What to do when a care-home resident keeps falling despite a plan

Recognise that another incident form is not a new plan

A resident has fallen three times in six weeks. Each event was recorded, and the care plan says they should use a walking aid and call for assistance. The latest fall happened while the aid was across the room and the resident was hurrying to the toilet. Repeating the same form and reminding staff to “monitor closely” would not answer why the plan is failing. The team needs a fresh assessment and a test of whether the promised support reaches the resident when they need it.

Respond to the latest fall using the home's clinical post-fall protocol. Check the person's condition, seek urgent assessment where indicated and document the event. The first-fall response covers that immediate pathway. This page begins with the broader pattern once immediate safety is addressed. Do not delay clinical help while analysing trends.

NICE NG249 says people in residential care settings should be offered comprehensive falls assessment and management. The guideline includes individual factors such as gait and strength, dizziness, cardiovascular findings, cognition, delirium, medicines, continence, footwear, vision and the environment. A repeat fall despite an existing plan is a strong prompt to ask whether the assessment is still current and whether the actions are effective. A numerical risk score alone cannot provide that answer.

Set an owner for the review and a near-term date. Bring the resident, care staff and relevant professionals into the discussion. A physiotherapist, occupational therapist, GP, pharmacist or falls service may each contribute according to the suspected factors. Record what remains uncertain. The home should not wait for a specialist appointment before making an obvious hazard safe or ensuring the current care plan is delivered.

Build a timeline that shows the pattern

Put the falls and near misses in order. For each event, record the date, time, location, activity, whether it was witnessed, the resident's condition, any injury, immediate response and what changed afterward. Include situations where the person nearly fell or needed sudden support. A chart can reveal that events cluster before breakfast, after a particular medicine, during bathroom visits or when a familiar worker is absent. It can also show no obvious pattern, which is itself useful.

Check the accuracy of the records. “Found on floor” is different from “tripped on carpet.” A worker's interpretation may have been copied into later reviews until it became treated as fact. Speak with the resident and witnesses. Ask what they experienced before falling, whether they called for help, where equipment was and what they were trying to do. Document the person's words and distinguish them from the team's hypothesis.

Overlay changes in health and routine. Has the resident recently returned from hospital, had an infection, experienced dizziness or changed mobility? Were medicines started, stopped or adjusted? Did continence needs, sleep or meal timing change? Has the room layout changed? Did a new walking aid arrive? A plan written before these changes may be obsolete even if it was thoughtful when created.

Look for the common conditions of the falls, not just the common location. Several falls in different rooms may all occur when the resident is rushing to the toilet. A sequence in one room may reflect poor lighting, but it may also reflect a new habit of getting up unassisted. The timeline should guide an assessment, not be used to declare one cause prematurely.

Test whether the existing controls were actually delivered

Read the current plan and observe a normal shift. If it says the walking aid should be within reach, is it there at night and after cleaning? If staff should respond to the call bell promptly, what happens during the busiest period? If the resident needs assistance to the toilet, is the rota able to provide it when they ask? If a sensor is part of the plan, is it switched on, positioned correctly and responded to? A control that exists on paper but fails in daily use is not an effective control.

Ask staff to explain the plan without prompting. Different answers reveal a handover problem. One worker may believe the resident can walk independently, while another insists on assistance. An agency worker may not know about the new aid. The manager should establish a clear, current instruction that respects the resident's choices and can be delivered across all shifts. An alert in an incident system cannot substitute for staff understanding.

Check whether promised referrals and reviews happened. Was the physiotherapy assessment completed? Did the pharmacist review medicines? Was a vision appointment arranged? Was the equipment delivered and adjusted? An action logged as “referred” is not the same as an assessment completed or a recommendation implemented. Record the status, next contact and interim support.

If the plan was delivered, ask whether it addressed the right factors. A walking aid may be correctly positioned but the resident may feel dizzy when standing. A call bell may be reachable but the person may avoid using it because they fear being a burden. The resident's explanation can change the intervention. Do not treat repeated falling as proof of non-compliance by the person.

Arrange a comprehensive reassessment

NICE NG249 recommendation 1.2.2 lists factors to consider in a comprehensive assessment, where appropriate. A healthcare professional or multidisciplinary team should decide what is needed for this resident. The care home can supply the timeline, baseline and observations. It should not perform examinations beyond staff competence or create a diagnosis from an incident chart.

Review mobility and strength. Has the person become weaker after illness or reduced activity? Is the walking aid suitable and used correctly? Is footwear secure? Does pain alter gait? A physiotherapist or occupational therapist can assess movement and equipment, with the resident involved in choosing practical supports. Any exercise or activity plan should match ability and preference and be monitored, not downloaded as a generic programme.

Review medical and medicine factors. Dizziness, blood pressure changes, vision problems, delirium, cognition, continence urgency and some medicines may contribute. The clinician or pharmacist can assess these and decide whether treatment should change. Care workers should not stop a medicine because a fall happened. Record the question, referral and professional answer. If the resident has new confusion or acute illness, escalate clinically rather than waiting for a routine falls meeting.

Review the environment and timing. Lighting, flooring, clutter, bathroom access, bed height and equipment positioning may matter. The assessment should consider how the resident actually moves at 03:00, not only a daytime demonstration. Check whether a proposed modification creates another risk. A lower bed or sensor may not be right for everyone, and bed rails require their own careful assessment. A device cannot replace a workable response from staff.

Agree interventions that preserve mobility and choice

The resident may value independence and be distressed by an instruction to wait for staff every time they stand. Explore what they want to do and what help they accept. NICE NG249 encourages discussion of safe movement, activity and ways to seek help in residential care. Preventing falls should not mean confining a person without a lawful, person-centred decision. Restriction can reduce strength, confidence and wellbeing.

Offer a combination of measures that addresses the assessed factors. This could include adjusted assistance times, a better-positioned aid, a clinical medicine review, lighting changes, footwear support or a referral for exercise and balance work. The exact combination is individual. Each action should have an owner, start date, review point and a way to test whether it happened. An intervention that depends on a worker being free at a particular time needs a staffing plan.

Discuss any alarm or sensor with the resident and relevant professionals. A movement alert may help staff respond but does not prevent a fall on its own. False alarms, delayed response and privacy concerns matter. Do not present technology as a guarantee of safety. Likewise, a “high risk” sticker or score has little value unless it changes the support the person receives.

If the resident declines an intervention, clarify their reasons and capacity for that decision. They may accept a less intrusive option. Record the discussion and professional advice. A capacitated choice can involve risk, and staff should work with the person to reduce avoidable harm while respecting that choice. A family member's preference is useful context but not an automatic override.

Make the plan measurable in everyday care

Write the revised plan as actions staff can follow. “Assist with mobility” is vague. “Offer support for the bathroom when the resident wakes, keep the aid within reach after evening care and check the call bell before leaving” is testable, if it reflects the assessed need and staffing. Specify what the person can do independently and what signs should trigger reassessment. Keep the instruction accessible in the normal care system, with the clinical assessment linked rather than buried in an incident folder.

Test the plan during the relevant shift. Ask a worker to show where the aid is and how they respond to a call. Ask the resident whether the arrangement works. Observe whether the action is possible when staff are busy. If it is not, revise resources or timing. A policy that requires continuous observation without providing staff cannot be relied upon as a control.

Track outcomes beyond the number of falls. Has the resident been able to move and participate as they wish? Are they more fearful? Have near misses changed? Are staff delivering the actions? Did a medicine review or therapy appointment happen? A short interval without a fall may be encouraging, but it does not prove the cause has been solved. Continue review at a frequency set by the care and clinical team.

If another fall occurs, compare it with the hypothesis behind the plan. Did the new action fail to happen, or did it happen but prove insufficient? Did a new factor appear? Use the answer to revise the plan, not to add another generic line saying “risk remains high.”

Keep the reporting decisions event specific

Each fall needs its own clinical and reporting assessment. A pattern of repeated falls does not automatically create one CQC notification for the series. A particular fall may cause a serious injury or other notifiable event. CQC's serious-injury guidance sets the criteria, and the registered person remains accountable for a required notification. Reassess if an injury becomes apparent later. Record the decision for each event.

Repeated falls can raise safeguarding questions if staff failed to provide agreed support, ignored clear risks or concealed incidents. That conclusion depends on facts and local safeguarding criteria. Seek advice or refer promptly when a concern is present; do not wait for a final root-cause report. Equally, do not label every fall in a frail resident as abuse. A careful review respects the person and enables fair learning.

