Sign Up
Login
Try for Free
Login
A medical source statement is a treating clinician's written opinion describing what a patient can still do despite mental impairment, rated across four work-related domains: understanding and memory, concentration and pace, social interaction, and adaptation. Clinicians complete one when a disability claim, most often Social Security, needs functional evidence from the treating source. Most combine a structured rating form with narrative comments, 2 to 6 pages.
Treating psychologists, psychiatrists, therapists, and other licensed clinicians
SSA disability examiners and judges, private disability insurers, claimant attorneys
600 to 1,200 words (form plus narrative) · 30 to 60 minutes by hand (clinical team estimate)
Structured functional-opinion form or narrative report (compare: capacity evaluation, FMLA certification, psychological evaluation report)
When a disability claim needs treating-source functional evidence, most often SSDI/SSI and private long-term disability
Opinion content is defined in SSA regulation; no law mandates a specific form, and HA-1152 is an appeals-stage tool
A medical source statement (MSS) is a treating clinician's written opinion about what a patient can still do despite mental impairment, and about impairment-related limits in the work-related mental activities the regulation names: understanding, remembering, and carrying out instructions; concentration, persistence, and pace; interacting with supervisors, coworkers, and the public; and adapting to workplace conditions and pressures (20 CFR 404.1513(a)(2)). The best-known version is SSA Form HA-1152, Medical Source Statement of Ability to Do Work-Related Activities (Mental), introduced by SSA's hearings office in 2002 for the appeals stage. You will also hear mental RFC, MRFC, mental impairment questionnaire, treating-source statement, and, from private disability insurers, attending physician statement.
One disambiguation saves most of the confusion: three different documents share the "mental RFC" nickname. A treating-source MSS is evidence you supply. Form SSA-4734-F4-SUP is an internal assessment completed by agency consultants, a "prior administrative medical finding," not a clinician's form. And the residual functional capacity determination itself is the adjudicator's finding, which no form decides. Since the rules changed for claims filed on or after March 27, 2017, a treating clinician's opinion gets no controlling weight; SSA weighs every opinion for supportability and consistency (20 CFR 404.1520c). An MSS is also not a full psychological evaluation report: the evaluation establishes diagnoses and test findings, while the MSS translates them into sustained work functioning, and unlike a capacity evaluation, which answers one specific legal decision, it addresses functioning across a workweek and a year.
Treating psychologists, psychiatrists, therapists, and counselors complete these statements when a patient applies for SSDI or SSI, usually at the request of the claimant's representative, and the same functional-opinion genre appears in private long-term disability claims as the attending physician statement and in state and veterans programs under other names. Pick the tool by the legal question. Job-protected leave runs through an FMLA certification; an employer's present-capacity question about a specific job is a fitness-for-duty evaluation; a benefits adjudicator asking about sustained work-related functioning over 12 months or more is asking for a medical source statement. Complete one when the chart already holds the evidence: dated scores, observed behavior, and a treatment relationship long enough to support longitudinal statements.
Claimant, clinician, and treatment relationship. Identify the claimant and claim, then establish yourself as the source: name, degree, license, specialty, and how long and how often you have treated. Under 20 CFR 404.1502 a licensed psychologist is an acceptable medical source, and other licensed clinicians acting within scope are medical sources whose opinions SSA evaluates. Pitfall: an unsigned or undated statement. SSA rejects proxy signatures, rubber stamps, and "dictated but not read" in the exams it purchases, and a reviewer discounts a treating-source statement with the same defects.
Diagnoses, symptoms, treatment, and response. State each diagnosis with its ICD-10 code, the clinical findings behind it, medication and therapy history, response, side effects, and prognosis. This is the "other medical evidence" that makes the ratings believable. Pitfall: letting the diagnosis argue the case. The regulatory question is what the claimant can still do; a diagnosis or a high score, standing alone, answers nothing about sustained work activity.
Understanding, remembering, and carrying out instructions. The first of the four work-related mental domains in 20 CFR 404.1513(a)(2): rate simple and complex instructions separately, address judgment, and attach the observed evidence for each rating. Pitfall: a checked box with a blank comments field. Under 20 CFR 404.1520c(c)(1), an opinion is persuasive in proportion to its supporting evidence and explanation; an unexplained "marked" carries little weight.
Concentration, persistence, and pace. Describe initiating, sustaining, sequencing, and completing tasks over a normal schedule, tolerance for ordinary pace, and recovery after interruption. Pitfall: treating off-task percentages and monthly-absence counts as required fields. They are conventions of attorney and vendor questionnaires, not elements of the regulation or the current HA-1152; give a number only when your records support one.
Social interaction. Address supervisors, coworkers, and public contact: accepting instructions, responding to criticism, managing conflict, and behavioral regulation, with the evidence for each. Pitfall: generalizing from the office. Cooperative behavior in a structured, supportive hour does not by itself predict a workweek; say whether the in-session presentation is representative and why.
Adaptation and workplace pressures. Cover response to change, ordinary work stress, decision-making, safety awareness, and independent planning. Pitfall: opining "disabled" or "unable to work." Disability is the adjudicator's conclusion; SSA instructs its own examiners to describe work abilities instead, and an ultimate-issue opinion gives the reviewer nothing to adopt.
Duration, fluctuation, and prognosis. Date the onset you can support, the duration already observed, the episodic pattern (frequency, length, recovery time, good-day and bad-day variation), and expected persistence, which matters because the statutory test looks at 12 months or longer. Pitfall: unsupported retroactive onset dates; tie every date to a record.
Supporting explanation and records reviewed. Link each limitation to findings, dated scores, observed behavior, treatment response, and collateral, and identify the records the opinion rests on. Pitfall: leaving a conflict with the longitudinal record unexplained. Consistency is one of the two most important regulatory factors, and an unaddressed contradiction is the standard reason a treating opinion is found unpersuasive.
MENTAL MEDICAL SOURCE STATEMENT (TREATING SOURCE) Claimant: [initials] Claim/file #: _________ Date: _________ Clinician: [name, degree, license #, specialty] Treatment relationship: first visit ____ last visit ____ Frequency and total visits: ____________________________________ Diagnoses (ICD-10): ____________________________________________ Clinical findings, treatment, and response: [medications, psychotherapy, dated measure scores, side effects] WORK-RELATED MENTAL ACTIVITIES (rate and give the evidence) 1) Understand, remember, and carry out instructions: [simple vs complex; judgment; evidence] 2) Concentration, persistence, and pace: [sustaining tasks over a normal schedule; evidence] 3) Social interaction (supervisors, coworkers, public): [evidence] 4) Adaptation and workplace pressures (change, stress, decisions): [evidence] Duration and course: [onset support, observed duration, episodic pattern, expected persistence, prognosis] What the claimant can still do: ________________________________ Basis: [records reviewed, testing, collateral] Dates: _________ Note: the agency decides disability; describe function, not the legal conclusion. Signature/credentials: __________________ Date signed: ________
Free to use and share, no signup. The PDF includes a one-page cheat sheet with section-by-section pitfalls and a pre-sign checklist; the DOCX is the blank template, ready to adapt.
Scenario: a treating psychologist completes a mental medical source statement for an SSDI claim at the reconsideration stage, at the request of the claimant's representative. Diagnoses are major depressive disorder and generalized anxiety disorder; treatment has run 14 months. All details are fictional.
Mental medical source statement · Program: SSDI, reconsideration stage · Prepared: 08/14/2026 at the request of the claimant's representative
Claimant: D.K., 41. Source: A. Whitmore, PhD, Licensed Psychologist (license on file). Treatment relationship: weekly to biweekly individual psychotherapy since 06/2025; 43 sessions through 08/2026. Records reviewed: my complete treatment record 06/2025 to 08/2026, PHQ-9 and GAD-7 logs, the primary care medication list, and a collateral interview with the claimant's spouse on 07/30/2026.
Diagnoses, treatment, and response: Major depressive disorder, recurrent, moderate (F33.1) and generalized anxiety disorder (F41.1), diagnosed at intake 06/2025. Treatment: weekly cognitive behavioral therapy with behavioral activation; sertraline prescribed by primary care since 09/2025 with partial response. PHQ-9: 21 (09/2025), 18 (01/2026), 16 (07/2026). GAD-7: 15 (07/2026). Attendance is consistent for afternoon appointments; three morning appointments were missed and rescheduled, documented as symptom-related.
1. Understanding, remembering, and carrying out instructions: D.K. understands and carries out simple one- and two-step instructions without support. Multi-step instructions require written prompts and re-explanation: in 9 of the last 12 sessions, three-step homework rationales had to be repeated (chart-documented), and D.K. maintains a written checklist for routine household sequences.
2. Concentration, persistence, and pace: Observed sustained task focus in session runs 15 to 25 minutes before redirection is needed. D.K. reports starting and abandoning household tasks most days, corroborated by the spouse's collateral report. Pace slows markedly under time pressure; a timed worksheet exercise on 06/18/2026 was abandoned at the halfway point with visible distress.
3. Social interaction: Cooperative and appropriate in the treatment setting. Tolerates brief, structured interactions such as store checkouts. After two extended family gatherings in 07/2026, D.K. experienced documented panic symptoms and two to three days of avoidance. Responds to perceived criticism with withdrawal lasting days, reported consistently across the treatment record.
4. Adaptation and workplace pressures: Routine changes, such as a rescheduled appointment or an unfamiliar building, produce documented anxiety spikes and avoidance. Decision-making under pressure deteriorates: D.K. defers routine decisions to the spouse during symptom exacerbations. No unsafe behavior observed; safety awareness is intact.
Duration, fluctuation, and prognosis: Symptoms at or near this severity are documented since intake, 14 months. Exacerbations lasting 2 to 4 days occur two to three times monthly, during which D.K. does not leave home and household functioning depends on the spouse. With continued treatment I expect meaningful but incomplete improvement over the next 12 months; the limitations described have already lasted more than 12 months and are expected to persist.
What the claimant can still do: Manage self-care and routine household finances with reminders; sustain simple, familiar tasks in low-demand settings for short periods; interact appropriately in brief, structured contacts.
Basis and limits: The findings above are drawn from 43 documented sessions, dated measure scores, and collateral, each cited by date in the record. The treatment setting is structured and supportive and may overstate workplace functioning; the longitudinal record is the better guide. Whether D.K. meets the statutory definition of disability is a determination reserved to the agency.
Signature: A. Whitmore, PhD, Licensed Psychologist · Date signed: 08/14/2026
This sample is fictional and for educational purposes. It does not describe a real patient, clinician, or claim.
Writing these after every session? BastionGPT drafts complete notes from bullets, dictation, or a transcript.
Generate a note from bulletsA medical source statement joins the claim file and the patient's designated record set: the patient can request it, opposing reviewers will read it, and it can resurface in later claims, hearings, or litigation, so write it as a document you can defend line by line from the chart. Two ownership points surprise clinicians. First, an opinion form completed at an attorney's request is not shielded as attorney work product: 20 CFR 404.1513 excludes the lawyer's legal analysis from the evidence but expressly keeps medical opinions and factual medical evidence in it. Second, retention follows the chart, and HIPAA sets no chart-retention period: state law and payer contracts govern, with APA record-keeping guidance suggesting at least seven years for adult records as professional convention. Keep the statement, the request that prompted it, and the fee agreement together in the record.
Sort the demands on this document into three piles before drafting. The four work-related mental domains and the "what the claimant can still do" frame are law, the definition in 20 CFR 404.1513(a)(2), and so is the weighing rule for claims filed on or after March 27, 2017, which makes supportability and consistency decisive. HA-1152 is agency practice for the appeals stage, policy rather than statute, and no rule makes it the exclusive format. Off-task percentages, monthly absence counts, and five-point severity scales are conventions of attorney and vendor questionnaires: useful when your records support them, safely left blank when they do not. The content is the requirement; the form is a convenience. A psychological evaluation report supplies the diagnostic and testing foundation that the statement's ratings then translate into sustained work functioning, and a capacity evaluation answers a different question entirely, one specific legal decision rather than work capacity over time. This page is general documentation guidance, not legal advice. Consult your attorney or licensing board; state rules vary.
The closest federal audit trail concerns the consultative examinations SSA purchases, the adjudicated neighbor of the treating-source statement, and its findings map directly onto the same failure modes. In the SSA Office of the Inspector General's July 2024 review of consultative examination oversight, 11 of 52 state Disability Determination Services oversight reports (21.2%) were missing required elements, and DDSs removed 47 CE providers from approved panels in fiscal year 2022, 44 of them for cause, with report-quality problems and late reports among the leading reasons. SSA's own report-review procedure (POMS DI 22510.020) checks completeness, internal consistency, documentation of evidence reviewed, and the examiner's actual signature. The BastionGPT Clinical Advisory Board sees the same errors most often in medical source statement reviews:
| Aspect | United States | Canada | Australia |
|---|---|---|---|
| Status | Opinion content is defined in law (20 CFR 404.1513(a)(2)); no statute mandates a specific form, and HA-1152 is an appeals-stage collection | No national mental-RFC form; CPP disability report content is set by law (CPP Regulations s. 68), with the principal report completed by a physician or nurse practitioner | No national mental-RFC form; DSP impairment is assessed under the legislated 2023 Impairment Tables, and the SA478 evidence checklist is expressly optional |
| Terminology | Medical source statement, mental RFC, mental impairment questionnaire; attending physician statement (private insurers) | CPP disability medical report (ISP-2519), Disability Tax Credit certification, attending physician statement (insurers) | DSP medical evidence, SA478 checklist, NDIS functional capacity report |
| What changes | The weighing rule turns on filing date: claims filed on or after March 27, 2017 get no treating-source controlling weight, and supportability plus consistency decide; SSA and private-insurer workflows differ | Physicians and NPs complete the principal CPP report, and a narrative may substitute for the form; psychologists provide supporting functional evidence and can certify DTC mental functions | Table 5 requires a medical practitioner's diagnosis, plus registered-psychologist evidence when the diagnosis is not from a psychiatrist; corroboration is statutory, and self-report alone is insufficient |
| Retention | No HIPAA chart-retention period; state law and payer contracts govern, with APA guidance suggesting 7+ years for adults as convention | Provincial; Ontario's psychology standard expects 10 years after the last professional contact for adult records (effective July 2024) | State and territory law; NSW requires 7 years after the last adult service and until age 25 for records of minors |
The four functional domains travel well as clinical content; the legal machinery around them does not. A US-style mental RFC questionnaire answers no Canadian or Australian statutory test, so match the document to the program before adapting it. For Australian clinicians writing functional evidence for support planning rather than income benefits, the NDIS functional capacity assessment page covers that distinct document family.
