Reliability and validity of the Spanish version of the World Health Organization-Five Well-Being Index in elderly

Authors

  • Ramona Lucas-Carrasco MD, PhD

    Corresponding author
    • Department of Methodology and Behavioral Sciences, University of Barcelona & SGR 822 Generalitat de Catalunya, Barcelona, Spain
    Search for more papers by this author

Correspondence: Ramona Lucas-Carrasco, MD, PhD, Department of Methodology and Behavioural Sciences, University of Barcelona, Passeig Vall d'Hebron 171, Barcelona 08035, Spain. Email: lucas.ramona@gmail.com

Abstract

Aims

The World Health Organization (WHO)-Five Well-being Index (WHO-5) is a short 5-item index designed to assess the level of emotional well-being. The positive questions of the WHO-5, shifted towards measuring cheerfulness and the level of energy, work in the screening of depression. This paper describes the psychometric properties of the Spanish WHO-5 in older persons.

Methods

A total of 199 participants from community centers and primary care centers participated in the study completing a battery of measures: WHO-5, Quality of Life (WHOQOL-BREF), depressive symptoms (Geriatric Depression Scale [GDS-15]), as well as health and sociodemographic information. Analysis was performed using standard psychometric methods.

Results

Internal consistency reliability was good (Cronbach's α = 0.86). Exploratory factor analysis showed a one-factor solution which accounted for 66% of the total variance of WHO-5. Moderate–high correlations were found between WHO-5 and the WHOQOL-BREF and GDS-15 were confirmed, indicating good convergent validity. Discriminative validity was confirmed by the ability of the WHO-5 to differentiate between healthy and unhealthy and depressed and non-depressed participants (as measured by the GDS-15).

Conclusions

The WHO-5 showed acceptable psychometric properties in elderly persons. It might be a useful tool to assess emotional well-being and to detect depressive symptoms among older persons in primary care and community centers.

THE CONCEPT OF well-being involves different dimensions related to the individual and society.[1] Among older persons, some studies have reported higher levels of well-being with increasing age among men, married people and those with higher income,[2] while others have found lower levels of well-being with increasing age.[3] Similarly, lower levels of well-being have been consistently associated with impaired physical health, chronic conditions and depressive symptoms.[4] Depression is associated with disability and higher use of health services;[5] however, in elderly persons, depression goes undetected in primary care and is poorly treated.[6] It has been established that primary care patients with major or minor depressive disorders respond well to psychotherapy and/or treatment with antidepressants.[7, 8]

The World Health Organization (WHO)-Five Well-being Index (WHO-5) is a short, 5-item index designed to assess the level of emotional well-being.[9] The positive questions of the WHO-5, shifted towards measuring cheerfulness and the level of energy, work in the screening of depression as successfully as the questions narrowly oriented on depressive symptoms,[10, 11] which could be hidden by patients because of shame and stigma associated with mental disorders.[10] The WHO-5 has been widely used to screen for depression in primary care.[12-14] The psychometric properties of the WHO-5 have been studied in elderly persons,[10, 12-17] patients with Parkinson's disease,[11] depressed psychiatric patients[18] and diabetics.[19] Overall, these studies reported acceptable psychometric properties: acceptable Cronbach's alpha coefficients, good test–retest reliability, discriminative validity and concurrent validity; and showed that the WHO-5 is useful for monitoring individual patient progress.[18] The WHO-5 is freely available in different languages (http://www.who-5.org/). Permission of use was requested and obtained from Professor Per Bech. Our objective was to examine the validity and reliability of the Spanish WHO-5 in a sample of community-dwelling elderly.

Methods

Participants

We used a convenience sample. A total of 199 elderly persons from community centers and primary care centers in Spain participated in the study. Participants were included if they were 65 years of age and older, knew how to read and write and were willing to provide written informed consent.

