Sign Up
Login
Try for Free
Login
The Morse Fall Scale is a nurse-rated screen for a patient's risk of falling, published by Janice Morse and colleagues in 1989, that scores six items for a total of 0 to 125, higher meaning higher risk. Hospital, rehabilitation, long-term care, and home health nurses use it to decide who needs which fall prevention measures. This page covers how to document the score and the interventions it triggers, with a fictional sample.
Registered nurses on admission and at reassessment in acute care, rehabilitation, long-term care, and home health, with licensed practical or vocational nurses where facility policy and state scope allow; no publisher qualification level or training certificate applies, and the nurse who watched the patient move records the result
Relieving and receiving nurses at handoff, charge nurses and nurse managers, attending physicians and advanced practice providers, pharmacists and physical therapists who receive the referrals, patient safety and quality staff, coders determining present-on-admission status after an injury, CMS and state surveyors, accreditation reviewers, and the post-fall reviewers and attorneys who reconstruct the sequence
4 to 10 chart lines (trigger and version, items that scored with the total, facility category, a measure for each scoring item, teaching and response, reassessment triggers) · 2 to 4 minutes by hand after the patient has been observed walking
Nurse-rated fall risk screen (six weighted items, some with graded responses, total 0 to 125, higher total meaning higher risk; risk categories set by facility policy from published examples, not by the instrument)
Admission and transfer assessments in acute care and rehabilitation, scheduled reassessment under unit policy, change of condition, after any fall, long-term care admission and periodic reassessment beside the MDS, home health start of care under agency policy; pediatric units use tools validated in children
Morse, Morse, and Tylko (Canadian Journal on Aging, 1989); copyrighted, free to use in clinical care with the author's permission per the Fall TIPS FAQ; carried as an example tool by AHRQ and the VA and mandated by no regulator or accreditor in the United States, Canada, or Australia; described here for documentation, no items or weights reproduced
The Morse Fall Scale is a nurse-rated screen for the likelihood that a hospitalized adult will fall. Janice M. Morse developed it with Robert M. Morse and Suzanne J. Tylko and published it in the Canadian Journal on Aging in 1989 (volume 8, pages 366 to 377), from a case-control study whose discriminant analysis correctly classified 80.5 percent of patients; her book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008) carries the scoring guidance and the level-of-risk chapter that facilities still cite. The scale weights six items, some with graded responses, for a total of 0 to 125, and a higher total means higher risk. Named here as topics in original words, the items ask whether the patient has fallen recently or during this admission; whether more than one active medical diagnosis is present; what the patient relies on to walk, because dependence on an aid, or on whatever is within reach, marks unsteadiness; whether an intravenous line or access device is in place, because it tethers and encumbers; how the patient actually walks when observed; and whether the patient's judgment of what they can safely do matches their real ability, the item that flowsheets usually label mental status. The item wording, the response anchors, and the point weights are the copyrighted text of the instrument and are not reproduced or paraphrased on this page; your facility's licensed form or approved electronic build is the source of record.
The load-bearing fact for documentation is that the total is a screening number whose meaning is set locally. The familiar bands, with 25 and 45 as the usual edges, appear as an example in AHRQ's Preventing Falls in Hospitals toolkit (Tool 3H) and in the VA's falls toolkit, whose policy template leaves the high-risk score blank for each unit to determine and points to Morse's book, which advises calibrating the cutoff to the unit's own patients; validation studies have found the best cutoff at 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), 51 in a Korean electronic record (Baek and colleagues, 2014), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and no regulator in the United States, Canada, or Australia names the scale or a threshold. The original validation reported a positive predictive value of 10.3 percent, so most patients the scale calls high risk never fall, and a 2026 VA study of 5,004 inpatients found the mean total above 45 in fallers and non-fallers alike while individual items still carried signal (Oppegaard and colleagues). Current guidelines have moved further: NICE (2025) advises against using falls risk prediction tools to predict who will fall, and the ACSQHC's 2025 hospital guideline treats every older inpatient as high risk and directs effort to the person's own risk factors. The lesson for the chart is the same in every setting: the items that scored, not the total, choose the interventions, and the record has to show that link. Locally altered forms (a medication item added, a rule that any in-facility fall makes the patient high risk, a 51-point band) are local instruments rather than the published scale, which Morse herself addressed in a 2006 commentary, so the note names the version or build it used. The Braden Scale is the structural peer on the skin side: a risk total that must never be mistaken for a plan.
