Skip to content

Improving Oral and Systemic Health in Individuals With Prediabetes Through Personalized Oral Hygiene Advice Provided by Dentists or by AI

Improving Oral and Systemic Health in Individuals With Prediabetes Through Personalized Oral Hygiene Advice Provided by Dentists or by Artificial Intelligence: A Randomized Clinical Trial

Status
Recruiting
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT06980701
Enrollment
148
Registered
2025-05-20
Start date
2025-04-12
Completion date
2026-04-18
Last updated
2025-12-03

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Gum Disease, Prediabetes

Keywords

gum disease, prediabetes, oral health, oral hygiene

Brief summary

Prediabetes is an intermediate stage before the development of diabetes, characterized by elevated blood glucose levels but lower than the diagnostic criteria of diabetes and is associated with multiple long-term complications. This systemic disease is mutually linked to inflammatory gum diseases through circulating inflammatory mediators. Controlling inflammatory gum diseases improves blood glucose levels and reduces long-term complications. While maintaining good oral hygiene through home care is essential for managing inflammatory gum diseases, close supervision of patients' home care is labor-intensive and expensive. Artificial Intelligence (AI) has been used to provide personalized advice on the adequacy of patients' home care (oral hygiene). The investigators hypothesize that the use of AI can improve home care, thereby enhancing both gum health and systemic health, similar to human dental professionals.

Detailed description

Prediabetes is an intermediate stage before the development of diabetes, characterized by elevated blood glucose levels but lower than the diagnostic criteria of diabetes and is associated with multiple long-term complications. This systemic disease is mutually linked to inflammatory gum diseases through circulating inflammatory mediators. The relation between oral health and prediabetes management has long been under-appreciated. People with prediabetes have a 2-3-fold greater risk for periodontitis compared to people without prediabetes. The progression and severity of periodontitis are also greater in prediabetic patients. According to the National Health and Nutrition Examination Survey, the severity of periodontitis is positively associated with the risk as well as the prevalence of prediabetes. A growing body of data indicates that oral inflammation has an impact on general diseases. Controlling inflammatory gum diseases improves blood glucose levels and reduces long-term complications. While maintaining good oral hygiene through home care is essential for managing inflammatory gum diseases, close supervision of patients' home care is labor-intensive and expensive. Nowadays, artificial intelligence (AI) can readily assist in the self-detection of diseases, including gum disease, allowing older adults to identify diseases early and prevent further complications. The use of AI-based mHealth has become increasingly effective in promoting periodontal health by adopting simple, AI-driven self-tests using smartphones. Another systematic review done by investigators' team found that AI-based mHealth for oral hygiene and gum disease monitoring showed clinical effectiveness across different clinical scenarios. The investigators' team has already launched an AI system for the detection of gum disease using smartphone intraoral photography, in which the system can detect colour changes of gum inflammation in specific sites in intraoral photography and diagnose as three simple situations (severe, mild and no inflammation). The AI system have high sensitivity 92% to identify disease from sites that have gingivitis, and high specificity 94% to identify healthy tissue from sites that have no gingivitis using professional intraoral photography. Moreover, the investigators have tested that the accuracy of colour captured by a smartphone is comparable to that captured by a professional single-lens reflective camera. The investigators' team already have applied the AI-powered smartphone photography among 38 older adults in 5 day-care centres of Hong Kong to test participants' gum health. The result is promising with accuracy of 96% sensitivity and 82% specificity. The present study will apply AI technology on disease detection and giving personalized oral health instruction (OHI) closely to the patients to maintain periodontal health and consequently prediabetic control. In this study, the hypothesis is that the use of AI can improve home care, thereby enhancing both gum health and systemic health, similar to human dental professionals.

Interventions

BEHAVIORALAI OHI group

The participants will receive personalized OHI such as toothbrush and interdental cleaning to specific areas provided by AI. An mHealth system will be used to detect intraoral photograph of anterior teeth and analysis of the photograph and label the gum condition as Healthy (green)/questionable (yellow)/diseased (red) within 2 minutes by AI. Then specific OHI to each particular site would be provided by AI according to tested results on the photograph

BEHAVIORALDental professionals OHI group

ll participants will receive personalized OHI by dental professionals. This instruction includes brushing and interdental cleaning in each particular dental site. If they have any personal concern or unclear points regarding oral hygiene practice, they can ask.

