Examining the Factor Structure and Discriminant Validity of the 12-Item General Health Questionnaire (GHQ-12) Among Spanish Postpartum Women

Aguado, Jaume; Campbell, Anne; Ascaso, Carlos; Navarro, Purificación; García-Esteve, Lluïsa; Luciano, Juan V.

doi:10.1177/1073191110388146

Cited by 40 publications

(68 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Se compararon los índices de ajuste del modelo original unidimensional, del bidimensional (p.ej., Campo-Arias, 2007Rocha, et al, 2011Villa, et al, 2013), del tridimensional (Graetz, 1991), y el unidimensional con efectos del método (Hankins, 2008), con el propósito de corroborar hallazgos recientes que cuestionan la multidimensionalidad del GHQ-12 (Abubakar & Fisher, 2011;Aguado, et al, 2012;Molina, et al, 2014;Smith, et al, 2013;Ye, 2009). La ausencia de una estructura factorial común a través de estudios dificulta la utilización del GHQ-12 como instrumento de tamizaje de la salud mental general y genera incertidumbre en la toma de decisiones en la práctica clínica.…”

Section: Discussionunclassified

“…Pero algunos estudios han encontrado que ese modelo muestra mejor ajuste que todos los modelos analizados (Ye, 2009), incluyendo dos modelos con efectos del método (Abubakar & Fischer, 2012). Sin embargo, en los modelos de dos y tres factores, se ha hallado que los factores están fuertemente correlacionados afectando el poder predictivo y de discriminación (Abubakar & Fischer, 2012;Aguado et al, 2012).…”

Section: Discussionunclassified

“…Los análisis de datos consistieron en analizar las varianzas de los ítems con intervalos de confianza (IC) de 95% con la intención de corroborar si los ítems negativos mostraban mayor varianza que los positivos; además, la obtención de la matriz de correlaciones Pearson de los ítems, incluyendo medias, desviaciones, asimetría y curtosis fueron incluidas para explorar la naturaleza de los tipos de respuesta ante ítems positivos versus negativos (Ye, 2009) dados los argumentos a favor de la unidimensionalidad del GHQ tanto como a favor de la multidimensionalidad. Los modelos evaluados se pueden dividir en tres tipos de acuerdo a varios autores (Aguado, et al, 2012): 1) el modelo unidimensional original y el modelo unidimensional con efectos del método (errores correlacionados para los ítems parafraseados en forma negativa; Hankins, 2008); 2) modelo bifactorial con factores correlacionados (p. ej., Cifre & Salanova, 2000); y 3) modelo trifactorial con factores correlacionados (p.…”

Section: Análisis De Datosunclassified

“…Sugieren también la opción de no considerar los ítems negativos al calificar el instrumento. Aguado et al (2012) exploran la estructura factorial y discriminante del GHQ-12 en mujeres españolas en fase postparto. En su estudio comparan el mejor ajuste de los datos de múltiples modelos, encontrando, nuevamente, que los modelos bi y tridimensionales no solo no sobresalen por su ajuste, si no que presentan correlaciones entre factores muy altas.…”

Section: Introductionunclassified

“…Confirman que la estructura multidimensional es un artificio de estilos de respuesta (debida a los ítems negativos) y se suman a la evidencia a favor del GHQ-12 como medida unidimensional. Debido a que Aguado et al (2012) solo estudiaron a mujeres y encontraron resultados semejantes a otros incluyendo a mujeres y hombres, la estructura factorial del GHQ-12, con efectos de método, parece adecuada para ambos sexos. En cuanto al poder discriminativo encontraron que solo el primer factor del modelo tridimensional discrimina y solo lo hace entre personas con diagnostico o sin él; por lo que no es más eficiente que la puntuación total del GHQ-12.…”

Section: Introductionunclassified

See 4 more Smart Citations

Estructura factorial del cuestionario de salud general GHQ-12 en población general de México.

