I love physical activity, but…
From a young age, competing in sports since I was 10, physical activity has been a cornerstone of my life. This passion led me through a journey in academia, starting in sports science and transitioning into health and medical research via evidence-based medicine/healthcare (EBM/EBHC). Initially, my research was guided by a strong belief in the transformative power of physical activity. This belief was built on my own activity experiences as well as a primitive and underdeveloped skill in research practice and evidence appraisal. My exposure to EBHC over the past 13 years, with one of its core tenets of rigorous methodology and the reduction of bias, has changed my perspective on the evidence base for physical activity’s role in health. Here’s why.
Physical activity for health — my evidence-based approach
In my research, I’ve examined the evidence addressing physical activity’s impact on chronic health conditions. One study I led, titled ‘Physical activity for the prevention and treatment of major chronic disease: an overview of systematic reviews,’ sought to elucidate the effectiveness of physical activity in both treating and preventing various chronic illnesses. We applied the key tenets of high-quality evidence appraisal and scrutinised data from 56 Cochrane systematic reviews, covering 829 randomised controlled trials (RCTs), and shed light on both the potential benefits and limitations of physical activity interventions across 20 diseases and over 300 outcomes.
A key finding of our study was the varying levels of (un)certainty in the evidence. We discovered that while physical activity has high-quality evidence supporting its benefits in certain conditions, such as osteoarthritis of the knee and hip, and COPD, its effectiveness is less certain in most others. For instance, high-quality evidence indicated no clinically relevant effect of physical activity on important outcomes for conditions like rheumatoid arthritis and postmenopausal osteoporosis. The study revealed that a significant portion of outcomes related to physical activity in chronic diseases had low, or very low certainty of evidence, suggesting a need for more research to reduce uncertainty.
We conclude that recommendations of physical activity in clinical guidelines for various chronic conditions do not always rest on solid, clinically relevant evidence. It underscores the importance of a nuanced understanding and application of physical activity in healthcare, reinforcing the critical need for an evidence-based approach.
In another of my research projects, we focused on the benefits and harms of physical activity in adults diagnosed with irritable bowel syndrome (IBS). This study was significant because current recommendations for people with IBS to engage in physical activity are based on relatively low-level evidence. Our objective was to assess various physical activity interventions, including their type, setting, and nature, and their impact on IBS.
We included 11 randomized controlled trials (RCTs) with 622 participants, examining interventions like yoga, treadmill exercise, and general physical activity support. The study’s findings suggested that physical activity might improve IBS symptoms, but not quality of life or abdominal pain, with the evidence quality rated as very low. This means that while physical activity, particularly yoga, may have some benefits, we can’t be confident about its efficacy due to the low certainty of the evidence.
The conclusions of our study underscore the need for higher-quality research in this area. It also emphasizes the importance of discussions with patients about the uncertainty of evidence surrounding physical activity as a part of IBS symptom management, ensuring fully informed decisions are made.
Even if we have the evidence, can we apply it?
This question becomes crucial when considering the quality of reporting in physical activity research. In my research, “Do Exercise Trials for Hypertension Adequately Report Interventions? A Reporting Quality Study,” we scrutinized the reporting quality of exercise interventions in hypertension treatment. Evaluating 24 RCTs, we found a universal shortfall in reporting all intervention components comprehensively. This gap in reporting details like adherence, provider information, and adverse events severely limits the ability to replicate studies and apply their findings in practice.
Concerns about reporting standards in physical activity studies are echoed beyond my research. A study published in the British Journal of Sports Medicine highlights similar issues across a range of physical activity research. This study underscores that the problem of inadequate reporting is not confined to a specific area but is a widespread issue in the field. Such gaps in reporting crucial details of research studies impede the replication of these studies and the application of their findings in clinical practice. Highlighting these issues is crucial in advocating for improved reporting standards, which are essential for the advancement of evidence-based practice in physical activity and healthcare.
The reporting and methodological issues are far from unique to physical activity research; they are pervasive issues across the broader spectrum of medical and health research. However, the EBM movement has made significant strides in improving both the quality of research methods and the transparency of reporting in these fields. This progress underscores the need for a similar focus in the physical activity research community. By adopting rigorous methodologies and enhancing reporting standards, we can elevate the quality and impact of our research, ensuring that findings are not only scientifically sound but also practically applicable in improving health outcomes.
An evidence-informed change in my perspective
We’ve had decades of research attempting to prove the benefits of physical activity for health. This in turn is used to promote physical activity uptake through the lens of “health benefits”. I don’t think this approach is working. As we adopt a more evidence-based approach, the health benefits of physical activity are not always as clear as once thought. The focus should shift towards understanding other benefits of physical activity and creating a stronger evidence base for how to increase uptake in those most in need.
