Twenty years ago, writing from India, one of us described distrust as a ‘third burden’ confronting medicine. Alongside infectious and cardiovascular disease, a quieter disorder was spreading: erosion of trust among physicians, patients, institutions, and society. The concern was not merely reputational.
Trust was part of medicine’s infrastructure; without it, even technically excellent care could fail.1
Two decades later, writing from Italy and within a very different healthcare environment, the other approached a related concern through two complementary perspectives: whether technological capability is beginning to outpace clinical judgement, and whether evidence intended to guide medicine can drift from discovery towards persuasion.23 These observations emerged in different countries, at different times and within health systems that differ in resources, financing, organisation and access. Yet they converge on the same concern.
That convergence now appears wider than the two countries from which we write. Work from China has described a vicious circle of patient–physician ‘mistrust’, in which fear, defensive behaviour, communication failure, poorer care and conflict reinforce one another, while identifying institutional conflicts of interest and the erosion of medical professionalism as important drivers.4 The manifestations differ, but the pattern is recognisable. Across wealthy and resource-constrained systems, public and private settings and very different cultures, medicine increasingly operates under pressures that can blur the boundary between what serves patients and what serves institutions, markets, professional incentives, technology or visibility.
Distrust may therefore have local expressions, but the forces that make medicine less trustworthy are increasingly global. The challenge is not simply how to persuade patients to trust medicine again. It is how to make medicine more deserving of trust.
Trust is an outcome; trustworthiness is a property
Trust cannot be demanded, manufactured by reassurance or repaired by communication alone. It is conferred. Patients trust physicians because they assume recommendations are guided primarily by their welfare.
Physicians trust evidence because they assume inquiry was designed to discover rather than merely confirm. Society trusts institutions because it assumes professional judgement remains sufficiently independent from the forces surrounding it.
When these assumptions weaken, distrust should not automatically be interpreted as ignorance, hostility towards science or a failure of communication. Sometimes distrust is a signal. The lesson is broader than either setting: trust is fragile when the incentives surrounding medicine become difficult to distinguish from the purposes of medicine.
Today, vulnerability extends across the whole chain through which knowledge becomes care: research priorities, evidence generation, publication, guidelines, clinical decisions, technology adoption, institutional incentives, communication and outcomes. Trust can weaken at every step. Medicine cannot determine whether society will trust it.
It can determine whether its evidence, institutions, technologies and professionals behave in ways that merit trust.
From local symptoms to a global disorder
India, Italy and China are not interchangeable examples. Their demography, access to care, financing, institutional structures and social expectations differ substantially. Yet this is precisely why their convergence matters.
Similar ethical tensions can arise through different mechanisms.
The geography of distrust is therefore less important than its architecture. Different systems can produce different symptoms of the same disease: a weakening alignment between professional purpose and patient welfare.
Evidence, technology and choice in cardiology
Modern medicine has become extraordinarily effective partly because of randomised trials, sophisticated imaging, devices, transcatheter interventions, remote monitoring and artificial intelligence. These achievements should be defended. But scientific and technological progress also expands the number of choices that must be interpreted, prioritised and sometimes declined.
Evidence can lose trustworthiness and relevance without becoming false. Choices of populations, comparators, endpoints, follow-up, analyses and language shape the reality that clinicians subsequently see.3 Relative effects can appear transformative when absolute benefit is modest; positive findings can receive greater visibility than negative ones and questions attractive to funders may crowd out questions more important to patients. For example, in a hypothetical trial, a 25% relative risk reduction could represent a fall in event risk from 4% to 3% over the reported follow-up.
Presenting both figures helps patients understand the magnitude of benefit and weigh it against treatment burden and potential harms. These concerns echo broader critiques of how evidence-based medicine can be distorted when methodological rigour is embedded within misaligned incentives.5 The problem is not industry itself, nor the existence of incentives, but the possibility that incentives begin to organise what is asked, emphasised, published, adopted and ultimately offered.
Technology creates a parallel challenge. The question can shift from ‘Why should we do this?’ to ‘Can we do this?’2 Anatomical feasibility can be mistaken for clinical appropriateness, and procedural success for patient benefit. Artificial intelligence further lengthens the chain of trust—from patient to physician to algorithm to data to the institution or company producing the system.
The convergence of human and artificial intelligence can strengthen medicine, but only if human accountability and judgement remain explicit.6 Technology may inform a decision, but it cannot assume moral responsibility for making it. The more medicine can do, the more important clinical judgement becomes.
Cardiology makes these tensions especially visible. New drugs, imaging technologies, implantable devices, transcatheter procedures and digital tools often move rapidly from clinical trials into guidelines and routine care, sometimes before their value is clear across the full range of patients encountered in practice. When relative treatment effects, surrogate endpoints or procedural success are allowed to stand in for outcomes that patients can feel or value, the distance between innovation and meaningful benefit can become a source of distrust.
