This cross-sectional study used data from the Chilean NHS 2016-2017; version 3 (Department of Epidemiology, Ministry of Health, Chile); which collected information on the social determinants, risk factors, and protective influences of various diseases (6). The study sample was representative of the Chilean population over a period of 15 years and included men and women from both rural and urban parts of the country. Pregnant women and individuals who refused to participate in the survey during the home visit were excluded from the study. The survey was carried out using a complex multi-stage clustered, stratified, randomized oversampling technique and had overall response and rejection rates of 67% and 9.8%, respectively.
Data collection included home interviews carried out between August 2016 and March 2017 by interviewers and previously calibrated nurses (kappa: 0.85, p-value < 0,01). The survey has 6233 respondents, of which 5520 underwent blood and laboratory testing and oral examination. The first, second, and third visits included interviews (Form F1); anthropometric measurement and testing (including oral examination) carried out by a nurse (Form 2); and application of an expanded mental health section to a sub-sample of participants by a trained interviewer (Form F3), respectively. The oral examination (Form F2) included evaluation of the following items: total number of remaining teeth (both jaws); absence of anterior teeth (yes/no); the total number of teeth with cavitated carious lesions (both jaws); and effective resolution of anterior edentulousness using removable dentures (yes/no; both jaws).
In form F3, selected sub-sections (screening, depression, social phobia, agoraphobia, alcohol abuse and dependence, suicidality, mania, psychosis, and use of mental health services) of the Composite International Diagnostic Interview (CIDI), a mental health diagnostic tool developed by the World Health Organization, were applied to a sub-sample of participants (n = 3403) that were >18 years of age by a trained interviewer (9). Older adults that exhibited cognitive impairment during the first visit were excluded.
The final study sample included 2953 survey participants that were >18 years of age. The exclusion criteria were as follows: missing data on the oral health item of interest; failure to undergo oral examination; missing data on the extended mental health section (CIDI); and missing data in the depressive symptoms section.
Depressive symptoms were recorded using an abbreviated version of the CIDI instrument (CIDI Short form; CIDI-SF) containing 30 questions focusing on the presence of dysphoria (sadness symptoms) and anhedonia (lack of interest or ability to enjoy), and a depression risk score was calculated if the patient met at least five out of seven complementary criteria (Diagnostic and Statistical Manual of Mental Disorders or DSM-IV minor criteria for depression).
The participants were diagnosed with depression (as per the CIDI-DSM IV criteria) if they exhibited [1] depressed mood and [2] reduction or loss of interest or pleasure for at least 2 weeks and met >3 of the following criteria: [1] significant increase or decrease in appetite resulting in substantial changes in weight; [2] suicidal ideation; [3] significant sleep disturbances; [4] psychomotor agitation or motor slow-down; [5] fatigue or loss of energy; [6] feelings of worthlessness or guilt; and [7] decreased concentration.
Of these, the last five symptoms must have been experienced all day or almost every day for at least two weeks to be considered in the score. Furthermore, these symptoms must have caused clinically significant discomfort and impairment of social, occupational, and other important aspects of the individual’s life. Therefore, a diagnosis of depression was made if the participant met at least five criteria. Participants with symptoms caused by substance abuse, drugs, medications, and grief or loss of a loved one were excluded.
The variables were defined based on the questions in specific sections of the forms used for corresponding NHS interviews. These sections include ‘Section IX’ for depressive symptoms (questions sd1_F1 to sd31_F1), ‘Section XV’ for oral health (questions sb1 to sb5) from Form 1, ‘Section V’ for oral examination (questions m5p1 to m5p10) from Form 2, and the ‘Depression Section’ (questions D1 to D88) from Form 3 (CIDI).
The independent variables included use of dental prosthesis; number of remaining teeth (both jaws); anterior tooth loss; number of decayed teeth; and self-perception of oral health (Table 1), while the dependent variables were history of depression; treatment for depression; suspected depression; diagnosed with depression in the last 12 months; and diagnosed with depression during the lifetime. The analysis models were adjusted for age, level of education, and sex.
Please add Table 1 here.
Descriptive statistics, including percentages for categorical variables and mean and dispersion measures for numerical variables, were generated. Logistic regression models were used to estimate OR and 95% CI. Directed acyclic diagrams (not shown) and relationship matrixes (heat plots) were used to examine the association between the variables and outcome measures. The models examining the association between suspected depression and self-perception of oral health were adjusted for sex, level of education, and age, while those exploring the relationship between prosthesis use and the number of remaining teeth were adjusted for the same factors as well as tobacco use. Potential confounding factors considered when examining a diagnosis of depression in the past 12 months as an outcome measure included sex, tobacco, and education, generating open backdoor paths if they do not condition them. The analysis carried out in this study respected the complex sampling and the expansion factors used, which is represented in the results through frequencies and expanded sample sizes. All analyses were performed using the statistical software STATA version 16.1 (Windows; STATA Corp. 2019. College Station, TX: StataCorp LLC.).
The NHS 2016-2017 survey was approved by the Scientific Ethics Committee, Faculty of Medicine, Pontificia Universidad Católica de Chile, and informed consent was obtained from all participants. An anonymized version of the database of volunteers has been made available for use for research purposes on the Chilean Ministry of Health website. The current study was approved by the Scientific Ethics Committee of Universidad de los Andes (ID: CEC2021059).