Services on Demand
Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.57 no.1 Campinas Jan./Feb. 2007
Hospital anxiety and depression scale: a study on thevalidation of the criteria and reliability on preoperative patients*
Escala hospitalaria de ansiedad y depresión: estudio de la validez de criterio y de la confiabilidad con pacientes en el preoperatorio
José Álvaro Marques MarcolinoI; Ligia Andrade da Silva Telles Mathias, TSAII; Luiz Piccinini FilhoIII; Álvaro Antônio Guaratini, TSAIV; Fernando Mikio SuzukiV; Luís Augusto Cunha AlliV
Adjunto do Departamento de Psiquiatria e Psicologia Médica da FCMISCMSP
IIDiretora do Serviço e Disciplina de Anestesiologia da ISCMSP e Faculdade de Ciências Médicas da Santa Casa de São Paulo; Responsável pelo Centro de Ensino e Treinamento, CET-SBA, ISCMSP; Professora Adjunta do Departamento de Cirurgia da FCMISCMSP
IIIProfessor-Assistente do Departamento de Cirurgia da FCMISCMSP, Diretor do Serviço de Anestesiologia do Hospital Santa Isabel
IVMestre em Medicina, Doutorando do Departamento de Cirurgia da FCMISCMSP
VGraduando do 3º Ano da FCMISCMSP
OBJECTIVES: A few studies have shown that patients are frequently under
strong distress in the preoperative period. The Hospital Anxiety and Depression
Scale (HADS) is an instrument used to evaluate anxiety and depression. The aim
of this study was to evaluate the validity of the criteria and reliability of
the Hospital Anxiety and Depression Scale (HADS) in preoperative patients.
METHODS: Seventy-nine patients admitted to the Surgery Department of Santa Casa de Misericórdia de São Paulo where evaluated, while the control group was composed of 56 companions. The following tools were applied: Demographics data questionnaire, Beck Anxiety and Depression Inventory, and HADS.
RESULTS: HADS internal consistency ranged from 0.79 to 0.84. The items of HADS demonstrated a positive correlation with the total score of the relating subscales. Spearman correlation between HADS-A and Beck Anxiety Inventory (BAI), and between HADS-D and Beck Depression Inventory (BDI) ranged from 0.6 to 0.7. The sensitivity and specificity ranged from 69.6% and 90.9%.
CONCLUSIONS: The subscales of HADS showed internal consistency indices recommended for screening tools. The items in HADS demonstrated a positive correlation with the total score of the anxiety and depression subscales. The moderate to strong correlation demonstrated for HADS-A and HADS-D may be related to the comorbidity between anxiety and depression. For the physician who uses HADS, the thought that anxiety and depression are separate concepts is still useful. The use of a simple tool, such as HADS, could reveal mood changes that can go unnoticed by the supporting team.
Key Words: PREANESTHETIC EVALUATION: psychological status.
Y OBJETIVOS: Algunos estudios han mostrado que los pacientes experimentan
una fuerte angustia en el período Preoperatorio. La Escala Hospitalaria
de Ansiedad y Depresión (HADS) es un instrumento para la evaluación
de la ansiedad y de la depresión. El objetivo de este trabajo fue el
de estudiar la validez de criterio y la confiabilidad de la Escala Hospitalaria
de Ansiedad y Depresión (HADS) en pacientes que están en el Preoperatorio.
MÉTODO: Se evaluaron 79 pacientes, internados en el Departamento de Cirugía de la Santa Casa de Misericordia de São Paulo y 56 acompañantes como grupo control. Se aplicaron los siguientes instrumentos: Cuestionario de datos sociodemográficos, Inventario de Ansiedad y de Depresión de Beck y la HADS.
RESULTADOS: La consistencia interna de la HADS varió de 0,79 a 0,84. Los ítems de la HADS se correlacionan positivamente con la puntuación total de las respectivas subescalas. La correlación de Spearman entre la HADS-A y el Inventario de Ansiedad de Beck (BAI) y HADS-D con el Inventario de Depresión de Beck (BDI) varió de 0,6 a 0,7. La sensibilidad y la especificidad variaron de 69,6% a 90,9%.
