Exploring the relationship between depression and cognitive anosognosia in Parkinson’s disease: a cross-sectional study

Introduction

Neuropsychiatric disorders are highly prevalent symptoms of Parkinson’s disease (PD), encompassing a range of disorders, including depression, anxiety, and apathy (Aarsland et al. 2009). Many researchers also classify cognitive impairment as a neuropsychiatric disorder (Weintraub and Burn 2011; Mueller et al. 2018). Cognitive decline is widespread, with up to 80% of individuals with PD experiencing some form of cognitive dysfunction as the disease progresses (Biundo et al. 2016). Cognitive profiles can vary significantly among individuals with PD (Kehagia et al. 2013; Biundo et al. 2016). However, the most common cognitive issue is the impairment of executive functions, likely due to the disruption of frontal-striatal networks caused by dopamine deficiency (Kudlicka et al. 2011). Neuropsychiatric conditions, including anosognosia – characterised by diminished awareness of one’s deficits or symptoms – are frequently diagnosed across a range of neurodegenerative disorders (Rosen 2011). In PD, anosognosia typically manifests as a reduced or complete lack of awareness regarding an individual’s cognitive or motor deficits, or both (Orfei et al. 2018). Since executive functions are the most commonly impaired cognitive domain in PD, being aware of these problems is particularly important. Diminished awareness of cognitive impairment is referred to as cognitive anosognosia (Barrett et al. 2005). In PD, cognitive anosognosia is observed in individuals and is related to substantial frontal dysfunction (Yoo et al. 2020).
Additionally, neuropsychiatric disorders are often comorbid with neurodegenerative diseases (Aarsland et al. 2009). Consequently, the relationship between depression and anosognosia has become a subject of considerable research interest (Starkstein et al. 1997; Smith et al. 2000; Verhülsdonk et al. 2013). Understanding this connection is crucial, as it may provide insight into the mechanism underlying depression in PD, potentially influencing treatment protocols. Previous studies have shown that cognitive anosognosia is associated with lower levels of depression, which might support the role of psychological factors in development of depression in PD (Yoo et al. 2020). However, the methodology of anosognosia assessment varies substantially among studies, and some limitations of previous research can be pointed out. First, in the Orfei et al. study (2018), the Anosognosia Questionnaire for Dementia was used, which was initially designed for assessing anosognosia in Alzheimer’s disease (Starkstein et al. 2006). The profile of cognitive problems is different in Alzheimer’s and Parkinson’s disease; therefore, administration of the test may result in biased data. In the Yoo et al. study (2020), anosognosia was assessed as the difference between subjective cognitive complaints and objective cognitive functioning, which might be considered a strength; however, it also resulted in a limitation because anosognosia was not measured as a longitudinal variable. To the best of the authors’ knowledge, there is also a lack of evidence of cognitive anosognosia in PD, particularly regarding executive functions. They appear to play a unique role in anosognosia due to their frontal localization, and are among the most profoundly impaired cognitive processes in PD (Wibawa et al. 2020).
Thus, based on previous studies’ limitations, more evidence is needed to generalise conclusions about the relationships between anosognosia and depression. The study aimed to investigate the association between depression and cognitive anosognosia in individuals with PD. We hypothesised that there is a negative association between depression and anosognosia in PD.

Material and methods

This study employed a cross-sectional design, with two assessment points: an initial recruitment phase followed by a neuropsychological assessment to explore the relationship between depression and cognitive anosognosia in individuals with idiopathic PD. The study project received approval from the ethics committee (approval no. 45/2020). These results are part of a larger project, the details of which are described elsewhere (Golińska et al. 2021).

Sample

The sample consisted of individuals with idiopathic PD. The inclusion criteria were as follows: 1) a diagnosis of idiopathic PD confirmed by a neurologist based on the current International Classification of Diseases 10th Revision (ICD-10 code: G20) (World Health Organization 1993); 2) being right-handed; 3) being a native speaker in Polish; 4) giving consent to further participation in the study. The exclusion criteria included: 1) advanced vision and hearing problems (preventing a standardized neuropsychological diagnosis); 2) major depressive episode and history of mental illness (including schizophrenia and bipolar disease); 3) addiction to alcohol or other psychoactive substances; 4) a history of ischaemic or haemorrhagic stroke; 5) deep brain stimulation treatment; 6) other neurological diseases such as epilepsy. No restriction on age, sex, or education was applied. Ninety-six participants with PD and their close relatives took part in the study. Among the 96 participants, 45 (46.9%) were classified as cognitively unimpaired (PD-normal), 39 (40.6%) as having mild cognitive impairment (PD-MCI), and 12 (12.5%) as having mild dementia (PDD), based on neuropsychological evaluation. A control group (n = 46) was included and matched to the clinical groups in terms of age and sex. However, significant differences in educational background were observed, with the control group being more educated than patients with dementia. The dementia group also exhibited notably lower education levels compared to the other clinical groups, aligning with trends reported in the literature. Due to the nature of the analysis, the clinical group was considered as a whole without further division.
Demographic characteristics for the whole sample and with subdivision into clinical subgroups are presented in Table 1. The levodopa equivalent daily dose (LEDD) data presented in Table 1 are based on 95 participants, as one individual did not report this information.

Procedure

The study was conducted in two stages: a neurological examination and a neuropsychological assessment. Participants with PD were recruited from medical centres in a major city in Poland. After providing informed consent, individuals diagnosed with idiopathic PD underwent a comprehensive neurological evaluation to confirm eligibility and assess disease severity. Neurologists used standardised scales to assess motor symptoms and calculate the daily LEDD. The neurological examination was conducted during the “on” medication phase to ensure consistent motor performance.
The neuropsychological assessment took place in a dedicated research room at one of the Polish universities. A close relative accompanied the participants. The evaluation order was pre-planned and unified across participants. The entire procedure lasted approximately three hours. Caregivers were asked to complete questionnaires, including those related to their observations of the participant’s executive functioning. Breaks were offered to ensure participants’ comfort.

Instruments

Data collection included a socio-demographic survey to gather participant background information, such as education level. We also used declarative psychological questionnaires, including the Geriatric Depression Scale (GDS), to evaluate depressive symptoms and the Dysexecutive Questionnaire to assess executive dysfunction. The study was part of a larger project; consequently, other cognitive functions measurements were also used during the procedures. Since this paper focuses solely on depression and anosognosia, the remaining tests are mentioned, but not described.
The Geriatric Depression Scale is a widely used self-report questionnaire to assess the severity of depressive symptoms in the elderly (Yesavage et al. 1982; Albiński et al. 2011). The questionnaire comprises 30 statements, each answered on a yes/no basis. Its simplicity is an advantage, especially for clinical groups with cognitive impairment. In this study, the test demonstrated good internal consistency, with a Cronbach’s α of 0.88. The total score serves as an indicator of depressive symptoms. A cut-off score of 10 may suggest mild depression, while a cut-off score of 20 may indicate major depression. However, a diagnosis of clinical depression may be given only based on depression criteria and not on the declarative questionnaire, which measures depression symptoms.
The Dysexecutive Questionnaire (DEX) consists of 20 questions in two versions: the self-report version (DEX-S) and the caregiver version (DEX-I). The questions are identical in both versions and focus on the patient’s executive functions. Responses are given on a 5-point Likert scale, and the scores from both versions are compared. This comparison makes the DEX a valuable tool for assessing insight into the executive dysfunctions of patients. The severity of anosognosia is determined by calculating the difference between the total score on the questionnaire completed by the caregiver and the total score on the questionnaire completed by the patient. The greater the difference in scores between the questionnaire completed by the patient and the one completed by the caregiver (e.g. underestimation in the self-report), the more severe the anosognosia is considered to be. However, there are no specific criteria or cut-off scores to determine this severity definitively based on this questionnaire. The test demonstrated good internal consistency, with a Cronbach’s α of 0.90 for self-assessment and 0.92 for caregiver’s assessment. This questionnaire is part of the Behavioral Assessment of the Dysexecutive Syndrome (Wilson et al. 1996). The DEX is considered to have strong ecological validity, as it captures executive deficits that manifest in real-life contexts beyond laboratory settings. It evaluates practical impairments in planning, emotional control, impulse inhibition, and social behaviour. Example items include: “I act without thinking, doing the first thing that comes to mind”, “I have difficulty thinking ahead or planning for the future”, and “I say things that are embarrassing or inappropriate to others”.

