Beyond PHQ-9 and GAD-7: choosing the right psychiatric rating scales

By Selam Koch, PMHNP-BC, founder of OneStep Scribe and practicing psychiatric nurse practitioner. Published July 2026.

The PHQ-9 and GAD-7 became the default psychiatric rating scales for a good reason. They are short, free, self-administered, and well validated, and they slot into almost any workflow. If you want the mechanics of documenting and trending those two, the PHQ-9 and GAD-7 documentation guide covers scoring, cadence, and the coding that goes with them. This article is about everything those two scales cannot see.

A depression scale measures depression. A generalized anxiety scale measures generalized anxiety. Run a full psychiatric panel and you are also managing mania, adult ADHD, post-traumatic stress, obsessive-compulsive symptoms, alcohol and drug use, antipsychotic side effects, and suicide risk. None of those is captured by a PHQ-9. Reaching for the same two forms out of habit is how a bipolar patient gets tracked with a depression-only instrument for a year. This guide is a practical map of which scale fits which target.

Three questions before you pick a scale

The instrument matters less than the reasoning behind it. Before you add a scale to an intake packet or an EHR template, answer three questions.

What decision will this score inform?

Match the scale to the job. A screener is tuned for sensitivity, so it accepts false positives to avoid missing cases, which makes a positive screen a prompt to evaluate and never a diagnosis. A severity scale quantifies how bad a known problem is now and trends it over time. A structured risk assessment organizes a clinical judgment; it does not predict the future. Using a screener as if it were diagnostic, or a one-time severity score as if it were a risk prediction, is the most common misuse I see.

Self-report or clinician-rated?

Self-report scales are efficient. The patient completes them in the waiting room or a portal, and they are excellent for trending because the same person answers the same items each time. Clinician-rated scales cost you interview minutes but capture severity a patient may minimize or lack insight into, which is exactly the situation in mania. The right choice depends on who has the more accurate view of the symptom.

Free or licensed?

Licensing status is not a footnote once a scale lives inside your EHR templates or an app you distribute. Some instruments are effectively in the public domain. Others are copyrighted and may require permission, registration, a fee, or use of only the official version, and some carry different terms for clinical versus commercial use. Status can also change, so verify the current license before you embed or reproduce any instrument, and keep the citation the author asks for. Treat the free-versus-licensed labels below as a starting point, not as legal clearance.

Scales by clinical target

Depression

The PHQ-9 remains the workhorse self-report depression scale, and it is freely available. When you want an alternative, the QIDS (Quick Inventory of Depressive Symptomatology) is a solid choice, available in both a self-report form and a clinician-rated form, and it maps cleanly onto the nine symptom domains of a major depressive episode. It is available at no cost for clinical use, though it is copyrighted, so confirm the current terms. Pick one and stay consistent, so the trend means something.

Anxiety

The GAD-7 covers generalized anxiety severity well and is free. Keep in mind what it does not measure. It is not a panic, social anxiety, OCD, or PTSD scale, even though anxiety features in all of those. If the clinical picture is panic or a phobia, a GAD-7 will trend the general anxious background but miss the target, so add a condition-specific measure when the diagnosis calls for one.

Bipolar disorder and mania

Bipolar management needs two different tools for two different jobs, and conflating them causes real harm. For screening, the MDQ (Mood Disorder Questionnaire) is the common self-report option: a short symptom checklist plus questions about symptom co-occurrence and resulting impairment. Be honest about its limits. The MDQ is a screener with modest positive predictive value, and performance drops for bipolar II and in lower-prevalence settings. A positive MDQ means "evaluate this person for a bipolar spectrum disorder with a full history and, when possible, collateral," not "this person has bipolar disorder." If you want the clinical reasoning that follows a positive screen, the bipolar versus borderline guide walks through the differential that so often sits underneath it.

For tracking manic symptoms once bipolarity is on the table, you need a mania-specific severity scale. The Young Mania Rating Scale (YMRS) is the widely used clinician-rated option, and the Altman Self-Rating Mania Scale (ASRM) is a short self-report alternative. Because insight is often impaired during an elevated episode, the clinician-rated YMRS frequently captures severity the patient will not report.

Adult ADHD

For adult ADHD, the ASRS v1.1 (Adult ADHD Self-Report Scale, developed with the World Health Organization) is the standard self-report screener, and its short Part A screening block flags who warrants a full evaluation. It is widely available at no cost, though the WHO terms of use are worth confirming. As with every screener, a positive ASRS is a doorway to a structured diagnostic assessment, developmental history, and functional impairment review, not a diagnosis on its own. The adult ADHD evaluation guide covers what that fuller workup includes. During treatment, track response with the specific target symptoms and functional outcomes you are trying to move, not the screener.

