{{ signal.label }}
{{ signal.source }}
{{ signal.message }}
{{ signal.action }}
| Score | Zone | Band | Intervention |
|---|---|---|---|
| {{ zone.rangeLabel }} | {{ zone.whoZone }} | {{ zone.label }} | {{ zone.interventionLabel }} |
AUDIT-10 is a screening instrument. Use the score with qualified professional judgment for diagnosis, withdrawal planning, or treatment decisions.
These cues explain why the current follow-up lane looks the way it does and where discussion is most likely to matter first.
{{ signal.message }}
{{ signal.action }}
| Focus area | Why it stands out | Suggested follow-up |
|---|---|---|
| {{ row.label }} | {{ row.reason }} | {{ row.action }} |
Questions 1 to 3 reflect consumption, Questions 4 to 6 reflect dependence symptoms, and Questions 7 to 10 reflect alcohol-related harm and outside concern.
| Domain | Score | Share | Band | What stands out |
|---|---|---|---|---|
| {{ domain.label }} | {{ domain.total }}/{{ domain.max }} | {{ domain.percentLabel }} | {{ domain.bandLabel }} | {{ domain.readout }} |
Use the item chart to see which answers are contributing most of the total and whether the pressure sits in heavy episodes, dependence symptoms, or harm-related items.
| # | Question | Answer | Score | Domain | Signal |
|---|---|---|---|---|---|
| {{ row.id }} | {{ row.text }} | {{ row.answer }} | {{ row.score }}/4 | {{ row.domainLabel }} | {{ row.signalLabel }} |
Alcohol screening is meant to turn a vague impression about drinking into something easier to judge. Frequency, quantity, loss of control, memory gaps, injuries, and outside concern do not all carry the same meaning, so a structured screen is useful when someone wants a clearer picture than “probably fine” or “probably too much.” This package uses the ten-item Alcohol Use Disorders Identification Test, usually shortened to AUDIT-10, to build that picture from the past year.
Once all ten answers are entered, the page calculates a total score from 0 to 40, places that score into one of four coded bands, and breaks the result into three grouped views: consumption, dependence, and harm. It also adds a gauge, a short result guide, higher-scored versus lower-scored item comparisons, and an answered-question table that can be copied or exported for later discussion.
That makes the tool useful in a few realistic situations. Someone can use it before a primary-care visit to summarise patterns they have struggled to describe out loud. A counselor or educator can use it as a teaching example of how screening differs from diagnosis. A person trying to cut down can use it as a structured before-and-after snapshot instead of relying on memory alone.
The result still needs restraint. AUDIT-10 is a screening instrument, not a diagnosis of alcohol use disorder and not a safety clearance. A lower total does not make intoxicated driving, injuries, blackouts, or withdrawal risk unimportant. A higher total does not tell you exactly what treatment is needed without clinical context.
This implementation also adds interpretation layers that belong to the package rather than to the core questionnaire. The three grouped subtotals are real question clusters, but the package’s none, mild, moderate, and severe subscore labels are internal grading bands derived from each cluster maximum. They are useful as a quick reading aid, but they should not be mistaken for official diagnostic cut points.
Answer with the last year in mind, because that is the recall frame stated in the tool itself and in the original AUDIT design. If you answer only from a recent holiday, dry month, or bad weekend, the result can look sharper or safer than your typical pattern really is.
The best first pass is to read the result in layers. Start with the total score and risk band. Then look at the grouped subtotals for consumption, dependence, and harm. After that, check the higher-scored items. That sequence tells you whether the main signal is frequent drinking, trouble stopping, or fallout such as guilt, memory loss, injury, or outside concern.
One more practical caution matters here because the package stores responses in the page state. Sharing a completed link can recreate the answer pattern. That design keeps scoring local to the browser, but it also means privacy depends partly on how you handle copied URLs and exports afterward.
AUDIT-10 uses ten items scored from 0 to 4, but not every item uses the same wording pattern. Questions 1 and 2 use drinking-frequency and typical-quantity ranges, questions 3 through 8 use frequency-style response ladders, and questions 9 and 10 use a three-level injury or concern structure scored as 0, 2, or 4. The package sums the chosen values directly, so the total score is simply the aggregate of all ten item scores.
The tool then groups the same answers into three domain-style clusters. Questions 1 to 3 form the consumption subtotal, questions 4 to 6 form the dependence subtotal, and questions 7 to 10 form the harm subtotal. Those grouped views help explain what is driving the total, but the package also adds its own quick-grading labels by dividing each subscale against its maximum and applying quartile-style boundaries.
