|Year : 2020 | Volume
| Issue : 1 | Page : 59-71
Development and validation of quality of life in Parkinson's disease instrument
Rajeev Aggarwal1, Vinay Goyal1, Ravindra Mohan Pandey2, Nand Kumar3, Sumit Singh4, Garima Shukla1, Madhuri Behari1
1 Department of Neurology, AIIMS, New Delhi, India
2 Department of Biostatistics, AIIMS, New Delhi, India
3 Department of Psychiatry, AIIMS, New Delhi, India
4 Artemis Hospital, Gurugram, Haryana, India
|Date of Web Publication||21-Jan-2020|
Dr. Madhuri Behari
Department of Neurology, Fortis Hospital, Vasant Kunj, New Delhi - 110 029
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Context: Parkinson's disease (PD) has devastating effects on quality of life (QoL), but there is no instrument that has been developed for Hindi-speaking persons with Parkinson's disease (PWP). Objective: The objective of this study was to develop and validate an instrument in Hindi language to measure health-related QoL (HRQoL) in PWP. Subjects and Methods: Literature review and interviews of stakeholders were done to create a pool of 68 items to develop a questionnaire. Self-rated global QoL item was also included in the questionnaire. Questionnaire was tested on 300 Hindi-speaking PWP. Item reduction was achieved through factor analysis and clinimetrics to finalize the QoL in PD (QLPD) instrument. Validity and reliability of the QLPD were tested. Results: “QLPD” is a 45-item instrument with nine subscales, namely, activities of daily living, mobility, psychological, fear, social, family, treatment, finance, and nonmotor symptom subscales. Internal consistency of QLPD's summary score and all subscales except treatment subscale was high (α = 0.74–0.94). Intraclass correlation coefficient between summary score and global QoL was 0.79. Summary score and subscale scores were significantly different (P < 0.0001) for predefined five categories on global QoL (very good to very bad). QLPD subscales exhibited good convergent and divergent validity with subscales of 39-item PD questionnaire and short form-36 scale. Higher Hoehn and Yahr stage, lower monthly per capita income, and higher levodopa equivalent daily dosage were found to be independently associated with poor HRQoL. Conclusion: QLPD is a valid and reliable instrument to measure HRQoL in Hindi-speaking PWP.
Keywords: Development, Parkinson's disease, quality of life, questionnaire, validation
|How to cite this article:|
Aggarwal R, Goyal V, Pandey RM, Kumar N, Singh S, Shukla G, Behari M. Development and validation of quality of life in Parkinson's disease instrument. Ann Indian Acad Neurol 2020;23:59-71
|How to cite this URL:|
Aggarwal R, Goyal V, Pandey RM, Kumar N, Singh S, Shukla G, Behari M. Development and validation of quality of life in Parkinson's disease instrument. Ann Indian Acad Neurol [serial online] 2020 [cited 2020 Feb 24];23:59-71. Available from: http://www.annalsofian.org/text.asp?2020/23/1/59/257659
| Introduction|| |
Parkinson's disease (PD) is the second most common neurodegenerative disorder of the elderly creating devastating burden on patients, caregivers, and the state with substantial reduction in patients' quality of life (QoL).,,, Many generic and disease-specific QoL scales are in use for persons with Parkinson's disease (PWP). Generic scales lack disease-specific items and have limited sensitivity and responsiveness as compared to disease-specific health-related QoL (HRQoL) scales. Thirty nine –item Parkinson's Disease Questionnaire (PDQ-39); Parkinson' s Disease Quality of Life Questionnaire; Parkinson's Disease Quality of Life Scale, and Scales for Outcomes in Parkinson's Disease-Psychosocial are widely used disease-specific HRQoL scales for PWP. Educational, socioeconomic, and cultural diversity in developing countries such as India poses challenges in adopting and translating these scales that have been developed for the Western population., Den Oudsten et al. have criticized these scales pronouncing them health status scales and not HRQoL scales. These scales lack domains on financial implication, efficacy of current treatment, and nonmotor symptoms (NMS). Keeping these aspects in mind, we present the development of Quality of Life in Parkinson's disease (QLPD) instrument in Hindi language, in the present communication.
