Jordi Alonso

The Spanish version of the Nottingham Health Profile: a review of adaptation and instrument characteristics

The increased interest in measuring health status implies a need for instruments that are appropriate and valid. Adaptation of existing instruments may be a cost-effective strategy. In this paper we describe the adaptation into Spanish of the Nottingham Health Profile (NHP), a self-administered perceived health questionnaire developed in the UK. The characteristics of the adapted questionnaire (validity, reliability, and sensitivity to change), as well as current and purposed applications, are discussed. Some of the principles that were applied in the described adaptation process may be useful for similar future research: involving the investigators that developed the original instrument; using a panel of lay individuals in the translation process; testing the characteristics of the adapted instrument by replicating previous studies with the original instrument, and organizing an international group for the development and use of the European versions of the NHP. Adaptation of health status measures is an opportunity for gaining comparability when measuring health, and for learning about cross-cultural differences in health-related quality of life.

The Spanish version of the Short Form 36 Health Survey: a decade of experience and new developments

OBJECTIVE The Short Form-36 Health Survey (SF-36) is one of the most widely used and evaluated generic health-related quality of life (HRQL) questionnaires. After almost a decade of use in Spain, the present article critically reviews the content and metric properties of the Spanish version, as well as its new developments. METHODS A review of indexed articles that used the Spanish version of the SF-36 was performed in Medline (PubMed), the Spanish bibliographic databases IBECS and IME. Articles that provided information on the measurement model, reliability, validity, and responsiveness to change of the instrument were selected. RESULTS Seventy-nine articles were found, of which 17 evaluated the metric characteristics of the questionnaire. The reliability of the SF-36 scales was higher than the suggested standard (Cronbachs alpha) of 0.7 in 96% of the evaluations. Grouped evaluations obtained by meta-analysis were higher than 0.7 in all cases. The SF-36 showed good discrimination among severity groups, moderate correlations with clinical indicators, and high correlations with other HRQL instruments. Moreover, questionnaire scores predicted mortality and were able to detect improvement due to therapeutic interventions such as coronary angioplasty, benign prostatic hyperplasia surgery, and non-invasive positive pressure home ventilation. The new developments (norm-based scoring, version 2, the SF-12 and SF-8) improved both the metric properties and interpretation of the questionnaire. CONCLUSIONS The Spanish version of the SF-36 and its recently developed versions is a suitable instrument for use in medical research, as well as in clinical practice.

Validation of a minimum outcome core set in the evaluation of patients with back pain

Study Design. Prospective study of patients with subacute osteoporotic fracture (SOF) or chronic low back pain (CLBP). Objective. To evaluate reliability, validity, and responsiveness of a purposefully brief outcome instrument. Summary of Background Data. A minimum standardized “core set” was proposed for monitoring patients with low back pain in 1998, but an assessment of metric properties was still lacking. Methods. The Core Set, SF-36, and Oswestry questionnaires were completed by 154 patients. Test–retest reproducibility was evaluated in a subsample of 43 stable patients with CLBP. Responsiveness was evaluated by estimating effect size (ES) of pre-postsurgery changes in 50 patients with SOF and 23 with CLBP. Results. The total Core Set showed good reproducibility with intraclass correlation coefficients on test–retest near the highest standard of 0.9, whereas internal consistency differed between patients with CLBP and those with SOF (Cronbach’s alpha of 0.92 and 0.64, respectively). Most correlations of the Core Set with SF-36 and Oswestry, previously hypothesized as high, were >0.65, demonstrating good construct validity. Sensitivity to change of the Core Set (ES 0.4–2.3) is similar to the Oswestry (ES 0.7 and 2.3). Conclusions. These findings support the potential usefulness of the Core Set when respondent burden is a major concern. However, subscale scores need to be further tested in other populations before they can be widely recommended.

Expert consensus in the development of a European health-related quality of life measure for children and adolescents: a Delphi study.

Aim: To determine the level of consensus among experts regarding content, structure and sources of content for a new European measure of health‐related quality of life in children and adolescents. Methods: A three‐round Delphi questionnaire was sent by e‐mail to 24 experts in quality of life measurement in 9 European countries. Consensus was considered reached when ≫90% of experts either agreed or disagreed with a given statement, or where median scores were over 6 on a scale of 1–10, and score dispersion was within predefined limits. Results: Completed questionnaires were received from 20 panellists in each round. It was agreed that the new instrument should be a multidimensional, profile measure with 30–49 items covering 5–8 dimensions, which should take no more than 10–15 min to complete. Agreement was also reached on 8 specific dimensions to include in the questionnaire (psychological well‐being, self‐esteem, body image, cognitive functioning, mobility, energy/vitality, social relations, family/home function). Consensus was against the use of individualized questionnaires. Focus groups with children, parents and workers in the field, literature and instrument reviews were considered appropriate sources for content.

