top of page

Join The Health Shift

Small, practical health insights—straight to your inbox.

Receive new articles, useful tools and occasional updates as The Health Shift grows.

Why Health Advice Feels So Confusing

Writer: Dr. Saartje Jooris
Dr. Saartje Jooris
May 5
7 min read

Updated: Sep 1

If you have ever tried to become healthier, you have probably experienced it. You decide to make a few changes, start reading and, half an hour later, feel less certain than when you began.


Illustration of an overwhelmed woman surrounded by conflicting health tips, research headlines and siIllustration of an overwhelmed woman surrounded by conflicting health tips, research headlines and signposts, showing how fragmented advice can create confusion, guilt and inaction.gnposts, showing how fragmented advice can create confusion, guilt and inaction.

One person recommends more protein, while another warns that most people already eat too much. Running is excellent for the heart, unless you are listening to someone concerned about injuries. One expert recommends fasting; another insists that breakfast matters. Coffee protects your health. Coffee disrupts your sleep.


We tend to assume that if the science were reliable, everyone would eventually arrive at the same answer. When they do not, it can seem as though nobody really knows anything—or that every recommendation is simply another opinion.


The reality is more interesting. Health advice conflicts for several reasons, and disagreement does not always mean that one side is wrong.


The same advice does not fit everyone

Illustration of a hand adding another health tip to an overflowing box, alongside a list showing how disconnected advice can be difficult to prioritise and sustain.

Many health recommendations are presented as though they apply equally to all of us. They rarely do.


A young athlete training several times a week has different nutritional needs from an older person who has lost muscle during an illness. Someone doing physical work throughout the day needs something different from someone who spends most of it behind a desk. Pregnancy, menopause, medication, digestive problems, disability and chronic disease can all change what is useful or safe.


I might encourage an older patient who has lost weight and muscle to add more protein to their meals. During another consultation, I may tell someone that the protein supplement they are considering is unnecessary because their usual diet already provides enough. Taken out of context, those recommendations sound contradictory. Once we put the people back into the story, they make sense.


This does not mean that everyone requires a completely unique set of health rules. The foundations of health are broadly shared. We all need nourishment, movement, recovery, connection and a reasonably safe environment. The details vary: how much, how often, in what form and under which circumstances. Research can show us what happens on average, while applying it to an individual requires us to consider how closely that person resembles the people who were studied. This question of generalisability is a recognised part of interpreting clinical research.


Unfortunately, much of that context disappears by the time a recommendation reaches a headline.


The question determines the answer

Conflicting advice also arises because people are often answering different questions.


Is running good for you? If we are asking about cardiovascular fitness, mood or long-term health, generally yes. If someone has developed a new injury, continuing their usual running programme this week may be unwise. A cardiologist, physiotherapist and orthopaedic surgeon may emphasise different aspects of running because they encounter different problems.


Coffee offers another example. Research may examine its relationship with cardiovascular disease, liver health, concentration, anxiety, pregnancy or sleep. A study suggesting that coffee consumption is associated with a lower risk of one disease does not mean that coffee improves every aspect of health. If it leaves you anxious or keeps you awake, that still matters.


Health is not a single outcome. Something may help in one area, make little difference in another and create a disadvantage elsewhere. Sunlight supports our circadian rhythm and allows the skin to produce vitamin D, while excessive ultraviolet exposure damages the skin. Whether the advice is “get more daylight” or “protect yourself from the sun” depends partly on the problem being discussed, the amount of exposure and the person receiving it.


To study health, researchers often have to narrow the question. They may examine whether a nutrient changes a blood marker, whether an exercise programme improves aerobic capacity or whether a treatment reduces one particular symptom.


That focus is necessary, but it can become misleading when the result is presented without its limits. A measurable improvement in one area does not automatically mean that the person’s health has improved overall. Researchers and regulators distinguish between changes in biomarkers and direct clinical benefits such as how someone feels, functions or survives precisely because one cannot always be assumed to predict the other.


Suppose a particular diet produces a slightly better glucose response but makes meals stressful, reduces dietary variety and cannot be sustained. Has it improved the person’s health? Perhaps according to the outcome being measured. The broader answer is less obvious.


A study showing a benefit also does not tell us how important that benefit should be. The effect may be real but small. It may apply mainly to people with a particular condition. It may improve a laboratory value without showing that people feel or function better. These distinctions rarely survive the journey from a scientific paper to a podcast title.


