Oral Care Reduces Pneumonia in Older Patients in

Nursing Homes
Takeyoshi Yoneyama, DDS, PhD, Mitsuyoshi Yoshida, DDS, PhD, Takashi Ohrui, MD, PhD,
Hideki Mukaiyama, DDS, Hiroshi Okamoto, DDS, PhD, Kanji Hoshiba, DDS, PhD,
Shinichi Ihara, DDS, Shozo Yanagisawa, DDS, Shiro Ariumi, DDS, Tomonori Morita, DDS,
Yasuro Mizuno, DDS, Takayuki Ohsawa, DDS, PhD, Yasumasa Akagawa, DDS, PhD,
Kenji Hashimoto, DDS, MD, PhD, Hidetada Sasaki, MD, PhD, and
Members of the Oral Care Working Group

OBJECTIVES: Aspiration of oral secretions and their bac-

teria is increasingly being recognized as an important fac-
tor in pneumonia. We investigated whether oral care low-
P neumonia is not only a common infection in older peo-
ple, it is also the most common cause of death from noso-
comial infection in this population.1 Bacterial pneumonia is
ers the frequency of pneumonia in institutionalized older thought to be due to organisms that inhabit the oropharynx,
people. and aspiration of oropharyngeal contents has been suggested
DESIGN: Survey. as the mechanism by which these bacteria reach the lower
respiratory tract.2 Many older patients in nursing homes
SETTING: Eleven nursing homes in Japan.
have poor oral health because of the difficulty of access to
PARTICIPANTS: Four hundred seventeen patients ran- professional dental care and insufficient personal oral hy-
domly assigned to an oral care group or a no oral care group. giene care.3 It is recognized that community-acquired
INTERVENTION: Nurses or caregivers cleaned the pa- pneumonia and lung abscesses can be the result of infection
tients teeth by toothbrush after each meal. Swabbing with by anaerobic bacteria, and dental plaque would seem to be a
povidone iodine was additionally used in some cases. Den- logical source of these bacteria, especially in patients with pe-
tists or dental hygienists provided professional care once a riodontal desease.4 However, to our knowledge, whether im-
week. proving oral hygiene would lower the risk in either of these
MEASUREMENTS: Pneumonia, febrile days, death from setting has not been studied. In the present study, we
pneumonia, activities of daily living, and cognitive functions. treated these patients with oral care to minimize respira-
RESULTS: During follow-up, pneumonia, febrile days, and tory infections, possibly caused by silent aspiration. A pre-
death from pneumonia decreased significantly in patients liminary experiment has been reported elsewhere.5
with oral care. Oral care was beneficial in edentate and den-
tate patients. Activities of daily living and cognitive functions SUBJECTS AND METHODS
showed a tendency to improve with oral care. The present study was performed with older patients in 11
CONCLUSION: We suggest that oral care may be useful nursing homes in Japan. Each nursing home has from 50
in preventing pneumonia in older patients in nursing homes. to 100 beds and serves as a long-term care facility for older
J Am Geriatr Soc 50:430433, 2002. patients who are physically handicapped or suffering from
Key words: silent aspiration; aspiration pneumonia; oral mental deterioration. Thus, to a large extent, they are de-
health; activities of daily living; cognitive function pendent on the service of caregivers for activities of daily
living (ADLs). The criterion for patient selection was that
physical symptoms and cognitive impairment must have
been stable for the preceding 3 months. During this 3-month
period no patient had acute disorders (e.g., severe infection,
From the Department of Geriatric and Respiratory Medicine, Tohoku Uni- heart failure, or stroke requiring special treatment and inten-
versity School of Medicine, Sendai, Japan. sive care). Chronic diseases suffered by the patients included
This study was supported by Funds for Comprehensive Research on Aging previous stroke, hypertension, arrhythmia, previous myo-
and Health from 1999 to 2000 of the Japan Welfare Ministry.
cardial infection, diabetes mellitus, and inactive gastric ul-
Address correspondence to Hidetada Sasaki, MD, PhD, Professor and
Chairman, Department of Geriatric and Respiratory Medicine, Tohoku
cer. Mental function varied from slight cognitive impair-
University School of Medicine, 11 Seiryo-machi, Aoba-ku, Sendai 980 ment to dementia. A few patients in each nursing home could
8574 Japan. E-mail: [email protected] not participate in the study because they could not give in-

JAGS 50:430433, 2002

2002 by the American Geriatrics Society 0002-8614/02/$15.00
JAGS MARCH 2002VOL. 50, NO. 3 ORAL CARE AND PNEUMONIA 431

formed consent. No patient had any chronic pulmonary dis- and gowns once a week, sponge bath once a week, and
ease such as chronic obstructive pulmonary disease, bronchial changing diapers as required. Cognitive impairment was
asthma, or pulmonary fibrosis. The patients fed themselves evaluated with Japanese versions of the Mini-Mental State
or needed help in eating; no patient had feeding tubes. The Examination (MMSE).7 Trained neuropsychologists mea-
patients were randomly selected from the same floor and sured MMSE every 6 months. ADLs were evaluated using
nursing team in each nursing home. Randomization was the modified Barthel Index.8 Both MMSE and ADL have a
made from a random-numbers table, and the list was held 30-point scale for healthy older people, with a score of 0
independently of the investigators. No dentist had been in indicating complete loss of cognition and dependence. We
charge of the institution before the study. Four hundred took axillary temperatures two times a day; a febrile day
seventeen patients were randomly assigned to an oral care was defined as the axillary temperature rising above 37.8C
group or a no oral care group in September 1996 and were at either of the two daily measurements. Patients who had
investigated for 2 years. Criteria for diagnosis of pneumo- febrile days for more than 7 cumulative days during 2 years
nia were a new pulmonary infiltrate seen on a chest radio- were assumed to be patients with fever. The Tohoku Uni-
graph and one of the following features: cough, tem- versity Ethics Committee granted ethical approval, and in-
perature greater than 37.8C, or subjective dyspnea. Two formed consent was obtained from the patients or their
radiologists who were not involved in the studies made the di- families.
agnosis of pneumonia. However, 51 patients were excluded All comparisons were made between the oral care and
from the analysis because they died from causes other than nonoral care groups or between dentate and edentate pa-
pneumonia during follow-up. Of the remaining 366 pa- tients using two-way analysis of variance. Statistical signif-
tients, 184 (mean age  standard deviation (SD) 82.0  icance was accepted as P  .05. All data are expressed as
7.8, 148 women and 36 men) had received oral care at mean  SD.
study entry; 182 (mean age 82.1  7.5, 145 women and
37 men) had not received oral care (Table 1).
Before the study, a physical examination and chest ra- RESULTS
diograph were performed. During follow-up, nurses or During follow-up, there were 54 (29%) patients with fe-
caregivers cleaned the patients teeth with a toothbrush for brile days in the nonoral care group and 27 (15%) in the
approximately 5 minutes after each meal. The brushing oral care group. The relative risk (RR) for patients who
was performed as usual daily toothbrushing without den- did not receive oral care compared with those receiving
tifrice, including brushing palatal and mandibular mucosa oral care was 2.45 (95% confidence interval (CI)  1.77
and tongue dorsum. If the toothbrush was not efficient, 3.40, P  .01). New pneumonia was diagnosed in 34 (19%)
the oropharynx was scrubbed with an applicator with a of the 182 patients who did not receive oral care and 21
minimum amount of povidone iodine (1%) in some cases, (11%) of the 184 patients who received oral care. The RR
and rinsing was not needed. In the nonoral care group, for patients who did not receive oral care compared with
several patients performed toothbrushing by themselves those receiving oral care was 1.67 (95% CI  1.012.75,
once a day or irregularly but none of them requested oral P  .05). There were 130 patients who could feed them-
care from caregivers. One hundred sixty-three patients selves in the oral care group and 116 in the nonoral care
used dentures. In both groups, dentures were cleaned with group and 54 patients who need help eating in the oral
a denture brush every day and with denture cleanser once care group and 66 in the nonoral care group. Frequency
a week. Dentists or dental hygienists administered profes- of pneumonia in patients feeding themselves and those
sional care such as plaque and calculus control as neces- needing help eating was 9% and 17% in the oral care group
sary once a week for the oral care group. Plaque score was and 13% and 29% in the nonoral care group, respectively.
measured in dentate patients following the Debris Index The RR of pneumonia for patients who needed help eating
reported by Greene and Vermillion6 (0  no debris and compared with patients who could feed themselves was
stain present, 1  soft debris covering not more than one- 2.72 (95% CI  1.275.82, P  .01) in the nonoral care
third of the tooth surface, 2  soft debris covering more group but was not significant in the oral care group. Of pa-
than one-third but not more than two-thirds of the exposed tients who suffered from pneumonia, 30 (16%) in the non
tooth surface, 3  soft debris covering more than two- oral care group died and 14 (7%) in the oral care group
thirds of the exposed tooth surface). Both groups were died because of pneumonia. The RR for death of patients
treated with usual nursing management of changing sheets due to pneumonia who did not receive oral care compared

Table 1. Comparisons Between Oral Care and No Oral Care Groups

Number Number of Number of Number of

of Age, Years, ADLs at Baseline, MMSE at Baseline, Patients Patients with Patients
Group Patients mean  SD F/M mean  SD mean  SD with Fever (%) Pneumonia (%) Dying (%)

Oral care 184 82.0  7.8 148/36 16.3  6.5 13.6  6.9 27** (15) 21* (11) 14** (7)
No oral care 182 82.1  7.5 145/37 16.2  6.7 13.9  6.9 54 (29) 34 (19) 30 (16)
*P  .05 and **P  .01 show significant differences between groups with oral care and no oral care.
SD  standard deviation; F/M  female/ male; ADLs  activities of daily living; MMSE  Mini-Mental State Examination.
432 YONEYAMA ET AL. MARCH 2002VOL. 50, NO. 3 JAGS

Table 2. Effects of Oral Care on Activities of Daily Living (ADLs) and Mini-Mental State Examination (MMSE)

Changes of ADL or MMSE Every 6 Months

6th Month 12th Month 18th Month 24th Month

Index Group Number of Patients Baseline Scores mean  standard deviation

ADL Oral care 170 16.9  6.3 0.4  2.9 1.4  3.7 1.4  3.4 2.1  4.0
No oral care 152 16.9  6.7 0.4  2.6 1.5  4.0 2.2  3.7 2.3  3.7
MMSE Oral care 170 14.3  6.9 0.3  4.2 0.7  4.9 1.3  6.4 1.5  4.9*
No oral care 152 15.0  8.4 0.4  3.7 1.1  4.8 2.0  5.8 3.0  5.9
*P  .05 shows a significant difference between groups with oral care and no oral care.
ADL  activities of daily living; MMSE  Mini-Mental State Examination.

with those receiving oral care was 2.40 (95% CI  1.54 tion was very old, and some of the subjects could not per-
3.74, P  .01) (Table 1). form basic ADLs, such as eating without help. Frequency
Mean changes from baseline in ADL and MMSE scores of pneumonia in patients who needed help eating was less
are shown in Table 2, excluding patients who died from than that in patients who could feed themselves. The large
pneumonia. ADL and MMSE scores showed a tendency to majority had neglected their oral health care for a long
improve with oral care, but only MMSE at the 24th month time. Thus, the plaque score was high, periodontal disease
was significant (P  .05). and caries were frequent, and dentures, when present, were
Comparisons were made between groups with dentate in poor condition.9 Oral status was similar to that reported
and edentate patients. Age (P  .01) and MMSE at base- in a study on institutionalized older people in various coun-
line (P  .01) showed significant differences (Table 3). tries.10 The present patients suffered from chronic diseases
However, a comparison between age-matched dentate pa- and needed long-term care facilities. Although febrile days
tients with edentate patients showed no significance base- may be caused by many conditions, such as small pulmo-
line MMSE difference. The RR of mortality in the non nary emboli and common colds, the present study suggested
oral care group compared with those in the oral care group that silent aspiration might be an important risk factor for
was 3.20 (95% CI  1.347.64, P  .05) in dentate pa- pneumonia in these patients.11
tients. Fewer edentate patients in the oral care group had Pneumonia as the old mans friend suggests a high
febrile days or pneumonia or died from pneumonia, similar prevalence of pneumonia in older people that results in
to the rates of dentate patients, although most of the differ- high mortality.12 In the present institutionalized patients,
ences between the oral care group and the nonoral care 80% died after contracting pneumonia (Table 1), but in
group were not significant, probably because of the small patients receiving oral care, the mortality due to pneumo-
number of edentate subjects (Table 4). Debris Indices were nia was about half of that in patients not receiving oral care.
2.6  0.8 in the oral care group and 2.5  0.9 in the non A comparison of mortality after pneumonia suggests that
oral care group at study entry and 2.3  0.8 in the oral oral care was much more effective in decreasing mortality
care group and 2.5  0.8 in the nonoral care group at the than were medical treatments for pneumonia.
end of the study. When each patient was categorized into Oral health can dramatically affected the quality of a
an improved or deteriorated group, oral care significantly persons life, affecting such things as chewing, swallowing,
reduced the Debris Index in contrast to nonoral care speaking, facial aesthetics, and social interation.13 MMSE
(2.81, 95% CI  1.395.69, P  .01). at the 24th month in patients receiving oral care was sig-
nificantly different from that in patients not receiving oral
DISCUSSION care, but we did not observe an overall improvement in
We have shown that patients receiving oral care had fewer ADL and MMSE scores due to oral care. We administered
febrile days than did patients not receiving oral care. We the MMSE to patients who were alive for 2 years irrespec-
also found that the risk of pneumonia in patients in a long- tive of pneumonia. Thus, patients suffering from pneumo-
term care facility followed up for 2 years was significantly nia might have ADL and MMSE degradation. Age-matched
reduced in patients receiving oral care. The study popula- baseline MMSE was also not significantly different between

Table 3. Physical Characteristics in Dentate and Edentate Patients

Patients Number of Patients Age, Years, mean  SD F/M ADLs at Baseline, mean  SD MMSE at Baseline, mean  SD

Dentate 208 79.6  7.8* 162/46 16.9  6.6 15.0  9.2*

Edentate 158 84.6  7.3 131/27 15.9  6.7 12.5  8.5
*P  .01 shows a significant difference between dentate and edentate groups.
SD  standard deviation; F/M  female/male; ADLs  activities of daily living; MMSE  Mini-Mental State Examination.
JAGS MARCH 2002VOL. 50, NO. 3 ORAL CARE AND PNEUMONIA 433

Table 4. Comparisons Between Oral Care and No Oral Care Groups in Dentate and Edentate Patients

Number ADLs at MMSE at Number of Number of Number of

of Age, Years, Baseline, Baseline, Patients with Patients with Patients
Patients Group Patients mean  SD F/M mean  SD mean  SD Fever (%) Pneumonia (%) Dying (%)

Dentate Oral care 109 79.9  7.9 82/27 17.1  6.3 14.8  8.5 13** (11) 12** (9) 8* (6)
No oral care 99 79.3  7.6 80/19 16.7  6.8 15.3  9.9 26 (26) 19 (21) 20 (20)
Edentate Oral care 75 84.3  7.4 63/12 15.8  6.5 12.7  7.8 14* (18) 9 (9) 6 (7)
No oral care 83 84.9  7.1 68/15 16.0  6.9 12.4  9.2 28 (34) 15 (20) 10 (13)
*P  .05 and **P  .01 show significant differences between groups with oral care and no oral care.
SD  standard deviation; F/M  female/male; ADLs  activities of daily living; MMSE  Mini-Mental State Examination.

