Abstract
Background:
Impairment in goal-directed behavior (GDB) contributes to apathy, a prevalent syndrome in Parkinson’s disease (PD). The Philadelphia Apathy Computerized Task (PACT) is a performance-based measure of GDB that may be less confounded by reduced patient insight, cognitive impairment, and care partner burnout.
Objective:
To examine how the PACT is related to patient function and care partner burden.
Methods:
PD patients with normal cognition (n=19) or mild cognitive impairment (n=14) and their care partners were recruited. Participants completed the PACT, a computerized paradigm consisting of subtasks specific to each component of GDB: initiation, motivation, and planning. Care partners completed the Zarit Burden Interview (ZBI) and the Penn Parkinson’s Daily Activities Questionnaire (PDAQ-15). The associations between mean latency on each PACT subtask and ZBI and PDAQ-15 scores, respectively, were tested using Spearman’s rank correlation coefficients. Significant associations were further delineated using multivariate regression with the following covariates: age, years of education, MoCA score, daily levodopa equivalency dose, UPDRS Part III score, and GDS-15 score.
Results:
Worse performance on the planning subtask of the PACT related to higher ZBI scores and lower PDAQ-15 scores when adjusting for covariates. Decreased initiation was associated with higher ZBI and decreased motivation with lower PDAQ-15.
Conclusions:
Specific components of the PACT are related to patient and care partner outcomes in PD. The main advantage of this measure is to minimize the confounds of poor insight and care partner distress. We propose future research directions to refine the PACT for potential use in research and clinical practice.
Introduction
Apathy is a prevalent and clinically impactful feature of Parkinson’s disease (PD) that leads to impaired activities of daily living[1] and greater care partner burden[2]. Historically described as a disorder of motivation[3], apathy has been recently conceptualized as a reduction in specific components of goal-directed behavior (GDB), namely initiation, motivation, and planning[4]. This quantitative definition is based on evidence that the prefrontal cortex (PFC)-basal ganglia network is integral to the generation and modulation of purposeful behavior[5]. In their review on the clinical features of apathy and the possible underlying and neural mechanisms, Pagonabarraga and colleagues proposed that when areas of the PFC-basal ganglia network are disrupted, there are negative clinical consequences, such as emotional blunting, negative affect, cognitive inertia, and lack of self-initiated behaviors[6].
It has been proposed that the PFC-basal ganglia network is directly related to the behavioral, emotional, and cognitive factors that contribute to different subtypes of apathy[3]. Levy and Dubois[4] subsequently adapted these subtypes as auto-activation, emotional-affective, and cognitive, which correspond to the components of GDB that are required to perform an action: initiation, motivation, and planning. Previous studies in patients who have structural PFC disease support the hypothesis that impairments in GDB involve localized regions of the PFC-basal ganglia network[7]. Considering the high prevalence of apathy in PD, a prototypical basal ganglia disorder, we aim to quantify it in people with PD (PwP).
Existing tools to measure apathy rely on surveys completed by the patient or a care partner, who can be confounded by reduced insight, care partner burnout, or cognitive impairment. Other potential barriers include differentiating apathy from bradykinesia and depression, and a tendency to focus on a single component of apathy (i.e., motivation) rather than consideration the overall reduction in GDB. To address these limitations, The Philadelphia Apathy Computerized Task (PACT) was developed to measure GDB[8], [9]. We focused on a theoretical model of apathy that results from a quantitative reduction in GDB. Based on existing literature indicating that three distinct components – initiation, motivation, and planning – are necessary to achieve GDB[4], we hypothesized that apathy results when any of these components are impaired (Figure 1). The PACT was designed to quantify initiation, motivation, and planning to directly measure GDB without confounding by care partner report or patient insight. The current study investigates whether performance on the PACT is related to functional outcomes. We hypothesized that reduction in any component of GDB – initiation, motivation, or planning – is related to higher care partner burden and impaired ability to perform instrumental activities of daily living (IADLs).
Figure 1.
Conceptual model of apathy. ACC: anterior cingulate cortex. dPFC: dorsolateral prefrontal cortex. OFC: orbitofrontal cortex.
Methods
Ethics
The University of Pennsylvania Institutional Review Board approved this study. Written informed consent was obtained from study participants prior to enrollment.
