Analysis of the PHQ-15 by individual items: a systematic review and meta-analysis

Background: The Patient Health Questionnaire-15 (PHQ-15) is among the most commonly used questionnaires to assess somatic symptoms in primary care settings. The aim of our study was to conduct a systematic review and meta-analysis of the studies that have analysed individual items from the PHQ-15 to determine the influence of differences in sociodemographic and clinical variables. Methods: This systematic review was conducted in accordance with the PRISMA statement. Studies analysing PHQ-15 by individual items were searched in Medline/PubMed, Embase, PsycINFO, Scopus from 1999 to November 2019. Results: Data were extracted from 13 studies. The most frequently reported symptoms were "pain in your arms, legs, or joints", "back pain", "headaches", and "feeling tired or having low energy". In patients with mental health disorders (depression, anxiety disorder, and somatoform disorder), the most commonly reported somatic symptoms were "trouble sleeping", "feeling tired or having low energy", and digestive symptoms. In the meta-analysis, females scored significantly higher than males on 3 of 14 items: "headaches" (odds ratio [OR] 2.16, p<.001), "dizziness" (OR 1.58, p<.001), and "stomach pain" (OR 1.46, p<.02), with a small to moderate effect size for these differences. Conclusions: Gender differences were detected for three items (headaches, dizziness, and stomach pain), although the magnitude of these differences was small. Further research is needed to better understand the relationship between individual items and other sociodemographic variables. In addition, more research is needed to determine measurement invariance across various demographic groups.

Introducción: El PHQ-15 (Patient Health Questionnaire) es uno de los cuestionarios más utilizados para la valoración de los síntomas somáticos en atención primaria. El objetivo del estudio fue realizar una revisión sistemática y un metaanálisis de los estudios previos que analizaron los ítems individuales del PHQ-15 y determinar su asociación con las variables clínicas y sociodemográficas. Metodología: Esta revisión sistemática se llevó a cabo de acuerdo a la declaración PRISMA. Los estudios que analizaron el PHQ-15 por ítems individuales fueron obtenidos en las bases de datos de Medline / PubMed, Embase, PsycINFO y Scopus; desde el 1999 hasta noviembre de 2019. Resultados: Los datos fueron extraídos de 13 estudios. Los síntomas más frecuentes fueron "dolor en los brazos, piernas o articulaciones", "dolor de espalda", "dolor de cabeza" y "sentirse cansado o con poca energía". Los pacientes con trastornos mentales (depresión, ansiedad y trastorno somatomorfo), presentaron con mayor frecuencia "problemas de sueño", "sentirse cansado o con poca energía" y síntomas gastrointestinales. En el metaanálisis, las mujeres obtuvieron puntuaciones significativamente mayores que los hombres en 3 de los 14 ítems: "dolor de cabeza" (odds ratio [OR] 2.16, p<.001), "mareo" (OR 1.58, p<.001) y "dolor de estómago" (OR 1.46, p<0,02), con un tamaño del efecto de pequeño a moderado. Conclusiones: Se detectaron diferencias entre ambos géneros en tres ítems (cefalea, mareo y dolor de estómago), aunque la magnitud de estas diferencias fue pequeña. Es necesario seguir investigando para lograr una mejor comprensión de la relación entre los ítems individuales y las variables sociodemográficas. Además, es necesario investigar la invarianza de medición en los diferentes grupos demográficos

STEPPS for Borderline Personality Disorder: A Pragmatic Trial and Naturalistic Comparison With Noncompleters.

Systems Training for Emotional Predictability and Problem Solving (STEPPS) is a group psychotherapy approach that is effective for patients with borderline personality disorder (BPD) in a public health care setting. The sample in this study comprised 118 outpatients with BPD who were asked to participate in a Spanish-adapted version of STEPPS for 18 months, a psychotherapy program that could be added to their usual psychiatric intervention. They were divided into an experimental group who participated in STEPPS, and a control group, who received treatment as usual. Several variables were collected and the Borderline Evaluation of Severity Over Time (BEST) scale was administered at pretest, Months 3 and 6, posttest (Month 18), and 2-year follow-up (Month 42), after which a post hoc data analysis was carried out. The STEPPS program improved the as-usual treatment provided previously, and the results were cost-effective. A higher educational level and good patient collaboration predicted better outcome.