1. Background
Parkinson’s disease (PD) is a neurodegenerative disorder, characterized by rigidity, tremor, bradykinesia, mask-like face and abnormal posture. It usually begins by the ages of 50 - 65 years and causes neuronal dysfunction. The prevalence is estimated to be 100 to 200 cases per 100,000 and the annual incidence is 20 cases per 100,000 [1, 2]. Quality of life (QoL) is a wide spreading concept which encompasses all dimensions of life including health. It has physical, physiological, social and spiritual dimensions [3]. According to the definition of WHO it is “individuals” perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” [4].
Usually, it is called “health-related quality of life” because of its close association with health [5]. In the cases of chronic disease, where the aim of treatment is to increase the level of satisfaction and improve QoL of patients rather than increase survival chance, the use of standard instruments for measuring QoL seems inevitable [5]. The PDQ-39 is a widely used questionnaire with 39 questions measuring the effects of Parkinson’s disease on QoL [6]. This questionnaire has been translated into several languages and utilized to measure QoL of patients with Parkinson’s disease [7-12]. To be useful in clinical research applications, tools for measuring quality of life, as well as other scientific measuring instruments must be relevant, measurable, reliable and valid [13].
2. Objectives
As far as the PDQ-39 is concerned, it has not been standardized in the Iranian context. Thus, this study aims to investigate the reliability and validity of the Persian duplicate of the PDQ-39.
3. Patients and Methods
3.1. Translation and Adaptation
In this cross-sectional study, the English version of the PDQ-39 was translated into Persian by a neurologist, an epidemiologist and a linguist. The Persian translation was in turn translated back into English by a different team including a neurologist, a nurse, and a linguist. After cultural adaptation, 25 patients completed the questionnaire to check comprehensibility, leading to the final Persian edition.
3.2. Sample Size
According to Fayers theory, in order to assess the validity and reliability of QoL questionnaires 100 to 400 patients are required [14].
3.3. Patients
From December 2010 to July 2011 all patients with Parkinson’s disease referred to Shahid Motahari specialist clinic affiliated to Shiraz University of Medical Sciences, included in study. Two patients were excluded because of comorbidities. Both of them suffered stroke and were disable. Finally 104 patients were included. All patients singed an informed consent, initially. Patients were examined by a neurologist and using the Hohen and Yahr index, the stage of the disease was determined. Afterward, the Persian version of the PDQ-39 and SF-36 (a questionnaire designed to assess the general QoL) were completed by patients assisted by a nurse. The questionnaire filled by interview for illiterate patients and patients with severe tremor.
3.4. Instrumrnts
3.4.1. Hohen and Yahr Index
The H and Y index is an acceptable indicator of the severity of Parkinson’s. It consists of five stages, with stage 1 (the least severe) and stage 5 (the most severe) [15].
SF-36 questionnaire is a general tool to assess quality of life. It was designed by Ware et al. consisting of eight dimensions: physical functioning, role limitations due to physical health, bodily pain, general health perceptions, vitality, social functioning, role limitations due to emotional health, and general mental health measures [16].
3.4.2. Parkinson Disease Questionnaire-39 (PDQ-39)
The PDQ-39 is a widely used questionnaire with 39 questions measuring the effects of Parkinson’s disease on QoL under 8 dimensions including mobility (10 questions), activities of daily living (6 questions), emotional wellbeing (6 questions), stigma (4 questions), social support (3 questions), cognitive impairment (4 questions), communication (3 items) and bodily discomfort (3 questions) [6].