Measures

WHO-5[15]

WHO-5 is a self-administered five-item scale. Each item assesses the degree of positive well-being during the past 2 weeks on a six-point Likert scale graded from 0 (at no time) to 5 (all of the time); the total score ranges from 0 to 25, with high scores indicating an increased sense of well-being. A cut-off point less than 13 has been associated with depression.[17]

WHO Quality of Life-BREF[20]

The World Health Organization Quality of Life (WHOQOL)-BREF is a generic quality of life (QoL) questionnaire comprising 26 items: 24 items covering four domains (physical health, psychological health, social relationships and environment) and two global questions about overall QoL and satisfaction with health. Each item is scored on a five-point scale, with higher scores indicating better QoL. Time frame for assessment is the past 2 weeks. The Spanish version has shown good psychometric properties in Spanish elderly.[21]

Geriatric Depression Scale[22]

The Geriatric Depression Scale 15-item questionnaire (GDS-15) measures depressive symptoms. Answers are reported on a yes/no scale with high scores indicating more severe depression. The time frame for the measure is the present. A cut-off score of 5 was used to divide the sample into non-depressed (GDS-15 < 5) and probably depressed (GDS-15 ≥ 5) participants.[23]

In addition, we obtained sociodemographic information; subjective perception of health (‘In general, do you consider yourself to be currently healthy or unhealthy?’); and a list of chronic health conditions (hypertension, arthritis, diabetes, depression, cancer, heart, lung, gastric, neurological, thyroid and kidney diseases, hearing and seeing problems, and other) reported on a yes/no scale.

Procedure

Participants completed measures at each participating centre. One research assistant was present while participants completed the questionnaire in case they needed assistance. All information was self-reported. All participants provided written informed consent. This study was part of a larger project, the WHOQOL-OLD which was aimed to develop a cross-cultural specific module to appraise QoL in older people. The project was financed by the European Commission Fifth Framework Programme (QLRT-2000-00320).[24]

Statistical analysis

Acceptability, reliability and validity were assessed using standard psychometric methods. To assess acceptability, we examined floor and ceiling effects and missing data for summary score. Floor and ceiling effects were present if more than 15% of respondents achieved the lowest or highest possible score. Internal consistency reliability was assessed using Cronbach's alpha coefficient (≥0.70).[25] To assess the construct validity of the WHO-5, exploratory factor analyses were performed. We used principal components method and factor loadings ≥0.40 as a criterion to define a ‘salient’ factor loading. Convergent validity assessed the association of the WHO-5 with the WHOQOL-BREF domains, and GDS-15. Spearman's rho correlation coefficients were used to explore related factors, and we considered a correlation of <0.3 to be small, 0.3–0.5 moderate and ≥0.5 high.[25] Subsequently, the differences between sex, age groups, marital status and education were examined. We also compared WHO-5 scores for subgroups of participants ‘a priori defined’[26] on the basis of self-perceived health (healthy vs unhealthy); number of chronic health conditions (<4 vs ≥4), based on the mean number of health problems reported by the Catalan Health Survey (2002) among elderly in the area; self-reported depression (no vs yes) from the list of chronic health conditions; and significant depressive symptoms as measured by the GDS-15 (non-depressed vs depressed). Student's t-tests were used to examine group differences. A P-value of <0.05 was regarded as statistically significant. All statistical calculations were performed with spss for Windows v19.0 (spss, Chicago, IL, USA).

Results

Sample characteristics

Nearly 62% of participants were female. Mean age for the entire sample was 74.6 years (SD ± 7.1), with no significant differences between men (73.8 ± 6.8) and women (75.1 ± 7.3) (t [189] = −1.191, P = 0.235). Sixty-six percent considered themselves to be ‘healthy’, but 95.3% reported having at least one or more chronic health conditions on the comorbid list, namely arthritis (57.6%), hypertension (47.1%), seeing problems (41.9%), hearing problems (23.6%), heart conditions (20.9%), and depression (18.8%) (see Table 1).