BastionGPT is specifically trained, tuned, and clinically tested on medical source statements.
See how clinicians use it day to day on the AI therapy notes page.
Many BastionGPT users report saving more than 90 minutes per day on documentation.
HIPAA-compliant with a signed BAA on every plan. Your data is never used to train models. BastionGPT drafts, you review and sign.
No regulation sets a length. Most treating-source statements run 2 to 6 pages: a structured form rating the four work-related mental domains plus narrative comments, or a narrative letter of roughly 600 to 1,200 words. SSA's paperwork estimate for its own HA-1152 form assumes 15 minutes per response, but that is a Paperwork Reduction Act assumption, not a measured completion time. Plan 30 to 60 minutes to do the supporting explanations well (clinical team estimate); the explanation, not the length, is what carries weight.
No. HA-1152, Medical Source Statement of Ability to Do Work-Related Activities (Mental), entered use in 2002 as a tool for administrative law judges and the Appeals Council, and no regulation makes it the exclusive format at any stage. SSA accepts a narrative letter, an attorney's questionnaire, or your own template, as long as the content addresses what the claimant can still do and the work-related mental limitations defined in 20 CFR 404.1513(a)(2). The familiar shape persists because reviewers can read it quickly.
Often, yes. For current claims, any licensed clinician acting within scope of practice is a "medical source" whose opinion SSA will evaluate. Licensed psychologists, like physicians, are also "acceptable medical sources," the narrower category that matters for establishing a medically determinable impairment. So a therapist's functional opinion is usable evidence, but the file still needs an acceptable medical source behind the diagnosis. Read the request's wording, too: some private insurers and programs direct their forms to specific professions.
Not for claims filed on or after March 27, 2017. Under 20 CFR 404.1520c, no medical opinion receives controlling or predetermined weight; supportability and consistency are the two most important factors. Claims filed before that date keep the older treating-source rule. The practical consequence: the explanation attached to each rating, and its fit with the longitudinal record, now do the work the treatment relationship used to do.
No. Disability is a conclusion the adjudicator reaches; SSA instructs its own consultative examiners to describe work-related abilities and not to opine on the statutory conclusion, and the same logic protects a treating-source statement. The same discipline applies to fields you cannot support: off-task percentages and monthly-absence estimates appear on many attorney questionnaires, but no regulation requires them and they are not on the current HA-1152. Complete them when your records support a number; leave them blank when they do not.
The legal question differs. An FMLA certification supports job-protected leave: a serious health condition, incapacity, and a leave schedule, usually over weeks or months. A fitness-for-duty evaluation answers an employer's present-capacity question about one specific job. A medical source statement answers a benefits question: sustained work-related functioning despite impairment, usually across 12 months or longer. Reusing one document for another purpose imports the wrong legal test, so adapt rather than forward.
There is no universal rule. In the US, state law, payer contracts, and your financial agreement govern; billers' forums show wide variation and no uniform coverage for standalone form completion, so set the fee policy before the request arrives. Service Canada pays up to CAD 85 toward the CPP disability medical report, with any excess potentially the patient's responsibility under the program's published process. Services Australia treats evidence gathered during a clinically necessary consultation as billable under the ordinary consultation framework, which does not make a standalone administrative report reimbursable.
Neither country uses the US mental-RFC model. Canada's CPP disability benefit requires a medical report whose content is set by regulation (CPP Regulations s. 68), completed under current instructions by a physician or nurse practitioner; psychologists supply supporting functional evidence and can certify the mental-functions route of the Disability Tax Credit. Australia's Disability Support Pension is assessed under legislated Impairment Tables: Table 5 requires diagnosis by a medical practitioner, with a registered psychologist's evidence required when the diagnosis is not from a psychiatrist, and the SA478 evidence checklist is expressly optional. For NDIS support planning rather than income benefits, see the NDIS functional capacity assessment page.
Yes. Give it your chart summary, measure log, and the request or form, and it drafts the statement around the four work-related domains with the supporting evidence attached to each rating, flags severity language your records do not support, and keeps the legal conclusion out. BastionGPT is HIPAA-compliant with a signed BAA on every plan, and your data is never used to train models.
Educational content, not legal or billing advice. Sample notes are fictional. Follow your organization's policies and your board, payer, and jurisdiction requirements.