Oral health-related quality of life and associated factors in a care-dependent and a care-independent older population

Abstract

Objectives

To examine relationships between oral health (OH) factors and general health (GH) factors (including physical, mental, and social health domains), and oral health-related quality of life (OHRQoL) in a care-independent and a care-dependent older population.

Methods

Care-independent participants (n = 109) were recruited from the Nijmegen dental school; care-dependent participants (n = 126) from residential aged care facilities. Data collected included: OHRQoL (Geriatric Oral Health Assessment Index (GOHAI)), age, gender, socioeconomic status, number of teeth and occluding pairs, presence of carious teeth, presence of removable dental prostheses, clinically assessed treatment need (CTN), self-reported GH, and, only for care-dependent participants: care-dependency level and health domain variables: physical, mental (SF-12: Physical and Mental Component Summary scores), and social (ENRICHD social support index). Multiple linear regression analyses were performed to assess the associations with GOHAI scores.

Results

Mean GOHAI scores of care-independent (51.6 ± 7.4) and care-dependent participants (52.1 ± 6.7) did not differ significantly despite considerably worse OH status of the latter. Regression models revealed significant ( p ≤ 0.05) associations between GOHAI scores and age, prosthodontic status, and CTN in care-independent participants (R 2 = 0.19) and only with CTN in care-dependent participants. (R 2 = 0.09). Self-reported GH was not significantly associated with GOHAI; when substituted by the health domain variables, only social support was significantly associated with GOHAI scores.

Conclusions

GOHAI outcomes are associated with different variables in care-independent and care-dependent older subjects. In care-dependent subjects, GOHAI outcomes are more strongly related to social support than to OH factors or other GH factors.

Clinical significance

OHRQoL outcomes should not be compared across care-dependent and care-independent populations without careful interpretation of these outcomes against specific factors that distinguish such populations, like health factors and living environment.

Introduction

Over the last two decades, outcomes from oral health-related quality of life (OHRQoL) questionnaires have become widespread and common indicators of self-reported oral health . The use of OHRQoL measures and the interpretation of its outcomes, however, is not without controversy . Several studies have shown that associations between clinical indicators of oral health and OHRQoL outcomes are weak in older populations . Time and again, older people have reported better OHRQoL than young or middle-aged adults despite their generally worse oral health conditions . Perhaps the most plausible explanation for the found discrepancies between oral health status and OHRQoL outcomes is that the pathways connecting these issues are mediated by personal and environmental variables . Apart from personal characteristics, changes in values and meanings influence people’s perceptions of and reactions to (oral) diseases or disorders, and shape processes of coping and adaption . Such changes include adaptation to oral impairments which often come with ageing, but also with health decline and other life changing events. It can therefore be assumed that in care-dependent (i.e. “having limited, health-associated abilities to meet self-care demands” ) older people, oral health impairments have less impact on QoL than in care-independent older adults. Changes in values and meanings amongst care-dependent institutionalized populations may also be induced by reduced social health. Social health can be measured by a broad range of aspects, of which the most important are related to the extent and quality of perceived social support . Lack of social support is associated with being institutionalized and has a proven negative effect on life satisfaction and QoL in general .

Although the effect of age on the relationship between oral health status and OHRQoL has been demonstrated previously , it is unclear if and how this relationship differs between groups with different general health status To our knowledge, only few studies have addressed this issue . Three of these studies reported a significant association between better (self-reported) general health and better OHRQoL, regardless of age and oral health status. Only one study has addressed the association between care-dependency and OHRQoL. Zenthöfer et al. [2014] found that care-dependency level was inversely related to OHRQoL, although the number of respondents in this study was too low to derive any solid conclusions . We found no studies that documented the associations between variables representing three separate health domains (physical, mental, and social) and OHRQoL.

In light of the findings presented above, we posed the questions (1) if OHRQoL in institutionalized, care-dependent older people has a different pattern of associations in terms of oral health and general health when compared to home dwelling care-independent older people, and (2) if, in care-dependent older people, the size of the associations between OHRQoL and separate health domains (physical, mental, and social health) differ.

Gaining more insight into association patterns of OHRQoL outcomes and oral health factors and general health factors in people with and without health impairments enables us to better understand how oral health and OHRQoL are related. It also tells us if OHRQoL outcomes can be meaningfully compared between groups of different general health or care-dependency status.