dentate and edentate patients. To be eligible for inclusion in dai Dental Association; Futyu City Dental Association; Yono
the pharmacological trial, dementia patients were required City Dental Association; and the Amagasaki City Dental As-
to have a MMSE score between 10 and 26 and either be sociation for their generous assistance with this study. We are
fully ambulatory or able to walk with an assisting device. 14 also thankful to ELVA Co. Ltd and Inoue Attachment Inc.
MMSE score was approximately 13 in the present group, for their help.
which may not be sufficient to draw conclusions about oral
care. Therefore, ADL and MMSE scores of more-active pa- REFERENCES
tients may improve with oral care, or a longer-term period 1. Niederman MS. Nosocomial pneumonia in the elderly patients: Chronic care
facility and hospital considerations. Clin Chest Med 1993;14:479490.
of care may be needed to improve them. 2. Kikuchi R, Watanabe N, Konno T et al. High incidence of silent aspiration in
A national survey of oral health in U.S. adults reported elderly patients with community-required pneumonia. Am J Respir Crit Care
that more than 41% of Americans aged 65 and older were Med 1994;150:251253.
totally edentate.13 Older edentate persons are supposed to 3. Scannapieco FA, Papandonatos GD, Dunford RG. Association between oral
conditions and respiratory disease in a national sample survey population.
have a lower risk of developing serious dentally derived sys- Ann Periodontol 1998;3:251256.
temic diseases than dentate patients, but the present results 4. Scannapieco FA, Mylotte JM. Relationships between periodontal disease and
suggest that even in older edentate patients, fever, pneumo- bacterial pneumonia. J Periodontol 1996;67:11141122.
nia, and death from pneumonia were observed as often as 5. Yoneyama T, Yoshida M, Matsui T et al. Oral care and pneumonia. Lancet
1999;354:515.
in dentate patients, and oral care in edentate patients pro- 6. Greene JC, Vermillion JR. The oral hygiene index: A method for classifying
duced benefits similar to those in dentate patients. oral hygiene status. J Am Dent Assoc 1957;61:171179.
We conclude that removal of latent oral infections and 7. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical
potential problems might be a way to reduce the incidence method for grading the cognitive state of patients for the clinician. J Psychiatr
Res 1975;12:189198.
of lower respiratory tract infection. In light of the central 8. Mahonery FI, Barthel DW. Functional evaluation: Barthel index. MD State
importance of oral health to an older patients overall well- Med J 1965;14:6165.
being, it is essential that all health providers support and 9. Ellen RP. Considerations for physicians caring for older adults with peri-
advocate the expansion of oral health benefits for older odontal disease. Clin Geriatr Med 1992;8:599616.
10. Mojon P, Budtz-Jrgensen E, Michel JP et al. Oral health and history of res-
adults, dentate or edentate. piratory tract infection in frail institutionalized elders. Gerodontology 1997;
14:916.
11. Yamaya M, Yanai M, Ohrui T et al. Interventions to prevent pneumonia
ACKNOWLEDGMENTS
among older adults. J Am Geriatr Soc 2001;49:8590.
We are very grateful to the Department of Removable Pros- 12. Osler W. The principles and practice of medicine. New York, NY: D Apple-
thodontics, Faculty of Dentistry Hiroshima University; the ton and Co, 1898.
13. Shay K, Ship JA. The importance of oral health in the older patient. J Am
Department of Dental and Oral and Maxillofacial Surgery, Geriatr Soc 1995;43:14141422.
Hamamatsu University School of Medicine; the Department 14. Rogers SL, Doody RS, Mohs RC et al. Donepezil improves cognition and
of Periodontology, Ohu University School of Dentistry; Sen- global function in Alzheimer disease. Arch Intern Med 1998;158:10211031.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.