| Subjects and Methods|| |
This study was done at a tertiary care teaching hospital in India after approval of IRB/EC of the hospital. We developed QLPD, following the principles laid down for patient-reported outcome measures.,,
Item pool generation
Literature review and semi-structured interviews of stakeholders with native language as Hindi created the item pool in Hindi language. After obtaining informed consent, 28 cognitively preserved (Mini–Mental State Examination score >24) PWP diagnosed on the basis of UK PD Brain Bank Clinical Diagnostic Criteria without any other disabling disease, six caregivers of PWP, eight movement disorder specialists, and one psychiatrist were interviewed by the same researcher (RA). The interviews were audio recorded with due permission and consent. The demographics; age, gender, duration of symptoms, family type (joint or nuclear), disease severity (as per Hoehn and Yahr stage) and socio-economic status (as per Kuppuswamy scale) of recruited PWP were recorded.
An expert panel consisting of four neurologists, one physiotherapist, one biostatistician, and one psychiatrist conceptualized and defined HRQoL in PD and identified 68 items from content analysis of interviews and literature review after several levels of deliberations to develop a questionnaire. An item on person's overall self-perception of their QoL (global QoL) based on 0–100 visual analog scale with five categories was also added in the questionnaire [Appendix 1]. Recall period for answering was kept at 4 weeks. (All items were translated from Hindi to English language for publication by translators fluent in English language: [Appendix 1].
Pilot testing of the questionnaire was done on 19 PWP who were recruited to fill the questionnaire followed by cognitive interview for its face validity, content validity, language, format, and options. Four neurologists and a psychiatrist also reviewed the questionnaire. At this stage, no item was considered for removal.
After pilot testing, the questionnaire was subjected for field testing on 300 consecutive PWP from the the same institution. Demographics, per capita monthly income (PCMI) and Levodopa equivalent daily dosage (LEDD) of these PWP were recorded. In addition to the questionnaire, all participants were requested to fill either PDQ-39 or 36-item short-form scale (SF-36). PDQ-39 is most commonly used PD specific scale to assess HRQoL in PWP. PDQ-39 contains 8 domains: Mobility, Activities of daily living, Emotional well being, Stigma, Social support, Cognitions, Communication and Bodily discomfort. Items of PDQ-39 assess how often PWP experiences difficulty using 5 point ordinal range where 0 means never and 4 means always. Each domain provides a domain score ranging from 0 to 100. Lower score reflects better QoL. SF-36 is most commonly used generic HRQoL for PWP that contains 36 items in 8 subscales: Physical function (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (VT), social function (SF), role emotion (RE) and mental health (MH). For each subscale, items are coded, summed and transformed on to a scale from 0 (worst HRQoL) to 100 (Best HRQoL) using a software provided by Optum® outcomes. SF-36 also provides two summary scores viz. Physical component summary score (PCS) and Mental component summary score (MCS). Scores are norm based: mean (SD) = 50 (10). Score interpretation: 50 is average; 0 to 49 are below average and 51 to 100 are above average (higher the score better will be QOL).
Proxies were used for collection of response in participants in advanced stage, who could not respond reliably due to infirmity/severe tremor. Questionnaires were reviewed for missing responses and participants were encouraged to answer all items before leaving.
Item reduction and construction of scales
Item-level analysis of data from field testing was done to find floor and ceiling effects and mean response of each item. We calculated impact of each item by multiplying mean response of the item with frequency of that item. Exploratory factor analysis of 68 items was done. Domains with eigenvalue >1 were retained and subjected to maximum likelihood method, followed by orthogonal varimax rotation to improve loading of items around domains. Items that loaded more than 0.4 were considered relevant. Before dropping an item, the remaining items were reviewed in relation to content validity of the questionnaire. The fate, frequency, mean response, and impact of all 68 items have been shown in [Appendix 1]. Extracted domains were subjected to confirmatory factor analysis to assign subscales. These subscales and global QoL item constituted QLPD [Appendix 2]. Cronbach's alpha (α) and item-scale correlation were used for internal consistency. Subscale scores and summary score were created on 0–100 scale with higher score representing poor HRQoL as follows:
- Subscale score = aggregate of observed scores of items in the subscale/maximum possible total score of that subscale × 100
- Summary score = aggregate of observed scores of items in all subscales/maximum possible total score of all subscale × 100.