Statistical analysis

Means, standard deviations, percentages, and the correlation coefficient were computed using IBM SPSS 28.0. Pearson’s r coefficient was calculated to examine the association between the study variables. The significance threshold was set at p = 0.05. The figures and subgroup-specific correlation analyses were generated using R software (version 4.4.3).

Results

Relationship between anosognosia and depression

To examine the relationship between anosognosia and depression, a Pearson’s correlation analysis was conducted (Table 2). Figure 1 visualizes the discrepancies in individual DEX scores. It demonstrates that caregivers tend to rate patients’ executive functioning as more impaired than patients rate themselves, indicating the presence of cognitive anosognosia in some individuals. While individual variability is evident, the overall pattern suggests a systematic underestimation of deficits by individuals with Parkinson’s disease. A statistically significant negative relationship was observed between anosognosia and depression, suggesting that higher levels of depressive symptoms are associated with lower levels of anosognosia. This relationship is further illustrated in Figure 2, which displays the association between depression and anosognosia for the entire sample as well as within three clinical subgroups: PD-normal, PD-MCI, and PDD. The negative association is most pronounced in the PD-normal group (r = –0.47**; CI [–0.68, –0.20]), whereas it is weaker and statistically nonsignificant in the PD-MCI subgroup (r = –0.10; CI [–0.42, 0.24]), and even slightly reversed in the PDD group (r = 0.14; CI [–0.54, 0.71]), although still statistically nonsignificant, as reflected by the broader confidence intervals. For descriptive statistics of the tests used and correlations by clinical subgroup, see Table S1.