The total banding is explicit in the code. Totals up to 7 are labeled Low Risk, 8 to 15 are Hazardous, 16 to 19 are Harmful, and 20 or above is Possible Dependence. The gauge, summary badges, and guide all use those same thresholds, so interpretation stays consistent across the result surfaces.
Item-level interpretation is also package-specific. The guide treats scores of 3 or 4 as higher-scored drivers, scores of 0 or 1 as lower-scored anchors, and then builds next-step suggestions from the total band plus whichever grouped subtotal stands out most. That is useful for reflection, but it is still a secondary interpretation layer built on top of the basic AUDIT scoring model.
| Result element | Range or rule | What the package does with it |
|---|---|---|
| Total score | 0 to 40 |
Drives the main badge, gauge, and risk band. |
| Low Risk | T <= 7 |
Shows the lowest band and maintenance-style advice. |
| Hazardous | 8 <= T <= 15 |
Prompts cut-down and tracking advice. |
| Harmful | 16 <= T <= 19 |
Prompts more urgent follow-up wording. |
| Possible Dependence | T >= 20 |
Shows the highest coded band and recommends professional assessment. |
| Higher-scored drivers | Items with score >= 3 |
Surface in the guide as the strongest current concerns. |
| Lower-scored anchors | Items with score <= 1 |
Surface as comparatively lighter areas in the same response set. |
| Grouped view | Questions used | Maximum | Package-specific interpretation note |
|---|---|---|---|
| Consumption | 1 to 3 | 12 | Package labels the subtotal from none to severe by quarter-range grading. |
| Dependence | 4 to 6 | 12 | Helps show loss-of-control and morning-drink signals inside the total score. |
| Harm | 7 to 10 | 16 | Shows guilt, memory loss, injury, and outside concern as one cluster. |
| Pattern flag | Highest group by margin | Not a numeric score | The guide calls the result Balanced, a leading cluster, or Low overall burden. |
The total band is the headline, but it is not the whole interpretation. Two people can land in the same range with different risk shapes. One may score mainly through frequent drinking and quantity. Another may score less on quantity but more on inability to stop, guilt, blackouts, or concern from others.
The most useful trust check is to compare the total band against the grouped subtotals and the higher-scored items. If the total seems only mid-range but questions about injury, memory loss, morning drinking, or outside concern carry weight, let those item-level details slow you down before you treat the result as routine.
| Output pattern | What it usually means here | What it does not mean |
|---|---|---|
| Higher consumption, lower dependence and harm | Drinking amount or frequency is the main driver right now. | It does not prove there are no consequences or no risk escalation ahead. |
| Dependence or harm clearly exceeds consumption | Control problems or consequences are doing more work than raw quantity. | It does not prove dependence by itself, but it is a stronger reason to seek assessment. |
| Low overall total with a yes-style response on injury or outside concern | A narrow but important problem signal is present. | It does not mean the result is safely dismissible because the total stayed low. |
The false-confidence mistake with AUDIT-10 is to treat the screen as a complete diagnosis or a guarantee of safety. The corrective step is simple: check whether the grouped profile and the highest-scored items match the story you are telling yourself. If they do not, trust the mismatch enough to slow down and get follow-up.
Someone scores often on drinking frequency and quantity but stays lower on guilt, injury, and concern from others. The total can still land in the Hazardous range because the screen is sensitive to repeated high-volume use, not only to visible fallout. In that case the grouped view helps explain why the result rose before the harm block did.
Another person may not drink most days, yet still score meaningfully on not being able to stop, failing expected duties, or needing a morning drink. The total and the dependence subtotal together tell a different story than quantity alone. That pattern is exactly why the package separates the grouped views instead of leaving only one overall number.
If the gauge and result card never appear, the usual cause is not a scoring error. One or more of the ten questions is still unanswered, so the assessment stays in progress mode. The corrective path is to use the question list, complete every item, and then let the page calculate the full result.
No. The highest band is labeled Possible Dependence because this is a screening result, not a standalone diagnosis.
The package follows the AUDIT item structure for those injury and concern questions, which use a three-level yes or timing pattern rather than a five-step frequency ladder.
Treat the grouped totals as helpful breakdowns, but treat the none, mild, moderate, and severe labels as package-defined reading aids rather than official cut points.
The tool only scores complete ten-item runs. If you do not see the final result, return to the question list and fill any unanswered item.