Scholarlink

Purpose <p>To evaluate the reliability and construct validity of the five-item World Health Organization Well-Being Index (WHO-5) in a nationwide sample of adults recently discharged from specialised inpatient mental health care in Norway.</p> Methods <p>A total of 2310 patients participated in the national survey in early 2024. The WHO-5 was assessed for missing data, floor effects, internal consistency, and factor structure using exploratory and confirmatory factor analysis (EFA, CFA). Measurement invariance across sex, age, and education was tested with multi-group CFA. Construct validity was evaluated through correlations with subjective well-being, self-reported mental and physical health, and treatment-related enablement. Item response theory (IRT) analyses assessed item discrimination and thresholds.</p> Results <p>Item-level missingness was low (&lt;1%), with moderate to high floor effects for two items. Internal consistency reliability was high (Cronbach’s alpha: 0.910). EFA supported a one-factor solution explaining 71.7% of the variance, confirmed by CFA; fit indices were strong for CFI/TLI/SRMR, while RMSEA was elevated, consistent with very low degrees of freedom in a five-item model. Configural, metric, and scalar invariance were supported across sex, age, and education. WHO-5 correlated strongly with life satisfaction, meaning in life, and self-perceived mental health, and moderately with physical health and treatment-related enablement. IRT indicated adequate to high discrimination; thresholds were ordered but closely spaced in the mid-range, and coverage primarily spanned low to moderate levels of well-being.</p> Conclusion <p>The WHO-5 demonstrated strong reliability, clear unidimensionality, and supported invariance in this population, supporting its use as a brief, generic PROM in mental health care. Observed floor effects and overlapping mid-range categories suggest scope for response-format refinement to improve measurement precision.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Psychometric performance of the WHO-5 well-being index in a nationwide sample of inpatients discharged from specialised mental health care

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月29日

查看全文
  • Hilde Hestad Iversen,
  • Marte Karoline Råberg Kjøllesdal,
  • Lina Harvold Ellingsen-Dalskau,
  • Mona Haugum,
  • Oyvind Bjertnaes

摘要

Purpose

To evaluate the reliability and construct validity of the five-item World Health Organization Well-Being Index (WHO-5) in a nationwide sample of adults recently discharged from specialised inpatient mental health care in Norway.

Methods

A total of 2310 patients participated in the national survey in early 2024. The WHO-5 was assessed for missing data, floor effects, internal consistency, and factor structure using exploratory and confirmatory factor analysis (EFA, CFA). Measurement invariance across sex, age, and education was tested with multi-group CFA. Construct validity was evaluated through correlations with subjective well-being, self-reported mental and physical health, and treatment-related enablement. Item response theory (IRT) analyses assessed item discrimination and thresholds.

Results

Item-level missingness was low (<1%), with moderate to high floor effects for two items. Internal consistency reliability was high (Cronbach’s alpha: 0.910). EFA supported a one-factor solution explaining 71.7% of the variance, confirmed by CFA; fit indices were strong for CFI/TLI/SRMR, while RMSEA was elevated, consistent with very low degrees of freedom in a five-item model. Configural, metric, and scalar invariance were supported across sex, age, and education. WHO-5 correlated strongly with life satisfaction, meaning in life, and self-perceived mental health, and moderately with physical health and treatment-related enablement. IRT indicated adequate to high discrimination; thresholds were ordered but closely spaced in the mid-range, and coverage primarily spanned low to moderate levels of well-being.

Conclusion

The WHO-5 demonstrated strong reliability, clear unidimensionality, and supported invariance in this population, supporting its use as a brief, generic PROM in mental health care. Observed floor effects and overlapping mid-range categories suggest scope for response-format refinement to improve measurement precision.