This is the second in a series examining how modern medicine’s instinct to intervene can, in certain circumstances, work against the very people it intends to help. The first piece looked at overdiagnosis - the transformation of normal human variation into pathology. This one follows the thread upstream, to ask a harder question: how do the rules that drive that transformation get written, and what happens to patients when those rules produce needless treatment?
There is a moment in any complex institution when the mechanism designed to serve a purpose quietly begins to serve itself instead. In medicine, that moment arrived gradually, and most people - patients and doctors alike - did not notice it happening.
Clinical guidelines are the architecture of modern medical practice. They determine who gets diagnosed, at what threshold, with what treatment, for how long. They are presented to patients and practitioners as the distilled wisdom of the best available evidence.
What they have become is something more complicated.
The Committee Room Problem
When a major medical body convenes a guideline committee to determine, say, the threshold at which cholesterol should be treated, or the blood pressure number that triggers medication, it faces a practical problem: the people with the deepest expertise in a given field are frequently also the people with the deepest financial relationships with the companies that manufacture treatments for that field.
This is not incidental. It is structural. Pharmaceutical companies have, for decades, invested heavily in cultivating relationships with the academic physicians most likely to sit on influential committees, through research funding, consultancy fees, speaking engagements, conference sponsorship, and the funding of the clinical trials that generate the evidence the guidelines are based upon. The result is a system in which the people writing the rules are often, to varying degrees, financially entangled with the industries that benefit from those rules being written in a particular way.
Studies have found that a significant majority of members on major cardiovascular and diabetes guideline panels have disclosed financial ties to pharmaceutical manufacturers. In some analyses, the proportion exceeds seventy percent.
These are not undisclosed relationships, conflict-of-interest declarations are now standard practice. But disclosure is not the same as independence, and the declaration of a conflict does not dissolve it.
A physician who has spent years collaborating with a drug company, whose research has been funded by it, who speaks at its events and consults for its strategy teams, will not approach the evidence in the same frame of mind as someone with no such relationship. That is simply human nature, and no amount of transparency paperwork changes it.
The consequences are predictable. Guidelines lower thresholds. Treatment eligibility expands. New categories of sub-clinical disease are invented - “pre-hypertension,” “pre-diabetes,” “borderline” this and “elevated” that, each one a step on the staircase from wellness to patient-hood, and each one a market opportunity.
When the American College of Cardiology and the American Heart Association revised the definition of hypertension in 2017, the new threshold brought an estimated 31 million additional Americans into the category of people requiring medical management. That is not a scientific finding. That is a policy decision with enormous commercial implications, made by a committee a significant portion of whose members had financial ties to companies selling antihypertensive drugs.
Evidence Built on Sponsored Foundations
The problem does not begin in the committee room. It begins earlier, in the design, conduct, and reporting of clinical trials themselves.
The majority of large-scale drug trials are funded by the manufacturers of the drugs being tested. This creates a well-documented set of distortions. Industry-sponsored trials are more likely to produce positive results than independently funded ones examining the same interventions. They are more likely to use comparators that flatter the new drug, comparing it to placebo rather than to an existing treatment, or using a suboptimal dose of the comparator.
They are more likely to report relative risk reductions (which sound impressive) than absolute risk reductions (which often do not). Plus, they are more likely, when results are unflattering, to remain unpublished, a phenomenon known as publication bias, which skews the entire published evidence base toward optimism.
A trial that shows a statin reduces the relative risk of a heart attack by 25% sounds transformative. But if the baseline risk in the study population was 4%, a 25% relative reduction means the absolute risk fell from 4% to 3%, that’s one person in a hundred benefiting over several years of daily medication.
The relative figure makes the front page. The absolute figure appears in a footnote, if at all. Guidelines built on selectively reported, industry-funded evidence, interpreted by committees with industry ties, are not a neutral scientific output. They are, at least in part, a commercial one.
When Treatment Becomes the Illness: The Rise of Iatrogenesis
This matters not as an abstract problem of research integrity, but because the guidelines produced by this system drive real interventions in real bodies and those interventions carry real risks.
Iatrogenesis - harm caused by medical treatment - is one of medicine’s oldest and most uncomfortable subjects. The word derives from the Greek iatros, meaning physician, and genesis, meaning origin. Physician-caused harm. It was a concept familiar to Hippocrates, whose instruction to first, do no harm was not a platitude but a serious warning: the act of intervention is never neutral, and its risks must always be weighed honestly against its benefits.
