SEARCH

SEARCH BY CITATION

Keywords:

  • ageing;
  • cross-sectional study;
  • epidemiology;
  • oral health related quality of life;
  • xerostomia

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

doi: 10.1111/j.1741-2358.2010.00420.x

Self-reported dry mouth in Swedish population samples aged 50, 65 and 75 years

Background:  Reduced salivary flow may have a negative impact on general well-being, quality of life and oral health.

Objectives:  To examine xerostomia in 50-, 65- and 75-year-olds, background factors and effect on Oral Impacts on Daily Performances (OIDP).

Methods:  In 1992, a questionnaire was sent to all 50-year-old persons (n = 8888) in two Swedish counties. In 2007, the same questionnaire was sent to all 65-year-olds (n = 8313) in the two counties and to all 75-year-olds (n = 5195). Response rate was for the 50, 65 and 75 year olds 71.4, 73.1 and 71.9%, respectively.

Results:  Xerostomia was higher in women than in men in all age groups. There was higher prevalence of xerostomia with increasing age in both sexes and it was more frequent at night than during daytime. ‘Often mouth dryness’ was 2.6–3.4 times more prevalent in those who reported an impact from OIDP. The highest odd ratios were for daytime xerostomia and for the variables burning mouth (17.1), not feeling healthy (4.5), daily smoking (4.4), and medication (4.1).

Conclusions:  The dramatic increase of xerostomia between age 50 and 75, especially amongst women, needs to be considered in the management of this age group.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Self-reported mouth dryness or xerostomia (these terms are used as synonyms in the following text) is commonly used in population studies since diagnosing hyposalivation by measurement of saliva flow rates is time-consuming. The amount of saliva needed for a person to experience oral comfort is on a case by case basis and studies have shown that xerostomia and hyposalivation are not always closely related1–4.

The prevalence of xerostomia varies a great deal between different population studies. A systematic review found a range from 1 to 65%5. Several studies have reported higher prevalence in elderly than in younger subjects6–9. In a longitudinal study of a large Swedish population sample, there was an almost linear increase in the prevalence of xerostomia from the age of 50–65 years and it was also found that xerostomia was more common amongst women. In addition, night-time xerostomia was significantly more common than xerostomia during the day, a finding that seldom has been described in previous studies10. It was suggested that a differentiation should be done between day and night-time dry mouth as differences were also demonstrated between the two types and related background variables10. For example, impaired health and smoking were found to be significantly associated with self-reported daytime but not with night-time mouth dryness.

Several studies have shown that the discomfort caused by xerostomia may have an influence on Oral Health-Related Quality of Life (OHRQL)11–13. It would be of interest to study this association further in addition to investigating if differences between day and night-time xerostomia also exist in subjects above 65.

It was therefore the first aim of this study to examine the prevalence of day and night-time xerostomia in a cohort of 75-year-old subjects from the same area in Sweden as the longitudinal sample presented previously10. The second and third aims were to investigate background factors to xerostomia and its possible effect on OHRQL, respectively.

It is hypothesised that xerostomia is more common at night than during daytime and that the prevalence is higher in the 75-year-old cohort than in the 50 and 65-year-old subjects and that xerostomia affects OHRQL.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

In 1992, a questionnaire was sent to all 50-year-old persons (born in 1942) in two counties in Sweden, Örebro and Östergötland, altogether 8888 persons, of whom 6346 responded. The response rate (71.4%) was the same in both counties14. In 2007, the same questionnaire was sent to all 65-year-old persons (born in 1942) who lived in the two counties (n = 8313). The response rate was 73.1%. Those who answered both the 1992 and 2007 questionnaires constituted the panel for a longitudinal study of 50- to 65-year-old subjects presented in a separate paper10.

In 2007, the same questionnaire was sent also to all 75-year-old subjects (born in 1932, n = 5195) in the two counties. The questionnaire was completed by 3735 persons, giving a response rate of 71.9%. This study will focus on the 50-year-old subjects examined in 1992 and the 65- and 75-year-old subjects examined in 2007 (Table 1).

