Oral health matters in cognitive impaired aged residents in geriatric care facilities: A cross‐sectional survey

Abstract Aims To investigate the oral health status of aged residents and explore the possible factors associated with oral health and the effect of cognitive impairment (CI) on it in geriatric care facilities (GCF) using oral health assessment tool (OHAT) in Shanghai, China. Background Oral health is closely associated with overall health. Problems like missing teeth, dental caries, dental pain, periodontal diseases, oral infections and dysphagia are common in GCFs. Furthermore, residents in GCFs with CI are becoming a dominant group and this prevalence increases with age. Detection of oral problems earlier and taking oral care actions is required for these special populations. Methods The study was an observational cross‐sectional study conducted in 42 GCFs. Data were collected from 657 subjects, including oral health assessment (OHAT), cognitive impairment (Mini‐Mental State Examination, MMSE) and respondents’ characteristics. The subjects were divided into CI group and non‐CI group based on MMSE. Oral health conditions were compared between the two groups. Results Oral health status in the CI group was significantly worse than that in the non‐CI group (p < .001) with four OHAT dimensions (tongue (p = .0007), saliva (p = .0011), natural teeth (p = .0155) and oral cleanliness (p < .001)). The worst dimension was natural teeth. Debris and plaque index (p < .001), oral odour (p < .001), chewing function (p = .0151) and swallowing function (p = .0405) were worse in CI group than those in non‐CI group. In the CI group, providing oral care was a protective factor in oral health (OR = 0.600 95CI% (0.39–0.92)) and wearing dentures was a risk factor (OR = 2.09, 95CI% (1.31–3.32)), while the similar effects were not found in non‐CI group. Conclusions Oral health status among aged residents in GCFs in China was worse among individuals with CI. Caregivers in GCFs should focus more on seniors’ oral health with CI. Relevance to Clinical Practice Residents who are suffering from CI are more vulnerable to have oral problems. Regular and proper oral health check‐ups in daily nursing work to define oral problems of residents are significant. Nursing staff should pay more attention to oral assessment and effective intervention.

surveyed residents in Germany had moderate or deep periodontal pockets (Ziebolz et al., 2017) and visible plaque was found in all residents in Victorian nursing homes and more than 25% of individuals had plaque covering over one-third of at least one tooth (Hopcraft et al., 2012a). A research in North Carolina discovered that denture hygiene was poor either, especially the upper dentures (especially unexposed surfaces) (Zimmerman et al., 2017). Meanwhile, oral care is not in compliance with the existing international evidence-based best practice guidelines and protocols; therefore, the residents do not regularly receive the best practice. A study in Canada found that 59% of care providers felt rushed in their last shift when providing oral care to residents and 19% did not complete oral care (Knopp-Sihota et al., 2015).
In GCFs, residents with dementia or cognitive impairment (CI) are becoming a dominant group and this prevalence increases with age. CI is a status of cognitive decline, the gradual loss of one's ability to learn, remember, pay attention and make decisions. Oral problems are more compromised in these residents and oral hygiene is unacceptable (Ziebolz et al., 2017) and may be at higher risk of developing oral diseases. Providing oral care is further complicated by cognitive or behavioural impairments, resistance and lack of cooperation, from a report of Australia research (Chalmers & Pearson, 2005); low priority of dental services among geriatric residents, poor support for caregivers and lack of staff, time, knowledge, GCF protocols, and regulations (Chalmers & Pearson, 2005;Hoben et al., 2017). This is an alarming phenomenon worldwide documented in numbers of studies reporting poor care and outcomes. Some exploratory mouth-care projects have been developed; however, the effects are rarely obvious and often temporary. Proper oral health among the institutionalized aged residents with CI around the world is not yet secured. As a result, improving oral health of these population becomes a vital topic for future studies.
In China, the similar dilemma exists. As the population ages, more elderly Chinese individuals are living in GCFs with declining cognitive function, exhibiting various types and degrees of CI, accompanied by chronic and aggressive mental and behavioural abnormalities, all of which complicate daily care (Chan et al., 2013). However, studies on the oral health of this specific group of GCF individuals were not reported and individuals with dementia or CI in GCFs have always been excluded from researches.
The purpose of this study was to explore the oral health situation of aged residents and the effects of CI on oral health. Results are useful not only to the nurses who working in geriatric fields, but also to interdisciplinary dental audience, for guiding future quality improvement efforts.

| Participants
We calculated a sample size of 531 to allow for stratification by GCFs depends on an α of 0.05, a relative error of sampling of 3% and non-response rate of 10% with assuming a prevalence of server oral health issues of 20%. A proportional sampling method (10%) with stratification factor of GCF was used to obtain a representative sample of aged residents in all of 42 GCFs in shanghai.
A total of 700 residents were obtained by using computer-generated random numbers. The inclusion criteria were over 60 years and stayed in GCF at least one month. The exclusion criteria were having any life-threatening, serious medical problems, inability to undergo assessment or cooperate. And 43 subjects were excluded because of having any of above situations. Finally, 657 subjects participated in oral health status assessments and completed the survey for the study.