Medical, surgical, orthopedic, and rehabilitation unit nurses score it on admission, on transfer, at the interval unit policy sets, when the patient's condition changes (a new sedating, opioid, or blood pressure medication, new confusion, a new line, a change in mobility), and after any fall; critical care nurses record it beside the RASS and the CAM-ICU, where sedation and delirium alter the self-appraisal and mobility items at once, and where a patient who does not walk can score low while transfers and toileting remain the moments of danger. Long-term care nurses use it, or a facility variant, at admission, with the MDS assessment cycle, with any change of condition, and after any fall, feeding the Section J fall items and the care plan that surveyors read under F689. Home health nurses record a fall risk assessment at the start of care under agency policy, sometimes the Morse and often a tool built for community-dwelling patients, and the OASIS assessment page covers what the item set itself asks. Charge nurses and nurse managers audit the entries; pharmacists and physical therapists receive the medication review and mobility referrals that scoring items generate; physicians and advanced practice providers document injuries and present-on-admission status after a fall, which rest on the clinical assessment and not on the score; patient safety and quality staff read the sequence after every event. The Morse page owns one job: the assessment, the items that scored, the facility's category named as policy, and the link from each scoring item to a specific measure and a reassessment trigger. Neighbors take over from there: a fall is reported through the incident report while the clinical reassessment stays in the chart, the risk status and measures travel in the transfer of care summary and the discharge summary, the medication review belongs in the medication management note, and the interdisciplinary plan the items feed belongs in the care coordination note. Pediatric units use a tool validated in children, such as the Humpty Dumpty Falls Scale, because the Morse was developed and validated in adults.
No regulator prescribes a Morse note format, and no authority in the United States, Canada, or Australia requires the Morse by name; what exists is a copyrighted instrument, facility policies that set the reassessment interval and the risk categories, and survey, accreditation, and coding rules that judge whether an identified risk led to action. What survives review is an entry that states the trigger and the version, names the items that scored in plain words with the total, attributes the category to policy, ties every scoring item to a specific measure that was actually delivered, records teaching and the patient's response, sets the next reassessment and the changes that bring it forward, and, after a fall, documents a new clinical assessment rather than a copied score. Each element below carries the pitfall that most often undermines it.
Trigger, timing, version, and rater. Open with why the assessment is being done: admission, transfer in, the scheduled interval under a named policy, a change in condition (a new opioid, sedative, or antihypertensive, new confusion, a new line, a change in how the patient moves), or a fall. Then the version (the published scale as built in the electronic record, or a named facility variant), the date and time, and the rater who watched the patient move. Every-shift scoring is a unit policy: the VA's falls toolkit names admission, transfer, change in condition, and after a fall as the triggers and leaves routine frequency to the facility, and neither CMS nor the Joint Commission sets an interval, so cite the policy rather than presenting the habit as a rule. A trigger named in one clause is what turns a repeated total into evidence of monitoring. Pitfall: A total time-stamped every twelve hours with no trigger, or a note that presents the unit's every-shift habit as a CMS or Joint Commission requirement.
The items that scored, in plain words, then the total. Write which of the six items scored and what each rests on in ordinary clinical facts: the date and place of the last fall, the active diagnoses, the aid the patient uses, the line that is in, what you saw when the patient walked (step length, steadiness, use of the hands, hesitation), and what the patient said about getting up alone against what you observed. Then the total against its maximum. Rate the patient you observed today: the gait and self-appraisal items are assessments of the patient, not of the chart, which AHRQ's toolkit says in as many words, and a total copied from the last shift is not a reassessment. The item topics can be named in your own words; the instrument's wording and weights belong to the licensed form. Pitfall: "Morse 85" with no items named, or a gait item rated from the previous entry while the patient lay in bed the whole shift.
Facility risk category, named as policy, with any override. Write the category the facility policy assigns to the total and cite the policy, because the categories are conventions layered on the instrument: the 25 and 45 edges most facilities use appear as an example in AHRQ's toolkit and in VA practice, Morse's own guidance is to calibrate the cutoff to the unit, and published optima run from 40 to 55 by population. Then record any clinical judgment override and its reason: a patient who does not walk and scores low on the mobility items but climbs out of bed, or a patient in the high band whose only scoring item has resolved. The category tells a reader which protocol tier applies; the items, not the tier, decide what is done, and a sentence such as high risk therefore alarm should be written only where that linkage is literally the organization's approved protocol. Pitfall: A category presented as the scale's rule, or a bedbound, agitated patient labeled low risk with no override recorded.
An intervention for each scoring item. For every item that scored, name the measure in place or started and how it is delivered: for a prior fall, the rounding interval, the call light within reach, and the teach-back; for multiple diagnoses, the medication review requested and from whom; for a walking aid, the aid within reach and the level of assistance for every walk; for a line, the pole on the side of the aid, tubing managed, disconnection when policy allows; for an unsteady gait, the gait belt, the footwear, the bed height, the therapy referral; for a mismatch between the patient's confidence and ability, the supervision plan, the toileting schedule, and who stays with the patient. Add the universal environmental checks with the time. Fall TIPS, the toolkit built on the scale, exists to make exactly this link, and in 37,231 patients on 14 units it was associated with 15 percent fewer falls and 34 percent fewer injurious falls (Dykes and colleagues, 2020). A bed or chair alarm is one measure among others, not the plan: CMS's long-term care guidance cautions against relying on position-change alarms as the sole intervention, and no authority reads a high total as a reason for restraint. Pitfall: "Fall precautions maintained" or "high fall risk bundle in place" with no measure that answers a specific scoring item.
