ABC charts for behaviour: how to record so the patterns actually show

An ABC chart records three things about a behaviour that challenges: what happened just before it (the antecedent), exactly what the person did (the behaviour), and what happened immediately afterwards (the consequence). Done well, and read together over weeks, a stack of ABC charts shows what the behaviour is for, which is the question every positive behaviour support plan has to answer. Done badly, they are a drawer of forms that say agitated, hit out, calmed down, and tell nobody anything.

The short answer

The ABC chart is the main data tool of positive behaviour support. Each chart is one incident. The antecedent column says what was going on in the minutes before, in enough detail that a reader could picture it. The behaviour column describes what the person did in observable terms, with how intense it was and how long it lasted. The consequence column says what happened next: what staff did, what the person got or avoided, and how the episode ended. A good chart also records the setting events for that day, the time, the place and who was present. Charts are analysed, not filed: someone reads them weekly, looks for what the antecedents and consequences have in common, and takes that to the PBS review. The rest of this guide takes each column in turn, gives a worked example, and shows how to analyse a month of charts in half an hour.

What an ABC chart is and where it comes from

The antecedent behaviour consequence chart comes from applied behaviour analysis. The idea is that behaviour is shaped by what comes before it and what follows it. If a person hits out and the demand they were resisting is withdrawn, the hitting has worked, and it will happen again. If a person shouts and a member of staff comes and sits with them, the shouting has worked. The chart captures those sequences so that the pattern can be seen and the environment changed.

In care homes the ABC chart is used for anyone whose behaviour challenges: people with learning disabilities, autistic people, people living with dementia, people with mental health conditions. It is required by most positive behaviour support plans and expected by NICE guideline NG11 as part of the functional assessment. It is not an incident form. An incident form records that something happened and what harm resulted, for reporting and safeguarding. An ABC chart records why, for understanding. Serious incidents need both.

At a glance: what goes in each column

Setting events: the column most charts do not have

A standard three-column ABC chart misses the thing that most often explains the behaviour: the background conditions that made the person more vulnerable that day. Poor sleep, pain, constipation, hunger, a change of staff, a missed visit, a noisy morning, a new resident, a change of medication. These are setting events. They do not trigger the behaviour directly but they lower the threshold at which a trigger will.

Add a setting events box to the top of the chart and fill it in every time, even if the answer is none obvious. Better still, record setting events on the daily log for every day, whether or not anything happened, because then the analysis can compare days with incidents to days without. In my experience the single most common finding when charts are analysed properly is that most incidents happened on days with two or more setting events present, and the plan then shifts toward preventing those events rather than managing the behaviour.

The antecedent: what happened just before

The antecedent is everything in the minutes before the behaviour. Where was the person, who was there, what was going on, what was said and by whom, what was asked of them, what changed. Write it as a scene, with the time and the place. The person reading it later should be able to see the room.

Common antecedents worth looking for: a demand or request, a transition from one activity to another, a change of plan, noise, crowding, someone standing too close, being ignored, another resident's behaviour, waiting, a word or tone that has meaning for the person. Also record what was not happening: the person had been alone for forty minutes, nobody had spoken to them since lunch.

Do not write became agitated as an antecedent. Agitation is the beginning of the behaviour, not what caused it. Go back further. What was happening when the agitation started? That is the antecedent.

The behaviour: observable, specific, countable

Describe what the person did in words that two people watching would both agree on. Not aggressive: hit toward J's face with an open hand, once, made contact with her shoulder. Not self-injurious: hit the side of his head with the heel of his hand, six or seven times. Not absconded: walked fast to the front door and pulled the handle for about thirty seconds. Not verbally abusive: shouted the same phrase repeatedly, and record the phrase.

Record the sequence if there was more than one behaviour. Record whether the person appeared to be targeting anyone. Record any injury to anyone, and cross-reference the incident form and body map. Avoid interpretation in this column. The person was frustrated goes in the antecedent or the analysis, not here.

The test is countability. If the behaviour is described specifically enough that you could say how many times it happened this month, it is described well enough.

Intensity and duration

A behaviour that happens ten times a month at low intensity for thirty seconds is a different problem from one that happens twice a month at high intensity for twenty minutes. The chart needs both measures or the analysis cannot tell them apart.

Intensity is best recorded on a simple scale defined for the person in their PBS plan, so that everyone uses it the same way. A five-point scale might run from one, a single low-force action with no contact, to five, sustained force causing injury requiring treatment. Write the definitions on the chart or the plan and train staff on them with examples.

Duration is from the first behaviour to the point the person was calm again, in minutes. Record it honestly. Ten minutes that felt like an hour is ten minutes.

