Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.56 no.5 Campinas Sept./Oct. 2006
Validity of the hospital anxiety and depression scale in patients with chronic pain*
Validez de la escala hospitalaria de ansiedad y depresión en pacientes con dolor crónico
Martha Moreira Cavalcante Castro, M.D.I; Lucas Quarantini M.D.II; Susana Batista-Neves M.D.II; Durval Campos Kraychete, TSA, M.D.III; Carla Daltro, M.D.IV; Ângela Miranda-Scippa, M.D.V
em Neurociências; Responsável pelo Setor de Psicologia do Centro
de Dor do HUPES da UFBA
IIPós-Graduando em Medicina e Saúde - CPgMS da UFBA
IIIAnestesiologista e Doutor em Medicina e Saúde da UFBA
IVProfessora de Medicina Interna da Escola Bahiana de Medicina e Saúde Pública
VProfessora Doutora do Departamento de Neuropsiquiatria da UFBA
OBJECTIVES: Several studies suggest a strong association between anxiety
and depression with chronic pain. That can be demonstrated using standard scales
to detect these symptoms. The objective of this study was to determine sensibility
and specificity of the Hospital Anxiety and Depression Scale (HAD) in patients
with chronic pain syndromes followed at the Pain Center of the Hospital Universitário
Professor Edgard Santos.
METHODS: A transversal, descriptive study was conducted with patients who sought to the Pain Center between March 2002 and July 2003. It was composed of interviews using the HAD Scale and the M.I.N.I. International Neuropsychiatric Interview Brazilian Version 5.0.0 (M.I.N.I. PLUS).
RESULTS: Ninety-one patients were evaluated. The HAD demonstrated that 61 patients (67%) presented anxiety, while 42 patients (46.2%) presented depression. HAD results showed that among patients with depression, 38 (90.5%) also had anxiety; while among those with anxiety, 38 (62.3%) also had depression. Statistical analysis showed that this association was statistically significant (p < 0.001). M.I.N.I. PLUS revealed an incidence of 40.7% in current mood changes and 47.3% of anxiety. As for HAD's sensibility and specificity, we found the following results: sensibility of 73.3% for depression and 91.7% for anxiety, and a specificity of 67.2% for depression and 41.8% for anxiety.
CONCLUSIONS: The HAD scale showed good sensibility to evaluate anxiety and depression symptoms, but did not demonstrate good specificity for the diagnosis of depression and anxiety.
Key Words: AVALIATION: psychological status; PAIN, Chronic.
Y OBJETIVOS: Diversos estudios sugieren una fuerte asociación entre
la ansiedad y la depresión con dolor crónico, lo que puede ser
evidenciado por la utilización de escalas estandarizadas para la detección
de esos síntomas. El objetivo de este estudio fue el de estimar la sensibilidad
y la especificidad de la Escala Hospitalaria de Ansiedad y Depresión
(HAD), en pacientes portadores de síndromes dolorosos crónicos
con seguimiento por parte del Centro de Dolor del Hospital Universitario Profesor
MÉTODO: Fue realizado un estudio descriptivo transversal en pacientes que se personaron en el Centro de dolor entre marzo de 2002 y julio de 2003, que constó de entrevistas utilizando la Escala HAD y el M.I.N.I International Neuropsychiatric Interview Brazilian Version 5.0.0 (M.I.N.I. PLUS).
RESULTADOS: Fueron evaluados 91 pacientes. La utilización de la HAD evidenció que 61 pacientes (67%) presentaron ansiedad y 42 pacientes (46,2%) presentaron depresión. Los resultados de la HAD mostraron que de los pacientes deprimidos, 38 (90,5%) eran también ansiosos; mientras que de los ansiosos, 38 (62,3%) también estaban deprimidos, siendo esa una asociación considerada como significativa por el análisis estadístico (p < 0,001). El M.I.N.I. PLUS reveló un 40,7% de trastorno del humor actual y un 47,3% de trastorno de ansiedad. En cuanto a la sensibilidad y a la especificidad de la HAD, se encontraron los siguientes resultados: sensibilidad 73,3% para la depresión y 91,7% para la ansiedad. Especificidad 67,2% para la depresión y 41,8% para la ansiedad.
CONCLUSIONES: La escala HAD mostró un buen nivel de sensibilidad para evaluar síntomas de ansiedad y depresión, sin embargo, no mostró una buena especificidad para diagnósticos de depresión y ansiedad.