Thus the aim of this study was to examine to what extent oral health factors and general health factors are associated with OHRQoL in care-dependent and care-independent older people.

Methods

Population and samples

Participants aged 65 years and over who were cognitively alert were recruited in two populations: a population of care-independent, home dwelling people (Group A) and a population of care-dependent institutionalized people with varying levels of care-dependency (Group B). Group A participants were recruited through convenience sampling of patients of the university dental clinic who visited the clinic for a regular check-up visit. Group B participants were recruited through contact managers of randomly chosen residential aged care facilities (RACFs) in South-East Netherlands. Purposive sampling was applied in both groups, aimed at achieving adequate numbers of subjects with regard to variables whose outcomes were known prior to sampling and that have been found to influence OHRQoL (i.e. gender, prosthodontic status (dentate/edentate) and levels of care-dependency) . The study was approved by the Medical Ethics Committee (CMO) of the Radboud University Medical Center Nijmegen (CMO ref. 2012/194). All recruits gave informed consent in writing to participate in the study.

Clinical data and care-dependency

Clinical data were obtained through clinical oral examination according to WHO criteria by final year calibrated dental students (all κ’s > 0.82; overall κ = 0.87; agreement = 90.1%) and calibrated final year dental hygiene students (all κ’s > 0.66; overall κ = 0.74; agreement = 84.4%). Data included number and position of teeth, presence of carious teeth, and presence of removable dental prostheses (RDPs). Clinical assessment of treatment need (y/n) was also recorded and comprised any need for professional dental treatment including reline, rebase or replacement of RDPs, and periodontal treatment. With regard to prosthodontic status, three categories were distinguished: people with at least one natural tooth and without RDPs, people with at least one natural tooth and one or more RDPs, and people without natural teeth with complete RDPs. With regard to health status and care-dependency, we distinguished between care-independent participants (Group A) with no major (general) health impairments according to the medical details in patients’ dental records and with no clinically indicated (general) health care need, and care-dependent participants (group B). Group B participants had care-dependency level 1 (low dependency) through 6 (high dependency) according to the Dutch national care-dependency classification system. In this system, the intensity and type of care needed is regularly determined by a medical authority and based on the intensity and type of clinically indicated functional impairments (in the physical and/or psychological and/or social domain) . People with care-dependency level 5 were excluded since this level comprises predominantly cognitive impairment.

Self-reported data

Data on OHRQoL, self-reported general health’( very bad/bad/moderate/good/very good ), and treatment demand ( y/n ) were obtained by a questionnaire. OHRQoL was measured using the validated Dutch version of the Geriatric Oral Health Assessment Index (GOHAI) . The GOHAI consists of 12 questions on experienced functional and psychosocial impacts of oral health, scored on a 1–5 point Likert scale with higher scores indicating better OHRQoL. Additional data were collected on age, gender, and socioeconomic status (SES). SES (high, middle, and low) was determined on the basis of the highest level of either education (high, middle, and low) or last held occupation (according to the ISCO-08 classification ).

Since we assumed, based on patient records and the absence of a medically assessed care-need indication, that group A participants generally would have good health, we assessed impairments in the physical, mental and social health domains only for group B. Physical and mental health were assessed through the SF-12(Short Form)health survey . Answers to SF-12 questions were used to compute a physical component summary (PCS) score and a mental component summary (MCS) score. The scores are calculated using indicator variable weights, with the mean scores set at 44.06 (PCS) and 49.50 (MCS), which are the age and sex-standardized norm scores for the 70–79 age group in the Dutch population . Higher scores indicate better health. Social support was used as a proxy measure for social health, and was assessed through the validated ENRICHD Social Support Index (ESSI) . The ESSI consists of seven items/questions that assess the four defining attributes of social support: emotional, instrumental, informational, and appraisal . Six out of seven ESSI items are scored on a 1–5 point Likert scale, higher scores reflecting more social support. The last item covers marital status, where 1 point (without partner) or 5 points (with partner) can be obtained. Total scores of 7–18 represent a low level of social support; 19–26 medium and 27–35 a high level of social support .

