Impact of Individual and Combined Sensory Impairment in Older Australians
Access status:
Open Access
Type
ThesisThesis type
Doctor of PhilosophyAuthor/s
Karpa, Michael JamesAbstract
Purpose: To estimate the prevalence and examine the clustering patterns of visual, auditory and olfactory impairments; to estimate the associations of olfactory impairment with neurodegenerative and other morbidities; to estimate the associations of visual and auditory impairments ...
See morePurpose: To estimate the prevalence and examine the clustering patterns of visual, auditory and olfactory impairments; to estimate the associations of olfactory impairment with neurodegenerative and other morbidities; to estimate the associations of visual and auditory impairments with morbidity and mortality using Cox regression; and to examine the associations of visual and auditory impairments with morbidity and mortality using structural equation modelling to identify potential indirect pathways and assess whether Cox regression underestimated the associations between VI, AI and mortality in a representative sample of older Australians. Methods: The Blue Mountains Eye Study (BMES) examined 3,654 persons aged 49+ during 1992-1994, and after 5 and 10 years. The Blue Mountains Hearing Study (BMHS) invited participants who attended the second cross-sectional survey of the Blue Mountains Eye Study (BMES 2). Persons who moved into the study area or study age group, identified from a repeat door-to door census in 1999, were also invited to participate. The Blue Mountains Hearing Study (BMHS) examined 2956 persons aged 49+ years (75.5% response) during 1997-2000. Vision, hearing and olfaction were assessed in BMES 3. Assessment was by interviewer administered structured questionnaire, clinical examination, audiometry, blood testing and the San Diego Odor Identification Test. A total of 1,497 (74.3% of all participants) had complete vision, auditory and olfactory data after BMES 3. Visual impairment (VI) was categorized as either: presenting visual impairment (PVI), VA less than 6/12 Snellen equivalent (<39 letters read correctly) in the better eye using current glasses; or correctable visual impairment (CVI), PVI less than 6/12 Snellen equivalent correctable to 6/12 or better after subjective refraction; or non-correctable visual impairment (NCVI), PVI correctable to less than 6/12 Snellen equivalent in the better eye, after subjective refraction. Olfactory impairment (OI) was defined by San Diego Odour Identification Test score with subjects classified as having no impairment (score 6, 7 or 8), mild impairment (4 or 5), moderate impairment (≤3), or any impairment (<6). Auditory impairment (AI) was defined as the pure-tone average (0.5-4kHz) of air-conduction hearing thresholds >25 decibels hearing level (dBHL). Cognitive impairment was defined as mini mental state exam (MMSE) scores <24. Log-linear models were used to assess the concomitant presence of the three sensory impairments (visual, auditory and olfactory). Observed frequencies of concomitant sensory impairments were compared to the expected frequencies estimated assuming they occurred independently (no clustering tendency). Multivariable adjusted logistic regression models were constructed to estimate associations between olfactory impairment and morbidities, including neurodegenerative conditions. Associations between visual impairment and mortality risk, and between hearing loss and mortality risk, were estimated using Cox regression and structural equation modelling (SEM). Odds ratios (OR), hazard ratios (HR) and 95% confidence intervals (CI) are presented. A p-value of less than 0.05 was considered statistically significant. Australian National Death Index data confirmed deaths until 2005. Results: After13 years from baseline, 1273 participants had died. After 5 years from BMES 2 (BMHS), 403 participants had died. At BMES 3, the prevalence of PVI, CVI and NCVI was 11%, 8% and 3% respectively. The prevalence of any OI was 27.0% and the prevalence of AI was 43%. The observed prevalence of having all three sensory impairments in persons with PVI (or NCVI) was 2.6 (or 3.0) times greater than predicted if they clustered independently. VI, AI and OI clustered differently in women compared to men. Inverse associations were observed between OI and body mass index (OR per 5 kg/m2 increase, 0.8, CI 0.7-0.9) and between moderate impairment and hypertension (OR 0.6, CI 0.4-0.9). There was no significant relationship with angina, previous myocardial infarction or diabetes. Persons with Parkinson disease had an increased likelihood of both mild (OR 9.8, CI 2.0-47.5) and moderate OI (OR 16.1, CI 3.8-68.2), as did persons with impaired cognitive function (OR 3.3, CI 1.3-8.6 and OR 3.7, CI 1.5-9.6, respectively). After adjusting for mortality risk markers using Cox regression, higher mortality was associated with NCVI (HR 1.35, CI 1.04-1.75). This association was stronger for ages <75 years (HR 2.58, CI 1.42-4.69). Structural equation