A retrospective study of 850 patients at Kyung Hee University Dental Hospital in Korea, published online in Clinical Oral Investigations on 13 October 2026, found that the number of high-confidence xerogenic medication classes a patient takes was independently linked to low stimulated saliva flow. In prediction models, adding two chairside measures, salivary pH and buffer capacity, produced the largest improvement in identifying hyposalivation.
Dry mouth can be easy to miss in a busy appointment, and a long medication list is easy to skim. As of October 2026, a new study from Korea gives dental professionals a useful way to think about who should have their saliva measured properly, and why the medication list on its own may not be enough.
The paper, "Xerogenic medications and clinical markers of salivary hypofunction," was written by Lee YH, Jeon S, Kim DH, Kim S, Kim TS, Lee SW and Noh YK, with affiliations at Seoul National University, Hanyang University and Kyung Hee University. It appeared online in Clinical Oral Investigations on 13 October 2026.
What did the researchers study?
The team looked back at 850 patients who had both a salivary flow assessment and an oral clinical evaluation at Kyung Hee University Dental Hospital. The group was older and mostly female: the mean age was 60.8 years, and 657 patients (77.3%) were women. The authors split the patients into four age bands: under 40, 40 to 64, 65 to 79, and 80 and older.
They analysed two standard measures from patients' existing salivary flow assessments:
- Unstimulated salivary flow rate (UFR): the resting flow, without stimulation.
- Stimulated salivary flow rate (SFR): the flow produced when the glands are activated.
Hyposalivation was defined by flow rate. The authors then ran the data through two kinds of analysis: conventional multivariable regression, and incremental gradient boosting models interpreted with SHapley Additive exPlanations (SHAP), a method that ranks how much each input contributes to a model's prediction.
What did the study find about age and sex?
Resting flow was lower in older age groups. Mean UFR was 0.57 mL/min in patients under 40 and 0.36 mL/min in patients 80 and older. The authors report that age-group differences in stimulated flow were less pronounced.
Even so, stimulated-flow hyposalivation was most common in the oldest group, at 17.3% of patients aged 80 and older. After adjustment, that group had 5.47 times the odds of SFR-defined hyposalivation compared with patients under 40, although the confidence interval was wide (1.44 to 20.78).
Female sex was associated with lower resting and lower stimulated flow.
How much did xerogenic medications matter?
This is the part most relevant to the medical history form. The authors counted "high-confidence" xerogenic medication classes for each patient. That count was independently associated with SFR-defined hyposalivation, with an odds ratio of 1.63 (95% CI 1.12 to 2.36).
In plain terms: in this cohort, a higher count of high-confidence xerogenic medication classes was independently associated with higher odds of SFR-defined hyposalivation. That fits a common clinical concern about long medication lists.
But the authors were careful not to stop there. Their conclusion is that objective salivary hypofunction "was not fully explained by age, systemic disease burden, or medication burden alone." In other words, the medication count alone did not fully account for who had low flow.
Pro Tip: Ask about dry mouth at the same time you review the medication list, not as a separate question later in the appointment. Pairing the two makes it easier to notice when a new prescription and a new complaint arrive together.
Why do salivary pH and buffer capacity matter?
The machine-learning results are where the study becomes practical. The authors built models step by step, adding groups of variables and checking how much each addition improved prediction. Adding salivary pH and buffer capacity produced the largest improvement in discrimination.
- For UFR-defined hyposalivation, the area under the ROC curve (AUROC) rose from 0.468 to 0.723.
- For SFR-defined hyposalivation, AUROC rose from 0.608 to 0.725.
An AUROC near 0.5 means a model does no better than chance; higher values mean better discrimination. Before the chairside markers were added, the model for resting-flow hyposalivation was essentially at chance level. After they were added, discrimination rose to a moderate level (AUROC 0.723).
The SHAP analysis ranked salivary buffer capacity as the leading contributor for both outcomes, with salivary pH also among the top predictors. The authors describe these markers as capturing "clinically relevant information about current salivary status,".
What does this mean for dental practices in Toronto and across Ontario?
The study was done in a Korean dental hospital, so its numbers describe that population, not yours. Its design also matters: a retrospective cross-sectional study can show associations at one point in time, but it cannot show that a given medication caused a given patient's low flow. Still, the clinical message translates well to a busy general practice in Toronto, Mississauga or Markham.
The authors' own summary is a "practical, layered approach." Read in that spirit, the study suggests three habits worth considering:
- Treat the medication list as a screening signal, not a diagnosis. A higher count of drying medication classes raised the odds of low stimulated flow, but it did not explain hypofunction on its own.
- Consider chairside salivary testing for patients who raise concern. In this study, pH and buffer capacity carried the most predictive weight of any variable group the authors added.
- Measure flow when the screen points that way. The study's conclusion is about identifying who "may require objective salivary flow assessment," not about replacing that assessment.
For practices with many older patients, the oldest age group's results are worth noting. Patients 80 and older had lower resting flow than the youngest group and the highest rate of stimulated-flow hyposalivation in this cohort.
Pro Tip: If your team adds chairside saliva testing, record the results in the chart in the same place each time. A baseline from this year is far more useful next year than a vague note that the patient "reports dry mouth."
What are the study's limits?
Three limits stand out from the abstract itself. First, the cohort came from a single dental hospital and skewed heavily female, which may not reflect a typical community practice. Second, the design was retrospective and cross-sectional, so it shows association rather than cause. Third, the confidence interval around the oldest group's odds ratio was wide, which means the size of that effect is uncertain even if its direction is clear.
The abstract does not list which medication classes were counted as "high-confidence" xerogenic, so clinicians should not read the study as a ranking of specific drugs. The full text, which requires a subscription, is the place to check those definitions.
EBIKO Dental will continue monitoring research on salivary hypofunction and chairside diagnostics as it is published. For practice supplies and more clinical news for Canadian dental professionals, visit EBIKO Dental.
Sources
- Lee YH, Jeon S, Kim DH, Kim S, Kim TS, Lee SW, Noh YK. "Xerogenic medications and clinical markers of salivary hypofunction." Clinical Oral Investigations, published online 13 October 2026. doi:10.1007/s00784-026-07184-0
- Europe PMC record: PMID 42867573
Frequently Asked Questions
Q: Does taking xerogenic medications mean a patient will have low saliva flow?
No. In the 2026 Clinical Oral Investigations study, the number of high-confidence xerogenic medication classes was linked to higher odds of low stimulated saliva flow, but the authors concluded that age, systemic disease burden and medication burden alone did not fully explain salivary hypofunction.
Q: Which chairside saliva tests were most useful in the study?
Salivary buffer capacity and salivary pH. Adding them to the prediction models produced the largest improvement, and buffer capacity ranked as the leading contributor for both resting and stimulated flow outcomes.
Q: Should salivary pH testing replace measuring saliva flow?
No. The authors describe pH and buffer capacity as a way to identify patients who may need objective salivary flow assessment, as part of a layered approach together with age, sex and medication exposure.