401
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 401–408 DOI 10.3233/BMR-140534 IOS Press
Development and validation of the Dutch version of the London Handicap Scale Catharina G.M. Groothuis-Oudshoorna,∗, Astrid M.J. Chorusb , G.H.W. Verripsb,c and Symone B. Detmarb a
MIRA Institute for Biomedical Technology and Technical Medicine, University of Twente, Enschede, The Netherlands b TNO, Leiden, The Netherlands c Academic Center Dentistry Amsterdam, Amsterdam, The Netherlands
Abstract. BACKGROUND: The London Handicap Scale (LHS) was found to be a valid and reliable scale for measuring participation restrictions in adults. OBJECTIVE: This paper describes the development and assesses the construct-related validity of a Dutch version of the London Handicap Scale (DLHS). METHODS: The DLHS was tested in 798 adults (mean age: 50.7 years, SD = 14.5, range 16 to 85) and validated with the ‘Impact on Participation and Autonomy’ (IPA) questionnaire, the Dutch version of the EQ-5D and questions concerning comorbidity and use of medical devices. The study population consisted of patients with rheumatoid arthritis, chronic obstructive pulmonary disease (COPD), epilepsy, laryngectomy and multiple sclerosis. RESULTS: Feasibility was satisfactory. Large correlations (ρ > 0.6) for the DLHS sum score were found with the IPA subscales ‘autonomy outdoors’, ‘perceiving problems’, ‘family role’, autonomy indoors’, ‘work and education’ and with the EQ-5D. The DLHS sum score differs significantly between subgroups based on the number of chronic diseases, number of medical devices and self-reported burden of disease or handicap (p < 0.001). CONCLUSIONS: Based on this evaluation the questionnaire seems feasible and valid for assessing differences in level of participation between subgroups of chronically ill or disabled persons in the Netherlands. Keywords: Disability, London handicap scale, Dutch, quality of life, construct-related validity
1. Background The reduction of participation restrictions due to a handicap is a key objective of disease management in patients with chronic conditions. Handicap is defined as the disadvantage of an individual as a result of ill health that limits or prevents the fulfillment of a role that is normal (depending on age, sex, and social and cultural factors) [1,2]. Instruments measuring participation restrictions are required for needs assess∗ Corresponding author: C.G.M. Groothuis-Oudshoorn, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands. Tel.: +31 53 4895397; E-mail: [email protected].
ments, quality assurance, and for evaluating interventions aimed at the reduction of handicap. Generic instruments of health-related quality of life are not developed to meet this purpose. The London Handicap Scale (LHS) was developed in the UK, as a generic outcome measure in order to translate the International Classification of Impairment, Disability and Handicap (ICIDH) construct of handicap defined by the World Health Organization [1, 3,4]. Since then, the terminology for handicap has been changed to participation restrictions as part of the International Classification of Functioning (ICF) model [2]. The result of the six items LHS is a preference based utility value, defining the handicap level
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
402
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
of an individual. The classification system of the LHS incorporates the ICF dimensions mobility, occupation, physical independence, social integration, orientation and economic self-sufficiency [5]. Amongst others, the scale has been validated for the use in rheumatologic patients and in stroke patients [6,7]. The reliability of the LHS was shown to be good [7,8]. Nowadays, the LHS has been translated into different languages and is used to address participation outcomes in several countries [5,9–13]. For use in the Netherlands the instrument should be translated into Dutch and evaluated. The estimation of a scoring function for the Dutch LHS to obtain utility values from the general population has been reported elsewhere [14]. The aim of this paper is to describe the adaptation and initial validation of the unweight version of the Dutch LHS (DLHS).
2. Methods 2.1. Instrument adaptation The LHS was translated into Dutch by a forwardbackward translation procedure, according to the principles of Guillemin et al. [15]. A professional Dutch translator made a draft of the Dutch translation. Subsequently, a professional English translator independently translated this back into English. Differences between both English versions of the questionnaire were discussed in a group, which consisted of both translators and two researchers. The differences were minor and could easily be overcome by slight adaptations of the Dutch text. The main discussion was focused on the translation of the word “charity”, mentioned in the last question of the LHS concerning economic self-sufficiency. In Dutch, charity (translated as ‘liefdadigheid’) is an old fashioned description with a negative association. Moreover, the use of the word charity is uncommon. However, it was decided to maintain the aspect of charity in the questionnaire since the concept could not be completely ignored and a better fitting description could not be found. 2.2. Feasibility The translated questionnaire was tested for feasibility in ten people with varying disorders in face-toface interviews. No difficulties were observed in filling in the questionnaire. Consulting colleague-researchers, who were not previously involved in the translation
process, carried out further checks on readability and clarity. Minor adjustments to the translation of the instruments were made according to the findings in this feasibility study. 2.3. Sample validation study In order to obtain a heterogeneous sample of chronically ill patients it was decided to focus on five patient groups, namely rheumatoid arthritis (RA), chronic obstructive pulmonary disease (COPD), epilepsy, laryngectomy, and multiple sclerosis (MS). Using those subgroups, a variety in disease consequences and therefore a wide range of participation restrictions were covered. For example, in the laryngectomy patient group major problems in social integration and less in mobility were expected, whereas major problems concerning mobility and physical independence were expected in the MS patient group. Patients were recruited in the Netherlands from March to May 2002 with the help of patient associations, and using advertisements in patient magazines and on Internet sites. Participants aged 16 years and above and able to read Dutch were recruited. The large differences in sample size between the patients groups were due to the different ways we were able to contact the participants and the prevalence of the different chronic diseases. We intended to include at least 50 participants per patient group in order to be able to detect large correlations per patient group (ρ > 0.5). We intended to obtain a sample size in total of at least 300 to be able to estimate, moderate (ρ > 0.3) and large correlations (ρ > 0.5) with a precision of 0.1 [16]. 2.4. Questionnaires To validate the Dutch version of the LHS, a set of questionnaires was used, namely: 1) the Dutch LHS; 2) the ‘Impact on Participation and Autonomy’ (IPA) questionnaire [17,18]; the Dutch version of the EQ5D [19,20] and 3) questions on socio-economic background, comorbidity, use of (medical) devices and selfreported burden of disease or handicap: ‘what is the burden of your chronic disease or handicap on your daily life’ (answered on a five-point Likert scale with the categories: ‘none’, ‘a little’, ‘moderate’, ‘severe’, ‘highly severe’). In addition, comorbidity was assessed using a list of 25 different chronic diseases as used in the Health Interview Survey (Gezondheidsenquete 1995, Statistics Netherlands). The questionnaires were sent to the participants by postal mail.
