Response Letter to Savikko et al

Abstract

We are pleased that Savikko et al. have studied the association between chewing ability and dementia in elderly persons living in long-term care (LTC). Although their findings do not agree with ours, their study adds to the existing literature concerning the complex association between oral health and cognition. As Savikko et al. point out, differences in the studied populations and their stages of cognitive decline may contribute to the differing results. Savikko et al. studied a LTC population, whereas our study was representative of the national population aged 77 and older (including people in LTC). The indicators used for cognition, chewing ability, and depression, for example, differ. Therefore, the prevalence rates for dependent and independent variables differed in the two studies. The Finnish study used a diagnosis of dementia confirmed by medical records, whereas the Swedish study used an indicator of cognitive impairment. In the Finnish LTC sample, 73.2% had dementia; in the Swedish population, 20.1% showed signs of cognitive impairment. The Finnish sample was older than the Swedish sample; there were also more women, more people who had had strokes, and more people with more than 8 years of education in the Finnish sample. The prevalence of chewing difficulty among the LTC residents was 28.9%, whereas it was only 20.8% in our national sample. The associations between some of the variables also differed in the two studies. The Finnish study reported a higher prevalence of depression in people without dementia (24.9%) than in those with dementia (18.1%). The relationship in the Swedish study was inverse; in people without cognitive impairment, 12.5% showed signs of depression, compared with 28.7% of persons with cognitive impairment (calculated from Table 1 in the article). In both studies, the association was significant. Because depression is an important confounding factor,1, 2 this may have contributed to the contradictory results. Education, another important confounding factor, also differed in the two populations. There was no significant association between education and dementia in the LTC sample, whereas in our study, more education was significantly protective for cognition. These different prevalence rates and associations reflect the complexity of the association between oral health and cognition, as Savikko et al. noted. Studies in different populations—at different ages and different levels of disability—can add to a better understanding of this association. Studies that follow people over time will provide even more insight. Ultimately, these epidemiological findings must be combined with clinical and animal experimental studies to understand mechanisms and causality. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Authors Contributions: Duangjai Lexomboon: literature review, writing drafts, and finalizing of the manuscript. Mats Trulsson, Inger Wårdh: comments on the manuscript. Marti G. Parker: comments and finalizing of the manuscript. Sponsor's Role: No role in any part of the original study or this letter.