Objectives To compare the clinical judgment of experienced neurologists after interviewing Parkinson's disease (PD) patients and their caregivers with the use of the Pill Questionnaire to determine the presence of impairments on activities of daily living (ADL).
Background ADL impairment is a criterion for the diagnosis of dementia associated with PD. The Pill Questionnaire has been recommended as a screening tool to assess ADL impairment in PD patients, but its usefulness and validity have not been fully investigated.
Methods We recruited idiopathic PD patients from 12 hospitals in Taiwan, and the patients underwent clinical assessments, a neuropsychological test battery and the Unified Parkinson Disease Rating Scale evaluation. The Pill Questionnaire was administered by study assistants. Patient and caregiver interviews were performed by experienced neurologists who were blinded to the Pill Questionnaire results.
Results In total, 284 PD patients (mean age 71.8±9 years, mean education 8.7±5.3 years, mean disease duration 5.4±5.3 years) were recruited. 63 patients showed ADL impairment by the Pill Questionnaire, and 108 patients showed ADL impairment by neurologists’ clinical interviews. κ Statistics showed moderate agreement between the two methods (κ=0.521, p<0.001). Of the 108 patients who were diagnosed with ADL impairment by neurologists, only 56 patients (51.9%) showed impairment according to the Pill Questionnaire. Most of the missed patients had milder cognitive impairment and lower motor disability.
Conclusions A comprehensive interview is essential to determine the presence of ADL impairment in PD patients, especially in patients with early PD.
- PARKINSON'S DISEASE
Statistics from Altmetric.com
Significant impairment of activities of daily living (ADL) that cannot be attributed to motor or autonomic symptoms is one of the major criteria for diagnosing dementia associated with Parkinson's disease (PDD).1 According to the concept of mild cognitive impairment (MCI) in Parkinson's disease (PD), cognitive deficits are not sufficient to interfere significantly with functional independence, although subtle difficulties in performing complex functional tasks may be present.2 It is challenging to decide which degree or threshold of ADL impairment represents PD-MCI or PDD.1–4 The Movement Disorder Society Task Force recommended the Pill Questionnaire as a screening tool to assess ADL impairment in PD patients.5 However, the usefulness and validity of the Pill Questionnaire have not been fully investigated.
The Pill Questionnaire is composed of two steps.1 ,5 First, the questionnaire asks patients to describe the doses and intake times of their PD medications. If the patients are able to describe the drugs, doses (milligram or colour of tablet), and timing of treatment spontaneously and clearly, they are considered to have no ADL impairment. For patients who need some help from the examiner, but who are successful in answering without clinically pertinent errors, consultation with a caregiver is required. The caregiver is asked to verify whether the patient can take their medications safely and reliably on their own and to determine the impact of the impairment on ADL. Difficulty describing the treatment regimen and having problems taking PD medications independently, as reported by the caregiver, are considered ADL impairments. However, it is difficult to discriminate ADL impairment due to cognitive decline independent of motor and autonomic symptoms in patients with PD6 or vascular dementia.7
The different levels of ADL must be distinguished.8 Basic daily activities, including bathing, eating and dressing, may be preserved when the first symptoms of cognitive impairment occur, but they may be affected early due to motor dysfunction. In contrast, complex activities, such as organising work, managing finances or using public transportation, are dependent on intact motor, memory, attention and executive functions. Motor symptoms of PD patients may limit complex ADLs even without mental decline.9
To date, there have been limitations in the screening tools used to detect significant ADL impairment in PD patients during clinical diagnosis. Recent studies have used scales that were originally designed for Alzheimer's dementia (AD) to assess ADL impairment in PD patients, including the clinical dementia rating scale (CDR)10–12 and disability assessment for dementia.13 The suitability of using scales designed for AD to assess ADL impairment in PD patients is uncertain. Thus, a newly developed PD Cognitive Functional Rating Scale was validated to measure the impact on ADL in PD patients according to the assessment of ADL impairment by CDR and consensus between researchers.12 In this study, we aimed to compare the clinical judgments of experienced neurologists after interviewing PD patients and their caregivers with the use of the Pill Questionnaire to determine the presence of ADL impairments in clinical practice.
