When Guidelines Collide, Doctors Must Choose: Prioritisation Is a Treatment in Its Own Right

Every day, in clinics and surgeries around the world, generalists make a decision that no guideline ever names: they choose which recommendation to follow first, which to postpone, and which to quietly abandon. A new Perspective published in the Journal of General Internal Medicine argues that this act of prioritisation should no longer be treated as an unavoidable compromise forced by short appointments and overstretched schedules. Instead, according to authors Waseem Jerjes and Azeem Majeed of Imperial College London, prioritisation is itself a clinical intervention, as consequential as any prescription, and it deserves the same scientific scrutiny, training, and recognition as the treatments it sequences.

The argument emerges from a growing mismatch between the architecture of medical evidence and the reality of modern patients. Clinical guidelines are overwhelmingly built for single diseases: one condition, one evidence base, one optimised pathway. But the patients walking into general practice and general internal medicine increasingly carry several long-term conditions at once, often bridging physical and mental health, cardiovascular risk, chronic pain, frailty, and social vulnerability. Large analyses of electronic health records show that multimorbidity clusters into recurring patterns that cut across these domains, and that the condition is associated with poorer function, worse quality of life, higher healthcare use, and earlier onset in deprived communities. For such a patient, the clinical problem is not only disease burden but healthcare burden: the appointments, tests, self-monitoring, medications, side effects, and contradictory advice that accumulate when each disease is managed by its own logic.

Consider a patient with type 2 diabetes, osteoarthritis, depression, hypertension, insomnia, and chronic kidney disease. The generalist seeing this person is not simply applying six evidence bases in sequence. The authors argue that the clinician is reconciling six partially incompatible logics of care, each with its own targets, monitoring schedules, and pharmacological demands. A diabetes guideline may push tight glycaemic control; the kidney disease may complicate the drugs used to achieve it; the depression may undermine the self-management every other condition depends on; the osteoarthritis may make exercise, the universal recommendation, painful. The question that actually matters in the consultation is not what each guideline advises, but what the next best thing to do for this person now happens to be.

That question becomes urgent when the mathematics of prevention is examined honestly. Modelling work cited in the Perspective shows that ignoring competing mortality can substantially overestimate the expected benefit of preventive interventions in older adults and people with multimorbidity. A statin, a blood pressure tablet, or a screening test may offer a genuine relative risk reduction, but if the competing risks of age, frailty, and other diseases intervene first, the absolute benefit shrinks, while the burden of taking the treatment remains fully present. The authors also highlight direct treatment disutility, the harm and inconvenience a treatment imposes in its own right, which can flip the balance between benefit and harm. Evidence, in other words, does not speak for itself; it must be adjudicated against life expectancy, burden, and the patient’s own goals.

And patients, it turns out, often adjudicate differently from their doctors. A systematic mixed-studies review found that patients with multimorbidity commonly prioritise symptom relief and the preservation of function and independence, whereas clinicians tend to prioritise longer-term risk reduction and surrogate markers. This divergence is not a failure of communication to be corrected but a structural feature of complex care that must be made visible. The Perspective is careful to distinguish prioritisation from its neighbouring concepts: it is not shared decision-making, though shared decision-making safeguards it; not deprescribing, though deprescribing is one of its tools; not treatment burden measurement, though burden is one of its inputs; and not Choosing Wisely, though it extends that movement’s logic. Its distinctive work is deciding what should come first now, what can wait, and what should stop.

Crucially, the authors insist that generalists already perform this work, but implicitly and without a language that grants it legitimacy. Qualitative studies of physicians caring for complex patients show that their decisions are shaped by disease interactions, patient resources, consultation constraints, organisational pressures, continuity of relationship, and the effort to steer care toward a workable goal. Physicians report that patient values matter, yet are not always elicited directly and are difficult to reconcile with protocols, multidisciplinary teams, and families. Even committed clinicians describe limited time and systems that reward the completion of checklists far more readily than the exercise of judgment. The result is a paradox the authors name directly: prioritisation is central to good generalist care, yet it remains under-described, under-taught, and under-valued.

