Who Counts as Unwell Enough? The Severity Thresholds Behind NHS Talking Therapies

For many people in England, the first substantive contact with NHS Talking Therapies is not a conversation but a questionnaire. Before a course of therapy begins, there are scores to be taken, thresholds to be checked and a decision to be made about whether the service is the right one. The language used to describe that process — 'mild to moderate anxiety and depression', 'stepped care', 'complexity' — is clinical and administrative at the same time. It is also easily heard as a judgement about how much a person deserves help.

That misunderstanding is worth taking seriously. NHS Talking Therapies, formerly known as Improving Access to Psychological Therapies, was designed to widen access to evidence-based treatments for common mental health difficulties. Its eligibility language is not intended to rank distress. It is intended to match people to a service that can meet their needs. Yet the matching process has edges, and some people fall on the wrong side of them.

What 'mild to moderate' is meant to signal

In NHS Talking Therapies in England, 'mild to moderate' generally refers to common mental health difficulties that are distressing and impairing but not so severe or complex that they require specialist secondary care. Anxiety, depression, panic, obsessive-compulsive difficulties, post-traumatic stress and some specific phobias may all be treated within the programme, depending on local provision and clinical assessment.

The phrase is therefore a description of service fit, not a measure of personal resilience. Someone with a score in the 'moderate' range may be more impaired than someone with a higher score who has strong support and stable circumstances. Severity is one dimension among several: duration, functional impact, risk, previous treatment, physical health and social context all matter.

The scores that structure the conversation

The two measures most familiar to patients are the PHQ-9, a nine-item depression questionnaire, and the GAD-7, a seven-item anxiety questionnaire. Both produce a score that can be banded. On the PHQ-9, scores of 5, 10, 15 and 20 are commonly used as thresholds for mild, moderate, moderately severe and severe depression. On the GAD-7, scores of 5, 10 and 15 are commonly used for mild, moderate and severe anxiety.

These bands are useful. They give clinicians a shared language and allow services to monitor whether treatment is helping. They are not diagnostic tests. A score of 11 rather than 9 may change the route a referral takes, but it does not change the reality of the person sitting in the waiting room. Nor does it capture everything that makes life difficult: bereavement, debt, caring responsibilities, discrimination, insecure housing or a long-term physical condition may all shape distress without appearing neatly in a questionnaire.

Stepped care, and the problem with steps

NHS Talking Therapies operates through a stepped-care model. At the lower step, people may be offered guided self-help, computerised cognitive behavioural therapy or group programmes. At the higher step, they may be offered individual cognitive behavioural therapy, counselling or other evidence-based therapies. Specialist services sit beyond that for people with more severe or complex difficulties.

The model assumes that people can move up and down according to need. In practice, the steps can feel like cliffs. Someone may be told they are 'not unwell enough' for a particular intervention, or 'too complex' for a Talking Therapies service, without a clear explanation of what happens next. Both phrases can be accurate shorthand for a service boundary. Both can also leave a person feeling dismissed.

When 'too complex' is a signpost, not a verdict

'Complexity' is not a single thing. It may refer to significant trauma, personality difficulties, an eating disorder, bipolar disorder, active substance misuse, severe self-harm or a combination of physical and mental health needs. In some cases, a Talking Therapies service will still offer treatment alongside other support. In others, it will signpost to a specialist team because the clinical need is better met there.

The important question is whether complexity has been assessed or assumed. A referral can be turned down on the basis of a brief questionnaire and a short telephone call. That may be proportionate, but it is not the same as a full clinical assessment. Patients are entitled to ask what specifically led to the decision, whether another service is available and how a future referral would be handled if circumstances change.

When 'not unwell enough' still deserves a plan

Subthreshold symptoms are real. Someone may score below a service's cut-off and still experience disturbed sleep, low mood, avoidance, irritability or reduced concentration that affects work and relationships. In a stretched system, it is tempting to treat such difficulties as a matter for self-management. That is not always adequate.

A sensible response is not to tell someone to try harder. It is to ask what support is available outside the service: GP review, social prescribing, voluntary-sector counselling, workplace adjustments, self-management programmes or a later re-referral if things worsen. The absence of a Talking Therapies place should not mean the absence of a plan.

Questions worth asking before and after referral

Patients can ask precise questions without becoming adversarial. A useful starting point is to request clarity on the measures and thresholds used by the local service. It is reasonable to ask:

• Which questionnaires will be used, and what scores does this service use when deciding eligibility?
• Is the decision based on the score alone, or on clinical judgement as well?
• If I am outside the criteria, what alternative support is recommended, and who will oversee it?
• What does 'complexity' mean in this service, and has it been assessed rather than assumed?
• If my circumstances change, how do I re-refer, and how long is the wait likely to be?
• Can I see a copy of the referral and assessment, and correct any factual errors?
• What reasonable adjustments are available if I have a disability, caring role or communication need?
• If I disagree with a decision, what is the feedback or complaints route, including the Patient Advice and Liaison Service?

What the scores cannot see

Measurement-based care has real advantages. It can make services more consistent and help clinicians notice change. But a score is a snapshot of the past two weeks, not a life story. The PHQ-9 and GAD-7 cannot see the effort it takes to attend an appointment, the effect of a hostile workplace or the exhaustion of managing a chronic illness. They cannot decide whether someone is worth helping.

Severity language is unavoidable in a public service with finite resources. What matters is that it is explained, applied consistently and accompanied by a route forward when a service is not the right fit. Clear thinking about thresholds should not make patients feel they must prove their suffering. It should help them ask better questions — and help services answer them plainly.