Consecutive idiopathic PD patients with or without cognitive decline were enrolled at 12 teaching hospitals in Taiwan. All subjects had been diagnosed with PD, according to the UK Parkinson's Disease Society Brain Bank criteria.14 Other inclusion criteria were age at onset greater than 50 years, no depressive disorder (Geriatric Depressive Scale15 score <6) and the caregiver was available to provide collateral history. Patients who had developed dementia within 1 year after PD onset or who had other diseases that could impair cognition (eg, stroke, hydrocephalus, epilepsy or major depressive disorder) were excluded. Each patient's age, sex, years of education and length of time since PD diagnosis were recorded. The study protocol was approved by the institutional review boards of each hospital.
The PD stage was evaluated using the modified Hoehn and Yahr (H & Y) Scale.16 ,17 Motor symptoms were assessed using the motor part of the Unified Parkinson's Disease Rating Scale (UPDRS).18 General cognitive functioning was evaluated using the Mini-Mental State Examination (MMSE)19 and Montreal Cognitive Assessment (MoCA).20 ,21 Following the recommendations for the diagnosis for PDD and PD-MCI by the Movement Disorder Task Force,2 ,5 we evaluated four cognitive domains in detail, as follows: attention and working memory (digit span),22 memory (12-item word recall test),23 executive function (category verbal fluency),24 and visuospatial function (5-point scale for cube copying25 and 16-point26 scale for clock drawing). The Pill Questionnaire was administered by study assistants. Patient and caregiver interviews were performed by experienced neurologists who were blinded to the Pill Questionnaire results.
Data analysis and statistics
All statistical analyses were performed using SPSS (V.17.0). The κ statistic was used to evaluate the agreement of ADL impairments as measured by the Pill Questionnaire with those assessed clinically by experienced neurologists. Landis and Koch characterised a κ value <0 as indicating no agreement, and κ values ranging from 0 to 0.20 as slight, 0.21 to 0.40 as fair, 0.41 to 0.60 as moderate, 0.61 to 0.80 as substantial and 0.81 to 1.00 as almost perfect agreement.27 Four subgroups of PD patients were defined according to the Pill Questionnaire and neurologists’ assessment results. To determine the differences between these four subgroups, the median values were examined using non-parametric Kruskal–Wallis tests or χ2 tests. The differences in demographic features and clinical characteristics between the patients who were assessed as having ADL impairments by the Pill Questionnaire and those who were diagnosed with ADL impairments by experienced neurologists were compared using non-parametric Mann–Whitney U tests or χ2 tests and adjusting for multiple comparisons. The significance level was set at p<0.05 initially and p<0.01 after adjusting for multiple comparisons. Results are reported as mean±SD unless otherwise indicated.
We recruited 284 patients (166 men, 118 women) who had been diagnosed with PD in 12 hospitals. Their mean age was 71.8±9 years, mean education 8.7±5.3 years and mean disease duration 5.4±5.3 years. When classified according to the modified H & Y Scale, 45 (15.8%) patients were stage 1.0, 15 (5.3%) were stage 1.5, 70 (24.6%) were stage 2.0, 41 (14.4%) were stage 2.5, 81 (28.5%) were stage 3.0, 22 (7.7%) were stage 4.0 and 10 (3.5%) were stage 5. The mean score of the motor part of the UPDRS was 27.7±15. The mean MMSE score was 23.2±5.6 (range 7–30), and the mean MoCA score was 18.9±6.4 (range 2–30). The mean score for verbal fluency was 10±3.7 (range 0–21). The mean score was 8.6±3.1 (range 1–14) for forward digit span and 4.8±2.8 (range 0–14) for backward digit span. For cube copying, the mean score was 2.4±2.0, and the mean score for clock drawing was 10.4±5.3.
Assessment of ADL impairment
According to the Pill Questionnaire, 63 patients showed impairment of ADL. Among the 63 patients, seven patients were considered to have no ADL impairment according to the neurologists’ clinical interview. In contrast, 108 patients were assessed as having impairment of ADL according to the neurologists’ clinical interview. Of these 108 patients, the Pill Questionnaire only identified ADL impairment in 56 patients (51.9%) (table 1).
Using the κ statistic, there was moderate agreement between the two methods for determining ADL impairment (κ=0.521, p<0.001). Using the neurologists’ judgment of ADL impairment as the gold standard, the sensitivity and specificity of the Pill Questionnaire were 51.9% and 96%, respectively.
The analysis of the differences between the four subgroups of PD patients as assessed by the Pill Questionnaire and the neurologists’ clinical interviews is presented in table 2.
Table 3 compares the demographic features and clinical characteristics between the patient group assessed as showing ADL impairment by both the Pill Questionnaire and the neurologists’ clinical interviews (n=56) and the patient group recognised as having ADL impairment only after an experienced neurologist interviewed the caregiver (n=52).
There were no significant differences between these two patient groups in age, sex, education years, body mass index, disease duration and modified H & Y scale. However, there was a lower score for the motor part of the UPDRS (p<0.001), higher MMSE score (p<0.001), higher MoCA score (p<0.001), higher delayed recall score (p<0.001), higher cube copy score (p=0.007) and higher clock drawing score (p<0.001) in PD patients with ADL impairment only recognised by experienced neurologists, compared with PD patients with ADL impairment recognised by both the Pill Questionnaire and the neurologists’ clinical interviews.
We classified our PD patients into PDD, PD-MCI and PD-normal cognition (PD-NC) groups according to the guidelines of the Movement Disorder Society Task Force2 ,5 and the assessment of significant ADL impairment by neurologists’ interviews. The criteria used for the diagnosis of PD-MCI were the level 1 abbreviated cognitive assessments including MoCA scores that were equal to or less than 2321 and impairment on at least two cognitive domains (1.5 SDs below the mean). Of the total 284 PD patients, 108 patients were diagnosed with PDD. Of the other 176 patients who did not have ADL impairment, 95 patients were diagnosed with PD-MCI. Of the 95 patients diagnosed with PD-MCI, seven patients showed ADL impairment according to the Pill Questionnaire. No ADL impairment was noted by the Pill Questionnaire in the 81 PD-NC patients.
Using the neurologists’ judgment of ADL impairment as the gold standard, our results suggest that the Pill Questionnaire misses approximately a half of probable PDD patients in clinical practice; however, the κ statistics showed moderate agreement between these two approaches. Most of the missed patients had milder cognitive impairment and motor symptoms than those identified by the Pill Questionnaire. A more detailed patient and caregiver interview should be considered to determine ADL impairment in PD patients who have minor motor and cognitive symptoms.
Determining ADL impairment due to mental decline in PD patients is difficult because the contributions of motor dysfunction and cognitive impairment to ADL impairment cannot be separated clearly. One previous study28 investigated the roles of the cognitive and motor components in determining the functional ability of PD patients. The results suggested that motor function is a significant predictor of physical ADL, and cognitive function is significantly correlated with instrumental ADL. The Pill Questionnaire examines the ability to manage medication, which is one type of instrumental activity of daily living, but it does not include all complex ADL.8
The value of the Pill Questionnaire in determining ADL impairment in PD patients is uncertain. One recent study that investigated the role of Pill Questionnaire showed that it was neither sensitive nor specific enough to be a screening or diagnostic tool for PD-MCI.29 According to our study, the Pill Questionnaire may be used to determine ADL impairment in more advanced PD patients, but it is not suitable for use in PD patients with less cognitive or motor impairment.
We reduced the potential bias of determining ADL impairment in PD patients by a single neurologist by using a multi-centre study design. Therefore, the results may represent the current clinical condition in Taiwan, and provide evidence for use of the Pill Questionnaire in clinical practice. However, our study was also subject to certain limitations. First, the number of advanced PD patients was limited. Second, although we excluded patients who developed dementia within 1 year after PD onset, it is possible that we recruited some patients with dementia with Lewy bodies (DLB) because the 1-year rule could not differentiate all patients with PDD from DLB. Third, the influence of drugs on cognition was difficult to evaluate in this study.
In conclusion, the present study showed that clinical assessments of ADL impairments in PD patients by experienced neurologists may diagnose more PDD patients than use of the Pill Questionnaire alone. Most of the missed patients had milder cognitive impairment and lower motor disability than those identified by the Pill Questionnaire. A more comprehensive interview is essential to determine the presence of ADL impairment in PD patients, especially in patients with early PD.
We thank Dr Chia-Lin Lee for his suggestions regarding the statistical analysis.
Contributors W-JL analysed the data and wrote the manuscript; all the coauthors collected the patients, discussed the results and corrected the manuscript; J-LF takes full responsibility for the data, analyses and interpretation, and conduct of the research. She had full access to all of the data and has the right to publish any and all data separate and apart from any sponsor.
Funding This study was funded by the Taiwan Dementia Society (TDS-99-01).
Competing interests None.
Patient consent Obtained.
Ethics approval IRB in each participated hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.