Validación de la escala Kessler 10 (K-10) en la detección de depresión y ansiedad en el primer nivel de atención. Propiedades psicométricas Translated title: Validity of the kessler 10 (k-10) scale in detecting depression and anxiety disorders in the primary care services. Psychometric properties

Abstract

De acuerdo con estudios realizados en diferentes países se estima que aproximadamente hay entre 30% a 50% de personas que presentan algún problema de salud mental que no es reconocido por el médico general. En virtud de esta situación se ha propuesto como estrategia a la atención primaria como base del sistema de salud, lo que permitiría la detección temprana de pacientes con algún trastorno psiquiátrico. Diferentes organismos señalan que una de las primeras acciones para cumplir con los objetivos en la atención de personas con algún trastorno mental, consiste en contar con una escala breve, autoaplicable, válida y confiable y de bajo costo. En este trabajo se presentan las características psicométricas de la escala Kessler (K-10) para detectar trastornos depresivos y ansiosos. La K-10 es un instrumento de tamizaje breve y de fácil aplicación por el personal del primer nivel de atención y ha sido utilizada en diferentes estudios a nivel poblacional. En Australia, en 1997, se aplicó la K-10 en una encuesta de salud, por medio del Consejo Nacional de Encuestas de Salud Mental. Material y métodos Se trata de un estudio de proceso metodológico, cuyo objetivo fue la validación de la escala de malestar psicológico K-10 de Kessler. El estudio se llevó a cabo en dos Centros de Salud del primer nivel de atención en la Ciudad de México. Los participantes fueron 280 personas que acudieron a la consulta externa de dichos centros. Se utilizaron los criterios del DSM-IV para la confirmación del diagnóstico de depresión y de ansiedad, por medio de la Mini International Neuropsychiatric Interview (MINI), en su versión computarizada, adaptada al español latinoamericano en el Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz. La validez diagnóstica se procesó utilizando los diagnósticos de la MINI para depresión y ansiedad como regla de oro y las puntuaciones obtenidas en la escala K-10 como predictor. Se calculó la sensibilidad y especificidad para todos los posibles puntos de corte con el fin de establecer el óptimo. Se calculó adicionalmente la eficiencia y las razones de máxima verosimilitud, así como el área bajo la curva ROC y los cocientes de probabilidad, positivo y negativo (LR+ y LR-). Resultados Del total de personas a quiénes se les aplicó la escala, el 78.9% (221) fueron mujeres y 21.1% (59) hombres. Estos valores representan la proporción en que los pacientes acuden a los servicios de primer nivel (IC 95%=±5.4%). El 70.6% de las mujeres presentaron mayor malestar psicológico en comparación con los hombres que representaron el 52.5% [χ²(1)=6.05,p=0.014]. En ninguna otra variable socio-demográfica se presentaron diferencias significativas. El instrumento tiene una alta precisión, ya que puede detectar hasta el 87% de los casos de depresión y un 82.4% de los casos de ansiedad. La escala se comparó con el MINI en español y presentó una prevalencia de 26.8% y 26.4%, respectivamente. Conclusiones El instrumento cumple con los criterios de validez y confiabilidad, por lo que se recomienda su uso por los médicos generales en el primer nivel de atención. Dado que esta recomendación sólo es una de las actividades que se requieren en la atención primaria para un buen manejo de los casos que se detecten, es necesario señalar que se requiere de un modelo de atención integral que incorpore tanto la detección como el tratamiento farmacológico y psicosocial.

