospitalisation risk in people with coronary heart disease – the CaReMATCH individual participant data meta-analysis
Article References: CaReMATCH individual participant data meta-analysis; published in the European Journal of Epidemiology.
Image Credits: AI Generated
DOI: 10.1007/s10654-026-01432-z
Keywords: coronary heart disease, cardiac rehabilitation, exercise therapy, hospitalisation, mediation analysis, individual participant data meta-analysis, cardiovascular prevention, exercise capacity, physical activity, secondary prevention