Sample
PwP with normal cognition (NC) or mild cognitive impairment (MCI) were recruited from a larger group of individuals enrolled in the Udall Center for Parkinson’s Research at the University of Pennsylvania. All participants were recruited from the Parkinson’s Disease and Movement Disorders Center at the University of Pennsylvania as a convenience sample. Inclusion criteria were a clinical diagnosis of PD and a care partner who could respond to written surveys about the participant. Participants were categorized as having NC or MCI based on consensus diagnosis by pairs of physician raters using a mix of global and domain-specific neuropsychological tests and functional measures described in detail previously[10]. We excluded those with dementia based on Movement Disorder Society criteria using the same consensus scoring system.
Measures
Demographic information was obtained from all participants. Participants completed the Montreal Cognitive Assessment (MoCA) and the 15-item Geriatric Depression Scale (GDS-15). Each patient’s medication regimen was converted to a daily levodopa equivalency dose (LED)[11]. The original Unified Parkinson’s Disease Rating Scale (UPDRS) Part III was performed by a qualified rater.
Computerized reaction times were obtained via the PACT to quantify initiation, motivation, and planning. See Supplementary Methods for specifics on the behavioral tasks associated with the PACT. Care partner burden was assessed using the Zarit Burden Interview (ZBI), where higher scores indicate greater burden. IADLs were assessed using the 15-item Penn Parkinson’s Daily Activities Questionnaire (PDAQ-15), completed by the care partner, where higher scores indicate better function.
Data analysis
Data analyses were performed using R Statistical Software (v4.2.2). Associations between mean latency on each PACT subtask and ZBI score and PDAQ-15 scores, respectively, were tested using Spearman’s rank correlation coefficients. We considered age, years of education, MoCA score, LED, UPDRS Part III score, GDS-15 score, and disease duration as potential confounders. If there was a statistically significant association between PACT performance and an outcome measure, this association was further investigated using linear regression with the aforementioned potential confounding factors as covariates. Log transformation of outcome measures were used for regression analyses to address the data skew.
Results
Participant demographic information is described in supplementary Table 1. Nineteen PD participants (57.4%) had NC, and 14 (42.4%) had MCI. Eighteen (54.5%) participants were male and 15 (45.4%) were female. The mean (SD) age was 70.3 (7.1) years in the NC group and 69.6 (6.1) years in the MCI group, p > 0.05. The median (IQR) GDS-15 score in the NC group was 2 (1–4.5) and 4 (1.25–5) in the MCI group, p > 0.05. Four participants were taking a cognition-enhancing medication, 13 were taking an antidepressant, and 9 were taking a benzodiazepine. There was no relationship between taking a centrally acting medication and performance on any PACT subtask.
Associations between PACT performance and functional outcome measures
Slower performance on each PACT subtask was positively correlated with ZBI scores (Figure 2a-c). When adjusting for age, disease duration, years of education, MoCA score, LED, UPDRS Part III score, and GDS-15 score, performance on the initiation and planning subtasks was still associated with log transformed ZBI scores (Figure 3a, 3c; supplementary Table 2). Performance on the motivation subtask was not associated with ZBI score when adjusting for covariates (Figure 3b).
Figure 2.
Spearman’s rank correlations between PACT and functional outcome measures (ZBI, PDAQ-15). *p < 0.05 * p < 0.01 ***p < 0.001. ZBI: Zarit Burden Interview, PDAQ-15: Penn Parkinson’s Daily Activities Questionnaire. Blue color indicates a positive correlation coefficient and red color indicates a negative one.
Figure 3.
Linear regression predicting log transformed functional outcome measures (ZBI, PDAQ-15) using PACT and covariates: age, disease duration, years of education, MoCA score, daily levodopa equivalency dose, UPDRS Part III score, GDS-15 score. *p < 0.05 **p < 0.01 ***p < 0.001. MoCA: Montreal Cognitive Assessment, UPDRS: Unified Parkinson’s Disease Rating Scale, GDS-15: 15-item Geriatric Depression Scale. Blue color indicates a positive beta coefficient and red color indicates a negative one.
PACT performance was negatively correlated with the PDAQ-15 (Figure 2d-f). Adjusting for age, disease duration, years of education, MoCA score, daily LED, UPDRS Part III score, GDS-15 score, motivation and planning was still associated with log transformed PDAQ-15 scores (Figure 3e, 3f; supplementary Table 2). Initiation was no longer associated with PDAQ-15 scores when adjusting for covariates (Figure 3d).
