Skip to content

Using the HAM-D in clinical practice

The HAM-D gives a clinician-rated, interview-based measure of depression severity. Learn when it adds value in practice, how to score it, and when a self-report measure fits better.

Use the Hamilton Depression Rating Scale (HAM-D) when a trained clinician needs a structured, interview-based rating of depression severity. Typical uses are a specialist assessment or documenting how symptoms are changing. For routine monitoring between visits, a self-report questionnaire such as the PHQ-9 is usually more practical. The two answer different questions, so neither score converts into the other.

What does the HAM-D measure?

Max Hamilton published a rating scale for depression in 1960, and it became the reference measure in antidepressant trials. A clinician rates each item from an interview and from observation, which differs fundamentally from self-report measures like the PHQ-9.

The 17-item version covers mood (depressed mood, guilt, suicidal ideation, work and activities), anxiety (psychic and somatic), sleep (early, middle, and late insomnia). It also covers somatic symptoms (gastrointestinal, general, genital, and weight loss), and other features (agitation, retardation, hypochondriasis, and insight).

Which HAM-D version should you use?

The HAM-D exists in several forms, including 6-, 17-, and 21-item versions and structured interview variants such as the GRID-HAM-D. They are not interchangeable. Record the exact edition in research and clinical notes, and compare only ratings made with the same one.

Survey Doctor uses the 17-item form hosted by the University of Florida, whose anchors follow the 1976 ECDEU manual version and which rates the past week. It does not calculate a HAM-D6 score or administer the GRID-HAM-D.

How do you administer the HAM-D?

A HAM-D rating is only as consistent as its interview. Raters should follow a documented training and structured interview process, and keep that process the same whenever scores will be compared.

Tell the patient which time frame you are rating, and record it with the edition. Cover every item rather than skipping those that seem irrelevant. Rate observation-based items from the interview itself, and consider whether a physical symptom has another cause before you score it.

How do you score and interpret the HAM-D?

Nine items are rated from 0 to 4: depressed mood, guilt, suicide, work and activities, retardation, agitation, both anxiety items, and hypochondriasis. The other eight are rated from 0 to 2: the three insomnia items, the gastrointestinal, general somatic, and genital items, weight loss, and insight. The 17 ratings add up to a total from 0 to 52.

Survey Doctor uses these ranges from Zimmerman and colleagues (2013), who studied 627 outpatients with major depressive disorder at one practice:

ScoreSeverity
0-7No depression
8-16Mild depression
17-23Moderate depression
24-52Severe depression

These ranges are not rules from Hamilton's original scale, and other publications use different boundaries, so state the convention you use. Trials commonly define response as a drop of at least 50% and remission as a total of 7 or below. A 2010 study of 846 inpatients largely supported those conventions, although on the 17-item form a total of 6 or below matched clinician-judged remission best.

The total also carries the scale's weighting:

Insomnia has three items, so severe sleep disturbance can raise a total well above what the other symptoms suggest.

Anxiety items contribute substantially, so anxious depression can score higher than its depressive symptoms alone would.

Somatic items can raise the total in patients with medical illness.

For repeated ratings, record the absolute and percentage change, but do not treat either as proof that treatment caused it. Compare only ratings made with the same edition, interval, interview method, and trained-rater process.

How should you handle the suicide item?

Any rating above zero on the suicide item (item 3) needs clinical attention, whatever the total.

If you need help now. In the US, Call 988 or text 988. Call 911 if you are in immediate danger. Outside the US, contact your local emergency number or find support in your country.

Item-level ratings can show whether sleep, anxiety, retardation, or agitation drove the total. They do not select medication or another treatment by themselves.

What are the HAM-D's strengths and limits?

Bagby and colleagues' 2004 review of 70 psychometric studies described the HAM-D as the gold standard for assessing depression for more than 40 years, and asked whether that should continue. They found adequate internal reliability and adequate convergent and discriminant validity. But many items contributed little to measuring severity, some had poor interrater and retest reliability, and content validity was poor.

Clinician rating adds interview and observation, but it is not automatically more accurate than self-report. Rater training, interview structure, and patient communication all affect results. The interview also takes far longer than a self-report questionnaire, which limits how often a busy practice can use it.

When should you use the HAM-D or an alternative?

The HAM-D adds value in a few settings:

Specialist evaluation. A clinician-rated inventory can add structured information when a specialist reviews a complex course. It does not confirm treatment resistance or authorize a specific treatment.

Documenting symptom patterns. Item ratings can support clinical notes when symptoms or functioning change.

Observed signs. The interview captures psychomotor change and behavior that a patient may not report. It should complement the patient's own account, not override it.

For routine monitoring, the PHQ-9 is usually more practical. The MADRS is another clinician-rated scale with a different item emphasis, and the CESD-R is a self-report option with a different symptom-group algorithm. When anxiety is the focus, use a measure built for it, such as the HAM-A (score guide) or the GAD-7. Choose by validated purpose rather than assuming one score converts to another.

How do you build the HAM-D into practice?

Decide when a HAM-D rating answers a useful clinical question rather than setting a fixed universal schedule. If another instrument is used between ratings, keep its trend separate from the HAM-D trend.

Before using the HAM-D routinely, train every rater, compare ratings periodically to check reliability, use a structured interview, and watch for drift over time. Record the total, each item score, the time frame, the edition, and the administration method so that later ratings can be compared. For patients who see their own result, the HAM-D score guide explains the ranges in plain language.

Track your mental health

Create an account to explore published assessments, automatic scoring, and score history

View plans

Get Survey Doctor updates

New assessments, features and articles, a few times a year. Unsubscribe at any time.

Survey Doctor provides published mental health questionnaires for informational purposes only. We work with physicians, but we are not your physician. A screening score is not a diagnosis, and nothing here is medical advice. Always talk to a qualified healthcare provider about your health.

Survey Doctor is operated by TeleEMG, LLC.