jarlife journal
Sample text

AND option

OR option

MALNUTRITION POINT-PREVALENCE FROM 2012 TO 2019 AND ASSOCIATED HEALTH-OUTCOMES IN ADULT PATIENTS IN RURAL HOSPITALS

E. Lopez1, M. Banbury2,3, E. Isenring4, S. Marshall5

1. MNutrDietPrac, Accredited Practising Dietitian, Faculty of Health Sciences and Medicine, Bond University, Australia; 2. BSc (Hons) Applied Human Nutrition & Dietetics. Accredited Practising Dietitian. Northern NSW Local Health District; 3. Faculty of Health Sciences and Medicine, Bond University, Australia; 4. PhD. Advanced Accredited Practising Dietitian. Bond University Nutrition & Dietetics Research Group, Faculty of Health Sciences and Medicine, Bond University, Australia; 5. BNutr&Diet (Hons), PhD. Accredited Practising Dietitian. Bond University Nutrition & Dietetics Research Group, Faculty of Health Sciences and Medicine, Bond University, Australia

Corresponding Author: Skye Marshall, Nutrition & Dietetics, Bond University, Robina, Queensland, 4226, Australia. Email: skye_marshall@bond.edu.au; Phone: 07 5595 3337.

J Aging Res Clin Practice 2019;8:91-97
Published online January 20, 2020, http://dx.doi.org/10.14283/jarcp.2019.16

Abstract

Background: Malnutrition negatively impacts hospitalised patients and the healthcare system. Objectives: 1) report point-prevalence of hospital malnutrition from 2012 to 2019; and 2) determine if there was an association between nutrition status and health-related outcomes. Design: Point-prevalence of malnutrition was determined by three (2012, 2014, and 2019) cross-sectional studies. Health-related outcomes, assessed by a prospective cohort study in 2014, were length of stay, in-hospital mortality, hospital readmission, infection, falls, fractures, and pressure wounds. Setting: three Australian rural hospitals. Participants: Adult inpatients. Measurements: Nutrition status was assessed with the Subjective Global Assessment (SGA) tool. Results: Malnutrition point prevalence was 39% in 2012 (n=62), 48% in 2014 (n=128), and 28% in 2019 (n=96); where the prevalence in 2019 was significantly lower than in 2014 (p<0.017). The 2019 (median age 70 years) sample was younger than the 2012 (median age 80 years) and 2014 (median age 78 years) samples (p<0.05). Mortality and falls rate were higher in the severely malnourished participants (p=<0.05); and severe malnutrition may predict mortality (Adjusted OR: 3.47 (95%CI: 0.94, 12.78] p=0.061). Conclusions: Nutrition status did not predict other health-related outcomes. The rate of malnutrition in rural hospitals was consistently high and may increase the risk of in-hospital mortality.

Key words: Malnutrition, hospitals, nutrition assessment, subjective global assessment, mortality.

Introduction

Protein-energy malnutrition (herein referred to as ‘malnutrition’) negatively impacts the patient and healthcare system alike (2, 3), a major concern as the prevalence has been reported internationally at 30-50% across inpatient and residential settings, and 1-25% across community settings (4-8). Malnutrition is the unintended loss of lean mass (muscle, immune and blood cells, viscera), with or without fat loss, due to inadequate intake, uptake, and/or utilisation of protein and energy to meet requirements (4, 9). Older adults are at greater risk of malnutrition due to their susceptibility of aetiological factors including psychological, socio-economic, and physiological changes and an overall increase in multi-morbidities and polypharmacy (9). A consequence of malnutrition is further morbidity, requiring increased healthcare resources including but not limited to hospital beds, multidisciplinary staff, and pharmaceutical and nutritional medicine (10). In particular, malnutrition increases risk of infection, pressure ulcers, poor wound healing, decreased response to medical treatment and pharmaceuticals, decreased respiratory function, decreased muscle repair, and overall functional impairment; leading to decreased quality of life and increased risk of mortality (3, 6, 11).
While several large-scale studies have reported the prevalence and outcomes of hospital malnutrition, the rural context requires specific examination as populations in rural areas are ageing more rapidly than in urban areas (11, 15-17). Rural areas face increased challenges in providing health and aged care due to the higher cost of establishing and delivering services, the limited availability of and access to health professionals, and less availability of informal care networks (12-14). Not only is access to health care more limited in rural areas, rural-dwelling older adults are also more in need of health and aged care services. A recent meta-analysis and meta-regression of international data found the prevalence of malnutrition in rural-dwelling older adults living at home was double that of urban-dwelling older adults (5). Therefore, the prevalence and health-related outcomes of malnutrition in rural hospitals is of interest, so that policies may appropriately support patients in the continuum of care from hospital to home or residential care.

Research aims

In adult patients admitted to three rural hospitals in Australia, the aims of this study were to: 1) report point-prevalence of malnutrition from 2012 to 2019; and 2) determine if there was an association between nutrition status and health-related outcomes.

Materials and Methods

Study design

The point-prevalence of malnutrition was assessed using three cross-sectional studies conducted in 2012, 2014, and 2019. The association between malnutrition and health-related outcomes was evaluated using a prospective observational study in 2014. Participants gave their verbal consent to participate in the study. The project was approved by the Human Research Ethics Committees in April 2018 (QA249) as a quality assurance project. This study has been reported according to the STROBE Statement for cohort studies (15) and was retrospectively registered with ANZCTR (ACTRN12619000342112) (19).

Setting and sample

All three hospitals within a rural government-funded local health district in northern New South Wales, Australia, were conveniently sampled in 2012. Reflecting the staffing resources available for each cross-sectional study, the medical, surgical, general (not diagnosis or treatment specific), and/or rehabilitation wards were sampled (Table 1).

Table 1 Rural hospitals and wards sampled by the three cross-sectional studies

Table 1
Rural hospitals and wards sampled by the three cross-sectional studies

* This sample was also that used in the prospective observational study.

The prospective cohort study was conducted using the 2014 sample due to availability of data and its larger sample size. Patients were eligible if they were 18 years or older and were admitted as inpatients to study sites during the recruitment phase of one to two weeks. No exclusion criteria were applied.

Participant characteristics and potentially confounding variables

Participant characteristics of age (years) and sex (male/female) were recorded for all participants. The 2014 sample were also described by comorbidities and medications. The number of active comorbidities were categorised into medical diagnostic groups: cancer, digestive, musculoskeletal, circulatory, respiratory, nervous, skin, reproductive, kidney, infectious, endocrine, injuries, ear, blood, and other. The number of medications were categorised into 24 drug classes based on the profile of medications recorded from the cohort (16).

Outcomes

Malnutrition was determined by the SGA tool which rates patients as A = well nourished, B = mild-moderate malnutrition, or C = severe malnutrition (17, 18).  The presence of malnutrition was the primary outcome to answer the first research question (point-prevalence) and the independent variable to answer the second research question (health-related outcomes).
The primary health-related outcome was length of hospital stay, defined by the number of days including the day of admission and discharge. Secondary health-related outcomes were in-hospital mortality (yes/no), hospital readmission (yes/no), in-hospital fall (yes/no), fall in subsequent hospital admissions (yes/no), pressure ulcer (yes/no), fracture acquired in hospital (yes/no), urinary tract or respiratory tract infection (yes/no). Health-related outcomes were measured from the day of hospital admission to three months post-discharge.

Data Collection

In 2012 and 2014, nutrition status was assessed by department dietitians over a 7-day period within each ward using the SGA. In 2019, nutrition status was assessed by one student-dietitian (EL) using the SGA over two weeks. Only the student dietitian in 2019 received training in correct SGA use, whereas department dietitians were expected to be competent in nutrition assessment due to years of experience. The SGA is comprised of two main components: medical and physical assessment. Changes in weight, dietary intake, gastrointestinal symptoms, and nutrition-related functional capacity were observed from a combination of patient records and patient interview for the medical component; while evidence of oedema, ascites, and loss of subcutaneous fat and muscle was assessed during a physical examination to inform the physical component. A patient who demonstrated negative changes to their oral intake and failed to meet their nutritional requirements with evidence of muscle and fat deterioration were classified as exhibiting a degree of malnutrition. Participant characteristics and the health-related outcomes were observed from the medical record.

Data Analysis

Data analysis was completed using IBM SPSS Statistics 25 [IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp]. Descriptive statistics were used to summarise characteristics and outcome data. Continuous variables were considered non-normal if their skewness and kurtosis divided by their standard error exceeded +2 or -2; parametric variables were reported as mean (standard deviation) and non-parametric as median (IQR). Participants were considered “malnourished” if they were rated as SGA B or C, and “well-nourished” if rated SGA A. To determine if there was a significant difference in age, nutrition status, and sex between the samples, the Kruskal-Wallis H Test was applied. If significance was observed, post-hoc testing using Mann-Whitney U was used for pairwise comparisons between cohorts. Here, the Bonferroni correction was applied at 0.05 level and adjusted for three groups; therefore, the cut-off value for significance for Mann-Whitney U tests was at p<0.017. If significance was observed, post-hoc testing using Mann-Whitney U was used between pairs of cohorts. Post hoc power analysis using G*Power (version 3.1.9.2) was conducted on independent group means which showed significant difference. Extreme outliers (interquartile range rule with multiplier of 3) were removed from continuous variables. Binary logistic regression was used to determine the effect of the sample (2012, 2014, or 2019) on nutrition status, with age and sex as a confounding variables.
For the 2014 cohort, differences between nutrition status and outcomes were tested by the Mann-Whitney U Test or Chi-squared test. Associations with health-related outcomes were tested according to the level of severity, with patients considered as “malnourished” (SGA rating B or C), and “severely malnourished” (SGA rating C). Multiple linear regression was used to determine the impact of nutrition status on LOS, and multiple logistic regression was used to determine the impact of nutrition status on secondary outcomes, accounting for participant characteristics which met assumptions for the model. Statistical significance was considered at the p<0.05 level two tailed unless otherwise indicated.

Results

Participant characteristics

A total of 286 participants were recruited; n=62 in 2012, n=128 in 2014, and n=96 in 2019 (Table 2). There was no difference in the sex ratio or the prevalence of malnutrition between the sexes in any sample. The 2019 sample (median age 70 years) was found to be significantly younger than the 2012 (median age 80 years) and 2014 (median age 78 years) samples (p < 0.017). In the 2014 sample, a circulatory condition was the most common comorbidity experienced (72%), followed by musculoskeletal and respiratory conditions (40% and 38% respectively) (Table 3). Cancer was more prevalent in malnourished (33%) than well-nourished participants (16%) (p=0.028); where prevalence and total number of other comorbidities between groups were similar. Both groups had a median of six classes of medications prescribed during hospitalisation; where the malnourished group had a higher range (IQR 5, 7.25) than the well-nourished group (IQR 3, 6.25) (p=0.024). Malnourished participants were also more likely to be prescribed nutritional supplements (p=0.008) and proton-pump inhibitors (p=0.013), and less likely to be prescribed medication for insomnia (p=0.037).

Table 2 Age, sex, and malnutrition point-prevalence of the 2012, 2014, and 2019 participant samples

Table 2
Age, sex, and malnutrition point-prevalence of the 2012, 2014, and 2019 participant samples

* Kruskal-Wallis H Test applied followed by Mann-Whitney U tests to determine significance between cohort years. The Bonferroni correction was applied at 0.05 level. Cut-off value for significance for Mann-Whitney U tests at p=0.017 (0.05/3) for 3 pairwise comparisons between cohorts. Significance for age found between pairs 2012 vs 2019 and 2014 vs 2019 cohorts; † No significance found between sex across cohorts; ‡ Malnourished = B (mild-moderate malnutrition) and C (severe malnutrition) rating combined; § Kruskal-Wallis H Test applied followed by Mann-Whitney U tests to determine significance between cohort years. The Bonferroni correction was applied at 0.05 level. Cut-off value for significance for Mann-Whitney U tests at p=0.017 (0.05/3) for 3 pairwise comparisons between cohorts. Significance for malnourished patients only found between 2014- and 2019-year groups; || No significance found for sex effect on malnutrition across cohorts.

Table 3 Comorbidity and medication characteristics of 2014 sample

Table 3
Comorbidity and medication characteristics of 2014 sample

* Comorbidity data in the medical record was unavailable for n=4 participants; † Comparison of well-nourished (SGA rating A) and malnourished (SGA rating B or C) groups; ‡  Number of comorbidities experienced by a single participant; data presented mean (SD); § PPI = proton pump inhibitor; ||  Number of medication classes taken by a single participant; data presented median (IQR)

Point-prevalence of malnutrition from 2012 to 2019

Across the three time-points, malnutrition according to the SGA shows a peak in the 2014 cohort (48%), which was significantly higher than in 2019 (28%), but not 2012 (39%) (Table 2). The prevalence of participants assessed as severely malnourished (SGA rating C) decreased over time from 15% in 2012, 13% in 2014, to 2% in 2019 (p=0.005); whereas the prevalence of well-nourished fluctuated from 61% in 2012, 52% in 2014, to 72% in 2019 (p=0.005). In a model adjusted for age and sex, regression analyses found that only age was a predictor of malnutrition, where each year of life increased the odds of malnutrition by 2% (OR: 1.020 [95%CI: 1.003, 1.036] p=0.018) but explained only 5% of variation in the model.
‘Effect size’ and ‘chance of impact’ was reported for comparisons between year groups for nutrition status to determine the magnitude of the difference between groups and whether or not the outcome was likely to have an actual impact. The pairwise comparison for nutrition status between 2014 and 2019 had a medium effect and 93% chance of impact. For age comparison, 2014 versus 2019 had a medium effect and 89% chance of impact. The 2012 versus 2019 comparison has a small effect size and a 35% chance of impact.

Association of malnutrition with health-related outcomes

In the 2014 sample, there were five extreme outliers for LOS that were removed. The average LOS was 12 (IQR: 6, 22) days and 60% of participants were readmitted to hospital within 3-months. Malnourished participants (SGA rating B or C) had a higher rate in-hospital mortality but this did not quite reach significace (21% versus 9%; p=0.059); groups did not differ on other outcomes (Table 4). Malnutrition was not a significant predictor of any health-related outcome in adjusted regression models.
Severely malnourished participants (SGA C) had a higher rate in-hospital mortality (38% versus 12%; p=0.006), and a lower rate of falls during subsequent admissions (0% versus 22%; p=0.035); but groups did not differ on other outcomes (Table 5). Severe malnutrition increased the risk of in-hospital mortality by 457% (OR: 4.57 [95%CI: 1.42, 14.66], p=0.011); but in a model adjusted for age, cancer diagnosis, and prescription of nutrition supplements (confounders which met assumptions), this was reduced to 347% with a trend for significance [OR: 3.47 (95%CI: 0.94, 12.78] p=0.061). Severe malnutrition was not a predictor of other health-related outcomes in adjusted or unadjusted models.

Table 4 Health-related outcomes of the 2014 sample according to well-nourished or malnourished

Table 4
Health-related outcomes of the 2014 sample according to well-nourished or malnourished

Data expressed as median (IQR) or n (%) ; * Comparison of well-nourished (SGA rating A) and malnourished (SGA rating B or C) groups;  † Test performed on log10 normalised data

Table 5 Health-related outcomes of the 2014 sample according to severely malnourished or not severely malnourished

Table 5
Health-related outcomes of the 2014 sample according to severely malnourished or not severely malnourished

Data expressed as median (IQR) or n (%); * Comparison of no severe malnutrition (SGA rating A or B) and severely malnourished (SGA rating C) groups; † Test performed on log10 normalised data

Discussion

This study has reported a consistently high prevalence of malnutrition in three rural hospitals in northern NSW from 2012 to 2019; however, in 2019 the prevalence was 11% and 20% lower than in the previous samples, and the prevalence of severe malnutrition was very low at 2%. The 2014 sample reported the highest prevalence of malnutrition in any Australian hospital (45%); which exceeds the rate reported in three remote Australian hospitals (42%) (19-21). The lowest prevalence reported in 2019 aligns with prevalence rates in Australian metropolitan hospitals. As nutrition status comparison between the 2014 and 2019 cohorts had a calculated medium effect size and high impact value, there is high confidence in the measured prevalence rates. Relevant for the Australian health care system, the SGA tool used to determine prevalence is synonymous with the International Classification for Diseases, 10th revision, Australian Modification (ICD-10-AM) classification of protein-energy malnutrition (22), and therefore directly linked to case-mix funding.
The differing rates in malnutrition prevalence over time is partially explained by the age of participants; however, the impact of age cannot account for the changes in prevalence alone. The lower rate of malnutrition in 2019 compared 2014 may also be due to variations in the sampled wards and hospital sites. The 2019 sample did not include a 34-bed rehabilitation ward; a setting which has previously been reported to have a high prevalence of malnutrition at 53% (23). There are also likely causes of variation in the prevalence of malnutrition over time which were not captured by this study, including demographics such as socio-economic status or ethnicity, inter-rater variability of SGA assessment, or changes in hospital policies and priorities to address malnutrition. Interestingly, the rate of malnutrition did not vary according to sex. A recent meta-analysis of worldwide data found that females had a 45% increased risk of malnutrition (OR: 1.45 [95%CI: 1.27, 1.66] p<0.00001) in the community setting, which included post-hospital samples.
It has been well established that malnutrition increases the risk of poor health outcomes in the hospital setting, (21, 24, 25). This study confirms that malnourished participants had higher rates of in-hospital mortality; however, only severe malnutrition was a predictor of this outcome. The clinical importance of this is still relevant despite the rate of severe malnutrition being reduced to only 2% in 2019, as there is a possibility of inter-rater variability in application of the SGA. To confirm if the risk of malnutrition-related in-hospital mortality has been eradicated with the decrease in severe malnutrition rates, health outcomes of the 2019 would need to be examined. A high rate of false positives in the 2014 sample would also explain why many participant characteristics usually associated with malnutrition were not significant predictors in multivariable models. Although severe malnutrition appeared to have a lower rate of falls in subsequent hospital admissions compared to better nourished participants, this is explained by very low rates of hospital readmission reflecting the high rate of in-hospital mortality in this group.
Of clinical significance, less than 50% of malnourished participants in the 2014 sample were provided with nutritional supplementation, and non-supplementation was a predictor of poor health-related outcomes in the adjusted models, including the high rate of hospital readmission within 3-months. Additionally, prescription of proton-pump inhibiters, which inhibit nutrient digestion, was higher in malnourished participants. Overall, the 2014 sample had a high rate of polypharmacy, a risk factor for malnutrition (26). Although comorbidities were highly prevalent, the only disease which was higher in the malnourished participants was cancer. Whilst malnutrition in cancer is known to highly prevalent (27); previous studies have identified that patients with other hypermetabolic conditions such as hepatic, cardiovascular, and gastrointestinal disease, depression, and dementia also have increased risk of malnutrition (28, 29).

Limitations

As discussed above, this study is limited by potential poor inter-rater reliability for the SGA assessment between samples, and not including further demographic data to explore variation in the multivariable models. In addition, as the SGA assessments were implemented cross-sectionally, the ratio of pre-existing malnutrition (i.e. admitted to hospital with malnutrition) to hospital acquired malnutrition within the reported prevalence is unclear. Finally, the 2014 sample may have been underpowered to detect differences in health-related outcomes, particularly risk of in-hospital mortality (p=0.059).

Conclusion

Although the prevalence of malnutrition decreased over time, the rate of malnutrition in the sampled rural hospitals was consistently high; and is associated with increased risk of in-hospital mortality. Research should continue to monitor the rate of malnutrition in acute hospitals in rural areas to evaluate the impact of health service policies and procedures to address this problem.

Key Question Summary

What is known about the topic? Malnutrition is highly prevalent in the acute hospital setting in Australia at 30-40% and up to 71% in older adults (1). The Australian health system faces unique challenges related to high proportions of older adults living in geographically rural and remote areas. The rates and health-related complications of malnutrition in rural and remote Australian hospitals is unexplored.
What does this paper add? This study reported the highest ever recorded prevalence of malnutrition in Australia, at 48% in 2014; which was associated with increased risk of death. However, in 2019 the prevalence has reduced to 28%, and severe malnutrition was almost eradicated (down to 2%).
What are the implications for practitioners? Although the prevalence of malnutrition decreased over time, the rate of malnutrition in the sampled rural hospitals was consistently high; and was associated with increased risk of in-hospital mortality. Less than 50% of malnourished participants were provided with nutritional supplementation, and non-supplementation was a predictor of poor health-related outcomes in adjusted models, including in-hospital mortality.

Acknowledgements: A great thank you to Bond University statistician Evelyn Rathbone and the staff at Tweed, Murwillumbah, and Byron Bay Hospitals for their support of this study.

Conflicts of Interest: MB is the nutrition and dietetics department manager for the sampled sites and oversaw the implementation of all three cross-sectional studies. MB was not involved in the analysis of results. All other authors declare no existing or potential conflict of interest.

Funding sources: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions: All authors contributed to the conception of the study. EL and MB contributed to data collection. EL and SM contributed to data analysis. EL drafted the manuscript, and all authors contributed to manuscript revision. All authors approve the final version of the manuscript.

References

1.    Banks M, Ash S, Bauer J, Gaskill D. Prevalence of malnutrition in adults in Queensland public hospitals and residential aged care facilities. Nutrition & Dietetics. 2007;64(3):172-8.
2.    Camilo ME. Disease-related Malnutrition: An Evidence-based Approach to Treatment. Clinical Nutrition. 2003;22(6):585.
3.    Marshall S, Bauer J, Isenring E. The consequences of malnutrition following discharge from rehabilitation to the community: a systematic review of current evidence in older adults. J Hum Nutr Diet. 2014;27(2):133-41.
4.    Marshall S. Protein-energy malnutrition in the rehabilitation setting: Evidence to improve identification. Maturitas. 2016;86:77-85.
5.    Crichton M, Craven D, Mackay H, Marx W, de van der Schueren M, Marshall S. A systematic review, meta-analysis and meta-regression of the prevalence of protein-energy malnutrition: associations with geographical region and sex. Age Ageing. 2018;48(1):38-48.
6.    Agarwal E, Ferguson M, Banks M, Batterham M, Bauer J, Capra S, et al. Malnutrition and poor food intake are associated with prolonged hospital stay, frequent readmissions, and greater in-hospital mortality: results from the Nutrition Care Day Survey 2010. Clin Nutr. 2013;32(5):737-45.
7.    Agarwal E, Marshall S, Miller M, Isenring E. Optimising nutrition in residential aged care: a narrative review. Maturitas. 2016;92:70-8.
8.    Pirlich M, Schütz T, Norman K, Gastell S, Lübke HJ, Bischoff SC, et al. The German hospital malnutrition study. Clin Nutr. 2006;25(4):563-72.
9.    Marshall S. Why is the skeleton still in the hospital closet? A look at the complex aetiology of malnutrition and its implications for the nutrition care team. J Nutr Health Aging. 2018;22:26-9.
10.    Curtis LJ, Bernier P, Jeejeebhoy K, Allard J, Duerksen D, Gramlich L, et al. Costs of hospital malnutrition. Clin Nutr. 2017;36(5):1391-6.
11.    Marshall S, Agarwal E. Comparing characteristics of malnutrition, starvation, sarcopenia and cachexia in older adults. Handbook of famine, starvation, and nutrient deprivation: from biology to policy. 2017:1-23.
12.    Caring for Older Australians, Volume 1. Canberra: Productivity Commission; 2011.
13.    Living Longer. Living Better. Aged Care Reform Package. Canberra: Australian Government Department of Health and Ageing; 2012.
14.    Caring for Older Australians, Volume 2. Canberra: Productivity Commission; 2011.
15.    Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Prev Med. 2007;45(4):247-51.
16.    Agarwal E, Ferguson M, Banks M, Vivanti A, Batterham M, Bauer J, et al. Malnutrition, poor food intake, and adverse healthcare outcomes in non-critically ill obese acute care hospital patients. Clinical Nutrition. 2019;38(2):759-66.
17.    Detsky AS, McLaughlin JR, Paker JP, Johnston N, Wittaker S, Mendelson RA, et al. What is subjective global assessment of nutritional status. JPEN. 1987;11:8-11.
18.    Marshall S, Craven D, Kelly J, Isenring E. A systematic review and meta-analysis of the criterion validity of nutrition assessment tools for diagnosing protein-energy malnutrition in the older community setting (the MACRo Study). Clin Nutr. 2018;37(6A):1902-12.
19.    Morris NF, Stewart S, Riley MD, Maguire GP. The burden and nature of malnutrition among patients in regional hospital settings: A cross-sectional survey. Clinical Nutrition ESPEN. 2018;23:1-9.
20.    Allard JP, Keller H, Jeejeebhoy KN, Laporte M, Duerksen DR, Gramlich L, et al. Decline in nutritional status is associated with prolonged length of stay in hospitalized patients admitted for 7 days or more: A prospective cohort study. Clinical Nutrition. 2016;35(1):144-52.
21.    Agarwal E, Ferguson M, Banks M, Batterham M, Bauer J, Capra S, et al. Malnutrition and poor food intake are associated with prolonged hospital stay, frequent readmissions, and greater in-hospital mortality: results from the Nutrition Care Day Survey 2010. Clin Nutr. 2013;32(5):737-45.
22.    Australian coding standards for I.C.D.-10-AM. Sydney: National Centre for Classification in Health; 2008.
23.    Marshall S, Young A, Bauer J, Isenring E. Malnourished older adults admitted to rehabilitation in rural New South Wales remain malnourished throughout rehabilitation and once discharged back to the community: a prospective cohort study Journal of Aging Research and Clinical Practice. 2015;4(4):197-204.
24.    Allard JP, Keller H, Jeejeebhoy KN, Laporte M, Duerksen DR, Gramlich L, et al. Malnutrition at Hospital Admission-Contributors and Effect on Length of Stay: A Prospective Cohort Study From the Canadian Malnutrition Task Force. JPEN J Parenter Enteral Nutr. 2016;40(4):487-97.
25.    Sanson G, Bertocchi L, Dal Bo E, Di Pasquale CL, Zanetti M. Identifying reliable predictors of protein-energy malnutrition in hospitalized frail older adults: A prospective longitudinal study. Int J Nurs Stud. 2018;82:40-8.
26.    Nutrition screening as easy as mna. A guide to completing the Mini Nutritional Assessment (MNA). Swizterland Nestle Nutrition Institute.
27.    Segura A, Pardo J, Jara C, Zugazabeitia L, Carulla J, de las Penas R, et al. An epidemiological evaluation of the prevalence of malnutrition in Spanish patients with locally advanced or metastatic cancer. Clin Nutr. 2005;24(5):801-14.
28.    Konturek PC, Herrmann HJ, Schink K, Neurath MF, Zopf Y. Malnutrition in Hospitals: It Was, Is Now, and Must Not Remain a Problem! Medical science monitor : international medical journal of experimental and clinical research. 2015;21:2969.
29.    Bonetti L, Terzoni S, Lusignani M, Negri M, Froldi M, Destrebecq A. Prevalence of malnutrition among older people in medical and surgical wards in hospital and quality of nutritional care: A multicenter, cross-sectional study. Journal of clinical nursing. 2017;26(23-24):5082-92.
29.    Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41, 1149-1160.

DENTITION AND MALNUTRITION RISK IN COMMUNITYDWELLING OLDER ADULTS

 

R. Zelig1,2, L. Byham-Gray1, S.R. Singer2, E.R. Hoskin3, A. Fleisch Marcus1, G. Verdino1, D.R. Radler1,2, R. Touger-Decker1,2

 

1. School of Health Professions’ Department of Nutritional Sciences at Rutgers University; 2. School of Dental Medicine, Department of Diagnostic Sciences, at Rutgers University; 3. School of Dental Medicine, Department of Restorative Dentistry, at Rutgers University

Corresponding Author: Rena Zelig, DCN, RDN, CDE, CSG, 65 Bergen Street #157, Newark, NJ 07107, (973)-972-5956, zeligre@shp.rutgers.edu

J Aging Res Clin Practice 2018;7:107-114
Published onlineAugust 7, 2018, http://dx.doi.org/10.14283/jarcp.2018.19

Abstract

Background and Objective: Older adults are at risk for both impaired oral health and suboptimal nutritional status. The objective of this study was to explore the relationships between malnutrition risk and missing teeth in community-dwelling older adults. Design: This was a retrospective cross-sectional analysis of data obtained from the electronic health records of 107 patients aged 65 and older who attended an urban northeast US dental school clinic between June 1, 2015 and July 15, 2016. Odontograms and radiographs were used to identify teeth numbers and locations; malnutrition risk was calculated using the Self-Mini Nutritional Assessment (Self-MNA). Relationships between numbers of teeth and malnutrition risk were assessed using bivariate logistic regression. Results: Participants (N=107) were 72.6 years (SD=5.6) of age; 50.5% were female. Mean Self-MNA score was 12.3 (SD=2.0) reflective of normal nutrition status; 20.6% were at risk for malnutrition, 4.7% were malnourished. Greater than 87% were partially or completely edentulous. Those with 10-19 teeth had lower Self-MNA scores (mean=11.6, SD=2.5) than those with 0-9 teeth (mean=12.7, SD=1.3) or 20 or more teeth (mean=12.6, SD=1.8) and had an increased risk for malnutrition (OR=2.5, p=0.076). Conclusion: The majority of this sample of older adults were partially edentulous and of normal nutritional status. Those with 10-19 teeth were more likely to be at risk for malnutrition. Further studies are needed to examine relationships between tooth loss and malnutrition risk and the impact of impaired dentition on the eating experience in a larger sample and to inform clinical practice.

Key words: Nutrition, Nutrition Assessment), MNA, Self-MNA, malnutrition, dentition, tooth loss, elderly.

 

 

Introduction

The relationships between nutrition and oral health are synergistic. The mouth is the entry way for food and fluid intake. If its integrity is impaired, the functional ability of an individual to consume an adequate diet may be adversely impacted. Older adults are a vulnerable population at high risk for both impaired oral health and malnutrition (1-4). Petersen (1) et al reported that 30% of older adults ages 65-74 globally were completely edentulous and many more were missing some of their natural teeth. Data from the United States (US) National Health and Nutrition Examination Survey (NHANES) 2011–2012, revealed that approximately 13% of those aged 65–74 and 26% of those aged 75 and over were edentulous (2).
Tooth loss may affect the ability of an individual to consume an adequate diet (5-9). Prior research has demonstrated relationships between nutritional status, nutrient intake, oral health and diet quality (5-9). The number and distribution of teeth impact masticatory function. Large population studies (5-9) have found that partial and complete edentulism are associated with changes in food and nutrient intake in older adults including decreased consumption of fruit, vegetables, dietary fiber, calcium, iron, and other vitamins.  Independent of its cause, inadequate intake in older adults leads to weight loss, malnutrition, and ultimately increased morbidity and mortality (10, 11).
The etiology of malnutrition in older adults is multifaceted and may be related to physiological and psychological changes that contribute to variable food and nutrient intake (12-14). Kaiser and colleagues (3) pooled analyses from multinational studies which used the Mini Nutritional Assessment (MNA) to evaluate malnutrition prevalence and risk and reported that 22.8% of the older adults studied were malnourished and 46.2% were at risk for malnutrition. Huhmann et al (4) explored malnutrition risk in community dwelling older adults using the Self-MNA , a validated self-administered version of the MNA and found that 27% of subjects were malnourished, 38% were at risk of malnutrition, and 35% had normal nutrition status (4).
A systematic review by Zelig et al (15) exploring the associations between missing teeth and nutritional status (determined by MNA score) in community dwelling older adults revealed conflicting findings. Significant associations were found in five of eight studies between missing teeth (16-18) or the use of dental prostheses (19, 20) and malnutrition risk. MNA scores were significantly lower in those with fewer teeth/limited occlusion as compared to those with more teeth and/or more posterior occluding teeth pairs.  However, other researchers did not report significant associations between occlusal status (21) or dental status (22) and MNA score.
Similarly, Toniazzo et al (23) systematically explored associations between malnutrition risk assessed by MNA or Subjective Global Assessment (SGA), and oral health status in elders and found that individuals with or at risk for malnutrition had significantly fewer teeth than those with normal nutritional status (23). However, similar to Zelig et al, a significant relationship between edentulism and malnutrition risk was not consistently identified.
Given the heterogeneity of the findings in this area (15, 23), the aim of this study was to explore the associations between nutritional status (as defined by the Self-MNA) and dentition status (missing teeth and edentulism with and without denture replacement) in older adults (=>65 years old) who came to the Rutgers School of Dental Medicine (RSDM) clinics in Newark, New Jersey (NJ), between June 1, 2015 and July 15, 2016.  Analyses from the NHANES surveys conducted from 1999-2010 have consistently shown that partial and complete edentulism are more common in females, in persons who are Black or Hispanic, and in those who are of a lower socio-economic status (24, 25). The population of patients treated at the RSDM provided a convenience sample of vulnerable individuals at high risk for impaired dentition status and malnutrition given their racial diversity and low socio-economic status (24-26). We hypothesized that Self-MNA scores will be significantly lower in those with fewer teeth/limited occlusion as compared to those with more teeth and/or more posterior occluding teeth pairs (15, 23).

 

Materials and Methods

Sample Design and Sample Selection

This was a retrospective cross-sectional analysis of data obtained from the electronic health records (EHR), (Axium, EXAN, Vancouver, BC, Canada) of patients who attended the RSDM clinic in Newark, NJ between June 1, 2015 and July 15, 2016. The EHR report contained select data provided by patients (n=192) during initial screening which included demographic characteristics, medical and dental history, and Self-MNA data. Patients were excluded (n=85) if they were younger than 65 years, height and/or weight were not available in the dental record, Self-MNA data were incomplete, and/or the number and location of teeth could not be accurately mined from the EHR. This study was approved by the Rutgers University Biomedical and Health Sciences Institutional Review Board.

Assessment of Nutritional Status

Data regarding nutritional status were obtained from the EHR using patient responses to the Self-MNA tool (Nestle Nutrition, available at http://www.mna-elderly.com/forms/Self_MNA_English_Imperial.pdf).
The original 18-item Mini Nutritional Assessment (MNA) was developed and validated in 1994 by Nestle Nutrition to detect risk for malnutrition in adults aged 65 and older (27).  A condensed version of the MNA, the Mini Nutritional Assessment Short Form (MNA-SF), containing only six questions, was validated to further facilitate rapid nutritional screening in older adults, as it takes only three to five minutes to administer and retains the diagnostic accuracy of the full MNA (28, 29). In 2013, the Self-Administered MNA (Self-MNA) was adapted from the MNA-SF and validated by Huhmann et al (4) to allow the patient to provide a self-assessment of their nutritional status (4). The Self-MNA requires that the patient or their caregiver complete the assessment prior to their appointment with the healthcare provider; it can then be reviewed with the healthcare provider to identify risk factors for malnutrition (4). The Self-MNA contains six questions which address recent changes in intake and weight, mobility, recent stress, illness, dementia and sadness, as well as body mass index (BMI). A score of 0-7 is considered malnourished, 8-11 is at risk of malnutrition and 12-14 reflects normal nutritional status (4).

Assessment of Dentition Status

Number and location of teeth were mined from the EHR using patient digital radiography and odontogram data. A research assistant, trained and calibrated with a prosthodontist, recorded tooth numbers as either “present” or “missing”. If the tooth surface had been restored with a permanent crown, implant, or fixed bridge the tooth was reported as “present.” The four third molars/wisdom teeth were not included as they are not consistently present in all individuals. The presence and location of these permanent natural or restored tooth surfaces was also used to categorize the number of anterior and posterior occluding pairs of teeth (AOP and POP, respectively), defined as the “presence of a natural (or restored permanent) tooth on the maxilla and corresponding mandible, excluding remaining roots or root caps (page 316)” (30). Number of natural or restored teeth was categorized in a manner similar to prior research into those with 0-9 teeth, 10-19 teeth and 20 or more teeth (17, 31, 32).

Data Analysis

All statistical analyses were conducted using the Statistical Package for Social Sciences (version 22.0, SPSS Inc., Chicago, IL). A sample size of 85 was determined to be needed to establish if a correlation existed between the number of remaining teeth and Self-MNA score, based on a two-tailed test with an a priori alpha of p≤0.05 and 80% power to detect  an small-moderate effect of at least 0.3 (33). Statistical significance was set at p<0.05.
Descriptive statistics were used to summarize all of the study variables. To test the hypothesis that Self-MNA scores will be significantly lower in those with fewer teeth / limited occlusion as compared to those with more teeth and/or more posterior occluding teeth pairs, Spearman’s Correlation Coefficient was used. Associations between categorical variables were evaluated using the chi-squared test; non-parametric Mann-Whitney and Kruskal-Wallis tests were used to compare groups. Simple logistic regression was used to assess the odds of being at risk for malnutrition or malnourished related to number of teeth present. To determine potential confounding by demographic characteristics, bivariate analyses were conducted in accordance to age, gender, race and ethnicity as well as prior medical history.

 

Results

Characteristics of the study sample

Records from 107 community dwelling older adults who came for care to the RSDM clinics in Newark, NJ between June 1, 2015 and July 15, 2016 were included in the analyses (Figure 1). Their mean age was 72.6 years (SD=5.6) with a range of 65-91 years. Table 1 describes demographic and clinical characteristics of participants.
The mean Self-MNA score of this sample was 12.3 (SD=2.0) reflective of normal nutrition status. Twenty percent (20.6%, n=22) were at risk for malnutrition and 4.7% (n=5) were malnourished according to their Self-MNA responses. A moderate to severe decrease in food intake was reported by 27.1% (n=29) of the patients; 32.7% (n=35) indicated a weight loss of two or more pounds in the past three months (Table 2). Of those who reported a weight loss of more than seven pounds in three months (n=14), all were either at risk for malnutrition (n=10) or malnourished (n=4).
Of the five who were categorized as having malnutrition; 80% (n=4) reported a severe or moderate decrease in food intake, a weight loss greater than seven pounds in three months, and that they had been stressed or severely ill in the past three months; 60% (n=3) of these also had positive responses for severe dementia and/or prolonged sadness; 40% (n=2) were unable to get out of bed or a chair without assistance.
Using the Centers of Disease Control and Prevention (34) (CDC) BMI categories, 21.5% (n=23) were in the normal weight range, 43.0% (n=46) were overweight and 35.5% (n=38) were obese. All those whose Self-MNA score reflected malnutrition (n=5) were obese (mean BMI = 31.6, SD =4.5) with higher BMI values than those who had a normal nutritional status (mean BMI = 29.0, SD=5.0) or who were at risk for malnutrition (mean BMI = 27.6, SD=4.7).

 

Table 1 Select demographic and clinical characteristics of the study sample (N=107)

Table 1
Select demographic and clinical characteristics of the study sample (N=107)

Note: Sample size for each question varied as participants were able to choose to not answer questions on the registration forms.

Table 2 Responses to Self-MNA questions by nutritional status category (N=107)

Table 2
Responses to Self-MNA questions by nutritional status category (N=107)

 

The majority of participants were partially edentulous (n=94, 87.8%); 4.7% (n=5) were completely edentulous and 7.5% (n=8) were fully dentate. Fifty-one (47.7%) had at least 20 teeth, 29.9% (n=32) had 10-19 teeth and 22.4% (n=24) had 0-9 teeth. Approximately one-third (35.5%, n=38) of the sample had no posterior occlusion and 7.5% (n=8) had complete posterior occlusion; approximately one quarter (26.2%, n=28) had no anterior occlusion and 43.0% (n=46) had complete anterior occlusion.
The mean number of natural or restored teeth decreased from 17.4 among those with normal nutritional status to 14.4 among those classified as having malnutrition; however, these differences were not statistically significant (p=0.656). Median values of teeth declined in a similar pattern, from 20 among those with normal nutritional status to 18 among those at risk for malnutrition and 15 among those categorized as having malnutrition (Figure 2).

 

Figure 1 Flow chart of study sample

Figure 1
Flow chart of study sample

 

Figure 2 Boxplot of number of natural or restored teeth by nutritional status category (N=107)

Figure 2
Boxplot of number of natural or restored teeth by nutritional status category (N=107)

 

Analyses of the relationships between dental and occlusal status and Self-MNA Score revealed no linear relationships between the Self-MNA score and the number of natural teeth (r=0.104, p=0.285), posterior occluding pairs (POP) of teeth (r=0.173, p=0.074) or anterior occluding pairs (AOP) of teeth (r=0.049, p=0.619).   Although mean differences were not significant (Kruskal-Wallis, p=0.116), those with 10-19 teeth had lower Self-MNA scores (mean=11.6, SD=2.5) than those with 0-9 teeth (mean=12.7, SD=1.3) or 20 or more teeth (mean=12.6, SD=1.8) (Figure 3). When MNA Score was dichotomized into those who were at risk for malnutrition or malnourished (25.2%, n=27) compared to those with normal nutritional status (74.8%, n=80), those with 10-19 teeth had 2.5 times the odds of being at risk for malnutrition/malnourished than those with 20 or more teeth (OR=2.5, p=0.076). Bivariate analyses yielded no statistically significant confounding effects of age, gender, race/ethnicity or medical history.

Figure 3 Boxplot of Self-MNA scores among those with different categories of natural or restored teeth (N=107)

Figure 3
Boxplot of Self-MNA scores among those with different categories of natural or restored teeth (N=107)

 

Discussion

The aim of this study was to explore the associations between nutritional status and dentition status among older adults who presented for care at the RSDM clinics in Newark, NJ.  The study hypothesis that Self-MNA scores will be significantly lower in those with fewer teeth / limited occlusion as compared to those with more teeth and/or more posterior occluding teeth pairs was not supported as a significant direct relationship between Self-MNA score and number of natural or restored teeth was not found.
The majority of the sample had some degree of tooth loss (87.8% partially edentulous; 4.7% completely edentulous). Although not statistically significant, there was a trend whereby the mean number of natural or restored teeth decreased from 17.4 in those with normal nutritional status to 14.4 in those classified as having malnutrition, suggesting a link between having fewer teeth and an increased risk of malnutrition. Those with 10-19 teeth had lower Self-MNA scores than those with 0-9 teeth or 20 or more teeth and, among those with 10-19 teeth the odds of being at risk for malnutrition/malnourished were 2.5 times those with 20 or more teeth. Interestingly, the patients with malnutrition had 10-19 teeth; none were fully dentate or fully edentulous.   Furuta et al (17) also noted a similar trend whereby those with 10-19 teeth had the lowest MNA-SF scores as compared to those with 0-9 teeth or greater than 20 teeth. While not significant, this finding suggests that those who are missing one third to two thirds of their natural teeth may be most at risk for becoming malnourished.
In contrast to our findings, Kikutani et al (16), Starr et al (18), and Furuta et al (17), reported significant associations between MNA scores and number of missing teeth whereby MNA scores were significantly lower in those with fewer teeth / limited occlusion as compared to those with more teeth and/or more posterior occluding teeth pairs of teeth. Kikutani et al found that individuals with more missing teeth and inadequate occlusion were more likely to be at risk for malnutrition, and those with functionally inadequate occlusion and no dentures had a 3.189 fold greater malnutrition risk than those with natural dentition and adequate function (16). Similarly, Starr et al demonstrated that individuals who were completely edentulous had significantly lower MNA scores than those who were partially or completely dentate (p = 0.028) (18). In bivariate models, Furuta et al found that those with 0-19 teeth had lower MNA scores than those with 20 or more teeth (p = 0.041), however, in a multivariate path analysis, no direct relationship was noted between oral health status and MNA score (17).
While the results of the current study showed similar trends they did not reach the level of statistical significance reported by others. Possible explanations for the variation in results lie in the differences in the populations studied. The subjects in our study were all patients of an urban northeast US dental school clinic in Newark, NJ, and were younger than participants in other studies (16-18). The combined 25% prevalence of malnutrition and risk for malnutrition was relatively lower than the 13.3% malnourished and additional 51.7% at risk for malnutrition reported by Kikutani et al (16), and 14.0% malnourished / 55.2% at risk of malnutrition reported by Furuta et al (17). Participants in the current study had more teeth (mean of 17.0 ±8.5 compared to 8.6 +/- 9.9) (17) and were less likely to be edentulous then those studied by Furuta et al (17). In contrast, the population studied by Starr et al represented healthy older adults who were free of medical conditions and medications at baseline, and as a whole had relatively high MNA scores with a mean of 8.0 or greater out of a maximum score of 9 using their version of the MNA tool (18).
Our findings are consistent with preceding studies, which were unable to find statistically significant relationships between number of teeth (35), occlusal status (21) or dental status (22) and MNA score. Systematic reviews of the evidence (15, 23), have similarly found much conflicting results due in part to the heterogeneity of the studies comparing these variables. Nutritional status and dentition status can be assessed in a variety of ways.  Varying clinical and demographic characteristics of the populations sampled such as age, health status and country of origin add to the complexity of this area of research.
Natural dentition and adequate function are important to maintain nutritional status. Missing teeth and inadequate occlusion affect masticatory function and can result in a decline in overall quality and quantity of intake (5-9) and/or replacement of difficult to chew foods with softer foods that may be more calorically dense and/or nutrient poor which could lead to changes in weight and overall nutritional status (36, 37). Further studies that address diet quality and quantity are necessary to better understand these relationships and guide implications for practice.

Strengths and Limitations

Although the study sample was adequately powered and representative of the population of older adults who attend the RSDM clinic, the majority were overweight/obese and of normal nutrition status. The findings are limited to a single institution and not generalizable to other community dwelling dental settings. Given the retrospective design of this study, a cause and effect relationship between variables could not be established.   Risk for malnutrition may be better assessed using a tool such as the MNA-SF, which considers oral health factors specifically when looking at inadequate intake.
Strengths include data collection over a 13-month timespan to assure an adequate sample size, the availability of all of the data in the EHR, and the verification of dentition status by a research assistant, who was trained and calibrated with a prosthodontist, using digital radiography and the patient’s odontogram. The subjective nature of all self-reported data, including all demographic characteristics, height, weight, and Self-MNA data is a potential limitation. Although the Self-MNA is a validated tool to assess malnutrition prevalence and risk, it lacks the details found in the MNA-SF  that prompts the healthcare professionals to assess the etiology behind a decline in food intake such as loss of appetite, digestive problems, chewing or swallowing difficulties (4).

 

Conclusion

The findings of this study did not support relationships between the number of natural or restored teeth and Self-MNA score in this sample of community dwelling older adults. Although not statistically significant, the mean number of natural or restored teeth declined as nutritional status declined, which may be clinically relevant. Those classified as having malnutrition had higher rates of weight loss, decreased intake and more frequently reported dementia and/or depression, and severe illnesses than those with a normal nutritional status. Similarly, those classified as being at risk of malnutrition were more likely to experience weight loss and a moderate decline in intake than those who were classified as being of normal nutritional status.

Implications for Research

Our findings add to the heterogeneity of research outcomes in this area. The conflicting findings between studies is in part due to the use of different measures of nutrition and dental status as well as variations in demographic characteristics. Further research with larger samples, in different populations of community dwelling older adults with varying health conditions is needed to better understand the associations between nutrition and oral health and increase the generalizability of results. Prospective studies to determine and compare changes in nutritional status over time with changes in dentition and denture usage can help to determine cause and effect. Qualitative research aimed at understanding how impaired dentition affects intake, the eating experience and overall nutritional status can also help to shed light on these relationships, and can provide the necessary data upon which to develop practice based interventions and guidelines for this vulnerable population.  Although the Self-MNA has been validated as a self-administered tool to determine malnutrition prevalence and risk in community dwelling older adults, research in this area may benefit from using the MNA-SF, which is administered by a healthcare professional as older adults may require more clarification of questions to elicit more accurate results. The MNA-SF also includes a question related to decline in intake attributed to chewing or swallowing difficulties.

Implications for Practice

The dental clinic setting may be an ideal location to perform nutritional status screenings. As reported by Greenberg et al, both oral healthcare professionals and patients were receptive to screening for various medical conditions at the time of a dental visit (38, 39). This could be expanded to include nutritional status screening to identify patients at risk for malnutrition who may not regularly attend visits with a primary care provider. The results of the current study revealed that over 25% of the sample of older adults in this research who came for care at the RSDM clinic had malnutrition or were at risk for malnutrition. Based on these results and previous studies that estimated the prevalence of malnutrition to be over one third in older adults (3, 4), the use of nutritional screening tools by oral healthcare professionals could help to provide timely referrals to primary care physicians or Registered Dietitian Nutritionists for these individuals. Referrals to community assistance programs (such as Meals on Wheels) could also be made as appropriate to prevent decline in nutrition status.

 

Ethical standards: This study was approved by the Rutgers University Biomedical and Health Sciences Institutional Review Board.

Funding agency: Sackler Institute for Nutritional Sciences. PI Rena Zelig received an early career investigator’s grant from the Sackler Institute for Nutrition Sciences for a project entitled “Exploring the Associations between Dentition Status, Nutritional Status, and the Eating Experience in Older Adults – A Mixed Methods Study.” The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Acknowledgements: Funding for this project was provided by the Sackler Institute for Nutritional Sciences of the New York Academy of Sciences. The authors wish to acknowledge Dr. Michael Conte and the Office of Clinical Affairs and the Information Technology Department at the Rutgers School of Dental Medicine, Newark, NJ as well as Steven Britton, DMD, and Veronica Jones, MPH, Research Assistants for their help with this study.

Conflict of Interest Disclosure: All authors report that they have no conflicts of interest to disclose related to this research and manuscript.

 

References

1.    Petersen PE, Kandelman D, Arpin S, Ogawa H. Global oral health of older people–call for public health action. Community Dent Health 2010;27(4 Suppl 2):257-67
2.     Dye BA, Thornton-Evans G, Li X, Iafolla TJ. Dental Caries and Tooth Loss in Adults in the United States, 2011–2012. NCHS data brief, no 197. Hyattsville, MD. National Center for Health Statistics. 2015. Available at:
https://www.cdc.gov/nchs/data/databriefs/db197.pdf. Accessed January 12, 2018.
3.     Kaiser MJ, Bauer JM, Ramsch C, et al. Frequency of malnutrition in older adults: a multinational perspective using the mini nutritional assessment. J Am Geriatr Soc  2010;58(9):1734-8 doi: 10.1111/j.1532-5415.2010.03016.
4.     Huhmann MB, Perez V, Alexander DD, Thomas DR. A self-completed nutrition screening tool for community-dwelling older adults with high reliability: a comparison study. J Nutr Health Aging 2013;17(4):339-44 doi: 10.1007/s12603-013-0015-x
5.     Sahyoun NR, Lin CL, Krall E. Nutritional status of the older adult is associated with dentition status. J Am Diet Assoc 2003;103(1):61-6 doi: 10.1053/jada.2003.50003
6.     Sheiham A, Steele J. Does the condition of the mouth and teeth affect the ability to eat certain foods, nutrient and dietary intake and nutritional status amongst older people? Public Health Nutr 2001;4(3):797-803
7.     Zhu Y, Hollis JH. Tooth loss and its association with dietary intake and diet quality in American adults. J Dent 2014;42(11):1428-35 doi: 10.1016/j.jdent.2014.08.012
8.     Iwasaki M, Taylor GW, Manz MC, et al. Oral health status: relationship to nutrient and food intake among 80-year-old Japanese adults. Community Dent Oral Epidemiol 2014;42(5):441-50
9.     Ervin RB, Dye BA. Number of natural and prosthetic teeth impact nutrient intakes of older adults in the United States. Gerodontology 2012;29(2):e693-702 doi: 10.1111/j.1741-2358.2011.00546.x
10.     Zajacova A, Ailshire J. Body Mass Trajectories and Mortality Among Older Adults: A Joint Growth Mixture-Discrete-Time Survival Analysis. Gerontologist 2013 doi: 10.1093/geront/gns164
11.     Lundin H, Saaf M, Strender LE, Mollasaraie HA, Salminen H. Mini nutritional assessment and 10-year mortality in free-living elderly women: a prospective cohort study with 10-year follow-up. Eur J Clin Nutr 2012;66(9):1050-3 doi: 10.1038/ejcn.2012.100
12.     Lorenzo-Lopez L, Maseda A, de Labra C, Regueiro-Folgueira L, Rodriguez-Villamil JL, Millan-Calenti JC. Nutritional determinants of frailty in older adults: A systematic review. BMC Geriatr 2017;17(1):108 doi: 10.1186/s12877-017-0496-2
13.     Favaro-Moreira NC, Krausch-Hofmann S, Matthys C, et al. Risk Factors for Malnutrition in Older Adults: A Systematic Review of the Literature Based on Longitudinal Data. Adv Nutr 2016;7(3):507-22 doi: 10.3945/an.115.011254
14.     van der Pols-Vijlbrief R, Wijnhoven HA, Schaap LA, Terwee CB, Visser M. Determinants of protein-energy malnutrition in community-dwelling older adults: a systematic review of observational studies. Ageing Res Rev 2014;18:112-31 doi: 10.1016/j.arr.2014.09.001
15.     Zelig R, Touger-Decker R, Chung  M, Byham-Gray L. Associations between Tooth Loss, with or without Dental Prostheses, and Malnutrition Risk in  Older Adults: A Systematic Review. Top Clin Nutr 2016;31(3):232-47
16.     Kikutani T, Yoshida M, Enoki H, et al. Relationship between nutrition status and dental occlusion in community-dwelling frail elderly people. Geriatr Gerontol Intl 2013;13(1):50-4 doi: 10.1111/j.1447-0594.2012.00855.x
17.     Furuta M, Komiya-Nonaka M, Akifusa S, et al. Interrelationship of oral health status, swallowing function, nutritional status, and cognitive ability with activities of daily living in Japanese elderly people receiving home care services due to physical disabilities. Community Dent Oral Epidemiol 2013;41(2):173-81 doi: 10.1111/cdoe.12000
18.     Starr JMH, R. J.;Macintyre, S.;Deary, I. J.;Whalley, L. J. Predictors and correlates of edentulism in the healthy old people in Edinburgh (HOPE) study. Gerodontology 2008;25(4):199-204
19.     Cousson PY, Bessadet M, Nicolas E, Veyrune JL, Lesourd B, Lassauzay C. Nutritional status, dietary intake and oral quality of life in elderly complete denture wearers. Gerodontology 2012;29(2):e685-92 doi: 10.1111/j.1741-2358.2011.00545.x
20.     McKenna G, Allen PF, Flynn A, et al. Impact of tooth replacement strategies on the nutritional status of partially-dentate elders. Gerodontology 2012;29(2):e883-90 doi: 10.1111/j.1741-2358.2011.00579.x
21.     Soini H, Routasalo P, Lauri S, Ainamo A. Oral and nutritional status in frail elderly. Spec Care Dent 2003;23(6):209-15
22.     Soini H, Routasalo P, Lagstrom H. Nutritional status in cognitively intact older people receiving home care services–a pilot study. J Nutr Health Aging 2005;9(4):249-53
23.     Toniazzo MP, Amorim PS, Muniz FW, Weidlich P. Relationship of nutritional status and oral health in elderly: Systematic review with meta-analysis. Clin Nutr 2017 doi: 10.1016/j.clnu.2017.03.014
24.     Dye BA, Li X, Thornton-Evans G. Oral Health Disparities as Determined by Selected Healthy People 2020 Oral Health Objectives for the United States, 2009–2010. NCHS data brief: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics, 2012.
25.     National Institutes of Health. National Institute of Dental and Craniofacial Research. Tooth Loss in Seniors (Age 65 and Over). Tooth Loss in Seniors (Age 65 and Over)  March 11, 2014. Available at: https://www.nidcr.nih.gov/DataStatistics/FindDataByTopic/ToothLoss/ToothLossSeniors65andOlder.htm. Accessed January 12, 2018.
26.     Bureau USC. QuickFacts. Newark city, New Jersey. Available at: https://www.census.gov/quickfacts/table/PST045215/3451000,34. Accessed January 12, 2018.
27.     Guigoz Y, Vellas B. The Mini Nutritional Assessment (MNA) for grading the nutritional state of elderly patients: presentation of the MNA, history and validation. Nestle Nutrition workshop series. Nestle Nutr Workshop Ser Clin Perform Programme 1999;1:3-11; discussion 11-2.
28.     Rubenstein LZ, Harker JO, Salva A, Guigoz Y, Vellas B. Screening for undernutrition in geriatric practice: developing the short-form mini-nutritional assessment (MNA-SF). J Gerontol A Biol Sci Med Sci 2001;56(6):M366-72
29.     Kaiser MJ, Bauer JM, Ramsch C, et al. Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for identification of nutritional status. J Nutr Health Aging 2009;13(9):782-8
30.     Yoshida M, Kikutani T, Yoshikawa M, Tsuga K, Kimura M, Akagawa Y. Correlation between dental and nutritional status in community-dwelling elderly Japanese. Geriatr Gerontol Intl 2011;11(3):315-9 doi: 10.1111/j.1447-0594.2010.00688.x
31.     Ostberg AL, Nyholm M, Gullberg B, Rastam L, Lindblad U. Tooth loss and obesity in a defined Swedish population. Scand J Public Health 2009;37(4):427-33
32.     Syrjala AMY, P.;Hartikainen, S.;Sulkava, R.;Knuuttila, M. Number of teeth and selected cardiovascular risk factors among elderly people. Gerodontology 2010;27(3):189-92
33.     Portney L, Watkins M. Foundations of Clinical Research, Applications to Practice. 3 ed. Upper Saddle River, NJ: Pearson Education, Inc, 2009.
34.     Centers for Disease Control and Prevention. Healthy Weight. Secondary Healthy Weight. Available at: http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html. Accessed January 12, 2018.
35.     Lopez-Jornet P, Saura-Perez M, Llevat-Espinosa N. Effect of oral health dental state and risk of malnutrition in elderly people. Geriatr Gerontol Int 2013;13(1):43-49 doi: 10.1111/j.1447-0594.2012.00853.x
36.     Musacchio EP, E.;Binotto, P.;Sartori, L.;Silva-Netto, F.;Zambon, S.;Manzato, E.;Corti, M. C.;Baggio, G.;Crepaldi, G. Tooth loss in the elderly and its association with nutritional status, socio-economic and lifestyle factors. Acta Odontol Scand 2007;65(2):78-86
37.     Ikebe KM, K.;Morii, K.;Nokubi, T.;Ettinger, R. L. The relationship between oral function and body mass index among independently living older Japanese people. Int J Prosthodont 2006;19(6):539-46
38.     Greenberg BL, Glick M, Frantsve-Hawley J, Kantor ML. Dentists’ attitudes toward chairside screening for medical conditions. J Am Dent Assoc 2010;141(1):52-62
39.     Greenberg BL, Kantor ML, Jiang SS, Glick M. Patients’ attitudes toward screening for medical conditions in a dental setting. J Public Health Dent 2012;72(1):28-35 doi: 10.1111/j.1752-7325.2011.00280.x