Journal of Public Health Dentistry . ISSN 0022-4006
Psychological distress and self-rated oral health among a convenience sample of Indigenous Australians Najith Amarasena, PhD1; Kostas Kapellas, PhD2; Alex Brown, PhD3; Michael R. Skilton, PhD4; Louise J. Maple-Brown, PhD5; Mark P. Bartold, PhD6; Kerin O’Dea, PhD7; David Celermajer, PhD4; Gary Douglas Slade, PhD2; Lisa Jamieson, PhD2 1 2 3 4 5 6 7
Australian Research Centre for Population Oral Health, School of Dentistry, The University of Adelaide, Adelaide, South Australia, Australia School of Dentistry, University of Adelaide, Adelaide, South Australia, Australia Indigenous Health Research, South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney, Sydney, New South Wales, Australia Menzies School of Health Research, Charles Darwin University, Darwin, Northern Territory, Australia Colgate Australian Clinical Dental Research Centre, University of Adelaide, Adelaide, South Australia, Australia Sansom Research Institute, University of South Australia, Adelaide, South Australia, Australia
Keywords Indigenous Australian; poor self-rated oral health; psychological distress. Correspondence Dr. Lisa Jamieson, School of Dentistry, University of Adelaide, 122 Frome St, Adelaide, SA 5005, Australia. Tel.: + 61-8-8303-4611; Fax: + 61-8-8303-4858; e-mail: [email protected]. Najith Amarasena is with the Australian Research Centre for Population Oral Health, School of Dentistry, The University of Adelaide. Kostas Kapellas is with the Dentistry, University of Adelaide. Alex Brown is with the Indigenous Health Research, South Australian Health and Medical Research Institute. Michael R. Skilton and David Celermajer are with the Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney. Louise J. Maple-Brown is with the Menzies School of Health Research, Charles Darwin University. Mark P. Bartold is with the Colgate Australian Clinical Dental Research Centre, University of Adelaide. Kerin O’Dea is with the Sansom Research Institute, University of South Australia. Gary Douglas Slade and Lisa Jamieson are with the School of Dentistry, University of Adelaide.
Abstract Objectives: This study sought to: a) estimate the frequency of poor self-rated oral health as assessed by a summary measure; b) compare frequency according to sociodemographic, behavioral, and psychological distress factors; and (3) determine if psychological distress was associated with poor self-rated oral health after adjusting for confounding. Methods: Data were from a convenience sample of Indigenous Australian adults (n = 289) residing in Australia’s Northern Territory. Poor self-rated oral health was defined as reported experience of toothache, poor dental appearance or food avoidance in the last 12 months. A logistic regression model was used to evaluate socio-demographic, behavioral, and psychological distress associations with poor self-rated oral health (SROH). Effects were quantified as odds ratios (OR). Results: The frequency of poor SROH was 73.7 percent. High psychological distress, measured by a Kessler-6 score ≥8, was experienced by 33.9 percent of participants. Poor SROH was associated with high levels of psychological distress, being older, being female, and usually visiting a dentist because of a problem. In the multivariable model, factors that were significantly associated with poor SROH after adjustment for other covariates included having a high level of psychological distress (OR 2.74, 95% CI 1.25–6.00), being female (OR 2.22, 95% CI 1.03–4.78), and usually visiting a dentist because of a problem (OR 3.57, 95% CI 1.89–6.76). Conclusions: Poor self-rated oral health and high levels of psychological distress were both highly frequent among this vulnerable population. Psychological distress was significantly associated with poor self-rated oral health after adjustment for confounding.
Received: 10/14/2013; accepted: 10/17/2014. doi: 10.1111/jphd.12080 Journal of Public Health Dentistry •• (2014) ••–••
Introduction Traditionally, measures in oral health epidemiology have been predominately clinical (1). However, these indices have © 2014 American Association of Public Health Dentistry
been criticized as inadequately reflecting the oral health status, concerns, and perceived needs of individuals (2). This criticism has led to the development of instruments measuring self-reported impacts of oral health, which enable greater 1
Distress and oral health in Indigenous Australians
insight into self-perceived oral health status and perceived need for oral health care (3). Although these instruments vary in their detail, all include assessments of three critical aspects of oral health: pain, appearance, and function. Locker et al. reported that oral health-related quality of life measures should incorporate “rate frequency, severity and importance; allow respondents to add supplemental items, and use global ratings which are summary variables and can reflect the differing values and preferences of a group of individuals” (4). The oral health impairment instrument used in this study does not meet these criteria, and thus is a reflection of selfrated oral health rather than oral health-related quality of life. The literature indicates that poor self-rated oral health may have substantial economic and social capital impacts (5). In general population settings, poor self-rated oral health has been associated with untreated dental decay, periodontal disease, difficulty paying a $100 dental bill, usually visiting a dentist because of a problem, cost barriers to receipt of dental care, and dental fear (6). In Indigenous Australian settings, it is important to consider other factors that are somewhat unique to this high-risk population. For example, levels of distress and poor psychological health are far higher among this group (7). It has been reported that the average Indigenous Australian family is confronted by seven major negative events every year: death, arrest, incarceration, hospitalization, alcohol abuse, extreme violence, and severe financial strains (8). Almost one in four Indigenous Australian families has even higher levels (9). By way of comparison, non-Indigenous Australian families typically experience less than one such event per year. Levels of physical activity are also reportedly lower (9), whereas levels of tobacco smoking are higher (10). These psychosocial and lifestyle factors have important associations with oral health outcomes (11-14). The aims of this analysis are: a) to estimate the frequency of poor self-rated oral health as assessed by a summary measure (including aspects of dental pain, dissatisfaction with dental appearance, and difficulty eating) among a convenience sample of Indigenous Australians; b) to compare frequency according to associations with sociodemographic, behavioral, and psychological distress factors; and c) to determine if psychological distress was associated with poor self-rated oral health after adjusting for confounding.