Most health decisions involve some combination of benefit, cost, dose, timing and context. Online discussions often isolate one part of that equation. One person looks through the lens of glucose, another through sleep, athletic performance, mental health or longevity. Each perspective may reveal something useful. Confusion begins when one perspective is presented as the whole picture.


Science changes because it is doing its job

Health advice also changes because science evolves.


This can undermine trust. If experts once recommended one thing and now recommend another, it is tempting to conclude that scientific guidance is arbitrary. Yet revising conclusions when better evidence becomes available is an essential part of science.


An early study may reveal an association. Later research can test whether that relationship is causal, examine it in different populations or uncover effects that were initially missed. Measurement methods improve, studies become larger and previously underrepresented groups are included. We may discover that the dose matters, that benefits level off or that an intervention helps one group more than another.


Individual responses can differ as well. Clinical research often reports an average effect, even though the balance of benefit and harm may vary among participants. Understanding this variation—known as heterogeneity of treatment effects—is important when translating research into decisions for individual people.


This process is rarely tidy. Individual studies sometimes contradict one another, and researchers may interpret the same evidence differently. Some questions remain genuinely uncertain for years. Replication, transparency and the accumulation of evidence allow findings to be tested, corrected and refined over time.


News and social media make the process appear more chaotic because they report studies one at a time. One week a food is associated with a benefit; the next week another study finds no effect. Science usually advances through the gradual accumulation of evidence, but we encounter it as a stream of isolated announcements.

A new study rarely overturns everything that came before it. More often, it adds a qualification or shifts our confidence slightly. This is why reputable guidelines are based on systematic assessments of the wider evidence and consider benefits, harms, feasibility, context and the priorities of the people affected.


We need a map

Most people already know that sleep matters, movement is beneficial and food influences health. Information is rarely the main problem. The difficulty lies in deciding what deserves our attention and how an isolated recommendation fits within the rest of our lives.


Two-panel illustration of an interconnected health map and a woman using it, showing how understanding the relationships between health foundations can support clearer, more sustainable choices.

Imagine arriving in a city you have never visited and asking three people for directions. One tells you to walk because it is only fifteen minutes away. Another says the metro is faster. A third recommends renting a bicycle.


All three may be right. Their advice depends on your destination, how much time you have, whether you can cycle, what you are carrying and whether it is pouring with rain.

Health works in much the same way. Individual recommendations are routes. To judge them, we need some idea of where we are, where we are trying to go and what the journey requires.


That is what I want to explore through The Health Shift. I am less interested in producing another list of habits than in helping you understand how the different parts of health connect. Why might a recommendation help? Does it apply to you? What else could it affect? Where does it fit among everything you are already trying to manage?


Understanding the map will not make every decision simple. Health is too personal and the evidence too incomplete for that. It can, however, make the contradictions less bewildering. Two recommendations may differ because they concern different people, circumstances, outcomes or time frames. A new finding may refine an older idea without making everything we previously knew obsolete. Something may benefit one aspect of health without being the right choice for the whole person.


Once we learn to ask those questions, we no longer have to reorganise our lives around every new headline. We can place it on the map, decide whether it is relevant and, quite often, continue along the road we were already taking.



Continue your Health Shift

I regularly explore practical ways to better understand and support your health—without rigid rules or health advice overload. Subscribe at the bottom of this page to receive new articles and practical shifts in your inbox.



Further reading

  • World Health Organization. Healthy living: What is a healthy lifestyle?

  • American College of Lifestyle Medicine. Lifestyle Medicine Core Competencies.

  • National Academies of Sciences, Engineering, and Medicine. Guiding Principles for the Care of Older Adults with Multimorbidity (illustrating interconnected health systems).

  • Ioannidis JPA. Why Most Published Research Findings Are False. PLoS Medicine. 2005.

  • GBD 2021 Risk Factors Collaborators. Global burden of 88 risk factors in 204 countries and territories, 1990–2021. The Lancet. 2024.

  • Rothwell PM. “External validity of randomised controlled trials: ‘To whom do the results of this trial apply?’” The Lancet. 2005.

  • Varadhan R, Segal JB, Boyd CM, Wu AW, Weiss CO. “A framework for the analysis of heterogeneity of treatment effect in patient-centered outcomes research.” Journal of Clinical Epidemiology. 2013.

  • FDA–NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource. National Center for Biotechnology Information.

  • National Academies of Sciences, Engineering, and Medicine. Reproducibility and Replicability in Science. 2019.

  • World Health Organization. Criteria for Use of Evidence to Inform Recommendations in World Health Organization Guidelines. 2023.

bottom of page