For decades, primary care medicine has asked patients to account for their own drinking—through questionnaires and direct conversation—trusting that self-knowledge and honesty would surface those most at risk. A decade of Swedish health records now gently challenges that trust: a blood biomarker called PEth, which accumulates in proportion to alcohol consumed over recent weeks, reveals hazardous drinking in nearly one in seventeen patients who had already been cleared by standard screening. The finding is less an indictment of any single tool than a reminder that human self-report and biologic
Blood biomarker test catches hazardous drinking missed by questionnaires
Nearly 6 percent of patients cleared by standard screening showed hazardous alcohol in their blood.
Mark
So the AUDIT questionnaire misses nearly 6 percent of hazardous drinkers? That seems like a small number. Why does it matter?
Mimi
Because in a busy primary care practice, that 6 percent compounds. If a doctor sees 200 patients a year, that's roughly 12 people drinking at dangerous levels who walk out thinking they're fine. Over a decade, that's 120 people per doctor.
Luke
But we should be careful here—the study is showing a discrepancy between what the AUDIT says and what the PEth test says. It's not proving the AUDIT is wrong. It's possible the PEth is flagging something that isn't actually clinically hazardous, or that the AUDIT is measuring something different and still valid.
Mimi
That's fair. But the PEth is an objective biomarker. It's not based on what someone tells you; it's based on what's in their blood.
Luke
True, but PEth reflects alcohol exposure over weeks, while AUDIT is asking about patterns and consequences. They're measuring different dimensions of the problem. A person could have high PEth from a binge last month but not have a drinking disorder.
Mark
So which method should doctors trust?
Mimi
The research suggests they shouldn't trust any one method alone. Use them together. If someone passes AUDIT but has high PEth, that's a signal to dig deeper.
Luke
The other striking finding is how much the methods disagree on prevalence. AUDIT flags 26.5 percent of patients, PEth flags 15 to 18 percent, and self-report only 8.5 percent. That's a huge range. It means we don't actually know how many people in primary care are drinking hazardously.
Mark
Which one is closest to the truth?
Luke
We don't know from this study. The researchers are saying the truth is probably somewhere in the middle, and that you need multiple measures to approach it. But that's an honest answer, not a satisfying one.
Mimi
It's the right answer, though. Drinking is complicated. How much you drink, how often, what it does to your health, whether you can stop—those are all different questions. One test can't answer all of them.
The Pulse
- Nearly 6% of patients who passed the standard AUDIT alcohol questionnaire were quietly flagged as hazardous drinkers by a blood biomarker test—people the system believed it had already checked.
- The three screening methods produced strikingly different counts: questionnaires caught 26.5% of hazardous drinkers, blood tests 15–18%, and simple self-reported weekly intake only 8.5%—each measuring something real, none measuring the same thing.
- Demographic fault lines complicate the picture further: men over-reported consumption relative to their biomarker levels, younger patients scored higher on questionnaires than their blood suggested, and education and income shaped how much people admitted to drinking.
- The tension is not between good tools and bad ones, but between methods that capture behavior as people narrate it and methods that capture behavior as the body records it.
- Swedish researchers are calling for a complementary approach—PEth blood tests used alongside questionnaires, not instead of them—so that biological evidence can catch what conversation and self-reflection leave behind.
For decades, primary care medicine has asked patients to account for their own drinking—through questionnaires and direct conversation—trusting that self-knowledge and honesty would surface those most at risk. A decade of Swedish health records now gently challenges that trust: a blood biomarker called PEth, which accumulates in proportion to alcohol consumed over recent weeks, reveals hazardous drinking in nearly one in seventeen patients who had already been cleared by standard screening. The finding is less an indictment of any single tool than a reminder that human self-report and biological reality do not always speak the same language.
Primary care physicians have long relied on two familiar instruments to identify dangerous drinking: the AUDIT questionnaire and a direct question about weekly consumption. Both are fast, practical, and widely trusted. But a study drawing on a decade of Swedish health records—2013 to 2023—suggests these tools are quietly missing a meaningful portion of the people they are meant to find.