Methods Participants were Indigenous adults aged 22 years and over who were recruited from two regional jurisdictions in the Northern Territory and correctional centers in Alice Springs and Darwin, as part of the PerioCardio study described in detail elsewhere (15). Data used in this paper were thus crosssectional and from a convenience sample, collected between June 2010 and January 2012. The participation rate was 100 percent, and there were no incomplete responses (due to 2
N. Amarasena et al.
questionnaire data being collected via interview). The recruitment approach included liaising with Aboriginal community champions previously involved in our research projects, engaging with key community stakeholder groups, encouraging word-of-mouth spread of knowledge about the study, advertisements in local newspapers and radio shows, and presentations made to local Aboriginal community groups. A snowballing technique was also employed, with participants asked to contact any Aboriginal friends, family and peers who may be interested in partaking. Data were collected through a questionnaire, which included information on sociodemographic characteristics, oral health status including oral health behaviors and perceptions, general health behaviors, including tobacco smoking status and physical activity, and psychological distress. Ethical approval for the study was obtained from the Human Research Ethics Committee of the Northern Territory Department of Health and Menzies School of Health Research, the Central Australian Human Research Ethics Committee, Northern Territory Correctional Services Research Committee, University of Adelaide Human Research Ethics Committee, and the Aboriginal Health Council of South Australia.
Dependent variable The dependent variable was poor self-rated oral health, the definition of which has been described in detail elsewhere (6). A summary self-rated oral health variable was created by combining three items: experience of toothache, experience of discomfort due to mouth appearance, and food avoidance. Experience of toothache was assessed by asking “How often during the last year did you have toothache?,” whereas experience of discomfort due to mouth appearance was assessed by asking “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth or false teeth?” Avoiding food because of oral health problems was assessed by asking “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?” For purposes of this analysis, those who answered “Very often,” “Often,” or “Occasionally” to any of the items were considered to have poor self-rated oral health because of oral health-related factors. Other possible responses were “Hardly ever” or “Never.” These are standard cutoffs used for analyses involving oral health impairment-type statements such as the Oral Health Impact Profile (16).
Main exposure variable Psychological distress was measured by an adapted Kessler-6 (K6) instrument for use among Aboriginal Australians (17,18). The original K6 is widely used as a simple measure of © 2014 American Association of Public Health Dentistry
N. Amarasena et al.
Distress and oral health in Indigenous Australians
psychological distress (19) and consists of six questions about depressive and anxiety symptoms. It has been widely used in Aboriginal and Torres Strait Islander peoples in jurisdictional and national population-based surveys. Brown et al. (16,17) undertook extensive developmental adaptation and translation of the K6 and Patient Health Questionnaire-9 (PHQ-9) depression instruments in five language groups in Central Australia. In brief, the modification process involved detailed qualitative work to explore the expression of emotional distress within the target population, and repeated translation and back-translation of the original instruments into a combined modified English version (Table 1). Based on evidence in the literature (19), the summary scores were trichotomized into “low” (0–2), “moderate” (3–7), and “high” (8+). The K6 scores in our study were determined using the same methods as that by Kessler et al. (19). Study participants were asked the respective questions during a face-toface interview using previously validated methodology specific to Indigenous Australians (16,17). Values from 0–4 corresponding to responses “never” – “always” were summed to obtain an individual participant score (range 0–24). Individuals with a score of 8 or higher were classified as “stressed/depressed.”
Table 1 Original K6 Items Compared with Those Moderated for Use among Aboriginal Australians Original K6 items (17) During the past 4 weeks, how much of the time did you feel . . . . .* 1. . . . so sad nothing could cheer you up 2. . . . nervous 3. . . . restless or fidgety
4. . . . hopeless
5. . . . that everything was an effort 6. . . . worthless
Modified for use among Aboriginal Australians (6) During the last 12 months, have you . . . . .† 1. . . . felt depressed or no good, your spirit was really sad, that nothing can make you feel better? 2. . . . been feeling frightened, scared or nervous inside? 3. . . . been feeling that you can’t sit still or rest or that you keep moving around too much? 4. . . . been feeling that life is hopeless, that your worries never will go away, that nothing will get better? 5. . . . felt that you couldn’t do anything, that it was all too hard? 6. . . . been feeling no good about yourself, that you are not a good person, that you are rubbish or worthless?
* Response options: “all of the time,” “most of the time,” “some of the time,” “a little of the time,” “none of the time.” † Response options: “very often,” “fairly often,” “occasionally,” “hardly ever,” “never.”
© 2014 American Association of Public Health Dentistry
Covariates Covariates included sociodemographic factors (age, sex, education, income, and employment status) and behaviors (brushing teeth previous day, time since last dental visit, usual reason for seeing a dentist, tobacco smoking, and physical activity). Education was dichotomized into receiving “less than 12 years education” and “receiving 12 + years education,” income was dichotomized into “annual income (per person) < $15,600” and “annual income per person $15,600 +” and employment status was dichotomized into “currently employed” and “currently unemployed.” Tobacco smoking was calculated based on the response to “Have you smoked more than 100 cigarettes in your life time?,” with responses dichotomized into “yes” or “no.” The WHO’s Global PhysicalActivity Questionnaire Version 2 was used to assess physical activity, with analysis conducted as per recommendations in the analysis guide (20). Physical activity was dichotomized into “low” and “moderate/high.”