PTSD

The PCL-5 (PTSD Checklist for DSM-5) is the go-to self-report measure for post-traumatic stress. It has twenty items keyed to the DSM-5 symptom clusters, it produces a severity total you can trend across trauma-focused treatment, and it is distributed at no cost by the National Center for PTSD. Two cautions: it presumes an identified index trauma, so it belongs after you have established that history, and the commonly cited provisional cutoff is a starting point to validate in your own setting, not a diagnostic line.

Obsessive-compulsive disorder

OCD has well-validated measures, and I will describe what they do rather than reproduce any item wording, since the leading instrument is copyrighted. The clinician-rated standard pairs a symptom checklist with a short severity scale that rates the time consumed by, distress from, interference from, resistance to, and control over obsessions and compulsions, yielding separate obsession and compulsion scores. There are also self-report OCD symptom inventories that patients complete quickly and that work well for trending. Confirm licensing before building any of these into templates, and obtain the official instrument from the source rather than transcribing items from memory.

Substance use

For alcohol, the AUDIT (Alcohol Use Disorders Identification Test) is the WHO-developed standard: ten items screening for hazardous and harmful drinking and possible dependence, usable as self-report or a brief interview. Its three-item consumption subset, the AUDIT-C, is handy when you need a faster screen. For other drugs, the DAST (Drug Abuse Screening Test, often the ten-item DAST-10) screens for problematic drug use over the past year. Both are brief and widely used; confirm the current use terms for each, since availability and licensing differ by publisher and version.

Antipsychotic monitoring

Antipsychotics require two monitoring streams that a mood scale will never cover. The AIMS (Abnormal Involuntary Movement Scale) is a brief clinician-administered exam for involuntary movements and is the standard way to screen for and track tardive dyskinesia. It is in the public domain. Run it at baseline and on a recurring, risk-driven schedule, and verify the current frequency recommendations for your agent and population. The second stream is metabolic monitoring: weight and body mass index, waist circumference, blood pressure, fasting glucose or HbA1c, and a lipid panel, obtained at baseline and on a defined follow-up schedule per current consensus guidance. Neither is a symptom rating scale, but a number you did not draw is a side effect you did not catch.

Suicide risk

The C-SSRS (Columbia-Suicide Severity Rating Scale) is a structured way to assess suicidal ideation and behavior. It is copyrighted, so I will not reproduce its items; instead, understand what it structures. It walks through the presence and severity of ideation, its intensity, and the presence and type of suicidal behavior, giving a consistent framework for asking and for documenting. Official versions are available at no cost for many settings, often with registration, and some uses call for training, so verify the terms for your context and use only the official version. Treat it as a clinical assessment aid that organizes your judgment, not as a test that predicts who will act. For how the assessment itself gets written into the chart, see the suicide risk assessment documentation guide.

The compact reference

This table pairs each target with a common instrument, what it measures, and a reasonable cadence. Cadence is a default to adjust to the patient and the setting, not a rule, and free-versus-licensed status must be verified before use.

Documenting a scale result the right way

A score is only as useful as the sentence around it: it needs a date, an interpretation, and a link to the decision it drove. Here is a short, fictional excerpt showing an MDQ and a PHQ-9 documented in a way that will not mislead the next reader, using the all-caps section heading the note structure uses.

Making it a system, not a habit

Scales earn their keep when the right one is repeated on a sensible cadence and read against the last result, which is the whole idea behind measurement-based care. That is also what makes the documentation stronger. When payer criteria are met, the administration and scoring of a brief, patient-completed standardized instrument (for example the PHQ-9, GAD-7, ASRS, PCL-5, AUDIT or AUDIT-C, DAST-10, MDQ, or ASRM) may be separately reportable, and the 96127 billing guide walks through those criteria. Clinician-rated scales that you administer during the visit, such as the YMRS, AIMS, C-SSRS, and the clinician-rated Y-BOCS, are generally part of the evaluation and management or psychiatric service and are not separately billed under a brief-assessment-instrument code. Verify the specific code and payer rules before you bill any of it.

Rating scales quantify; they never replace the narrative. A YMRS total tells you the mania is severe, but your mental status exam is what shows it. For language that makes an abnormal exam specific rather than generic, the MSE vocabulary bank is the companion to any of these numbers.

This article is educational and reflects one clinician's approach. It is not clinical, legal, or billing advice for any specific patient, and it does not guarantee any outcome or reimbursement. Rating-scale licensing, validated cutoffs, and monitoring intervals change over time and vary by version and setting, so verify the current terms, guidance, and citation for every instrument before use. Coding and payer rules vary and must be confirmed with the payer and a certified coder. The patients described are fictional. Clinical judgment stays with the treating clinician. CPT is a registered trademark of the American Medical Association. Always verify requirements with current clinical guidelines, your payers, and your compliance advisors.