| GCF characteristics
Corresponding information on GCFs was obtained from administrators based on face-to-face interviews, including facility size, private or public ownership, established date, doctor-patient ratio, nursepatient ratio, nurse assistant-patient ratio and GCF's policy on regular oral health assessment.

| Demographical characteristics
Sex, age, marital status, BMI, concomitant disease, cognitive impairment, Barthel Index, length of GCF-stay, payment with medical insurance and regular family visits were collected.  (Folstein et al., 1975). Participants were asked to solve 30 tasks in six dimensions: orientation, registration, attention, calculation, recall and language. Correctly executed exercises were awarded 1 point, with a total score ranging from 0 to 30, MMSE scores of <24 are indicative of cognitive impairment or dementia. This scoring is considered the gold standard for dementia screening.

| Oral characteristics
Trained nursing staff performed the following non-invasive dental examinations and measurements with a headlight in the participant rooms. Double checks were performed to ensure the integrity and accuracy of the information. Residents were examined using OHAT, it is a simple indicator of overall oral health validated by  and is widely translated to different versions and applied. It consists of eight items: lips, tongue, gum and tissues, saliva, natural teeth, dentures, oral cleanliness and dental pain. This tool is not requiring patients to express themselves, so persons with cognitive disorders could also be evaluated. Scores on each item range from 0 (healthy) to 2 (unhealthy). The total score ranges from 0 to 16, can be classified into three categories: "healthy mouth" (0-3); "changing mouth" (4-8); "unhealthy mouth" (9-16). The oral environment refers to plaque index and oral odour. Plaque index was measured visually using a probe to examine for soft accumulations, with scores ranging from 0 to 3 (Oltdrn, 1972). Oral odour was also examined by instructing participants to open their mouths and make an "ah~" sound for 5 s; the odour was then judged and recorded in five degrees. Oral function scoring included determination of chewing function in 3 degrees and swallowing function in 5 degrees.
Information on oral health behaviours, including brushing, rinsing, oral care procedure, was collected via interview and observation.

| Data collection
Data collection was conducted from February to October in 2019.
In each GCF, all the research staff were trained and calibrated.
Participant's cognitive evaluation was completed by trained nurse first and then the nurse continued to finish the oral assessment.
Double checks between two trained nurses were performed to ensure the integrity and accuracy of the information. Data collection for this article was conducted from February to October in 2019.

| Ethical considerations
This study was approved by the ethics committee. Written informed consent was signed by all participants or their legally authorized representatives if resident was unable to. We also considered the freedom to withdraw consent during the process, the absence of any detriment in doing so and the anonymity of data.

| Statistical analysis
Continuous variables were expressed as mean with standard deviation or median with 25 percentile and 75 percentile and compared using t test or Wilcoxon signed-rank test, as per the normality of data. Categorical variables were expressed as N (%) and compared using chi-square or Fisher's exact tests. For oral health and oral health-related behaviours, CI group was compared with non-CI group based on MMSE, respectively. A multivariate logistic regression analysis was performed to identify the possible associations between factors and oral health status. Sex, age, cognitive impairment, education level, regular family visit, institution ownership, providing oral care or not, denture use, length of time to wear dentures, chewing function and swallowing function were entered in the model and selected by a method of stepwise. In the multivariate analysis, we also considered the potential interactions among variables. Values of p < .05 were considered significant. All statistical analyses were performed by SAS 9.4(SAS, Cary, NC).

| Measuring oral health status using OHAT
The overall distribution of OHAT scores in two groups is presented in Table 2. Overall, oral health status in the CI group was significantly worsen than that in the non-CI group (p < .001), only 281 (68.2%) individuals had "healthy mouth," and 131 (31.8%) showed signs of "changing mouth" or "unhealthy mouth." On all eight OHAT dimensions, the two groups showed significant differences, with the CI group exhibiting poorer oral health, especially regarding to tongue (p = .0007), saliva (p = .0011), natural teeth (p = .0155) and oral cleanliness (p < .001), the proportions scored as "changing mouth" and "unhealthy mouth" were 22.8%, 23.8%, 64.8% and 53.9% in the CI group and 13.0%, 13.5%, 55.8% and 41.6% in the non-CI group, respectively. The worst dimension was natural teeth, in CI group, 24.3% had over 4 decayed or broken teeth/roots, or less than 4 teeth. The second worsen dimension was oral cleanliness, 10.7% residents with CI showed food particles/tartar/plaque in most areas of the mouth or severe halitosis (Table 3).
Concerning on the influences of oral conditions on OHAT, the CI group showed significantly worsen in oral plaque index (p < .001), oral odour (p < .001), chewing function (p = .0151) and swallowing function (p = .0405) than the non-CI group. In CI group, 58.7% of residents were in the "healthy mouth", with no dental plaque; however, in "unhealthy mouth," 77.7% of residents were with plaque; of these, 33.3% had dental plaque. Similarly, as oral health transitions into the unhealthy category, the severity of oral odour increases.
In seniors with "healthy mouth," 56.2% had no oral odour, whereas among those with "unhealthy mouth," 100% exhibited oral odour and 25% had severe oral odour (

| Multivariate analysis of the factors associated with OHAT
A multivariate logistic regression analysis was performed to explore the potential influence factors of oral health status. We found that residents in GCFs who receiving oral care had better oral health status (With oral care/without oral care, OR is 0.63, 95CI% (0.43-0.94)). Denture use was a risk factor, those who wearing dentures had worse oral health (No denture use/dentures-use, OR is 1.95, 95CI% (1.34-2.86)). However, CI was a protective factor in the initial regression analysis. Then a stratified analysis was implemented and the results showed that (

| D ISCUSS I ON
According to our study, oral health status was not desired overall and the situation was even worse among individuals with CI. Oral health status was positively correlated with oral environment and oral function and the CI group showed a poor situation. Furthermore, oral health behaviours were inadequate and most GCFs failed to meet In this study, institutional elderly oral health was generally not optimistic and was worsened with CI. Oral problems already occurred among 31% residents with CI, who required medical interventions immediately. Natural teeth appear to be the most problematic domain, as tooth loss, tooth decay or root damage was more prevalent, which was consistent with previous reports that elderly patients with CI were more likely to have fewer remaining natural teeth (Cocco et al., 2018).
We also found oral cleanliness as a worsen dimension especially in CI group, while non-communicable diseases are likely to increase physical inability and self-care dependence as showed in previous studies (Chen et al., 2019;Wu et al., 2016). In dental pain, CI groups have no difference with cognitively healthy residents, perhaps they could not express themselves properly and existing abnormal behaviours (yelling/screaming; facial convulsion) influenced nurses' assessment. The CI group showed poorer care than the non-CI group in gums and tissues and saliva categories, indicating that nurses should evaluate gum tissue and saliva due to the use of long-term psychotropic drugs, which may lead to reduced saliva secretion with accompanying symptoms like dry mouth, gum swelling and discomfort (Janssens et al., 2017).
Poorer oral condition is associated with diminished status of oral functions, in CI group, among the residents with "healthy mouth," 68% retained normal chewing function and 75.5% had normal swallowing function. However, 55.6% of those in the "unhealthy mouth" group experienced total chewing loss and 55.6% had severe swallowing disorder. Similar as some studies highlighted the importance of oral health as a predictor of oral functions and frailty in older age (Hakeem et al., 2019). On the other hand, chewing and swallowing dysfunction, which results in more food residue left in the mouth, affects the oral environment, the appearance of plaque and oral odour and hence accelerates the deterioration of oral health and even leads to malnutrition.
According to previous studies, increased oral health behaviour helps to improve oral health. However, oral health behaviour is relatively insufficient in GCFs. In our investigation, fewer than 10% of elderly residents received any oral health assessments and 90% were not concerned about oral problems. Even a basic twice-a-day tooth-brushing habit seemed difficult to be adopted, companied by the situation that less than 50% (47.35%) of patients completed the tooth cleaning twice activities and the quality of brushing varied, which was the same case in other researches (Hopcraft et al., 2012a).
For those requiring oral care, GCFs have also failed to provide oral care that meets requirements. Our study shows providing oral care is a protective factor in oral health. However, there was only 26.3% of the residents received oral care in GCFs. Interviews with nursing staff revealed difficulties implementing cleaning activities. It was also common for them to encounter sudden resistance, which could interrupt the caring process. Nurses were unwilling to provide oral  Note: Distribution of scores of per item are expressed as N (%); total score is expressed mean ± SD. the score of each item is from 0-3, the oral health condition is from healthy to unhealthy.
The limitations of this study are related to the recruitment of residents only in Shanghai, which may limit the generalizability of the results. In the process of data collection, cognitive evaluation and oral assessment were conducted by the same nurse, thereby resulting in a possible evaluation bias. Furthermore, the cross-sectional nature of this study prevented examination of causality between oral health status and cognition impairment. We are planning future longitudinal follow-up studies with a larger sample size to confirm our findings.

| CON CLUS IONS
Oral health status among aged residents in GCFs was worse among individuals with CI. In the future, oral-related quality indicators should be part of the process of accreditation in the quality of care in GCFs and systematic oral health surveillance is necessary, especially in CI population.  have the awareness to focus on oral assessment and appropriate intervention.

ACK N OWLED G M ENTS
The authors thank all the staff of GCFs participating in this study for their ongoing efforts to promote the quality of care in GCFs settings.
This study was funded by a grant from Standardization Pilot Program of Shanghai Quality and Technical Supervision Bureau (S18-04-017).
The funding source has not influenced the content of the paper.

CO N FLI C T O F I NTE R E S T
All authors declare there are no conflicts of interest.

AUTH O R CO NTR I B UTI O N
Zhang Lingjuan and Gu Liyan designed the study; Chen Wenyao performed the experiments. Li Xianchen contributed to the data analysis; Chen Lan and Gu Liyan contributed to the writing of the paper.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analysed during this study are included in this article.