Analytic methods The number of participants who identified as being Aboriginal or Torres Strait Islander and who provided complete responses was 289 (a priori sample size calculation). All subsequent analyses pertained to these participants only. Blocks of explanatory variables that were significant at the P < 0.2 level in bivariate analysis were entered into a binary logistic regression model in three steps. The dependent variable of these models was poor self-rated oral health. Correlations between explanatory variables were weak to moderate, ranging from rho = −0.60 to +0.29. Psychological distress was entered in Model 1, with the main effect presented as an odds ratio (OR) with its 95 percent confidence interval (95 percent CI). The sociodemographic factors were entered into Model 2, and behavioral factors entered into Model 3. Data were analyzed using SAS version 9.3 (SAS, Cary, NC, USA). An additional validation step was undertaken using an alternative definition to define poor self-rated oral health as a positive report of all three impacts: toothache, dissatisfaction with appearance, and food avoidance. This more stringent case definition necessarily created a lower frequency of impairment; the main objective in its use was to determine if similar associations emerged from a similar multivariable analysis.
Results Approximately 40 percent of participants experienced toothache in the past year, whereas just under 55 percent experienced dissatisfaction with their dental appearance and 48 percent had avoided food because of problems with their 3
Distress and oral health in Indigenous Australians
teeth, mouth, or false teeth (Table 2). Almost three-quarters had reported one or more poor self-rated measures of oral health in the previous year, with just under one-third having reported all three poor self-rated measures of oral health. The mean K6 score was 6.0, with a standard deviation of 5.3. Scores ranged from 0 to 22 (possible range 0–24). Less than 12 percent of participants scored 13+, which is the recommended cutoff point for serious psychological distress (21). Participants were evenly distributed across the psychological distress categories, with approximately one-third in each. The age range was 22–73, with a mean age of 39.6 years (standard deviation = 10.3 years). Around 54 percent of participants were male, and approximately 31 percent was incarcerated. Around three-quarters had received less than 12 years’ education, and the average annual income for 52 percent of participants was less than $15,600 (median annual income in Australia at this time was $48,684). Just under half the participants were employed. Around 87 percent of participants reported brushing their teeth the previous day, whereas just under 73 percent reported last visiting a dentist more than a year ago. Around 71 percent of participants reported usually seeing a dentist because of a problem. Around 73 percent of participants reported having smoked 100 or more cigarettes in their lifetime, and 80 percent had low levels of physical activity. Experience of toothache was associated with usually visiting a dentist because of a problem (Table 2). Dissatisfaction with dental appearance was associated with psychological distress, being female, being incarcerated and problem-based dental attendance. Avoidance of food was associated with psychological distress, being female, and problem-based dental attendance. A higher frequency of poor self-rated oral health (as defined by the summary measure) was observed among those with high psychological distress, older participants, females, and those who usually attend for dental care because of a problem. Psychological distress and problembased dental attendance were also associated with poor selfrated oral health when the more stringent definition for the summary measure was used. A positive correlation was noted in relation to psychological distress and all poor self-rated measures of oral health (individual components, summary measure, and stringent summary measure), with increasing levels of poor self-rated oral health observed with increasing levels of psychological distress. In multivariable analysis, psychological distress remained significantly associated with poor self-rated oral health after controlling for sociodemographic factors significant at the P < 0.2 level in bivariate analysis (Table 3, Model 2). Although addition of significant dental behavior-related factors attenuated the odds slightly, psychological distress remained significantly associated with poor self-rated oral health (Table 3, Model 3). Other variables significant in the final model included being female and problem-based dental attendance. 4
N. Amarasena et al.
Three items remained significantly associated with poor self-rated oral health when measured in the more stringent way (presence of all three self-rated measures of oral health); high psychological distress (OR 5.17, 95 percent CI 1.85– 14.43), being female (OR 2.85, 95 percent CI 1.07–7.58), and usually visiting a dentist because of a problem (OR 10.83, 95 percent CI 4.06–28.87) (results not tabled).
