A high-risk patient is someone whose combination of medical conditions, medications, social circumstances, or functional limitations puts them at significantly greater odds of hospitalization, complications, or death compared with the average person seeking care. The term has no single universal definition because it shifts depending on context: a surgeon evaluating an elderly person before a heart operation, a health plan trying to predict who will need expensive care next year, and a primary care clinic triaging follow-up calls after a hospital discharge may each define “high risk” differently. What unites these uses is the idea that certain patients carry a measurably heavier burden of need, and that identifying them early enough can sometimes change their trajectory.
Why the Definition Depends on Who Is Asking
Health systems, insurers, and clinicians all flag patients as high risk, but they do it for different reasons and using different yardsticks. A hospital readmission algorithm, for instance, might focus on a narrow list of conditions where timely outpatient care has the best chance of preventing a return trip: heart failure, chronic obstructive pulmonary disease (COPD), diabetes, and sickle cell disease are common targets. The logic is practical: some hospitalizations, like major trauma or surgery for cancer, are driven by events that case management cannot easily prevent in the short term. Algorithms therefore concentrate on conditions where better ambulatory care could realistically make a difference.1BMJ. Case finding for patients at risk of readmission to hospital: development of algorithm to identify high risk patients
An insurer or Medicaid program, on the other hand, may define high risk almost entirely in terms of cost. One large university health system study sorted more than 9,500 Medicare and Medicaid patients into four risk tiers and found that people in the higher tiers had both more healthcare episodes and higher charges in the following year.2PubMed Central. Identifying Patients at Risk of High Healthcare Utilization A clinician, meanwhile, might think of risk in terms of frailty, medication safety, or the likelihood that a patient will deteriorate between visits. These perspectives overlap considerably, but they are not identical, and the label “high risk” can follow a patient through the system even when the original reason no longer applies.
Multiple Chronic Conditions Are the Single Biggest Driver
If one factor towers above the rest in determining who becomes a high-risk patient, it is the number of chronic conditions a person carries at the same time. The medical term is multimorbidity, and its impact on healthcare costs and resource use is not simply additive; it climbs in a near-exponential curve. Each additional chronic disease brings more doctor visits, more medications, more emergency department trips, and more hospital admissions.3PubMed Central. Multimorbidity in chronic disease: impact on health care resources and costs
Among older adults, multimorbidity is the norm rather than the exception. A German study of elderly patients found that about 62% were multimorbid, with women and people receiving nursing care for disability disproportionately represented. The combinations involved are staggeringly diverse: out of more than 15,000 possible three-disease combinations, nearly all of them showed up in the data. Yet a handful of conditions dominated. Triads built from just six diseases — hypertension, lipid disorders, chronic low back pain, diabetes, osteoarthritis, and chronic ischemic heart disease — covered the disease picture for about 42% of the multimorbid group.4PubMed Central. Which chronic diseases and disease combinations are specific to multimorbidity in the elderly? Results of a claims data based cross-sectional study in Germany
The trajectory matters, too. A large study following middle-aged and older patients at community health centers found that while roughly half started without multimorbidity, about a third of those eventually shifted into it over the study period. The most common pathways led toward cardiometabolic clusters or combined mental and physical conditions.5JAMA Network Open. Trajectories of Chronic Disease and Multimorbidity Among Middle-aged and Older Patients at Community Health Centers This suggests that risk is not static; people migrate into higher-risk categories over time, which creates a window for earlier intervention.
Frailty as a Risk Multiplier
Chronic disease counts alone do not capture the full picture. Two people can have the same list of diagnoses, yet one walks independently and cooks dinner while the other struggles to get out of a chair. Frailty — a state of diminished physiological reserve that leaves a person vulnerable to sudden decline from even minor stressors — is an independent predictor of bad outcomes, and researchers have built electronic tools to measure it from health records.
A study applying an electronic frailty index to older adults found that very frail individuals had roughly four times the mortality risk, more than five times the rate of acute care visits, and about double the odds of being readmitted within 90 days, all compared with robust older adults.6PubMed Central. Application of an Electronic Frailty Index to Identify High-Risk Older Adults Using Electronic Health Record Data In the context of cardiac surgery, frailty assessment has shown similar predictive value for 30-day mortality, adding useful information beyond the standard surgical risk scores that focus mainly on cardiac anatomy and procedure complexity.7European Journal of Cardio-Thoracic Surgery. Comprehensive assessment of frailty for elderly high-risk patients undergoing cardiac surgery The takeaway for patients and families: if your doctor mentions frailty, it is not just a polite way of saying “old.” It is a measurable condition that meaningfully changes the risk calculus for procedures, medications, and care plans.
