Adaptación y validación de las escalas QPC "Quality in Psychiatric Care" en el ámbito de la hospitalización

Abstract

[spa] INTRODUCCIÓN: Los países occidentales comparten cada vez más el interés en evaluar la calidad de la atención en el ámbito de la salud mental. Sin embargo, se sabe poco sobre cómo perciben la calidad de los cuidados, los pacientes y profesionales en unidades de hospitalización de psiquiatría. Son pocos los instrumentos que se conocen y que estén adecuadamente validados entorno a la calidad de la atención en el ámbito de la salud mental, y especialmente en el entorno hospitalario. Uno de los más relevantes es el “Quality in Psychiatric Care” (QPC).

OBJETIVOS: Los objetivos generales fueron la traducción y adaptación al castellano de la escala “Quality in Psychiatric Care-Inpatient” (QPC-IP) y de la escala “Quality in Psychiatric Care-Staff” (QPC-IPS) y el análisis de las propiedades psicométricas. Otro objetivo fue analizar la invarianza configuracional de los ítems y las subescalas de los instrumentos utilizados.

METODOLOGÍA:

La investigación se organizó en tres estudios. La primera fase de los estudios 1 y 2 consistió en el proceso de traducción y retrotraducción de las escalas “Quality Psychiatric Care” del idioma sueco al castellano.

En la segunda fase, para el estudio 1 (Spanish QPC-IP) se reclutaron 150 pacientes que cumplían con los criterios de inclusión: mayores de 18 años, diagnosticados de trastorno mental según la clasificación DSM-V y estar ingresados en el momento del estudio.

En la segunda fase, para el estudio 2 (Spanish QPC-IPS) se reclutaron a 163 profesionales que cumplían con los criterios de inclusión: profesionales asistenciales que estaban activos laboralmente y que participaran voluntariamente en el estudio.

El tipo de muestreo utilizado fue no probabilístico, dado que la población incluida en ambos estudios fueron los que voluntariamente quisieron participar.

En el estudio 1 y 2 se realizó un análisis descriptivo de los ítems de las escalas y de las variables sociodemográficas y clínicas de la muestra. La fiabilidad se realizó analizando la consistencia interna y la estabilidad temporal, administrando la escala de nuevo a los 7-14 días. Se realizó el análisis de la correlación de la escala Spanish QPC-IP con el cálculo del coeficiente de correlación de Pearson con el fin de determinar la validez convergente y discriminante. Sin embargo, en el estudio 2, la validez convergente se analizó mediante el análisis de correlación de Spearman (rho) con la escala de satisfacción laboral de NTP 394 administradas al mismo tiempo. Posteriormente, se realizó un análisis factorial confirmatorio.

Los análisis del estudio 3 se llevaron a cabo en una muestra total de 578 participantes (incluidos los pacientes suecos que completaron la Sweden QPC- IP). En la primera fase, se utilizó el análisis factorial confirmatorio y en la segunda fase, la estructura de medición fue confirmada a través del Modelo Exploratorio de Ecuaciones Estructurales (ESEM).

Para el análisis estadístico de los datos del estudio 1 y 2 se utilizó el paquete estadístico IBM® SPSS® Statistics 22.0 y el EQS® versión 6.2. para el análisis factorial confirmatorio. Los estudios 1, 2 y 3 fueron aprobados por el Comité de Ética e Investigación Clínica de la Fundation Research de Sant Joan de Déu, CEIC PIC-128-15 de Barcelona (España).

[eng] INTRODUCTION:

Western countries are sharing increasing interest in evaluating and improving care quality in the setting of mental health. Nonetheless, little is known about how quality is perceived by patients and professionals in psychiatric inpatient units. There are a number of challenges, one of which is the fact that there is no general consensus on how to define quality, as it is considered a multidimensional concept. Secondly, unlike the concept of patient satisfaction, care quality includes the perspectives of all interested parties.

OBJECTIVES:

The general objectives were the translation and adaptation to Spanish of the “Quality in Psychiatric Care-Inpatient” (QPC-IP) questionnaire and of the Spanish of the “Quality in Psychiatric Care-Inpatient Staff” (QPC-IPS) and the analysis of psychometric properties in a sample of patients admitted to psychiatric inpatient units in two leading hospitals in the province of Barcelona (study 1 and study 2) . Study 3 analysed the configuration invariance of the items and domains of the instruments with the Swedish items and domains of the “Quality in Psychiatric Care-Inpatient” (Sweden QPC-IP) questionnaire.

