Why Single-Disease Guidelines Struggle With Multimorbidity

A clinical guideline usually asks how to diagnose and manage one condition, while a person may live with diabetes, heart failure, chronic kidney disease, arthritis, depression, and frailty at the same time. Applying each guideline separately can create a combined plan that no study tested and no one intentionally designed.

The problem is not that single-disease guidelines are careless. They synthesize evidence for a focused question. Their trials often include people with fewer conditions and medicines than those seen in routine care. Recommendations can interact when several conditions coexist: one medicine worsens another symptom, one dietary instruction conflicts with another, and several modest tasks become an unmanageable daily workload.

NICE guideline NG56, published in 2016 and still listed as current at the research date, addresses multimorbidity as a care problem in its own right. It asks clinicians to reduce treatment burden, adverse events, and fragmented care while aligning decisions with what matters to the person.

What counts as multimorbidity#

NICE defines multimorbidity as two or more long-term health conditions. The definition is broader than a count of organ diagnoses, and it can include a learning disability, frailty, chronic pain, sight or hearing impairment, and alcohol or substance misuse alongside conditions such as diabetes or schizophrenia.

Two diagnoses can create little additional complexity, while another combination can be difficult to manage. The number alone does not measure burden. Severity, interaction, and social context determine whether a multimorbidity approach is useful. So do cognitive function, care fragmentation, and the person's goals.

Multimorbidity occurs at younger ages as well as older ages and is more common in communities facing socioeconomic disadvantage. Mental and physical conditions often coexist. Treating it as synonymous with old age misses many people and can delay coordinated care.

How trial eligibility creates an evidence gap#

Randomized trials use eligibility criteria to protect participants, reduce confounding, and create a population in which an effect can be measured; they may exclude severe kidney or liver disease, frailty, cognitive impairment, interacting medicines, another active illness, or limited life expectancy.

Those choices can be reasonable for the trial. They also mean that the estimated benefit and adverse-event rate may not transfer directly to the person on the page in front of you, who has several of the excluded features. Pharmacokinetics, competing outcomes, adherence, and baseline risk can differ.

Guideline panels grade evidence and often discuss subgroups, but they cannot generate missing trials, so a recommendation can be internally valid for the studied population and uncertain in a person with multimorbidity. The response should be calibrated judgment, not dismissal of all guidance.

Stacking creates more than a long medicine list#

Imagine you are holding separate plans for heart failure, diabetes, osteoporosis, chronic lung disease, and depression. Each may recommend medicines, monitoring, and exercise. Each may recommend diet, education, appointments, and escalation criteria. Together they can require many daily doses, conflicting fluid or nutrition advice, and repeated blood tests. They can require several specialty visits and complex sick-day rules.

Medicine interactions are only one layer. A diuretic can worsen urgency and falls. A pain medicine can affect kidney function or blood pressure. Sedation can worsen function. Treating one numerical target aggressively can increase dizziness or hypoglycemia.

There are also disease-disease interactions. Arthritis may limit recommended activity. Cognitive impairment may make a complex insulin plan unsafe. Depression can reduce capacity for self-care. A single-disease document cannot rank all of these relationships for your patient.

Treatment burden is a clinical outcome#

Treatment burden is the work the health system hands to the patient and the family. It includes getting and organizing the medicines, doing the monitoring, and getting to the appointments. It also includes the parts nobody counts: the transport, the insurance calls, the exercises and the diets that contradict each other, the devices to learn, and telling the same history to every new service.

Capacity is the time, energy, and cognition available to do that work. It is also the money, support, and physical ability. When workload exceeds capacity, tasks go undone. Calling the result nonadherence without examining the mismatch hides the design problem.

A plan can improve a disease-specific marker while reducing overall quality of life. Conversely, simplifying one low-value task can free capacity for a treatment with larger benefit. Reducing burden is therefore not merely convenience; it can improve the reliability of the whole plan.

Why preventive treatment needs a time horizon#

Some treatments relieve symptoms within hours or days. Others reduce the probability of a future event over years. Time to benefit asks how long treatment usually takes to produce the outcome that matters. Competing risk asks whether another illness may make that outcome less likely to occur first.

For a robust person with a long horizon, a small annual reduction can accumulate into meaningful benefit. For someone with advanced frailty, severe competing illness, or major treatment harm, the same delayed benefit may be less likely to arrive. This is a probability judgment, not a claim that age or disability makes prevention worthless. Estimates are often uncertain, and life expectancy should not be treated as a precise countdown. The useful conversation compares expected benefit, current burden, harm, and the person's priorities.

