Alcohol use disorder trials turn daily self-reported drinking into endpoints, check it against biomarkers, and watch liver safety and withdrawal. Capture holds the diary, the visit exams and the labs in one audit-trailed study. Test free, pay only at go-live.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Heavy drinking days, last 28 days
7 / 28
What an alcohol use disorder trial needs from its EDC
The study data
Timeline followback (TLFB) is the standard way to recover day-by-day drinking over a recall window, usually up to 30 days at a time. Interviewers prompt with calendar anchors, and each day gets a number of standard drinks. Trials then derive percent heavy drinking days, percent days abstinent, drinks per drinking day and responder categories. Heavy drinking is commonly defined as four or more drinks for women and five or more for men in a day, and gram thresholds appear in EMA and FDA guidance text with different values. Whatever your protocol says, put the definition on the form specification and the SAP.
The 2015 FDA guidance recommended two primary endpoint types, complete abstinence or no heavy drinking days. A 2025 summary reports that FDA has qualified a reduction of at least two levels on the WHO risk drinking level metric as a primary endpoint. Check the FDA source before relying on that wording. For the EDC, the point is that the daily grain must be preserved. A form that stores only "drinks last week" cannot be rederived into any of these endpoints. Capture lets a participant questionnaire collect daily entries, with a table section on site forms for interviewer-collected calendar days, so any endpoint can be derived from the export.
Missing days must stay missing. Dropouts, inpatient days and days a participant could not recall should be coded as such, not as zero drinks. Use a dropdown or single-choice code on each day for "no data" or "controlled environment", and add an edit check that raises a query when a day is left blank. Exports are wide CSV or Excel with an automatic data dictionary, so the analyst can see the coding.
Days covered by recall
TLFBNDays with no data
TLFBMISSMissing days stay missing, not zero
Heavy drinking days
HDDUsing the threshold set for this study
7 CalculatedBreath alcohol at visit
BACGamma-GT
GGTAbove reference range: confirm
Protocol to build
| Protocol element | What the data looks like | Where it lives in Capture |
|---|---|---|
| Daily drinking | Standard drinks per day over a recall window | Daily participant questionnaire, or interviewer calendar table on a site form |
| Severity and craving | Diagnostic criteria count; craving scale | Site form and participant questionnaire; licences stay with study documents |
| Biomarkers | Breath alcohol, PEth, urine drug screen | Site form with range edit checks; lab results template |
| Liver safety | ALT, AST, GGT, bilirubin by visit | Lab form with range rules and queries |
| Withdrawal assessment | Clinician-rated scale at detox | Site eCRF form with calculated total |
| Mood and suicidality | Depression and suicidality screens | PHQ-9 template and C-SSRS template |
| Behavioural therapy sessions | Session count, attendance, fidelity | Site form per session, or psychotherapy trial pattern |
| Concomitant substances | Tobacco, cannabis, other drugs | Concomitant medications template |
Capture does not supply instrument wording or verify licences. Clinician-rated forms have no licence controls in Capture.
Build a daily drinking diary and a TLFB calendar form, enter sample data, and export it to check your endpoints. Free sandbox, no credit card.
Operations
Detox and withdrawal change what the first week of a trial looks like. Many AUD trials require a short abstinence before randomisation, which adds a lead-in epoch. Clinician-rated withdrawal scales are repeated several times a day in inpatient settings; a repeating-row form with date-time and a calculated total keeps these together, and minutes between two date-times can be calculated automatically. Seizure history, delirium tremens and benzodiazepine use belong on the medical history and concomitant medication forms.
Liver safety is the second axis. Many participants start with elevated enzymes, so baseline needs to be clear, and changes relative to baseline have to be flagged consistently. Edit checks with the priority you choose raise automatic queries when a value breaks the rule. Cases that reach your protocol thresholds go through the serious adverse event workflow. An edit check is a data-management prompt, not a clinical alert.
Retention strategies include frequent short contacts. A daily diary on the participant's own phone, opened from a secure link without an app, doubles as the retention touchpoint and the data source. Reminders go by email or SMS. For combined designs with counselling, record session attendance and keep the pharmacotherapy and psychotherapy arms distinct. See EDC for smoking cessation trials for a close substance-use pattern, and EDC for behavioural health studies for the wider field.
Before first participant
Standard drink size, thresholds by sex, grams or drinks, written in the SAP and the form spec.
A distinct code for no data and controlled environment days.
Breath alcohol and PEth timing and acceptable ranges.
Baseline-relative thresholds and who reviews a query.
Lead-in epoch and scale frequency if inpatient.
Rating scales and craving measures documented in study files.
Yes, through a daily participant questionnaire on a schedule, or a calendar-style table on a site form for interviewer-collected days. Keeping the daily grain lets you derive heavy drinking days, abstinence and other endpoints from the export.
Calculated fields cover specific built-in formulas, so the level derivation belongs in your analysis programme. Capture stores the daily data and exports it with a data dictionary.
Use a distinct code for no data on each day and an edit check that raises a query on blanks. The coding appears in the data dictionary.
Yes, on a laboratory form with range edit checks that raise automatic queries. Clinical response procedures stay in the protocol.
Check the rights holder for terms. Capture does not supply instrument wording or verify licences; its licence controls exist only for participant questionnaires.
The sandbox is free with every feature. You pay only when you go live; plans are on the pricing page.
Free sandbox with every feature. No credit card, and you pay only when you go live.