Translated abstract

According to studies conducted in different countries, it is estimated that approximately 30% to 50% of people with mental health problems are not recognized by the general practitioner. Given this situation, it has been proposed that the practitioner at the primary care services must play a decisive role in the early detection of cases by establishing a definitive diagnostic and a timely treatment. Several organizations have pointed out that one of the first actions that need to be implemented to fulfill the aims in the care of people with mental disorders is to prepare the first-contact doctors and to have a brief, low cost, self-applied, valid and reliable scale. The studies mention that using screening tests at the primary care level is crucial for the success of the programs. The detection and recognition of psychiatric symptomatology rates vary depending on the type of scale applied. The tools that have been widely used are the Goldberg's General Health Questionnaire (GHQ), Zung Self-Rating Depression Scale, Beck Depression Inventory, the Depression Symptom Checklist (DS 20), the Hopkins Symptom Checklist (SCL), the Hamilton Depression Scale, the Center for Epidemiologic Studies Depression Scale (CES-D), the Montgomery-Asberg Depression Rating Scale, the Geriatric Depression Scale (GDS), the self-administered computerized assessment (PROQSY), the criteria of the 3rd revised edition of the Diagnostical and Statistical Manual of Mental Disorders (DSM-III-R), the Structured Clinical Interview for DSM-IV (SCID), and the criteria of the Symptom Driven Diagnostic System for Primary Care (SDDS-PC), among others. The preliminary results confirm the existence of a high percentage of possible psychiatric cases (46.9%), but only 4% of cases are referral. The low capability of the general practitioner at the primary care level in detecting these pathologies has been confirmed as well. These scales have been applied in different scenarios and to different types of population. Although the dominating criteria for choosing the tool are sensitivity and specificity, some authors mention that strategies for adequately handling cases, such as the confirmation of the diagnosis and follow-up of the patients, are required once the treatment has started. In this paper, we present the psychometric characteristics of the Kessler (K-10) scale in detecting depression and anxiety disorders in the primary care. Material and methods The study is a methodological process that aims to validate the Kessler Psychological Distress scale (K-10). It was conducted in two health care centers of primary care level in Mexico City. The subjects were 280 individuals who requested attention at the mentioned centers and to whom the K-10 test was applied after giving their informed consent. Later on, the computerized version of the International Neuropsychiatric Interview (MINI), which uses the diagnostic criteria of the DSM-IV, was applied to the subjects in order to confirm the diagnostics for depression and anxiety. The MINI is a version adapted to Latin American Spanish by the National Institute of Psychiatry Ramon de la Fuente Muñiz. The diagnostic accuracy was processed following the MINI diagnoses for depression and anxiety closely, and the scores on the scale K-10 as a predictor. The sensitivity and specificity were calculated for all possible cut points in order to establish the optimal cut off point. The efficiency and maximum likelihood ratios were also calculated. The area under the ROC curve as well as the probability quotients, positive and negative (LR+ and LR-), were also calculated. The K-10 is a brief screening tool that can be easily applied by the primary care personnel which measures the psychological distress of a person during the four weeks prior to the application. It consists of ten questions with Likert-like answers that range from 1 to 5 and are categorized in a five level ordinal scale: Always, Very Often, Sometimes, Rarely, Never; where «Never» has an assigned value of 1, and «Always» has assigned value of 5. It has a minimum score of 10 and a maximum of 50. The ranges of the instrument are four levels: low (10-15), moderate (16-21), high (22-29) and very high (30-50). The instrument showed an internal consistency of 0.90 and it has been used in various population studies promoted by the World Health Organization as well as government organizations in Australia, Spain, Colombia and Peru. Results Out of 280 individuals to whom the tool was applied, 78.9% (221) were female and 21.1% (59) male. These values represent the proportion of patients attending the primary care services (95% confidence interval=±5.4%). The mean age of women was 39 years, and the mean age of men was 41. The 70.6% of the women manifested more psychological distress than men (52.5%)[χ2(1)=6.05,p=0.014. No other socio-demographic variable showed significant differences. The instrument is highly precise, it can detect up to 87% of depression cases, and 82.4% of anxiety cases. The scale was compared with the MINI and it presented a prevalence of 26.8% and 10.6%, respectively. Of the total of depression cases, 26.4% also presented anxiety; these represent a co-morbidity of 5.4%. The construct validity presented one factor alone that explains the 53.4% of the total variance, this is why the scale is considered as one-dimensional. In other words, the scale only measures the construct of the psychological distress. The internal consistency was α=0.901. Once the sensitivity and specificity for all cut off points had been determined using the MINI as a golden rule, it was observed that the cut off point for maximum sensitivity and specificity corresponded to 21 for the diagnosis of depression, and 22 for anxiety. Conclusions The K-10 is a good instrument for the detection of depression and anxiety cases at the primary care level which meets the criteria of validity and reliability. However, given that only one diagnosis was considered for all the range of anxiety disorders, the scale must be chosen carefully for all the other disorders that are not included in this paper. The use of the instrument is recommended for the general practitioners at the primary care level, mainly for diagnosing depression. Various studies in which other screening instruments have been used for the detection of depressive disorder at primary care point out that any screening method are useful in making the diagnosis. By using these instruments, the depression diagnosis at primary care level increases from 10% to 47%. The latter supports the fact that the selection of a good instrument turns out to be effective in detection, treatment and clinical outcomes of the entity. Since this recommendation is only one of the activities required in primary care level for good handling of detected cases, it is noteworthy to mention that a comprehensive care model that encompasses both the detection as well as the pharmacological and psychosocial treatments is required.