Pain is a subjective and unpleasant phenomenon. Its concept has changed considerably over the years 1. The idea that pain was a punishment caused by the wrath of gods, or by dark spirits who took over one's body, was substituted by the hypothesis of changes in the neurophysiological mechanisms of perception involving cognitive and behavioral activities 2.
Several studies stress that, even in fundamentally organic conditions, the influence of psychological aspects are relevant for the complaints of pain. Thus, chronic pain syndromes favor depression, or depression may precede or predispose people to complaints of chronic pain 3. Anxiety is another psychiatric disorder associated with chronic pain. Anxiety can be situational (when associated with specific circumstances), or constant. Therefore, it is necessary to consider the context in which the emotion occurred, besides its characteristics, to determine which the clinical manifestations are proportional to the intensity and duration of the triggering factors or not. Some authors demonstrated that patients with chronic pain present disproportional preoccupation, tension, nervousness, and apprehension regarding their disease, leading to anxiety levels that are higher than in the general population 4.
The objective of this study was to determine the sensibility and specificity of the Hospital Anxiety and Depression (HAD) Scale in chronic pain patients.
After approval by the Ethics Committee on Research of the Maternidade Climério de Oliveira da Universidade Federal da Bahia, a transversal, descriptive study was conducted. Patients who sought voluntarily the Pain Center of the Hospital Universitário Professor Edgard Santos from March 2002 to July 2003 were evaluated. After signing the Informed Consent, patients were measured against the Hospital Anxiety and Depression (HAD) Scale this scale has 14 items, seven for anxiety and seven for depression and, as a cutting point, 8 for anxiety and 9 for depression 5 and the M.I.N.I. International Neuropsychiatric Interview M.I.N.I. PLUS, a short standard diagnostic interview, compatible with the criteria of the DSM-IV and CID-10, to determined current psychiatric disorders 6. A team composed of three interviewers was trained to use this tool. The Kappa index (proportion of agreement beyond probability divided by the agreement potential) was used to obtain reliability. Every disorder patients fulfilled the criteria for, according to this tool, was recorded; antisocial and behavioral disorders were excluded.
The Statistical Package for the Social Sciences Program (SPSS) was used to build a data bank and for statistical calculations 7. To evaluate the validity indicators and predictive value, the following formula was used: Sensibility = a/a+c; Specificity = d/b+d; Positive Predictive Value = a/a+b; Negative Predictive Value = d/c+d 8. We also used the 2 x 2 table where the number of patients with and without disease diagnosed by the M.I.N.I. was placed on the vertical column and the positive or negative values diagnosed by the HAD on the horizontal. Values were expressed in percentages and based on a 95% reliable interval. The results of the categorical variables were expressed as proportions and the Pearson's Chi-square test was used to study them. A p < 0.05 was considered significant.
Ninety-one patients with ages varying from 18 to 50 years (42.8 ± 7.7 years), 87.9% females and 12.1% males, were studied.
After the HAD was applied, anxiety was diagnosed in 61 (67%) patients, 55 (90.2%) females and 6 (9.8%) males; depression was diagnosed in 42 (46.2%) patients, 39 (92.9%) females and 3 (7.1%) males. When HAD results were analyzed, we observed that among patients with depression, 38 (90.5%) also presented anxiety; and among patients with anxiety, 38 (62.3%) also presented depression. This association was statistically significant (p < 0.001) (Table I).
The Kappa index analysis among observers in the M.I.N.I. was 1.0. After evaluation by the M.I.N.I., we did not find any mental disorder in 22 patients (24.1%), just one type of disorder in 15 (16.4%), and two or more mental disorders in the majority of them, i.e., 54 (59.3%).
M.I.N.I. results showed that 37 patients (40.7%) had current and 25 (27.5%) had past mood disorders. Anxiety was diagnosed in 43 patients (47.3%) and, among those, social phobia was diagnosed in 12 patients (13.2%), while only three patients (3.3%) had anxiety and depression.
When the results of mood disorders obtained using the M.I.N.I. were compared with the diagnosis of depression obtained by the HAD, we observed that among the 25 patients with depression by the M.I.N.I. only 19 had depression according to the HAD; of the 10 patients with dysthymia by the M.I.N.I., only nine had depression according to the HAD; one patient with a manic episode by the M.I.N.I. was diagnosed with depression by the HAD; and one patient with a hypomanic episode did not present anxiety or depression by the HAD (Table II).
When the number of patients with anxiety according to the M.I.N.I. was compared to the number of patients with anxiety by the HAD, we observed that, among the diagnosis of social phobia, agoraphobia, and specific phobias, only one of them was not detected as anxiety by the HAD; while panic, generalized anxiety, obsessive-compulsive, and post-traumatic stress disorders were similar in both tools (Table III).
Table IV shows the validity and predictive value indicators obtained by the analysis of the HAD scale.
This study demonstrated that the profile of patients with chronic pain is similar to what is described in the literature, being more frequent in female patients and in this age group 9.
Several authors emphasized the predominance of psychological aspects, such as mood and anxiety disorders, in patients with chronic pain 3. The evaluation of these disorders is frequently done using scales that only measure the clinical symptoms but do not diagnose the conditions 10,11. However, this study evaluated the specificity and sensibility of the HAD scale for anxiety and depression using the M.I.N.I. as the gold standard.
The greatest advantage of HAD is the absence of confounding symptoms common to clinical disorders and depression, such as fatigue, loss of appetite, and sleep disorders. This tool can also evaluate subjective symptoms that are more specific for depression showing, therefore, in its final result, the need to use or not other more suitable diagnostic methods to identify and treat the condition 12.
In comparing anxiety and depression diagnosed by the HAD scale, the majority of the patients with depression (n = 38, 90.5%) also presented anxiety, and the majority of the patients with anxiety (n = 38, 62.3%) also presented symptoms of depression. This demonstrated a high correlation among the symptoms of anxiety and depression in the same patient 13.
On the other hand, the investigation of mental disorders in this population with the M.I.N.I. showed that most patients (n = 54, 59.3%) had one or more diagnosis. This showed that this is a population that has psychological disorders, demonstrating the need for further diagnostic evaluation studies, since psychiatric diseases can interfere with adequate pain control 14,15.
Some authors suggested that chronic pain is directly related with depression, and anxiety is part of the fear and ignorance regarding the diagnosis 16. However, this study contradicts the reports found in the literature. Our sample had 43 patients (47.3%) with anxiety and 37 patients (40.7%) with depression. This small difference might be secondary to the fact that it was a randomized sample, and that patients were referred for evaluation in different moments of their follow-up. The number of patients being treated for up to three months was significant, suggesting that they presented an acute reaction profile, with great expectations of a new treatment option associated with a prior history of several therapeutic attempts that had failed 2.
Likewise, the high prevalence of social phobias, agoraphobia, and specific phobia (for a total of 29.7%) may be due to the constant presence of pain, which favors antisocial behavior, either for fear of increased pain intensity or lack of sympathy from the patients' social environment 4.
Comparing the frequency of depression in the M.I.N.I. with that of the HAD showed that the former had good diagnostic sensibility, except for the manic episodes. This suggests that scales in clinical practice may be useful to demonstrate symptoms of depression, since there was a good correlation in 29 patients (out of a total of 37). As for anxiety, of 43 patients diagnosed by the M.I.N.I., only three of them were not detected by the HAD, showing that this scale is more sensitive regarding anxiety than symptoms of depression.
Searching PubMed's data base, we found only three reports on the specificity and sensibility of scales for anxiety and depression 5,17,18. These authors demonstrated that these scales have good sensibility (around 90%) but less specificity (around 50%) to detect anxiety and depression 17; however, the population in these studies did not include individuals with chronic pain. Thus, by comparing the results of the works mentioned with ours it is clear that this is an original study comparing HAD and M.I.N.I. in patients with chronic pain. Our study demonstrated that HAD is a high sensibility tool to detect anxiety and depression; its use, associated with a follow-up with pain specialists, can be extremely useful in detecting patients who need more specific psychiatric evaluation.
Besides, since this population presents a higher incidence of psychiatric comorbidities, a high pain index, and several changes in their daily life and the difficulties to face them, identifying adequately the mental disorder of each one would lead to an improvement of the therapeutic plan, their quality of life, and pain management 19-24.