modelling revealed greater effects of NCVI on mortality risk (HR 5.25, CI 1.97-14.01 for baseline ages <75), with both direct (HR 2.16, CI 1.11-4.23) and indirect effects (HR 2.43, CI 1.17-5.03). Of the mortality risk markers examined, only disability in walking demonstrated a significant indirect pathway for the link between VI and mortality. Disability in walking acted both directly on mortality and via an association with self-rated health. Using Cox regression, hearing loss was associated with increased risk of both cardiovascular (HR 1.36, CI 1.08-1.84) and all-cause mortality (HR 1.39, CI 1.11-1.79) after adjustment for age and sex, but not after multivariable adjustment. Structural equation modelling pathway analysis, however, revealed a higher all-cause mortality risk (HR 2.58, CI 1.64-4.05) in persons with hearing loss, which was mediated by two variables: cognitive impairment (HR 1.45, CI 1.08-1.94) and disability in walking (HR 1.63, CI 1.24-2.15). These variables increased mortality both directly and indirectly through effects on self-rated health. Conclusions: In this representative population of older Australians, over one in ten persons had VI, over one in four persons had OI and almost one in two persons had AI. The prevalence of VI, AI and OI increased with increasing age. The prevalence of AI and OI was higher in males. The prevalence of VI was higher in females. Visual, auditory and olfactory impairments aggregated mutually and dependently. Visual impairment and AI were significantly associated with morbidity and mortality. Visual impairment predicted mortality by both direct and indirect pathways. Auditory impairment predicted mortality via indirect pathways. Disability in walking, which can substantially influence general health, represented a major indirect pathway for both VI and AI. Auditory impairment was also associated with increased all-cause mortality via cognitive impairment and self-rated health. Adjustment for these co-variables using Cox regression underestimated the associations between VI and AI and mortality. Olfactory impairment was inversely associated with BMI and hypertension. Olfactory impairment was significantly higher among persons with Parkinson disease and cognitive impairment. It is important to recognise that persons with sensory impairments are at increased risk of important comorbidities and mortality. Dependent clustering of sensory impairments suggest the possibility of a common underlying mechanism and that separate hearing and vision services may not adequately support older persons.
See less
See morePurpose: To estimate the prevalence and examine the clustering patterns of visual, auditory and olfactory impairments; to estimate the associations of olfactory impairment with neurodegenerative and other morbidities; to estimate the associations of visual and auditory impairments with morbidity and mortality using Cox regression; and to examine the associations of visual and auditory impairments with morbidity and mortality using structural equation modelling to identify potential indirect pathways and assess whether Cox regression underestimated the associations between VI, AI and mortality in a representative sample of older Australians. Methods: The Blue Mountains Eye Study (BMES) examined 3,654 persons aged 49+ during 1992-1994, and after 5 and 10 years. The Blue Mountains Hearing Study (BMHS) invited participants who attended the second cross-sectional survey of the Blue Mountains Eye Study (BMES 2). Persons who moved into the study area or study age group, identified from a repeat door-to door census in 1999, were also invited to participate. The Blue Mountains Hearing Study (BMHS) examined 2956 persons aged 49+ years (75.5% response) during 1997-2000. Vision, hearing and olfaction were assessed in BMES 3. Assessment was by interviewer administered structured questionnaire, clinical examination, audiometry, blood testing and the San Diego Odor Identification Test. A total of 1,497 (74.3% of all participants) had complete vision, auditory and olfactory data after BMES 3. Visual impairment (VI) was categorized as either: presenting visual impairment (PVI), VA less than 6/12 Snellen equivalent (<39 letters read correctly) in the better eye using current glasses; or correctable visual impairment (CVI), PVI less than 6/12 Snellen equivalent correctable to 6/12 or better after subjective refraction; or non-correctable visual impairment (NCVI), PVI correctable to less than 6/12 Snellen equivalent in the better eye, after subjective refraction. Olfactory impairment (OI) was defined by San Diego Odour Identification Test score with subjects classified as having no impairment (score 6, 7 or 8), mild impairment (4 or 5), moderate impairment (≤3), or any impairment (<6). Auditory impairment (AI) was defined as the pure-tone average (0.5-4kHz) of air-conduction hearing thresholds >25 decibels hearing level (dBHL). Cognitive impairment was defined as mini mental state exam (MMSE) scores <24. Log-linear models