Single-item QoL scales are valid and reliable tool for measuring QoL,, so considering global QoL, a close approximation of true QoL, we did intraclass correlation (ICC) between summary score and global QoL score. For discriminative validity, we compared mean summary score and subscale scores for five categories of participants on the basis of global QoL (0–20, very bad; 21–40, bad; 41–60, neither good nor bad; 61–80, good; and 81–100, very good) using one-way analysis of variance. Multitrait–multimethod correlation matrix was used for convergent and divergent (discriminant) validity of QLPD subscales with subscales of PDQ-39 and SF-36. We attributed strong correlation when Spearman's correlation coefficient (ρ) was ≥0.6 between two subscales, moderate correlation for 0.4 ≤ ρ < 0.6, while low correlation for ρ < 0.4. Strong correlation coefficients between similar subscales of QLPD and PDQ-39 and/or SF-36 were considered evidence of convergent validity while divergent validity was assumed when a subscale of QLPD had the highest correlation with similar subscales of PDQ-39 and/or SF-36, i.e., lower correlation with other unrelated subscales of PDQ-39 and/or SF-36. We did regression analysis to determine the effect of demographics and clinical severity on HRQoL. For this purpose, we divided participants into two groups Group 1: Hoehn and Yahr [H and Y] stage ≤ 2.5; and Group 2: H and Y stage ≥3. We did multivariate linear regression of summary score as dependent variable and age at onset of symptoms, duration of symptoms, H and Y stage, PCMI, and LEDD as explanatory variables. STATA version 9.0 (Statacorp, 4905 Lakeway drive, College station, Texas 77845 USA) and SPSS version 16.0 (SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6412 USA) software were used for statistical analysis.
| Results|| |
Demographic summary of PWP who participated in the study is shown in [Table 1]. Mean (range) clinical experience of nine interviewed specialists was 19.2 (8–32) years for treating PWP. The duration of interviews ranged from 15 to 45 min.
|Table 1: Characteristics of persons with Parkinson's disease who participated in the development of quality of life in Parkinson's disease|
Click here to view
In the pilot testing of initial questionnaire, face validity and content validity of the questionnaire were graded as very good as the stakeholders did not suggest any new item. From 300 PWP for field testing, five treatment-naive persons were excluded from the study, as questions regarding treatment satisfaction and/or expenditure were irrelevant for them. Final analysis was done on data collected from 295 PWP. Out of 295 PWP, 136 also filled PDQ-39 scale while 114 filled SF-36 scale. Item-level analysis of data revealed that only 0.85% of data was missing which was replaced with item's mean value. The mean response of all items was in the range of 0.18–2.88 (maximum possible = 4) while the impact of items ranged from 0.01 to 2.11 (maximum possible = 4). Exploratory factor analysis of 68 items yielded eight domains with eigenvalue >1 that accounted for 80.3% variance of the data. After applying maximum likelihood with eight domains and orthogonal varimax rotation, 41 items loaded >0.4 in seven domains. These domains were motor, financial, fear and social, psychological, NMS, treatment, and family with variance 46%, 10%, 6%, 5%, 4%, 3%, and 3%, respectively. None of the items in the 8th domain reached benchmark of 0.4 and accounted for only 2.9% variance and so it was dropped. Of the remaining 27 items with loading <0.4, 11 causal items on tremor, family disturbance, consistency of effects of medicine, constipation, fatigue, sleep, body aches, sensory symptoms, swallowing, restless leg syndrome, and sexual satisfaction were retained, as these do not follow psychometric principles. These 11 items were distributed among seven domains on the basis of their face validity. The remaining 16 items were deleted as they neither were causal items nor loaded >0.4. This left us with 52 items. To further reduce items, seven more items on social embarrassment, family responsibilities, spat in family, dependency on medication, vision-related problems, and weird dreaming were removed as they were either not applicable to large population, or their content was already represented in other items in different domains. Hence, the final questionnaire contained 45 items.