01. Oliveira AS, Bermudez CC, Souza RA et al Pain impact in life of patients with temporomandibular disorder. J Appl Oral Sci, 2003;11:138-143. [ Links ]
02. Pimenta CAM Dor crônica, terapia cognitiva comportamental e o enfermeiro. Rev Psiquiatria Clin, 2001;28:288-294. [ Links ]
03. Oliveira JT Aspectos comportamentais das síndromes de dor crônica. Arq Neuro-Psiquiatria, 2000;58:(Suppl2A):360-365. [ Links ]
04. Pagano T, Matsutani LA, Ferreira EA et al Assessment of anxiety and quality of life in fibromyalgia patients. São Paulo Med J, 2004;122:252-258. [ Links ]
05. Botega NJ, Ponde MP, Medeiros P et al Validação da escala hospitalar de ansiedade e depressão (HAD) em pacientes epilépticos ambulatoriais. J Bras Psiq, 1998;47:285-289. [ Links ]
06. Amorim P M.I.N.I. International Neuropsychiatric Interview (M.I.N.I.): validação de entrevista breve para diagnóstico de transtornos mentais. Revista Brasileira de Psiquiatria, 2000; 22:106-115. [ Links ]
07. Norusis MJ Statistical Package for the Social Sciences (SPSS), Chicago IL, versão 9.0, 1998. [ Links ]
08. Pereira MG Epidemiologia: Teoria e Prática, 5ª Ed, Rio de Janeiro, Guanabara-Koogan, 1995;358-376. [ Links ]
09. Kurita GP, Pimenta CAM Adesão ao tratamento da dor crônica: estudo de variáveis demográficas, terapêuticas e psicossociais, Arq Neuro-Psiquiatria, 2003;61:416-425. [ Links ]
10. Matta APC, Moreira Filho PF Sintomas depressivos e ansiedade em pacientes com cefaléia do tipo tensional crônica e episódica. Arq Neuro-Psiquiatria, 2003;61:991-994. [ Links ]
11. Pinto D Uma análise lingüística da adaptação brasileira do Inventário Beck de Depressão, J Bras Psiq, 1997;46:249-253. [ Links ]
12. Botega NJ, Bio MR, Zomignani A et al Transtornos do humor em enfermaria de clínica médica e validação de escala de medida (HAD) de ansiedade e depressão. Rev Saúde Pública, 1995; 29:355-363. [ Links ]
13. Keogh E, McCracken LM, Eccleston C Do men and women differ in their response to interdisciplinary chronic pain management? Pain, 2005;114:37-46. [ Links ]
14. Haythornthwaite JA, Sieber WJ, Kerns RD Depression and the chronic pain experience. Pain, 1991;46:177-184. [ Links ]
15. Hiller W, Heuser J, Fichter MM The DSM-IV nosology of chronic pain: a comparison of pain disorder and multiple somatization syndrome. Eur J Pain, 2000;4:45-55. [ Links ]
16. Averill PM, Novy DM, Nelson DV et al Correlates of depression in chronic pain patients: a comprehensive examination, Pain, 1996;65:93-100. [ Links ]
17. Teissedre F, Chabrol H A study of the Edinburg Postnatal Depression Scale (EPDS) on 859 mothers: detection of mothers at risk for postpartum depression. Encephale, 2004;30:376-381. [ Links ]
18. Lopez-Alvarenga JC, Vazquez-Velazquez V, Arala-Martinez D et al Accuracy and diagnostic utility of the Hospital Anxiety and Depression Scale (HAD) in a sample of obese Mexican patients. Rev Invest Clínica, 2002;54:403-409. [ Links ]
19. Fisher BJ, Haythornthwaite JA, Heinberg LJ et al Suicidal intent in patients with chronic pain. Pain, 1999;55:259-266. [ Links ]
20. Botega NJ Suicídio e Tentativa de Suicídio, em: Almeida OP, Dratcu L, Laranjeira R Manual de Psiquiatria. Rio de Janeiro, Guanabara Koogan, 1996;221-227. [ Links ]
21. Botega NJ Suicídio e Tentativa de Suicídio, em: Lafer B, Almeida OP, Fráguas Jr R et al Depressão no Ciclo da Vida. Porto Alegre, Artmed, 2000;157-165. [ Links ]
22. Botega NJ Depressão no Paciente Clínico, em: Botega NJ Prática Psiquiátrica no Hospital Geral: Interconsulta e Emergência. Porto Alegre, Artes Médicas, 2002;339-351. [ Links ]
23. Botega NJ, Rapeli CB Tentativa de Suicídio, em: Botega NJ Prática Psiquiátrica no Hospital Geral: Interconsulta e Emergência. Porto Alegre, Artes Médicas, 2002;365-377. [ Links ]
24. Portnoi AG O Enfrentamento da Dor, em: Teixeira MJ, Braum Filho JL, Márquez JO et al - Dor: Contexto Multidisciplinar. Curitiba, Maio, 2003;205-212. [ Links ]
Dra. Martha Moreira Cavalcante Castro
Centro de Dor do Hospital Universitário Professor Edgard Santos
Rua Augusto Viana S/N - Canela
40110-160 Salvador, BA
Submitted for publication
06 de outubro de 2005
Accepted for publication 02 de junho de 2006
* Received from Hospital Professor Edgard Santos da Universidade Federal da Bahia (HUPES-UFBA), Salvador, BA