TYPE A BEHAVIOR WITH ERCTA SCALE IN NORMAL SUBJECTS AND CORONARY PATIENST

INTRODUCTION
Type A behaviour pattern (TABP) has its origin in Friedman and Rosenman's (1974) description of behaviours made from observations carried out in the 1950s. These authors propose that the TABP implies trait characteristics in the person which, in interaction with certain environmental events, result in certain behaviours. Among the characteristics included in the TABP are: competitiveness, aggressiveness, irritability, work orientation, worrying about deadlines, urgency, etc. From a physical point of view this manifests itself in general tension, an explosive style of speech, a state of alertness, urgent behavior style and irritability, among other characteristics (cf. Friedman and Rosenman, 1974; Matthews, 1988). In recent years, it has been attempted to integrate the vectors of isolated behaviour patterns in more organised models -or typologies- that include also, for example, behaviours that supposedly increase the probability of cancer (Grossarth-Maticek and Eysenck, 1990, Grossarth et al., 1988: cf. the review by Eysenck, 1991). These models, however, have yet to be tested in wide-ranging and well-controlled studies (Amelang et al., 1996).
Among the TABP scales that have been translated and used extensively in our country, we should mention that of Bortner (Bortner Rating Scale) (Bortner, 1969; cf. Flores and cols., 1985; Del Pino and cols., 1992), the Framingham Type A Scale (Haynes et al., 1978; cf. Del Pino and cols., 1990; García Fernández-Abascal, 1994), and that of Jenkins (Jenkins Activity Survey) (Jenkins, Zyzanski and Rosenman, 1979; cf. García FernándezAbascal, 1994).
Based on the existing measures, we developed a new instrument (see Appendix), with a small number of items, 8, and a short application time, around five minutes, and whose main objective is the detection (screening) of TABP in large groups of subjects (Rodríguez Sutil et al., 1996). A measure of Type A, as Powell (1987) points out, may be considered as valid if it demonstrates its relationship to other validated measures of the TABP, and if it can be shown that it helps to predict the appearance of a coronary disorder. Nevertheless, it should be stressed that the validity of the TABP for predicting coronary illness has been seriously questioned in recent years (cf. Foreyt, 1990; Miller et al., 1991). We attempt to show the validity of our instrument, on the one hand, through its construct and factorial validity, and on the other, by means of the simultaneous application to a group of subjects of our questionnaire and the Framingham scale, mentioned above. Subsequently, we compare the scores of two different groups of subjects: normal, and with various cardiovascular illnesses.

METHOD
Subjects
The total sample comprises 476 subjects in two groups. The first and most numerous (Group 1) consists of 316 incidental subjects, with a mean age of 23.5 years (S.D. = 4.58), due to the presence of a large number of students. The second group (Group 2) is made up of patients that were examined by the Cardiology Unit of the Gregorio Mara–on Hospital in Madrid by the two last-named authors of this article (individual heteroapplication). The subjects in this group are 154 inor out-patients suffering from well-documented coronary disorders or other cardiovascular complaints. Their mean age is 57.9 (S.D. = 13.3). In terms of gender the sample is for the most part feminine (300, as against 176 males), especially in Group 1, made up mostly of students; in Group 2 the ratio was 117 men to 38 women.

Measures
The measurement instrument used was the ERCTA-a, the "screening" instrument of the TABP, designed by the first two authors of the research team. It comprises 8 items with a 5-point response scale (see APPENDIX 1). The difficulty observed on applying the first tests to the clinical sample with regard to the understanding of the language used to formulate the questions led us to design a parallel form, ERCTA-b (see APPENDIX 2), using simpler language. The correlation between the two scales, obtained with 163 subjects from Group 1, is sufficiently high (r = .880), with the two scales showing, for these subjects, similar means and standard deviations, as it can be observed in Table 1. We also applied the Framingham Type A Scale (Haynes et al., 1978) to Group 1 subjects, in conjunction with the scales ERCTA-a (N=243) and ERCTA-b (N=155). Subjects were also asked to respond to a brief questionnaire on general health matters (smoking, weight, etc.), whose relationship to the scores in the scales will be analysed in a later work.