Validity of the Spanish version of the Chronic Liver Disease Questionnaire (CLDQ) as a standard outcome for quality of life assessment

The Chronic Liver Disease Questionnaire (CLDQ) measures the impact on quality of life of chronic liver diseases, regardless of underlying etiology. The aim of this study was to develop a Spanish version of the CLDQ, and to assess its acceptability, reliability, validity, and sensitivity to change. The forward and back‐translation method by bilingual translators, with expert panel and pilot testing on patients, was used for the adaptation. The final version was self‐administered, together with the Short Form‐36 Health Survey (SF‐36), on 149 consecutive patients with chronic liver disease. Child‐Turcotte‐Pugh scores were evaluated by a physician. To assess reproducibility and responsiveness the CLDQ was readministered to a subsample of stable patients and to those who had received a liver transplant. Validity was evaluated via exploratory factor analysis, the CLDQ pattern across severity groups, and correlation coefficients with “itching” and SF‐36 scores. Cronbachs alpha and Intraclass Correlation Coefficient for CLDQ global score were 0.93 and 0.90, respectively, demonstrating good reliability. Validity was supported by correlations of the CLDQ with SF‐36 and “itching,” and CLDQ severity gradient (global score means were 5.5, 5.2, 5.0, and 4.5 in patients with no cirrhosis, cirrhosis Child‐Turcotte‐Pugh A, B, and C, respectively; P = 0.012). Responsiveness was shown by a high CLDQ improvement in patients who had received liver transplant (mean change = −1.4; P < 0.001). In conclusion, the Spanish CLDQ is reliable, valid, responsive, and equivalent to the original. These findings support its use as a standard outcome for patients with chronic liver diseases within the whole severity range, from “no cirrhosis” to transplant recipients, both in Spanish and international studies. Liver Transpl 12:95–104, 2006.

Construcción de un filtro geográfico para la identificación en PubMed de estudios realizados en España

Introduccion y objetivos El volumen de PubMed (Medline) hace necesarias herramientas especificas para la recuperacion de informacion (filtros). La actual configuracion de la opcion «Limits» en PubMed no permite restringir las busquedas segun criterios geograficos. Por ello, se plantearon como objetivos: a) la construccion de un filtro geografico para la recuperacion en PubMed de estudios realizados en la poblacion espanola, y b) la evaluacion de su rendimiento. Metodos Se ha construido un filtro de segunda generacion, basado en la revision de estrategias previas, para la seleccion de 3 bloques: a) el termino «Espana» y sus variants en diversos idiomas; b) otros toponimos correspondientes principalmente a comunidades autonomas y provincias, y c) acronimos correspondientes a los servicios autonomicos de salud. Se evaluo el rendimiento del filtro (sensibilidad y especificidad) en una muestra de conveniencia (ensayos clinicos indizados en PubMed bajo el termino MeSH «Myocardial Infarction»), para lo cual se tomo como patron de referencia la revision manual de las referencias. Tambien se comparo el rendimiento del filtro con el de la busqueda simple «spain[ad]». Resultados El filtro ha recuperado 74 (98,7%) de las referencias pertinentes y ha demostrado una sensibilidad (88,1%) muy superior a la de la busqueda simple (45,8%), alcanzandose una especificidad excelente en uno y otro caso (100%). Conclusiones Se dispone de un filtro eficiente para la identificacion de estudios realizados en poblacion espanola en PubMed.

Validity and reliability of the Spanish Version of the Psychological General Well-Being Index