The research compared patient self-reports against PEth, a blood biomarker that accumulates in proportion to alcohol consumed over the preceding weeks. The gap it exposed was striking: nearly 6 percent of patients who scored below the AUDIT danger threshold—people the standard process had effectively cleared—showed hazardous biomarker levels in their blood. They had passed the questionnaire. The blood told a different story.
The three methods also diverged sharply in how many patients they flagged overall. AUDIT identified hazardous drinking in 26.5 percent of the primary care population. PEth blood tests caught between 15 and 18 percent. But when doctors simply asked patients how much they drank each week, only 8.5 percent admitted to hazardous levels. Each number reflects something true; none of them reflects the whole truth.
Patterns in who gets missed—and who gets caught—added further complexity. Men reported drinking more than women at equivalent biomarker levels. Patients with higher education or income disclosed greater weekly consumption. Younger patients scored higher on AUDIT even when their blood markers resembled those of older drinkers. Self-report, the data suggests, is shaped not only by actual drinking but by perception, willingness to disclose, and the social textures of a clinical conversation.
The researchers' conclusion is not that questionnaires should be abandoned, but that they should no longer stand alone. PEth tests and self-report measures used together—each checking the other—offer a more complete picture than either can provide in isolation. The goal is to add a layer of biological verification capable of reaching the drinkers that words, however carefully solicited, continue to miss.
Doctors in primary care have long relied on the same two tools to identify patients drinking at dangerous levels: they hand out questionnaires like the Alcohol Use Disorders Identification Test, or AUDIT, and they ask patients directly how many drinks they consume in a typical week. It's straightforward, it's fast, and it has become standard practice. But a decade of Swedish health records suggests these methods are missing a significant slice of the problem.
Researchers examining data from 2013 to 2023 compared what patients reported about their drinking against a blood test that measures phosphatidylethanol, or PEth—a biomarker that accumulates in the bloodstream in proportion to alcohol consumption over the preceding weeks. The comparison revealed a troubling gap: nearly 6 percent of patients who scored below the danger threshold on the AUDIT questionnaire—meaning they appeared safe by that measure—actually showed hazardous levels of alcohol in their blood. These were people the standard screening process had cleared, yet the biomarker test flagged them as at-risk drinkers.
The three methods diverged dramatically in how many patients they identified as drinking hazardously. When researchers defined hazardous use as either a PEth level above 210 nanograms per milliliter, an AUDIT score of 8 or higher, or self-reported consumption of 10 or more drinks per week, the numbers told very different stories. The AUDIT questionnaire caught the most patients—26.5 percent of the primary care population. The PEth blood test identified between 15 and 18 percent. But when doctors simply asked patients how much they drank each week, only 8.5 percent admitted to hazardous levels. Each method was measuring something real, but none of them were measuring the same thing.
The researchers also noticed patterns in who was most likely to be missed or caught by each approach. At the same PEth biomarker levels, men reported drinking more than women did, and patients with higher education or income tended to report greater weekly consumption. Younger patients, meanwhile, scored higher on the AUDIT questionnaire even when their biomarker levels were comparable to older patients'. These differences suggest that self-reporting is shaped not just by actual drinking but by how people perceive their drinking, how willing they are to disclose it, and possibly by demographic factors that influence both drinking patterns and how those patterns are expressed in conversation with a doctor.
The implication is clear: no single screening method captures the full picture. A patient might pass the questionnaire and minimize their drinking in conversation, yet still show objective evidence of hazardous consumption in their blood. Conversely, someone might score high on AUDIT or report heavy drinking but have biomarker levels that suggest less chronic alcohol exposure. The Swedish researchers argue that primary care should move toward a complementary approach—using PEth tests and self-report measures together, each one checking the other, rather than treating them as interchangeable alternatives. The goal is not to replace the questionnaire or the conversation, but to add a layer of biological verification that can catch the drinkers the standard tools miss.
Notable Quotes
No single measure captures the full picture of a patient's drinking, supporting a multimethod approach where PEth biomarker tests and self-reports are used as complementary rather than interchangeable tools.— Swedish researchers
Want the full story?
Read the original at News-Medical ↗
Science & Space