Table 1.   Number of participants and participation rate (%) in the three cohorts.
Age group Examined year Total number Women/Men Responders Participation rate
50-year-old199288883184/3162634671.4
65-year-old200783133080/2998607873.1
75-year-old200751951987/1748373571.9

The original study in 1992 was approved by the Ethics Committee in the Örebro and Östergötland region, Sweden, but as a consequence of new regulations, an approval by an ethical committee of the follow-up examinations in 2007 was not required.

Analyses of non-response

The collected data from the 50- and 65-year-olds have been analysed and the results presented previously. There were only small deviations from a random distribution10. Regarding the 75-year-olds, there was no significant differences as regards county, whilst the female/male ratio differed significantly between the responders (53.2% vs. 46.8%) and non-responders (58.7% vs. 41.3%) (p ≤ 0.01).

Questionnaire

The questionnaire comprised 53 questions. The questions were divided into socio-economic conditions (e.g. age, gender, occupation), general health (e.g. self-reported health, physician visits, tobacco habits, drug consumption), and oral conditions (e.g. satisfaction with teeth and dental appearance, oral problems, oral hygiene habits, number of teeth). This study has focussed on answers to two questions regarding perceived mouth dryness. The wordings of these questions were: (1) ‘Does your mouth usually feel dry in the daytime?’ and (2) ‘Does your mouth usually feel dry at night?’ Both questions had four response alternatives (yes often, yes sometimes, no seldom, no never). The complete questionnaire design, originally used in a study of 50-year-old subjects in 1992, has previously been described and discussed14.

At the examination in 2007 questions were added to study the relationship between xerostomia and OHRQL using the instrument Oral Impacts on Daily Performances (OIDP). The eight-item OIDP comprises the following questions: ‘During the previous six months, how often have problems with teeth or mouth caused you any difficulty with: (1) eating and enjoying food, (2) speaking and pronouncing clearly, (3) tooth cleaning, (4) sleep and relaxation, (5) smiling and showing teeth without being embarrassed, (6) being emotionally stable, (7) being sociable, (8) performing daily chores’. Each item was scored from 1, i.e. every, or almost every, day, to 5, i.e. never, and was assessed using a 5-point scale: (1) ‘‘never affected’’, (2) ‘‘less than once a month’’, (3) ‘‘once or twice a month’’, (4) ‘‘once or twice a week’’, (5) ‘‘every or nearly every day’’15.

Statistical methods

All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS, Release 15). The Mann–Whitney U-test was used for testing differences between genders and between the 65 and 75 year group. The Wilcoxon signed rank test was used for testing differences between subjects at 50–65 years of age and between day- and night-time xerostomia at each occasion.

Logistic regression model (Forward Conditional Method) was computed using daytime and night-time xerostomia as dependent variables. The independent variables used were dichotomised (Table 2). The selection of the independent variables was according to previous papers10,16, i.e. a Spearman correlation analysis was first performed between the dependent variable (1 = never dry mouth; 2 = often dry mouth) and all recorded variables (n = 72). Numerous significant correlations were exhibited. From the groups of variables that were interrelated, one representative variable was chosen for the final selection.

Table 2.   Dichotomisation of independent variables (description 1 and 2) used in the stepwise logistic regression models.
Variable Description 1 Description 2
GenderManWoman
Working hoursFull-timeLess than full-time
Place of birthSwedenOutside Sweden
EducationUniversityHigh-/elementary-/lower-school
HealthyYes/on the wholeNo/absolutely not
Use of medicine last 2 weeksYesNo
SmokingDailyNot daily, stopped, never
Chewing abilityVery goodRather good/not so good/bad
ToothacheDuring the last yearMore than 1 year ago/never/do not remember
Number of teethAll or almost all remainingMany missing and no teeth
Removable complete or partial dentureYesNo
Burning mouthNo problemsSome problems/rather many/great problems
Taste changes
Sensitive teeth
TMJ pain
Difficulty wide opening
Bruxism
Gum bleeding
Bad mouth smell
Dental restorations