Teaching, preferences, and refusals. Record the teaching given to the patient and family, in their words where you can, with a teach-back of the one behavior that matters most (call before getting up), and the patient's stated preferences. When a measure is declined, chart the offer, the reason given, the education, the alternatives offered or accepted, who was notified under policy, and the plan as revised; CMS's long-term care guidance expects a resident's choice to be honored while the team works to mitigate the risk, and a signed waiver does not end the duty. A family member at the bedside is part of the plan, not a scored item, so it changes the measures and the supervision, never the number. Pitfall: "Refused bed alarm" with nothing after it, or a daughter staying overnight charted as if the total had dropped.
Reassessment triggers and a changed total. Close with the next scheduled reassessment under the policy and the changes that bring it forward: transfer, a new sedating, opioid, or blood pressure medication, new confusion, a change in mobility, a line placed or removed, and any fall. When the total changes, say which item changed and why in clinical terms, because two items can move in opposite directions and hide each other: a line removed lowers the total on the same day that new confusion raises the risk, and a note that reports only the lower number tells the next reader the opposite of the truth. De-escalate measures item by item when an item resolves, and carry the scoring items, the measures in place, and the next reassessment into the handoff and the transfer of care summary. Pitfall: A total that fell because a line came out read as lower risk, while the patient's new confusion went unrecorded.
After a fall: the clinical entry and the incident report. A fall triggers a new assessment, not an edit. Chart the injury check and vital signs, the neurological checks where the head may have struck, the provider notification with time and response, the imaging or treatment ordered, a fresh Morse in which the fall now scores, the contributing factors you can see (toileting urgency, footwear, lighting, a line, a medication given), the revised measures, the patient's and family's understanding, and the next reassessment. Report the event through the organization's safety reporting system separately; AHRQ's toolkit keeps the clinical post-fall assessment and the root-cause review as different activities, and the chart should not carry the incident narrative or speculation about blame. Whether the chart cites the report number is a local risk-management decision. The injury diagnosis and present-on-admission status that coders need come from the provider's note; the fall's time and circumstances come from yours. The incident report page covers the safety document itself. Pitfall: The pre-fall score copied forward after the event, the incident report pasted into the progress note, or a post-fall entry with no revised plan.
MORSE FALL SCALE DOCUMENTATION BLOCK (risk assessment and linked plan) Date / time: [ ] Setting: [ ] Rater: [ ] Trigger: [admission / transfer / scheduled per policy (name it) / change in condition (state the change) / after a fall] Version: [published scale as built in the EHR / facility variant (name it)] Items that scored (plain words, no weights): [recent or in-facility fall: date and place / more than one active diagnosis: which / walking aid: which / IV line or access: which / observed gait: what you saw / self-appraisal of ability against observed ability] Items that did not score and why: [ ] Total: [ ] of 125 Prior total and date: [ ] Item that changed and why: [ ] Facility risk category: [ ] per [policy name]; categories are policy, not the scale's rule Clinical judgment override: [none / reason] Interventions by scoring item (what is actually in place): prior fall: [rounding interval; call light in reach; teach-back] diagnoses: [medication review requested: by whom, when] walking aid: [aid within reach; assist level for every walk] IV line: [pole placement; tubing managed; disconnection per policy] gait: [gait belt; footwear; bed height; therapy referral] self-appraisal: [supervision plan; toileting schedule; who stays] universal: [bed low and locked; path clear; lighting; belongings in reach] Teaching and response: [patient and family teach-back / preferences / declined: offer, reason, education, alternatives, who was notified] Reassessment: [next per policy; earlier for transfer, new sedating or blood pressure medication, new confusion, mobility change, any fall] After a fall (if applicable): [injury check; vitals; provider notified, time; new score; contributing factors; revised plan; safety report filed separately] Communication: [provider, pharmacy, therapy, charge nurse; handoff; transfer summary] Clinician signature / credentials: Date:
Free to use and share, no signup. The PDF includes a one-page cheat sheet with element-by-element pitfalls and a pre-sign checklist; the DOCX is the blank documentation block, ready to adapt. Neither reproduces the scale's items, anchors, or point weights.
Scenario: a US community hospital medical unit, where the admitting nurse documents a 77-year-old woman's Morse Fall Scale on the evening of admission for pneumonia, names the items that scored in plain words, attributes the risk category to hospital policy, ties each scoring item to a specific measure, records a declined bed exit alert with the alternatives, and two days later documents a change-of-condition reassessment in which the total fell while the risk rose. All details are fictional.