3.5. Validity, Reliability and Statistical Analysis
The internal consistency of this instrument was evaluated using Cronbach’s alpha coefficient. Cronbach’s alpha values fall between zero and one and the closer to 1 the higher the reliability. Also, values higher than 0.7 are defined as acceptable and those were higher than 0.8 are excellent [17]. The convergent validity of the two measures was assessed using Spearman's correlation coefficient. The values between 0 and 0.25 indicate a very weak correlation or no correlation at all; those between 0.25 and 0.50 indicate a weak correlation; those from 0.50 and 0.75 show good and medium correlations; and values of 0.75 and above are excellent [18]. In this study, the assumption was that the following pairs would be significantly correlated. Mobility (PDQ-39) and physical function (SF-36); activity of daily living (PDQ-39) and physical role (SF-36); wellbeing (PDQ-39) and mental health (SF-36); social support (PDQ-39) and social function (SF-36); and bodily discomfort (PDQ-39) and bodily pain (SF-36). The discriminant validity was calculated by comparing the H and Y-based scores of QoL at different stages of the disease and using the Kruskal-Wallis test. The hypothesis here was that the questionnaire must discriminate the scores at different stages of the disease.
Construct validity was determined by item-convergent validity and item-discriminant validity. Spearman’s correlation coefficient was used to evaluate item-convergent validity, with a value equal or above 0.4 considered acceptable. Item-discriminant validity was present when an item had a significantly higher correlation with its own dimension, compared with the other dimensions [19]. All statistical evaluations were made assuming a two-sided test based at 0.05 level of significance using SPSS-16.
4. Results
A total of 104 patients contributed to the analysis. The mean ± SD age of participants was 65.0 ± 11.6 years with a male predominance (60.6%). According to H and Y criterion 17 patients (16.3%) in stage I, 40 (38.5%) in stage II, 33 (30.8%) in stage III, 13 (12.5%) in IV and two were in stage V. The demographic and clinical characteristics of participants are presented in Table 1.
| Variable | Value |
|---|---|
| Age, y | 65.00 ± 11.6 |
| Onset of disease | 60.67 ± 12.1 |
| Disease duration | 4.19 ± 4.3 |
| H and Y stage | 2.4 ± 0.9 |
| Sex, No. (%) | |
| Male | 63 (60. 6) |
| Female | 41 (39.4) |
| Marital status, No. (%) | |
| Single | 3 (2.9) |
| Married | 81 (77.9) |
| Separation | 2 (1.9) |
| Widowed | 18 (17.3) |
Demographic and Clinical Characteristics of Study Participantsa
Table 2 shows the value of Coronbach’s α coefficients and Mean ± SD of quality of life score for different dimensions. The patients had most problem in mobility dimension (highest score) and the lowest in social support dimensions (lowest score). Except for stigma dimension, Coronbach’s α coefficients were higher than 0.7 for other dimensions.
| Dimensions | Mean ± SD | Cronbach’s alpha |
|---|---|---|
| Mobility | 57.55 ± 17.0 | 0.92 |
| Activities of daily living | 34.63 ± 21.0 | 0.88 |
| Emotional well-being | 32.04 ± 16.3 | 0.85 |
| Stigma | 48.61 ± 20.9 | 0.64 |
| Social support | 24.51 ± 19.4 | 0.74 |
| Cognition | 33.81 ± 18.2 | 0.84 |
| Communication | 41.83 ± 17.8 | 0.82 |
| Bodily discomfort | 38.41 ± 19.1 | 0.71 |
| PDQ-39 summary index | 44.81 ± 15.6 |
Mean and Cronbach’s α-Coefficients for Each Dimensions of PDQ-39 Questionnaire (N = 104)
Item-convergent validity and item-discriminant validity are presented in Table 3. For item-convergent validity all coefficients were more than 0.4 except in one item. The scores of PDQ-39’s dimensions among persons at different H and Y stages were shown in Table 4.
| Dimensions | Number of Items | Item-Convergent Validity (Range of Correlation) | Item-Discriminant Validity (Range of Correlation) |
|---|---|---|---|
| Mobility | 10 | 0.49 - 0.81 | 0.27 - 0.51 |
| Activity of daily living | 6 | 0.56 - 0.83 | 0.25 - 0.49 |
| Emotional well-being | 6 | 0.65 - 0.82 | 0.20 - 0.38 |
| Stigma | 4 | 0.38 - 0.62 | 0.22 - 0.46 |
| Social support | 3 | 0.58 - 0.79 | 0.29 - 0.50 |
| Cognition | 4 | 0.52 - 0.68 | 0.31 - 0.47 |
| Communication | 3 | 0.58 - 0.72 | 0.23 - 0.41 |
| Bodily discomfort | 3 | 0.44 - 0.59 | 0.26 - 0.34 |
Item Scaling Tests: Convergent and Discriminant Validity for PDQ-39 Dimensions
| Dimensions | Hohen and Yahr Scale | P Value | |||
|---|---|---|---|---|---|
| Stage 1 (N = 17) | Stage 2 (N = 40) | Stage 3 (N = 32) | Stage 4, 5 (N = 15) | ||
| Mobility | 24.00 ± 15.0 | 28.00 ± 11.4 | 42.75 ± 14.4 | 55.76 ± 11.4 | 0.0001 |
| Activities of daily living | 16.44 ± 9.0 | 31.56 ± 16.3 | 42.75 ± 14.4 | 62.15 ± 17.4 | 0.0001 |
| Emotional well-being | 31.77 ± 11.3 | 39.37 ± 12.1 | 50.08 ± 15.0 | 60.98 ± 14.7 | 0.0001 |
| Stigma | 25.00 ± 16.0 | 41.09 ± 15.3 | 50.00 ± 15.4 | 67.94 ± 11.0 | 0.0001 |
| Social support | 28.88 ± 11.8 | 38.75 ± 15.3 | 54.50 ± 19.6 | 64.70 ± 11.7 | 0.0001 |
| Cognition | 28.33 ± 10.9 | 40.46 ± 14.5 | 55.50 ± 17.4 | 65.00 ± 11.0 | 0.0001 |
| Communication | 40.47 ± 12.1 | 44.79 ± 15.3 | 57.87 ± 13.9 | 66.27 ± 17.3 | 0.0001 |
| Bodily discomfort | 31.11 ± 20.1 | 37.91 ± 16.9 | 57.43 ± 15.3 | 58.64 ± 16.8 | 0.0001 |
| PDQ-39 summary index | 27.28 ± 10.3 | 35.80 ± 9.8 | 53.00 ± 15.6 | 61.38 ± 11.3 | 0.0001 |
Mean and Significance Level of Dimensions of PDQ-39 at Different Stages of Diseasea
The quality of life score was positively associated with the stage of disease. That is patients with the higher stage had lower QoL score. Convergent validity assessed by calculating spearman correlation coefficient between different dimensions of PDQ-39 and SF-36. There were acceptable correlations between the same dimensions of two questionnaires. The mobility, activity of daily living, emotional well-being, social support and bodily discomfort of PDQ-39 questionnaire were considered as equal to physical function, physical role, mental health, social function and bodily pain dimensions of the SF-36 questionnaire respectively.
5. Discussion
In order to be beneficial in clinical research, an instrument requires acceptable reliability and validity. In this study, found that the Persian version of the PDQ-39 had a very good internal consistency. Cronbach’s alpha coefficient varied between 0.64 and 0.92 except for stigma dimension (Cronbach’s alpha = 0.64). This may be due cultural probsems and adapting appropriate coping strategies. It was similar to Peto et al.’s study on the original questionnaire [20].
Cronbach’s alpha coefficient for physical discomfort and social support dimensions were 0.71 and 0.74, respectively which were acceptable. Findings were similar to the other studies carried out in other parts of the world. Cronbach’s correlation coefficients for some other countries varied between 0.69 to 0.94 in South Korea [7], 0.63 to 0.94 in Spain [9], 0.69 to 0.94 in UK [20], 0.48 to 0.91 again in China [21], 0.64 to 0.90 in Singapore [8] and 0.51 to 0.96 in the USA [12]. Comparing to these studies, the coefficients obtained are acceptable.
There were acceptable correlation between 38 out of 39 questions and their related dimensions (Table 3). Also the correlation between each of the questions and the other dimensions was lower than the correlation between their related ones. These demonstrate that each question is able to measure its dimensions and also to discriminate the other ones. The correlation was less than 0.4 in the case of question numbered 25 in stigma dimension.
It seems that this is due to cultural maladjustment. May be it is because of different concept and viewpoints in Iranian patients.
According to the result of the present study the higher stage of Parkinson’s the higher the score of QoL reflected the lower QoL. The mean of scores on all dimensions of the questionnaire is higher at higher stages, which certainly indicates the lower QoL in patients with more severe stages of Parkinson’s. This finding sheds light on the discriminant power of the PDQ-39. These results can be compared to a study by Zhang et al. in China [21]. The scores of the questionnaire’s summary and the dimensions of mobility and activities of daily living were significantly different. In another study in China by Luo et al. the correlation was significant only in relation with mobility, activities of daily living, emotional health and social support [22]. However, in our study the relationship was significantly different in all aspects. In Martinez-Martin and Frades Payo’s study in Spain [9], there was a significant difference between the score on QoL in all dimensions and the score on the PDQ-36’s summary at the disease’s stages based on the H and Y index.
Dimensions of mobility, activities of daily living, emotional wellbeing, social support, and physical discomfort in the PDQ-39 correlated to an acceptable extent to the dimensions of physical function, physical role, mental health, social functioning, and bodily pain, respectively, in the SF-36. Despite these, some dimensions of the former, namely cognitions, stigma and cognitions bore no significant correlation with any dimension in the latter (Table 5). Correlation coefficients were all negative as the higher scores of the PDQ-39 and the lower scores of the SF-36 signify lower QoL. The results of the current study are consistent with those of Luo et al.’s [22] and Zhang and Chan’s in China [21].
| PDQ-39 dimensions | SF-36 Dimension | |||||||
|---|---|---|---|---|---|---|---|---|
| Physical Function | Role-Physical | Bodily Pain | General Health | Social Function | Vitality | Role-Emotional | Mental Health | |
| Mobility | -0.72a | -0.54a | -0.36 | -0.40b | -0.47b | -0.42 | -0.41 | -0.32 |
| Activities of daily living | -0.52a | -0.69a | -0.18 | -0.26 | -0.55a | -0.51 | -0.37 | -0.30 |
| Emotional well-being | -0.22 | -0.36b | -0.28 | -0.25 | -0.46a | -0.68 | -0.33 | -0.68a |
| Stigma | -0.28 | -0.29 | -0.24 | -0.18 | -0.45b | -0.38b | -0.22 | -0.32 |
| Social support | -0.30b | -0.27 | -0.35b | -0.27 | -0.71a | -0.51a | -0.39b | -0.36b |
| Cognition | -0.33b | -0.32b | -0.21 | -0.27 | -0.42b | -0.43 | -0.24 | -0.21 |
| Communication | -0.41b | -0.31b | -0.15 | -0.34 | -0.56a | -0.44a | -0.35 | -0.38a |
| Bodily discomfort | -0.46b | -0.32b | -0.67a | -0.37b | -0.34 | -0.38b | -0.13 | -0.16 |
| PDQ-39 summary index | -0.59a | -0.56a | -0.39b | -0.45b | -0.68a | -0.47b | -0.42b | -0.46a |
Spearman’s Rank Correlation Between PDQ-39 and SF-36 Dimensions
In conclusion, the results of this study demonstrated that the validity and reliability of the Persian version of the PDQ-39 are acceptable, and it can thus be recommended as a self-reported questionnaire for measuring the QoL of patients with Parkinson’s disease. We hope that this instrument can be easily used by researchers and clinicians to better evaluate the QoL of patients with Parkinson. Thereby enhancing their health status and wellbeing the authors would be glad to provide the enthusiasts with this instrument without charge.