METHODOLOGY:

The questionnaires underwent a process of transcultural adaptation to Spanish, and in the second phase a metric analysis of the questionnaires was conducted by means of a cross-sectional study. For study 1 (Spanish QPC-IP) 150 patients were recruited, all of whom met the established inclusion criteria: over 18 years of age, with a diagnosis of a mental disorder according to the DSM-V classification, and at the time of the study being an inpatient of one of the psychiatric units. In the second phase, for study 2 (Spanish QPC-IPS) a total of 163 professionals were recruited, all of whom met the following inclusion

criteria: healthcare professionals, currently active, working in the psychiatric care units, and who wished to participate voluntarily in the study.

The IBM® SPSS® Statistics 22.0 package was used for the statistical analysis of the data from studies 1 and 2, while EQS® 6.2 was used for the confirmatory factor analysis.

In study 1 and in the study 2, a descriptive analysis of the items in the Spanish QPC-IP questionnaire and of the socio-demographic and clinical variables of the sample was conducted. Reliability was assessed by analysing internal consistency and temporal stability, after 7-14 days. The correlation analysis for the Spanish QPC-IP questionnaire was conducted with Pearson correlation coefficient, with the aim of determining the convergent and discriminant validity. In study 2, the convergent validity was analysed by Spearman's rank correlation coefficient (rho) of the Spanish QPC-IPS with the NTP 394 job satisfaction questionnaire administered at the same time. A confirmatory factor analysis was performed.

Study 3 was conducted in the first half of 2019. The total sample consisted of 578 participants. The configuration invariance study was conducted in two phases: in the first phase of the analysis, the confirmatory factor analysis was used; and in the second phase, the measuring structure was confirmed with the exploratory structural equation model (ESEM).

Studies 1, 2 and 3 were approved by the Independent Ethics Committee of the Sant Joan de Déu Research Foundation CEIC PIC-128-15 of Barcelona (Spain).

RESULTS: Study 1: the results of the first phase yielded an instrument adapted and translated into Spanish by a process of translation and back-translation by independent translators. A Cronbach's Alpha of 0,94 was obtained for the entire

QPC-IP instrument and values of 0,52-0,89 for the dimensions of the questionnaire. The intraclass correlation coefficient for the QPC-IP questionnaire was 0,69, whilst values of 0,62-0,74 were obtained in the individual dimensions, indicating an acceptable level of temporal stability. In terms of convergent and discriminant validity, dimension D1 (Therapeutic relationship) and dimension D2 (Patient participation) were more strongly correlated with the overall questionnaire (r=0,902 and r=0,892, respectively), and the stronger correlation between the sub-questionnaires was represented by dimension D1 (Therapeutic relationship) and dimension D3 (Support received) (r=0,744). The confirmatory factor analysis for the Spanish QPC-IP questionnaire revealed that the six factors are identical to those of the original Swedish questionnaire and the model fit indices are adequate.

Study 2: The result obtained in the first phase was satisfactory in all four stages. A Cronbach's Alpha of 0.92 was obtained for the entire instrument and values of 0,45-0,85 for the dimensions of the questionnaire. With regard to test- retest reliability, the intraclass correlation coefficient was 0,91, whilst in the individual dimensions, values of 0,68-0,89 were obtained, indicating a good level of temporal stability. The convergent validity showed a positive correlation (0,58) with the NTP 394 questionnaire. The confirmatory factor analysis revealed that the six factors are the same as those of the original Swedish questionnaire and showed good model fit. The results of the Spanish version show adequate results in terms of validity and reliability.

Study 3: The first phase of study 3 shows that the confirmatory factor analysis of the latent model of the three groups individually does not have a good fit. Nevertheless, the model fit index for the group of Swedish patients is adequate. In the second phase, the ESEM analysis indicates a better model fit, although the group of Spanish patients and professionals obtains lower scores than the adequate fit. The factor loadings for each group were significant. Factor 3 (item 24) showed a greater factor loading in all three study samples. In contrast,

factor 2 (items 1 and 14) showed a low factor loading.

CONCLUSIONS: This research has made it possible to adapt and validate the Spanish QPC-IP and the Spanish QPC-IPS questionnaires. The purpose of these instruments is to measure care quality from the perspective of patients admitted to psychiatric inpatient units and from the perspective of mental healthcare professionals in hospital settings.