et al. 2016

View full text Add to dashboard Cite

RESUMENOBJETIVO: conocer la estructura factorial del GHQ-12 en México y explorar su utilidad para identificar a personas que perciben tener problemas de salud. METODO: participaron 1093 adultos sin diagnóstico de enfermedades crónicas; 462 fueron varones, con una edad media de 32.2 años (DT=13.2) y 631 mujeres, con una edad media de 29.1 años (DT=12.0). Se realizaron comparaciones de correlaciones, medias y varianzas para explorar los efectos de la redacción de ítems en forma negativa versus positiva. Posterior al AFE, se compararon los índices de ajuste de los modelos multidimensionales, del unidimensional original y del unidimensional con 'efectos del método'. Se comparó la percepción general de salud y los puntajes del GHQ-12 para mujeres y hombres. RESULTADOS: el modelo unidimensional con 'efectos de método' (errores correlacionados) mostró el mejor ajuste a los datos. Los puntajes de ítems positivos, negativos y el total fueron más altos para participantes con problemas de salud, pero solo las varianzas e intervalos de confianza de los ítems positivos fueron consistentes y homogéneos. CONCLUSION: se apoya que el GHQ-12 es una medida unidimensional. Aplicado en población mexicana muestra una estructura factorial semejante a la de países orientales y occidentales. Se sugiere utilizar el puntaje de los ítems positivos del instrumento como un método general de tamizaje del bienestar/malestar psicológico.PALABRAS CLAVE: Análisis factorial confirmatorio, efectos de redacción, estructura factorial, GHQ-12, México. ABSTRACT OBJECTIVES: 1. To know the factor structure of the GHQ-12 in Mexican adult population. 2. To explore its usefulness to identify individuals who report health problems. METHOD: The study was carried out with a sample composed of 1093 adults without diagnosis of chronic diseases, divided into 462 males with an average age of 32.2 (DT = 13.2) and 631 females with an average age of 29.1 (DT = 12.0). In order to explore worded items effects positively and negatively, correlations, means, and variances were compared. Following the EFA, the adjustment of index of multidimensional models (the 'original' unidimensional model and the unidimensional with method effects) was compared. Also comparisons between self-reported health and GHQ-12 scores for both sexes were performed. RESULTS: The unidimensional model with 'method effects' obtained the best adjustment. Scores from positive, negative, and total items were higher for participants with self-reported health problems; however, only variances and confidence intervals with positive items were found homogeneous and consistent. CONCLUSION: GHQ-12 is confirmed as a unidimensional measure. Structural factor is similar to that of western and eastern countries when applied to Mexican population. Using the scores of GHQ-12 positive items as a screening method of psychological welfare/psychological is advised.

show abstract

Section: Discussionunclassified

Section: Análisis De Datosunclassified

Section: Introductionunclassified

See 3 more Smart Citations

Estructura factorial del cuestionario de salud general GHQ-12 en población general de México.

et al. 2016

View full text Add to dashboard Cite

show abstract

Assessment of Patient-Reported Outcome Measures for Maternal Postpartum Depression Using the Consensus-Based Standards for the Selection of Health Measurement Instruments Guideline

et al. 2022

View full text Add to dashboard Cite

IMPORTANCE Maternal depression is frequently reported in the postpartum period, with an estimated prevalence of approximately 15% during the first postpartum year. Despite the high prevalence of postpartum depression, there is no consensus regarding which patient-reported outcome measure (PROM) should be used to screen for this complex, multidimensional construct. OBJECTIVETo evaluate psychometric measurement properties of existing PROMs of maternal postpartum depression using the Consensus-Based Standards for the Selection of Health Measurement Instruments (COSMIN) guideline and identify the best available patient-reported screening measure. EVIDENCE REVIEW This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline. PubMed, CINAHL, Embase, and Web of Science were searched on July 1, 2019, for validated PROMs of postpartum depression, and an additional search including a hand search of references from eligible studies was conducted in June 2021. Included studies evaluated 1 or more psychometric measurement properties of the identified PROMs. A risk-of-bias assessment was performed to evaluate methods of each included study. Psychometric measurement properties of each PROM were rated according to COSMIN criteria. A modified Grading of Recommendations Assessment, Development, and Evaluation approach was used to assess the level of evidence supporting each rating, and a recommendation class (A, recommended for use; B, further research required; or C, not recommended) was given based on the overall quality of each included PROM. FINDINGS Among 10 264 postpartum recovery studies, 27 PROMs were identified. Ten PROMs (37.0%) met the inclusion criteria and were used in 43 studies (0.4%) involving 22 095 postpartum women. At least 1 psychometric measurement property was assessed for each of the 10 validated PROMs identified. Content validity was sufficient in all PROMs. The Edinburgh Postnatal Depression Scale (EPDS) demonstrated adequate content validity and a moderate level of evidence for sufficient internal consistency (with sufficient structural validity), resulting in a recommendation of class A. The other 9 PROMs evaluated received a recommendation of class B. CONCLUSIONS AND RELEVANCEThe findings of this systematic review suggest that the EPDS is the best available patient-reported screening measure of maternal postpartum depression. Future studies should focus on evaluating the cross-cultural validity, reliability, and measurement error of the EPDS to improve understanding of its psychometric properties and utility.

show abstract

Screening Tools and Methods of Identifying Perinatal Depression

Mann¹,

Evans²

2015

Identifying Perinatal Depression and Anxiety

View full text Add to dashboard Cite

Examining the Factor Structure and Discriminant Validity of the 12-Item General Health Questionnaire (GHQ-12) Among Spanish Postpartum Women

Cited by 40 publications

References 50 publications

Estructura factorial del cuestionario de salud general GHQ-12 en población general de México.

Estructura factorial del cuestionario de salud general GHQ-12 en población general de México.

Assessment of Patient-Reported Outcome Measures for Maternal Postpartum Depression Using the Consensus-Based Standards for the Selection of Health Measurement Instruments Guideline

Screening Tools and Methods of Identifying Perinatal Depression

Contact Info

Product

Resources

About