Karnofsky Performance Status Scale

Learn the Karnofsky Performance Status Scale (100–0): definitions, how to score consistently, ECOG crosswalk, clinical uses in oncology, and common limitations. Companion to the CalcMD KPS calculator.

Karnofsky Performance Status Scale

Select the score that best matches the patient's functional level over the prior weeks. Scores run from 100 (normal) to 0 (dead).

Karnofsky score *

Choose the single best-fit descriptor; use the same anchor over time when following response or decline.

Disclaimer: This tool is for education and clinical support only. Karnofsky scoring is a judgment-based measure; it does not replace examination, staging systems, or shared decision-making. Always apply institutional protocols and specialist input when determining treatment eligibility or intensity.

Karnofsky Performance Status Scale Explained

Overview

The Karnofsky Performance Status (KPS) scale is an 11-point ordinal measure (100 in steps of 10 down to 0) that describes a patient's ability to carry out ordinary activities and need for assistance. It was introduced in the 1940s and remains common in oncology for prognosis, trial eligibility, and communication among clinicians.

Purpose: Summarize global functional capacity and need for care in patients, especially those with cancer or other serious illness.

Standard score definitions

Score	Criteria
100	Normal; no complaints; no evidence of disease.
90	Able to carry on normal activity; minor signs or symptoms of disease.
80	Normal activity with effort; some signs or symptoms of disease.
70	Cares for self; unable to carry on normal activity or do active work.
60	Requires occasional assistance, but is able to care for most personal needs.
50	Requires considerable assistance and frequent medical care.
40	Disabled; requires special care and assistance.
30	Severely disabled; hospitalization is indicated although death is not imminent.
20	Very sick; hospitalization necessary; active supportive treatment necessary.
10	Moribund; fatal processes progressing rapidly.
0	Dead.

Approximate ECOG mapping

Trials and charts often cross-walk KPS to the Eastern Cooperative Oncology Group (ECOG) scale. The mapping below is widely used in practice but is conventional—not part of Dr. Karnofsky's original work.

KPS range	Typical ECOG
90–100	0
70–80	1
50–60	2
30–40	3
0–20	4

Clinical use

Eligibility and stratification in cancer clinical trials
Prognostic discussions and multidisciplinary tumor boards
Serial documentation of response to therapy or disease progression
Triggering supportive care, rehabilitation, or palliative services when scores fall

Limitations

Subject to inter-rater variability and the anchor period used (e.g., past week vs. “usual”).
Does not capture specific symptoms, cognition, or frailty components unless reflected in function.
Should be interpreted with disease-specific staging, labs, imaging, and patient goals—not in isolation.

Background and purpose

The Karnofsky Performance Status (KPS) scale is one of the most familiar tools in cancer medicine for summarizing how well a patient can carry out ordinary activity and how much help they need. Scores are assigned in steps of ten, from 100 (no functional limitation attributable to illness) down to 0 (death). Although it was developed decades ago, it remains embedded in clinical trials, chart documentation, tumor boards, and conversations about prognosis and treatment intensity.

Unlike symptom checklists or disease-specific staging systems, KPS is deliberately global: it integrates cancer burden, comorbid conditions, treatment side effects, nutrition, mood, cognition, and social support into a single ordinal label. That breadth is both a strength (it mirrors what clinicians mean by “how the patient is doing”) and a weakness (it is subjective and can drift between raters or over time if anchors are not explicit).

Historical context

The scale is named for David A. Karnofsky, who with colleagues advanced systematic evaluation of chemotherapeutic agents and patient functional status in the mid‑20th century. The descriptors in common use today follow the familiar pattern of anchoring each score to ability to work, self‑care, need for assistance, and need for hospitalization. Over time, KPS became a standard language in oncology registries and cooperative group studies, often appearing alongside TNM stage and histology.

As trial design evolved, many protocols began to require minimum performance status for cytotoxic therapy or novel agents; KPS (or its close relative, ECOG) became a routine gate for safety and interpretability of results. Even when newer measures of frailty, patient‑reported outcomes, or geriatric assessment are available, KPS often remains the lingua franca in multidisciplinary notes.

Structure of the scale

KPS is an ordered categorical scale, not a continuous physiologic measurement. A change from 80 to 70 is not guaranteed to represent the same “amount” of decline as a change from 50 to 40, and the scale should not be averaged like laboratory values. The clinically meaningful unit is usually the 10‑point step, though some analyses collapse scores into broader bands (for example 100–80, 70–50, <50) for modeling or reporting.

When assigning a score, the rater chooses the single best descriptor that fits the patient’s usual function over a defined interval (commonly the preceding one to two weeks, unless a protocol specifies otherwise). If two adjacent levels seem equally plausible, many teams document the more conservative (lower) score or clarify the time window and principal limiting factor in the note.

Standard score definitions

Score	Descriptor (commonly used wording)
100	Normal; no complaints; no evidence of disease.
90	Able to carry on normal activity; minor signs or symptoms of disease.
80	Normal activity with effort; some signs or symptoms of disease.
70	Cares for self; unable to carry on normal activity or do active work.
60	Requires occasional assistance, but is able to care for most personal needs.
50	Requires considerable assistance and frequent medical care.
40	Disabled; requires special care and assistance.
30	Severely disabled; hospitalization is indicated although death is not imminent.
20	Very sick; hospitalization necessary; active supportive treatment necessary.
10	Moribund; fatal processes progressing rapidly.
0	Dead.