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

The prevalence of reported mouth dryness was significantly higher in women than in men in all age groups (p < 0.001) except for night-time xerostomia at age 50 were it was higher but not significantly so. There was higher prevalence of xerostomia with increasing age in both sexes (< 0.001) for differences between each consecutive age group. In all age groups, xerostomia was more frequent at night than during daytime (Tables 3 and 4; p < 0.001) except for women at age 50. Approximately a fifth of the 50-year-old subjects had experienced xerostomia (‘yes, often’ or ‘yes, sometimes’) during daytime, and a quarter at night. Amongst the 75-year-old subjects, the corresponding proportions were a third and more than half, respectively (Fig. 1).

Table 3.   Daytime xerostomia. Distribution (%) of answers at age 50 (n = 8888), 65 (n = 8313) and 75 (n = 5195) year-old subjects to the question ‘Does your mouth usually feel dry in the daytime’ with four response alternatives.
Answer Women Men
50 yearsn = 3105 65 years n = 2969 75 yearsn = 1864 50 yearsn = 3096 65 yearsn = 2906 75 yearsn = 1675
Yes, often4.98.111.02.13.95.7
Yes, sometimes21.626.032.615.819.325.0
No, seldom33.333.527.741.539.836.1
No, never40.232.428.640.637.133.1
Table 4.   Night-time xerostomia. Distribution (%) of answers at age 50 (n = 8888), 65 (n = 8313) and 75 (n = 5195) year-old subjects to the question ‘Does your mouth usually feel dry at night’ with four response alternatives.
Answer Women Men
50 yearsn = 2975 65 yearsn = 3001 75 yearsn = 1897 50 yearsn = 2977 65 yearsn = 2934 75 yearsn = 1694
Yes, often5.917.222.34.611.515.3
Yes, sometimes24.135.141.021.833.236.0
No, seldom28.123.317.435.729.625.7
No, never41.824.419.338.025.722.9
image

Figure 1.  Distribution (%) of day and night-time xerostomia at age 50-, 65- and 75-year-olds reporting ‘yes, often’ mouth dryness.

Download figure to PowerPoint

The mean pooled prevalence of night and daytime ‘often’ xerostomia (the highest alternative of the answers recorded for the two questions) was for both sexes 6.6% at age 50, 15.9% at age 65 and 20.9% at age 75. The difference between women and men was also here significant at age 50 (p < 0.05) and at age 65 and 75 (p < 0.001) (Table 5).

Table 5.   Distribution (%) of answers amongst 50 (n = 8888), 65 (n = 8313) and 75 (n = 5195) year-old subjects to the questions ‘Does your mouth usually feel dry in daytime’ or ‘Does your mouth usually feel dry at night-time’ with four response alternatives. The prevalence figures for day and night are pooleda.
Answer Total Women Men
50 yearsn = 6310 65 yearsn = 5960 75 yearsn = 3611 50 yearsn = 3168 65 yearsn = 3018 75 yearsn = 1907 50 yearsn = 3142 65 yearsn = 2942 75 yearsn = 1704
  1. aThe highest value recorded for the two questions.

Yes, often6.615.920.97.919.124.65.312.616.7
Yes, sometimes27.836.541.429.637.543.725.935.438.8
No, seldom33.726.820.629.623.916.937.829.824.7
No, never31.920.917.132.819.514.831.022.319.7

‘Often dry mouth’ at both night and day was significantly more common amongst women and increased more or less linearly from age 50–75 in both genders and was highest in 75-year-old women (8.6%). ‘Never mouth dryness’ at both night and day was most common at age 50 (≈30%) and decreased in both sexes up to age 75. There were no significant differences between women and men at age 50 and 65 but was so at 75 (Table 6).

Table 6.   Frequency and percentage distribution (% of the total sample) of ‘yes, often’ and ‘no, never’ mouth dryness at both night and day in the different groups. p denotes differences between women and men.
Age group Gender Yes, often night and day p No, never night and day p
Frequency % Frequency %
50 yearsWomen782.7<0.00196633.2NS
Men351.291831.3
65 yearsWomen1796.1<0.00158319.7NS
Men802.865222.5
75 yearsWomen1598.6<0.00127915.00.024
Men714.333520.1

Approximately 26–30% of the individuals, with small differences between gender and age groups, reported at least one impact from the eight-item questionnaire in OIDP (Table 7). The response alternative ‘often xerostomia’ was 1.5–3.3 times more prevalent in those who reported an impact than in those who did not (Table 8).

Table 7.   Impact according to the 8-item questions in OIDP in 65- and 75-year-old subjects in 2007.
Age group Gender   Frequency %
65 yearsWomenNo impact211170.4
Impact88729.6
Total2998100.0
MenNo impact208171.7
Impact82128.3
Total2902100.0
75 yearsWomenNo impact138173.5
Impact49826.5
Total1879100.0
MenNo impact121773.8
Impact43326.2
Total1650100.0
Table 8.   Comparison between individuals with or without impact of xerostomia at age 65 and 75 according to OIDP regarding the response alternative ‘often daytime or night-time xerostomia’. p refers to the difference between individuals with or without impact.
Age group Gender Impact Yes, often daytime p Yes, often night-time p
Frequency % Frequency %
65 yearsWomenNo impact (n = 2111)1115.3< 0.00129013.7< 0.001
Impact (n = 887)12514.121524.2
MenNo impact (n = 2081)462.2< 0.0011909.1< 0.001
Impact (n = 821)607.314117.2
75 yearsWomenNo impact (n = 1381)1047.5< 0.00126619.3< 0.001
Impact (n = 498)9218.514128.3
MenNo impact (n = 1217)483.9< 0.00115012.3< 0.001
Impact n = 433)4510.39922.9

The logistic regression models showed somewhat different results for the three age groups and for day and night-time xerostomia (Tables 9 and 10). Age group 75 had the fewest significant associations between daytime xerostomia and the independent variables but also the highest OR (17.1 for burning mouth). Daily smoking exhibited the strongest association in the youngest age group (OR 4.4 at age 50) but was non-significant at age 75. The association with ‘not feeling healthy’ increased with age (Table 9).

Table 9.   Logistic regression model (Forward Conditional Method) for daytime xerostomia, presenting independent variables significantly associated with the dependent variable at age 50, 65 and 75 years. Dependent variable 1 = never dry mouth daytime, 2 = often dry mouth daytime. OR = odds ratio, CI = confidence interval for OR. n = number of individuals in group 1/group 2.
  50 yearsn = 2505/218 65 yearsn = 2040/352 75 yearsn = 1089/301
OR 95% CI p OR 95% CI p OR 95% CI p
Female gender2.31.5–3.4<0.0012.81.9–4.0<0.0012.31.5–3.5<0.001
Born outside Sweden2.61.2–5.80.018NSNS
Not feeling healthy2.81.8–4.3<0.0013.42.4–4.9<0.0014.52.9–6.9<0.001
Working <full timeNS1.71.1–2.70.028NS
Medicine usage3.82.5–5.6<0.0014.12.5–6.6<0.0012.91.5–5.90.002
Daily smoking4.43.0–6.4<0.0012.41.6–3.7<0.001NS
Impaired chewing1.81.2–2.80.0031.61.1–2.40.0222.41.6–3.7<0.001
Burning mouth2.91.4–6.10.0055.63.2–10.0<0.00117.16.9–42.4<0.001
Taste changes3.41.9–6.2<0.0013.42.0–5.7<0.001NS
TMJ pain2.01.2–3.50.011NSNS
Difficulty wide opening2.01.1–3.40.016NSNS
Gum bleeding2.51.7–3.6<0.001NS2.41.5–3.9<0.001
BruxismNS1.61.1–2.30.024NS
Bad mouth smellNS2.21.5–3.1<0.001NS
Reduced no. of teethNS1.91.3–2.80.001NS
Problem w. restorationsNS1.91.1–3.30.018NS
BlistersNSNS2.01.1–3.60.032
Nagelkerke R20.380.460.45
Table 10.   Logistic regression model (Forward Conditional Method) for night-time xerostomia presenting independent variables significantly associated with the dependent variable at age 50, 65 and 75 years. Dependent variable 1 = never dry mouth night-time, 2 = often dry mouth night-time. OR = Odds ratio, CI=confidence interval for OR. n = number of individuals in group 1/group 2.
  50 yearsn = 2375/312 65 yearsn = 1487/853 75 yearsn = 755/683
OR 95% CI p OR 95% CI p OR 95% CI p
Female genderNS1.71.4–2.1<0.0011.81.3–2.4<0.001
Lower education2.51.6–3.8<0.001NSNS
Not feeling healthy2.41.7–3.5<0.0011.91.4–2.5<0.0013.42.4–4.8<0.001
Medicine usage2.31.7–3.0<0.0012.21.7–2.8<0.0012.71.8–4.0<0.001
Daily smoking1.61.2–2.20.002NSNS
Impaired chewing1.71.3–2.40.0011.91.5–2.4<0.0012.41.7–3.4<0.001
Burning mouth2.41.4–4.20.0022.41.4–4.00.00112.14.6–32.0<0.001
Taste changes2.61.6–4.2<0.0013.22.1–5.0<0.001NS
TMJ painNSNS2.31.2–4.60.018
Difficulty wide opening2.01.3–3.00.0021.51.1–2.10.024NS
Gum bleeding1.61.2–2.10.0031.61.3–2.1<0.0012.01.3–3.00.001
BruxismNSNS2.31.4–3.70.001
Bad mouth smellNS1.51.2–2.00.0011.81.2–2.70.003
Tooth sensitivity1.71.2–2.30.0011.41.1–1.80.006NS
BlistersNS1.41.0–2.00.05NS
Wearing dentureNSNS1.71.1–2.60.009
Nagelkerke R20.240.280.40

The associations between night-time xerostomia and the independent variables were in general smaller than for daytime xerostomia. Only burning mouth and medicine usage demonstrated OR > 2 in all three age groups for night-time xerostomia (Table 10).

Nagelkerke R2 for daytime xerostomia ranged from 0.38 to 0.46 and for night-time xerostomia from 0.24 to 0.40 (Tables 9 and 10).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

As expected the prevalence of reported mouth dryness increased more or less linearly with age, as demonstrated by the results of the large cross-sectional samples aged 50, 65 and 75 years in this study (Fig. 1). The prevalence of xerostomia was, with some exceptions, also higher during night- than daytime as well as being more common in women than in men in all age-groups. This finding can be explained by the fact that salivary flow follows the circadian rhythm and is lowest during night-time17 and the results of the study were thus in line with our hypotheses and corroborated earlier cross-sectional studies2,3,10.

One reason for the varying results is the different methods used in collecting the data, for example the wording of questions for measuring dry mouth symptoms18. The prevalence presented by Nederfors et al. was based on the question ‘Does your mouth usually feel dry?’19. They found that 19% of the men at age 50 and 28% at age 60 answered affirmatively. The corresponding figures in women were 25 and 36%, respectively. Our results were based on a similar question but divided between dryness during the day or at night, to be answered with one of four alternatives. The results from this study are smaller when focussing on the answer ‘Yes often’, but higher when combining ‘Yes often’ and ‘Yes sometimes’ (Tables 3, 4 and Fig. 1).