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
Number of chronic diseases is defined as one, two or three and more chronic diseases from a list of 25 possible comorbidities in addition to the chronically disease diagnoses used in the study. To assess the use of medical devices, the participant was asked whether he/she uses (one or more) medical devices with respect to mobility, self-care, house keeping, communication, work, leisure, or medicine intake. The number of medical devices used was defined as the total of seven questions regarding the use of medical devices responded with “yes”. The IPA questionnaire is a Dutch generic multi-scale instrument addressing the personal impact of illness on participation and autonomy of people with chronic conditions, and the related experience of problems as defined in the ICF domain of participation [17]. It consists of six subscales: indoor autonomy (7 items, range 0–28), outdoor autonomy (5 items, range 0–20), family role (7 items, range 0–28), social relations (6 items, range 0–24), work and educational opportunities (6 items, range 0–24) and perceiving problems (8 items, range 0–16). The IPA has been validated for various chronic diseases including stroke and RA and has excellent test-retest reliability and internal consistency [18]. The EQ-5D is an often-used generic health-related quality of life measure. It is based on a classification system that classifies patients into 243 possible health states according to the dimensions ‘Mobility’, ‘Self-care’, ‘Usual activities’, ‘Pain/discomfort’, ‘Anxiety/depression’. It consists of five questions, one for each dimension with answering categories according to a three-point Likert scale referring to the level of severity with that dimension. The responses on these five questions were converted into one utility score using the York A1 tariff set [19,20]. 2.5. Analyses Construct validity was assessed by means of correlations between the LHS and the EQ-5D and different subscales of the IPA, which were expected to be greater than 0.6 [21,22]. Based on conceptual similarity and past experiences [18,23], we expected large correlations between the DLHS sum score and EQ-5D and IPA scales ‘autonomy indoors’, ‘family role’, ‘autonomy outdoors’, ‘work and education’. Construct validation was further evaluated by comparing the LHS sum score between different subgroups based on respectively ‘the number of chronic diseases’ (categorized into “one”, “two” and “three or more” chronic
403
diseases), ‘use of medical devices’ (number of medical devices), ‘having a job’ (“yes” and “no”), ‘getting occupational disability benefit’ (“no”, “partly” and “full”) and ‘self-reported burden of disease or handicap’. The following five hypotheses were tested, each with and without adjustment for age, sex and education: 1) The use of more medical devices implies a higher LHS sum score, 2) Suffering from more chronic diseases results in a higher DLHS sum score, 3) Employed patients have a lower DLHS sum score than unemployed patients, 4) Getting occupational disability benefit implies a higher LHS sum score, 5) A higher score on self-reported burden of disease or handicap question, results in a higher DLHS sum score. The characteristics of the sample were summarized using descriptive statistics. Item responses on the DLHS were recoded and transformed such that the total score ranges between 0 and 100 (= no disadvantage). Internal consistency reliability of the DLHS sum score was estimated by means of Cronbach’s alpha, which should be around 0.8 and not below 0.7 [22]. Floor and ceiling effects of the DLHS were examined by percentages of respondents answering the lowest and highest possible scores. Ceiling and floor effects were considered present if these percentages exceed 15–20% [24,25]. Spearman rank correlation coefficients were used to analyze the relation between the DLHS sum score, the IPA subscales, EQ-5D and the number of medical devices. The ANCOVA was used to analyze differences of the DLHS sum score between subgroups as defined in hypotheses 2 to 5, with adjustment for age, sex and education. In case of missing values we used list-wise deletion. For analysis of the data, SPSS (version 19.0) was used.
3. Results In total 803 questionnaires were collected. Five questionnaires were excluded from the analyses since no chronic disease was reported. The prevalence of RA in the sample was 49.2%. COPD was reported by 40.6% of the study population. 7.6% had epilepsy and 7.7% had a laryngectomy. 5.6% of the sample had MS. Due to comorbidity in the sample these percentages do not count to 100%. The non-response in the data was minor (average of less than 2 % over all variables). In only 11 of the 798 questionnaires one or more questions of the DLHS were missing. The mean age of the sample was 50.7 (SD = 14.5), 66.4% were female and 73.7% had a partner (Table 1).
404
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale Table 1 Characteristics of the study population (n = 798)
Mean Age, yrs (sd) Sex (%) Female Male Having partner (%) Yes No Paid job (%) Yes No Unpaid work (%) Yes No Occupational disability benefit (%) No Partly Full Education (%) Lower Middle Higher Number of chronic Diseases (%) One Two Three of more Median number of Medical devices, range 0–7.
Total study population (n = 798) 50.7 (14.5)
MS (n = 65)
RA (n = 395)
COPD (n = 326)
epilepsy (n = 61)
laryngectomy (n = 62)
42.2 (10.7)
52.4 (13.2)
50.1 (16.1)
46.6 (13.4)
62.3 (7.9)
530 (66.4) 268 (33.6)
45 (69.2) 20 (30.8)
308 (78.0) 87 (22.0)
223 (68.4) 103 (31.)
41 (67.2) 20 (32.8)
530 (66.4) 268 (33.6)
586 (73.7) 209 (26.3)
55 (84.6) 11 (15.4)
277 (70.1) 117 (29.7)
232 (71.4) 93 (28.6)
41 (67.2) 20 (32.8)
48 (77.4) 14 (22.6)
231 (29.4) 556 (70.6)
28 (43.1) 37 (56.9)
91 (23.0) 299 (75.7)
103 (32.0) 219 (68.0)
19 (31.1) 42 (68.9)
7 (11.3) 51 (82.3)
188 (24.0) 595 (76.0)
19 (29.2) 46 (70.8)
94 (24.2) 295 (75.8)
67 (20.9) 253 (77.6)
18 (29.5) 42 (68.9)
9 (14.5) 51 (82.3)
428 (54.5) 97 (12.4) 260 (33.1)
18 (27.7) 15 (23.1) 32 (49.2)
198 (51.0) 52 (13.2) 138 (34.9)
207 (64.3) 29 (8.9) 86 (26.4)
34 (55.7) 29 (6.6) 86 (37.7)
36 (58.1) 3 (4.8) 21 (33.9)
84 (10.7) 473 (60.3) 228 (29.0)
1 (1.5) 35 (55.4) 29 (44.6)
33 (8.4) 246 (62.3) 111 (28.1)
37 (11.5) 198 (61.3) 88 (27.2)
9 (15.3) 36 (59.0) 14 (23.0)
15 (24.2) 35 (56.5) 8 (12.9)
192 (24.1) 197 (24.7) 409 (51.3) 2 (range 0–7)
– 37 (56.9) 28 (43.1) 3 (range 0–6)
66 (16.7) 62 (15.7) 267 (67.6) 2 (range 0–7)
79 (24.2) 65 (19.9) 182 (55.8) 2 (range 0–7)
19 (29.5) 15 (24.6) 28 (45.9) 0 (range 0–6)
– 13 (21.0) 49 (79.0) 1 (range 0–5)
Table 2 Item score distributions of the DLHS (n = 798), frequencies (%) Level of disadvantage (1 = none, 6 = most) Mobility (n = 796) Physical independence (n = 794) Occupation (n = 795) Social Integration (n = 796) Orientation (n = 794) Economic self-sufficiency (n = 791) ∗ All
1
2
3
4
5
6
218 (27.3) 273 (34.2) 44 (5.5) 231 (28.9) 584 (73.2) 282 (35.3)
370 (46.4) 198 (24.8) 153 (19.2) 424 (53.1) 157 (19.7) 242 (30.3)
196 (29.6) 235 (29.4) 395 (49.5) 82 (10.3) 52 (6.5) 177 (22.2)
9 (1.1) 77 (9.6) 163 (20.4) 59 (7.4) 1 (0.1) 68 (8.5)
1 (0.1) 7 (0.9) 38 (4.8) 0 (0.0) 0 (0.0) 21 (2.6)
2 (0.3) 4 (0.5) 2 (0.3) 0 (0.0) 0 (0.0) 1 (0.1)
correlations are statistically significant, p < 0.001.
Almost 30% were employed and 24% performed unpaid work. Education of the sample was distributed as low: 10.7%, middle: 60.3%, and high: 29%. The median number of medical devices in the sample was two (range 0–7). Large floor effects were found for all items of the DLHS (Table 2). The highest level of handicap disadvantage with respect to social integration and orientation were not observed. The lowest DLHS sum score in the sample was 40. The highest possible score for the DLHS sum score of 100 was obtained for 1.9% of the sample. To compare, 9.9% of the sample obtained the
highest EQ-5D. Cronbach’s alpha for the DLHS was 0.75. The DLHS sum score ranged from 40 to 100, with a mean value of 78.0. The mean EQ-5D was 0.59 and ranged from −0.43 to one. The larynchetomy patients had the highest DLHS sum score, EQ-5D and lowest IPA scales scores except for IPA social relations (Table 3). The largest correlations (> 0.6) between the subscales of the IPA and the DLHS sum score were found for the IPA scales ‘autonomy outdoors’, ‘perceiving problems’, ‘family role’, ‘autonomy indoors’ and ‘work and education’, as was hypothesized (Ta-
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
405
Table 3 Mean summary outcome measures of the respondents (n = 798) Patient group
DLHS sum score (SD) IPA score (SD) Autonomy indoors Family role Autonomy outdoors Social relations Work and education Perceiving problems EQ-5D (SD)
Total study population (n = 798) 78.0 (11.9)
MS (n = 65)
RA (n = 395)
COPD (n = 326)
epilepsy (n = 61)
74.9 (11.7)
75.6 (10.6)
77.8 (12.6)
78.2 (14.6)
82.8 (12.5)
6.0 (5.2) 11.2 (6.3) 8.8 (4.4) 6.4 (3.9) 11.0 (5.8) 6.3 (3.5) 0.59 (0.31)
6.6 (5.4) 12.5 (5.7) 10.4 (4.4) 6.1 (3.7) 12.8 (5.5) 7.6 (3.4) 0.58 (0.26)
7.3 (5.0) 12.7 (5.8) 9.6 (4.1) 6.5 (3.7) 12.0 (5.5) 6.9 (3.3) 0.48 (0.30)
6.1 (5.7) 11.5 (6.7) 9.3 (4.6) 6.6 (4.0) 10.7 (6.2) 6.5 (3.5) 0.59 (0.31)
4.8 (4.9) 9.6 (7.1) 8.0 (4.8) 7.1 (4.2) 10.9 (6.4) 5.9 (3.8) 0.70 (0.29)
4.0 (5.4) 8.3 (7.0) 6.1 (4.1) 6.9 (4.5) 10.3 (5.6) 4.5 (3.1) 0.78 (0.26)
Table 4 Construct validity of the DLHS∗ Scale IPA Autonomy indoors Family role Autonomy outdoors Social relations Work and education Perceiving problems EQ-5D
laryngetomy (n = 62)
Table 5 Mean (SD) of DLHS sum score for different subgroups
DLHS sum score −0.69 −0.72 −0.79 −0.51 −0.69 −0.75 0.66
∗ The
trends in the means are statistically significant (p 0.001, tested with ANOVA) even after correction for age, sex and education (tested with ANCOVA).
