Abstract
Older adults are prone to have multiple chronic health conditions. The purpose of this study was to investigate whether older adults with age-related macular degeneration (AMD) who experience Charles Bonnet syndrome (CBS) are at a higher risk of developing depression and mild cognitive impairment (MCI). A total of 42 participants (31 females, 11 males; age: 68–99 years, M = 85.5 years; visual acuity [VA] in the better eye ranging from 20/70 to 20/1200) diagnosed with AMD were recruited in a vision rehabilitation center. They completed the Montreal Cognitive Assessment (MoCA) in its blind version, the Geriatric Depression Scale, and responded to questions designed to determine whether they experienced visual hallucinations consistent with CBS. Participants were then categorized into whether or not they experienced CBS, were at risk of depression or were at risk of MCI. Participants in the group experiencing CBS did not statistically show a higher likelihood of developing depression and/or MCI than those without CBS. Overall, the risk of depression (30%) was consistent with previous studies. In our sample of 42 older adults with visual impairment, 62% failed the MoCA suggesting they were at risk of cognitive decline. Our study was not able to replicate previous reports of a possible relationship between CBS and MCI in older adults with AMD; however, the observed level of possible cognitive impairment warrants further investigation. Future studies should include participants with other ocular pathologies to investigate whether a relationship among CBS, MCI, and/or depression may exist independently of AMD.
Older adults diagnosed with age-related macular degeneration (AMD) often experience visual hallucinations, known as Charles Bonnet syndrome (CBS) (Scott, Schein, Feuer, & Folstein, 2001; Teunisse, Cruysberg, Verbeek, & Zitman, 1995). Depression is also a comorbidity of AMD (Rovner & Casten, 2002; Rovner, Casten, Hegel, Leiby, & Tasman, 2007) and is often a precursor to mild cognitive impairment (MCI) (Jorm, 2000; Paterniti, Verdier-Taillefer, Dufouil, & Alpérovitch, 2002; Rovner, Casten, & Leiby, 2009; Sachs-Ericsson, Joiner, Plant, & Blazer, 2005). Since depression appears to be a common comorbidity in both MCI and AMD, a relationship among depression, MCI, and CBS can be hypothesized. This study examined whether the probability of individuals with AMD experiencing CBS is higher in those showing signs of depression and lower cognitive function when compared to individuals with AMD who do not experience CBS.
AMD
According to Statistics Canada (2007), by 2026, older adults will make up 21% of the population compared to 13% in 2000. One of the most commonly diagnosed forms of vision loss in the elderly is AMD, which is characterized by progressive destruction of the central retina or macula (Stuen & Faye, 2003). Numerous studies have reported that AMD is the leading cause of eye disease in persons 65 years of age and older (Casten, Rovner, & Edmonds, 2002; Rovner & Casten, 2002; Tolman, Hill, Kleinschmidt, & Gregg, 2005; Watson, 2001) causing a loss of visual acuity (VA) which affects the perception of detail, color, and contrast sensitivity (Stuen & Faye, 2003; Watson, 2001). These losses have an impact on the ability to read and drive. Also affected is the ability to recognize faces and to perceive depth which can lead to issues in safety and mobility (e.g. decreased safety in going to the grocery store), problems of isolation (e.g. decreased interest in participating in social activities), and decreased ability to complete household and/or leisure tasks (e.g. cooking, cleaning, sewing, and knitting). Many of those diagnosed with AMD also report experiencing visual hallucinations called CBS.
CBS
The definition currently accepted by most professionals in diagnosing CBS is the presence of complex, persistent, or repetitive visual hallucinations without hallucinations in other sensory modalities. These individuals do not have psychiatric or sleep disorders, they do not abuse substances, and they have been diagnosed with an acquired visual loss (Cole, 1992; Eperjesi & Akbarali, 2004; Menon, 2005). The distinction from hallucinations of a psychiatric nature is that individuals with CBS are fully aware that their hallucinations are not real. Although there appears to be no consensus among investigators as to the role of visual impairment in the development of CBS, there is support that having a vision impairment is necessary for its diagnosis (Menon, 2005).