From trustworthy medicine to the trustworthy physician
If distrust is the disease, reassurance cannot be the treatment. The ultimate safeguard is the trustworthy physician: a professional capable of integrating evidence, judgement, integrity, restraint and responsibility within systems that support rather than undermine those qualities. Four domains remain essential, but they should be understood as expressions of a broader professional commitment rather than as separate technical checklists.
Trustworthy evidence requires intellectual honesty: prespecified analyses distinguished from exploration, absolute effects presented alongside relative effects when clinically meaningful, negative studies treated as information rather than failure and uncertainty communicated rather than hidden. Research should ask not only who benefits, but who does not, when benefit becomes too small and when intervention should stop.
Trustworthy decisions require judgement. Every consequential intervention should survive several questions: Can we do it? Does evidence support it?
Will this patient meaningfully benefit? Does that benefit matter to this patient? What are the alternatives, including doing less?
And why are we doing it? Frailty, cognition, competing risks, functional trajectory, affordability, burden of treatment and patient preferences belong inside appropriateness, not at its margins. This is consistent with contemporary patient-centred cardiovascular care, which places patients’ goals, values and meaningful outcomes at the centre of clinical decisions.7 For a patient with severe aortic stenosis and advanced frailty, for example, the discussion should connect the feasibility of valve intervention with the likelihood of achieving the patient’s own goals, such as walking independently or remaining at home.
Trustworthy institutions must make it easier to do the right thing. Disclosure of conflicts is necessary but insufficient if incentives themselves remain misaligned. Institutions should examine how payment, procedural volume, research funding, promotion, performance metrics and technology adoption influence professional behaviour.
Guideline processes should communicate uncertainty and preserve independence. Academic systems should value replication, negative findings, de-implementation and independent investigation alongside novelty.
Trustworthy technology requires proportionality and accountability. Approval establishes that a technology may be used; evidence may show that it works. Neither alone establishes when it adds meaningful value to a particular patient.
Scientific inquiry should accompany diffusion rather than end when adoption begins. Validation across relevant populations, surveillance for bias and drift, transparency about limitations and identifiable human accountability are essential. For example, when an artificial intelligence alert conflicts with the clinical assessment, a trainee should examine the underlying data and the tool’s limitations, seek senior review when needed and explain the reasoning behind the final decision.
These four domains ultimately depend on the trustworthy physician and on professional character. Medicine is not simply the delivery of technical services. In a JACC Editor’s Page nearly three decades ago, Parmley invoked a definition of professionalism in which integrity gives professional service its value and compensation must never be confused with the meaning of the work itself.8 That language may sound idealistic in a healthcare environment dominated by metrics, productivity and market pressures, but its core claim should remain non-negotiable.
Patients ultimately encounter not ‘evidence’, ‘institutions’, or ‘technology’ in the abstract; they encounter a physician who must decide what those forces mean for this person, now.
Professionalism therefore requires more than being well intentioned. It requires the competence to know what can be done, the intellectual independence to question what is promoted, the courage to say no when an intervention adds little value, the humility to communicate uncertainty, the honesty to acknowledge conflicts and the fidelity to keep patient welfare primary when other incentives compete for attention. Restraint is not therapeutic nihilism.
Sometimes it is the most demanding expression of expertise.
A new social contract for global medicine
Shared decision-making is essential but incomplete if the information being shared is already shaped by selective framing or unacknowledged incentives. What is needed is shared epistemic and professional responsibility: researchers should describe uncertainty honestly; sponsors should preserve scientific independence; journals should resist exaggeration; guideline committees should distinguish strong evidence from inference; institutions should make incentives visible; physicians should translate population evidence into individual meaning and patients should receive an intelligible account of benefits, harms, alternatives, uncertainty and the option of not intervening.
Not every adverse outcome is malpractice, not every commercial relationship implies corruption and not every innovation is unnecessary. Society must tolerate uncertainty and recognise that medicine cannot abolish risk. A distrustful society also requires corrective measures of its own, which we will address elsewhere.
At the same time, the asymmetry of knowledge gives medicine the greater obligation. Professional authority is justified because society expects knowledge to be exercised with competence, independence, honesty, judgement and restraint.
Twenty years ago, the warning came from India: distrust might become medicine’s third burden. Similar concerns have since been described in China, while in Europe and elsewhere they are visible in debates about evidence, commercialisation, technology, overtreatment and professional autonomy.1–4 The details remain local, but the diagnosis is increasingly global.
Medicine does not need to become identical across countries to remain trustworthy. It needs to remain recognisable across countries in its moral purpose. The task is not to eliminate scepticism, but to reduce the conditions that make scepticism reasonable.
Evidence should seek truth rather than persuasion; technology should serve judgement rather than replace it; institutions should align incentives with patient welfare and physicians should remain capable of choosing what is meaningful over what is merely possible. In cardiology, this requires every new drug, device, image, procedure and algorithm to be judged by whether it improves outcomes that matter to the individual patient, not simply by technical availability or its effect on a measurable endpoint.
The vaccine against distrust is not trust. It is trustworthiness. And the final bearer of that trustworthiness, at the bedside, remains the physician.