RESULTS
Scores on the total scale
In a previous study (Rodríguez Sutil, et al., 1994), the distribution of scores on the ERCTA scale for the total sample -after eliminating Item 8, for which the theoretical score for each individual may range from 7 to 35- was approximately normal, with a mean of 24.36 that coincided with the median, 24.00, and a standard deviation of 3.81. If we consider, as in other works (cf. Miller et al., 1991) that the proportion of TABP in the population is around 50%, we could take a score of 24 as orientative. In the current sample, the mean is slightly lower (22.94; s.d.= 3.73), perhaps due to the high proportion of young people and women, with an approximately normal distribution. As for the parallel form (ERCTA-b) the mean is 24.60 (S.D. 3.86), somewhat higher due to the abundance of clinical subjects.
Factorial analysis
The construct validity of the scales was established by means of an analysis of principal components with varimax rotation, using the SPSS program FACTOR. With the ERCTA-a -applied to 397 subjects- we obtained, in a first analysis, 2 factors, following Kaiser's criterion of eigenvalues greater than 1. Overall, these explained 49.8% of total variance. Table 2 shows this factorial solution. In Factor 1, items with relevant weight are numbers 2, 4 and 7, which we may consider as making up a positive feature of work orientation (activity, professional goals, attention to work). Factor 2, on the other hand, may represent a negative feature of work tension (stress, perfectionism, competitiveness, hurry). Item 8, which in previous studies appeared to be isolated, saturates positively in Factor 2 and negatively in Factor 1, a result that appears to be coherent with the above descriptions. That is, subjects that experience stress are those that also have greater difficulty in expressing their emotions.
The ERCTA-b scale -applied to 252 subjects- also gave us 2 factors, again following Kaiser's criterion of eigenvalues greater than 1. Overall, these two factors could explain 48.9% of the total variance. Table 3 shows this factorial solution.
As it can be observed, this second table repeats point by point the factorial structure of the first one, thus giving additional support to the previously proposed interpretation. According to these data, moreover, Item 8 appears to be even more clearly associated with Factor 2.
The analyses of elements and the calculation of the alpha coefficient of reliability-internal consistency were carried out using the SPSS program RELIABILITY. The homogeneity scores of the elements attained high values for both scales. The alpha coefficient, with Item 8 eliminated, reached a value of .6834, for ERCTA-a, and .7073, for ERCTA-b, which are quite high values, if we consider that it is a short-length test. When we combine the two scales -with 171 subjects- the alpha coefficient rises to .8389, considering 14 items.
Correlations with the Framingham scale
In Table 4 it can be seen that the ERCTA scales correlate in a moderate and significant way with one of the commonest TABP measurement instruments, Framingham's Type A Behaviour Scale. It is noteworthy, nevertheless, that the highest correlations occur for Factor 2. This seems to suggest that it is this factor, and the items making it up, that best represents the characteristics of the TABP, so that this pattern should be understood, above all, as a negative feature of self-induced work stress.
Intergroup comparisons
Finally, in Table 5, we present the differences of means between the two study groups. These differences, as indicated, are significant. They are, in fact, sufficiently important for us to consider their utility for the clinical field. We should, however, qualify this situation: according to the direct experience of the last two authors with patients, these subjects may have been given ample information by health staff about the possible causes of their illness, lifestyle, and how they should change for the good of their future health. We still lack, therefore, the best form of criteria validation, which would be the prediction of coronary disorders.

DISCUSSION AND CONCLUSIONS
The results of the factorial matrix suggest that there are two main variables relevant to the TABP, which we have called: orientation towards work and work tension or work stress. These two components are similar to those of "competitive drive" and "impatience", which, according to Matthews (1982), are the only two components associated with the subsequent appearance of coronary disorders, on the Framingham scale, of a total of five factors. As it has been seen, the relationship with this scale is especially obvious in Factor 2. This leads us to think that the TABP, as an indicator of coronary risk, is impregnated chiefly by stress and hostile tendencies. Item 8, which in previous studies appeared isolated, saturates positively in Factor 2 and negatively in Factor 1, suggesting that those subjects that experience stress, are those that also have greater difficulty in expressing their emotions. Let us refer to two studies coming from Scandinavian countries. In the most recent, carried out in Sweden by Orth-Gomer in 1994, it was found that the relationship between TABP and coronary risk is only effective in those subjects that lack appropriate social support. Finns Venalainen and Salonen (1992), meanwhile, from a psychodynamic perspective, showed that Type A subjects usually have a more narcissistic, exploitative and distant personality than other people. In view of this data, it does not seem appropriate to eliminate Item 8, in spite of what other authors (cf. Del Pino et al., 1992) suggest, or we ourselves have suggested in earlier works.
One of the general conclusions we can draw is that, as several authors point out (Powell, 1987; Matthews, 1988, Miller et al., 1991, among others), it is necessary to separate the components of the TABP in order to make a more precise prediction of the appearance of cardiovascular disorders in the studied population.
Finally, we should emphasise the ample support the ERCTA scales (a and b) have received, given their factorial structure, their correlation with other TABP measures and their capacity for differentiating diagnostic groups.

REFERENCES
Amelang, M., Schmidt-Rathjens, C. and Matthews, G. (1966). Personality, cancer and coronary heart disease: Further evidence on a controversial issue. British Journal of Health Psychology 1, 191-205.