Patient: L.M., 77 · Setting: Community hospital, medical unit; admission assessment and change-of-condition reassessment · Clinician: R. Delgado, RN · Note date: 09/14/2026
Trigger, version, and rater (09/14/2026, 20:10): Admission fall risk assessment under the hospital's fall prevention policy NSG-207, which requires the Morse Fall Scale on admission, every shift, with any change in condition, and after any fall; the interval is hospital policy, not a regulatory requirement. Published scale as built in the electronic record, rated by R. Delgado, RN, at 20:10 on 09/14/2026 after walking L.M. from the stretcher to the bathroom and back with her cane. Admitted from the emergency department at 19:30 with community-acquired pneumonia; no prior score this admission.
Items that scored and total: Five of the six items scored. Prior fall: L.M. fell in her kitchen on 08/22/2026 reaching for a cupboard, no injury, confirmed by her daughter. Active diagnoses beyond the admitting one: heart failure, atrial fibrillation on an anticoagulant, and osteoarthritis of both knees. Walking aid: a single-point cane brought from home. Intravenous access: a peripheral line in the left forearm for antibiotics and a diuretic. Observed gait: short, slow steps, trunk bent forward, eyes on the floor, one hand on the cane and the other reaching for the wall on the return, with two stops to steady herself. The self-appraisal item did not score: she says she cannot walk without the cane, asked for help before standing, and described the near-miss that preceded the August fall accurately. Total 85 of 125.
Facility risk category: High fall risk under NSG-207, which classifies totals of 45 and above as high risk, following the example bands in the AHRQ toolkit; the threshold is the hospital's policy choice, not a rule from the scale or from a regulator. No clinical judgment override is needed today, because the category and the observed risk agree. The measures below answer the individual items; the policy's high-risk tier is applied without being restated.
Interventions by scoring item: Prior fall: hourly rounding on the unit schedule, call light clipped within reach and demonstrated, teach-back completed (below). Multiple diagnoses and medications: pharmacist medication review requested at 20:40 through the admission order set, with attention to the timing of the diuretic and the anticoagulant's bearing on injury risk; attending informed of the August fall during the admission call at 20:55. Walking aid: cane kept at the bedside on her right, one-person standby assist for every walk entered as a nursing intervention, physical therapy evaluation requested for 09/15/2026. Intravenous access: pole kept on the cane side, tubing routed away from her feet, line capped between infusions per policy. Gait: gait belt at the bedside for assisted walks, nonslip socks applied, bed in low position with wheels locked, bathroom path cleared and the night light on. Universal checks completed at 20:20: belongings, water, and glasses within reach; two upper side rails raised at her request for repositioning, per policy, and not a restraint. Bed exit alert offered and declined (below).
Teaching, preferences, and the declined alert: L.M. and her daughter (present; health care proxy) were taught the one rule that matters most: call and wait for help before getting up. L.M. repeated it back correctly and agreed. She declined the bed exit alert, which she recalled from an admission last year as noisy and embarrassing when it sounded; the offer, her reason, the education about how the alert is used, and the alternatives were documented, and she accepted the alternatives: the room across from the nurses' station, hourly rounding with a bathroom offer, and her daughter staying until 22:00 each evening. Charge nurse informed of the declined alert at 21:05 per policy. The daughter's presence is recorded as part of the plan and does not change the score.
Reassessment plan: Rescore each shift per NSG-207 and immediately with any new sedating, opioid, or blood pressure medication, new confusion, a change in her gait or aid, removal of the line, or any fall. Handoff at 23:00 carried the five scoring items, the measures in place, and the declined alert.
Reassessment entry (09/16/2026, 21:30, change in condition): Change-of-condition reassessment by R. Delgado, RN, after two events on the evening shift: the peripheral line was removed at 15:00 when antibiotics were changed to oral, and since a low-dose opioid was started for pleuritic chest pain on the evening of 09/15/2026 L.M. has twice been found standing at the bedside without calling, at 19:40 and 21:10, saying she did not want to bother anyone and thought she could manage. Rescored after walking her to the bathroom with the cane and standby assist: the prior fall, multiple diagnoses, walking aid, and observed gait items still score, gait unchanged from admission; the intravenous access item no longer scores; the self-appraisal item now scores, because her account of what she can do safely no longer matches what was observed. Total 80 of 125, down from 85. The lower total reflects the removed line and not lower risk; the new scoring item is the more dangerous one, and the category remains high risk under NSG-207. Measures revised item by item: line measures discontinued; for the self-appraisal item, the bed exit alert was re-offered and accepted (on and tested at 21:20), bathroom offers scheduled every two hours while awake and at 02:00, her daughter will stay overnight tonight at her own offer, and one-person assist applies to every transfer; attending and pharmacist notified at 21:35 of the timing of the opioid doses and the change in judgment, medication review requested; the other measures continue, as does the therapy plan from the 09/15/2026 evaluation. L.M. re-taught to call first, repeated it back, and agreed to the alert. Reassess at 06:00 and with any further medication change, transfer, or fall. No fall has occurred, so no safety report is filed; if one occurs, the clinical assessment will be charted here and the event reported through the safety reporting system separately.
