Validation of Five Short Versions of the Geriatric Depression Scale in the Elder Population in Taiwan

Transcription

1 156 Taiwanese Journal of Psychiatry (Taipei) Vol. 28 No Original Article Validation of Five Short Versions of the Geriatric Depression Scale in the Elder Population in Taiwan Wei-Chih Chin, M.D. 1,2 Chia-Yih Liu, M.D. 1,2,3 Chin-Pang Lee, M.D. 1,2 Chun-Lin Chu, M.D. 1,2* Objectives: The objective of the present study was to examine the feasibility of the use of five different versions of 4-items and 5-items Geriatric Depression Scale short form in the ethnic Chinese elderly population in Taiwan. Methods: We used the data bank from our previous epidemiological study. We used the data of 388 subjects and compared the scores of GDS-15, two GDS-5, and two GDS-4. Results: All of the GDS short forms showed good correlations to each other. GDS- 15 showed the highest area under receiver operating characteristic curve (AUC = 0.828), followed by GDS-5 (2006 Molloy et al. s version) (AUC = 0.816), GDS-5 (1999 Hoyl et al. s version) (AUC = 0.816), GDS-4 (1994 D Ath et al. s version ) (AUC = 0.756), and GDS-4 (1995 van Marwijk et al. s version ) (AUC = 0.741). Conclusions: All versions of short form GDS can be used in screening major depressive episode among older Taiwanese adults. Key words: geriatric depression, DSM, sensitivity, specificity, Taiwanese (Taiwanese Journal of Psychiatry [Taipei] 2014; 28: ) Introduction The Geriatric Depression Scale (GDS) is a useful screening tool for detecting geriatric depression. The 30-item of yes/no questionnaire is a valid instrument for assessing depressive symptoms in the geriatric populations [1, 2]. Since the GDS is a screening, not a diagnostic tool, convenience is an important consideration. But the original version is time-consuming for the elderly. To reduce fatigue and worsened concentration of older individuals, some short forms of the geriatric depression scale has been developed gradually. Sheikh and Yesavage in 1986 extended the first short form of the GDS with only 15 selected items, the GDS-15. The scores on the GDS-15 have shown a high correlation with those on the original GDS screening tool [3]. 1 Departments of Psychiatry, Chang Gung Memorial Hospital-Linkuo, Tauyuan, Taiwan 2 Departments of Psychiatry, School of Medicine, Chang Gung University, Guei-Shan, Tauyuan, Taiwan 3 Department of Chinese Medicine, School of Medicine, Chang Gung University, Guei-Shan, Tauyuan, Taiwan Received: January 13, 2014; revised: May 15, 2014; accepted: August 1, 2014 *Corresponding author. No. 5, Fu-Hsing Road, Guei-Shan County, Tao-Yuan 333, Taiwan Chun-Lin Chu <56phantom@gmail.com>

2 Chin WC, Liu CY, Lee CP, et al. 157 Although being extensively used with fair validity for screening, GDS-15 is still considered too lengthy by some experts. So, various shorter and simpler forms of geriatric depression scale, including GDS-10, GDS-5, GDS-4 and GDS-1, have been published. Among them, GDS-1 has been suggested to have limited clinical use due to low reliability [4, 5]. And others (GDS-10, GDS-5 and GDS-4) are thought to be good screening instruments for major depression, but the smaller number of questions would be the most desirable for practical purposes [6]. Comparing GDS-5 and GDS-4, Cheng et al. have shown that both versions are excellent alternatives to the GDS-15 [7], whereas Weeks et al. found that GDS-5 has showed better test characteristics than two versions of GDS-4 [8]. Besides, Rinaldo et al. found that GDS-5 is as effective as GDS-15 as a depression screening tool in three different settings (the community, the hospital, and the nursing home) [9]. Therefore, we in the present study intended to focus on if any differences between GDS-5 and GDS-4 exist. The GDS has been translated into different languages and validated in many countries. In Taiwan, translated and adapted Chinese versions GDS-30 and GDS-15 with satisfactory internal consistency and validity have been performed for both the Geriatric Depression Scale [10-12]. But few investigators have ever pursued other shorter forms of Chinese versions GDS, such as GDS-5 and GDS-4, in Taiwanese samples. To the best of our knowledge, little evidence exist about using shorter form GDS in ethnic Chinese population, with only two articles in Hong Kong [13, 14]. Meanwhile, most studies were compared their results with GDS-15 as standard, rather than precise diagnosis of the patients. And comparison of the efficacy among various forms in Chinese version is absent as well. In this study, we intended to assess the feasibility of five different short forms, including GDS-15, two of GDS-5 and two of GDS-4, as screening tools in Taiwanese elderly in a community setting with comparing psychiatric diagnosis (Table 1). Methods We used the data bank from our previous study. Detailed descriptions of the general methodology, including locality, population and methods of subject recruitment, have been reported elsewhere [15] and are briefly summarized below. The study was conducted in conjunction with the Kinmen Neurological Disorders Survey (KINDS), and the data were collected in During the surveillance, a Chinese version of the GDS-15 was used to screen for depression. Because many of the subjects were illiterate, the GDS-15 was asked by a trained medical student one to three days before the psychiatrist s visit. A modified Psychiatrist Diagnostic Assessment (mpda) was used to obtain information for the clinical diagnosis of depression. The PDA is a semi-structured comprehensive psychiatric interview schedule designed specifically for the ethnic Chinese people in Taiwan [16], it was developed for systematic gathering of information for diagnosis according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III). Study subjects The data recruited in this study included 80 subjects. They were originally diagnosed as having major depressive disorder (MDD), and other non-mdd subjects chosen as one-to-four ratio. In this study, we used a shortened version to focus on depressive disorders, and some of its questions about depression were modified to cover the crite-