CONCLUSIONES: Las subescalas de la HADS presentaron índices de consistencia interna recomendables para instrumentos de screening. Los ítems de la HADS se correlacionaron positivamente con la puntuación total en las subescalas de ansiedad y de depresión. La correlación de moderada para fuerte entre la HADS-A y la HADS-D, puede estar relacionada con la comorbidad entre ansiedad y depresión. Para el clínico que utiliza la HADS, continua siendo útil el razonamiento que hace la ansiedad y la depresión como constructos separados. La utilización de un instrumento sencillo como la HADS podría revelar casos de trastorno del humor que pueden pasar desapercibidos por el equipo asistencial.
The overall frequency of mood disorders in patients admitted to hospitals varies from 20% to 50%, depending on the population (sociodemographic characteristics, and type, severity, and chronicity of the disease), and on methodological definitions (inclusion criteria, research tools, cutting point, "case" definition, and others). Despite the suffering they cause to patients and their clinical implications, mood disorders are not recognized as such by physicians and they affect at least one third of the patients 1-3.
Patients who undergo surgical procedures are frequently under strong preoperative distress1. The extension of this distress can be influenced by the presence of prior psychiatric disorders, such as depression, anxiety, and other minor psychiatric disorders 2. It has been described that the incidence of preoperative anxiety varies from 11% to 80% in adults 3.
Preoperative anxiety and depression can also cause reactions that result in an increase in the intraoperative consumption of anesthetics and in a greater postoperative demand for analgesics 4,5. Besides, preoperative anxiety and depression seem have a profound influence on the immune system and on the development of infections, and facilitates the development of other changes on the long range 6.
There are reports on the literature on several tools to assess anxiety and depression, such as the Hamilton Anxiety Scale 7, the State Trait Anxiety Inventory (STAI) I and II7, Beck Anxiety and Depression Inventories 7, and the Hospital Anxiety and Depression Scale (HADS) 8-20. Most of them were created to be used by patients with psychiatric disorders. Beck Anxiety and Depression Inventories is probably the anxiety and depression self-evaluation tool used more often in research and clinical settings. Their 21 items form a questionnaire developed to evaluate the presence and severity of depression symptoms 7.
The Hospital Anxiety and Depression Scale 8 was developed initially to identify anxiety and depression symptoms in patients admitted to clinical and not to psychiatric hospitals, followed by its use with other types of patients 9-20, in outpatients 10-12, and in healthy individuals 21-23. To prevent the interference of somatic disorders on the scale, anxiety and depression symptoms related with physical diseases were excluded, which is an important difference between HADS and other scales. It does not have items, such as weight loss, anorexia, insomnia, fatigue, a gloomy outlook on the future, headache, dizziness, etc. that could also be caused by physical diseases. If there are comorbidities, the psychological symptoms, more so than somatic symptoms, determine the mood disorders present in other clinical diseases. Since this is a study with patients who had been admitted to the hospital for surgery, the presence of somatic symptoms of anxiety and depression could be misinterpreted as signs and symptoms of the underlying disease or its treatment. Besides, the scale is easy and fast to apply, and can be done by the patient (which was the case in this study) or by the interviewer (patients who cannot read or who present visual or motor deficits).
The notion of measurement is an essential component of scientific research. In social sciences, measurement is more commonly seen as a process that articulates abstract concepts with empirical indicators. In general, there are two basic properties indispensable for empiric measurements: validity and reliability 24. The validation of the Portuguese translation of HADS had already been done in patients admitted to a medical award 25, but it had not been used to validate the presence of preoperative anxiety and depression.
The aim of this study was to study and validate the criteria and reliability of the Portuguese translation of the Hospital Anxiety and Depression Scale in patients admitted to the surgical ward.
This study was approved by the Ethics Commission of the Hospital Central da Irmandade da Santa Casa de Misericórdia de São Paulo (ISCMSP). Eighty patients admitted to the Surgical Ward of ISCMSP, older than 16 years, physical status ASA I and II, who were going to small and medium elective surgeries were included. Patients with malignancies, psychiatric disorders; hearing, visual, and speaking deficits; and patients taking psychoactive drugs were excluded. Eighty companions of those patients were asked to be part of the control group.
