NHANES is a comprehensive continuous cross-sectional study conducted in the United States designed to evaluate the nutritional status and overall well-being of the U.S. population. To achieve this, a sample of approximately 5,000 non-institutionalized civilians, selected through a stratified multistage probability sampling method, is recruited on an annual basis to ensure national representation. Data from the NHANES 2009–2014 were included in this study. Participants were invited to complete a detailed home interview and a health examination in a mobile examination center (MEC), including a periodontal examination (age ≥ 30 years) and a depression screening (age ≥ 18 years) [18]. All NHANES protocols that generated the data were approved by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention (CDC). A total of 30,468 participants from the selected NHANES cohorts were divided into three groups in our analysis. Data inclusion process is shown in Fig. 1. The study excluded participants who had incomplete data on exposure variables including smoking, alcohol consumption, or depression, as well as those who did not undergo periodontal examinations, had abnormal BMI (including absent BMI and extreme BMI values), or were pregnant. Finally, we screened the following three groups as the smoking-depression group (10,164 individuals), drinking-depression group (6,900 individuals, alcohol consumption), and drinking-smoking group (6,849 individuals, the frequency of drinking).

Flowchart of the sample selection from NHANES 2009–2014

All oral health assessments were conducted within a specifically designated room at the mobile examination center (MEC), equipped with a portable dental chair, light, and compressed air. Trained dentists performed the periodontal examination, measuring clinical attachment loss (CAL) and probing depth at six interproximal sites per tooth (distal-facial, mesial, proximal, distal-lingual, mesial-lingual, and proximal-lingual), excluding third molars, and then a health technician entered the collected data. Only individuals aged 30 years and above were considered eligible for participation in these periodontal examinations. The proportion of sites with a CAL ≥ 3 mm is commonly used as an indicator of the severity of periodontitis. In our study, we identified periodontitis in each subject with at least one site exhibiting a CAL ≥ 3 mm [19, 20].

Depression was assessed with the Patient Health Questionnaire-9 (PHQ-9), including nine items about the frequency of depressive symptoms over the previous two weeks. PHQ-9 total scores ranged from 0 to 27 and the cutoff score for depression identification was 10 or more [21]. The drinking and smoking data were collected by computer-assisted personal interview (CAPI) technology. Daily alcohol consumption in the past 12 months was categorized into three levels: low-intensity (1 drink or less), moderate (between 1 and 3 drinks), heavy drinking (4 drinks or more) [22]. A ‘drink’ was defined as containing 14g alcohol. The frequency of weekly alcohol consumption was categorized as ‘5 days or more per week’ (regular drinkers), ‘between 2 and 4 days per week’ (occasional drinkers), and “1 day or less per week’ (non-drinkers). Smokers were defined as those who reported smoking at least 100 cigarettes during their lifetime, either currently (current smoker) or not currently (former smoker) [18].

The sociodemographic variables included in this study were age (in years), gender (male, female), race/ethnicity (non-Hispanic white, non-Hispanic black, non-Hispanic Asian, Mexican Americans, and other race), education level (less than 9th Grade, 9-11th Grade, high school grade/ General Education Development or Diploma or equivalent, some college or Associate of Arts Degree, college graduate or above, missing), marital status (married/living with partner, widowed/divorced/separated, missing), history of diabetes (no, yes, missing), and family monthly poverty level (≤ 1.30, 1.30–1.85, 1.85–5, > 5, missing) [23, 24], as per the poverty guidelines outlined by the U.S. Department of Health and Human Services (HHS) and the recommendations provided by the National Health and Nutrition Examination Survey (NHANES) [25]. Body mass index (BMI) was classified as low weight (< 18.5 kg/m2), normal weight (18.5 to 25 kg/m2), overweight (25 to 30 kg/m2) and obesity (≥ 30 kg/m2) [26]. Diagnosed diabetes mellitus, damaging the healing process of periodontitis, was identified by self-report questionnaires [21].

Considering the complicated sample design, all analyses incorporated the NHANES sampling weights to ensure the representativeness of noninstitutionalized civilian resident population of the United States. Descriptive data were presented as mean (standard deviation) for continuous variables and number (percentage) for categorical variables in terms of sociodemographic and periodontal characteristics. The multivariate logistic regression model was used to analyze the independent effects and multiplicative interaction between depression, smoking, and drinking based on initial model and adjusted model (adjusted for age, gender, ethnicity, education, household income and history of diabetes). We analyzed the interaction effect across different gender (male and female) and age groups (≤ 45 and > 45) [27]. Odds ratio (OR) and 95% confidence interval (CI) were calculated. All statistical analyses were performed using the SPSS 24.0 statistical software package (IBM SPSS Statistics for Windows, Version 24.0) and R (versionversion 4.3.1). The statistical significance was set at P < 0.05 using the two-sided test.