The phrase “normal activity” historically emphasizes employment‑like or full daily activity; “active work” distinguishes heavy occupational demands from basic self‑care. Lower scores emphasize dependence, medical complexity, and whether inpatient level care is clinically appropriate, independent of insurance or bed availability.

How to apply KPS in practice

Anchor the time period

Specify whether the score reflects the usual status over the past week, the best status during a treatment cycle, or the current bedside impression during an acute event. Acute issues (fever, delirium, opioid sedation, uncontrolled pain) can transiently depress KPS; documenting the context prevents misleading comparisons across visits.

Separate function from willingness

KPS measures capacity and need, not personal preference. A patient who could work but chooses not to should not automatically be scored as 70 solely on that basis; conversely, a patient who pushes through symptoms may be scored lower if objective limitations (falls, debility, need for assistance) are present. When cognition is impaired, collateral history from caregivers becomes essential.

Integrate comorbidity and treatment effects

Progressive cancer is only one driver of performance status. Heart failure, chronic lung disease, neurologic disability, infection, anemia, malabsorption, and polypharmacy all belong in the same global picture KPS is meant to capture. This is why KPS often correlates with survival even after adjustment for stage: it encodes host reserve beyond tumor metrics.

Serial measurement

Single snapshots are useful for trial screens; trajectory is often more informative clinically. A decline from 90 to 70 during first‑line therapy may prompt dose modification, supportive care intensification, or imaging to evaluate progression. A rise after treating a reversible problem (obstruction, hypercalcemia, infection) validates response to supportive management even before radiographic reassessment.

Relationship to ECOG performance status

The ECOG (Zubrod) performance status is a simpler 0–4 scale widely used in oncology trials. Because many protocols and publications report ECOG while others use KPS, clinicians frequently cross‑walk between them. Commonly cited banding (used for convenience in practice and secondary analyses) maps approximately as follows:

KPS range	Typical ECOG equivalent
90–100	0 — Fully active
70–80	1 — Restricted but ambulatory
50–60	2 — Ambulatory, limited self‑care, not working
30–40	3 — Limited self‑care, confined to bed or chair much of the day
10–20	4 — Completely disabled / moribund
0	Death (ECOG is not applied after death)

These mappings are conventional, not biologically exact. A patient may “fit” one KPS level yet feel closer to an adjacent ECOG category depending on whether the dominant problem is fatigue versus pain versus neurologic impairment. Trials and databases should always record the scale actually used rather than relying on retrospective conversion alone.

Clinical applications

Treatment selection and intensity

KPS frequently informs whether a patient is likely to tolerate combination chemotherapy, prolonged radiation courses, surgery, or cellular therapies. It is not a substitute for organ‑specific testing, but it often aligns with resilience to toxicity. Patients with lower KPS may still benefit from selected regimens when goals are palliative and burdens are acceptable; the scale should guide discussion, not replace shared decision‑making.

Prognosis and risk communication

In many malignancies and advanced disease states, lower KPS is associated with shorter survival and higher complication rates after intervention. When discussing prognosis, pairing KPS with stage, histology, molecular features, and patient priorities yields a more nuanced picture than any single number. KPS can also help identify patients who may benefit from early integration of specialized supportive services.

Clinical trials and eligibility

Protocols often specify thresholds such as KPS ≥70 or ≥80. Consistency in how sites assign scores affects accrual, safety signals, and generalizability. Sponsors sometimes provide rater training or central adjudication for subjective endpoints; even without that infrastructure, teams improve fidelity by using the same time window and documenting interfering factors.

Documentation and care coordination

Because KPS is widely recognized, it travels well between primary oncology, radiation oncology, surgery, emergency medicine, and inpatient services. A concise note (“KPS 60, limited by dyspnea and debility”) orients consultants quickly and supports handoffs, discharge planning, and equipment or home‑care orders.

Limitations and pitfalls

Inter‑rater variability: Different clinicians may disagree by one or two levels unless anchors and interval are standardized.
Cultural and occupational bias: “Normal activity” and “work” are interpreted against social roles that vary by age, baseline fitness, and socioeconomic context.
Insensitive to specific symptoms: Two patients scored 80 may differ markedly in pain, dyspnea, or anxiety; KPS should be complemented when those domains drive decisions.
Not a frailty instrument: KPS overlaps with frailty but does not replace comprehensive geriatric assessment when treatment decisions hinge on physiologic reserve in older adults.
Acute confounders: Sedation, encephalopathy, or hospital bedrest can artifactually lower scores; repeat assessment after reversible factors are treated is often appropriate.

Practical tips for consistent scoring

State the look‑back period in the chart (for example, “KPS 70 based on usual function over the past 7 days”).
Note the principal limiting factor (tumor, treatment toxicity, comorbidity, psychosocial barrier) when it materially affects interpretation.
Reassess after meaningful clinical changes rather than on a rigid calendar alone.
When both scales appear in the same record, ensure KPS and ECOG are mutually consistent with the crosswalk you use locally, or explain any intentional mismatch.
Use KPS 0 strictly for deceased patients in analytic datasets; living patients should not be assigned 0 regardless of severity.

Medico‑legal and educational note

This article summarizes widely used clinical concepts for education and communication. Performance status is a clinical judgment that must be individualized; it does not mandate any specific treatment, establish disability benefits, or replace institutional policies, informed consent, or specialist consultation.