Approximately 14%–20% of all individuals with a vision impairment will develop CBS (Crumbliss, Taussig, & Jay, 2009; Fitzgerald, 1971; Lepore, 1990; Rovner, 2002), with AMD appearing to be the ocular pathology in which this occurs most frequently. Furthermore, CBS is most often associated with higher degrees of vision loss, in particular bilaterally (Gilmour, Schreiber, & Ewing, 2009; Menon, 2005; Teunisse et al., 1995); however, improvements in vision sometimes result in a decrease of hallucinations (Menon, 2005).
CBS is not widely recognized or reported as clients hesitate disclosing their experiences because they are concerned that others will think they are mentally ill (Cohen, Bulik, Tadavoni, & Quentel, 2003; Fernandez, Lichtshein, Vieweg, & Victor, 1997; Menon, 2005). This makes it important to increase awareness of CBS in order for health-care professionals to correctly diagnose and reassure individuals, facilitating disclosure without fear (Tan & Au Eong, 2004). A correct diagnosis is critical in order to be referred (Eperjesi & Akbarali, 2004; Gilmour et al., 2009) to vision rehabilitation experts where the occurrence of CBS is more widely recognized, rather than to mental health professionals (O’Farrell, Lewis, McKenzie, & Jones, 2010). When learning about CBS, individuals get a sense of relief in knowing that their hallucinations are not psychiatric in nature, and they are spared the stigmatizing experience of having to undergo psychological assessment and counseling (Abbott, Connor, Artes, & Abadi, 2007).
Depression associated with aging
Depression is one of the most common mental health conditions in older adults. Physicians may have difficulty in diagnosing depression in the elderly, and this may result in the condition not being treated. But when depression is recognized, older persons are three times more likely to receive adequate treatment (Crawford, Prince, Menezes, & Mann, 1998). Those who show signs of even a mild form of depression are more likely to suffer from a major depression at a later date, potentially causing a higher frequency of functional disability (Horowitz, Reinhardt, & Boerner, 2005). It is also important to realize that older adults may experience signs of depression differently than their younger counterparts, suggesting that their depression can result from feelings of decreased productivity, reduced social contacts, and financial resources (O’Donnell, 2005).
Depression associated with age-related vision loss
There are numerous studies linking age-related vision loss to the onset of depression (Horowitz, 2003; Lee, Beaver, Jogerst, & Daly, 2003; Lupsakko, Mantyjarvi, Sulkava, & Kautiainen, 2002; O’Donnell, 2005; Raasch, Leat, Kleinstein, Bullimore, & Cutter, 1997). Prevalence tends to generally hover around 30% (Overbury & Wittich, 2011). Impaired vision is a key risk factor for the onset of depressive symptoms in older adults because after living with intact vision for most of their lives, an impairment of this type can have serious emotional implications. Moreover, when depression coincides with vision loss, it can persist for a much longer period of time (Horowitz, 2003). Feelings of denial, anger, grief, and depression occur as individuals try to adjust and accept that their vision cannot be restored (Jack & McGalliard, 1999). Depression can occur as the ability to conduct daily activities, particularly reading and driving, decreases, leaving individuals feeling that they are unable to be part of their community or society. In particular, losing the ability to drive can decrease one’s sense of self-worth and can represent the beginning of the end of independence (Horowitz, 2003).
MCI
MCI is a term describing individuals who show a noticeable decline in cognitive function, for example, short-term memory loss (Chertkow et al., 2008; Gauthier et al., 2006; Petersen et al., 1999; Steffens et al., 2006). According to the Canadian Study of Health and Aging Working Group (1994), dementia has been estimated to be prevalent in 8% of the older adult population, whereby 16% of cognitive impairment may occur without dementia in older adults. The effect of MCI in daily living is that it can impair one’s ability to function independently; however, whether it is a precursor to Alzheimer’s disease (AD) is currently a topic of debate in the literature (Chertkow, 2002; Morris et al., 2001).