Table 1. Participant characteristics
Sociodemographic and health status variablesTotal sample, n = 191
  1. GDS-15, Geriatric Depression Scale; SD, standard deviation; WHO-5, World Health Organization-Five Well-being Index.
Age: Mean (SD)74.6 (7.1)
Range65–95
Sex, n (%) 
Male73 (38.2)
Female118 (61.8)
Marital status, n (%) 
Married98 (51.3)
Other partnered (other than married)93 (48.7)
Education, n (%) 
Less than primary school47 (24.6)
Primary school finished & higher144 (75.4)
WHO-5: Mean (SD)15.6 (5.2)
Range0–25
Self-perceived health, n (%) 
Healthy126 (66.0)
Unhealthy65 (34.0)
Chronic health conditions: Mean (SD)4.6 (3.2)
GDS-15: Mean (SD)2.7 (3.1)
Range0–14
GDS-15 ≥ 5 (%)22.5%

Acceptability and reliability

There were no floor/ceiling effects or missing data. The Cronbach's alpha coefficient was 0.86.

Factor structure

Factor analysis showed that the Kaiser–Meyer–Olkin statistic (KMO) was greater than 0.5 (KMO = 0.861) and the Bartlett's test of sphericity was significant (P < 0.0001), indicating an underlying structure in the scale. We found a single-factor structure that accounted for 66% of the variance (Table 2).

Table 2. Factor matrix of the Spanish version of the World Health Organization-5 Well-being Index
Item Factor
Item 1I have felt cheerful and in good spirits0.850
Item 2I have felt calm and relaxed0.841
Item 3I have felt active and vigorous0.769
Item 4I woke up feeling fresh and rested0.746
Item 5My daily life has been filled with things that interest me0.850

Validity

Convergent validity

Spearman's rank correlation coefficients between WHO-5 and WHOQOL-BREF domains were high (r > 0.50) except on the social domain (r = 0.372, P < 0.001). Correlation between WHO-5 and the GDS-15 was also high and negative (r = −0.587, P < 0.001). We also found a significant negative correlation with number of chronic health problems (r −0.444, P < 0.001) (Table 3).

Table 3. Convergent validity of the World Health Organization-5 Well-being Index total score with other measures
Association with:Spearman's rho
  1. *All P-value <0.001.
  2. WHOQOL-BREF, World Health Organization Quality of Life-BREF.
WHOQOL-BREF domains 
Physical0.587*
Psychological0.604*
Social0.372*
Environment0.562*
Geriatric Depression Scale-15−0.587*
Number of chronic health problems−0.444*

Discriminative validity

We did not find significant differences in WHO-5 scores in relation to sociodemographic variables (sex, age, marital status, and education).

A priori defined groups

WHO-5 scores were higher in participants reporting themselves healthy compared to unhealthy (t [189] = 4.854, P < 0.001); those with a lower number of chronic health conditions (<4 vs ≥4) (t [189] = 5.552, P < 0.001), non-depressed (GDS-15 < 5) compared to depressed (GDS-15 ≥ 5) (t [189] = 9.252, P < 0.001) and those who self-reported not having depression (no vs yes) (t [189] = 5.174, P < 0.001) (Table 4).

Table 4. Discriminative validity of the World Health Organization-5 Well-being Index: sociodemographic and health data
 Mean (SD)T-test; d.f.; P-value
  1. d.f., degrees of freedom; SD, standard deviation.
Sex  
Male16.4 (5.2)1.794; 189; 0.074
Female15.0 (5.1)
Age groups  
60–7915.8 (5.0)1.291; 189; 0.198
80+14.7 (5.5)
Marital status  
Married15.9 (4.5)0.855; 189; 0.394
Non-married15.2 (5.8)
Education  
Primary school finished & higher15.9 (4.8)1.534; 189; 0.127
Less than primary school14.5 (6.0)

A priori defined groups

Self-perceived health:

