Care Home Dependency Tools Explained

This guide explains how dependency tools are used to plan and evidence staffing in care homes in England. It is general guidance on compliance documentation, not clinical advice or a staffing prescription. Decisions about staffing levels for a specific service remain yours. References to legislation are current as of the last reviewed date above.

A dependency tool is a structured way of scoring how much support each resident needs, so that the total support need in your home can be converted into staffing hours. It answers the question CQC actually asks at inspection: not "how many staff do you have?" but "how did you decide that number, and how do you know it is still right?"

Most small homes already make this judgement well. The gap is usually evidential. The manager knows that Room 4 now needs two carers for transfers and that nights got heavier after two admissions in March — but none of that is written down in a form anyone else can follow. A dependency tool turns professional judgement into a record.

What a dependency tool actually does

A dependency tool breaks each resident's needs into categories — typically mobility and transfers, personal care, continence, nutrition and hydration, medication, cognition and behaviour, communication, and night-time need. Each category is scored, the scores are totalled, and residents fall into bands such as low, medium, high, or very high dependency.

The banding then maps to indicative care hours. Add the hours across all residents, add the time that is not resident-specific (handover, medication rounds, cleaning where care staff cover it, training, breaks), and you have a staffing requirement you can defend.

The output is not a rota. It is the evidence base underneath the rota.

The legal basis: Regulation 18

There is no legal minimum staffing ratio for care homes in England. There is no rule that says one carer per six residents, or any other number. What the law requires is sufficiency, judged against the needs of the people you actually care for.

Regulation 18(1) of the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014 states: "Sufficient numbers of suitably qualified, competent, skilled and experienced persons must be deployed in order to meet the requirements of this Part."

Three words in that sentence do the work. Sufficient is relative to need, so it changes as your residents change. Suitably qualified, competent, skilled and experienced means the mix matters, not just the headcount — three staff who cannot use the hoist are not three staff for a hoist-dependent resident. Deployed means on the floor where the need is, not merely on the payroll.

Because the standard is relative, the only way to show you meet it is to show what you measured it against. That is the dependency tool's job.

Regulation 17(2) then requires systems to "assess, monitor and improve the quality and safety of the services provided…" and to "assess, monitor and mitigate the risks relating to the health, safety and welfare of service users and others…" A dependency assessment that is completed once and never revisited satisfies neither.

Which tool should you use?

There is no CQC-approved or mandated dependency tool. Providers use a range of them, and CQC's interest is in whether yours is appropriate for your service and consistently applied — not which brand it carries.

In practice homes use one of three routes:

A published sector tool. Several sector bodies, local authorities, and commissioning groups publish dependency and staffing tools. Check the licence terms before adopting one, and check it was designed for residential or nursing care rather than another setting.
A tool built into your care system. Many care management platforms score dependency from care-plan data. Convenient, but confirm you can export the scoring and the rationale — you need to hand CQC the evidence, not a login.
Your own structured assessment. Perfectly acceptable, provided the categories are relevant, the scoring is consistent between assessors, and the link from score to hours is written down.

Whichever route you take, avoid reproducing a copyrighted clinical scale inside your own paperwork without permission. Reference it and use it under its terms instead.

Using the tool so it holds up

The tool is the easy part. These habits are what make the evidence credible.

Assess every resident on admission, and re-assess on change. A dependency score taken at admission and left alone describes a resident who no longer exists. Re-assess after any significant change — a fall, a hospital discharge, a new diagnosis, weight loss, a change in continence or cognition, a new behaviour that needs two staff.

Re-run the whole home on a fixed cycle. There is no regulatory-set frequency — this is a practical judgement, not a compliance minimum. A monthly cycle suits most homes; a stable service with low turnover may reasonably stretch to quarterly. What is not defensible is "when we remember."

Record the conversion, not just the score. The step inspectors most often find missing is the arithmetic between total dependency and rostered hours. Write down the hours each band attracts and the non-resident-specific time you add. Someone who has never seen your home should be able to follow it.

Write down when you overrode the tool. Tools inform judgement; they do not replace it. If the tool said 4.5 care staff for the late shift and you rostered 5 because two residents were end-of-life, record that reasoning. A documented override is a strength. An undocumented gap between tool output and rota is a finding.

Show what happens when the numbers move. If dependency has risen for three months and staffing has not, the tool has become evidence against you. Show the escalation: what you raised, with whom, and what changed.

What CQC asks about staffing

Expect questions in three layers. First, the method — how do you calculate staffing, and can you show it? Second, the currency — when was this last reviewed, and does it reflect the residents here today? Third, the triangulation, which is where most homes come unstuck: your dependency data will be read against call-bell response times, accident and incident logs, staff and resident feedback, and the actual rota including agency use.

If the tool says one thing and the incident log says another, the incident log wins. Dependency evidence is strongest when it is visibly connected to the rest of your governance rather than filed on its own.

Your dependency data also feeds the Provider Information Return, which asks about staffing and dependency directly, and it is standing evidence for the Well-Led key question. For how that fits the wider governance picture, see our guide to Well-Led evidence.

Common weaknesses worth checking this month

Scores completed by different assessors with no shared definition of "moderate" or "high"
No documented link between total dependency and rostered hours
Night-time need scored as a fraction of daytime need, with no separate assessment
Dependency reviewed on paper but never discussed at governance meetings
Agency and bank use rising while dependency scores stay flat — a sign the tool is not being updated

Reviewing the tool as part of a scheduled audit cycle keeps it current. Our guide to running a care home audit covers the method, and the free audit schedule generator will put a staffing and dependency review on the calendar alongside your other audits.

This is general guidance based on the published Health and Social Care Act 2008 (Regulated Activities) Regulations 2014 and CQC guidance current at the date above. There is no mandated dependency tool and no statutory staffing ratio in England — requirements can change, so always check the current guidance on cqc.org.uk and the regulations on legislation.gov.uk for your specific situation. Not legal advice.

Sources

Dependency scores, rotas, audits, and the evidence tying them together tend to live in different places until the week CQC calls. InspectReady keeps them organised and current in one place, built for independent care homes rather than enterprise chains. Start your free trial — 14 days free, no card required.