Study design
The presented observational study was carried out as part of the project "Oral Health and Quality of Life in Old Age: A Cross-Sectional Pilot Project in Germany and Poland" in cooperation between Municipal Council and the University Hospital Carl Gustav Carus in Dresden, Germany. The Polish population was examined over the period of 24 months after obtaining the consent of the Bioethics Committee of Wroclaw Medical University. The study was supported by Municipal Council resources as part of the project "Oral Health and Quality of Life of Elderly Residents of Wroclaw) (financial agreement P/ZJU/1/2015-2017) in accordance with the STROBE guidelines (Strengthening the Reporting of Observational Studies in Epidemiology) [12].
Participants
The study involved randomly selected participants (n = 1338), both men and women, aged between 65 and 99. However, 41.0% (n = 549) did not respond to the study invitation, 6.6% (n = 88) did not provide written consent to take part in the study and 15.0% (n = 201) were excluded due to the presence of systemic diseases in which periodontal probing, leading to transient bacteremia, might have posed a risk to the patient's overall health, particularly in the subjects diagnosed with cardiovascular diseases (patients with heart valves, after a heart transplant, with congenital heart diseases or infective endocarditis), blood diseases (thrombocytopenia, haemophilia, von Willebrand disease), viral diseases (B and C type hepatitis, AIDS/HIV), as well as subjects with Multi-Drug Resistant Organisms (MDRO). That is why 500 participants were ultimately included in the study (Fig.1).
The inclusion criteria consisted of age – 65 and more, place of living (local resident), the ability to communicate and provision of written consent to participate in the survey. The exclusion criteria included concurrent systemic diseases, in which periodontal probing, leading to transient bacteremia, might have posed a risk to the patient's overall health, failure to provide written consent to participate in the survey and the occurrence of mental disorders which render the completion of a questionnaire impossible. All participants involved were obliged to provide written informed consent and completed questionnaire, as well as demonstrate cooperation during the clinical oral examinations. Participants who did not fulfil the inclusion criteria were excluded from the study.
Ethical permission
The study protocol was approved by the Bioethics Committee of Wroclaw Medical University (permission no. KB 420/2015) under the Declaration of Helsinki. Participation in the study was voluntary and anonymous, and the collected data were treated confidentially.
Sample size estimation
Data on the total number of residents aged 65 and more were derived from Statistics Poland (Demographic Yearbook of Poland 2015) [13]. The sample size was calculated based on the data concerning the number of individuals in this age group who live in the city. With such an assumption, a 95% level of confidence and ±5% margin of error, a minimum sample size consisted of 384 participants[14].
Oral examination
Evaluation of oral health was conducted by two examiners. Each examiner was asked to study the same group of 10 patients. The findings were then compared with those of the experienced supervisor. The value of the inter-examiner kappa coefficient amounted to 0.874 while the intra-examiner kappa coefficient was 0.870.
Oral examination was performed using artificial light, a plane dental mirror and a ball-ended probe (WHO CPI probe). Coronal and root caries were evaluated according to the World Health Organisation (WHO) criteria, DMFT values and their components. Periodontal health was assessed in individuals who had more than 2 natural teeth by measuring bleeding on probing (BoP) of 6 sites, gingival pocket depth (PD), clinical attachment loss (CAL) [15] and tooth mobility according to the Miller index. Dental prosthetic status was assessed by taking into account the number of functional tooth units of natural and artificial teeth on implant-supported and fixed prostheses (non-functional tooth units), as well as the information on whether the participants wore partial or complete removable dentures. Based on the clinical oral examination, a patient's treatment needs were determined and categorised on a 5-grade scale: no treatment need, preventive treatment, prompt treatment, immediate treatment or referral.
Moreover, xerostomia (oral dryness) was examined using the Challacombe scale (Clinical Oral Dryness Score, CODS). The level of dryness was categorised based on the number of symptoms observed as mild (1-3), moderate (4-6), or severe (7- 10) [16, 17].
Questionnaire study
QoL was evaluated using three inventories – Euro-Quality of Life (EQ-5D-3L), Oral Health Impact Profile-14 (OHIP-14) and Patient Health Questionnaire (PHQ-9), which were validated for the Polish population [18,19, 20].
EQ-5D-3L is a brief, five-item instrument (mobility, self-care, usual activities, pain/discomfort, anxiety/depression) with three response options (no problems, some problems, extreme problems), resulting in 243 health states. Moreover, a visual analogue scale (EQ-VAS) was used for the evaluation of the health-related quality of life (HRQoL) [18, 21].
OHIP-14 consisted of 14 questions related to problems occurring within 1 year: 1) trouble with pronouncing words, 2) worsened taste, 3) pain, 4) discomfort while eating, 5) self-consciousness, 6) emotional tension, 7) unsatisfactory diet, 8) interrupted meals, 9) difficulty with relaxing, 10) embarrassments, 11) irritability, 12) inability to complete everyday tasks, 13) reduced satisfaction with life, 14) complete inability to function. The questions corresponded to 7 dimensions: functional limitation, pain, psychological discomfort, physical disability, psychological disability, social disability and handicap. The frequency of occurrence within 1 year was assessed using the five-point Likert scale: 0 – never, 1 – rarely, 2 – occasionally, 3 – frequently, 4 – very often. All values were summed to calculate the total OHIP-14 score which could vary between 0 and 56; whereby the higher the OHIP-14, the poorer the OHRQoL [22].
PHQ-9 is a questionnaire consisting of 9 basic questions and one additional question. The answer to each question is scored on a scale ranging from 0 to 3, depending on the frequency of occurrence of a given symptom in the last 2 weeks (3 points indicate the most frequent occurrence of the symptom). The Polish version of PHQ-9, used in the presented study, is available at http://www.phqscreeners.com by Pfizer Inc. Diagnosis includes the following severity stages: 0-4 - none, 5–9 - mild depression, 10–14 - moderate depression, 15–19 - moderately severe depression, 20–27 - severe depression [23].
Statistical analysis
The obtained questionnaire results and results of clinical trials were statistically analysed using STATISTICA v. 12 software (StatSoft, Inc.). Qualitative, nominal and ordinal variables are shown in tables in terms of sample size (n), set size (N) and proportion (%). The Pearson's chi-squared (χ²) test of independence was used for the assessment of the relationship strength between nominal and ordinal variables. In the case of quantitative variables, arithmetic means (M), standard deviations (SD), medians (Me), lower quartiles (Q1) and upper quartiles (Q3) were calculated and ranges of values from smallest to largest were provided. The normality based on the empirical distributions of the quantitative variables was checked using the Kolmogorov-Smirnov test. Hypotheses concerning the lack of differences between the mean values in two subgroups (women/men) were verified using the non-parametric Mann–Whitney U test of significance. The relationship strength between the EQ-5D-3L, EQ-VAS, OHIP-14, and PHQ-9 psychometric scale scores and sociodemographic and clinical factors was determined by calculating Spearman's linear correlation coefficient values (rho) and regression coefficient values (b). A t-test based on the Student's t-distribution with n-1 degrees of freedom was used for verifying the significance of the Pearson's correlation coefficient. Furthermore, the reliability analysis of each psychometric scale used in the study was conducted by calculating both the Cronbach's alpha coefficient value and the mean correlation between the items and the final score. In all statistical tests, p<0.05 was taken as the critical value.