The Psychological General Well-Being Index (PGWBI) was adapted for use in Spain. The American version was translated and back-translated independently by three bilinguals and the comprehensibility of a consensuated version was tested in a pilot study. The construct validity and internal consistency of the questionnaire were tested in 104 patients (52 high blood pressure patients attending a primary health care centre and 52 end-stage renal disease patients undergoing hemodialysis). Concurrently to the PGWBI, patients were administered the General Health Questionnaire (GHQ-12) and three psychosocial categories of the Sickness Impact Profile (SIP-PD). Patients in the dialysis group were administered the PGWBI on two occasions, two weeks apart. The PGWBI was moderately to highly correlated with the GHQ-12 (r=-0.71) and with the SIP-PD (r=-0.69). Overall internal consistency as measured by Cronbachs α coefficient was 0.94. By dimensions, PGWBI α coefficients ranged from 0.56 (Self-control) to 0.88 (Anxiety). No differences were found in mean PGWBI scores between the two groups of patients studied. Two-week test-retest intraclass correlation coefficient was 0.76. the similarity of the operating characteristics supports the contention that the Spanish and American versions of the PGWBI are conceptually equivalent. More research is needed on the reliability and responsiveness of the different dimensions of the questionnaire. Our results suggest that only global scores should be used when making individual comparisons.

Reliability of the Spanish version of the Nottingham health profile in patients with stable end-stage renal disease

OBJECTIVE Since reproducibility of results is a basic prerequisite of health status measures for its use in prospective and evaluative studies, the reliability of the Spanish version of the Nottingham Health Profile (NHP), a multi-dimensional perceived health status measure, was assessed in a sample of stable end-stage renal disease (ESRD) patients. METHODS The NHP was administered on two occasions four weeks apart to a group of hospital hemodialysis program patients who were clinically stable according to their physicians. Correlations of scores and agreement of first and second administrations were assessed together with internal consistency. Afterwards, analyses were repeated taking into account the time (before, during or after the dialysis) and the method of administration (self vs interviewer), and the interviewer. RESULTS Spearman correlation coefficients (rs) between responses to the first and to the second administration were > 0.6 for all of the six dimensions of the NHP (range = 0.69-0.85) and in every sub-group analyzed (P < 0.01). Agreement percent (AP) between items was > 0.4 (0.48-0.65). Internal consistency was 0.91 for the whole profile and > 0.5 (0.58-0.86) when analyzed by individual dimensions. Reliability did not vary significantly with the time nor the method of administration (self or interviewer). CONCLUSIONS Overall, results suggest that the Spanish version of NHP is sufficiently reliable to be used in ESRD patients. While a higher reliability would have been achieved by a shorter retest period, the study provides a realistic approximation to the reliability of the questionnaire in actual research and clinical applications.

Analyzing health outcomes through international comparisons.

There is increasing interest in the conduct of international studies that compare health care outcomes. This paper draws from the literature in sociology, anthropology, and psychology to examine the problems that can be encountered when collecting primary data for comparative purposes. Issues of study design, selection, development, and adaptation of instruments, problems encountered in collecting primary data, and techniques for analyzing comparative data are discussed. The paper is intended as a primer for health services researchers conducting comparative outcomes research in international settings.

Valores poblacionales de referencia del perfil de salud CHIP-AE a partir de una muestra representativa de adolescentes escolarizados

Objetivo: El perfil de salud CHIP-AE (Child Health and Illness Profile, Adolescent Edition) es un instrumento generico para adolescentes de 12 a 19 anos que ha sido adaptado para uso en Espana. El objetivo del estudio fue obtener los valores poblacionales de referencia de la version espanola del CHIP-AE. Metodos: Se administro el CHIP-AE a una muestra representativa de adolescentes escolarizados de Barcelona, mediante muestreo por conglomerados, estratificado segun la titularidad del centro (publico o concertado) y el indice de capacidad economica familiar (bajo, medio y alto). Se estandarizaron las puntuaciones a una media de 20 y desviacion estandar (DE) de 5. Se calcularon las medias y los percentiles. Las medias se compararon por edad, genero y nivel socioeconomico mediante analisis de la varianza. Resultados: Las proporcion de respuesta fue del 81% (n = 902). Las puntuaciones presentaron un amplio rango de distribucion, y aunque en general fueron algo sesgadas hacia las puntuaciones de buena salud, sugieren que la muestra de poblacion general no esta exenta de problemas de salud. El 25% presento puntuaciones por debajo de 17,2 en la dimension de bienestar, lo que indica un tamano de efecto de 0,56 unidades estandarizadas de DE. La distribucion de las puntuaciones de las muestras de referencia de Barcelona fueron muy similares a las originales de Baltimore (Estados Unidos), con diferencias minimas en riesgo individual. Conclusiones: El CHIP-AE recoge de manera sistematica las dimensiones propias de la salud de los adolescentes. Los resultados permitiran establecer comparaciones con adolescentes de otras regiones, y/o con diferentes problemas de salud y analizar las desigualdades en salud durante la adolescencia.