Sponsors

The University of Hong Kong
Lead SponsorOTHER

Study design

Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
TRIPLE (Subject, Investigator, Outcomes Assessor)

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No

Inclusion criteria

* \- Subjects who are \>18 years of age and able to give informed consent. * \- Subjects who are mentally and cognitively healthy. * \- Subjects who have at least 6 anterior maxillary or mandibular natural teeth including incisors and canine. * \- Subjects who are with prediabetic state with impaired HbA1c 5.7% to 6.4%, impaired fasting plasma glucose level 5.6mmol/L to 6.9 mmol/L and/or impaired plasma glucose level after 2h OGTT 7.8 mmol/L to 11.0 mmol/L. * \- Subjects who can speak, read, or understand Cantonese to complete the satisfaction questionnaire. * \- Subjects who can practice oral hygiene procedure (regular tooth brushing and interdental cleansing/flossing/brushing) daily on their own independently.

Exclusion criteria

* \- Subjects who have less than 6 anterior maxillary or mandibular natural teeth with or without dental prostheses in those area. * \- Subjects who are with a current diagnosis or clinical history of T2DM. * \- Subjects who have mental illness, or similar problems that unable to complete the satisfaction questionnaire. * \- Subjects who cannot perform oral hygiene procedure (regular tooth brushing and interdental cleansing/flossing/brushing) by any condition of oral cavity such as tumor or maxillomandibular fixation.

Design outcomes

Primary

MeasureTime frameDescription
2-h PG during 75-g OGTT level at 9-month9-monthGlycaemic level at 9-month follow-up is evaluated by 2-h PG during 75-g OGTT (plasma glucose level after 2-hour 75-g oral glucose tolerance test). Prediabetes is defined as 2-h PG during 75-g OGTT of 140-199 mg/dL (7.8-11.0 mmol/L)
Gum inflammation at 3-month3-monthGum inflammation will be evaluated using BPE score (from 0 to 4), which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 3-month follow up
Gingival health at 3-month3-monthGingival health will be assessed by gingival index (Löe H 1967), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 3-month follow-up
Oral hygiene status at 3-month3-monthOral hygiene status will be assessed by plaque index (Silness and Loe, 1965), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 3-month follow-up
HbA1c level at 3-month3-monthGlycaemic level at 3-month follow-up is evaluated by HbA1c level (glycated haemoglobin that measures glycaemic control over the past 2-3 months). Prediabetes is defined as HbA1c of 5.7-6.4% (39-47 mmol/mol)
FPG level at 3-month3-monthGlycaemic level at 3-month follow-up is evaluated by FPG (fasting plasma glucose level that measures the blood sugar levels). Prediabetes is defined as 100-125 mg/dL (5.6-6.9 mmol/L)
2-h PG during 75-g OGTT level at 3-month3-monthGlycaemic level at 3-month follow-up is evaluated by 2-h PG during 75-g OGTT (plasma glucose level after 2-hour 75-g oral glucose tolerance test). Prediabetes is defined as 2-h PG during 75-g OGTT of 140-199 mg/dL (7.8-11.0 mmol/L)
Gum inflammation at 9-month9-monthGum inflammation will be evaluated using BPE score (from 0 to 4), which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 9-month follow up
Gingival health at 9-month9-monthGingival health will be assessed by gingival index (Löe H 1967), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 9-month follow-up
Oral hygiene status at 9-month9-monthOral hygiene status will be assessed by plaque index (Silness and Loe, 1965), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group) at 9-month follow-up
HbA1c level at 9-month9-monthGlycaemic level at 9-month follow-up is evaluated by HbA1c level (glycated haemoglobin that measures glycaemic control over the past 2-3 months). Prediabetes is defined as HbA1c of 5.7-6.4% (39-47 mmol/mol)
FPG level at 9-month9-monthGlycaemic level at 9-month follow-up is evaluated by FPG (fasting plasma glucose level that measures the blood sugar levels). Prediabetes is defined as 100-125 mg/dL (5.6-6.9 mmol/L)
Gum inflammation at baselinebaselineGum inflammation will be evaluated using BPE score (from 0 to 4), which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group)
Gingival health at baselinebaselineGingival health will be assessed by gingival index (Löe H 1967), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group)
Oral hygiene status at baselinebaselineOral hygiene status will be assessed by plaque index (Silness and Loe, 1965), with a scale from 0 to 3, which will be examined by a blinded assessor (a calibrated dentist who blinded to participants' group)
HbA1c level at baselinebaselineGlycaemic level at baseline is evaluated by HbA1c level (glycated haemoglobin that measures glycaemic control over the past 2-3 months). Prediabetes is defined as HbA1c of 5.7-6.4% (39-47 mmol/mol)
FPG level at baselinebaselineGlycaemic level at baseline is evaluated by FPG (fasting plasma glucose level that measures the blood sugar levels). Prediabetes is defined as 100-125 mg/dL (5.6-6.9 mmol/L)
2-h PG during 75-g OGTT level at baselinebaselineGlycaemic level at baseline is evaluated by 2-h PG during 75-g OGTT (plasma glucose level after 2-hour 75-g oral glucose tolerance test). Prediabetes is defined as 2-h PG during 75-g OGTT of 140-199 mg/dL (7.8-11.0 mmol/L)