Discussion In our convenience sample of Indigenous Australian adults, the level of poor self-rated oral health was high – 1.7 times the estimate reported for a national sample (6). Psychological distress was also high compared with national-level estimates (22) and remained significantly associated with poor selfrated oral health after adjusting for other covariates. We acknowledge that the low threshold of one or more self-rated measures of oral health resulted in high frequency (73.7 percent), although even when the more stringent definition of all three self-rated measures of oral health was employed, it was notable that psychological distress, being female, and usually visiting a dentist because of a problem were still associated with adverse impacts. It was concerning that poor self-rated oral health was reported so frequently in this group of Indigenous Australians, particularly when it has been argued that Australians in this generation have had, on the whole, the best opportunities for good oral health among those born in the 20th century (23). Clearly, Indigenous Australians have not been privy to the same social and economic benefits that result in favorable perceptions of oral health, an issue that rents at the very fabric of a society given evidence that oral health impairments impact on ability to enjoy social occasions, operate effectively in the home environment, and function in the workplace/ wider community (24). The frequency of periodontal disease and untreated dental caries in this group was also high (25). Psychological distress has been associated with other oral health outcomes. Wright et al. (26) reported that stressrelated behaviors such as jaw-clenching contributed to experience of subsequent toothache among patients with posttraumatic stress disorder. In 2005, Sanders and Spencer (27) reported that perceived stress was a risk indicator for poor self-rated oral health. In Spain, psychological distress has been associated with periodontal disease among adolescents (28). In an Ethiopian immigrant population in Israel, psychological distress was associated with both experience of dental caries and periodontal disease (29). To the best of our knowledge, this is the first time that psychological distress has been linked with a poor self-rated oral health in an Australian Aboriginal group. There has been increased attention paid to the high levels of psychological distress experienced by Australia’s Indigenous population in recent times. Although the Indigenous © 2014 American Association of Public Health Dentistry
31.9 (26.5-37.4) 33.7 (28.2-39.2)
© 2014 American Association of Public Health Dentistry 38.2 (28.0-48.4) 39.5 (33.0-46.1) 35.9 (24.1-47.8) 43.4 (32.6-54.2) 45.5 (34.2-56.7) 39.2 (30.8-47.7) 42.5 (33.0-51.9) 39.4 (22.6-56.2) 37.4 (31.1-43.8) 37.5 (30.6-44.4) 38.6 (27.1-50.0) 44.4 (37.2-51.6)* 22.4 (12.9-31.8) 38.6 (31.9-45.2) 39.1 (27.5-50.7) 33.3 (23.8-42.9) 36.0 (16.9-55.1)
76.0 (71.1-81.0) 24.0 (19.0-28.9) 51.8 (44.1-59.6) 48.2 (40.4-55.9) 51.8 (45.9-57.7) 48.2 (42.3-54.1) 13.1 (9.0-17.2) 86.9 (82.8-91.0) 72.9 (67.5-78.2) 27.1 (21.8-32.5) 70.7 (65.3-76.2) 29.3 (23.8-34.7) 72.9 (67.8-78.1) 27.1 (21.9-32.2) 79.7 (72.5-86.9) 20.3 (13.1-27.5)
38.7 (31.8-45.7)
31.1 (25.7-36.5) 68.9 (63.5-74.3)
38.1 (30.4-45.8)
39.2 (30.6-47.8)
45.7 (39.9-51.4) 54.3 (48.6-60.1)
42.3 (34.0-50.7) 35.0 (27.1-42.8)
49.8 (44.0-55.6) 50.2 (44.4-56.0)
53.7 (43.5-63.9) 48.0 (28.1-67.9)
57.7 (50.9-64.4) 44.9 (33.1-56.7)
59.7 (52.6-66.8)* 40.0 (28.8-51.2)
53.7 (46.5-60.8) 54.3 (42.5-66.0)
63.6 (47.1-80.2) 52.4 (45.9-59.0)
53.8 (45.2-62.5) 59.6 (50.1-69.1)
57.8 (47.1-68.6) 58.7 (47.4-69.9)
53.1 (46.3-59.8) 57.8 (45.6-70.0)
44.9 (34.5-55.3)
58.7 (51.7-65.8)*
47.7 (39.8-55.7)
62.6 (54.0-71.2)*
55.6 (47.1-64.0) 53.1 (44.9-61.4)
55.2 (44.7-65.7) 39.1 (29.1-49.2)
67.7 (58.3-77.1)*
54.3 (48.4-60.2)
Frequency dissatisfied appearance‡ (95% CI)
51.0 (40.9-61.2) 40.0 (20.5-59.5)
51.2 (44.4-58.0) 40.6 (28.9-52.2)
54.3 (47.1-61.5)* 30.3 (19.9-40.7)
46.6 (39.5-53.7) 50.0 (38.2-61.8)
63.6 (47.1-80.2) 45.6 (39.0-52.1)
48.5 (39.8-57.1) 54.3 (44.7-63.9)
48.2(37.3-59.1) 57.9 (46.7-69.1)
49.5 (42.8-56.3) 43.8 (31.5-56.0)
41.6 (31.3-51.9)
51.6 (44.4-58.7)
41.9 (34.1-49.8)
56.5 (47.7-65.2)*
51.5 (43.0-59.9) 45.5 (37.2-53.7)
48.9 (38.4-59.4) 35.9 (26.0-45.7)
59.4 (49.5-69.3)*
48.4 (42.5-54.3)
Frequency avoid eating food¶ (95% CI)
67.4 (57.9-76.8) 68.0 (49.5-86.5)
75.2 (69.4-81.1) 70.5 (60.3-80.7)
80.6 (75.0-86.3)* 57.0 (46.0-68.0)
71.4 (65.1-77.8) 79.5 (70.1-88.8)
80.0 (66.7-93.3) 72.8 (67.1-78.6)
71.5(68.0-82.7) 75.4 (64.1-78.9)
78.8 (70.1-87.6) 81.0 (72.3-89.8)
73.5 (67.6-79.4) 73.9 (63.5-84.3)
76.9 (71.0-82.8)
66.7 (56.9-76.5)
66.9 (59.5-74.3)
81.8 (75.2-88.4)*
79.2 (72.5-85.8)* 68.3 (60.7-75.9)
75.8 (67.0-84.7) 63.5 (53.9-73.2)
82.7 (75.1-90.2)*
73.7 (68.6-78.8)
Frequency poor self-rated oral health§ (95% CI)
41.5 (27.9-55.1) 42.9 (16.3-69.5)
49.5 (39.7-59.3) 35.3 (19.0-51.6)
58.6 (48.1-69.1)* 12.8 (2.2-23.5)
43.9 (33.9-53.8) 46.4 (27.7-65.2)
53.3(27.7-78.9) 57.8 (48.4-67.2)
43.5 (31.0-56.1) 57.7 (44.1-71.3)
52.6 (36.4-68.9) 60.5 (44.6-76.4)
46.2 (36.6-55.8) 43.8 (26.3-61.2)
50.0 (37.2-62.8)
55.0 (45.1-64.9)
49.0 (39.2-58.8)
60.3 (47.6-73.1)
60.0 (48.8-71.2) 47.1 (36.3-57.8)
46.3 (30.9-61.8) 25.5 (12.9-38.2)
64.6 (50.9-78.3)*
29.5 (24.5-35.0)
Frequency poor self-rated oral health (stringent)• (95% CI)
95% CI, 95% confidence interval.
* P < 0.05. † A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you have toothache?” ‡ A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth, or false teeth?” ¶ A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?” § Experienced toothache; felt uncomfortable about the appearance of teeth, mouth or false teeth; or avoided eating because of problems with teeth, mouth or false teeth “very often,” “fairly often,” or “occasionally.” • A response of “Very often,” “Often,” or “Occasionally” to the items “How often during the last year did you have toothache?,” “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth or false teeth?,” and “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?”