Confirmatory factor analysis of financial, psychological, treatment, and family domains revealed single dimensionality. Motor domain was divided into activities of daily living (ADL) and mobility domain. 'Fear and social' domain was divided into two separate social and fear domains. NMS domain did not show single dimensionality on confirmatory factor analysis as it consisted of causal items. We decided to retain all NMS items, though they did not comply with factor analysis and an item on overall health satisfaction into NMS domain. These nine domains formed nine subscales. Finally, we had 45 items in nine subscales plus one global QoL which constituted QLPD [Appendix 2].
Internal consistency (reliability) and scoring of quality of life in Parkinson's disease
Cronbach's alpha (α) of QLPD was 0.94 while α of eight subscales was in the range of 0.74–0.91 [Table 2]. Item-scale correlations of subscales found only four items that did not correlate with their subscales as α improved after removing these items. These four items were retained considering content validity. Subscale scores were widely distributed and had full range of scores from 0 to 100 in all subscales except NMS (0–86) [Table 2]. Mean (standard deviation [SD], range) summary score of QLPD (n = 295) was 35.3 (16.6, 1–83). Reliability testing of summary score with global QoL revealed ICC coefficient, 0.79 (95% confidence interval [CI]; 0.74–0.83; P < 0.0001). Discriminant validity was demonstrated as mean summary score and subscale scores had statistically significant difference (F(4, 290) =55.1 P < 0.0001) across five global QoL categories [Table 3].
|Table 2: Subscales of quality of life in Parkinson's disease instrument: Internal consistency and their scores tested on 295 persons with Parkinson's disease|
Click here to view
|Table 3: Mean summary score and subscale scores across the global quality of life categories of quality of life in Parkinson's disease inst|
Click here to view
Multitrait–multimethod correlation matrix showed good convergent and divergent validity. [Table 4] shows the Spearman's correlation coefficient among subscales of the QLPD (n = 295), QLPD versus SF-36 (n = 114), and QLPD versus PDQ-39 (n = 136). For convenience, the subscales of QLPD and PDQ-39 were prefixed with q and p, respectively. All subscales of QLPD had low-to-moderate correlation with other subscales of QLPD except qADL with qmobility and qNMS, and qsocial with qfear. Summary score had high correlation with global QoL. In regression analysis model, univariate analysis of age at onset of symptoms, duration of symptoms, PCMI, and LEDD with summary score as dependent variable explained variance of 2%, 8%, 12%, and 14%, respectively. Younger age at onset (β: −0.23 [95% CI: −0.41–−0.06]; P < 0.05), longer duration of symptoms (β: 0.88 [95% CI: 0.53–1.22]; P < 0.001), lower PCMI (β: −0.68 [95% CI: −0.89–−0.47], P < 0.001), and higher LEDD (β: 0.03 [95% CI: 0.02–0.04]; P < 0.001) were associated with higher summary score (poor HRQoL). Clinical severity of disease (H and Y stage) explained 18% variance. Mean (SD) summary scores for Group 1 (H and Y ≤2.5) and group 2 (H and Y ≥3) were 30 (14.8) and 44.7 (15.7), respectively. Higher staging of H and Y stage (β: 14.7 [95% CI: 11.09–18.34], P < 0.001) revealed poor HRQoL. In combined model, to know the significant determinant of HRQoL, multivariate linear regression showed that higher H and Y stage, lower PCMI, and higher LEDD (adjusted R2 = 0.37) were associated with poorer HRQoL.