When the Medications Themselves Become the Problem
High-risk patients tend to take a lot of drugs, and the drugs themselves become a source of danger. Polypharmacy — generally defined as using five or more medications at once — is nearly universal among people with multiple chronic conditions. The risk of harmful drug interactions and adverse drug events rises steeply with each additional medication. A Spanish study examining multimorbidity and adverse drug events found that the odds of an adverse event in patients with the highest level of multimorbidity were more than 45 times those in patients with the lowest level, with polypharmacy and the involvement of multiple specialists contributing independently.8PubMed Central. Multimorbidity, polypharmacy, referrals, and adverse drug events: are we doing things well?
The specific combination of drugs matters enormously. A study of primary care patients aged 65 and older found that certain medication pairings carried dramatically higher risks of emergency hospital admission, with the highest-risk combinations conferring roughly seven times the odds of admission compared with the lowest-risk group. Loop diuretics (commonly prescribed for fluid retention in heart failure), certain anti-nausea medications, and drugs for anemia were among the medication classes most strongly linked to hospital visits.9PLOS ONE. Combinations of medicines in patients with polypharmacy aged 65–100 in primary care: Large variability in risks of adverse drug related and emergency hospital admissions Research using a drug-interaction screening tool also found that inpatients had a higher rate of contraindicated drug combinations than outpatients, and those combinations were significantly associated with the adverse events that caused hospitalization in the first place.10PubMed Central. Using INTERCheck to Evaluate the Incidence of Adverse Events and Drug–Drug Interactions in Out- and Inpatients Exposed to Polypharmacy
For patients and caregivers, this means regular medication reviews are not optional housekeeping. They are one of the most important safety interventions available. If you or a family member takes more than a handful of daily medications and sees multiple specialists, asking a pharmacist or primary care physician to do a thorough interaction check is a concrete step toward reducing risk.
Social Circumstances That Clinical Records Miss
A patient’s risk profile is not written entirely in their diagnoses and prescriptions. Whether someone has stable housing, can afford their medications, has a partner or support network, and can read and follow discharge instructions all affect whether they end up back in the hospital. These are often called social determinants of health, and until recently, most risk prediction tools ignored them completely.
Research among Veterans Affairs patients showed that adding patient-reported social and behavioral factors — things like marital status, smoking, resilience, depression symptoms, medication affordability, and health literacy — improved the accuracy of hospitalization predictions beyond what electronic health record data alone could achieve.11JAMA Network Open. Patient-Reported Social and Behavioral Determinants of Health and Estimated Risk of Hospitalization in High-Risk Veterans Affairs Patients A cardiovascular outcomes study reinforced this, finding that among racial and ethnic minorities, clinical models alone underpredicted hospitalization by 20% and cardiovascular hospitalization by 70%. Adding social determinants brought predictions much closer to reality.12PubMed Central. Social Determinants of Health Improve Predictive Accuracy of Clinical Risk Models for Cardiovascular Hospitalization, Annual Cost, and Death
A systematic review of studies incorporating social determinants into electronic health records found that individual-level data, meaning information collected directly from a patient, tended to improve predictive models meaningfully. Area-level data, like a neighborhood’s average income or crime rate, contributed much less.13Journal of the American Medical Informatics Association. Social determinants of health in electronic health records and their impact on analysis and risk prediction: A systematic review The implication is straightforward: zip codes are crude proxies. Actually asking patients about their lives produces better predictions and, potentially, better-targeted help.
Mental Illness as an Overlooked Risk Factor
Severe mental illness — which in clinical terms usually means schizophrenia, bipolar disorder, or major depression — is one of the most potent yet underappreciated contributors to high-risk status. People with these conditions die from cardiovascular disease at about twice the rate of the general population.14PubMed. Severe Mental Illness and Cardiovascular Disease: JACC State-of-the-Art Review The excess risk does not come from the psychiatric diagnosis alone. It stems from a combination of higher rates of smoking, obesity, and metabolic disease; side effects of psychiatric medications that promote weight gain and diabetes; and systematic disparities in the physical healthcare these patients receive.15PubMed Central. Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care
When severe mental illness coexists with chronic physical conditions like heart disease, COPD, diabetes, cancer, or liver disease, hospital use increases further, though the specific impact of each combination is still being mapped.16PLoS ONE. The impact of comorbid severe mental illness and common chronic physical health conditions on hospitalisation: A systematic review and meta-analysis In practice, mental health conditions often cause a cascading effect: depression makes it harder to manage diabetes; uncontrolled diabetes leads to more emergency visits; repeated emergency visits without consistent follow-up make both conditions worse. The result is that a patient with two seemingly manageable conditions can end up in the highest-risk tier.