Nonincarcerated Education
Psychological distress and self-rated oral health among a convenience sample of Indigenous Australians Najith Amarasena, PhD1; Kostas Kapellas, PhD2; Alex Brown, PhD3; Michael R. Skilton, PhD4; Louise J. Maple-Brown, PhD5; Mark P. Bartold, PhD6; Kerin O’Dea, PhD7; David Celermajer, PhD4; Gary Douglas Slade, PhD2; Lisa Jamieson, PhD2 1 2 3 4 5 6 7
Australian Research Centre for Population Oral Health, School of Dentistry, The University of Adelaide, Adelaide, South Australia, Australia School of Dentistry, University of Adelaide, Adelaide, South Australia, Australia Indigenous Health Research, South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney, Sydney, New South Wales, Australia Menzies School of Health Research, Charles Darwin University, Darwin, Northern Territory, Australia Colgate Australian Clinical Dental Research Centre, University of Adelaide, Adelaide, South Australia, Australia Sansom Research Institute, University of South Australia, Adelaide, South Australia, Australia
Keywords Indigenous Australian; poor self-rated oral health; psychological distress. Correspondence Dr. Lisa Jamieson, School of Dentistry, University of Adelaide, 122 Frome St, Adelaide, SA 5005, Australia. Tel.: + 61-8-8303-4611; Fax: + 61-8-8303-4858; e-mail: [email protected]. Najith Amarasena is with the Australian Research Centre for Population Oral Health, School of Dentistry, The University of Adelaide. Kostas Kapellas is with the Dentistry, University of Adelaide. Alex Brown is with the Indigenous Health Research, South Australian Health and Medical Research Institute. Michael R. Skilton and David Celermajer are with the Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney. Louise J. Maple-Brown is with the Menzies School of Health Research, Charles Darwin University. Mark P. Bartold is with the Colgate Australian Clinical Dental Research Centre, University of Adelaide. Kerin O’Dea is with the Sansom Research Institute, University of South Australia. Gary Douglas Slade and Lisa Jamieson are with the School of Dentistry, University of Adelaide.
Abstract Objectives: This study sought to: a) estimate the frequency of poor self-rated oral health as assessed by a summary measure; b) compare frequency according to sociodemographic, behavioral, and psychological distress factors; and (3) determine if psychological distress was associated with poor self-rated oral health after adjusting for confounding. Methods: Data were from a convenience sample of Indigenous Australian adults (n = 289) residing in Australia’s Northern Territory. Poor self-rated oral health was defined as reported experience of toothache, poor dental appearance or food avoidance in the last 12 months. A logistic regression model was used to evaluate socio-demographic, behavioral, and psychological distress associations with poor self-rated oral health (SROH). Effects were quantified as odds ratios (OR). Results: The frequency of poor SROH was 73.7 percent. High psychological distress, measured by a Kessler-6 score ≥8, was experienced by 33.9 percent of participants. Poor SROH was associated with high levels of psychological distress, being older, being female, and usually visiting a dentist because of a problem. In the multivariable model, factors that were significantly associated with poor SROH after adjustment for other covariates included having a high level of psychological distress (OR 2.74, 95% CI 1.25–6.00), being female (OR 2.22, 95% CI 1.03–4.78), and usually visiting a dentist because of a problem (OR 3.57, 95% CI 1.89–6.76). Conclusions: Poor self-rated oral health and high levels of psychological distress were both highly frequent among this vulnerable population. Psychological distress was significantly associated with poor self-rated oral health after adjustment for confounding.
Received: 10/14/2013; accepted: 10/17/2014. doi: 10.1111/jphd.12080 Journal of Public Health Dentistry •• (2014) ••–••
Introduction Traditionally, measures in oral health epidemiology have been predominately clinical (1). However, these indices have © 2014 American Association of Public Health Dentistry
been criticized as inadequately reflecting the oral health status, concerns, and perceived needs of individuals (2). This criticism has led to the development of instruments measuring self-reported impacts of oral health, which enable greater 1
Distress and oral health in Indigenous Australians
insight into self-perceived oral health status and perceived need for oral health care (3). Although these instruments vary in their detail, all include assessments of three critical aspects of oral health: pain, appearance, and function. Locker et al. reported that oral health-related quality of life measures should incorporate “rate frequency, severity and importance; allow respondents to add supplemental items, and use global ratings which are summary variables and can reflect the differing values and preferences of a group of individuals” (4). The oral health impairment instrument used in this study does not meet these criteria, and thus is a reflection of selfrated oral health rather than oral health-related quality of life. The literature indicates that poor self-rated oral health may have substantial economic and social capital impacts (5). In general population settings, poor self-rated oral health has been associated with untreated dental decay, periodontal disease, difficulty paying a $100 dental bill, usually visiting a dentist because of a problem, cost barriers to receipt of dental care, and dental fear (6). In Indigenous Australian settings, it is important to consider other factors that are somewhat unique to this high-risk population. For example, levels of distress and poor psychological health are far higher among this group (7). It has been reported that the average Indigenous Australian family is confronted by seven major negative events every year: death, arrest, incarceration, hospitalization, alcohol abuse, extreme violence, and severe financial strains (8). Almost one in four Indigenous Australian families has even higher levels (9). By way of comparison, non-Indigenous Australian families typically experience less than one such event per year. Levels of physical activity are also reportedly lower (9), whereas levels of tobacco smoking are higher (10). These psychosocial and lifestyle factors have important associations with oral health outcomes (11-14). The aims of this analysis are: a) to estimate the frequency of poor self-rated oral health as assessed by a summary measure (including aspects of dental pain, dissatisfaction with dental appearance, and difficulty eating) among a convenience sample of Indigenous Australians; b) to compare frequency according to associations with sociodemographic, behavioral, and psychological distress factors; and c) to determine if psychological distress was associated with poor self-rated oral health after adjusting for confounding.