|Table 4: Spearman's correlation coefficient “ρ” between subscales of quality of life in Parkinson's disease instrument, Parkinson's disease questionnaire-39, and short form-36|
Click here to view
| Discussion|| |
We developed QLPD instrument, a disease-specific HRQoL instrument for Hindi-speaking Indian PD population. From literature review and interviews of stakeholders involved in the care of PWP, comprehensive coverage of all relevant items related to HRQoL was ensured, but item reduction was a challenge. It was difficult to ascertain the cutoff line for inclusion of items on the basis of impact. Second, many items on NMS and falls had low impact due to the lower prevalence in our population due to the differential representation of various stages, but when present, they significantly and adversely affected QoL. Keeping content validity in mind, clinimetrics based on clinical sensibility; and psychometrics based on factor analysis guided us to finalize items. We made it a point to include patients of varied background and all severity of disease, so that the instrument would be relevant to all categories of Indian PWP. Our sample for field testing was a heterogeneous group of PWP, including persons of both gender, wide age range (23–85 years), wide duration of illness (3 months–32 years), and wide severity of PD as gauged by H and Y stage of disease (1-5).
QLPD is comprehensive in approach with all relevant domains of HRQoL in PWP. QLPD gives emphasis on NMS in addition to motor function as it has 13 items on NMS while 10 items on motor functions. This makes it more relevant and holistic as it takes in account both motor and NMS to measure QoL, since past studies,, have shown higher impact of NMS on HRQoL.
Treatment and finance are two unique subscales in QLPD. Successful management of symptoms and meeting the patients' expectations has direct relationship with patients' HRQoL. Regularly, new technologies for the management of PD are emerging. HRQoL has been recommended as an important end point for efficacy measurement of any new treatment strategy; therefore, it is required that PD-specific HRQoL instruments should be sensitive to treatment efficacy as well as patients' satisfaction. Financial implication of the disease is of paramount importance for PWP, especially in low-income countries in the absence of medical insurance. The direct cost of treatment due to expenditure on medications, consultations, and visits to the hospital as well as indirect costs in the form of loss of wages of self and caregiver also impact overall well-being. Lower PCMI was found to be associated with poor HRQoL in this study. Similar results have been published by Ray et al.
We chose 5-point Likert scale as it covers good range of possible responses with adequate scope for change in score with change in status of the patient while keeping a balance between precision and responsiveness. To develop subscale scores and summary score, we did not use weights as it has not been shown to be better method than nonweighted system.
The internal consistency of QLPD and its subscales except treatment subscale was very good. Low internal consistency of treatment subscale may be attributed to its three items with wide-ranging content. The presence of ceiling or floor effects indicates that scale will have poor discrimination resulting in reduced sensitivity and responsiveness. Summary score and subscale scores in our cohort of PWP were widely distributed without any floor or ceiling effect.
Validity of a scale implies ability of the scale to measure what it intends to measure. An ICC of 0.79 between global QoL and summary score in QLPD supports the patient's overall perceived QoL which is highly correlated with estimated QoL from nine subscales. Interpretability of the QLPD is good as it provides a scale profile on 0–100 scales that includes a summary score, nine subscale scores, and one global QoL item.
Construct validity of quality of life in Parkinson's disease and its subscales
The convergent validity of QLPD's subscales with similar subscales and divergent validity with unrelated subscales of SF-36 and PDQ-39 were good. ADL subscale of QLPD had high correlation with mobility, family and NMS subscales of QLPD, and ADL and mobility subscales of PDQ-39. High correlation between ADL and mobility subscales may be attributed to items of these subscales which are largely motor features whereas high correlation between ADL and NMS subscales of QLPD may be due to parallel deterioration of motor as well as NMS with advancing disease. Fulfilling family responsibilities largely depends on patient's ability to move and ability to do household chores and/or occupational works that are primarily dependent on physical abilities that may explain high correlation of ADL subscale with family subscale. Psychological subscale of QLPD has high correlation with emotion subscale of PDQ-39, mental health, and mental component summary of SF-36 that primarily consists of items on anxiety and depression, hence highly correlated to each other. Fear subscale of QLPD had high correlation with emotion and stigma subscales of PDQ-39 and mental health subscale of SF-36. Patients who are apprehensive and worrying in nature (neuroticism trait) curtail their social participation affecting their social life. The common elements among these subscales were fearfulness, worrying nature, stigma, and emotional issues.
