Research Article PANIC DISORDER PHENOMENOLOGY IN URBAN SELF-IDENTIFIED CAUCASIAN–NON-HISPANICS AND CAUCASIAN–HISPANICS Michael Hollifield, M.D., 1n M. Rosina Finley, M.D., 2 and Betty Skipper, Ph.D. 2 The epidemiology of panic disorder is well known, but data about some phenomenological aspects are sparse. The symptom criteria for panic disorder were developed largely from rational expert consensus methods and not from empirical research. This fact calls attention to the construct validity of the panic disorder diagnosis, which may affect accuracy of epidemiological findings. Seventy self-identified Non-Hispanic–Caucasian (Anglo) and Hispanic–Cau- casian (Hispanic) people who were diagnosed with DSM-III-R panic disorder with or without agoraphobia were invited to complete a Panic Phenomenological Questionnaire (PPQ), which was constructed for this study from the Hamilton Anxiety Scale Items and The DSM-III-R panic symptoms. Fifty (71%) subjects agreed to participate, and there was no response bias detected. Seven symptoms on the PPQ that are not in the DSM-IV diagnostic criteria were reported to occur with a high prevalence in this study. Furthermore, many symptoms that occurred with a high frequency and were reported to be experienced as severe are also not included in current nosology. A few of the DSM-IV criterion symptoms occurred with low prevalence, frequency, and severity. Cognitive symptoms were reported to occur with higher frequency and severity during attacks than autonomic or other symptoms. There were modest differences between ethnic groups with regard to panic attack phenomena. Further research using multiple empirical methods aimed at improving the content validity of the panic disorder diagnosis is warranted. This includes utilizing consistent methods to collect data that will allow for rational decisions about how to construct valid panic disorder criteria across cultures. Depression and Anxiety 18:7–17, 2003. & 2003 Wiley-Liss, Inc. Key words: anxiety; panic disorder; phenomenology; ethnicity; explanatory model; nosology; diagnosis; validity; measurement; culture INTRODUCTION The prevalence of panic disorder is 0.4–4.2% in community samples and 5–8% in primary care settings [Dick et al., 1994; Eaton et al., 1994; Faravelli et al., 1989; Hollifield et al., 1990; Joyce et al., 1989; Katon, 1986; Lee et al., 1997; Myers et al., 1984; Von Korff et al., 1985; Weissman et al., 1997]. The ratio of panic disorder prevalence in primary care clinics to commu- nity samples is similar in developing and developed countries [Hollifield et al., 1994]. Some phenomenological aspects of panic are not as well understood as the epidemiology. The DSM-IV panic disorder field trial demonstrated the order of DEPRESSION AND ANXIETY 18:7–17 (2003) 1 Departments of Psychiatry and Family and Community Medicine, University of New Mexico Health Sciences Center, Albuquerque, New Mexico 2 Department of and Family and Community Medicine, Uni- versity of New Mexico Health Sciences Center, Albuquerque, New Mexico n Correspondence to: Dr. Michael A. Hollifield, Departments of Psychiatry and Family and Community Medicine, University of New Mexico Health Sciences Center, 2400 Tucker Avenue, N.E., Albuquerque, NM, 87131. E-mail: mhollifield@salud.unm.edu Received for publication 12 April 2002; Accepted 5 February 2003 DOI 10.1002/da.10100 Published online in Wiley InterScience (www.interscience.wiley. com). & & 2003 WILEY-LISS, INC.