Methods Participants were Indigenous adults aged 22 years and over who were recruited from two regional jurisdictions in the Northern Territory and correctional centers in Alice Springs and Darwin, as part of the PerioCardio study described in detail elsewhere (15). Data used in this paper were thus crosssectional and from a convenience sample, collected between June 2010 and January 2012. The participation rate was 100 percent, and there were no incomplete responses (due to 2
N. Amarasena et al.
questionnaire data being collected via interview). The recruitment approach included liaising with Aboriginal community champions previously involved in our research projects, engaging with key community stakeholder groups, encouraging word-of-mouth spread of knowledge about the study, advertisements in local newspapers and radio shows, and presentations made to local Aboriginal community groups. A snowballing technique was also employed, with participants asked to contact any Aboriginal friends, family and peers who may be interested in partaking. Data were collected through a questionnaire, which included information on sociodemographic characteristics, oral health status including oral health behaviors and perceptions, general health behaviors, including tobacco smoking status and physical activity, and psychological distress. Ethical approval for the study was obtained from the Human Research Ethics Committee of the Northern Territory Department of Health and Menzies School of Health Research, the Central Australian Human Research Ethics Committee, Northern Territory Correctional Services Research Committee, University of Adelaide Human Research Ethics Committee, and the Aboriginal Health Council of South Australia.
Dependent variable The dependent variable was poor self-rated oral health, the definition of which has been described in detail elsewhere (6). A summary self-rated oral health variable was created by combining three items: experience of toothache, experience of discomfort due to mouth appearance, and food avoidance. Experience of toothache was assessed by asking “How often during the last year did you have toothache?,” whereas experience of discomfort due to mouth appearance was assessed by asking “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth or false teeth?” Avoiding food because of oral health problems was assessed by asking “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?” For purposes of this analysis, those who answered “Very often,” “Often,” or “Occasionally” to any of the items were considered to have poor self-rated oral health because of oral health-related factors. Other possible responses were “Hardly ever” or “Never.” These are standard cutoffs used for analyses involving oral health impairment-type statements such as the Oral Health Impact Profile (16).
Main exposure variable Psychological distress was measured by an adapted Kessler-6 (K6) instrument for use among Aboriginal Australians (17,18). The original K6 is widely used as a simple measure of © 2014 American Association of Public Health Dentistry
N. Amarasena et al.
Distress and oral health in Indigenous Australians
psychological distress (19) and consists of six questions about depressive and anxiety symptoms. It has been widely used in Aboriginal and Torres Strait Islander peoples in jurisdictional and national population-based surveys. Brown et al. (16,17) undertook extensive developmental adaptation and translation of the K6 and Patient Health Questionnaire-9 (PHQ-9) depression instruments in five language groups in Central Australia. In brief, the modification process involved detailed qualitative work to explore the expression of emotional distress within the target population, and repeated translation and back-translation of the original instruments into a combined modified English version (Table 1). Based on evidence in the literature (19), the summary scores were trichotomized into “low” (0–2), “moderate” (3–7), and “high” (8+). The K6 scores in our study were determined using the same methods as that by Kessler et al. (19). Study participants were asked the respective questions during a face-toface interview using previously validated methodology specific to Indigenous Australians (16,17). Values from 0–4 corresponding to responses “never” – “always” were summed to obtain an individual participant score (range 0–24). Individuals with a score of 8 or higher were classified as “stressed/depressed.”
Table 1 Original K6 Items Compared with Those Moderated for Use among Aboriginal Australians Original K6 items (17) During the past 4 weeks, how much of the time did you feel . . . . .* 1. . . . so sad nothing could cheer you up 2. . . . nervous 3. . . . restless or fidgety
4. . . . hopeless
5. . . . that everything was an effort 6. . . . worthless
Modified for use among Aboriginal Australians (6) During the last 12 months, have you . . . . .† 1. . . . felt depressed or no good, your spirit was really sad, that nothing can make you feel better? 2. . . . been feeling frightened, scared or nervous inside? 3. . . . been feeling that you can’t sit still or rest or that you keep moving around too much? 4. . . . been feeling that life is hopeless, that your worries never will go away, that nothing will get better? 5. . . . felt that you couldn’t do anything, that it was all too hard? 6. . . . been feeling no good about yourself, that you are not a good person, that you are rubbish or worthless?
* Response options: “all of the time,” “most of the time,” “some of the time,” “a little of the time,” “none of the time.” † Response options: “very often,” “fairly often,” “occasionally,” “hardly ever,” “never.”
© 2014 American Association of Public Health Dentistry
Covariates Covariates included sociodemographic factors (age, sex, education, income, and employment status) and behaviors (brushing teeth previous day, time since last dental visit, usual reason for seeing a dentist, tobacco smoking, and physical activity). Education was dichotomized into receiving “less than 12 years education” and “receiving 12 + years education,” income was dichotomized into “annual income (per person) < $15,600” and “annual income per person $15,600 +” and employment status was dichotomized into “currently employed” and “currently unemployed.” Tobacco smoking was calculated based on the response to “Have you smoked more than 100 cigarettes in your life time?,” with responses dichotomized into “yes” or “no.” The WHO’s Global PhysicalActivity Questionnaire Version 2 was used to assess physical activity, with analysis conducted as per recommendations in the analysis guide (20). Physical activity was dichotomized into “low” and “moderate/high.”