were used to assess the concomitant presence of the three sensory impairments (visual, auditory and olfactory). Observed frequencies of concomitant sensory impairments were compared to the expected frequencies estimated assuming they occurred independently (no clustering tendency). Multivariable adjusted logistic regression models were constructed to estimate associations between olfactory impairment and morbidities, including neurodegenerative conditions. Associations between visual impairment and mortality risk, and between hearing loss and mortality risk, were estimated using Cox regression and structural equation modelling (SEM). Odds ratios (OR), hazard ratios (HR) and 95% confidence intervals (CI) are presented. A p-value of less than 0.05 was considered statistically significant. Australian National Death Index data confirmed deaths until 2005. Results: After13 years from baseline, 1273 participants had died. After 5 years from BMES 2 (BMHS), 403 participants had died. At BMES 3, the prevalence of PVI, CVI and NCVI was 11%, 8% and 3% respectively. The prevalence of any OI was 27.0% and the prevalence of AI was 43%. The observed prevalence of having all three sensory impairments in persons with PVI (or NCVI) was 2.6 (or 3.0) times greater than predicted if they clustered independently. VI, AI and OI clustered differently in women compared to men. Inverse associations were observed between OI and body mass index (OR per 5 kg/m2 increase, 0.8, CI 0.7-0.9) and between moderate impairment and hypertension (OR 0.6, CI 0.4-0.9). There was no significant relationship with angina, previous myocardial infarction or diabetes. Persons with Parkinson disease had an increased likelihood of both mild (OR 9.8, CI 2.0-47.5) and moderate OI (OR 16.1, CI 3.8-68.2), as did persons with impaired cognitive function (OR 3.3, CI 1.3-8.6 and OR 3.7, CI 1.5-9.6, respectively). After adjusting for mortality risk markers using Cox regression, higher mortality was associated with NCVI (HR 1.35, CI 1.04-1.75). This association was stronger for ages <75 years (HR 2.58, CI 1.42-4.69). Structural equation modelling revealed greater effects of NCVI on mortality risk (HR 5.25, CI 1.97-14.01 for baseline ages <75), with both direct (HR 2.16, CI 1.11-4.23) and indirect effects (HR 2.43, CI 1.17-5.03). Of the mortality risk markers examined, only disability in walking demonstrated a significant indirect pathway for the link between VI and mortality. Disability in walking acted both directly on mortality and via an association with self-rated health. Using Cox regression, hearing loss was associated with increased risk of both cardiovascular (HR 1.36, CI 1.08-1.84) and all-cause mortality (HR 1.39, CI 1.11-1.79) after adjustment for age and sex, but not after multivariable adjustment. Structural equation modelling pathway analysis, however, revealed a higher all-cause mortality risk (HR 2.58, CI 1.64-4.05) in persons with hearing loss, which was mediated by two variables: cognitive impairment (HR 1.45, CI 1.08-1.94) and disability in walking (HR 1.63, CI 1.24-2.15). These variables increased mortality both directly and indirectly through effects on self-rated health. Conclusions: In this representative population of older Australians, over one in ten persons had VI, over one in four persons had OI and almost one in two persons had AI. The prevalence of VI, AI and OI increased with increasing age. The prevalence of AI and OI was higher in males. The prevalence of VI was higher in females. Visual, auditory and olfactory impairments aggregated mutually and dependently. Visual impairment and AI were significantly associated with morbidity and mortality. Visual impairment predicted mortality by both direct and indirect pathways. Auditory impairment predicted mortality via indirect pathways. Disability in walking, which can substantially influence general health, represented a major indirect pathway for both VI and AI. Auditory impairment was also associated with increased all-cause mortality via cognitive impairment and self-rated health. Adjustment for these co-variables using Cox regression underestimated the associations between VI and AI and mortality. Olfactory impairment was inversely associated with BMI and hypertension. Olfactory impairment was significantly higher among persons with Parkinson disease and cognitive impairment. It is important to recognise that persons with sensory impairments are at increased risk of important comorbidities and mortality. Dependent clustering of sensory impairments suggest the possibility of a common underlying mechanism and that separate hearing and vision services may not adequately support older persons.
See less
Date
2015-09-30Licence
The author retains copyright of this thesis. It may only be used for the purposes of research and study. It must not be used for any other purposes and may not be transmitted or shared with others without prior permission.Faculty/School
Sydney Medical SchoolAwarding institution
The University of SydneyShare