Discussion

The hypothesis that depression negatively correlates with cognitive anosognosia has been supported. This finding is in line with results obtained in other research (Orfei et al. 2018; Yoo et al. 2020). Conde-Sala et al. (2014) also found that anosognosia may be positively associated with patients’ quality of life. To the best of our knowledge, studying the relationship between insight into cognitive difficulties in individuals with neurodegenerative disorders and depression could enhance our understanding of the development of depression not only among neurodegenerative patients, but also in neurologically healthy individuals, as it may reveal common mechanisms underlying the interplay between self-awareness, cognitive functioning, and emotional regulation.
There are several mechanisms behind depression in Parkinson’s disease. According to Aarsland et al. (2012), the aetiology remains unclear. However, researchers have identified potential contributing factors based on previous studies, including Parkinson-related changes in brain structures and neurotransmitters, levels of neuroinflammatory and neurotrophic factors, deep brain stimulation, psychosocial factors, and even the experience of pain (Aarsland et al. 2012). In summary, a multifaceted approach to depression in Parkinson’s divides these factors into biological and psychological categories. The question of whether PD is more closely associated with brain dysfunction or with psychological factors remains unresolved. For example, Tandberg et al. (1997) evaluated various clinical and demographic variables. The results revealed that biological factors are more likely to contribute to depression severity. Notwithstanding, situational variables also play an important role in depression development in some individuals (Tandberg et al. 1997). The authors identified two situational factors: independence in activities of daily living and being a nursing home resident. Additionally, Garlovsky et al. (2016), in their systematic review, highlighted psychological factors as important determinants of depression development in PD. They described the relationship between depression and coping mechanisms, personality, and illness cognitions, where illness cognition was defined as individuals’ beliefs about PD. These findings highlight a growing body of evidence emphasising the role of psychological factors in the development of depression within neurodegenerative diseases, as well as the significance of cognitive appraisal of an individual’s current circumstances, often referred to as “Parkinson’s beliefs”. Notably, cognitive appraisal necessitates an accurate assessment of one’s abilities, which may differ substantially between individuals with anosognosia and those with intact insight.
Anosognosia may make it difficult or even impossible to assess one’s cognitive problems, actual skills and capabilities, and limitations resulting from the disease (Sunderaraman and Cosentino 2017). In Alzheimer’s disease, it has been proven that anosognosia is related to behavioural and neuropsychiatric problems. Additionally, in Alzheimer’s disease, it has been found that anosognosia is more common in individuals with executive dysfunction, highlighting the role of executive functions in the development of anosognosia. The association between executive functions and anosognosia may be explained by the involvement of frontal brain regions, particularly the frontal poles (Stuss and Levine 2002), inferior frontal gyrus, anterior cingulate cortex, and medial temporal lobe (Hallam et al. 2020), all of which are crucial for metacognition.
According to Lazarus and Folkman’s transactional stress and coping theory (1984), the impaired insight into executive functions may lead to a reduced ability to form cognitive interpretations as part of cognitive and emotional appraisal. Additionally, following Beck et al. (1967), the cognitive approach to depression assumes an interpretation of stressful events. Based on the results of our study, we hypothesise that many of the psychological factors identified as significant contributors to depression in PD may stem from individuals’ core beliefs, not solely about the disease itself but, as suggested by Nolen-Hoeksema (1991), more broadly conceptualised as response styles that either heighten or mitigate the risk of depression. Our findings suggest that impaired insight into executive functions, which are crucial for decision-making, problem-solving, and planning (Collins and Koechlin 2012), likely extends to processes involved in confronting new situations or challenges with personal beliefs. This may support the relevance of the cognitive theory of depression in the development of depression in PD.

Limitations and future directions

Research on anosognosia face methodological challenges that require careful consideration. The limitations of this study should be taken into account: convenience sampling (a sample with higher education, mostly women), power issues, single indicators of studied constructs, and no LEDD control. Some researchers also claim that declarative measurement of depression in individuals with potentially reduced insight may lead to biased data. From this perspective, observational depression scales are often recommended in anosognosia research. However, in our opinion, this approach may also present a methodological trap. Caregivers are typically the most common evaluators of mood disorders in patients. However, providing feedback on a relative’s depression symptoms can be challenging due to the high prevalence of apathy in PD (Pedersen et al. 2009) and other neurodegenerative disorders (Radakovic and Abrahams 2018). Another limitation of the study is the lack of normalization for the DEX test. The DEX test does not allow for an assessment of whether the individual result is high or low. The outcomes in this study are based on group means. Apathy symptoms can imitate depression, and caregivers may lack the knowledge needed to distinguish between the two, which remains a challenge even for professional mental health specialists. Furthermore, it has been found that the risk of apathy increases with the severity of anosognosia (Starkstein et al. 2010). Observational scales, while useful, do not offer insight into the subjective perspective and depressive thoughts of the individual. Nevertheless, the declarative scales also might be biased by impaired insight. One potential solution for future studies is to incorporate more objective measures of depression, such as biological indicators. New anosognosia measurement procedures should also be developed in the future, for example, based on experimental design.

Disclosures

This project was funded by the National Science Centre (grant Preludium 17, No. 2019/33/N/HS6/00173), PI: Paulina Golińska.
The study was approved by the Bioethics Committee of the University of Gdańsk (Approval No. 45/2020).
The authors declare no conflict of interest.
Supplementary material is available on the journal’s website.

References