Discussion
The current study examined whether GDB related to meaningful functional outcomes in PwP and their care partners. We found that planning was related to care partner burden and IADLs even after adjusting for important potential confounding factors, such as cognitive, psychiatric and motor scores, and dopamine replacement therapy. Additionally, initiation was related to care partner burden and motivation was related to IADL function.
It is notable that only planning was associated with both care partner burden and cognitive function. The planning subtask may involve multiple executive functions (e.g., working memory, rule-finding, and set-shifting), which may explain why it relates to both patient and care partner outcomes. On the other hand, initiation and motivation are single components of complex GDBs relevant for daily functioning, such as handling an unfamiliar problem or learning to use a new household gadget. Those who need a strong cue to initiate a complex task but otherwise can follow through (i.e., have preserved planning) may be able to maintain relative independence. Conversely, patients who can initiate with minimal cuing and appear motivated but consistently lack the executive function to bring the task to completion may have worse functional outcomes, particularly if their care partner must take over. We also consider the possibility that the initiation and motivation portions of the PACT was unintentionally easier to perform compared to the planning subtask, thereby decreasing our ability to measure more subtle deficits in initiation and motivation.
The PACT was easy to administer and appeared to be well understood even in those with MCI. The relationship between GDB and functional outcomes seems to be unaffected by the presence of depressive symptoms. Though we created the PACT for a laptop computer, we increasingly see other paradigms being successfully administered via wearable sensors. Existing wearable technologies in PwP focus on motor signs (e.g., tremor, limb bradykinesia, gait) rather than neuropsychiatric and nonmotor symptoms. In the future, modifying the PACT to be administered on a wearable device can enable the assessment of apathy in real time in various settings. This can help characterize not only the degree of apathy, but also fluctuations in GDB, which may provide a clue into when nonmotor “on” and “off” states may be occurring.
Future studies are required to determine the clinimetric properties of the PACT and should focus on addressing the following issues. First, considering the comorbidity of depression and apathy in patients with PD [12], [13], the ability of the PACT to measure apathy in the presence and absence of a clinical diagnosis of depression should be investigated. This would also help clarify whether poor performance on the motivation subtask is entirely due to reduced GDB (i.e., suggestive of pure apathy) or is also related to anhedonia. In our study, depressive symptoms were not associated with care partner burden, but independently associated with worse IADLs, though our participants had overall low GDS-15 scores. Second, the ability of the PACT to measure apathy in the presence of dementia should be determined. Third, future studies should incorporate comparisons to established measures of apathy. For the current participants, the “motivation/initiative” item of the UPDRS Part I and the Neuropsychiatric Inventory-Questionnaire (NPI-Q) “apathy/indifference” item did not associate with outcome measures. Lastly, it is important to determine whether the PACT can detect changes in apathy, which can be evaluated through longitudinal studies.
This study has several limitations. First, due to our small sample size, the relationships we found between the PACT and functional outcomes, should be considered preliminary and investigated in a larger cohort. Second, although we included many important potential confounders in our analyses, we did not account for all the comorbid conditions that can be seen in PD, such as fatigue and sleep disorders. Third, care partner reports of burden and functional ability can be inherently biased by burnout. Future studies could utilize a performance-based measure of IADLs. Fourth, although we included 12 practice trials for each portion of the PACT to ensure that participants understood the paradigms, we did not utilize performance validity tests (PVTs). Incorporating PVTs in future studies would help ensure that only valid test trials are analyzed. We also did not include any measures of processing speed in our study design. Lastly, measures of culture and health disparities were not assessed, which constrains the generalizability of our findings. Future studies should include nuanced measures of education, culture, and health disparities, as these factors can impact response to performance-based tasks, cognitive reserve, and care partner expectations.
Despite these limitations, the PACT offers a unique performance-based approach to investigate the specific components of GDB that may be impaired in apathy. The main advantage of the PACT is that it does not depend on informant report, thereby minimizing the effects of poor insight or care partner distress on apathy measurement. Use of the PACT could be a useful, objective outcome measure in clinical trials, longitudinal studies of wearable sensors, and potentially clinical practice.
Supplementary Material
Use of subjective report to measure apathy in Parkinson’s disease (PD) is confounded by reduced insight and caregiver burnout
Computerized measurements of apathy are feasibly obtained in PD patients with and without cognitive impairment
Performance is related to instrumental activities of daily living and caregiver burden
Acknowledgment:
We are grateful to Chinwe Nwadiogbu for her assistance in data collection.