Treatment and finance subscales contributed around 14% of total variability in the data and had low-to-very low correlations with most of the subscales of QLPD, PDQ-39, and SF-36. The items of these two subscales are unique and are not available in other scales. In QLPD, global QoL item is moderately (0.4–0.6) correlated with all subscales of QLPD which suggests that all subscales of QLPD contribute to overall QoL in PD.
The development of QLPD had many limitations and challenges. Being a single-center study has its own strength, as it ensures uniformity of data collection and assessment. The number of participants in Stage 5 was low (3%) as patients who are very disabled do not like to visit hospitals due to obvious reason. This was observed in many of the studies published earlier., Despite these limitations, QLPD is the first disease-specific HRQoL instrument that has been developed after defining HRQoL and conceptualization of QoL and does not measure just frequency and severity of different symptoms but actually measures the perception of persons about impact of the disease on their lives. QLPD is an improvement over other known HRQoL questionnaires as it has finance, treatment, family, and NMS subscales that have not been considered in earlier PD-specific scales. QLPD also includes global QoL item that is instrumental in estimation of patient's self-perception about his/her overall QoL.
| Conclusion|| |
We have developed QLPD, a new HRQoL instrument for Hindi-speaking PWP. QLPD has good internal consistency, construct validity, precision of scores, and interpretability. It will be useful if it can be translated into other Indian languages and tested in PWP conversant with those languages.
The authors would like to thank all the participant persons with PD, their caregivers, and specialists (Dr. CS Aggarwal MD, DM, Dr. Meena Gupta MD, DM, Dr. SS Raghavan, MD, DM, and Dr. KS Anand, MD, DM, from both public and corporate tertiary care hospital settings of New Delhi). We are thankful to QualityMetric Incorporated, an OptumInsight company, for permitting us to use 36-item Short-Form Questionnaire (Hindi version) free of cost.
Financial support and sponsorship
This was a self-funded study.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Agrawal V, Goyal V, Shukla G, Behari M. Predictors of caregivers' burden of Parkinson's disease in India: Experience of a tertiary care centre in India. J Parkinsonism Restless Legs Syndr 2012;2:59-65.
Whetten-Goldstein K, Sloan F, Kulas E, Cutson T, Schenkman M. The burden of Parkinson's disease on society, family, and the individual. J Am Geriatr Soc 1997;45:844-9.
Dowding CH, Shenton CL, Salek SS. A review of the health-related quality of life and economic impact of Parkinson's disease. Drugs Aging 2006;23:693-721.
Spottke AE, Reuter M, Machat O, Bornschein B, von Campenhausen S, Berger K, et al.
Cost of illness and its predictors for Parkinson's disease in Germany. Pharmacoeconomics 2005;23:817-36.
Peto V, Jenkinson C, Fitzpatrick R, Greenhall R. The development and validation of a short measure of functioning and well being for individuals with Parkinson's disease. Qual Life Res 1995;4:241-8.
de Boer AG, Wijker W, Speelman JD, de Haes JC. Quality of life in patients with Parkinson's disease: Development of a questionnaire. J Neurol Neurosurg Psychiatry 1996;61:70-4.
Welsh M, McDermott MP, Holloway RG, Plumb S, Pfeiffer R, Hubble J. Development and testing of the Parkinson's disease quality of life scale. Mov Disord 2003;18:637-45.
Marinus J, Visser M, Martínez-Martín P, van Hilten JJ, Stiggelbout AM. A short psychosocial questionnaire for patients with Parkinson's disease: The SCOPA-PS. J Clin Epidemiol 2003;56:61-7.
El Rhazi K, Nejjari C, Serhier Z, Tachfouti N, Berraho M, Zakaria Y, et al.