ble 4). These correlations were larger than corresponding correlations between the EQ-5D and the IPA subscales (ranging from −0.34 for ‘social relations’ to −0.64 for ‘autonomy indoors’). The correlation between the DLHS sum score and the EQ-5D was 0.66. The (Spearman) correlation between the number of medical devices and the DLHS sum score was −0.48 (p < 0.001), implying that the number of medical devices is negatively related to the DLHS sum score. The difference in the total LHS score between ‘one chronic disease’ and ‘three or more chronic diseases’ was 7.5 (Table 5). The DLHS sum score was significantly different between subgroups based on the number of chronic diseases (F = 31.6, p < 0.001). Employed respondents had a significantly larger DLHS sum score than unemployed respondents (F = 106.0, p < 0.001). More occupational disability benefit implied a lower DLHS sum score (F = 74.2, p < 0.001). Finally, a higher score on the general question regarding the experienced burden of the disease resulted in a lower DLHS sum score (F = 168.0, p < 0.001). The largest differences for the DLHS sum scores between different subgroups were found for categories based on the self-reported burden of disease or handicap question, namely from 95.3 (‘none’) to 63.9 (‘highly se-
Number of Chronic diseases∗ One (n = 190) Two (n = 196) Three or more (n = 401) Having a job∗ Yes (n = 231) No (n = 545) Occupational disability benefit∗ No (n = 423) Partly (n = 96) Full (n = 256) Self-reported burden of disease or handicap∗ None (n = 12) A little (n = 225) Moderate (n = 303) Severe (n = 161) Highly severe (n = 82)
DLHS sum score (sd) 82.4 (11.5) 80.0 (10.9) 74.9 (11.7) 84.3 (10.0) 75.3 (11.7) 81.9 (11.5) 78.0 (9.8) 71.4 (10.3)
95.3 (5.4) 87.8 (8.0) 77.9 (9.1) 70.3 (8.7) 63.6 (9.9)
vere’). All differences remained highly significant after correcting for age, sex and education. Moreover, posthoc tests revealed that all pairwise comparisons of the differences where significant (p < 0.02) except the difference in DLHS sum score for subjects between one or two chronic diseases. The correlation between the number of medical devices and the DLHS sum score was slightly higher compared to the EQ-5D for all patient groups (Table 6). Both for the DLHS sum score and the EQ-5D there was a significant difference between the different categories of subjective disease burden question (p < 0.001). The same holds true for the number of chronic diseases except for the laryngetomy and MS patients.
4. Discussion The aim of this article was to demonstrate the feasibility and (construct) validity of the Dutch translation
−0.39∗ – 0.65 (0.22)† 0.48 (0.29)†
– 0.78 (0.14)∗ 0.63 (0.15)∗ 0.41 (0.29)∗ 0.21 (0.26)∗
−0.63∗
– 77.0 (10.2) 72.0 (13.1)
– 86.9 (7.7)∗ 76.4 (6.3)∗ 65.3 (10.1)∗ 63.8 (12.1)∗
†: p < 0.01; *: p < 0.001; #: p < 0.05.
Correlation with ‘number of (medical) devices’ Number of chronic diseases One Two Three or more Self-reported burden of disease or handicap None A little Moderate Severe Highly severe
MS (n = 65) DLHS EQ-5D −0.43∗ 0.60 (0.24)∗ 0.54 (0.28)∗ 0.44 (0.31)∗ 0.93 (0.14)∗ 0.68 (0.19)∗ 0.56 (0.23)∗ 0.34 (0.30)∗ 0.16 (0.29)∗
−0.45∗ 81.7 (9.7)∗ 77.3 (9.3)∗ 73.7 (10.5)∗ 97.5 (3.2)∗ 84.7 (8.0)∗ 76.9 (8.3)∗ 70.8 (7.9)∗ 63.6 (9.0)∗
RA (n = 395) DLHS EQ-5D
96.7 (2.7)∗ 88.9 (7.5)∗ 78.1 (9.8)∗ 69.3 (8.4)∗ 63.3 (9.1)∗
83.0 (12.6)∗ 80.6 (11.8)∗ 74.4 (11.8)∗
−0.51∗
1.0 (0.0)∗ 0.80 (0.16)∗ 0.63 (0.22)∗ 0.37 (0.30)∗ 0.23 (0.29)∗
0.75 (0.26)∗ 0.66 (0.24)∗ 0.49 (0.32)∗
−0.48∗
COPD (n = 326) DLHS EQ-5D
96.7 (3.3)∗ 89.2 (7.9)∗ 72.5 (11.0)∗ 75.0 (8.8)∗ 61.7 (10.8)∗
87.1 (8.4)∗ 81.8 (12.7)∗ 70.3 (14.6)∗
−0.59∗
0.87 (0.09)∗ 0.85 (0.12)∗ 0.71 (0.24)∗ 0.58 (0.22)∗ 0.37 (0.36)∗
0.82 (0.11)† 0.80 (0.24)† 0.57 (0.33)†
−0.50∗
epilepsy (n = 61) DLHS EQ-5D
88.3 (11.8)∗ 90.5 (9.0)∗ 83.6 (6.6)∗ 66.7 (3.7)∗ 50.0 (10.0)∗
– 86.7 (8.6) 81.7 (13.2)
−0.34∗
0.79 (0.09)∗ 0.90 (0.11)∗ 0.82 (0.12)∗ 0.58 (0.21)∗ −0.03 (0.51)∗
– 0.91 (0.12)# 0.75 (0.27)#
−0.29∗
laryngetomy (n = 62) DLHS EQ-5D
Table 6 Correlation and means (SD) of the DLHS sum score and the EQ-5D for each patient group with respect to the number of medical devices, the number of chronic diseases and the self-reported handicap of disease or handicap
406 C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
of the LHS, the DLHS. More questionnaires have been developed for the ICF construct of participation, e.g. the IPA. These are however more lengthy than the LHS. Moreover the LHS can be used as a utility measure for participation in cost-effectiveness studies as a scoring function for the DLHS to obtain utility values is also derived [14]. Feasibility of the DLHS was shown to be satisfactory. Construct validity was supported by the distribution of the DLHS scores, which have been shown to be dependent on the number of chronic diseases. There seemed to be ceiling effects in the distribution of the DLHS data (Table 2). This might be due to the small number of severely handicapped patients in the sample. The large ceiling effects found in this study are in accordance with some other studies [6,9,26] but not Harwood and Ebrahim [27]. The distribution of the sum score of the DLHS, which will be used in practice, had no ceiling effects. Evidence for construct validity was given by correlating the DLHS with different subscales of the IPA. The correlations between almost all predefined pairs were large. However, a few correlations between notpredefined pairs were even slightly larger. This can be explained from the fact that the IPA scales themselves were dependent and had correlations ranging from 0.44 (autonomy indoors – social relations) to 0.77 (autonomy outdoors – perceiving problems). Due to the lack of a gold standard, evidence regarding the validity of the DLHS remains circumstantial and depends in our case on the choice of the IPA. However, the IPA is a well validated and reliable Dutch instrument especially developed for the ICF domain of participation. To further assess the construct validity we showed that the DLHS is suitable for assessing differences between different subgroups of chronically ill patients. In order to do so we showed that the sum score of the DLHS was significantly different between subgroups of the sample based on respectively ‘the number of chronic diseases’, ‘having a job’, ‘getting occupational disability’ and ‘self-reported burden of disease or handicap’. Additionally, use of medical devices is positively related with the DLHS sum score. The difference in the DLHS sum score between the different subgroups of the sample based on ‘number of chronic diseases’ was larger than one, which is at least more than one level change for one dimension of the DLHS. The variable ‘number of chronic diseases’ is a rough measure for severeness of disease. However, it was the best at our disposal in this study since the sample was heterogeneous with very different diagnosis groups.