Statistical analysis

SPSS version 22.0 (SPSS Inc., Chicago, IL, USA) was used for data analyses. Given the robustness of parametric tests to deviations from data normality, and given the higher power of these tests , we chose to use parametric tests for analyses of GOHAI outcomes. One-way ANOVA, followed by Welch F tests in case of non-homogeneous variances, followed by posthoc tests (Tukey or Tamhane’s T2 in case of non-homogeneous variance) were used to assess significant differences (p < 0.05) in mean GOHAI scores between categories. Continuous variables (age, number of teeth and number of occluding pairs) were correlated with GOHAI scores using Pearson’s correlation coefficients. Multiple linear regression analyses were performed to examine, for both groups A and B, the associations between OHRQoL (dependent variable) and three sets of explanatory variables which were added in three stages using the ‘enter’ method. In order to check potential violation of assumptions for linear regression tests, homoscedasticity (equal variances) and normality of standardized residuals were checked through Q-Q-plots . Multicollinearity (inter-dependency of variables) was tested through calculation of variance inflation factors (VIFs): variables with VIFs ≥ 2.5 were not included in the models. The first model included background variables with a known potential confounding effect (age, gender, and SES). In a second model, clinically assessed variables that represented oral health (including prosthodontic status) were added. In a third model, variables that represented general health (including care-dependency) were added. In order to compare differences in pattern of associations (between GOHAI outcomes and OH and GH variables) between group A and B, predicted GOHAI outcomes according to the third models for group A and B were calculated for all respondents (i.e. n = 235) based on the models’ respective constants and partial regression coefficients. Differences in association patterns were analyzed using a Pearson correlation test and through calculation of the mean of the absolute differences between predicted GOHAI outcomes based on Model A and Model B. Differences were visualized through a scatterplot of paired predicted outcomes, where points should lie close to the line x = y in case of small differences.

Additional regression analysis was carried out in Group B in order to examine the associations between three specific health domains (physical, mental, and social) and OHRQoL.

Minimal sample size was determined to allow for multiple linear regression with a maximum of 9 (Group A) and 12 (Group B) independent variables, based on the presumption that a minimum of 10 participants for each independent variable is required for meaningful outcomes .

Methods

Population and samples

Participants aged 65 years and over who were cognitively alert were recruited in two populations: a population of care-independent, home dwelling people (Group A) and a population of care-dependent institutionalized people with varying levels of care-dependency (Group B). Group A participants were recruited through convenience sampling of patients of the university dental clinic who visited the clinic for a regular check-up visit. Group B participants were recruited through contact managers of randomly chosen residential aged care facilities (RACFs) in South-East Netherlands. Purposive sampling was applied in both groups, aimed at achieving adequate numbers of subjects with regard to variables whose outcomes were known prior to sampling and that have been found to influence OHRQoL (i.e. gender, prosthodontic status (dentate/edentate) and levels of care-dependency) . The study was approved by the Medical Ethics Committee (CMO) of the Radboud University Medical Center Nijmegen (CMO ref. 2012/194). All recruits gave informed consent in writing to participate in the study.

Clinical data and care-dependency

Clinical data were obtained through clinical oral examination according to WHO criteria by final year calibrated dental students (all κ’s > 0.82; overall κ = 0.87; agreement = 90.1%) and calibrated final year dental hygiene students (all κ’s > 0.66; overall κ = 0.74; agreement = 84.4%). Data included number and position of teeth, presence of carious teeth, and presence of removable dental prostheses (RDPs). Clinical assessment of treatment need (y/n) was also recorded and comprised any need for professional dental treatment including reline, rebase or replacement of RDPs, and periodontal treatment. With regard to prosthodontic status, three categories were distinguished: people with at least one natural tooth and without RDPs, people with at least one natural tooth and one or more RDPs, and people without natural teeth with complete RDPs. With regard to health status and care-dependency, we distinguished between care-independent participants (Group A) with no major (general) health impairments according to the medical details in patients’ dental records and with no clinically indicated (general) health care need, and care-dependent participants (group B). Group B participants had care-dependency level 1 (low dependency) through 6 (high dependency) according to the Dutch national care-dependency classification system. In this system, the intensity and type of care needed is regularly determined by a medical authority and based on the intensity and type of clinically indicated functional impairments (in the physical and/or psychological and/or social domain) . People with care-dependency level 5 were excluded since this level comprises predominantly cognitive impairment.