Patients were asked, always by the same researchers (two 3rd year Medical students), to sign an informed consent. Those who agreed to take part in the study were asked to answer the following tools the day before the surgery, prior to the pre-anesthetic evaluation:
a) Sociodemographic questionnaire;
b) Hospital Anxiety and Depression Scale (HADS) 8, which has 14 items, 7 of which are aimed at evaluating anxiety (HADS-A) and seven for depression (HADS-D). Each item receives a score that ranges from zero to three, achieving a maximal score of 21 points for each scale (Chart 1);
The answers to HADS were used to assess the frequency of anxiety and depression. Zigmond and Snaith's 8 cutting points were adopted for both subscales:
- HAD-anxiety: without anxiety, from 0 to 8, with anxiety > 9;
- HAD-depression: without depression, from 0 to 8, with depression >9.
c) Beck Depression Inventory (BDI) 26: Composed of 21 items, each one with four possibilities in ascending order of severity of depression;
d) Beck Anxiety Inventory (BAI) 27: Composed of 21 items, each one with four possibilities that reflect, in ascending order, the severity of each symptom.
Beck Depression and Anxiety Inventories were considered the gold standard to determine the sensibility and specificity of HADS 26,27. We adopted a cutting point of 11 for BDI and 10 for BAI.
Before the beginning of the procedures, both medical students were trained on how to use the HADS.
The results were submitted to a descriptive analysis. To determine the reliability, the measure of the internal consistency of patients' answers was assessed by Cronbach's alpha index and Spearman correlation among HADS items. For the validity of the criteria, the indexes obtained by the answers to HADS were compared with those obtained with BAI and BDI, scales that were considered the gold standard, to determined the sensibility and specificity.
A difference was considered statistically significant when p < 0.05. The tests used are part of the SPSS (Statistical Package for the Social Sciences) for Windows 10 28.
The study population was composed of 79 patients, 43 females (54.4%) and 36 males (45.6%). The control group was composed of 56 companions, 42 females (75%) and 14 males (25%).
In the evaluation of the reliability for the patients, the anxiety and depression HADS subscales had a Cronbach's alpha index (internal consistency) of 0.84 and 0.83, respectively. For the control group, Cronbach's alpha index was of 0.84 and 0.79 (anxiety and depression) (Table I).
The Spearman correlation between each item and final scores of the subscales of anxiety and depression of HADS was measured (Table II).
When evaluating the validity of the criteria, Spearman correlation between HADS-A and BAI was 0.68, and between HADS-D and BDI was 0.67. The correlation between HADS total score and BAI was 0.66 and with BDI was 0.69. In the control group, Spearman correlation between HADS-A and BAI was 0.70, and between HADS-D and BDI was 0.68. The correlation of HADS total score with BAI was 0.65 and with BDI was 0.69. The study of HADS performance as compared to the other anxiety and depression scales was based on the sensitivity and specificity of the tools that categorize the individuals as having a disorder or not.
To determine HADS sensitivity and specificity, the diagnosis provided for by Beck scales was considered as the standard. Table III shows the results of the sensitivity and specificity for patients and control group.
The subscales of HADS presented indices of internal consistency, as measured by Cronbach's alpha index, recommended for screening tools. According the Nunnaly 29, this value should be at least 0.6 and should be above 0.8. The data of this study were higher than those of Botega et al. 25. On a review of the literature and using 15 studies, Bjelland et al. 30 demonstrated that HADS internal consistency ranged from 0.67 to 0.93, what reassures and strengthens the cohesion of the items of this tool.
The items of the HADS had a positive and significant correlation with the total score, both for the anxiety and depression subscales. This study also found a significant correlation among the items in each scale and the alternative, but with lower scores. This suggests the possibility that the subscales have convergent validities and do not discriminate anxiety and depression. Botega et al. 25 obtained similar results.
The distinction between anxiety and depression is very useful in clinical practice. It can guide the therapeutic approach with the use of drugs specific for the symptoms. In theory, the distinction between anxiety and depression can be controversial. There is an ongoing discussion on whether anxiety and depression are separate entities, belong to different categories, or different dimensions 31. Population studies demonstrated the correlation between both dimensions 32. It is also known that the correlation between scales can be increased when it is calculated based on a population that has mixed disorders of anxiety and depression 33.
Some authors have proposed using the sum of all 14 items of HADS, producing one measure of morbidity 34. Others have demonstrated the practical value of using two subscales. A study with 568 cancer patients using factorial analysis demonstrated two factors that corresponded to both HADS scales 35. A revision by Bjelland et al.30 also supported the two-factor structure. The moderate to strong correlation between HADS-A and HADS-D found in this study may be related to the comorbidity between anxiety and depression. Burns and Eidelson 36 argued that the correlation between any valid and reliable measure of depression and anxiety should be at least 0.70, not for the presence of shared symptoms but because they share a common cause.