407
In the translation process of the LHS to the DLHS, the discussion was mainly on the word “charity”. It has been decided to maintain this aspect in the questionnaire. An implication of retaining this item might be that respondents will scarcely select the worst category of disadvantage of ‘Economic self-sufficiency’. A footnote in the questionnaire explaining this aspect may be worthwhile for future users. In our study we have not tested the (inter- and intrarater) reliability of the DLHS. There is some evidence for reliability of the English and Turkish LHS especially when the instrument is used for comparing at a group level [7,8,13]. Further research should be done to study the reliability of the (Dutch) instrument. It would also be of interest to study whether caretakers could fill in the instrument instead of the patient himself as this becomes necessary in case of severe handicapped patients. A limitation of the study is the representativeness of the sample for the chronically ill part of the Dutch population due to the sampling process. Chronically ill patients with either access to Internet and/or a membership to patient organizations may have a higher intrinsic level of participation as compared to patients who do not have those facilities. On the other hand, they may experience a higher level of disability, resulting in a larger need to become a member of a patient organization. Therefore, it is not clear whether this selection bias may result in an under or overestimation of the consequences of the handicap on participation. Also, the sample is largely based on RA and COPD, making up almost 90% of the sample. However, the English or other translated versions of the LHS has been validated and used in other significant conditions like stroke [7, 9,28], cataract [29] and cancer [30]. Furthermore, the study population was relatively young (mean age of 50.7) and less handicapped (mean DLHS sum score of 78). 5. Conclusions It can be concluded that, based on the evaluation among 798 chronically ill patients, the DLHS seems a feasible and valid scale for assessing differences in level of participation between subgroups of chronically ill or disabled patients in the Netherlands. Additional research is needed to further validate the DLHS in other important patient groups, older age groups and more severely handicapped patients. Moreover the reliability of the DLHS has not been tested and this should be borne in mind while applying the instrument as well as the use of it by caretakers.
408
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
References
46: 1417-1432. Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Oxford: Blackwell Science; 2002. [17] Cardol M, de Haan RJ, van den Bos GA, de Jong BA, de Groot IJ. The development of a handicap assessment questionnaire: the Impact on Participation and Autonomy (IPA). Clinical Rehabilitation 1999; 13: 411-419. [18] Cardol M, De Haan RJ, De Jong BA, Van Den Bos GAM, De Groot IJM. Psychometric properties of the questionnaire ‘Impact on Participation and Autonomy’(IPA). Arch Phys Med Rehabil 2001; 82: 210-216. [19] Dolan P. Modeling valuations for EuroQol Health States. Medical care, 1997; 35(11): 1095-1108. [20] Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol Group. Ann Med 2001; 33(5): 337-343. [21] Nunnally J, Bernstein I. Psychometric theory: McGraw-Hill; 1994. [22] Streiner DL, Norman GR. Health measurement scales, a practical guide to their development and use. New York: Oxford University Press; 2003. [23] Sibley A, Kersten P, Ward CD, White B, Methta R, George S. Measuring autonomy in disabled people: validation of a new scale in a UK population. Clin Rehab 2006; 20: 793-803. [24] Holmes W, Shea J. Performance of a new, HIV/AIDS targeted quality of life (HAT-QoL) instrument in asympotomatic seropostive individuals. Qual Life Res 1997; 6: 561-571. [25] McHorney CA, Tarlov AR. Individual-patient monitoring in clinical practice: are available health status surveys adequate? Qual Life Res 1994; 4(4): 293-307. [26] Dubuc N, Haley S, Ni P, Kooyoomjian J, Jette A. Function and disability in late life: comparison of the Late-Life Function and Disability Instrument to the SF-36 and London Handicap Scale. Disabil Rehabil 2004; 26(6): 362-370. [27] Harwood RH, Ebrahim S. A comparison of the responsiveness of the Nottingham extended activities of daily living, London handicap scale and SF-36. Disabil Rehabil 2000; 22(17): 786793. [28] Hershkovitz A, Beloosesky Y, Brill S, Gottlieb D. Is a day hospital rehabilitation programme associated with reduction of handicap in stroke patients? Clin Rehabil 2004; 18(3): 261266. [29] Foss AJE, Harwood R, Osborn F, Gregson RM, Zaman A, Masud T. Falls and health status in elderly women following second eye cataract surgery: a randomised control trial, Age and Ageing 2006; 35:66-71. [30] Lo RSK, Kwok TCY, Cheng JOY, Yang H, Yuan HJ, Harwood R, Woo J. Cross-cultural validation of the London Handicap Scale and comparison of handicap perception between Chinese and UK populations. Age and Ageing 2007; 36: 544-548. [16]
[1]
[2] [3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
World Health Organization. International Classification of Impairments, Disabilities and Handicaps. A manual of classification relating to the consequences of disease. Geneva: WHO; 1980. World Health Organization. International Classification of Functioning, Disability and Health. Geneva: WHO; 2002. Harwood RH, Rogers A, Dickinson E, Ebrahim S. Measuring handicap: the London Handicap Scale, a new outcome measure for chronic disease. Qual Health Care 1994; 3: 11-16. Harwood RH, Ebrahim S. Manual of the London handicap scale. Nottingham: University of Nottingham; 1995. Perenboom RJ, Chorus AM. Measuring participation according to the International Classification of Functioning, Disability and Health (ICF). Disability and Rehabilitation 2003; 25(11-12): 577-587. Harwood RH, Carr AJ, Thompson PW, Ebrahim, S. Handicap in inflammatory arthritis. British Journal of Rheumatology 1996; 35: 891-897. Harwood RH, Gompertz P, Ebrahim S. Handicap one year after a stroke: validity of a new scale. J Neurol Neurosurg Psychiatry 1994; 57: 825-829. O’shea SD, Taylor NF, Paratz JD. Measuring changes in activity limitation and participation in people with COPD. Int J of Therapy and Rehabil 2005; 12(6): 264-268. Westergren A, Hagell P. Initial validation of the Swedisch version of the London Handicap Scale. Quality of Life Research 2006; 15: 1251-1256. Lo R, Harwood R, Woo J, Yeung F, Ebrahim S. Cross-cultural validation of the London Handicap Scale in Hong Kong Chinese. Clin Rehab 2001; 15: 177-185. Bent N, Tennant A, Swift T, Posnett J, Scuffham P, Chamberlain MA. Team approach versus ad hoc health services for young people with physical disabilities: a retrospective cohort study. Lancet 2002; 360: 1280-1286. D’alisa S, Baudo S, Mauro A, Miscio G. How does stroke restrict participation in long-term post-stroke survivors? Acta Neurol Scand 2005; 112: 157-162. Kutlay S, Küçükdeveci AA, Yanik B, Elhan A, Öztuna D, Tennant A. The interval scaling properties of the London Handicap Scale: an example from the adaptation of the scale for use in Turkey. Clin Rehabil 2011; 25:248-255. Groothuis-Oudshoorn CGM, Chorus AMJ, Beekum WT van, Detmar SB, Hout WB van den. Modelling and estimation of valuations for the Dutch London Handicap Scale. J Health Econ 2006; 25(6): 1119-1138. Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation of health-related quality of life measures: literature review and proposed guidelines. J Clin Epidemiol 1993;
Copyright of Journal of Back & Musculoskeletal Rehabilitation is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 401–408 DOI 10.3233/BMR-140534 IOS Press
Development and validation of the Dutch version of the London Handicap Scale Catharina G.M. Groothuis-Oudshoorna,∗, Astrid M.J. Chorusb , G.H.W. Verripsb,c and Symone B. Detmarb a
MIRA Institute for Biomedical Technology and Technical Medicine, University of Twente, Enschede, The Netherlands b TNO, Leiden, The Netherlands c Academic Center Dentistry Amsterdam, Amsterdam, The Netherlands
Abstract. BACKGROUND: The London Handicap Scale (LHS) was found to be a valid and reliable scale for measuring participation restrictions in adults. OBJECTIVE: This paper describes the development and assesses the construct-related validity of a Dutch version of the London Handicap Scale (DLHS). METHODS: The DLHS was tested in 798 adults (mean age: 50.7 years, SD = 14.5, range 16 to 85) and validated with the ‘Impact on Participation and Autonomy’ (IPA) questionnaire, the Dutch version of the EQ-5D and questions concerning comorbidity and use of medical devices. The study population consisted of patients with rheumatoid arthritis, chronic obstructive pulmonary disease (COPD), epilepsy, laryngectomy and multiple sclerosis. RESULTS: Feasibility was satisfactory. Large correlations (ρ > 0.6) for the DLHS sum score were found with the IPA subscales ‘autonomy outdoors’, ‘perceiving problems’, ‘family role’, autonomy indoors’, ‘work and education’ and with the EQ-5D. The DLHS sum score differs significantly between subgroups based on the number of chronic diseases, number of medical devices and self-reported burden of disease or handicap (p < 0.001). CONCLUSIONS: Based on this evaluation the questionnaire seems feasible and valid for assessing differences in level of participation between subgroups of chronically ill or disabled persons in the Netherlands. Keywords: Disability, London handicap scale, Dutch, quality of life, construct-related validity
1. Background The reduction of participation restrictions due to a handicap is a key objective of disease management in patients with chronic conditions. Handicap is defined as the disadvantage of an individual as a result of ill health that limits or prevents the fulfillment of a role that is normal (depending on age, sex, and social and cultural factors) [1,2]. Instruments measuring participation restrictions are required for needs assess∗ Corresponding author: C.G.M. Groothuis-Oudshoorn, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands. Tel.: +31 53 4895397; E-mail: [email protected].