Secondary

MeasureTime frameDescription
IL6 level at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsInflammatory markers, i.e., IL6 from serum, saliva and plaque samples (pg/mL) will be assessed at baseline, 3-month and 9-month follow-ups
IL8 level at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsInflammatory markers, i.e., IL8 from serum, saliva and plaque samples (pg/mL) will be assessed at baseline, 3-month and 9-month follow-ups
Body weight at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsBody weight (measured with standard procedures, in kilograms) will be assessed at baseline, 3-month and 9-month follow-ups
Percentage body fat at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsPercentage body fat (using bioelectrical impedance analysis, %) will be assessed at baseline, 3-month and 9-month follow-ups
Shannon diversity index of oral and gut microbiota at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsOral and gut microbiota will be analyzed from stool, plaque and salivary sample. The diversity of the oral and gut microbiota samples will be measured by the Shannon diversity index.
Concentration of short chain fatty acid in stool samples at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsFecal metabolites will be measured through the concentration of short chain fatty acid in stool samples
3-day food record at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsDietary intake will be assessed using a 3-day food record to determine meal patterns, including macronutrient intake, energy intake, and meal timing
Chinese version of Chrono-nutrition Profile Questionnaire at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsChrono-nutrition behaviors will be evaluated using the Chinese version of the Chrono-nutrition Profile Questionnaire (CP-Q), which includes six distinct aspects: breakfast skipping, timing of the largest meal, evening eating habits, evening latency, nighttime eating behaviors, and eating window.
Chinese version of Munich Chronotype Questionnaire at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsChronotype will be assessed by Chinese version of Munich Chronotype Questionnaire (MCTQ). The computed variables of workdays and work-free days from this questionnaire include sleep onset (SOw, SOf, hh:mm), local time of getting out of bed (GUw, GUf, hh:mm), sleep duration (SDw, SDf, hh:mm), total time in bed (TBTw, TBTf, hh:mm), mid-sleep (MSW, MSF, hh:mm). The computation of the variables include: Average weekly sleep duration (hh:mm) =(SDw x WD + SDf x FD)/7, Chronotype (hh:mm) =If SDf ≤ SDw: MSF; If SDf \> SDw: MSF - (SDf - SDweek)/2, Weekly sleep loss (hh:mm) = If SDweek \> SDw: (SDweek - SDw) x WD; If SDweek ≤ SDw: (SDweek - SDf) x FD, Relative social jetlag (hh:mm) = MSF - MSW, Absolute social jetlag (hh:mm) = \| MSF - MSW \|, Average weekly light exposure (hh:mm) = (LEw x WD + LEf x FD)/7
Chinese version of international physical activity questionnaire short form at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsPhysical activity (PA) levels will be evaluated by a Chinese version of international physical activity questionnaire short form (IPAQ). Total PA (min/wk) = 2x time spent on vigorous + moderate + walking, MET (Metabolic Equivalent of Task, min/wk) = 8 x vigorous +4x moderate + 3.3 x walking
C-reactive protein level at baseline, 3-month and 9-month follow-upsat baseline, 3-month and 9-month follow-upsInflammatory markers, i.e., C-reactive protein from serum, saliva and plaque samples (pg/mL) will be assessed at baseline, 3-month and 9-month follow-ups

Countries

Hong Kong

Contacts

Primary ContactWalter Y.H. Lam, Prof.
retlaw@hku.hk(852) 28590306
Backup ContactMandy M Ho, Prof.
mandyho1@hku.hk(852) 39176973

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026