Cross-cultural adaptation difficulties in health quality of life scales for developing countries: Example of St-George respiratory questionnaire validation in morocco. Rev Epidemiol Sante Publique 2009;57:179-89.
Molzahn AE, Kalfoss M, Schick Makaroff K, Skevington SM. Comparing the importance of different aspects of quality of life to older adults across diverse cultures. Age Ageing 2011;40:192-9.
Den Oudsten BL, Van Heck GL, De Vries J. Quality of life and related concepts in Parkinson's disease: A systematic review. Mov Disord 2007;22:1528-37.
Guyatt GH, Cook DJ. Health status, quality of life, and the individual. JAMA 1994;272:630-1.
Gill TM, Feinstein AR. A critical appraisal of the quality of quality-of-life measurements. JAMA 1994;272:619-26.
Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinson's disease: A clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181-4.
Kumar N, Gupta N, Kishore J. Kuppuswamy's socioeconomic scale: Updating income ranges for the year 2012. Indian J Public Health 2012;56:103-4. [Full text]
Aggarwal R, Goyal V, Pandey RM, Kumar N, Singh S, Shukla G, et al.
What should constitute a health related quality of life scale for Parkinson's disease? J Clin Diagn Res 2016;10:OC35-9.
Ware JE Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473-83.
Segovia J, Bartlett RF, Edwards AC. An empirical analysis of the dimensions of health status measures. Soc Sci Med 1989;29:761-8.
Yohannes AM, Dodd M, Morris J, Webb K. Reliability and validity of a single item measure of quality of life scale for adult patients with cystic fibrosis. Health Qual Life Outcomes 2011;9:105.
Rothman ML, Beltran P, Cappelleri JC, Lipscomb J, Teschendorf B, Mayo/FDA Patient-Reported Outcomes Consensus Meeting Group. Patient-reported outcomes: Conceptual issues. Value Health 2007;10 Suppl 2:S66-75.
Behari M, Srivastava AK, Pandey RM. Quality of life in patients with Parkinson's disease. Parkinsonism Relat Disord 2005;11:221-6.
Hinnell C, Hurt CS, Landau S, Brown RG, Samuel M; PROMS-PD Study Group. Nonmotor versus motor symptoms: How much do they matter to health status in Parkinson's disease? Mov Disord 2012;27:236-41.
Li H, Zhang M, Chen L, Zhang J, Pei Z, Hu A, et al.
Nonmotor symptoms are independently associated with impaired health-related quality of life in Chinese patients with Parkinson's disease. Mov Disord 2010;25:2740-6.
Uitti RJ. Treatment of Parkinson's disease: Focus on quality of life issues. Parkinsonism Relat Disord 2012;18 Suppl 1:S34-6.
Winfree KN, Pretzer-Aboff I, Hilgart D, Aggarwal R, Behari M, Agrawal SK, et al.
The effect of step-synchronized vibration on patients with Parkinson's disease: Case studies on subjects with freezing of gait or an implanted deep brain stimulator. IEEE Trans Neural Syst Rehabil Eng 2013;21:806-11.
Martinez-Martin P, Deuschl G. Effect of medical and surgical interventions on health-related quality of life in Parkinson's disease. Mov Disord 2007;22:757-65.
Ray J, Das SK, Gangopadhya PK, Roy T. Quality of life in Parkinson's disease – Indian scenario. J Assoc Physicians India 2006;54:17-21.
Fayers PM, Machin D. Quality of Life: Assessment, Analysis, and Interpretation. Chichester, New York: John Wiley; 2000.
Jenkinson C. Why are we weighting? A critical examination of the use of item weights in a health status measure. Soc Sci Med 1991;32:1413-6.
Dubayova T, Nagyova I, Havlikova E, Rosenberger J, Gdovinova Z, Middel B, et al.
Neuroticism and extraversion in association with quality of life in patients with Parkinson's disease. Qual Life Res 2009;18:33-42.
[Table 1], [Table 2], [Table 3], [Table 4]