In an era of overdiagnosis and expanded treatment thresholds, iatrogenesis does not arrive dramatically. It accumulates quietly, in the aggregate. It is the muscle damage and glucose dysregulation in patients taking statins for a risk that would never have harmed them. It is the surgical complications - incontinence, impotence - in men treated for prostate cancers that would never have progressed.
It is the osteoporosis drug that causes the rare but devastating atypical femoral fracture. It is the blood pressure medication that makes an elderly person dizzy, and the fall that follows, and the hip fracture and the long slow decline that a hip fracture in old age so often initiates.
None of these patients was harmed by negligence. They were harmed by guidelines, by the expansion of treatment criteria into populations where the probability of benefit was low and the probability of harm was high. They were treated in good faith, by doctors following the rules, on the basis of evidence that told an incomplete story, produced and curated by a system with financial incentives pointing firmly in one direction.
This is the downstream cost of guideline capture. It is not paid by the committee members. It is paid by patients.
Share HWSS Newsletter & Healthcare News
The Cascade Effect
There is a further dimension that rarely features in guideline deliberations: the diagnostic cascade. An investigation leads to a finding. The finding prompts further investigation. Further investigation produces more findings, some of them incidental, some ambiguous, all of them demanding a response. Each response carries its own risks. The cascade is self-propagating, and at each step the patient moves further from the ordinary life they were living before someone decided to look.
A middle-aged man attends a routine health check. His cholesterol is borderline elevated; he is started on a statin. A liver function test is ordered as a precaution. A mildly abnormal result triggers an ultrasound. The ultrasound reveals a small lesion on the liver, almost certainly benign, but impossible to ignore. A follow-up MRI is arranged. Months pass. More scans.
An eventual biopsy. The lesion is benign, as it almost certainly always was but by now this man has spent a year living in a medical shadow, has had a needle inserted into his liver, has missed work, has lain awake at three in the morning thinking about his children. His original cholesterol reading, the thing that started all of this, would most likely never have killed him.
This is not a hypothetical. It is a well-documented phenomenon, studied and named. And it begins, almost invariably, with the same originating step: a threshold crossed, a number noticed, an intervention initiated in a person who, left alone, would have been fine.
The Trust That Gets Spent
There is a slower, less visible harm at work here too. Every patient who takes a medication for years and suffers side effects they are told are unrelated to it - only to later discover the connection is well established - loses a measure of trust in the system that prescribed it. Every person who undergoes treatment for a cancer that would never have harmed them, and lives with the consequences of that treatment, has reason to feel that the bargain they were offered was not honestly described.
Every individual told that the guidelines recommend intervention, without being told who wrote those guidelines and what interests shaped them, has been given an incomplete picture of the decision they are being asked to make.
Medicine’s authority rests on public trust. That trust is not a fixed quantity. It is spent, or earned, in thousands of individual consultations, and in the honesty of the institutions that set the terms on which those consultations proceed. A system that allows commercial interests to quietly colonise its evidence base and its rule-making apparatus is not only producing worse outcomes for patients - it is slowly eroding the legitimacy on which the whole enterprise depends.
A Different Kind of Courage
Reforming this will require something that institutions find genuinely difficult: the courage to acknowledge that the rules have been shaped by interests other than patient welfare, and to build structures that meaningfully separate the two.
That means guideline committees composed predominantly of clinicians and scientists with no financial relationships to the relevant industry.
It means mandatory independent replication of pivotal trials before new treatment thresholds are adopted.
It means presenting evidence to patients in absolute rather than relative terms, as a matter of ethical obligation rather than optional good practice.
And it means a medical culture willing to take iatrogenesis seriously - to count, clearly and honestly, the harm done by treatment, not just the harm it prevents.
None of this requires cynicism about medicine, or about the many physicians who practise it with integrity and genuine commitment to their patients. It requires only the recognition that good intentions are not sufficient protection against a system whose incentive structures point the wrong way.
The doctor in the consulting room is, in most cases, simply following the rules. The question is who wrote them, and why.
Until that question is answered honestly, the rules will continue to serve more than one master. And the patients sitting across the desk will continue to pay the price.
The first piece in this series — on overdiagnosis and the medicalisation of normal human variation — is available here. A further piece, examining how relative risk statistics are used to make modest treatment effects sound transformative, will follow.
If this resonated, share it with someone who should read it.
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.