Medicine-count thresholds are prompts, not verdicts#

NICE recommends a multimorbidity approach for adults of any age prescribed 15 or more regular medicines because adverse-event and interaction risk is likely higher. It says to consider the approach for 10 to 14 regular medicines, and for fewer when particular risk exists.

The threshold does not mean the fifteenth medicine is wrong or that a person taking nine needs no review. Some people need many evidence-based medicines and tolerate them well. Others experience substantial burden with fewer.

Counting should include more than tablets. Inhalers, eye drops, injections, creams, as-needed medicines, and non-drug treatments can all add work, and the review asks whether each item has a current indication, expected benefit, acceptable harm, feasible schedule, and monitoring plan.

Frailty changes interpretation without erasing the person#

Frailty describes reduced reserve and vulnerability to stressors. It is associated with falls, hospitalization, adverse drug events, and recovery difficulty. Acute illness can make a person appear more frail, so NICE cautions against relying on physical-performance tools during acute illness.

Frailty can alter blood-pressure targets, glucose safety, operative decisions, and tolerance of medicines. It also makes function, independence, and caregiver support especially important outcomes. A frailty score should start a conversation, not replace it. People with similar scores can have different values, supports, symptoms, and treatment goals.

What a multimorbidity review asks#

NICE begins with what matters to the person. Priorities may include maintaining independence, staying in work, participating in family or social life, preventing a particular event such as stroke, reducing medicine harm, lowering treatment burden, or lengthening life.

The review then identifies how conditions and treatments affect one another, which services are involved, and which tasks are difficult. It examines symptomatic treatments for actual benefit and harm. It considers preventive treatments in light of time horizon and preferences. It also looks for useful treatments that may be missing, not only medicines to stop; the result is an individualized management plan that records priorities, agreed changes, monitoring, responsibility, and coordination. Without ownership, a list of recommendations can leave each specialist assuming someone else will reconcile the whole.

Deprescribing is an intervention with risks#

Deprescribing means reducing or stopping a medicine when its likely harm or burden exceeds benefit for that person. It includes verifying the indication, dose, duration, and current response; considering withdrawal or rebound; agreeing on the change; and monitoring afterward.

Some medicines can be stopped directly in selected situations. Others require tapering. Stopping an anticoagulant, insulin, or steroid without a plan can cause serious harm. So can stopping an antiseizure medicine, a psychiatric medicine, or cardiovascular therapy. Even a low-value medicine may need staged withdrawal.

NICE recommends discussing reduction or stopping when a symptom treatment provides uncertain benefit or causes harm, followed by review and possible restart. Reversibility is useful: a monitored trial can tell you whether simplification helps this person, without committing you to a decision that can never be revisited.

Conflicting guidelines require prioritization#

When recommendations conflict, clinicians can identify the outcome each is trying to prevent, estimate absolute benefit and harm, and consider interaction. A blood-pressure target that reduces long-term stroke risk may need adjustment if it causes recurrent falls. A strict diet for one condition may be relaxed if it worsens malnutrition.

Shared decision-making is not asking the patient to solve a technical conflict alone. Clinicians should explain options and uncertainty, make a recommendation, and invite the person to weigh outcomes. You cannot infer values from age, diagnosis, or family preference. When evidence is thin, write down your reasoning. It is what lets the plan be revisited as function, disease, evidence, or goals change.

Coordination is part of treatment#

Fragmented care can generate duplicate tests, incompatible instructions, and prescription cascades. One clinician or team needs responsibility for maintaining the integrated plan, even while specialists manage particular conditions.

Useful coordination includes a current medicine list, named indications, and a clear monitoring schedule. It includes who responds to abnormal results and what changes during illness. Pharmacists, nurses, and therapists may all contribute. So may social workers, caregivers, and specialists. Caregivers can provide essential information and support, but the person's consent, capacity, privacy, and preferences remain central. When capacity is impaired, applicable legal and ethical processes guide decisions.

A practical review can start small#

A complete overhaul is not always feasible or desirable. Start with the issue causing the most harm or burden: dizziness, hypoglycemia, pain, conflicting instructions, missed refills, or too many appointments. Identify one change, define what success and harm would look like, and set follow-up.

Then reconcile the rest of the plan over time. Ask which treatment has the largest expected benefit, which is not working, which causes harm, and which task the person cannot realistically complete. Review after hospitalization or major diagnosis because priorities and medicines often change.

The aim is not the shortest list. It is the most coherent plan the person can use.