Bortner, R. W. (1969). A short rating scale as a potential measure of pattern A behaviour. Journal of Chronic Disease 22, 87-91.

Del Pino, A., Borges, A., Díaz, S., Suárez, D. and Rodríguez. I. (1990). Propiedades psicométricas de la escala Tipo A de Framingham (ETAF) (Psychometric properties of the Framingham Type A scale). Psiquis, 11, 19-30.

Del Pino Pérez, A., Gaos Meizoso, M. T., Díaz del Pino, S., Díaz Cruz, F. and Suárez Santana, D. (1992).
Propiedades psicométricas de la escala de Bortner para la medida del patrón de conducta Tipo "A" (Psychometric properties of the Bortner scale for the measurement of the Type A behaviour pattern). Psiquis, 13, 11-26.

Eysenck, H. J. (1991). Personality, stress and disease: An interactionist perspective. Psychological Inquiry, 2, 221-232.

Flores, T., Valdés, M., Treserra, J., García-Esteve, LL. and Núñez, P. (1985). Cuestionario Bortner para la detección del patrón A de conducta: validación española (Bortner questionnaire for the detection of Type A behaviour pattern: Spanish validation). Revista del Departamento de Psiquiatría, 4, 227-238.

Foreyt, J. P. (1990). Behavioral Medicine. Review of Behaviour Therapy, 12, 138-177. Friedman, M. and Rosenman, R. H. (1974). Type A Behaviour and your Heart. New York: Knopf. García Fernández-Abascal, E. (1994). Intervención Comportamental en los Trastornos Cardiovasculares (Behavioural Intervention in Cardiovascular Disorders). Madrid: Fundación Universidad-Empresa.

Grossarth-Maticek, R. and Eysenck, H. J. (1990). Personality, stress and disease: Description and validation of a new inventory. Psychological Reports, 66, 355-373.

Grossarth-Maticek, R., Eysenck, H. J. and Vetter, H. (1988). Personality type, smoking habits and their interaction on predictors of cancer and coronary heart disease. Personality and Individual Differences, 9, 479-495.

Grupo de Estudio EPCUM (1993). Estudio de Prevalencia y Prevención de las Enfermedades Cardiovasculares en Madrid (Study on the Prevalence and Prevention of Cardiovascular Disorders in Madrid). Ayuntamiento de Madrid. Haynes,

S. G., Levine. S., Scotch. N., Feinleib, M. and Kannel, W. B. The relationship of psychosocial factors to coronary heart disease in the Framingham study. I. Methods and risk factors. American Journal of Epidemiology, 107, 362-383.

Jenkins, C. D., Zyzanski, S. J. and Rosenman, R. H. (1979). The Jenkins Activity Survey for Health Prediction. New York: The Psychological Corporation.

Matthews, K. A. (1982). Psychological Perspectives on the Type A Behaviour Pattern. Psychological Bulletin, 91, 293-323

Matthews, K. A. (1988). Coronary heart disease and Type A behaviours: Update on and alternative to the Booth-Kewley and Friedman quantitative review. Psychological Bulletin, 104, 373-380.

Miller, T. Q., Turner, Ch. W., Tindale, R. S., Posavac, E.J. and Dugoni, B.L. (1991). Reasons for the trend toward null findings in research on Type A behaviour. Psychological Bulletin. 110, 469-485.

Orth-Gomer, K. (1994). Individual and environmental resources for health promotion: Interactive effects. Advances, 10, 29-31. Powell, L. H. (1987). Issues in the Measurement of Type A Behaviour Pattern. In S.V. Kasl and C.L. Cooper (comps.) Stress and Health: Issues in Research Methodology. New York: John Wiley & Sons.

Rodríguez Sutil, C., Gil-Corbacho, P. and Martínez Arias, R. (1996). Presentación de la Escala Retiro de Patrón de Conducta Tipo A (ERCTA) (Presentation of the ERCTA scale). Psicothema, 8, 1, 207-213.

Venalainen, E. and Salonen, J. T. (1992). Characteristics of Type-A men in a psychodynamically oriented interview. Nordic Journal of Psychiatry, 46, 329-334.

View this Table:
[in this window]
[in a new window]

Table 1. Comparisom between the scales ERCTA-a and ERCTA-b.

View this Table:
[in this window]
[in a new window]

Table 2. Rotated factorial matrix.Varimax. ERCTA-a

View this Table:
[in this window]
[in a new window]

Table 4. Correlations between the ERCTA scales (a and b) and the Framingham scale*

View this Table:
[in this window]
[in a new window]

Table 5. Comparison of results between patients and non-patients..

Copyright 1998 by the	Psychology in Spain, Vol 2 No 1, 43-47
Colegio Oficial de Psicólogos