Self-reported data

Data on OHRQoL, self-reported general health’( very bad/bad/moderate/good/very good ), and treatment demand ( y/n ) were obtained by a questionnaire. OHRQoL was measured using the validated Dutch version of the Geriatric Oral Health Assessment Index (GOHAI) . The GOHAI consists of 12 questions on experienced functional and psychosocial impacts of oral health, scored on a 1–5 point Likert scale with higher scores indicating better OHRQoL. Additional data were collected on age, gender, and socioeconomic status (SES). SES (high, middle, and low) was determined on the basis of the highest level of either education (high, middle, and low) or last held occupation (according to the ISCO-08 classification ).

Since we assumed, based on patient records and the absence of a medically assessed care-need indication, that group A participants generally would have good health, we assessed impairments in the physical, mental and social health domains only for group B. Physical and mental health were assessed through the SF-12(Short Form)health survey . Answers to SF-12 questions were used to compute a physical component summary (PCS) score and a mental component summary (MCS) score. The scores are calculated using indicator variable weights, with the mean scores set at 44.06 (PCS) and 49.50 (MCS), which are the age and sex-standardized norm scores for the 70–79 age group in the Dutch population . Higher scores indicate better health. Social support was used as a proxy measure for social health, and was assessed through the validated ENRICHD Social Support Index (ESSI) . The ESSI consists of seven items/questions that assess the four defining attributes of social support: emotional, instrumental, informational, and appraisal . Six out of seven ESSI items are scored on a 1–5 point Likert scale, higher scores reflecting more social support. The last item covers marital status, where 1 point (without partner) or 5 points (with partner) can be obtained. Total scores of 7–18 represent a low level of social support; 19–26 medium and 27–35 a high level of social support .

Statistical analysis

SPSS version 22.0 (SPSS Inc., Chicago, IL, USA) was used for data analyses. Given the robustness of parametric tests to deviations from data normality, and given the higher power of these tests , we chose to use parametric tests for analyses of GOHAI outcomes. One-way ANOVA, followed by Welch F tests in case of non-homogeneous variances, followed by posthoc tests (Tukey or Tamhane’s T2 in case of non-homogeneous variance) were used to assess significant differences (p < 0.05) in mean GOHAI scores between categories. Continuous variables (age, number of teeth and number of occluding pairs) were correlated with GOHAI scores using Pearson’s correlation coefficients. Multiple linear regression analyses were performed to examine, for both groups A and B, the associations between OHRQoL (dependent variable) and three sets of explanatory variables which were added in three stages using the ‘enter’ method. In order to check potential violation of assumptions for linear regression tests, homoscedasticity (equal variances) and normality of standardized residuals were checked through Q-Q-plots . Multicollinearity (inter-dependency of variables) was tested through calculation of variance inflation factors (VIFs): variables with VIFs ≥ 2.5 were not included in the models. The first model included background variables with a known potential confounding effect (age, gender, and SES). In a second model, clinically assessed variables that represented oral health (including prosthodontic status) were added. In a third model, variables that represented general health (including care-dependency) were added. In order to compare differences in pattern of associations (between GOHAI outcomes and OH and GH variables) between group A and B, predicted GOHAI outcomes according to the third models for group A and B were calculated for all respondents (i.e. n = 235) based on the models’ respective constants and partial regression coefficients. Differences in association patterns were analyzed using a Pearson correlation test and through calculation of the mean of the absolute differences between predicted GOHAI outcomes based on Model A and Model B. Differences were visualized through a scatterplot of paired predicted outcomes, where points should lie close to the line x = y in case of small differences.

Additional regression analysis was carried out in Group B in order to examine the associations between three specific health domains (physical, mental, and social) and OHRQoL.

Minimal sample size was determined to allow for multiple linear regression with a maximum of 9 (Group A) and 12 (Group B) independent variables, based on the presumption that a minimum of 10 participants for each independent variable is required for meaningful outcomes .

Results

Sample characteristics are displayed in Table 1 . GOHAI scores in Groups A and B were similar: mean Group A: 51.6 ± 7.4 (range 29–60), mean Group B: 52.1 ± 6.7 (range 26–60); Clinical oral health outcomes of care-dependent participants were significantly worse than those of care-independent participants (16.8 ± 8.2 vs . 10.7 ± 5.1 missing teeth, 5.4 ± 5.0 vs. 9.4 ± 3.5 occluding pairs of natural teeth; 59% vs. 37% subjects with clinically assessed treatment need, 57% vs. 11% subjects with one or more carious teeth). Perceived general health of care-dependent participants was also worse than that of care-independent participants: 24% of Group B subjects reported ‘bad’ or ‘very bad’ health vs. 1% of Group A subjects, whereas ‘good’ or ‘very good’ general health was reported by 44% of Group B subjects vs. 79% of Group A subjects.