My time in the health and medical field has also heightened my awareness of other critical health drivers, notably social determinants of health — the conditions under which we live. These factors are challenging to study and address, and it appears that the research community often opts for more straightforward studies focusing on individual lifestyle elements, such as activity levels, diet, and sleep. Yet, this approach risks missing the broader and more relevant factors impacting health.
Two recent studies are good examples of the evidence informing these changes in my perspective.
The PrAISED trial is one of the most comprehensive efforts to provide the best possible conditions for evaluating the efficacy of physical activity in improving health outcomes that I’ve come across. A large, independently funded study in individuals with early dementia or mild cognitive impairment, it employed a detailed, clinically pragmatic implementation strategy that ensured participants received the “required dose” of physical activities targeting relevant health outcomes (self-reported daily living activities, physical activity levels, quality of life, balance, functional mobility, fear of falling, frailty, cognition, mood, carer strain, incidents of falls). Despite this optimal setup, participants engaged in the physical activity program did not show greater improvements in any of the measured outcomes when compared to the control group (usual care plus a falls risk assessment).
One key takeaway from the PrAISED trial is that physical activity may not be effective in restoring function or reducing health declines in populations like those with early dementia or mild cognitive impairment. This finding could potentially extend to other groups as well. For me, there is another key takeaway (and one I think echoed by the authors) which is the need to move away from the prevailing research focus on “exercise as medicine”. The authors conclude that “physical activity should be promoted for enjoyment, social inclusion, and enhancing relationships, highlighting its value beyond just health benefits.” I would agree. I would stress that these claims can also be better evidenced. This will require a move away from interventional and quantitative research to address.
A second recent study involved Finnish twins and explored the relationship between leisure-time physical activity (LTPA), biological ageing, and mortality. The study employs a twin design to isolate the effects of LTPA from genetic and early-life environmental influences. It categorizes participants into four LTPA groups and reveals that both low and high levels of activity are associated with faster biological ageing. Importantly, when the study adjusts for factors such as education, smoking, and alcohol consumption, the link between LTPA and mortality risk significantly diminishes and becomes highly uncertain (the risk of early mortality could be decreased or increased with higher levels of physical activity).
It’s important to note that while providing valuable insights, the study does have certain limitations that should be considered. Firstly, its reliance on self-reported data for physical activity levels may introduce recall bias or inaccuracies in reporting. Additionally, the study’s specific cultural and demographic context, centred on a Finnish population, may limit the generalisability of its findings to other populations with different genetic, environmental, or lifestyle backgrounds. While the twin study design is powerful for controlling genetic and familial environmental factors, it may not fully account for all variables that influence health outcomes, such as socioeconomic status or access to healthcare resources, which could impact the interpretation of the relationship between physical activity, biological ageing, and mortality. Finally, the study has not yet been formally peer-reviewed. Though peer-review is no guarantee of quality by any means.
If the findings hold, this study exemplifies the limitations of research that focuses narrowly on physical activity, neglecting the complex interplay of various health determinants. It serves as a reminder that while physical activity is important, its impact is intertwined with, and often overshadowed by genetic, familial, and broader societal factors.
Addressing physical activity for health through an EBHC lens has led me to question the overall effectiveness of a research approach that isolates individual lifestyle factors. If we ask ourselves, honestly, what has this siloed approach achieved? Are societies getting healthier and avoiding disease and ill health? I’m not sure. Some might argue (ill) health would be even worse than it currently is without the (siloed) research to date. Again, I’m not so sure.
Paradoxically, the emphasis on individual lifestyle factors may have contributed to more ill health through a rise in over-medicalisation and inequitable care, with those least in need often receiving unnecessary healthcare. It has also allowed those responsible for public health to neglect the more challenging social determinants, conveniently drawing our collective gaze to the lower-hanging fruit of single lifestyle factors and the individual as the change agent.
The continued focus on evidencing an association between human behaviours and health needs to change. The “decline in health” approach to physical activity research may also have run its course. The PrAISED trial informs us that even under the best of conditions, physical activity interventions alone may not overcome the decline in health for some conditions. This mirrors findings from my work as well as additional recent trials. Where we have evidence of an effect, in most cases the effect is small. And when we want to translate any effective evidence into practice, we struggle due to poor reporting and insufficient implementation evidence. The Finnish twin study underscores the importance of integrating the study of social determinants with traditional health research. We must adopt a more holistic approach that considers the myriad factors contributing to health, moving beyond the simplicity of focusing solely on individual lifestyle elements. These are the areas where I now focus my physical activity research.
My 23-year journey in health research, the last 13 underpinned by the principles of EBHC, has led to a crucial understanding: a healthy life is a complex interplay of behaviours, genetics, social environment, and more. To effectively improve societal health, our research, policies, and interventions must reflect this complexity.