The Perspective does not leave the concept abstract. It points to a growing body of implementation research that moves goal-oriented care from theory toward operational reality. Grudniewicz and colleagues have proposed strengthening the Chronic Care Model by subordinating disease management to what matters most to the person over time, rather than abandoning disease management altogether. Tinetti and colleagues tested a patient priorities-aligned care model for older adults with multiple conditions and reported potentially favourable differences in treatment burden and shared prescribing decisions compared with usual care, although the differences did not reach statistical significance. Schuttner and colleagues used multicriteria decision analysis to demonstrate something clinically profound: the right next step changes depending on whether the patient’s aim is mortality reduction, symptom relief, or preserved function. Prioritisation, in other words, is not a fixed ranking but a dynamic computation that depends on the goal.

Making prioritisation explicit, the authors argue, requires deliberate elicitation of what matters now. They describe a simple clinical sequence: identify what matters most to the patient at this moment, select the next action with the greatest likely benefit and lowest burden relative to that goal, and actively stop, simplify, or defer care that does not serve it. Supporting tools are emerging. Person-centred care approaches built around extended consultations, personalised goal setting, and planned follow-up have been developed in general practice, and one research group has even produced a prioritisation algorithm to identify which patients are most likely to benefit from a longer, goal-setting appointment in time-constrained settings. Communication and goal-setting tools such as structured wellbeing journals have shown promise in supporting more meaningful care-planning conversations with older adults. The lesson, the authors stress, is not that every consultation must become long or formalised, but that a good consultation may sometimes be one in which fewer targets are addressed and the patient leaves with a clearer plan, lower burden, and fewer contradictory demands.

The hardest test of prioritisation is whether it leads to de-intensification. Adding treatment is easy to justify, because each addition can be defended by a disease-specific rationale; stopping requires accepting responsibility for uncertainty. Yet the authors argue that prioritisation which never results in stopping is performative rather than real. The evidence here is genuinely mixed, and they are candid about it. Deprescribing interventions generally reduce medication counts, and a meta-analysis found benefits on potentially inappropriate prescribing and adverse drug reactions in older adults, but evidence for broader outcomes remains inconsistent, particularly in vulnerable multimorbid populations. Structured medication reviews, meanwhile, are hampered by time pressure, organisational complexity, unclear roles, and poor data integration, according to qualitative findings from the DynAIRx study. This is the landscape in which generalists operate: enough evidence to justify action, never enough to eliminate judgment.

The Perspective closes with a warning about how the SPPiRE analysis complicates even well-intentioned priority elicitation: when general practitioners prioritised pain, opioid intensification became more likely despite guideline concerns about chronic pain pharmacotherapy in older adults. Eliciting priorities, the authors conclude, is necessary but not sufficient; priorities must still be interpreted through a generalist lens weighing immediate suffering against long-term harm. The wider implications reach into education, where trainees should learn to justify not only what they do but what they decline to do; into quality measurement, which should reward goal-concordant sequencing and burden reduction rather than indiscriminate completion; and into system design, where altered specialist treatments should be documented and reconciled so that patients are not left as messengers between competing experts. In complexity, the authors write, the central clinical task is deciding what deserves attention first, what can wait, and what no longer serves the patient. Prioritisation, they argue, is the treatment decision.

Subject of Research: Prioritisation as a clinical intervention in generalist care for patients with multimorbidity

Article Title: Prioritisation Is Treatment: What Generalists Do When Guidelines Collide

Article References: Jerjes, W., & Majeed, A. (2026). Prioritisation Is Treatment: What Generalists Do When Guidelines Collide. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10861-1

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10861-1

Keywords: generalism, multimorbidity, prioritisation, clinical guidelines, polypharmacy, deprescribing, treatment burden, shared decision-making, goal-oriented care, primary care, Choosing Wisely, patient priorities