  
Healthy16.8 (4.5)4.854; 189; <0.001
Unhealthy13.2 (5.5)
Chronic health problems  
<417.8 (3.5)5.552; 189; <0.001
≥413.9 (5.5)
Geriatric Depression Scale -15  
Non-depressed <517.1 (4.3)9.252; 189; <0.001
Depressed ≥510.2 (4.1)
Self-reported depression  
No16.4 (4.4)5.174; 189; <0.001
Yes11.8 (6.4)

DISCUSSION

The aim of this study was to examine the psychometric properties of the WHO-5 Spanish version on elderly persons. Internal consistency found in this study was acceptable and similar to that previously reported by other authors.[11, 16, 17, 27] Results of the exploratory factor analysis on this study were similar to findings reported by other authors[16, 17] and the factor structure we found was consistent with the hypothesized factor structure for the WHO-5.[17] Convergent validity was good, as shown by the high correlations between WHO-5 and WHOQOL-BREF domains, and between WHO-5 and GDS-15. In previous studies, other authors reported moderate–high correlations between the WHO-5 and other depression screening scales, for example, between the WHO-5 and the Hamilton Depression scale[17] in primary care patients; and between the WHO-5 and Zung's Self-Rating Depression Scale. Similarly, other authors found moderate–high correlations between the WHO-5 and the Short-Form 36 Health Survey questionnaire (SF-36) in diabetic patients in Japan[16] and between the WHO-5 and the Depression Anxiety Stress Scale – Depression and some of the subscales of the SF-36.[18]

In terms of discriminative validity, regarding associations between the WHO-5 and sociodemographic variables (age, sex, level of education) we did not find significant differences. Thus, our findings do not support some previous studies associating lower levels of well-being with increasing age.[3, 28] Finally, health, those with a lower number of comorbidities, and non-depressed had higher levels of well-being, confirming that lower levels of well-being are associated with depressive symptoms, impaired physical health and chronic conditions.[12] Overall, the WHO-5 score for the total sample in this study was similar to scores in samples from Germany,[15] Japan,[16] and Thailand.[17] A cut-off point less than 13 has been associated with depression.[17, 27] If we take this cut-off as reference, we found that those probably depressed (GDS-15 ≥ 5) and those self-reporting having depression on the comorbid condition list scored lower. The WHO-5 may be a suitable instrument to screen for low emotional well-being and depressive affect among older adults.[12, 27] The use of the WHO-5 could improve the capacity to detect depression in primary care centers; it might be useful for physicians, nurses and social workers. The questionnaire can easily be scored by hand.[13] Persons who score positively for depression should be further appraised in order to confirm a diagnosis of depression or to rule out normal distress or physical causes of depression;[13] the consequences of ignoring social and medical support among depressed elderly have been emphasized by several authors.[29, 30]

Limitations of the study

We used a convenience sample and it is difficult to say with any precision how representative these results are for the wider population. Yet, basic sociodemographic characteristics, such as sex and educational level, did not differ from those reported on the Catalan Health Survey (2002). We did not collect information on the stability of the measure (test–retest).

To our knowledge, this is the first study to examine the psychometric properties of the WHO-5 in elderly Spanish persons; our findings show good psychometric properties. Given its brevity, it might be useful in measuring depressive symptoms in clinical practice, primary care, nursing homes, and health surveys. As the WHO-5 is available and has been validated in different languages, it might also be useful in cross-cultural studies.

Acknowledgments

We wish to thank all participants who took part in the study. We would also like to thank the professionals who provided information about the study to participants.

The study was funded by the European Commission Fifth Framework, QLRT-2000-00320, and was carried out under the auspices of the World Health Organization Quality of Life Group (WHOQOL Group). The funder did not have any role in the analysis of the data or in the preparation of the manuscript.

Dr Lucas-Carrasco had full access to the data and was responsible for carrying out the study, data analysis, and interpretation of results and preparation of the manuscript.

Ancillary