Table 1
Group characteristics and mean GOHAI scores of care-independent home dwelling older people (group A) and care-dependent older people in residential aged care facilities (group B).
Group A (n = 109) Group B (n = 126)
Group characteristics GOHAI
mean (SD)
GOHAI
mean (SD)
GOHAI score 51.6 (7.4) 52.1 (6.7)
Age (years); mean (SD) 73.1 (5.4) 85.4 (7.1)
Gender; [n (%)]
-female 52 (48) 50.3 (8.0) 73 (58) 52.2 (6.9)
-male 57 (52) 52.7 (6.8) 53 (42) 51.9 (6.6)
SES (Group A: n = 107); [n (%)]
– low 19 (18) 48.8 (8.7) 48 (38) 51.3 (6.4)
– medium 54 (50) 51.4 (7.7) 50 (40) 51.5 (7.6)
– high 34 (32) 53.0 (6.0) 28 (22) 54.2 (5.4)
Prosthodontic status; [n (%)]
– dentulous: natural teeth only 44 (40) 54.4 (4.6) a 32 (25) 52.1 (7.2)
– dentulous + partial and/or complete RDPs 22 (20) 50.7 (7.4) ab 36 (29) 52.0 (5.8)
– edentulous; complete RDP in both jaws 43 (39) 49.1 (8.8) b 58 (46) 52.2 (7.1)
Missing teeth; mean (SD)* 10.7 (5.1) 16.8 (8.2)
Occluding pairs of natural teeth; mean (SD)* 9.4 (3.5) 5.4 (5.0)
At least one carious tooth; [n (%)]*
– yes 7 (11) 52.5 (5.1) 39 (57) 50.2 (5.8) a
– no 59 (89) 53.2 (6.0) 29 (43) 54.4 (6.6) b
Clinically assessed treatment need; [n (%)]
– yes 40 (37) 48.8 (8.6) a 74 (59) 50.4 (6.2) a
– no 69 (63) 53.7 (6.2) b 52 (41) 54.4 (6.9) b
Self-reported treatment demand
(Group A: n = 108); [n (%)]
– yes 45 (42) 48.1 (8.2) a 35 (28) 46.2 (7.4) a
– no 63 (58) 54.8 (5.3) b 91 (72) 54.3 (4.9) b
Self-reported general health
(Group A: n = 107); [n (%)]
– very bad or bad 1 (1) 59.0 (-) 31 (24) 50.6 (7.8)
– moderate 21 (20) 47.6 (4.5) a 40 (32) 52.0 (6.2)
– good 74 (69) 52.0 (6.8) ab 50 (40) 52.8 (6.7)
– very good 11 (10) 55.6 (5.0) b 5 (4) 54.8 (3.9)
Care dependency level; [n (%)] NA NA
– 1 26 (21) 52.2 (8.3)
– 2 31 (24) 53.3 (6.5)
– 3 19 (15) 50.1 (8.1)
– 4 29 (23) 53.2 (4.9)
– 6 21 (17) 50.3 (5.7)
SF-12 physical component (PCS); [n (%)] NA NA
– Lower than 44.06 104 (83) 52.2 (6.5)
– Higher than 44.06 22 (17) 51.5 (8.1)
SF-12 mental component (MCS); [n (%)] NA NA
– Lower than 49.50 62 (49) 51.0 (7.0)
– Higher than 49.50 64 (51) 53.2 (6.4)
Social Support (ESSI) NA NA
(Group B: n = 125); [n (%)] 33 (26) 48.9 (6.5) a
– low 49 (39) 51.9 (7.4) ab
– medium 43 (35) 54.6 (5.0) b
– high
Only gold members can continue reading. Log In or Register to continue

Stay updated, free dental videos. Join our Telegram channel

Jun 19, 2018 | Posted by in General Dentistry | Comments Off on Oral health-related quality of life and associated factors in a care-dependent and a care-independent older population

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos