About

The Barthel Index quantifies a patient's performance across 10 activities of daily living (ADL), yielding a composite score from 0 (total dependence) to 100 (full independence). Developed by Mahoney and Barthel in 1965, it remains the most widely validated ordinal scale in stroke rehabilitation and geriatric medicine. Misscoring even one item can shift a patient between dependency categories, altering discharge planning, staffing ratios, and insurance reimbursement. The tool uses the original 10-item weighted version where items like bathing contribute 0 - 5 points while ambulation contributes 0 - 15 points, reflecting their relative clinical weight.

This calculator applies the Modified Barthel Index scoring convention (Shah et al., 1989) mapped to the original 0 - 100 scale. Note: the index measures what a patient does, not what they can do. Scores should reflect observed performance over 24 - 48 hours. The scale has a known ceiling effect above 85 points and does not capture cognitive or communicative function.

Formulas

The Barthel Index is computed as the unweighted sum of 10 individually weighted ADL item scores:

BI = 10∑i=1 s_i

Where BI = Barthel Index total score (0 - 100), s_i = score for the i-th ADL item, and each s_i takes a value from its item-specific ordinal set (e.g., feeding: {0, 5, 10}, bathing: {0, 5}). The maximum possible score is 5 + 5 + 10 + 10 + 10 + 10 + 10 + 15 + 15 + 10 = 100.

Dependency classification uses a piecewise mapping:

Class = {

Total if BI ≤ 20Severe if 21 ≤ BI ≤ 60Moderate if 61 ≤ BI ≤ 90Slight if 91 ≤ BI ≤ 99Independent if BI = 100

Reference Data

ADL Item	Weight Range	Score Options	Clinical Notes
Feeding	0 - 10	0, 5, 10	Includes cutting food; 5 = needs help cutting or spreading
Bathing	0 - 5	0, 5	Independent means unassisted shower or bath
Grooming	0 - 5	0, 5	Teeth, hair, shaving, face care
Dressing	0 - 10	0, 5, 10	Includes buttons, zips, laces, braces
Bowel Control	0 - 10	0, 5, 10	5 = occasional accident (<1×/week)
Bladder Control	0 - 10	0, 5, 10	5 = occasional accident or needs catheter help
Toilet Use	0 - 10	0, 5, 10	Transfer on/off, wiping, flushing
Transfers (Bed↔Chair)	0 - 15	0, 5, 10, 15	0 = unable, no sitting balance
Mobility (Level Surface)	0 - 15	0, 5, 10, 15	5 = wheelchair independent ≥50 m
Stairs	0 - 10	0, 5, 10	May use handrail or aid for independent score
Dependency Classification Thresholds
Total Dependence	0 - 20 points
Severe Dependence	21 - 60 points
Moderate Dependence	61 - 90 points
Slight Dependence	91 - 99 points
Independent	100 points

Frequently Asked Questions

The original Barthel Index (Mahoney & Barthel, 1965) uses a 0-100 scale with unequal item weights (5, 10, or 15 points per item). The Modified Barthel Index (Shah et al., 1989) introduced 5-level scoring per item for finer granularity. This calculator implements the original 0-100 weighted version, which remains the most commonly cited in stroke outcome literature and insurance assessments.

The weights reflect Mahoney and Barthel's clinical judgment about each item's contribution to overall independence. Mobility and transfers are gatekeeper activities - a patient who cannot transfer cannot use the toilet, bathe, or dress independently regardless of those specific skills. Bathing received 5 points because it was considered supplementary; supervision alone often suffices.

Patients scoring above 85 may still have significant functional limitations that the index cannot detect. For example, a patient scoring 95 (only losing 5 points on bathing) might have severe cognitive impairment, aphasia, or poor community mobility. The Barthel Index does not capture instrumental ADLs (shopping, cooking, finances). For higher-functioning patients, the Functional Independence Measure (FIM) or Lawton IADL Scale provides better discrimination.

Scores must reflect actual performance over the preceding 24-48 hours, not capability or best performance. If a patient can walk independently but refuses due to fear of falling, score the observed behavior (dependent). This is a frequent source of inter-rater disagreement. Document the context when scoring.

Inter-rater reliability is generally high (Cohen's kappa 0.85-0.95) for trained assessors using direct observation. Reliability drops significantly when scoring is based on chart review alone or when raters are untrained. Telephone-administered Barthel assessments show moderate agreement (kappa ≈ 0.75). Always specify the assessment method when reporting scores.

Yes. A score below 40 at 2 weeks post-stroke strongly predicts institutional care rather than home discharge. Scores above 60 at admission correlate with home discharge in approximately 80% of cases. However, discharge planning should integrate social factors (caregiver availability, home accessibility) that the index does not measure.

The MCID is generally accepted as a change of 1.85 points on the 20-point version (equivalent to approximately 9 points on the 0-100 scale). Changes smaller than this threshold may represent measurement error rather than true functional change. When tracking rehabilitation progress, ensure re-assessments use the same rater and method to minimize noise.