Another comparison of interest is the prevalence of dryness of the mouth in 70-year-old subjects. To the question ‘Does your mouth feel distinctly dry?’ 16% of the men and 25% of the women answered positively20. A longitudinal population study of 70-year-old subjects found that complaints of dry mouth increased with age; but this was significant only for women. Reported mouth dryness was significantly associated with lower salivary flow rate. However, in the population there was no decline in stimulated salivary flow with increasing age from 70 years6. In a study of 32-year-old subjects, the prevalence of xerostomia was found to be 10%, with no gender difference. It was also suggested that xerostomia later on in life may develop differently amongst females than amongst males8 which has been confirmed in several studies. This suggestion was to some extent confirmed in the present study: the differences between women and men were quite moderate at age 50 but developed to be substantial at age 75. A smaller gender difference at age 50 compared to age 65 was also found in a previous study10. Our results are also largely in agreement with two Swedish cross-sectional population studies, where similar gender and age differences in xerostomia were demonstrated3,19. In a study of individuals aged 20–80 years, women reported a higher prevalence of xerostomia than men (independent of age) and that the symptom of dry mouth was strongly associated with increasing age19. In the other study, unstimulated and stimulated whole salivary flow rates were measured in 1427 dental patients aged from 20 to 69 years and hyposalivation was correlated to age above 50 years and female gender3.

‘Often mouth dryness’ at both night and day was relatively seldom reported at age 50 but increased with age and the prevalence was 8.6% in women aged 75. There was no significant gender difference in those who reported ‘never mouth dryness’ at both day and night-time at age 50 and 65 but at age 75 women had a significantly lower prevalence of ‘never mouth dryness’ than men. This finding again underlines the gender and age related differences in reported mouth dryness and that women are at a higher risk.

The strong associations between xerostomia and burning mouth, taste changes and para- and dysfunction of the masticatory system corroborate earlier findings and demonstrate the complexity of these conditions21–23. Many of these associations could well be deemed as not being independent of a dry mouth but rather caused by it. It could equally be argued that these variables mirror underlying physio-pathological processes, which cause dry mouth. Smoking, a truly independent variable, has been considered a risk factor for dry mouth18. The present results (OR 1.6–4.4 at age 50 and 65) support this opinion.

There are several reports claiming that the main causes for xerostomia and hyposalivation are systematic diseases and medication19,20,24,25 which also was supported in this study (variables ‘medicine usage’ and ‘not feeling healthy’). These findings may however be debated because other or additional factors may also play a role. In this regard, it has been suggested that both quantity and quality of saliva will be influenced by the menopause in women, thus being potentially important for xerostomia. Some studies on healthy women have reported higher salivary secretion before menopause than after26 whilst others did not find any difference27. The different composition of saliva in peri- and post-menopausal women has been suggested to be depending on the amount of oestrogen28. Other possible factors influencing xerostomia are age-related changes in the composition of saliva29 and geriatric malnutrition30.

A weighted average of 21%, with a range of 12–39%, was calculated for estimates of prevalence of xerostomia from epidemiological studies of older populations18. The great variation in many studies has been discussed with respect to the assessment of xerostomia and its consequences for the interpretation of the results. Furthermore, most studies have not considered the significant difference in prevalence between daytime and night-time xerostomia reported in the present and a previous Swedish study10. This finding is not surprising though it is well known that salivary secretion is lower at night than during daytime17. It seems necessary to take that finding into account in the analyses of xerostomia since the logistic regression analyses indicated that day and night-time xerostomia may be different conditions or at any rate have a somewhat different background.

Impaired OHRQL has previously been shown to be associated with xerostomia8,13,31,32. In this study, the prevalence of xerostomia was significantly more common in subjects who had an impact on OIDP compared to those who had not, thus corroborating these findings. The instrument for measuring OHRQL (OIDP) has earlier been described, used and validated33. From the results of this study it was clear that xerostomia was affecting OHRQL measured with the OIDP instrument. This was especially obvious for daytime xerostomia. The corresponding figures for OIDP impact on night-time xerostomia was substantially lower (results not shown) and it could be discussed whether OIDP is suitable to use for night-time comparison since most of the items in OIDP is dealing with daytime activities.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References
  •  The dramatic increase of self-reported mouth dryness between age 50–75, especially amongst women, needs to be considered in the management of this age group.
  •  Burning mouth, impaired health, daily smoking and medicine usage are strong predictors for self-reported dry mouth.
  •  Xerostomia affects oral health related quality of life, shown as a negative impact on daily performances.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References