High-Risk Children Are a Different Population Entirely
The concept of high risk in pediatrics looks nothing like its adult counterpart. Children with medical complexity — those with congenital or acquired multisystem diseases, severe neurological conditions, or dependence on technology like ventilators or feeding tubes — represent a small fraction of the pediatric population but account for a dramatically outsized share of hospital stays, costs, and in-hospital deaths.17PubMed Central. Children with medical complexity: an emerging population for clinical and research initiatives
Estimates of how many children qualify vary widely depending on which classification criteria are used, ranging from under 1% to more than 11% of the pediatric population. Regardless of definition, children who meet the criteria are more likely to be on government insurance, to be male, and to live in urban areas. Across all classification approaches, they have substantially greater odds of mortality and heavy healthcare use than other children.18JAMA Pediatrics. Prevalence of Children With Medical Complexity and Associations With Health Care Utilization and In-Hospital Mortality Their families often become de facto care coordinators, managing interactions with multiple specialists, school systems, insurers, and home-health providers simultaneously.
Being “High Risk” Is Often Temporary
One of the most important and least understood aspects of high-risk status is that for many patients, it does not last. An analysis of health system data found that roughly 3% of adult patients met “super-utilizer” criteria at any given time, accounting for about 30% of total charges. But fewer than half of those patients still met the criteria seven months later, and only about 28% remained at the end of a full year.19PubMed. For many patients who use large amounts of health care services, the need is intense yet temporary
This instability has major consequences for how programs are designed. If you build an intervention around this year’s costliest patients, many of them would have improved next year anyway, simply because the crisis that drove their utilization — a bad flare, an acute illness, a destabilizing life event — resolved on its own. This phenomenon, called regression to the mean, has fooled many well-intentioned programs into thinking they were reducing costs when the natural ebb of acute need was doing most of the work.20PubMed Central. Assessing regression to the mean effects in health care initiatives It is one reason researchers increasingly emphasize identifying “rising-risk” patients — people not yet in crisis but on a trajectory toward it — rather than focusing exclusively on those already at the top of the cost curve. One Medicaid analysis identified about 13.6% of enrollees as rising-risk, compared with only 0.64% who were currently high-cost claimants.21The American Journal of Managed Care. Preventing Tomorrow’s High-Cost Claims: The Rising-Risk Patient Opportunity in Medicaid
When Algorithms Carry Built-In Bias
Increasingly, health systems use automated algorithms to flag high-risk patients. These tools scan electronic health records, sometimes incorporating thousands of variables, to assign patients a risk score.22PubMed. A probabilistic topic model for clinical risk stratification from electronic health records They can process far more data than a human clinician reviewing a chart, but they can also encode biases at enormous scale.
A landmark study published in Science examined a widely used commercial algorithm that affected millions of patients and found significant racial bias. At any given risk score, Black patients were considerably sicker than white patients. The root cause was that the algorithm used healthcare spending as a proxy for health need. Because of systemic inequities, less money was spent on Black patients for equivalent levels of illness. The algorithm therefore interpreted lower spending as lower need, systematically undercounting Black patients’ risk. The researchers estimated that correcting this bias would nearly triple the proportion of Black patients flagged for additional support, from about 18% to roughly 47%.23PubMed. Dissecting racial bias in an algorithm used to manage the health of populations This finding reshaped the conversation about algorithmic risk stratification and serves as a cautionary reminder: a tool is only as fair as the data it was trained on.
What Actually Helps High-Risk Patients
Identifying someone as high risk is only useful if it leads to a meaningful change in their care. The intervention with the most robust evidence is complex care management: a dedicated team (typically a nurse, social worker, or care coordinator) that works intensively with a patient to manage appointments, medication, social needs, and communication between providers. A randomized trial in a Medicaid population found that complex care management reduced total medical expenditures by about $7,700 per patient per year, cut inpatient bed days, and lowered hospital admissions and specialist visits compared with usual care.24The American Journal of Managed Care. Impact of Complex Care Management on Spending and Utilization for High-Need, High-Cost Medicaid Patients
Community health workers — non-clinical staff who share a patient’s community, language, or lived experience — have shown particular promise for socially vulnerable patients. In one program targeting low-income urban patients with heart failure, community health worker support was associated with a steep drop in heart failure-related readmissions, from about 0.64 per patient in the year before to 0.07 after, along with similar reductions in emergency department visits.25PubMed Central. Community Health Workers Reduce Rehospitalizations and Emergency Department Visits for Low-Socioeconomic Urban Patients With Heart Failure A separate randomized trial of a community health worker intervention found that while overall 30-day readmission rates were similar between the intervention and control groups, the intervention significantly reduced multiple readmissions among those who were readmitted, cutting the rate of recurrent readmissions roughly in half. Patients in the intervention group were also more likely to complete a primary care follow-up visit within two weeks of discharge.26JAMA Internal Medicine. Patient-Centered Community Health Worker Intervention to Improve Posthospital Outcomes
These results are encouraging but not uniform. A pilot study of a similar program found a modest and statistically non-significant difference in readmission rates, partly because many patients did not receive the intended number of follow-up calls.27International Journal for Quality in Health Care. Feasibility and evaluation of a pilot community health worker intervention to reduce hospital readmissions The pattern across studies suggests that intensity and follow-through matter: a phone call or two after discharge is not enough. Sustained, relationship-based support with a trusted person who helps patients navigate both medical and social systems seems to be the key ingredient.
The Burden on Patients and Their Providers
Being labeled high risk does not just trigger extra services. It also imposes extra work on the patient. Ethnographic research with multimorbid patients in primary care found that managing polypharmacy, keeping track of potential interactions, and maintaining medication adherence functioned as a form of unpaid labor that patients and families described as burdensome and anxiety-provoking.28PubMed Central. Safety work and risk management as burdens of treatment in primary care: insights from a focused ethnographic study of patients with multimorbidity When patients approach the end of life and begin transitioning between care settings — hospital to home, home to hospice, hospice to hospital again — the challenge intensifies. Patients and caregivers in that situation frequently report a lack of information about new care settings and difficulty accessing support during transitions, which amplifies feelings of uncertainty and distress.29PubMed Central. Experiences of transitioning between settings of care from the perspectives of patients with advanced illness receiving specialist palliative care and their family caregivers: A qualitative interview study
There is also a tension between how patients and clinicians view risk during these transitions. Research on end-of-life care transitions found that patients living at home tended to tolerate increasing levels of risk and developed their own strategies for coping with escalating health and social problems. Hospital staff, by contrast, tended to be risk-averse when making discharge decisions, sometimes prolonging hospital stays or resisting home discharge because of safety concerns the patient was willing to accept.30PubMed Central. Managing risk during care transitions when approaching end of life: A qualitative study of patients’ and health care professionals’ decision making
Genomics and the Future of Risk Prediction
A newer frontier in risk identification involves polygenic risk scores, which estimate a person’s genetic predisposition to diseases like heart disease, diabetes, or certain cancers based on thousands of tiny genetic variants. In theory, these scores could identify high-risk individuals decades before symptoms appear. In practice, translating them into clinical action is still evolving. Researchers have argued that a polygenic score is clinically useful only when it is folded into an absolute risk model — one that also accounts for age, sex, family history, and lifestyle — and linked to a specific threshold that triggers guideline-based care.31PubMed Central. Polygenic risk scores in the clinic: Translating risk into action A high genetic score without a clear action plan is just information. The challenge is building clinical workflows where “your genetic risk for X is elevated” leads reliably to screening, monitoring, or prevention rather than just worry.
Meanwhile, the health system’s capacity to care for growing numbers of high-risk patients is itself under pressure. A large study within the Veterans Health Administration found that as primary care practitioners’ patient panels grew fuller, their odds of burnout and intention to leave clinical practice both increased.32JAMA Network Open. Panel Fullness, Burnout, and Intention to Leave Among Veterans Health Administration Primary Care Practitioners High-risk patients, by definition, require more time, more coordination, and more follow-up per visit. If the workforce shrinks while the patient population ages and accumulates chronic conditions, the gap between need and capacity will only widen. Tools, teams, and algorithms can help, but only if there are enough clinicians willing and able to act on what the tools find.