Frequency comes from counting the charts, which is why every incident needs one, however minor. A month with three charts and twenty incidents that nobody wrote up is a month with no data.

The consequence: what happened next

The consequence is what happened immediately after the behaviour, and it is where the function of the behaviour shows. What did staff do? What did the person get, or get away from? How did the episode end? What did the person do afterwards?

Be honest here even where it is uncomfortable. If a member of staff shouted, write it. If the demand was dropped, write it. If the person was given a drink and a biscuit and sat with for half an hour, write it, because that may be exactly what the behaviour was for. The consequence column is not a record of staff performance; it is a record of what the behaviour achieved, and the analysis needs the truth.

Record which strategies from the PBS plan were used and which were not. Record any PRN medication and cross-reference the MAR. Record when the person was calm and what they did next, because that is often the clearest sign of function: a person who goes to their room and closes the door after an outburst in a noisy lounge is telling you something.

Time, place, people and who completed it

Every chart records the date, the time the behaviour started, the location, who was present including other residents, and who completed the chart and when. The completion time matters: a chart written at the end of a shift six hours later is a different quality of evidence from one written within the hour. Aim for within an hour, and never later than the end of the shift.

Where more than one member of staff saw the incident, the one who was closest completes the chart and the other reads and countersigns it. Where the accounts differ, both are recorded. Where the person themselves can say what happened, their account goes on the chart too, in their words, and it is often the most useful column of all.

A worked example

Tom is 29, autistic, with a learning disability. His PBS plan says head-hitting is usually escape from noise or from a cancelled plan. Here is a chart completed properly.

Setting events: slept about four hours (night log). No bowel movement for two days (bowel chart). Agency worker J on shift, first time with Tom. Mum phoned at 10:00 to cancel Sunday visit.

Antecedent: 14:20, lounge. TV on at high volume, three other residents present, one of them shouting at the TV. J came in and stood directly in front of Tom, about a metre away, and told him the minibus was not coming this afternoon. Tom had been humming and rocking for about two minutes before she spoke; J had not seen the amber signs.

Behaviour: hit the side of his head with the heel of his hand six or seven times over about thirty seconds. J stepped closer and said his name twice. Tom hit toward her face once with an open hand and made contact with her shoulder. No injury. Intensity 3 of 5. Total duration about two minutes.

Consequence: J stepped back and left the room. M, who knows Tom, turned the TV off and asked the other residents to move to the dining room. Tom walked to his room and closed the door. Nobody followed. Quiet after five minutes. Came out at 15:00 and asked for squash. Strategies used: noise reduced, one person, stepped back. Not used: headphones not offered, timetable not shown, what next card not offered. No PRN.

Completed by M at 15:10, countersigned by J. Incident form completed for the contact with J.

What the example tells the reader

Read as a single chart, that tells the analyst four things without any guesswork. Three setting events were present. The amber signs were there for two minutes and were not acted on because the worker did not know them. The trigger was a cancelled plan delivered by someone standing close in front of him, both of which are named triggers in his plan. The behaviour stopped when the noise stopped and he could leave, which is consistent with the escape function in his functional assessment.

The actions that follow are not about Tom. They are: agency workers are introduced to Tom's amber signs before they work with him, cancellations are shown on the timetable rather than announced, and the too loud button is practised more. One chart, properly written, changes the plan. A vague chart changes nothing.

Analysing a stack of charts

Once a month, or weekly for someone with frequent incidents, take every chart for the period and read them together. This takes about half an hour for twenty charts if they are written well. Work through the following.

1. Count the incidents by behaviour, and record the frequency, the average intensity and the total duration. Compare with last period.
2. Sort the charts by time of day and day of the week. Look for clusters.
3. List every setting event mentioned and count how many charts have each one. Compare with days when no incident happened, using the daily log.
4. List every antecedent and group them: demands, transitions, noise, proximity, cancellations, waiting, other residents. Count each group.
5. List every consequence and group them: demand withdrawn, left the room, given attention, given an item, PRN, nothing changed. Count each.
6. Look at which PBS strategies were used and which were not, and whether incidents where the plan was followed were shorter or less intense.
7. Write a short summary: the behaviour, the frequency and trend, the most common setting events, the most common antecedents, the most common consequences, and what that suggests about function.
8. Take the summary to the PBS review with two or three proposed changes.

The point is the summary. Twenty charts in a drawer are data. A page that says most head-hitting follows cancellations announced verbally, on days with poor sleep, and ends when he can leave the room, is information.

Frequency charts and the trend

Alongside the ABC charts, keep a simple count per behaviour per week or month, and plot it. A line going down is the evidence that the PBS plan is working. A line going up is the evidence that something has changed and the plan needs review. A line that is flat with a plan that has been in place for six months is a plan that needs rethinking.

Plot intensity and duration as well as frequency where you can, because a plan can reduce the number of incidents while the ones that remain get worse, and that is a different problem. Show the graph to staff. People who fill in charts and never see the result stop filling them in. If you want to see how a monthly frequency view and the underlying charts look in one place, book a demo and bring a month of your own charts to compare.

ABC charts and the PBS plan

The chart is the evidence; the plan is the decision. The functional assessment in the PBS plan is based on the charts, and the review of the plan reads the charts to check whether the assessment still holds and whether the strategies are being followed. Every ABC chart should therefore be linked to the person's plan, and the plan should say who reads the charts and how often.

The PBS plan example shows how the charts feed the review. In a well-organised record, the ABC charts sit beside the plan so that the reviewer can filter them by antecedent, by consequence, by strategy or by time of day without retyping anything. A system that holds ABC charts and PBS plans in the same record saves the half-hour of sorting; on paper, a summary sheet with a row per chart does the same job.

ABC charts, incident forms and body maps

An ABC chart is not an incident report, and the two must not be confused. The incident report records that an event happened, who was hurt, what was done, and who was told, for the purposes of reporting, safeguarding and notification. The ABC chart records the sequence for the purposes of understanding. A serious incident needs both, and each should reference the other.

Injuries are recorded on a body map, with a photograph where appropriate and consented, in the daily log. The ABC chart references the body map rather than describing the injury in detail. A daily log with body maps and photos makes this straightforward; on paper, the body map is filed with the incident form and the chart notes its date.

Where an incident involves restraint or any restrictive intervention, that is recorded separately again, with the lawful basis, the duration and the debrief, and the ABC chart notes that it happened. The incident reporting guide covers the reporting side.

ABC charts in dementia and mental health settings

The chart is the same in any setting, but the antecedents and functions differ. In a dementia unit, the commonest antecedents are personal care, unfamiliar staff, pain, noise, and being unable to find something or someone; the function is often escape from distress or a communication of pain, and the analysis should look for health causes first. A chart that shows a sudden increase in distress during personal care over a week points to a urinary tract infection or a pressure area before it points to a behaviour plan.

In mental health services, antecedents may include contact with family, medication changes, news, anniversaries, and the person's own thoughts, which they may be able to describe. The person's own account is essential and the chart should have space for it. Function may be about seeking help, expressing distress, or testing whether staff will respond, and the consequence column is where that shows.

ABC charts on paper and on screen

A paper ABC chart works if the form has the right columns, including setting events, intensity, duration and strategies used, and if the completed charts are kept together per person and summarised monthly on a single sheet. The weakness of paper is the analysis: sorting twenty charts by antecedent means reading all twenty every time, and the summary sheet depends on someone having the half hour.

A digital chart, entered on a phone or tablet within the hour, carries the time stamp and the author automatically, prompts for each column so nothing is left blank, and can be filtered by antecedent, consequence, strategy, time of day or worker in seconds. Kiwi keeps ABC charts inside the person's record beside the PBS plan and the daily log so the setting events are already there; any system that does the same will change how often the analysis actually happens. The record of the analysis and the changes it led to is the evidence an inspector asks for, whichever format holds it.

Reading the chart with the person

Where the person can take part, the chart is stronger for it. After the episode has passed, and only when they are ready, ask them what happened and what they wanted, and write their answer in their words. Some people can say exactly what they were trying to escape or get; some can point to a picture; some will tell you the next day. Their account often corrects the staff account, and it is the most direct evidence of function there is.

Doing this also changes the relationship. A person whose behaviour is recorded about them, in a form they never see, experiences the chart as surveillance. A person who is asked what happened and sees their words written down experiences it as being listened to. In mental health services in particular, that difference matters, and the daily notes should reflect the same approach.

Training staff to complete ABC charts

Nobody writes a good ABC chart without being shown one. Training should use a real, anonymised chart written badly and the same incident written well, and ask staff to spot the difference. It should cover the person-specific intensity scale, the setting events list, and the rule about completing within the hour. It should explain what the charts are for and show the analysis, so staff can see that what they write changes the plan.

Make it routine to look at the charts in team meetings: here are this month's numbers, here is what they show, here is what we changed. When staff see their charts being read, the quality of the charts goes up.

The ABC chart checklist

• Date, start time, location, who was present, who completed it and when.
• Setting events for the day, or none identified.
• Antecedent written as a scene: what was happening, who said what, what changed.
• Behaviour in observable, countable terms, in sequence.
• Intensity on the person's agreed scale and duration in minutes.
• Consequence: what staff did, what the person got or avoided, how it ended, what they did next.
• PBS strategies used and not used.
• Any PRN, cross-referenced to the MAR.
• Any injury, cross-referenced to the incident form and body map.
• The person's own account where they can give one.
• Completed within the hour and never later than the end of the shift.
• Countersigned by a second witness where there was one.

Common mistakes

• Vague behaviour. Aggressive, challenging, kicked off. Cannot be counted, so cannot be analysed.
• Antecedent starts with the agitation. Became upset, then hit out. What happened before he became upset?
• Consequence is calmed down. How? What did staff do? What did he get? That is the whole point.
• No setting events. The single biggest predictor is missing from every chart.
• Charts only for big incidents. Minor episodes are not recorded, so the frequency is wrong and the pattern is hidden.
• Written at the end of the shift. Six hours later the detail has gone.
• Never analysed. A drawer of charts and a review that says continue current plan.
• Blame in the consequence column. Staff stop writing what they actually did, and the data becomes fiction.

What good looks like on inspection day

The inspector asks how the home understands a resident's behaviour. The manager brings up the PBS plan and the ABC charts for the last quarter. The charts are specific: the inspector can picture each incident from the antecedent column. Each has setting events, intensity and duration, and strategies used. The monthly analysis is a page with numbers and a trend, and the last one led to two changes in the plan, both dated.

The inspector asks a care worker to describe the last incident they recorded. The worker describes the room, what was said, what the person did, and what they did in response, and it matches the chart. The inspector asks what the person's behaviour is for and the worker says escape from noise, mostly, and that is what the charts show. The inspector asks whether minor incidents are recorded and the worker says every one, because the numbers matter.

That is inspection-ready evidence: not a home where nothing happens, but a home that knows exactly what happens, why, and what it is doing about it.

Final conclusion

An ABC chart is a small form that does a large job. It works when the antecedent is a scene, the behaviour is countable, the consequence is honest, the setting events are recorded, and someone reads the charts together and writes down what they show. It fails when any of those is missing, and most charts in most homes are missing at least two. Fix the form, train the staff with real examples, analyse monthly, and take the summary to the PBS review. The behaviour will not stop because you recorded it well. But you will finally know why it happens, and that is where every good plan begins.

Frequently asked

What is an ABC chart for behaviour?

An ABC chart records one incident of behaviour that challenges in three parts: the antecedent (what happened just before), the behaviour (exactly what the person did) and the consequence (what happened immediately afterwards). Read together over time, the charts show what the behaviour achieves for the person, which is the basis of a positive behaviour support plan.

What should be written in the antecedent column of an ABC chart?

Everything that was happening in the minutes before the behaviour: time, place, who was present, what was said and by whom, what was asked of the person, what changed. Write it as a scene the reader could picture. Do not write became agitated; that is the start of the behaviour, and the antecedent is what happened before it.

What are setting events on an ABC chart?

Background conditions that made the person more vulnerable that day without directly triggering the behaviour: poor sleep, pain, constipation, a change of staff, a cancelled visit, a noisy morning. Standard three-column charts leave them out. Adding a setting events box, and recording them daily on the log, is usually the single biggest improvement to the analysis.

How soon after an incident should an ABC chart be completed?

Within the hour where possible, and never later than the end of the shift. Detail disappears quickly, and a chart written six hours later is much weaker evidence. The chart records who completed it and when, and a second witness countersigns where there was one.

How do you analyse ABC charts?

Take all the charts for the month, count incidents by behaviour with intensity and duration, sort by time and day, list and count the setting events, antecedents and consequences, check which PBS strategies were used, and write a one-page summary of what the pattern suggests about function. Take that summary and two or three proposed changes to the PBS review.

Is an ABC chart the same as an incident report?

No. An incident report records that an event happened, who was hurt, what was done and who was told, for reporting and safeguarding. An ABC chart records the sequence for understanding. A serious incident needs both, each referencing the other, with injuries recorded on a body map.

Should minor behaviours be recorded on ABC charts?

Yes. If only serious incidents are charted, the frequency is wrong and the pattern is hidden. Every episode of a behaviour named in the person's PBS plan gets a chart, however brief, so that the monthly count is real and the trend can be seen. A quick chart for a thirty-second episode takes two minutes.

Sources

• NICE guideline NG11 Challenging behaviour and learning disabilities: prevention and interventions
• BILD and the PBS Academy: Positive behavioural support competence framework
• CQC: Right support, right care, right culture guidance for providers supporting autistic people and people with a learning disability
• Restraint Reduction Network training standards
• Skills for Care: positive behaviour support resources for adult social care
• Challenging Behaviour Foundation: information sheets on understanding challenging behaviour and ABC recording