3 158 Short Form GDS and Screening Table 1. Criterion of various geriatric depression scale short forms Item GDS-15 GDS-5 (Hoyl et al [18]) GDS-5 (Molloy et al [19]) GDS-4 (D Ath et al [20]) GDS-4 (van Marwijk et al. [6]) 1 Are you basically satisfied with your life? 2 Have you dropped many of your activities and interests? 3 Do you feel that your life is empty? 4 Do you often get bored? 5 Are you in good spirits most of the time? 6 Are you afraid that something bad is going to happen to you? 7 Do you feel happy most of the time? 8 Do you often feel helpless? 9 Do you prefer to stay at home, rather than going out and doing new things? 10 Do you feel you have more problems with memory than most people? 11 Do you think it is wonderful to be alive? 12 Do you feel pretty worthless the way you are now? 13 Do you feel full of energy? 14 Do you feel that your situation is hopeless? 15 Do you think that most people are better off than you are? ria for depressive disorders according to the revised version of the DSM-III, the DSM-III-R. In this study, we recruited 338 participants to get total scores 5 on the Chinese version GDS- 15, and another 50 participants with the total scores < 5 as the population of this study under randomized selection. All subjects (n = 388) were interviewed by board-certified psychiatrists with the use of the modified PDA. The diagnosis of depressive disorders was made according to the DSM-III-R criteria. Nevertheless, the Chinese versions GDS-5 and GDS-4 were not a real scale at operated time. The final scores of those scales were calculated from the scores in items of the original Chinese versions GDS-15 items actually. We compared scores of five different short forms, including GDS-15, two forms of GDS-5 (by Hoyl et al. [18] and by Molloy et al. [19]), as well as two forms of GDS-4 (by D Ath et al. [20] and by van Marwijk et al. [6]). The institutional review board of Chang Gung Memorial Hospital approved the study protocol without the need of obtaining signatures for the consent forms from study subjects.

4 Chin WC, Liu CY, Lee CP, et al. 159 Table 2. Correlation of each GDS short forms GDS-4 (D Ath et al 1994 [20]) GDS-4 (van Marwijk et al [6]) GDS-5 (Hoyl et al [18]) GDS-5 (Molloy et al [19]) GDS-15 GDS-4 ( D Ath et al 1994 [20]) r = 0.819*** r = 0.812*** r = 0.901*** r = 0.810*** GDS-4 (van Marwijk et al [6]) r = 0.904*** r = 0.820*** r = 0.828*** GDS-5 (Hoyl et al [18]) r = 0.883*** r = 0.862*** GDS-5 (Molloy et al [19] ) r = 0.859*** * p < 0.05; ** p < 0.01; *** p < GDS short forms, Geriatric Depression Scale short forms Statistical analysis Descriptive and analytical statistics were computed. Internal consistency of each scale was measured with Cronbach s alpha. We calculated specificity, sensitivity, positive predictive value (PPV) and negative predictive value (NPV) of different cut-off points of each scale. The receiver operating characteristic (ROC) curve and area under curve (AUC) of each scale were shown. The diagnostic performance of each scale was evaluated using the ROC curve analysis. In a ROC curve, the true positive predictive value (sensitivity) is plotted in function of the false positive rate (100-specificity) for different cut-off points. Each point on the ROC curve represents a sensitivity/specificity pair corresponding to a particular decision threshold. The accuracy of the test depends on how well the test separates the group being tested into those with and without the disease in question. Accuracy is measured by the area under the ROC curve. An area of 1 represents a perfect test; an area of 0.5 represents a worthless test [17]. We determined the best cut off point of each scale, and calculated their specificity, sensitivity, positive predictive value, and negative predictive value. We computed all study data with Statistical Package for Social Science software version 10 for Windows (SPSS, Inc., Chicago, Illinois, USA). The differences between groups were considered significant if p-values are smaller than Results We included 388 subjects were included in the study. There were 177 male and 211 female, aged from 65 to 93 years old with an average of 74.3 ± 6.8 years. Among them, 80 were diagnosed as having major depressive disorder (83.8% educational year < 1), 308 without a diagnosis of major depressive disorder (78.6% educational year < 1). The age, sex and educational year were no significant difference between the two groups. We compared five different forms of GDS, all of the GDS short forms showed good correlations to each other (Table 2). The ROC curve of each scales were shown in Figure 1. Table 3 shows the Cronbach s alpha, specificity, sensitivity, positive predictive value and negative predictive value of each scale. Discussion This is the first article comparing various short forms of Chinese version GDS in Taiwanese population. All short forms of Chinese version

5 160 Short Form GDS and Screening Figure 1. The receiver operating characteristic curve of each scales. GDS 93: 15 item form, GDS-5_H: 5 items form by Hoyl et al [18]; GDS-5_M: 5 items form by Molloy et al [19]; GDS-4_vM: 4 items form by van Marwijk et al [6]; GDS-4_D: 4 items form by D Ath et al [20] Table 3. The Cronbach s alpha, specificity, sensitivity, positive predictive value and negative predictive value of each GDS short forms sensitivity specificity PPV NPV Cronbach s alpha GDS-4 (D Ath et al [20]) GDS-4 (van Marwijk et al [6]) GDS-5 (Hoyl et al [18]) GDS-5 (Molloy et al [19]) GDS GDS short forms, Geriatric Depression Scale short forms; PPV, positive predictive value; NPV, negative predictive value GDS showed fairly AUC, sensitivity, and specificity for detecting major depressive disorder. All previous Western studies revealed that most short forms have fair screening ability [18-20, 22] for geriatric patients with major depressive disorders. The result of our study showed better findings than those of Western studies, possibly because our study had more participants than other studies [18, 20, 22]. Reviewing previous Taiwanese studies, we have found an article using Chinese version GDS in Taiwan, also showing better AUC, sensitivity and specificity [21]. The study from the originally GDS-Chinese translation group [21] showed that GDS (long form-30 items) have an AUC of 0.97, the sensitivity , and specificity with the cut-off point 15/16.

6 Chin WC, Liu CY, Lee CP, et al. 161 Low positive predictive values (PPV) and high negative predictive values (NPV) of the scales in this study are an issue of concern. They might be due to low prevalence of the illness. In a critical review in British Medical Journal, Loong stated the PPV of a diagnostic test will fall as the prevalence of the disease falls while the NPV will rise [23]. For example, in a study of Brazilian version of the Geriatric Depression Scale (GDS) among primary care patients, the PPV of Brazilian versions of GDS-15, GDS-10 and GDS-4 are 51%, 46% and 41%, respectively. The NPVs are 97%, 94%, and 96% [24]. So, the condition is acceptable for a screening test used in a low prevalence rate disease. Thus, we suggest that Chinese version GDS is a useful screening tool in Taiwanese population. Our results also showed that all Chinese version GDS short forms had good sensitivities but relatively poor specificity in detecting major depressive disorder. The findings mean that they are more suitable to be used for screening depression. That is, like our previous KIND study [15], GDS short form should be used as a screening tool in a two-stage epidemiological studies with psychiatric diagnosis in the second stage. Besides, those versions with more items showed higher AUC in our study, compatible with results of other studies [4]. Previous study reported that GDS-4 has lower sensitivity and specificity than GDS-5, and suggested to use GDS-5 instead of GDS-4 before [12]. Limitation of the study The readers are warned not to over-interpret our study results because our study has several limitations. We did not directly perform Chinese version GDS-5 and GDS-4, and the results were calculated from data of Chinese version GDS-15. Only 50 subjects with GDS score being smaller than 5 received psychiatric interview for control group. Interviewing all the participants was impossible due to the huge sample size. We did not do physical examination, laboratory work-ups, or image studies on our study subjects. We determined study subjects general medical condition and cognitive dysfunction only with interview and reading their medical records. Summary of the study In this study, we found that either longer versions or shorter versions of GDS could reliably detect major depressive episode among older Taiwanese adults. Although shorter versions are less reliable and informative, they are more convenient for clinical use or epidemiological studies. Considering both time-saving and reliability, we recommend that based on the study finding, GDS- 5 is the short form of choice, which is superior to GDS-4 and has little difference in performing time. We also further recommend that the cut-off point can be further lowered to 0/1 in GDS-4 or GDS-5 to increase sensitivity to 100% in the screening stage if in a two-stage epidemiological study. Acknowledgement We declare no potential financial conflicts of interest in writing this report. References 1. Allen J, Annells M: A literature review of the application of the Geriatric Depression Scale, Depression Anxiety Stress Scales and Post-traumatic Stress Disorder Checklist to community nursing cohorts. J Clinl Nurs 2009; 18:

7 162 Short Form GDS and Screening 2. Yesavage JA, Brink TL, Rose TL, et al.: Development and validation of a geriatric depression screening scale: A preliminary report. J Psychiat Res 1982; 17: Sheikh JI, Yesavage JA: Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinl Gerontol 1986; 5: Almeida OP, Almeida SA: Short versions of the Geriatric Depression Scale: A validity for the diagnosis of a major depressive episode according to ICD- 10 and DMS-IV. Int J Geriat Psychiatry 1999; 14: Castelo MS, Coelho-Filho JM, Carvalho AF, et al.: Validity of the Brazilian version of the Geriatric Depression Scale (GDS) among primary care patients. Int Psychogeriatr 2010; 22: van Marwijk HW, Wallace P, de Bock GH, Hermans J, Kaptein AA, Mulder JD: Evaluation of the feasibility, reliability and diagnostic value of shortened versions of the Geriatric Depression Scale. Br J Gen Pract 1995; 45: Cheng ST, Yu EC, Lee SY, et al.: The geriatric depression scale as a screening tool for depression and suicide ideaiton: a repication and extention. Am J Geriatr Psychiatry 2010; 18: Weeks SK, McGann PE, Michaels TK, Penninx BW: Comparing Various Short-Form Geriatric Depression Scales Leads to the GDS-5/15. J Nurs Scholarsh 2003; 35: Rinaldo P, Patrizia M, Benedetti C, et al.: Validation of the Five-Item Geriatric Depression Scale in Elderly Subjects in Three Different Settings. J Am Geriatr Soc 2003; 51: Lee H-cB, Chiu HFK, Kowk WY, et al.: Chinese elderly and the GDS short form: A preliminary study. Clin Gerontologist 1993; 14: Yeh TL, Liao IC, Yang YK, Ko HC, Chang CJ, Lu FH: Geriatric Depression Scale (Taiwanese and Mandarin Translations). Clin Gerontologist 1995; 15: Liu CY, Lu CH, Yu S, Yang Y-Y: Correlations between socres on Chinese Versions of long and short forms of the Geriatric Depression Scale among elderly Chinese. Psychol Rep 1998; 82: Cheng ST, Chan AC: A brief version of the Geriatric Depression Scale for the chinese. Psychol Assess 2004; 16: Cheng ST, Chan AC: Comparative performance of long and short forms of the Geriatric Depression Scale in mildly demented Chinese. Int J Geriatr Psychiatry 2005; 20: Liu HC, Wang SJ, Fuh JL, et al.: The kinmen neurological disorders survey (KINDS): A study of a Chinese population. Neuroepidemiology 1997; 16: Hwu HG: Psychiatrist Diagnostic Assessment: establishment and inter-rater reliability. Chinese Psychiatry (Taipei) 1988; 2: Zweig MH, Campbell G: Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993; 39: Hoyl MT, Alessi CA, Harker JO, et al.: Development and testing of a five item version of the Geriatric Deprssion Scale. J Am Geriatr Soc 1999; 47: Molloy DW, Standish TI, Dubois S, et al.: A short screen for depression: The AB Clinician Depression Screen (ABCDS). Int Psychogeriat 2006; 18: D Ath P, Katona P, Mullan E, Evans S, Katona C: Screening, detection and mangement of depression in elderly primary care attenders. I: The acceptablitiy and performance of the 15 item Geriatric depression Scale (GDS15) and the development of short versions. Fam Prac 1994; 11: Liao YC, Yeh TL, Yang YK, et al.: Reliability and Validation of the Taiwan Geriatric Depression Scale. Taiwanese Journal of Psychiatry (Taipei) 2004; 18: Shan A, Herbert R, Lewis S, Mahendran R, Platt J, Bhattacharyya B: Screening for depression among acutely ill geriatric inpatients with a short Geriatric Depression Scale. Age and Aging 1997; 26: Tze-Wey L: Understanding senstivity and specificity with the right side of the brain. Br Me J 2003; 327: Castelo MS, Coelho-Filho JM, Lima JWO, Noleto

8 Chin WC, Liu CY, Lee CP, et al. 163 JC, Ribeiro KG, Siqueira-Neto JI: Validity of the Brazilian version of the Geriateric Depression Scale (GDS) among primary care patients. In Psychogeriat 2009; 22: