PHQ-9 and GAD-7 in private practice: scoring, RCI and when to re-rate

The PHQ-9 scores 0–27 (nine items, 0–3 each); the GAD-7 scores 0–21 (seven items, 0–3 each). Add all items for the total. Clinical cut-points are well established, but the score alone tells you less than the change over time — and whether that change is statistically reliable rather than noise.

Scoring the PHQ-9

Each of the nine items asks how often the client has been bothered by a symptom over the past two weeks: 0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day. Add all nine items for a total between 0 and 27.

The standard severity bands used in NHS Talking Therapies and widely adopted in private practice are:

• 0–4 Minimal or no depression
• 5–9 Mild depression
• 10–14 Moderate depression
• 15–19 Moderately severe depression
• 20–27 Severe depression

A score of 10 or above is the most commonly used threshold for a clinically significant presentation. Item 9 — thoughts of self-harm or being better off dead — should always be reviewed individually regardless of the total score; your safeguarding and risk protocols apply.

Scoring the GAD-7

Seven items, same 0–3 frequency scale, two-week recall window. Total 0–21.

• 0–4 Minimal anxiety
• 5–9 Mild anxiety
• 10–14 Moderate anxiety
• 15–21 Severe anxiety

A score of 8 or above is often used as a clinical threshold in primary care; 10 or above is the NHS Talking Therapies cut-point for caseness. As with the PHQ-9, the total is a starting point for clinical conversation, not a substitute for it.

Why the score alone isn't enough

A client who moves from PHQ-9 19 to PHQ-9 14 has dropped five points. That looks meaningful — but is it? All psychometric tools carry measurement error. The Reliable Change Index (RCI) tells you whether a score change is large enough to be a genuine shift rather than random fluctuation.

For the PHQ-9, the widely cited reliable change threshold is 5 points or more (Jacobson & Truax method, using published test-retest reliability and standard deviation figures from Kroenke et al., 2001). For the GAD-7, reliable change is typically 4 points or more (Spitzer et al., 2006 data). Both thresholds appear in NHS Talking Therapies guidance. These citations should be independently verified before publication.

Recovery, in NHS Talking Therapies terms, means moving from above the clinical threshold to below it and achieving reliable improvement. Private practitioners aren't obliged to use that exact definition, but it's a clinically coherent benchmark and useful when communicating outcomes to GPs or insurers.

When to re-rate

There's no single rule, but common practice in UK services is:

• Baseline: before or at the first session, before any intervention begins
• Mid-treatment: around session 4–6 for a short-term contract, or every 4–6 weeks for open-ended work
• End of treatment: at or just before the final session
• Follow-up: 1–3 months post-ending if your contract allows

Re-rating every session can feel burdensome and may sensitise clients to symptom monitoring in unhelpful ways. Re-rating too rarely loses the signal. For most private-practice contracts of 8–20 sessions, baseline, mid-point and end-of-treatment covers the essentials.

If a client's score rises significantly between reviews — particularly if PHQ-9 item 9 changes — that warrants immediate clinical attention and documentation.

Presenting outcomes to clients

Sharing scores can strengthen the therapeutic alliance and support collaborative goal-setting, but framing matters. A graph of scores over time is usually more meaningful than a number in isolation. Presenting it as "one way we can notice together whether things are shifting" positions the measure as a tool rather than a judgement.

Some clients find repeated symptom questionnaires distressing or feel reduced to a number. It's reasonable to use measures less formally, or to supplement them with idiographic tools such as personalised outcome measures, where that fits the work better.

Limitations and trade-offs

The PHQ-9 and GAD-7 were developed and validated primarily in primary care populations, not across the range of presentations common in private practice — complex trauma, personality difficulties, neurodivergence, or personal-development work. They measure symptom frequency, not meaning, functioning, or quality of life. A client doing important therapeutic work may score similarly or higher at mid-treatment than at baseline. Treating the score as the primary indicator of progress in complex or longer-term work can distort clinical thinking. The WSAS (Work and Social Adjustment Scale) is a useful adjunct for measuring functional impact alongside symptom severity.

Documenting outcomes in private practice

For insurers, GPs, and your own clinical governance, a clear record of baseline and follow-up scores — with dates, who administered the measure, and any clinical response to significant changes — is good practice. Referencing objective outcome data in letters to GPs or insurers may support the clinical picture. Your supervision log is also a reasonable place to record how you're using outcome measures and any reflections on their fit with your practice.

Where Sorca fits

Sorca sends PHQ-9, GAD-7, and WSAS questionnaires to clients via the free companion app between sessions, includes RCI tracking, and produces a monthly printable outcomes report with a CSV export — useful for insurer reporting or clinical governance reviews. Nothing enters a client record until you save it. There's a 3-day free trial at sorca.life — no card required.