This sample is fictional and for educational purposes. It does not describe a real patient or record; the scores, dates, and details are invented to show documentation structure and are not clinical guidance. No scale items, anchors, or point weights are reproduced.
Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript.
Generate a note from bulletsUnited States: no LAW names the Morse Fall Scale, and the rules that do exist judge what the score led to. In long-term care, 42 CFR 483.25(d) (LAW, surveyed as F689) requires the facility to ensure that the resident environment remains as free of accident hazards as is possible and that each resident receives adequate supervision and assistance devices to prevent accidents; the interpretive guidance in Appendix PP of the State Operations Manual sends surveyors to whether foreseeable hazards and each resident's own risks were identified and addressed, expects the causes of a fall to be evaluated and the care plan revised afterward, cautions against relying on position-change alarms as the fall prevention plan, and warns that restraints used to prevent falls can cause serious harm. The MDS 3.0 (LAW as to the assessment) collects fall history at admission (J1700), falls since the prior assessment (J1800), and falls by injury level (J1900), and the RAI Manual version 1.20.1 effective October 1, 2025 clarified the definitions of a fall and a major injury; none of it names a screening tool. In hospitals, the Medicare Conditions of Participation require a nursing assessment for every inpatient (42 CFR 482.23) and complete, dated, timed, and authenticated records (42 CFR 482.24) without prescribing a tool or an interval. Under PAYER POLICY, section 5001(c) of the Deficit Reduction Act of 2005 placed Falls and Trauma (fractures, dislocations, intracranial injuries, crushing injuries, burns, and other injuries) among the hospital-acquired condition categories for which Medicare pays no additional amount when the condition was not present on admission, for discharges on or after October 1, 2008, with the present-on-admission indicator reported since October 1, 2007; the determination rests on the provider's injury diagnosis and the documented time of the fall, never on a risk score, and the provision is distinct from the Hospital-Acquired Condition Reduction Program, which scores a patient safety composite and infection measures. CMS added a Hospital Harm: Falls with Injury electronic clinical quality measure (CMS1017) to the Hospital Inpatient Quality Reporting Program for the 2026 reporting year, which counts inpatient falls with moderate or major injury and, again, names no tool. In home health, OASIS collection is required (LAW and PAYER POLICY), and the current OASIS-E2 item set, effective April 1, 2026, contains no fall risk tool item (the former M1910 is gone) but still collects falls since the start or resumption of care at transfer and discharge (J1800 and J1900) and asks in the intervention synopsis (M2401) whether falls prevention interventions were in the plan of care whenever a standardized, validated multi-factor fall risk assessment showed risk, so the agency chooses the tool under its own policy, often one built for community-dwelling patients rather than the inpatient-derived Morse (the OASIS assessment page covers the items). Accreditation is CONVENTION with teeth. The Joint Commission replaced the hospital and critical access hospital National Patient Safety Goals chapter with National Performance Goals on January 1, 2026; under Goal 11, NPG.11.02.01 requires the hospital to assess and manage the patient's risks for falls and to implement fall risk reduction interventions based on the patient population, the setting, and the individual patient's assessed risks. The fall reduction National Patient Safety Goal, NPSG.09.02.01, continues in 2026 for the nursing care center program and the Joint Commission's assisted living and home care programs, with elements of performance that assess risk, implement interventions based on the assessed risk, educate staff, educate the patient or resident and family on individualized strategies, and evaluate the effectiveness of the program; no Joint Commission requirement names the Morse or an interval. AHRQ's Preventing Falls in Hospitals toolkit carries the scale as Tool 3H beside STRATIFY and directs users to plan care from the patient's own risk factors; the VA's National Center for Patient Safety falls toolkit names admission, transfer, change in condition, and after a fall as the triggers and leaves each unit to set its own score parameters; NDNQI and Magnet track falls with injury as a nurse-sensitive indicator and name no tool. In litigation and survey defense (CONVENTION), the questions a reviewer asks are what was known, what was done about it, and what changed after the event, so a chart of repeated totals with a generic precautions phrase is exposed at each step; no published dataset counts how often that pattern is targeted, this page does not claim one, and negligence standards and incident report privilege are state law questions for counsel.