Analytic methods The number of participants who identified as being Aboriginal or Torres Strait Islander and who provided complete responses was 289 (a priori sample size calculation). All subsequent analyses pertained to these participants only. Blocks of explanatory variables that were significant at the P < 0.2 level in bivariate analysis were entered into a binary logistic regression model in three steps. The dependent variable of these models was poor self-rated oral health. Correlations between explanatory variables were weak to moderate, ranging from rho = −0.60 to +0.29. Psychological distress was entered in Model 1, with the main effect presented as an odds ratio (OR) with its 95 percent confidence interval (95 percent CI). The sociodemographic factors were entered into Model 2, and behavioral factors entered into Model 3. Data were analyzed using SAS version 9.3 (SAS, Cary, NC, USA). An additional validation step was undertaken using an alternative definition to define poor self-rated oral health as a positive report of all three impacts: toothache, dissatisfaction with appearance, and food avoidance. This more stringent case definition necessarily created a lower frequency of impairment; the main objective in its use was to determine if similar associations emerged from a similar multivariable analysis.
Results Approximately 40 percent of participants experienced toothache in the past year, whereas just under 55 percent experienced dissatisfaction with their dental appearance and 48 percent had avoided food because of problems with their 3
Distress and oral health in Indigenous Australians
teeth, mouth, or false teeth (Table 2). Almost three-quarters had reported one or more poor self-rated measures of oral health in the previous year, with just under one-third having reported all three poor self-rated measures of oral health. The mean K6 score was 6.0, with a standard deviation of 5.3. Scores ranged from 0 to 22 (possible range 0–24). Less than 12 percent of participants scored 13+, which is the recommended cutoff point for serious psychological distress (21). Participants were evenly distributed across the psychological distress categories, with approximately one-third in each. The age range was 22–73, with a mean age of 39.6 years (standard deviation = 10.3 years). Around 54 percent of participants were male, and approximately 31 percent was incarcerated. Around three-quarters had received less than 12 years’ education, and the average annual income for 52 percent of participants was less than $15,600 (median annual income in Australia at this time was $48,684). Just under half the participants were employed. Around 87 percent of participants reported brushing their teeth the previous day, whereas just under 73 percent reported last visiting a dentist more than a year ago. Around 71 percent of participants reported usually seeing a dentist because of a problem. Around 73 percent of participants reported having smoked 100 or more cigarettes in their lifetime, and 80 percent had low levels of physical activity. Experience of toothache was associated with usually visiting a dentist because of a problem (Table 2). Dissatisfaction with dental appearance was associated with psychological distress, being female, being incarcerated and problem-based dental attendance. Avoidance of food was associated with psychological distress, being female, and problem-based dental attendance. A higher frequency of poor self-rated oral health (as defined by the summary measure) was observed among those with high psychological distress, older participants, females, and those who usually attend for dental care because of a problem. Psychological distress and problembased dental attendance were also associated with poor selfrated oral health when the more stringent definition for the summary measure was used. A positive correlation was noted in relation to psychological distress and all poor self-rated measures of oral health (individual components, summary measure, and stringent summary measure), with increasing levels of poor self-rated oral health observed with increasing levels of psychological distress. In multivariable analysis, psychological distress remained significantly associated with poor self-rated oral health after controlling for sociodemographic factors significant at the P < 0.2 level in bivariate analysis (Table 3, Model 2). Although addition of significant dental behavior-related factors attenuated the odds slightly, psychological distress remained significantly associated with poor self-rated oral health (Table 3, Model 3). Other variables significant in the final model included being female and problem-based dental attendance. 4
N. Amarasena et al.
Three items remained significantly associated with poor self-rated oral health when measured in the more stringent way (presence of all three self-rated measures of oral health); high psychological distress (OR 5.17, 95 percent CI 1.85– 14.43), being female (OR 2.85, 95 percent CI 1.07–7.58), and usually visiting a dentist because of a problem (OR 10.83, 95 percent CI 4.06–28.87) (results not tabled).