For the physician who uses HADS, the thought that anxiety and depression are separate concepts is still useful. Anxiety involves feelings of fear, worries, and apprehension, while depression is dominated by sadness, sorrow, and hopelessness.
To assess the validity of the criteria, HADS subscales were compared with BAI and BDI, considered the gold standards. In this case, sensibility, which represents the capacity of the scale to identify a proportion of sick individuals compared with the gold standard, ranged from 70.8% and 80.6%. Specificity, or the capacity of the scale to identify healthy individuals when compared with the gold standard, ranged from 69.6% to 90.9%.
When HADS was compared with other tools frequently used to assess anxiety and depression, such as Beck's depression and anxiety scales, HADS correlation ranged from 0.6 to 0.7, what can be considered a medium to strong correlation. In the review undertaken by Bjelland et al 30, the correlation between HADS and BDI ranged from 0.6 to 0.73. According to this data, the conclusion is that in this study the validity of HADS criteria can be considered good to very good.
This study confirmed the assumption that HADS is a good screening tool to distinguish anxiety and depression and to detect cases of anxiety and depression among surgical patients. Using a simple tool such as HADS can uncover cases of mood disorders that would go unnoticed by the supporting team.
This study was part of the Programa Institucional de Bolsas de Iniciação Científica do Conselho Nacional de Pesquisas (PIBIC CNPq) awarded by the Faculdade de Ciências Médicas da Santa Casa de São Paulo for the 2003/2004 biennium.
01. Mathews A, Ridgeway V - Personality and surgical recovery: a review. Br J Clin Psychol, 1981;20:243-260. [ Links ]
02. Clark DA, Cook A, Snow D - Depressive symptom differences in hospitalized, medically ill, depressed psychiatric inpatients and nonmedical controls. J Abnorm Psychol, 1998;107:38-48. [ Links ]
03. Maranets I, Kain ZN - Preoperative anxiety and intraoperative anesthetic requirements. Anesth Analg, 1999;89:1346-1351. [ Links ]
04. Parris WC, Matt D, Jamison RN et al - Anxiety and postoperative recovery in ambulatory surgery patients. Anesth Prog; 1988;35:61-64. [ Links ]
05. Caumo W, Schmidt AP, Schneider CN et al - Risk factors for preoperative anxiety in adults. Acta Anaesthesiol Scand, 2001;45:298-307. [ Links ]
06. Linn BS, Linn MW, Klimas NG - Effects of psychophysical stress on surgical outcome. Psychosom Med, 1988;50:230-244. [ Links ]
07. Andrade L, Gorenstein C - Escalas de Avaliação Clínica em Psiquiatria e Psicofarmacologia. São Paulo, Lemos Editorial, 2000;139. [ Links ]
08. Zigmond AS, Snaith RP - The hospital anxiety and depression scale. Acta Psychiatr Scand, 1983;67:361-370. [ Links ]
09. Kabak S, Halici M, Tuncel M et al Functional outcome of open reduction and internal fixation for completely unstable pelvic ring fractures (type C): a report of 40 cases. J Orthop Trauma, 2003; 17:555-562. [ Links ]
10. Vage V, Solhaug JH, Viste A et al Anxiety, depression and health-related quality of life after jejunoileal bypass: a 25-year follow-up study of 20 female patients. Obes Surg. 2003;13:706-13. [ Links ]
11. Al-Ruzzeh A, Mazrani W, Wray J et al - The clinical outcome and quality of life following minimally invasive direct coronary artery bypass surgery. J Card Surg, 2004;19:12-16. [ Links ]
12. Brandberg Y, Arver B, Lindblom A - Preoperative psychological reactions and quality of life among women with an increased risk of breast cancer who are considering a prophylactic mastectomy. Eur J Cancer, 2004;40:365-374. [ Links ]
13. Brilstra EH, Rinkel GJ, van der Graaf Y et al - Quality of life after treatment of unruptured intracranial aneurysms by neurosurgical clipping or by embolisation with coils. A prospective, observational study. Cerebrovasc Dis, 2004;17:44-52. [ Links ]
14. Christensen FB - Lumbar spinal fusion. Outcome in relation to surgical methods, choice of implant and postoperative rehabilitation. Acta Orthop Scand, 2004;75:(Suppl):2-43. [ Links ]