Keep the resident and representatives informed according to the person's wishes and lawful sharing arrangements. Explain what the team has learned, what remains uncertain and how the plan will change. If a service failure caused harm, consider duty of candour requirements with appropriate advice. A communication log should show questions, responses and promised follow-up, not just that a call was made.

Commissioners or local partners may have reporting expectations under a contract. The manager should check them alongside statutory duties. A generic article cannot determine every local route. It can require the provider to name the decision maker, evidence and deadline.

Review service patterns without losing the resident's story

Several residents may fall at a similar time or place. The provider should look for service-wide factors such as staffing, lighting, bathroom access, floor maintenance or equipment availability. A group trend can identify a hazard that individual care plans miss. Keep the individual resident review as well. An aggregate graph cannot explain why one person falls when standing after a medicine change.

Use a consistent incident dataset so comparisons are possible: time, activity, location, witness status, injury, immediate action, suspected factors and outcome of actions. Mark unknowns rather than filling them with guesses. Review near misses as well as injuries. CQC inspection material shows that providers are expected to learn from incidents and put measures into care plans, but an inspection example is not a clinical protocol. NICE NG249 remains the main primary guidance for assessment and management.

Check whether the organisation's learning loop closes. Were the same environmental defects reported before? Did a manager approve an aid that was never delivered? Did staff receive the updated plan? Did a specialist recommendation reach the rota? A repeated pattern after “action completed” can expose a governance gap. Track the corrective action to an observed result.

Share lessons without publishing identifiable resident details more widely than needed. A staff briefing might say that night-time equipment positioning is being checked across the home, while the resident-specific clinical findings remain in their care record. Governance evidence should demonstrate action and follow-up without duplicating sensitive health information unnecessarily.

Example: three bathroom-related falls despite a walking-aid plan

A resident has three falls over six weeks. The plan says they should use a walking aid and call for help. The manager plots the events and finds that all occurred on early morning bathroom trips. Staff confirm the aid was sometimes left outside the room after cleaning. The resident says they avoid the bell because they do not want to wait. The latest fall receives immediate clinical assessment under the home's protocol.

The manager arranges a comprehensive falls reassessment. A clinician reviews possible dizziness and medicines; a therapist reviews transfers and the aid; the care team observes the night routine. They agree with the resident on an accessible aid position, a realistic early morning support arrangement and a way to call that the resident is willing to use. The plan states who checks the aid after cleaning and how staff respond. A supervisor tests it during the relevant shift and checks with the resident whether it helps.

The manager assesses injury, safeguarding and notification thresholds separately for each fall and records the decisions. Later review asks whether the actions occurred and whether falls or near misses changed. If another event happens despite delivered controls, the team reopens the clinical and environmental hypotheses. This example does not prescribe a specific therapy, observation period or restriction.

Where Complys may fit

The live Complys care home guide discusses incident records, actions and evidence of learning. It does not verify clinical falls assessment, gait measurement, movement-sensor integration or fall prediction. A product owner must confirm any proposed workflow claim before publication.

If available, an administrative system could link incidents to action owners, due dates and management checks. The resident's clinical assessment and care plan remain in the authorised care system. A useful governance view would show whether a referral occurred, equipment was delivered, staff understood the revision and the outcome was reviewed. It should not reduce the person to a risk score or automatically declare the plan effective.

CTA: Review Complys for Care and ask how current incident and action workflows can support follow-through while the clinical team owns assessment and care planning. Related tool opportunity: A repeat-falls review worksheet with event timeline, resident account, delivered versus planned controls, clinical referrals, revised actions and measured follow-up. It must not calculate a fall prediction or clinical diagnosis. Suggested internal links out: First-fall response, care home CQC guide, future pressure-ulcer owner when immobility or injury affects skin. Suggested links in: First-fall response after a recurring pattern is identified, broad care incident guide and future falls cluster page. Cannibalisation boundary: This page owns repeated falls despite a plan, including the pattern, delivery test, multidisciplinary reassessment and revised outcome review. Draft 78 owns the first immediate response after a fall without obvious injury. Neither should be split by room, injury type or resident age without a distinct clinical task.