Lauren Massimo, PhD, CRNP received funding from NIA R01AG076832
Nabila Dahodwala, MD, MS received funding from Parkinson’s Foundation (PDF-TRG1724)
Funding Sources:
Parkinson’s Foundation PDF-TRG-1724, NINDS P50 NS053488, NIA R01AG076832
Footnotes
Declarations of interest: none
Declaration of Conflict of Interest
The authors of this manuscript do not have competing interests to disclose.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- [1].Laatu S, Karrasch M, Martikainen K, and Marttila R, “Apathy is associated with activities of daily living ability in Parkinson’s disease,” Dement Geriatr Cogn Disord, vol. 35, no. 5–6, pp. 249–255, 2013, doi: 10.1159/000347129. [DOI] [PubMed] [Google Scholar]
- [2].Leroi I, Harbishettar V, Andrews M, McDonald K, Byrne EJ, and Burns A, “Carer burden in apathy and impulse control disorders in Parkinson’s disease,” Int J Geriatr Psychiatry, vol. 27, no. 2, pp. 160–166, 2012, doi: 10.1002/gps.2704. [DOI] [PubMed] [Google Scholar]
- [3].Marin RS, “Apathy: a neuropsychiatric syndrome,” J Neuropsychiatry Clin Neurosci, vol. 3, no. 3, pp. 243–254, 1991, doi: 10.1176/jnp.3.3.243. [DOI] [PubMed] [Google Scholar]
- [4].Levy R. and Dubois B, “Apathy and the functional anatomy of the prefrontal cortex-basal ganglia circuits,” Cerebral Cortex, vol. 16, no. 7, pp. 916–928, 2006, doi: 10.1093/cercor/bhj043. [DOI] [PubMed] [Google Scholar]
- [5].Kimura M. et al. , “Goal-directed, serial and synchronous activation of neurons in the primate striatum,” Neuroreport, vol. 14, no. 6, pp. 799–802, 2003, doi: 10.1097/00001756-200305060-00004. [DOI] [PubMed] [Google Scholar]
- [6].Pagonabarraga J, Kulisevsky J, Strafella AP, and Krack P, “Apathy in Parkinson’s disease: Clinical features, neural substrates, diagnosis, and treatment,” Lancet Neurol, vol. 14, no. 5, pp. 518–531, 2015, doi: 10.1016/S1474-4422(15)00019-8. [DOI] [PubMed] [Google Scholar]
- [7].Levy R. and Czernecki V, “Apathy and the basal ganglia,” J Neurol, vol. 253, no. SUPPL. 7, pp. 54–61, 2006, doi: 10.1007/s00415-006-7012-5. [DOI] [PubMed] [Google Scholar]
- [8].Massimo L, Evans LK, and Grossman M, “Differentiating Subtypes of Apathy to Improve Person-Centered Care in Frontotemporal Degeneration,” J Gerontol Nurs, vol. 40, no. 10, pp. 58–65, 2014, doi: 10.3928/00989134-20140827-01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Massimo L. et al. , “Apathy in Frontotemporal Degeneration: Neuroanatomical Evidence of Impaired Goal-directed Behavior,” Front Hum Neurosci, vol. 9, no. November, pp. 1–10, 2015, doi: 10.3389/fnhum.2015.00611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Pigott K. et al. , “Longitudinal study of normal cognition in Parkinson disease,” Neurology, vol. 85, no. 15, pp. 1276–1282, Oct. 2015, doi: 10.1212/WNL.0000000000002001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Tomlinson CL, Stowe R, Patel S, Rick C, Gray R, and Clarke CE, “Systematic review of levodopa dose equivalency reporting in Parkinson’s disease,” Movement Disorders, vol. 25, no. 15, pp. 2649–2653, Nov. 2010, doi: 10.1002/mds.23429. [DOI] [PubMed] [Google Scholar]
- [12].Pluck GC and Brown RG, “Apathy in Parkinson ‘ s disease,” pp. 636–642, 2002. [DOI] [PMC free article] [PubMed]
- [13].Starkstein SE, Mayberg HS, Andrezejewski P, Leiguarda R, and Robinson G, “Reliability, validity, and clinical correlates of apathy in Parkinson’s disease,” J Neuropsychiatry Clin Neurosci, vol. 4, no. 2, pp. 134–139, 1992, doi: 10.1176/jnp.4.2.134. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.