15. Keller M, Sommerfeldt S, Fischer C et al - Recognition of distress and psychiatric morbidity in cancer patients: a multi-method approach. Ann Oncol, 2004;15:1243-1249. [ Links ]
16. Brady S, Thomas S, Nolan R et al - Pre-coronary artery bypass graft measures and enrollment in cardiac rehabilitation. J Cardiopulm Rehabil, 2005;25:343-349. [ Links ]
17. Hobby JL, Venkatesh R, Motkur P - The effect of psychological disturbance on symptoms, self-reported disability and surgical outcome in carpal tunnel syndrome. J Bone Joint Surg Br, 2005; 87:196-200. [ Links ]
18. Gustafsson M, Ahlstrom G - Emotional distress and coping in the early stage of recovery following acute traumatic hand injury: A questionnaire survey. Int J Nurs Stud, 2006;43:557-565. [ Links ]
19. Johansson M, Thune A, Nelvin L et al - Randomized clinical trial of day-care versus overnight-stay laparoscopic cholecystectomy. Br J Surg, 2006;93:40-45. [ Links ]
20. Muszbek K, Szekely A, Balogh EM et al - Validation of the Hungarian translation of Hospital Anxiety and Depression Scale. Qual Life Res, 2006;15:761-766. [ Links ]
21. Kliszcz J, Nowicka-Sauer K, Trzeciak B et al - The level of anxiety, depression and aggression in nurses and their life and job satisfaction. Med Pr, 2004;55:461-468. [ Links ]
22. Weigl V, Rudolph M, Eysholdt U et al - Anxiety, depression, and quality of life in mothers of children with cleft lip/palate. Folia Phoniatr Logop, 2005;57:20-27. [ Links ]
23. Andrews B, Hejdenberg J, Wilding J - Student anxiety and depression: Comparison of questionnaire and interview assessments. J Affect Disord. 2006 Jun 6; [Epub ahead of print]. [ Links ]
24. Carmines EG, Zelller RA - Reliability and validity assessment. Sage University paper series on quantitative applications in the social science. Beverly Hills and London: Sage publications; 1979;7-17. [ Links ]
25. Botega NJ, Bio MR, Zomignani MA et al - Transtornos de humor em enfermarias de clínica médica e validação de escala de medida (HAD) de ansiedade e depressão. Rev Saúde Pública, 1995;29:355-363. [ Links ]
26. Gorenstein C, Andrade L - Inventário de depressão de Beck: propriedades psicométricas da versão em português. Rev Psiq Clin, 1998;25:245-250. [ Links ]
27. Cunha JA - Manual da versão em português das Escalas Beck, Casa do Psicólogo, Psychological Corporation, 2001. [ Links ]
28. Norusis MJ - SPSS for windows. Professional statistics. Release 6.0. Chicago, 1993. [ Links ]
29. Nunnally JC - Psychometric Theory. New York, McGraw Hill, 1978. [ Links ]
30. Bjelland I, Dahl AA, Haug TT et al - The validity of the hospital anxiety and depression scale: an updated literature review. J Psychosom Res, 2002;52:69-77. [ Links ]
31. Stavrakaki C, Vargo B - The relationship of anxiety and depression. A review of the literature. Br J Psychiatry, 1986; 149: 7-16. [ Links ]
32. Rose G, Barker DJ Epidemiology for the uninitiated. What is a case? Dichotomy or continuum? Br Med J, 1978;2:873-874. [ Links ]
33. Snaith RP, Taylor CM - Rating scales for depression and anxiety: a current perspective. Br J Clin Pharmacol, 1985; 19:(Suppl1)17S-20S. [ Links ]
34. Wilkinson MJ, Barczak P - Psychiatric screening in general practice: comparison of the general health questionnaire and the hospital anxiety depression scale. J R Coll Gen Pract, 1988; 38:311-313. [ Links ]
35. Moorey S, Greer S, Watson M et al - The factor structure and factor stability of the hospital anxiety and depression scale in patients with cancer. Br J Psychiatry, 1991;158:255-259. [ Links ]
36. Burns DD, Eidelson RJ - Why are depression and anxiety correlated? A test of tripartite model. J Consult Clin Psychol, 1998; 66:461-473. [ Links ]
Dr. José Álvaro Marques Marcolino
Rua Monte Alegre, 428 conjunto 53
05040-000 São Paulo, SP
Submitted em 05
de janeiro de 2006
Aceito para publicação em 11 de novembro de 2006
* Received from Hospital Central da Irmandade da Santa Casa de Misericórdia de São Paulo (HCISCMSP), São Paulo, SP