ments, quality assurance, and for evaluating interventions aimed at the reduction of handicap. Generic instruments of health-related quality of life are not developed to meet this purpose. The London Handicap Scale (LHS) was developed in the UK, as a generic outcome measure in order to translate the International Classification of Impairment, Disability and Handicap (ICIDH) construct of handicap defined by the World Health Organization [1, 3,4]. Since then, the terminology for handicap has been changed to participation restrictions as part of the International Classification of Functioning (ICF) model [2]. The result of the six items LHS is a preference based utility value, defining the handicap level
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
402
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
of an individual. The classification system of the LHS incorporates the ICF dimensions mobility, occupation, physical independence, social integration, orientation and economic self-sufficiency [5]. Amongst others, the scale has been validated for the use in rheumatologic patients and in stroke patients [6,7]. The reliability of the LHS was shown to be good [7,8]. Nowadays, the LHS has been translated into different languages and is used to address participation outcomes in several countries [5,9–13]. For use in the Netherlands the instrument should be translated into Dutch and evaluated. The estimation of a scoring function for the Dutch LHS to obtain utility values from the general population has been reported elsewhere [14]. The aim of this paper is to describe the adaptation and initial validation of the unweight version of the Dutch LHS (DLHS).
2. Methods 2.1. Instrument adaptation The LHS was translated into Dutch by a forwardbackward translation procedure, according to the principles of Guillemin et al. [15]. A professional Dutch translator made a draft of the Dutch translation. Subsequently, a professional English translator independently translated this back into English. Differences between both English versions of the questionnaire were discussed in a group, which consisted of both translators and two researchers. The differences were minor and could easily be overcome by slight adaptations of the Dutch text. The main discussion was focused on the translation of the word “charity”, mentioned in the last question of the LHS concerning economic self-sufficiency. In Dutch, charity (translated as ‘liefdadigheid’) is an old fashioned description with a negative association. Moreover, the use of the word charity is uncommon. However, it was decided to maintain the aspect of charity in the questionnaire since the concept could not be completely ignored and a better fitting description could not be found. 2.2. Feasibility The translated questionnaire was tested for feasibility in ten people with varying disorders in face-toface interviews. No difficulties were observed in filling in the questionnaire. Consulting colleague-researchers, who were not previously involved in the translation
process, carried out further checks on readability and clarity. Minor adjustments to the translation of the instruments were made according to the findings in this feasibility study. 2.3. Sample validation study In order to obtain a heterogeneous sample of chronically ill patients it was decided to focus on five patient groups, namely rheumatoid arthritis (RA), chronic obstructive pulmonary disease (COPD), epilepsy, laryngectomy, and multiple sclerosis (MS). Using those subgroups, a variety in disease consequences and therefore a wide range of participation restrictions were covered. For example, in the laryngectomy patient group major problems in social integration and less in mobility were expected, whereas major problems concerning mobility and physical independence were expected in the MS patient group. Patients were recruited in the Netherlands from March to May 2002 with the help of patient associations, and using advertisements in patient magazines and on Internet sites. Participants aged 16 years and above and able to read Dutch were recruited. The large differences in sample size between the patients groups were due to the different ways we were able to contact the participants and the prevalence of the different chronic diseases. We intended to include at least 50 participants per patient group in order to be able to detect large correlations per patient group (ρ > 0.5). We intended to obtain a sample size in total of at least 300 to be able to estimate, moderate (ρ > 0.3) and large correlations (ρ > 0.5) with a precision of 0.1 [16]. 2.4. Questionnaires To validate the Dutch version of the LHS, a set of questionnaires was used, namely: 1) the Dutch LHS; 2) the ‘Impact on Participation and Autonomy’ (IPA) questionnaire [17,18]; the Dutch version of the EQ5D [19,20] and 3) questions on socio-economic background, comorbidity, use of (medical) devices and selfreported burden of disease or handicap: ‘what is the burden of your chronic disease or handicap on your daily life’ (answered on a five-point Likert scale with the categories: ‘none’, ‘a little’, ‘moderate’, ‘severe’, ‘highly severe’). In addition, comorbidity was assessed using a list of 25 different chronic diseases as used in the Health Interview Survey (Gezondheidsenquete 1995, Statistics Netherlands). The questionnaires were sent to the participants by postal mail.
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
Number of chronic diseases is defined as one, two or three and more chronic diseases from a list of 25 possible comorbidities in addition to the chronically disease diagnoses used in the study. To assess the use of medical devices, the participant was asked whether he/she uses (one or more) medical devices with respect to mobility, self-care, house keeping, communication, work, leisure, or medicine intake. The number of medical devices used was defined as the total of seven questions regarding the use of medical devices responded with “yes”. The IPA questionnaire is a Dutch generic multi-scale instrument addressing the personal impact of illness on participation and autonomy of people with chronic conditions, and the related experience of problems as defined in the ICF domain of participation [17]. It consists of six subscales: indoor autonomy (7 items, range 0–28), outdoor autonomy (5 items, range 0–20), family role (7 items, range 0–28), social relations (6 items, range 0–24), work and educational opportunities (6 items, range 0–24) and perceiving problems (8 items, range 0–16). The IPA has been validated for various chronic diseases including stroke and RA and has excellent test-retest reliability and internal consistency [18]. The EQ-5D is an often-used generic health-related quality of life measure. It is based on a classification system that classifies patients into 243 possible health states according to the dimensions ‘Mobility’, ‘Self-care’, ‘Usual activities’, ‘Pain/discomfort’, ‘Anxiety/depression’. It consists of five questions, one for each dimension with answering categories according to a three-point Likert scale referring to the level of severity with that dimension. The responses on these five questions were converted into one utility score using the York A1 tariff set [19,20]. 2.5. Analyses Construct validity was assessed by means of correlations between the LHS and the EQ-5D and different subscales of the IPA, which were expected to be greater than 0.6 [21,22]. Based on conceptual similarity and past experiences [18,23], we expected large correlations between the DLHS sum score and EQ-5D and IPA scales ‘autonomy indoors’, ‘family role’, ‘autonomy outdoors’, ‘work and education’. Construct validation was further evaluated by comparing the LHS sum score between different subgroups based on respectively ‘the number of chronic diseases’ (categorized into “one”, “two” and “three or more” chronic
403
diseases), ‘use of medical devices’ (number of medical devices), ‘having a job’ (“yes” and “no”), ‘getting occupational disability benefit’ (“no”, “partly” and “full”) and ‘self-reported burden of disease or handicap’. The following five hypotheses were tested, each with and without adjustment for age, sex and education: 1) The use of more medical devices implies a higher LHS sum score, 2) Suffering from more chronic diseases results in a higher DLHS sum score, 3) Employed patients have a lower DLHS sum score than unemployed patients, 4) Getting occupational disability benefit implies a higher LHS sum score, 5) A higher score on self-reported burden of disease or handicap question, results in a higher DLHS sum score. The characteristics of the sample were summarized using descriptive statistics. Item responses on the DLHS were recoded and transformed such that the total score ranges between 0 and 100 (= no disadvantage). Internal consistency reliability of the DLHS sum score was estimated by means of Cronbach’s alpha, which should be around 0.8 and not below 0.7 [22]. Floor and ceiling effects of the DLHS were examined by percentages of respondents answering the lowest and highest possible scores. Ceiling and floor effects were considered present if these percentages exceed 15–20% [24,25]. Spearman rank correlation coefficients were used to analyze the relation between the DLHS sum score, the IPA subscales, EQ-5D and the number of medical devices. The ANCOVA was used to analyze differences of the DLHS sum score between subgroups as defined in hypotheses 2 to 5, with adjustment for age, sex and education. In case of missing values we used list-wise deletion. For analysis of the data, SPSS (version 19.0) was used.
3. Results In total 803 questionnaires were collected. Five questionnaires were excluded from the analyses since no chronic disease was reported. The prevalence of RA in the sample was 49.2%. COPD was reported by 40.6% of the study population. 7.6% had epilepsy and 7.7% had a laryngectomy. 5.6% of the sample had MS. Due to comorbidity in the sample these percentages do not count to 100%. The non-response in the data was minor (average of less than 2 % over all variables). In only 11 of the 798 questionnaires one or more questions of the DLHS were missing. The mean age of the sample was 50.7 (SD = 14.5), 66.4% were female and 73.7% had a partner (Table 1).
404
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale Table 1 Characteristics of the study population (n = 798)
Mean Age, yrs (sd) Sex (%) Female Male Having partner (%) Yes No Paid job (%) Yes No Unpaid work (%) Yes No Occupational disability benefit (%) No Partly Full Education (%) Lower Middle Higher Number of chronic Diseases (%) One Two Three of more Median number of Medical devices, range 0–7.
Total study population (n = 798) 50.7 (14.5)
MS (n = 65)
RA (n = 395)
COPD (n = 326)
epilepsy (n = 61)
laryngectomy (n = 62)
42.2 (10.7)
52.4 (13.2)
50.1 (16.1)
46.6 (13.4)
62.3 (7.9)
530 (66.4) 268 (33.6)
45 (69.2) 20 (30.8)
308 (78.0) 87 (22.0)
223 (68.4) 103 (31.)
41 (67.2) 20 (32.8)
530 (66.4) 268 (33.6)
586 (73.7) 209 (26.3)
55 (84.6) 11 (15.4)
277 (70.1) 117 (29.7)
232 (71.4) 93 (28.6)
41 (67.2) 20 (32.8)
48 (77.4) 14 (22.6)
231 (29.4) 556 (70.6)
28 (43.1) 37 (56.9)
91 (23.0) 299 (75.7)
103 (32.0) 219 (68.0)
19 (31.1) 42 (68.9)
7 (11.3) 51 (82.3)
188 (24.0) 595 (76.0)
19 (29.2) 46 (70.8)
94 (24.2) 295 (75.8)
67 (20.9) 253 (77.6)
18 (29.5) 42 (68.9)
9 (14.5) 51 (82.3)
428 (54.5) 97 (12.4) 260 (33.1)
18 (27.7) 15 (23.1) 32 (49.2)
198 (51.0) 52 (13.2) 138 (34.9)
207 (64.3) 29 (8.9) 86 (26.4)
34 (55.7) 29 (6.6) 86 (37.7)
36 (58.1) 3 (4.8) 21 (33.9)
84 (10.7) 473 (60.3) 228 (29.0)
1 (1.5) 35 (55.4) 29 (44.6)
33 (8.4) 246 (62.3) 111 (28.1)
37 (11.5) 198 (61.3) 88 (27.2)
9 (15.3) 36 (59.0) 14 (23.0)
15 (24.2) 35 (56.5) 8 (12.9)
192 (24.1) 197 (24.7) 409 (51.3) 2 (range 0–7)
– 37 (56.9) 28 (43.1) 3 (range 0–6)
66 (16.7) 62 (15.7) 267 (67.6) 2 (range 0–7)
79 (24.2) 65 (19.9) 182 (55.8) 2 (range 0–7)
19 (29.5) 15 (24.6) 28 (45.9) 0 (range 0–6)
– 13 (21.0) 49 (79.0) 1 (range 0–5)
Table 2 Item score distributions of the DLHS (n = 798), frequencies (%) Level of disadvantage (1 = none, 6 = most) Mobility (n = 796) Physical independence (n = 794) Occupation (n = 795) Social Integration (n = 796) Orientation (n = 794) Economic self-sufficiency (n = 791) ∗ All
1
2
3
4
5
6
218 (27.3) 273 (34.2) 44 (5.5) 231 (28.9) 584 (73.2) 282 (35.3)
370 (46.4) 198 (24.8) 153 (19.2) 424 (53.1) 157 (19.7) 242 (30.3)
196 (29.6) 235 (29.4) 395 (49.5) 82 (10.3) 52 (6.5) 177 (22.2)
9 (1.1) 77 (9.6) 163 (20.4) 59 (7.4) 1 (0.1) 68 (8.5)
1 (0.1) 7 (0.9) 38 (4.8) 0 (0.0) 0 (0.0) 21 (2.6)
2 (0.3) 4 (0.5) 2 (0.3) 0 (0.0) 0 (0.0) 1 (0.1)
correlations are statistically significant, p < 0.001.
Almost 30% were employed and 24% performed unpaid work. Education of the sample was distributed as low: 10.7%, middle: 60.3%, and high: 29%. The median number of medical devices in the sample was two (range 0–7). Large floor effects were found for all items of the DLHS (Table 2). The highest level of handicap disadvantage with respect to social integration and orientation were not observed. The lowest DLHS sum score in the sample was 40. The highest possible score for the DLHS sum score of 100 was obtained for 1.9% of the sample. To compare, 9.9% of the sample obtained the
highest EQ-5D. Cronbach’s alpha for the DLHS was 0.75. The DLHS sum score ranged from 40 to 100, with a mean value of 78.0. The mean EQ-5D was 0.59 and ranged from −0.43 to one. The larynchetomy patients had the highest DLHS sum score, EQ-5D and lowest IPA scales scores except for IPA social relations (Table 3). The largest correlations (> 0.6) between the subscales of the IPA and the DLHS sum score were found for the IPA scales ‘autonomy outdoors’, ‘perceiving problems’, ‘family role’, ‘autonomy indoors’ and ‘work and education’, as was hypothesized (Ta-
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
405
Table 3 Mean summary outcome measures of the respondents (n = 798) Patient group
DLHS sum score (SD) IPA score (SD) Autonomy indoors Family role Autonomy outdoors Social relations Work and education Perceiving problems EQ-5D (SD)
Total study population (n = 798) 78.0 (11.9)
MS (n = 65)
RA (n = 395)
COPD (n = 326)
epilepsy (n = 61)
74.9 (11.7)
75.6 (10.6)
77.8 (12.6)
78.2 (14.6)
82.8 (12.5)
6.0 (5.2) 11.2 (6.3) 8.8 (4.4) 6.4 (3.9) 11.0 (5.8) 6.3 (3.5) 0.59 (0.31)
6.6 (5.4) 12.5 (5.7) 10.4 (4.4) 6.1 (3.7) 12.8 (5.5) 7.6 (3.4) 0.58 (0.26)
7.3 (5.0) 12.7 (5.8) 9.6 (4.1) 6.5 (3.7) 12.0 (5.5) 6.9 (3.3) 0.48 (0.30)
6.1 (5.7) 11.5 (6.7) 9.3 (4.6) 6.6 (4.0) 10.7 (6.2) 6.5 (3.5) 0.59 (0.31)
4.8 (4.9) 9.6 (7.1) 8.0 (4.8) 7.1 (4.2) 10.9 (6.4) 5.9 (3.8) 0.70 (0.29)
4.0 (5.4) 8.3 (7.0) 6.1 (4.1) 6.9 (4.5) 10.3 (5.6) 4.5 (3.1) 0.78 (0.26)
Table 4 Construct validity of the DLHS∗ Scale IPA Autonomy indoors Family role Autonomy outdoors Social relations Work and education Perceiving problems EQ-5D
laryngetomy (n = 62)
Table 5 Mean (SD) of DLHS sum score for different subgroups
DLHS sum score −0.69 −0.72 −0.79 −0.51 −0.69 −0.75 0.66
∗ The
trends in the means are statistically significant (p 0.001, tested with ANOVA) even after correction for age, sex and education (tested with ANCOVA).
ble 4). These correlations were larger than corresponding correlations between the EQ-5D and the IPA subscales (ranging from −0.34 for ‘social relations’ to −0.64 for ‘autonomy indoors’). The correlation between the DLHS sum score and the EQ-5D was 0.66. The (Spearman) correlation between the number of medical devices and the DLHS sum score was −0.48 (p < 0.001), implying that the number of medical devices is negatively related to the DLHS sum score. The difference in the total LHS score between ‘one chronic disease’ and ‘three or more chronic diseases’ was 7.5 (Table 5). The DLHS sum score was significantly different between subgroups based on the number of chronic diseases (F = 31.6, p < 0.001). Employed respondents had a significantly larger DLHS sum score than unemployed respondents (F = 106.0, p < 0.001). More occupational disability benefit implied a lower DLHS sum score (F = 74.2, p < 0.001). Finally, a higher score on the general question regarding the experienced burden of the disease resulted in a lower DLHS sum score (F = 168.0, p < 0.001). The largest differences for the DLHS sum scores between different subgroups were found for categories based on the self-reported burden of disease or handicap question, namely from 95.3 (‘none’) to 63.9 (‘highly se-
Number of Chronic diseases∗ One (n = 190) Two (n = 196) Three or more (n = 401) Having a job∗ Yes (n = 231) No (n = 545) Occupational disability benefit∗ No (n = 423) Partly (n = 96) Full (n = 256) Self-reported burden of disease or handicap∗ None (n = 12) A little (n = 225) Moderate (n = 303) Severe (n = 161) Highly severe (n = 82)
DLHS sum score (sd) 82.4 (11.5) 80.0 (10.9) 74.9 (11.7) 84.3 (10.0) 75.3 (11.7) 81.9 (11.5) 78.0 (9.8) 71.4 (10.3)
95.3 (5.4) 87.8 (8.0) 77.9 (9.1) 70.3 (8.7) 63.6 (9.9)
vere’). All differences remained highly significant after correcting for age, sex and education. Moreover, posthoc tests revealed that all pairwise comparisons of the differences where significant (p < 0.02) except the difference in DLHS sum score for subjects between one or two chronic diseases. The correlation between the number of medical devices and the DLHS sum score was slightly higher compared to the EQ-5D for all patient groups (Table 6). Both for the DLHS sum score and the EQ-5D there was a significant difference between the different categories of subjective disease burden question (p < 0.001). The same holds true for the number of chronic diseases except for the laryngetomy and MS patients.
4. Discussion The aim of this article was to demonstrate the feasibility and (construct) validity of the Dutch translation
−0.39∗ – 0.65 (0.22)† 0.48 (0.29)†
– 0.78 (0.14)∗ 0.63 (0.15)∗ 0.41 (0.29)∗ 0.21 (0.26)∗
−0.63∗
– 77.0 (10.2) 72.0 (13.1)
– 86.9 (7.7)∗ 76.4 (6.3)∗ 65.3 (10.1)∗ 63.8 (12.1)∗
†: p < 0.01; *: p < 0.001; #: p < 0.05.
Correlation with ‘number of (medical) devices’ Number of chronic diseases One Two Three or more Self-reported burden of disease or handicap None A little Moderate Severe Highly severe
MS (n = 65) DLHS EQ-5D −0.43∗ 0.60 (0.24)∗ 0.54 (0.28)∗ 0.44 (0.31)∗ 0.93 (0.14)∗ 0.68 (0.19)∗ 0.56 (0.23)∗ 0.34 (0.30)∗ 0.16 (0.29)∗
−0.45∗ 81.7 (9.7)∗ 77.3 (9.3)∗ 73.7 (10.5)∗ 97.5 (3.2)∗ 84.7 (8.0)∗ 76.9 (8.3)∗ 70.8 (7.9)∗ 63.6 (9.0)∗
RA (n = 395) DLHS EQ-5D
96.7 (2.7)∗ 88.9 (7.5)∗ 78.1 (9.8)∗ 69.3 (8.4)∗ 63.3 (9.1)∗
83.0 (12.6)∗ 80.6 (11.8)∗ 74.4 (11.8)∗
−0.51∗
1.0 (0.0)∗ 0.80 (0.16)∗ 0.63 (0.22)∗ 0.37 (0.30)∗ 0.23 (0.29)∗
0.75 (0.26)∗ 0.66 (0.24)∗ 0.49 (0.32)∗
−0.48∗
COPD (n = 326) DLHS EQ-5D
96.7 (3.3)∗ 89.2 (7.9)∗ 72.5 (11.0)∗ 75.0 (8.8)∗ 61.7 (10.8)∗
87.1 (8.4)∗ 81.8 (12.7)∗ 70.3 (14.6)∗
−0.59∗
0.87 (0.09)∗ 0.85 (0.12)∗ 0.71 (0.24)∗ 0.58 (0.22)∗ 0.37 (0.36)∗
0.82 (0.11)† 0.80 (0.24)† 0.57 (0.33)†
−0.50∗
epilepsy (n = 61) DLHS EQ-5D
88.3 (11.8)∗ 90.5 (9.0)∗ 83.6 (6.6)∗ 66.7 (3.7)∗ 50.0 (10.0)∗
– 86.7 (8.6) 81.7 (13.2)
−0.34∗
0.79 (0.09)∗ 0.90 (0.11)∗ 0.82 (0.12)∗ 0.58 (0.21)∗ −0.03 (0.51)∗
– 0.91 (0.12)# 0.75 (0.27)#
−0.29∗
laryngetomy (n = 62) DLHS EQ-5D
Table 6 Correlation and means (SD) of the DLHS sum score and the EQ-5D for each patient group with respect to the number of medical devices, the number of chronic diseases and the self-reported handicap of disease or handicap
406 C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
of the LHS, the DLHS. More questionnaires have been developed for the ICF construct of participation, e.g. the IPA. These are however more lengthy than the LHS. Moreover the LHS can be used as a utility measure for participation in cost-effectiveness studies as a scoring function for the DLHS to obtain utility values is also derived [14]. Feasibility of the DLHS was shown to be satisfactory. Construct validity was supported by the distribution of the DLHS scores, which have been shown to be dependent on the number of chronic diseases. There seemed to be ceiling effects in the distribution of the DLHS data (Table 2). This might be due to the small number of severely handicapped patients in the sample. The large ceiling effects found in this study are in accordance with some other studies [6,9,26] but not Harwood and Ebrahim [27]. The distribution of the sum score of the DLHS, which will be used in practice, had no ceiling effects. Evidence for construct validity was given by correlating the DLHS with different subscales of the IPA. The correlations between almost all predefined pairs were large. However, a few correlations between notpredefined pairs were even slightly larger. This can be explained from the fact that the IPA scales themselves were dependent and had correlations ranging from 0.44 (autonomy indoors – social relations) to 0.77 (autonomy outdoors – perceiving problems). Due to the lack of a gold standard, evidence regarding the validity of the DLHS remains circumstantial and depends in our case on the choice of the IPA. However, the IPA is a well validated and reliable Dutch instrument especially developed for the ICF domain of participation. To further assess the construct validity we showed that the DLHS is suitable for assessing differences between different subgroups of chronically ill patients. In order to do so we showed that the sum score of the DLHS was significantly different between subgroups of the sample based on respectively ‘the number of chronic diseases’, ‘having a job’, ‘getting occupational disability’ and ‘self-reported burden of disease or handicap’. Additionally, use of medical devices is positively related with the DLHS sum score. The difference in the DLHS sum score between the different subgroups of the sample based on ‘number of chronic diseases’ was larger than one, which is at least more than one level change for one dimension of the DLHS. The variable ‘number of chronic diseases’ is a rough measure for severeness of disease. However, it was the best at our disposal in this study since the sample was heterogeneous with very different diagnosis groups.
407
In the translation process of the LHS to the DLHS, the discussion was mainly on the word “charity”. It has been decided to maintain this aspect in the questionnaire. An implication of retaining this item might be that respondents will scarcely select the worst category of disadvantage of ‘Economic self-sufficiency’. A footnote in the questionnaire explaining this aspect may be worthwhile for future users. In our study we have not tested the (inter- and intrarater) reliability of the DLHS. There is some evidence for reliability of the English and Turkish LHS especially when the instrument is used for comparing at a group level [7,8,13]. Further research should be done to study the reliability of the (Dutch) instrument. It would also be of interest to study whether caretakers could fill in the instrument instead of the patient himself as this becomes necessary in case of severe handicapped patients. A limitation of the study is the representativeness of the sample for the chronically ill part of the Dutch population due to the sampling process. Chronically ill patients with either access to Internet and/or a membership to patient organizations may have a higher intrinsic level of participation as compared to patients who do not have those facilities. On the other hand, they may experience a higher level of disability, resulting in a larger need to become a member of a patient organization. Therefore, it is not clear whether this selection bias may result in an under or overestimation of the consequences of the handicap on participation. Also, the sample is largely based on RA and COPD, making up almost 90% of the sample. However, the English or other translated versions of the LHS has been validated and used in other significant conditions like stroke [7, 9,28], cataract [29] and cancer [30]. Furthermore, the study population was relatively young (mean age of 50.7) and less handicapped (mean DLHS sum score of 78). 5. Conclusions It can be concluded that, based on the evaluation among 798 chronically ill patients, the DLHS seems a feasible and valid scale for assessing differences in level of participation between subgroups of chronically ill or disabled patients in the Netherlands. Additional research is needed to further validate the DLHS in other important patient groups, older age groups and more severely handicapped patients. Moreover the reliability of the DLHS has not been tested and this should be borne in mind while applying the instrument as well as the use of it by caretakers.
408
C.G.M. Groothuis-Oudshoorn et al. / Development and validation of the Dutch version of the London Handicap Scale
References
46: 1417-1432. Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Oxford: Blackwell Science; 2002. [17] Cardol M, de Haan RJ, van den Bos GA, de Jong BA, de Groot IJ. The development of a handicap assessment questionnaire: the Impact on Participation and Autonomy (IPA). Clinical Rehabilitation 1999; 13: 411-419. [18] Cardol M, De Haan RJ, De Jong BA, Van Den Bos GAM, De Groot IJM. Psychometric properties of the questionnaire ‘Impact on Participation and Autonomy’(IPA). Arch Phys Med Rehabil 2001; 82: 210-216. [19] Dolan P. Modeling valuations for EuroQol Health States. Medical care, 1997; 35(11): 1095-1108. [20] Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol Group. Ann Med 2001; 33(5): 337-343. [21] Nunnally J, Bernstein I. Psychometric theory: McGraw-Hill; 1994. [22] Streiner DL, Norman GR. Health measurement scales, a practical guide to their development and use. New York: Oxford University Press; 2003. [23] Sibley A, Kersten P, Ward CD, White B, Methta R, George S. Measuring autonomy in disabled people: validation of a new scale in a UK population. Clin Rehab 2006; 20: 793-803. [24] Holmes W, Shea J. Performance of a new, HIV/AIDS targeted quality of life (HAT-QoL) instrument in asympotomatic seropostive individuals. Qual Life Res 1997; 6: 561-571. [25] McHorney CA, Tarlov AR. Individual-patient monitoring in clinical practice: are available health status surveys adequate? Qual Life Res 1994; 4(4): 293-307. [26] Dubuc N, Haley S, Ni P, Kooyoomjian J, Jette A. Function and disability in late life: comparison of the Late-Life Function and Disability Instrument to the SF-36 and London Handicap Scale. Disabil Rehabil 2004; 26(6): 362-370. [27] Harwood RH, Ebrahim S. A comparison of the responsiveness of the Nottingham extended activities of daily living, London handicap scale and SF-36. Disabil Rehabil 2000; 22(17): 786793. [28] Hershkovitz A, Beloosesky Y, Brill S, Gottlieb D. Is a day hospital rehabilitation programme associated with reduction of handicap in stroke patients? Clin Rehabil 2004; 18(3): 261266. [29] Foss AJE, Harwood R, Osborn F, Gregson RM, Zaman A, Masud T. Falls and health status in elderly women following second eye cataract surgery: a randomised control trial, Age and Ageing 2006; 35:66-71. [30] Lo RSK, Kwok TCY, Cheng JOY, Yang H, Yuan HJ, Harwood R, Woo J. Cross-cultural validation of the London Handicap Scale and comparison of handicap perception between Chinese and UK populations. Age and Ageing 2007; 36: 544-548. [16]
[1]
[2] [3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
World Health Organization. International Classification of Impairments, Disabilities and Handicaps. A manual of classification relating to the consequences of disease. Geneva: WHO; 1980. World Health Organization. International Classification of Functioning, Disability and Health. Geneva: WHO; 2002. Harwood RH, Rogers A, Dickinson E, Ebrahim S. Measuring handicap: the London Handicap Scale, a new outcome measure for chronic disease. Qual Health Care 1994; 3: 11-16. Harwood RH, Ebrahim S. Manual of the London handicap scale. Nottingham: University of Nottingham; 1995. Perenboom RJ, Chorus AM. Measuring participation according to the International Classification of Functioning, Disability and Health (ICF). Disability and Rehabilitation 2003; 25(11-12): 577-587. Harwood RH, Carr AJ, Thompson PW, Ebrahim, S. Handicap in inflammatory arthritis. British Journal of Rheumatology 1996; 35: 891-897. Harwood RH, Gompertz P, Ebrahim S. Handicap one year after a stroke: validity of a new scale. J Neurol Neurosurg Psychiatry 1994; 57: 825-829. O’shea SD, Taylor NF, Paratz JD. Measuring changes in activity limitation and participation in people with COPD. Int J of Therapy and Rehabil 2005; 12(6): 264-268. Westergren A, Hagell P. Initial validation of the Swedisch version of the London Handicap Scale. Quality of Life Research 2006; 15: 1251-1256. Lo R, Harwood R, Woo J, Yeung F, Ebrahim S. Cross-cultural validation of the London Handicap Scale in Hong Kong Chinese. Clin Rehab 2001; 15: 177-185. Bent N, Tennant A, Swift T, Posnett J, Scuffham P, Chamberlain MA. Team approach versus ad hoc health services for young people with physical disabilities: a retrospective cohort study. Lancet 2002; 360: 1280-1286. D’alisa S, Baudo S, Mauro A, Miscio G. How does stroke restrict participation in long-term post-stroke survivors? Acta Neurol Scand 2005; 112: 157-162. Kutlay S, Küçükdeveci AA, Yanik B, Elhan A, Öztuna D, Tennant A. The interval scaling properties of the London Handicap Scale: an example from the adaptation of the scale for use in Turkey. Clin Rehabil 2011; 25:248-255. Groothuis-Oudshoorn CGM, Chorus AMJ, Beekum WT van, Detmar SB, Hout WB van den. Modelling and estimation of valuations for the Dutch London Handicap Scale. J Health Econ 2006; 25(6): 1119-1138. Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation of health-related quality of life measures: literature review and proposed guidelines. J Clin Epidemiol 1993;
Copyright of Journal of Back & Musculoskeletal Rehabilitation is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
