MDQ PDF: Mood Disorder Questionnaire and Screening Documentation

Looking for an MDQ PDF? Open the Mood Disorder Questionnaire and scoring overview hosted by UW Medicine. The two-page file contains the form and instructions. The MDQ records a person’s reported history of symptoms associated with bipolar spectrum disorders. A positive screen calls for a comprehensive assessment; it does not establish a diagnosis.

For a usable clinical record, preserve the symptom count, whether symptoms occurred together, and the reported impact. Then document the clinician’s interpretation and the next responsible person. The examples below show why “MDQ 8” is an incomplete handoff.

For how the three parts combine into a positive screen, see our MDQ scoring guide.

Sources checked October 27, 2026. This guide covers adult MDQ results and follow-up documentation. In England, NICE advises against using questionnaires to identify bipolar disorder in adult primary care. NICE CG185, recommendation 1.2.3

Find the MDQ PDF and check reproduction terms

The UW Psychiatry Consultation Line resources page lists the MDQ under bipolar-spectrum resources. Use that page to find the source file again if a saved link stops working. Keep the questionnaire and its accompanying overview together so that the scoring method remains identifiable.

The UW-hosted form carries the notice “Permission for use granted by RMA Hirschfeld, MD.” Mental Health America’s MDQ source information states that use requires the author’s permission. For a new hosted copy, translation or software implementation, confirm permission covering that use and retain it with the implementation record. The original results worksheet below accompanies the linked questionnaire.

When selecting a copy, record the source URL, retrieval date, language and any version label printed on it. For a translated form, match the validation evidence to its language and assessment setting. Preserve the exact language administered separately from the language used to write the note. Mental Health America links a Spanish adaptation and its research.

MDQ scoring: keep all three components

The linked scoring overview requires all three conditions for a positive screen:

Use the scoring overview accompanying the PDF when checking the form. Identify any different scoring method explicitly rather than silently applying a modified rule to a standard-form result.

Why a negative screen can miss bipolar II

Miller and colleagues’ 2004 clinical validation compared MDQ results with DSM-IV-based SCID diagnoses in 37 patients with bipolar-spectrum illness and 36 with unipolar depression. Sensitivity was 0.69 for bipolar I and 0.30 for the combined bipolar II/not-otherwise-specified group. The latter figure is not an estimate for bipolar II alone. This small study reported a between-group comparison of P=0.06, so the difference is suggestive rather than conclusive. Lower self-rated severity on the impact question accounted for almost half of the false-negative results. The authors identified potentially weaker detection in milder bipolar-spectrum presentations and people with impaired insight. Miller et al., sensitivity and specificity study

When the clinical history raises concerns, a negative MDQ result cannot exclude bipolar disorder. Preserve the reported impact and relevant episode history for further assessment. Continue to apply all three components of the identified scoring rule.

For documentation, make each input inspectable. A structured field that stores only a number cannot show whether the other answers were supplied. Likewise, a blank response is not evidence that the patient selected “no.” If the information needed for classification is absent, preserve the incomplete state and assign clarification.

These fictional record examples make the difference concrete:

The count is the same in every row. The source information and resulting record are different. That is why an intake system should retain component responses alongside its classification.

What the validation studies can and cannot tell you

The original 2000 validation included 198 patients at five outpatient clinics primarily treating mood disorders. A research professional, blinded to the questionnaire result, conducted a telephone diagnostic interview using the bipolar module of the Structured Clinical Interview for DSM-IV. The study reported sensitivity of 0.73 and specificity of 0.90 at a symptom-count threshold of seven. These estimates describe that clinic sample and interview method. Hirschfeld and colleagues, original validation

The 2003 general-population study selected 711 people, stratified by MDQ score, from 85,358 adult survey respondents; 695 received the abbreviated SCID telephone interview. Reported sensitivity was 0.281 and specificity 0.972. The large survey population and the smaller diagnostic-interview sample are different denominators. Hirschfeld and colleagues, general-population validation

Sensitivity concerns detection among people with the reference diagnosis; specificity concerns correctly classifying those without it. Neither percentage gives an individual patient’s probability of having bipolar disorder after a positive result. Predictive value also depends on how common the condition is in the population assessed. These two studies support recording the setting and method, and they show why a negative screen cannot close a diagnostic question by itself.

MD-01: a complete screening record

Fictional adult outpatient mental-health example, October 27, 2026.

Source packet

• At 09:00, Avery completes the English form linked above without assistance. All 13 symptom items are answered; eight are marked yes. The same-period response is yes and the impact response is moderate.
• At 09:20, Avery tells the assessing clinician that the answers concern earlier periods. Avery can place one period approximately in 2024 but cannot yet give its duration. Today’s presentation has not been described in this packet.
• Avery agrees that the team may invite sibling Casey to contribute relevant history through the clinic’s approved process. Casey has not yet been contacted and has supplied no account.
• The clinician records that the screen warrants fuller diagnostic assessment. The supplied plan is to clarify the episode timeline, consider other explanations, and obtain the agreed collateral account. No diagnosis or medication decision is supplied.
• Care coordinator Morgan accepts responsibility for arranging the assessment. In this scenario, the team agrees to check booking progress at 15:00 on October 28. If no appointment is confirmed, Morgan will return the task to the duty clinician to decide the next step. This is the scenario’s agreed checkpoint.

Finished screening entry

27 October 2026, 09:20. MDQ completed by Avery at 09:00 using the English UW-hosted form; no assistance reported. All symptom items answered: 8/13 yes; same-period response yes; reported impact moderate. Positive screen under the accompanying three-part rule. Avery attributes responses to earlier periods and approximately dates one to 2024; duration remains to be clarified. The clinician plans further diagnostic assessment, including episode chronology and alternative explanations. Avery agreed to invite sibling Casey to provide collateral history; no collateral account received. Morgan accepted the assessment-booking task, with a progress check at 15:00 on 28 October and return to the duty clinician if booking remains unresolved. Diagnostic assessment and current safety-assessment findings are not supplied in this source packet.

This entry retains the source, all scoring components and an accepted next action. It keeps three states distinct: a positive screen, a planned diagnostic assessment and collateral history still awaited. Add current assessment findings and actions when the clinician supplies them, with their own date and source.

The biopsychosocial assessment example provides a broader structure for history, functioning and formulation. The MDQ result can be one attributed source within that assessment.

MD-02: when a missing answer becomes a false result

Fictional correction exercise. At 10:00, an imported form contains eight yes responses and moderate impact, but its same-period field is blank. An automated mapping populates the result as “positive” from the count alone. A generated narrative then changes that into “bipolar disorder confirmed.” Both transformations exceed the source.

At 10:20, the reviewer checks the scanned form and identifies the missing field. At 10:40, the patient clarifies that the endorsed experiences occurred in different periods and answers the co-occurrence question no. The clarification is a new dated response; it does not make the original blank field a completed answer retrospectively.

Reopen the destination after saving. A corrected paragraph does not prove that an imported diagnosis flag, result field or already-sent referral was corrected. Check each affected location and document who received an amended communication. Preserve the original source and the reason for change through the record system’s correction process.

Practice: what is wrong with “MDQ 8, bipolar ruled out after clarification, no follow-up needed”?

It hides the same-period and impact responses, turns a screening classification into a diagnostic exclusion, and deletes the clinician’s retained assessment plan. A faithful correction states the three component responses, dates the clarification, explains the changed screening classification and preserves the open assessment task. There is no supplied basis for “no follow-up needed.”

Lifetime history, current presentation and collateral sources

Keep the date of completion separate from the dates of reported experiences. “Completed today” does not mean “all symptoms present today.” An approximate date, uncertainty about duration and an uncontacted collateral source each belong in their own place. A later clarification can improve the history without changing what was known at the earlier visit.

NIMH explains that diagnostic assessment considers symptom severity, duration, frequency and lifetime history, with attention to family history and other possible causes. Thyroid illness and effects of prescribed or other drugs can resemble or influence mood symptoms. These considerations belong in the clinician’s reasoning; the questionnaire cannot resolve them. NIMH bipolar disorder overview

For example, distinguish “Avery reports a period in 2024” from “Casey independently confirms that period.” In MD-01 only the first statement is available. A permission to contact Casey is not evidence that contact occurred. The medical history form guide offers related prompts for recording source, uncertainty and later corrections.

If accounts conflict, preserve both with dates and attribution. Ask the clinician to document how the difference affects the assessment. Avoid resolving disagreement by selecting whichever account creates the most complete-looking record.

Copyable MDQ results documentation template

Copy these blank fields into your approved documentation environment. They are an original record structure, separate from the questionnaire. Complete only the fields supported by the encounter, and keep uncollected information identifiable.

The source form, screening classification, clinical interpretation and follow-up each have separate fields. This makes a saved record easier to check against the original material. In MD-02, the correction section would identify the count-only mapping, the original blank response, the patient’s later clarification and the retained assessment plan.

Keep completed forms and notes in the organization’s approved clinical systems. Use its access, retention and sharing controls; confirm the intended recipient before sending a result. The public copy button contains only blank prompts and asks for no patient responses. For wider record-handling questions, see the medical records software guide.

From screening result to completed follow-up

The record should explain why the next step was chosen and who accepted it. NICE’s specialist-assessment guidance includes a detailed episode history, functioning, comorbidity, treatment history and corroborative accounts. Its urgent-referral recommendation concerns suspected mania, severe depression or danger to self or others. Current urgent concerns require the clinical response pathway rather than waiting for a questionnaire or routine booking task. NICE CG185, recommendations 1.2.2 and 1.3.2–1.3.5

In MD-01’s fictional continuation, Morgan checks at the agreed time on October 28. The receiving service has acknowledged the request but has not offered an appointment. Morgan records the acknowledgment, keeps booking unresolved and returns the task to the duty clinician as planned. The clinician assigns Morgan a further contact attempt and an agreed next-day checkpoint; the duty coordinator accepts absence cover. No assessment is recorded as completed.

The example’s purpose is to show a closed communication loop around open clinical work. Someone has accepted the next action, yet the assessment still needs to occur. A handoff that says only “referral received” conceals that difference. Use the referral letter guide to separate sending, receipt, acceptance and completion in the receiving workflow.

Record patient communication as an event too: who explained the plan, when, by which agreed route and what remains outstanding. Do not mark “patient informed” merely because a message was drafted. A task’s owner and checkpoint should remain visible until its actual endpoint is reached.

Review the draft and the saved record

Before signing or transferring a result, compare it with the source:

1. Identity and provenance: correct patient, form, date, language and respondent; assistance attributed.
2. Completeness: answered symptom items distinguished from missing fields; no blank-to-no conversion.
3. Classification: all three components visible and matched to the identified scoring instructions.
4. Clinical meaning: screening result separated from diagnosis, current presentation and risk assessment.
5. History: earlier periods, current findings and collateral accounts kept distinct.
6. Follow-up: actual decisions, accepted owner, checkpoint and unresolved work preserved.
7. Destination: saved fields, narrative, attachments and tasks agree after transfer or correction.

These are original documentation checks. They can be applied to a manually written entry or an AI-assisted draft. A polished sentence can still be wrong if it changes an unanswered field into a negative response or turns “assessment requested” into “assessment completed.”

When evaluating a drafting tool, use synthetic inputs such as MD-02 and retain the source, raw draft, correction and saved result. Record observed errors and corrections. The clinical note examples show related source-fidelity checks; the AI clinical notes guide covers draft review and transfer. Vero publishes this guide and offers clinical-documentation software.