MCI and AMD
Several authors have explored the relation between cognitive loss and AMD (Clemons, Rankin, McBee, & Age-Related Eye Disease Study Research Group, 2006; Klaver et al., 1999; Pham, Kifley, Mitchell, & Wang, 2006). Such a relationship is plausible since both disorders are commonly related to aging and degenerating neurological functioning (Clemons et al., 2006). One of the proposed links is Amyloid-beta, a string of peptides found in both the drusen in AMD patients and the brain plaques in AD patients (Johnson et al., 2002). An analysis of cognitive functioning in those with AD and their level of vision loss revealed that the loss of cells in the nervous system is the distinguishing feature of both diseases (Wong et al., 2002). Furthermore, this study found that individuals with lowered cognitive functioning were more likely to have lowered VA as compared to those without cognitive impairment. Pham et al. (2006) agreed with these results, reporting a significant correlation between late AMD and cognitive impairment, demonstrating that 18% of older adults with AMD had cognitive impairment as compared to 2.6% of individuals without AMD.
MCI and depression
Several studies have shown that depression can occur with cognitive decline (Ganguli, Du, Dodge, Ratcliff, & Chang, 2006; Wilson, Mendes De Leon, Bennett, Bienias, & Evans, 2004), and that the decline in cognitive function is more rapid when depression is present. In one investigation, cognitive decline occurred at a rate of 20% greater in individuals presenting with depressive symptoms as compared to those without (Wilson et al., 2004). However, little is known about the association between depression and cognitive loss since recruitment for studies on depression typically excludes those with cognitive impairment. Steffens et al. (2006) classified individuals into four categories of cognitive functioning: those without any loss of cognition, those with cognitive impairment, those without any form of dementia, and those with a diagnosed form of dementia. Although this study placed only 16% of older adults in the category of cognitive impairment without dementia, the same sample also showed a 50% increased risk of developing dementia subsequently.
MCI and CBS
Pliskin et al. (1996) conducted a study to test the neuropsychological function of participants exhibiting symptoms of CBS to determine whether a relationship exists between MCI and CBS. They hypothesized that individuals with CBS actually had an underlying and undiagnosed pathology within the central nervous system that could be indicative of early onset dementia. Their results supported their theory, as the likelihood of showing signs of dementia was higher in persons experiencing CBS (associated with a visual disease), suggesting a potential connection between early onset dementia and a diagnosis of vision loss. However, no data are currently available as to whether this relationship generalizes to persons experiencing depression and/or AMD.
Hypotheses
Given the possible relationship among depression, MCI, and the experience of CBS in older adults with AMD, our study examined three hypotheses. The first proposed that there would be a larger proportion of people at risk of depression among those who have CBS. The second hypothesis proposed that there would be a larger proportion of people with MCI among those who have CBS, and the third hypothesis predicted that there would be a larger proportion of people with depression and MCI among those who have CBS as opposed to those who do not.
Method
Participants
A total of 42 participants (31 females, 11 males), ranging in age from 68 to 99 years (M = 85.5 years), were recruited from the clientele of the MAB-Mackay Rehabilitation Centre (MMRC) between November 2011 and July 2012; 22 of whom experienced CBS, and 20 of whom did not. Their VAs ranged from 20/70 to 20/1200 in the better eye, with 23 being partially sighted (20/70–20/180) and 19 being legally blind (VA < 20/200). Inclusion criteria required participants to have a diagnosis of AMD and to speak and understand English or French. The clients could be diagnosed with scotomas of any size, a VA of any level, and a loss of contrast sensitivity of any severity. Conditions excluding a client were a diagnosis of AD, clinical depression, other psychiatric illness, or prescriptions for anti-psychotic medication or selective serotonin reuptake inhibitors (SSRI) medications. The study protocol followed the tenets of the Declaration of Helsinki (Williams, 2008). Ethics approval was obtained from the Center for Interdisciplinary Research in Rehabilitation of Greater Montreal.
Materials
This study used the Montreal Cognitive Assessment (MoCA) questionnaire, which screens for MCI. Both the English and French versions were used (Nasreddine et al., 2005). The MoCA is designed to assess various areas of cognition to determine whether any impairment is present. It is a 30-item test which can be adapted for the visually impaired by omitting the first eight questions. The adapted version is scored out of 22 with a cut-off score of 18, indicating a referral for additional neuropsychological testing as the person may be at risk of MCI (Wittich, Phillips, Nasreddine, & Chertkow, 2010). Participants were categorized by their scores as those scoring 18 and higher and those scoring lower than 18. The Geriatric Depression Scale, which is used to screen older adults for symptoms of depression, was used to identify participants who were at risk of depression. Both the English and French versions of the test were used (Clement, Nassif, Leger, & Marchan, 1997; Yesavage et al., 1982).