Canada and Australia require a validated process and name no tool, and the evidence explains why the items matter more than the total. In Canada, HSO 5060:2025, Preventing Falls and Reducing Injuries from Falls, is a Required Safety Practice of the Health Standards Organization used by Accreditation Canada's programs (CONVENTION with accreditation force), the successor to the Required Organizational Practice that asked organizations to define a fall, screen on admission and after a change in status or a fall, act on identified risk, and review every fall, and it names no instrument; Ontario's long-term care regulation, O. Reg. 246/22 under the Fixing Long-Term Care Act, 2021 (LAW in Ontario long-term care), requires each home to run a falls prevention and management program and to complete a post-fall assessment when a resident falls; and the Registered Nurses' Association of Ontario best practice guideline on preventing falls and reducing injury from falls (fourth edition, 2017; CONVENTION) recommends multifactorial assessment and intervention rather than reliance on a score. In Australia, the NSQHS Comprehensive Care Standard (second edition) requires systems for preventing falls and harm from falls that are consistent with best-practice guidelines, equipment and devices that promote safe mobility, and information for patients, carers, and families (actions 5.24 to 5.26; CONVENTION with accreditation force); the Commission's 2025 falls guidelines for hospitals, residential aged care, and community care advise considering every older person in hospital to be at high risk and building prevention on the person's individual risk factors and a post-fall review rather than on a calculated score; and a third edition of the standards is in development, so no new requirement should be asserted until it is final. The reference points everyone cites bind no one in the three countries but set the direction: NICE NG249 (April 2025), replacing the 2013 guideline, states in recommendation 1.1.1, "Do not use falls risk prediction tools to predict a person's risk of falling," and the World Falls Guidelines (2022) recommend against scored screening tools in hospitals in favor of multifactorial assessment and tailored intervention. On the evidence: Morse's 1989 study reported sensitivity of 78 percent, specificity of 83 percent, a positive predictive value of 10.3 percent, a negative predictive value of 99.3 percent, and inter-rater reliability of .96; at a cutoff of 45, Chow and colleagues (2007) found sensitivity of only 31 percent in 954 Chinese inpatients, Baek and colleagues (2014) found 51 the best cutoff in a Korean electronic record (sensitivity 0.72, specificity 0.91, area under the curve 0.77), Mao and colleagues (2024) found 40 in obstetric and gynecologic wards, and Schwendimann and colleagues (2006) found 55 in a Swiss hospital; Aranda-Gallardo and colleagues (2013) pooled 14 studies, ranked STRATIFY above the Morse on diagnostic odds ratio (7.64 for STRATIFY), and concluded that the instruments behave very differently across populations and settings; Oppegaard and colleagues (2026) found the mean total above 45 in both fallers and non-fallers among 5,004 VA inpatients, Zhang and colleagues (2026) found no significant difference in totals between matched fallers and non-fallers in 4,887 acute care patients, and Jiang and colleagues (2025) found an area under the curve of 0.825 in 206,846 Chinese inpatients, so discrimination runs from poor to good depending on population, workflow, and threshold; Morris and colleagues (2022) found education the only intervention with a significant reduction in hospital falls (rate ratio 0.70) and found risk assessment tools, alarms, and sensors on their own not associated with fewer falls; and Fall TIPS, which links each scoring item to a matched intervention, was associated with 15 percent fewer falls and 34 percent fewer injurious falls in 37,231 patients (Dykes and colleagues, 2020). On versions: Morse's 2006 commentary, The Modified Morse Fall Scale, addressed the spread of locally altered forms, and the medication items, automatic high-risk rules after an in-facility fall, and 51-point bands found in local variants make those forms different instruments whose validity is not inherited from the published scale, so the note names the version or build; the adult scale is not validated in children, where the Humpty Dumpty Falls Scale and similar pediatric tools apply; and a patient who does not walk can score low on the mobility items while remaining at risk during transfers and toileting, which is why the record carries a clinical judgment override beside the number.
The Morse Fall Scale is a copyrighted instrument. Janice M. Morse developed it with Robert M. Morse and Suzanne J. Tylko and published it in the Canadian Journal on Aging in 1989, and her book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008) sets out its use, so the item wording, anchors, and weights sit under the author's and the publishers' rights rather than in the public domain. The author's published position, carried on the Fall TIPS FAQ maintained by the Brigham and Women's Hospital team whose toolkit is built on the scale, is that use of the scale is free of charge but permission from the author is required, with requests directed to Dr. Morse at the University of Utah; Morse's 2006 commentary in the International Journal of Nursing Practice objected to modified versions circulating under the scale's name, so a facility that changes items, weights, or response logic has created a local instrument and should say so. No public statement grants blanket permission for websites, calculators, custom forms, or electronic builds, and the many calculators that reproduce the anchors show market practice, not a license; by contrast, the Johns Hopkins Fall Risk Assessment Tool is licensed for a fee by the Johns Hopkins Health System Corporation and the Hendrich II Fall Risk Model is licensed and patented through AHI of Indiana, so a free-with-permission scale, a licensed one, and a patented one sit side by side in the same market. BastionGPT is not affiliated with, or endorsed by, the authors or the rights holder. This page reproduces no test items, stimuli, norms, or scoring materials.
