Apathy is prevalent in this sample of older adults, has a high incidence rate, and is stable in 46.4% of the sample at 1 year follow-up and 8.2% at 13 years follow-up. It does not affect cognitive decline but is significantly associated with functional decline. As well, age, race, and education were independent predictors of the risk of cognitive decline.
It is important to note the limitations of this study. The racial / ethnic composition of the Baltimore ECA study sample was mainly Whites and African-Americans and, as such, the results might not be generalizable to all racial / ethnic groups within the US or globally. The use of GHQ items to assess apathy is a weakness since use of these items in this context has not been validated elsewhere or against existing apathy measures. However, the use of the psychiatrists’ ratings of apathy based on the SCAN items to validate the use of the GHQ apathy items lends credibility to the results, since it is validated against a clinical judgment. Furthermore, the use of the GHQ-apathy items allowed us to examine the prevalence, incidence, and correlates of apathy in a community-dwelling sample with a scale that was not designed for and validated in a clinical sample. In addition, the method used to identify functional decline was limited in that it was dichotomous and failed to capture the various levels of functioning. However, since the study involved secondary data analysis, the researchers had to make use of what was available in the data set despite limitations. The results however, provide support for the idea that apathy can negatively impact functional ability, as observed by other apathy researchers (Freels et al., 1992
; Starkstein et al., 2001
; Lechowski et al.
Studies on apathy in older adults who were referred to an outpatient memory disorder clinic for neurological and neuropsychiatric assessments due to memory complaints, have reported current prevalence rates ranging from 28.3% (Clarke et al., 2007
) to 86.4% (Thomas et al., 2001
). The prevalence of apathy observed in the current study (23.7%) is similar to that reported by Starkstein et al (2006)
and Turro-Garriga et al
(2009) in the existing longitudinal studies and lower than the range observed in clinic samples, likely reflecting the composition of this less cognitively impaired population-derived sample. However, two studies on community-dwelling older adults using Neuropsychiatric Inventory (NPI) to assess apathy have reported prevalence estimates between 17–37% for the subset of individuals with dementia (Onyike et al., 2007
, Lyketsos et al., 2002
) and 3.1 (Onyike et al., 2007
) to 15% (Lyketsos et al., 2002
) for individuals with mild cognitive impairment (MCI). Therefore, the apathy estimate observed in the current study falls within the category of community-dwelling individuals with dementia. It is possible that the measurement used herein to classify individuals as demented was not as sensitive as the measures used in the aforementioned studies. As such, despite our attempts to remove individuals with dementia at baseline (i.e., individuals with MMSE ≤ 23 at baseline excluded), it is likely that some individuals with borderline dementia or MCI/cognitive impairment no dementia (CIND) were misclassified as non-demented. This assumption is supported by Kukull and colleagues (1994)
finding that the MMSE cut-off score (≤ 23) had a sensitivity of 0.63 and a specificity of 0.96 in a community-dwelling sample with cognitive complaints. In addition, the variations of measures used to ascertain apathy status may account for the observed differences in the prevalence of apathy. Depression diagnosis, and ADL and IADL impairment were associated with apathy; these results are in agreement with the current literature on apathy (Freels et al., 1992
; Lavretsky et al., 1999
; Starkstein et al., 2001
; Starkstein et al., 2006
; Turro-Garriga et al.
The observed incidence of apathy at 1-year follow up in individuals without apathy at baseline (i.e., 22.6%) is similar to the 21.7% reported by Turro-Zarriga and colleagues (2009) in which the NPI was used to ascertain apathy status. This finding lends support to the use of the GHQ-apathy related items to ascertain apathy status in the current study. The observation of persistent apathy in 46.4% of the sample at 1-year follow-up was slightly lower than the 51.7% ascertained with the NPI in the longitudinal study on apathy conducted by Turro-Zarriga and colleagues (2009) or the 50% reported by Starkstein et al (2006)
in which the 10-item Apathy Scale was used to ascertain severity of apathy. The fact that the study by Turro-Zarriga and colleagues consisted of individuals with mild to severe AD might account for the slight difference in the percentage of the two study samples with persistent apathy at 1 year follow-up. On the other hand, Lechowski and colleagues (2009)
, using the NPI to ascertain apathy status, found that only 22.1% of their sample of women with AD had persistent apathy at 18-months follow-up. The current study is the first to report the persistence of apathy at 13 years follow-up and using a non-specialized clinical sample. Unfortunately, the number of individuals with persistent apathy at 13 years follow-up was small (n=22), and hence lacked the study power to conduct in-depth inferential analyses. Given that apathy is highly prevalent across a number of disorders and related to significant cognitive and functional decline, it is important to identify and further characterize this high risk group (i.e., individuals with persistent apathy) in terms of socio-demographic, clinical, and genetic factors. An initial attempt at this line of research was carried out by Lechowski and colleagues (2009)
in their study involving females with AD in outpatient care that found persistent apathy was associated with “rapid loss of autonomy in activities of daily living”. It would be important to know if these results extend to other dementia groups, males and general community samples.
Individuals who had apathy at baseline were more likely to experience functional decline in instrumental (IADL) and basic (ADL) activities of daily living at one and 13 years follow-up compared to those who did not have apathy. This greater risk of functional decline in individuals who had apathy, which persisted despite adjustment for age, sex, race, education, and the changing state of apathy and depression, is consistent with results observed by Starkstein et al., (2006)
, Turro-Zarriga et al.
, (2009), and Lechowski et al., (2009)
in their respective longitudinal studies. These more recent results advance our knowledge of the effects of apathy that mainly had been based on the observation of statistically significant associations in cross-sectional studies (e.g., Landes et al., 2005
; Onyike et al., 2007
; Clarke et al., 2008
). These cross-sectional studies have also indicated significant associations between apathy and poor cognitive functioning (Landes et al., 2005
; Onyike et al., 2007
; Clarke et al., 2008
). The existing longitudinal studies, albeit limited in that they utilized dementia patients in outpatient care at various dementia clinics, have provided supporting results that indicated a causal relationship between apathy and cognitive functioning (Robert et al., 2006
; Starkstein et al., 2006
; Lechowski et al., 2009
; Turro-Zarriga et al.
, 2009). More specifically, such studies have shown that apathy at baseline was associated with significant decline in cognitive functioning at one year follow-up. Although the current study also found a statistically significant relationship between apathy and cognitive decline at one year follow-up this was not observed at 13-year follow-up. Furthermore, the significant relationship between apathy and cognitive decline at one year follow-up did not hold true when the changing state of apathy, depression, and functional status were taken into account using GEE analyses. The observations from this study might suggest the potential benefits of the reversal of apathy on cognitive functioning. Further investigations into the longitudinal relationship between apathy and cognitive functioning are needed.
The discrepancy between the results from this study and the existing longitudinal studies that have also taken into account the repeated measure of apathy could be related to the differences in how cognitive decline and apathy were measured. It is possible that the GHQ apathy-related items were not sensitive enough to capture all individuals with apathy (i.e., misclassification of individuals with apathy as non-apathetic), thereby diluting the measure of effect. Furthermore, survival bias and the observation of a transient apathy group (i.e. individuals with apathy at baseline but not at follow-up) might account for the lack of significant association between baseline apathy and cognitive decline at 13-year follow-up. In other words, it is possible that individuals with transient apathy sought treatment and the remaining individuals with apathy who survived through the 13-year follow-up interview were the healthiest of the stable apathetic group. This line of reasoning is supported by the observation that across apathy groups, those who died by the 13-year follow-up interview were on average 7 years older and possibly had more health problems.
In summary, the current study supports the notion that apathy is an important predictor of functional decline and possibly cognitive decline in older adults. Given the high prevalence of apathy across a number of disorders, it is important to gain better understanding of apathy and its long-term impact on cognitive and functional status. This enhanced knowledge can improve treatment and prevention of cognitive and functional declines.