There is currently no objective test for CBS; however, several researchers have used responses to specific questions to screen for visual hallucinations. An example of one question is “Sometimes people with poor eyesight see things which they know are not actually there. Has that ever happened to you?” (Lannon, Stevenson, White, Reinhardt-Rutland, & Jackson, 2006). Other studies include questions such as “Apart from blurred vision, have you noticed anything else unusual about your vision? Have you had any unusual visual experiences?” (Menon, 2005). If the individual responds positively additional questions follow in an attempt to determine the type of visual experience, for example, “What image did you see?” (Menon, 2005). These are followed up with statements asking the individual to describe what they saw, whether the image was static or dynamic, how the individual felt about the visions, and whether or not they have disclosed having hallucinations to others (Menon, 2005; Wagner-Lampl & Oliver, 2000). Participants were assigned to the CBS group if they answered yes to any of the above questions while those answering no to all questions were assigned to the non-CBS group.
Procedure
Individuals with a diagnosis of AMD were selected based on the information contained in their file. Once the criteria of the study were met, a clinician other than the primary investigator contacted the clients by telephone to inform them of the research study and to inquire if they were interested in participating. If the clients responded positively, an appointment was scheduled to proceed. Written informed consent was obtained prior to the administration of the questionnaires. Depending on the level of functional vision, participants were able to use visual aids such as closed circuit televisions (CCTVs) or magnifiers to complete the questionnaires. If, due to the participants’ severity of vision loss, the use of optical devices was not helpful, the questionnaires were verbally administered. If, upon completion of the Geriatric Depression Scale and the MoCA, any participant showed signs of depression or MCI, they were referred to the clinical coordinator of the visually impaired seniors program for follow-up.
Results
Pearson’s correlation coefficients on the raw scores of all measures are displayed in Table 1. Increased age was statistically significantly correlated with increased logMAR acuity (poorer VA). All participants were categorized into groups, according to the presence or absence of CBS, and proportional distribution across groups was compared using chi-square nonparametric statistics. Their distribution across groups is displayed in Table 2.
Pearson’s correlation coefficients for all variables.
MoCA: Montreal Cognitive Assessment; GDS: Geriatric Depression Scale; MAR: minimum angle of resolution.
p < .05 (two tailed); **p < .01 (two tailed).
Distribution of participants across all grouping variables.
CBS: Charles Bonnet syndrome; MoCA: Montreal Cognitive Assessment.
When testing whether the proportion of individuals at risk of depression would be higher among those with symptoms of CBS, no statistically significant relationship was apparent, χ2(1, n = 20) = .30, p = NS. Furthermore, when examining the proportion of participants who failed the MoCA among those with symptoms of CBS, no statistically higher proportion was noted, χ2(1, n = 22) = .87, p = NS. Finally, when testing whether the proportion of individuals both at risk of depression and MCI would be higher among persons who reported symptoms of CBS compared to those without, no significant relationship was found, χ2(1, n = 42) = .81, p = NS.
Discussion
This study examined the potential relationship among depression, MCI, and CBS in older adults diagnosed with AMD. Examining the first two variables, the results demonstrated no relationship between CBS and depression. Nonetheless, when examining the results of the Geriatric Depression Scale alone, it showed that 33% of all participants presented with symptoms of depression. All the participants (average age = 85.5 years) expressed significant changes in their ability to join in activities they had enjoyed prior to their vision impairment, such as driving a car, playing cards, reading the newspaper, and volunteering, not to mention not being able to clearly see the faces of their grandchildren, other family members, and friends. This result was consistent with previous studies indicating that roughly one in three individuals with vision impairment have signs of depression (Brody et al., 2001; Horowitz, 2003; Overbury & Wittich, 2011). Older adults with vision loss are two to five times more likely to suffer from depressive symptoms than their sighted peers as they attempt to come to terms with their impairment while trying to maintain their independence. This suggests that vision loss is a more significant risk factor for depression than any other age-related health problem (Horowitz, 2003; Rovner & Casten, 2002; Rovner et al., 2007).
Regarding the second hypothesis, previous studies indicated a possible relationship between CBS and MCI (Menon, 2005; Pliskin et al., 1996), although our study was unable to replicate these findings. Interestingly, 26 of the 42 participants (62%) failed the MoCA. This unusually high percentage could have been due to the advanced age of participants as studies have shown a possible correlation between aging and decreased cognitive function (Clemons et al., 2006; Petersen et al., 1999). We have previously reported on a different sample of older visually impaired adults with similarly high failure rates on the MoCA, whereby we speculated that test anxiety may contribute to the high failure rate (Wittich, Murphy, & Mulrooney, 2014). It is worth repeating in this context that visual ability was not a factor in the cognitive screen used in our study, as the blind version of the MoCA does not require functional vision and is administered verbally only (Wittich et al., 2010).
Our results are clinically significant and consistent with previous research showing higher rates of MCI in individuals with AMD as compared to in those without (Clemons et al., 2006; Rovner et al., 2009). According to previous work, older individuals who showed signs of MCI were twice as likely to have signs of early onset AMD suggesting a significant association (Baker et al., 2009; Pham et al., 2006). Furthermore, individuals with vision impairment tend to experience a loss of independence and a decrease in social activities which has been shown to contribute to a loss of cognitive function (Clemons et al., 2006). On a more qualitative note, participants who experienced CBS reported during administration of the questionnaire that their hallucinations were primarily of people, usually children. Some participants saw animals, mainly dogs, while a few experienced geometric shapes. Participants reported that all their experiences were pleasant with the exception of one participant who reported seeing children playing in the middle of the street and feared they would get hit by cars. Most of the participants were not aware these hallucinations had an actual diagnostic name. However, with more detailed explanations, they appeared relieved that the hallucinations were not related to mental health. Individuals experiencing CBS are more likely to confide in professionals who are knowledgeable about the syndrome. Ideally, those diagnosed with CBS could have their hallucinations reduced by learning rehabilitative techniques, participating in peer-support groups, or improving visual functioning (Eperjesi & Akbarali, 2004).
Limitations and future directions
It is estimated that only approximately 20% of individuals diagnosed with AMD also experience CBS (Lannon et al., 2006) but reliable data on prevalence are unlikely to exist due to issues with disclosure. Previous studies suggest that older adults are more likely to disclose the occurrence of visual hallucinations to someone they trust (Cohen et al., 2003; Fernandez et al., 1997). Therefore, responding to questionnaires administered by a researcher with whom they have not had the time to develop a trust relationship may have resulted in participants not willingly admitting to experiencing CBS. This raises questions about whether our non-CBS group may actually have included individuals who chose not to disclose. Gilmour et al. (2009) reported that CBS was twice as likely to occur in those with visual acuities ranging between 20/300 and 20/800. In our sample, only nine CBS participants (41%) had acuities within this range. Studies with larger samples will be required to determine the strength of the relationship between acuity and the experience of CBS. Furthermore, our participants were chosen from only one rehabilitation center. Ideally, recruitment could be facilitated by including other establishments to increase the sample size and to include a control group. Such an approach would allow future studies to detect smaller effects that were missed in the current data set. Finally, this study involved older adults with a diagnosis of AMD and, as such, it excluded other important visual diagnoses such as glaucoma, diabetic retinopathy, and cataracts.
It has been suggested that improving older adults’ functional capabilities by providing low-vision devices can decrease the occurrence of hallucinations (Anderson, Pankow, & Luchins, 2004; Eperjesi & Akbarali, 2004). Therefore, we can speculate that it is appropriate for rehabilitation centers to include questions pertaining to CBS on intake forms or global evaluation interviews as it is at this initial stage of rehabilitation that a relationship of trust is forged between the client and the low-vision rehabilitation specialist. Determining whether a client has CBS can be instrumental in the development of their individual intervention plan and can alert other members of the interdisciplinary team in order to optimize the client–clinician relationship, thereby ultimately improving the client’s quality of life.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