The numbers behind these errors are specific. Morse's development study classified 80.5 percent of patients correctly, with sensitivity of 78 percent, specificity of 83 percent, and a positive predictive value of only 10.3 percent, so about nine of ten patients the scale calls high risk never fall (Morse, Morse, and Tylko, 1989). Among 5,004 VA inpatients, 3.5 percent fell, the mean total of fallers and non-fallers alike sat above the conventional 45, and the total did not usefully separate the two groups while prior falls, an assistive device, an unsteady gait, and impaired self-appraisal still did (Oppegaard and colleagues, 2026); among 4,887 acute care patients, totals did not differ significantly between matched fallers and non-fallers (Zhang and colleagues, 2026), while a study of 206,846 inpatients reported an area under the curve of 0.825 (Jiang and colleagues, 2025). The best cutoff was 51 in a Korean electronic record (Baek and colleagues, 2014), 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and at 45 the scale identified only 31 percent of fallers in 954 Chinese inpatients (Chow and colleagues, 2007). A 2022 systematic review found risk assessment tools, alarms, and sensors on their own not associated with fewer falls in hospitals and education the only intervention with a significant effect (Morris and colleagues), while the Fall TIPS toolkit, which ties each scoring item to a matched measure, was associated with 15 percent fewer falls and 34 percent fewer injurious falls in 37,231 patients (Dykes and colleagues, 2020). No published audit counts how often a chart carries a Morse total with no linked intervention; the F689 survey pathway, the accreditation requirements, and the hospital-acquired condition rules described under compliance considerations are what turn that omission into a deficiency or a payment and defensibility problem. The BastionGPT Clinical Advisory Board sees the same errors most often in Morse Fall Scale documentation reviews:
BastionGPT is specifically trained, tuned, and clinically tested on nursing fall risk assessments and safety documentation.
See how clinicians use it day to day on the AI medical notes page.
Many BastionGPT users report saving more than 90 minutes per day on documentation.
HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign.
The scale scores six weighted items, some with graded responses, for a total of 0 to 125, and a higher total means higher risk. Named as topics in original words, the items ask whether the patient has fallen recently or during this admission; whether more than one active medical diagnosis is present; what the patient relies on to walk, because dependence on an aid or on whatever is within reach marks unsteadiness; whether an intravenous line or access device is in place, because it tethers and encumbers; how the patient actually walks when you watch; and whether the patient's judgment of what they can safely do matches what you observe, the item that flowsheets usually label mental status. The item wording, the response anchors, and the point weights are the copyrighted text of the instrument, so this page does not reproduce or paraphrase them; score from your facility's licensed form or approved electronic build and write the basis for each scoring item in ordinary clinical words. The total is a screening number: the items that scored, not the sum, tell the next nurse what to do.
Whichever your facility's policy says, and the note should cite that policy. The bands most facilities use, with 25 and 45 as the edges, appear as an example in AHRQ's Preventing Falls in Hospitals toolkit and reflect common VA practice, but Morse's own guidance is to calibrate the high-risk cutoff to the unit's patients, the VA's falls toolkit policy template leaves the number blank for each unit to determine, and no regulator, accreditor, or payer in the United States, Canada, or Australia has adopted 45 or any other value. Validation studies explain why: the best cutoff was 51 in a Korean hospital's electronic record (Baek and colleagues, 2014), 40 in obstetric and gynecologic wards (Mao and colleagues, 2024), and 55 in a Swiss hospital (Schwendimann and colleagues, 2006), and in a 2026 VA study the average total of the patients who did not fall was already above 45. The 51-point high-risk band traces to local and adapted versions, not to Morse's published scale. Document the total, the facility category, and the policy that defines it, record any clinical judgment override, and let the items that scored, not the tier, choose the measures.
No regulator sets a Morse interval, and none requires the Morse at all. CMS's long-term care rule requires supervision and assistance devices to prevent accidents and a care plan revised after a fall; the MDS collects fall history and falls with injury; the Joint Commission's 2026 hospital National Performance Goals require assessing and managing fall risk with interventions based on the population, the setting, and the individual patient's assessed risk, and its nursing care center, assisted living, and home care goal NPSG.09.02.01 requires assessment, interventions based on assessed risk, education, and evaluation; the OASIS-E2 item set asks whether falls prevention interventions followed a standardized multi-factor assessment; none names a tool or an interval. The triggers that the VA toolkit, AHRQ, Canadian accreditation practice, and the Australian guideline agree on are admission, transfer, a change in condition (a new sedating, opioid, or blood pressure medication, new confusion, a new or removed line, a change in mobility), and any fall. Every shift is a workflow many hospitals choose; chart it as your policy, name the trigger for each repeat, and make sure each entry links a scoring item to a measure, because a column of totals with no linked action is the defect reviewers find.