Discussion In our convenience sample of Indigenous Australian adults, the level of poor self-rated oral health was high – 1.7 times the estimate reported for a national sample (6). Psychological distress was also high compared with national-level estimates (22) and remained significantly associated with poor selfrated oral health after adjusting for other covariates. We acknowledge that the low threshold of one or more self-rated measures of oral health resulted in high frequency (73.7 percent), although even when the more stringent definition of all three self-rated measures of oral health was employed, it was notable that psychological distress, being female, and usually visiting a dentist because of a problem were still associated with adverse impacts. It was concerning that poor self-rated oral health was reported so frequently in this group of Indigenous Australians, particularly when it has been argued that Australians in this generation have had, on the whole, the best opportunities for good oral health among those born in the 20th century (23). Clearly, Indigenous Australians have not been privy to the same social and economic benefits that result in favorable perceptions of oral health, an issue that rents at the very fabric of a society given evidence that oral health impairments impact on ability to enjoy social occasions, operate effectively in the home environment, and function in the workplace/ wider community (24). The frequency of periodontal disease and untreated dental caries in this group was also high (25). Psychological distress has been associated with other oral health outcomes. Wright et al. (26) reported that stressrelated behaviors such as jaw-clenching contributed to experience of subsequent toothache among patients with posttraumatic stress disorder. In 2005, Sanders and Spencer (27) reported that perceived stress was a risk indicator for poor self-rated oral health. In Spain, psychological distress has been associated with periodontal disease among adolescents (28). In an Ethiopian immigrant population in Israel, psychological distress was associated with both experience of dental caries and periodontal disease (29). To the best of our knowledge, this is the first time that psychological distress has been linked with a poor self-rated oral health in an Australian Aboriginal group. There has been increased attention paid to the high levels of psychological distress experienced by Australia’s Indigenous population in recent times. Although the Indigenous © 2014 American Association of Public Health Dentistry
31.9 (26.5-37.4) 33.7 (28.2-39.2)
© 2014 American Association of Public Health Dentistry 38.2 (28.0-48.4) 39.5 (33.0-46.1) 35.9 (24.1-47.8) 43.4 (32.6-54.2) 45.5 (34.2-56.7) 39.2 (30.8-47.7) 42.5 (33.0-51.9) 39.4 (22.6-56.2) 37.4 (31.1-43.8) 37.5 (30.6-44.4) 38.6 (27.1-50.0) 44.4 (37.2-51.6)* 22.4 (12.9-31.8) 38.6 (31.9-45.2) 39.1 (27.5-50.7) 33.3 (23.8-42.9) 36.0 (16.9-55.1)
76.0 (71.1-81.0) 24.0 (19.0-28.9) 51.8 (44.1-59.6) 48.2 (40.4-55.9) 51.8 (45.9-57.7) 48.2 (42.3-54.1) 13.1 (9.0-17.2) 86.9 (82.8-91.0) 72.9 (67.5-78.2) 27.1 (21.8-32.5) 70.7 (65.3-76.2) 29.3 (23.8-34.7) 72.9 (67.8-78.1) 27.1 (21.9-32.2) 79.7 (72.5-86.9) 20.3 (13.1-27.5)
38.7 (31.8-45.7)
31.1 (25.7-36.5) 68.9 (63.5-74.3)
38.1 (30.4-45.8)
39.2 (30.6-47.8)
45.7 (39.9-51.4) 54.3 (48.6-60.1)
42.3 (34.0-50.7) 35.0 (27.1-42.8)
49.8 (44.0-55.6) 50.2 (44.4-56.0)
53.7 (43.5-63.9) 48.0 (28.1-67.9)
57.7 (50.9-64.4) 44.9 (33.1-56.7)
59.7 (52.6-66.8)* 40.0 (28.8-51.2)
53.7 (46.5-60.8) 54.3 (42.5-66.0)
63.6 (47.1-80.2) 52.4 (45.9-59.0)
53.8 (45.2-62.5) 59.6 (50.1-69.1)
57.8 (47.1-68.6) 58.7 (47.4-69.9)
53.1 (46.3-59.8) 57.8 (45.6-70.0)
44.9 (34.5-55.3)
58.7 (51.7-65.8)*
47.7 (39.8-55.7)
62.6 (54.0-71.2)*
55.6 (47.1-64.0) 53.1 (44.9-61.4)
55.2 (44.7-65.7) 39.1 (29.1-49.2)
67.7 (58.3-77.1)*
54.3 (48.4-60.2)
Frequency dissatisfied appearance‡ (95% CI)
51.0 (40.9-61.2) 40.0 (20.5-59.5)
51.2 (44.4-58.0) 40.6 (28.9-52.2)
54.3 (47.1-61.5)* 30.3 (19.9-40.7)
46.6 (39.5-53.7) 50.0 (38.2-61.8)
63.6 (47.1-80.2) 45.6 (39.0-52.1)
48.5 (39.8-57.1) 54.3 (44.7-63.9)
48.2(37.3-59.1) 57.9 (46.7-69.1)
49.5 (42.8-56.3) 43.8 (31.5-56.0)
41.6 (31.3-51.9)
51.6 (44.4-58.7)
41.9 (34.1-49.8)
56.5 (47.7-65.2)*
51.5 (43.0-59.9) 45.5 (37.2-53.7)
48.9 (38.4-59.4) 35.9 (26.0-45.7)
59.4 (49.5-69.3)*
48.4 (42.5-54.3)
Frequency avoid eating food¶ (95% CI)
67.4 (57.9-76.8) 68.0 (49.5-86.5)
75.2 (69.4-81.1) 70.5 (60.3-80.7)
80.6 (75.0-86.3)* 57.0 (46.0-68.0)
71.4 (65.1-77.8) 79.5 (70.1-88.8)
80.0 (66.7-93.3) 72.8 (67.1-78.6)
71.5(68.0-82.7) 75.4 (64.1-78.9)
78.8 (70.1-87.6) 81.0 (72.3-89.8)
73.5 (67.6-79.4) 73.9 (63.5-84.3)
76.9 (71.0-82.8)
66.7 (56.9-76.5)
66.9 (59.5-74.3)
81.8 (75.2-88.4)*
79.2 (72.5-85.8)* 68.3 (60.7-75.9)
75.8 (67.0-84.7) 63.5 (53.9-73.2)
82.7 (75.1-90.2)*
73.7 (68.6-78.8)
Frequency poor self-rated oral health§ (95% CI)
41.5 (27.9-55.1) 42.9 (16.3-69.5)
49.5 (39.7-59.3) 35.3 (19.0-51.6)
58.6 (48.1-69.1)* 12.8 (2.2-23.5)
43.9 (33.9-53.8) 46.4 (27.7-65.2)
53.3(27.7-78.9) 57.8 (48.4-67.2)
43.5 (31.0-56.1) 57.7 (44.1-71.3)
52.6 (36.4-68.9) 60.5 (44.6-76.4)
46.2 (36.6-55.8) 43.8 (26.3-61.2)
50.0 (37.2-62.8)
55.0 (45.1-64.9)
49.0 (39.2-58.8)
60.3 (47.6-73.1)
60.0 (48.8-71.2) 47.1 (36.3-57.8)
46.3 (30.9-61.8) 25.5 (12.9-38.2)
64.6 (50.9-78.3)*
29.5 (24.5-35.0)
Frequency poor self-rated oral health (stringent)• (95% CI)
95% CI, 95% confidence interval.
* P < 0.05. † A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you have toothache?” ‡ A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth, or false teeth?” ¶ A response of “Very often,” “Often,” or “Occasionally” to the item “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?” § Experienced toothache; felt uncomfortable about the appearance of teeth, mouth or false teeth; or avoided eating because of problems with teeth, mouth or false teeth “very often,” “fairly often,” or “occasionally.” • A response of “Very often,” “Often,” or “Occasionally” to the items “How often during the last year did you have toothache?,” “How often during the last year did you feel uncomfortable about the appearance of your teeth, mouth or false teeth?,” and “How often during the last year did you have to avoid eating some foods because of problems with your teeth, mouth or false teeth?”
Nonincarcerated Education