Follow the definition in your facility's approved build of the scale and chart what is actually present. The access item exists because a line or device tethers and encumbers a patient who tries to get up, and facility builds and training materials commonly treat a capped or locked peripheral device as access in place; do not invent a local exception because nothing is infusing, and if your build reads it differently, say so in the note. For a patient on bed rest or in a wheelchair, the mobility items rate how the patient actually moves, so a patient who does not walk can score low on them while transfers, toileting attempts, and delirium make the real risk high; the VA's falls toolkit notes that wheelchair users score low on those items even though transfers put them at risk. Score the items as observed, then document a clinical judgment override with its reason and the supervision and transfer measures it triggers, so the record shows the risk that the number cannot.
Because two items can move at once and cancel. A fall during the admission makes the prior-fall item score, so all else equal a post-fall total should rise; if it fell, another item must have resolved at the same time, most often a line removed or an aid no longer needed, and the note must say which item changed and why rather than leave a lower number to imply lower risk. The same logic runs the other way: a line removed on the day a new opioid makes a patient overestimate what she can do produces a lower total and a more dangerous patient, as the sample on this page shows. Write the changed items in clinical terms, keep the category attributed to policy, de-escalate only the measures tied to the item that resolved, add the measures the new item calls for, and record the trigger. After a fall, never copy the pre-fall score: perform a new assessment, in which the fall now scores, and attach the revised plan; an unexplained lower post-fall total should send you back to both entries to reconcile them.
Chart the offer, the patient's stated reason where appropriate and any capacity concern, the education given about the specific risk, the alternatives offered and which were accepted, who was notified under policy, and the plan as revised so it still mitigates the risk as far as it can; do not write that the patient refused fall precautions and stop there. In long-term care, CMS survey guidance expects the resident's choice to be respected while staff work with the resident to understand the choice and build a plan that honors it and mitigates the risk, and a signed waiver does not end the facility's duty; in hospitals the same documentation protects the record in a post-fall review. Family presence is an intervention, not a scored item: no Morse item represents a sitter or a relative, so a daughter staying overnight changes the supervision plan and never the number, and the note should say who is staying, for how long, and what they have agreed to do (call for staff, not lift). The sample on this page shows a declined bed exit alert documented with its alternatives and then accepted two days later when the risk changed.
Two documents, two purposes. The chart carries the clinical event: the time and how the patient was found, the injury check and vital signs, neurological checks where the head may have struck, the provider notification with time and response, imaging or treatment ordered, a new Morse in which the fall now scores, the contributing factors you can observe, the revised measures, the patient's and family's understanding, and the next reassessment. The safety report carries the organization's analysis: the event reconstruction, system factors, and the post-fall huddle or root-cause review, which AHRQ's toolkit treats as a separate activity from the clinical reassessment. Keep the incident narrative and any speculation about blame out of the progress note, and treat whether the note cites the report number as a local risk-management decision, because incident report privilege and negligence standards are state law. For coding, the injury diagnosis and its present-on-admission status come from the provider's documentation and drive the hospital-acquired condition payment provision for Falls and Trauma; your note supplies the time and circumstances. The incident report page covers the safety document itself.
Ask first. The scale is copyrighted: it was published by Morse, Morse, and Tylko in the Canadian Journal on Aging in 1989 and is set out in Morse's book Preventing Patient Falls (SAGE, 1997; second edition, Springer Publishing, 2008), and the Fall TIPS FAQ, maintained by the Brigham and Women's Hospital team whose toolkit is built on the scale, states that use of the Morse Fall Scale is free of charge but permission from the author is required, with requests directed to Dr. Morse at the University of Utah. No public statement grants blanket permission for websites, calculators, custom forms, or electronic builds, and the many pages that reproduce the anchors show market practice rather than a license. Modifying the scale is a second question: Morse's 2006 commentary addressed altered versions circulating under her scale's name, and a form with a medication item added or a rule that any in-facility fall makes the patient high risk is a local instrument whose validity is not inherited, so name it as a variant. For comparison, the Johns Hopkins Fall Risk Assessment Tool may not be used without a paid license from the Johns Hopkins Health System Corporation, and the Hendrich II Fall Risk Model is licensed and patented through AHI of Indiana. What needs no permission is the concept: fields for which items scored and the total, your own documentation prompts in original words, and the structure on this page, which reproduces none of the instrument.
Yes. Give it the facts (the trigger and version, which items scored and what each rests on, the total and any prior total, the facility category and policy, the measure in place for every scoring item, the teaching and any refusal with the alternatives, the notifications, and the next reassessment) and it drafts the full entry: trigger and version, items in plain words with the total, category attributed to policy, an intervention for each scoring item, teaching and response, and the reassessment and communication lines, ready for your review. It can also check a finished entry for a total with no items named, a cutoff written as the scale's rule, a score with no linked intervention, a bundle with no item behind it, items rated from the chart, a changed total left unexplained, or a post-fall entry with a copied score. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models.
The instrument facts and compliance claims on this page trace to these sources, last verified September 2026:
Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements.