VOL 4 No. 4 December 1989
children must address ontogenetic factors, the chronologic unfolding of biologic maturation, and the contributions of socialization, the process whereby the social and cultural milieu shape patterns of pain and illness behavior. Fortunately, there has been a relatively prompt and substantial response to the urgent needs of children. These two books provide a remarkable state of the art integration of the substantial achievements in our understanding of pediatric pain. Given the complexity of pain during childhood, it is not surprising that two sets of investigators working independently in two different locations at roughly the same time should have written books that are essentially nonredundant and strikingly complementary. Each book has its unique features. While the most widely accepted definitions of pain have focused upon experiential qualities, and there is a heavy emphasis on self-report in assessing pain, the phenomenology of pain throughout childhood received little systematic exploration until Ross and Ross devised structured methods for interviewing children that yielded systematic information from adolescents down through the preschool years. This careful attention to self-report in children provides a remarkably clear picture of the constraints that develomental factors have on the experience and expression of pain. The anecdotal illustrations that generously pepper the Ross and Ross volume very effectively illustrate the changes in affective, cognitive, and behavioral response patterns that characterize different age levels. The McGrath and Unruh book has a fine balance between chapters, addressing such general issues as the history of relevant concepts, ethics, and service delivery, and material devoted to a careful exploration of particular sources of pain and syndromes. One exceptional and original feature of the McGrath and Unruh book is a remarkable and extensive chapter on the history of our medical understanding and childhood disciplinary practices. The book also is an excellent resource for information on specific forms of pain and clinical problems; it provides descriptive characteristics, probable origins, response to different treatments, including critical analyses of the existing literature and findings, and direction as to where the deficiencies exist. While providing an excellent general background for pharmacologic interventions, the Ross and Ross book is not as comprehensive in
applying these approaches to particular pain syndromes as the McGrath and Unruh book. The balance between biomedical and psychologic perspectives on pain achieved in the latter book is one of its striking features. One could quibble about various features of both books. The structural organization and use of headings in the McGrath and Unruh book is exemplary and will make it easy for readers to locate specific types of material, but the absence of an integrated set of references and author index is sometimes frustrating. The Ross and Ross book provides a fine summary of our understanding of the broad range of determinants of pain, but at times the accounts are agonizingly brief. The two books combined provide the most comprehensive understanding of the field available today. Both books will be of great interest to a broad range of research and clinician readers. The research summaries are state-of-the-art for investigators pursuing the field, and the critical methodological analyses of existing studies are fascinating to read. It is obvious that both sets of investigators have extensive research experience in clinical settings. Both books provide clinical guidelines for the practitioner, with the McGrath and Unruh book providing concrete summaries of this material at the end of each chapter. The books provide fine examples of scientist/professionals in action. We clearly have two sets of professionals who recognize that responsible caring requires careful collection of data in the course of service delivery. Children stand to benefit enormously from these books.
An Antidote to Self-Deception Dennis
Assessment of Chronic Pain
By P. Karoly and M.P. Jensen Published by Pergamon Press, New York 198’7, 163 pages, paper: $13.95; hard cover: $23.50 Dennis C. Turk, PhD, is Professor of Psychiatry and Anesthesiology and Director of the Pain Evaluation and Treatment Institute at the University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.
It appears to be a truism that the assessment of patients with chronic pain syndromes is essential both for research and in guiding clinical decision making and treatment planning. Although this statement appears on the surface to be self-evident, I am often surprised when I read clinical studies to see how little attention seems to be given to the selection of assessment procedures that are used as dependent variables and the basis for conclusions regarding treatment efficacy. I am equally surprised when I discuss assessment with clinicians how little thought goes into the development of a comprehensive assessment battery. Furthermore, I am continually amazed at how little relationship there seems to be between the data acquired during assessment and decisions regarding treatment as well as decisions during treatment. At times I get the feeling that assessments are mentioned, performed, and filed, but are rarely used in treatment planning or even looked at, once placed in the bowels of patients’ ever-expanding files. Does the investigator or clinician adopt instruments or procedures because they meet specific conceptual and essential psychometric requirements or because they are: (a) available, (b) familiar to the clinician or investigator, (c) easy to administer, and (d) reimbursable? Karoly and Jensen raise these issues in their brief book (130 pages of text). The authors draw our attention to clinical thinking about assessment in the hope that raising these concerns will provide a source to serve as a “deterrence to selfdeception.” In the introductory chapter, Karoly and Jensen raise the fundamental question concerning the purposes of assessment. In short, what questions do I want answered and why? In the next set of chapters, they consider what assessment strategies (instruments, procedures) are available to achieve these purposes and describe a range of instruments and procedures. When they begin to evaluate these various instruments and procedures, their coverage becomes somewhat spotty. In this brief volume, they could not be expected to describe and critically evaluate the entire range of available instruments and procedures as well as the relevant research on the available psychometrics (ie, demonstrated to have satisfactory reliability, validity, and utility for intended purposes) and the normative base for each of these. Psychometric considerations and appropriateness of extrapola-
Journal of Pain and Symptom Managknent
tion to the population of interest are extremely important, and greater attention should have been devoted to these points. Consider how infrequently the questions above are given adequate thought and attention. To answer the first question, “what do I want to know?” requires that a model of chronic pain and chronic pain patients must be articulated and that there are treatment options that will vary with the assessment outcomes. This point is discussed in some depth by Karoly and Jensen and is the strength of the volume. Once the relevant variables and construct are identified, they must be operationalized. Accomplishment of satisfactory operationalization requires a series of steps, including surveying the assessment literature and identifying instruments that, at least on the surface, match the variables and constructs of interest. Next, the psychometric properties of these instruments or procedures must be studied. That is, has the developer of the instrument (or procedure) included the appropriate content in the instrument so that it covers the domain (content validity)? Has the developer demonstrated that the instrument is related to analogous constructs (concurrent validity) and unrelated to inappropriate constructs (eg, intelligence, reading ability-convergent and discriminant validity)? Has the instrument been shown to predict outcomes of interest (ie, predictive validity)? Has the instrument been shown to be sensitive to change subsequent to intervention? And is the instrument more economical (eg, time, cost, effort required) than already existing instruments or procedures (ie, ecological validity)? Is the instrument reliable, ie, has it been shown to be internally consistent, stable over time, and do individual judges agree on the coding and scoring? Once the psychometric data have been deemed acceptable, the question to ask is whether the original sample upon which the norms were developed is appropriate for extrapolation to the population of interest. For example, are norms of instruments originally developed on psychiatric populations (eg, MMPI) appropriate for chronic pain patients? Unless the questions listed above are considered and answered satisfactorily by the assessment procedures, we are more likely to be deceived into drawing inappropriate conclusions about research and the successfulness of our clinical interventions. Karoly and Jensen’s book
Vo,l: 4 No. 4 December 1989
draws our attention to these issues. They have provided “a conceptual rationale and a comprehensive review of empirically tested procedures for the clinical assessment of chronic pain in adults” (p. 124). The authors have done a superb job of accomplishing their first aim, “conceptual rationale for assessment” (in Chapters 1 and 9). Because of brevity, the authors’ second aim, “comprehensive review of empirically tested is somewhat less satisfactory. procedures,” Many topics and discussion of procedures are covered in a superficial fashion. Some of the conclusions regarding specific procedures are posed as if there was more supportive data than there really is (eg, the generally positive recommendation about thermography, the utility of surface electromyography for musculoskeletal disorders). I was also somewhat concerned about the lack of attention given to problems of patient motivation and interjudge reliability of many medical and physical therapy examination procedures (eg, performing examination of spinal mobility, straight leg raising, reading plain radiographs). These procedures need the same psychometric consideration as psychological assessment instruments.
I would also have liked the authors to have presented more information about psychometrics and specific guidelines for developing and evaluating instruments and procedures. A useful addition would have been a discussion of the use of assessment throughout the treatment process. Assessment should be viewed as an ongoing process that is not completed once treatment is initiated. In my opinion, the review of specific instruments and procedures is significantly less important than the discussion of conceptual models and issues in assessment, as new instruments will appear regularly, whereas how we think about assessment will remain a constant. In sum, Karoly and Jensen have provided a valuable service to the field by suggesting appropriate strategies for thinking about assessment. Despite some limitations in depth of coverage of specific assessment instruments and Multimethod Assessment of Chronic procedures, Pain should be required reading for all clinicians and clinical investigators. I hope that this volume will find a place on the shelf and, more important, in the hands of “painiacs,” as careful consideration of the issues raised will indeed serve as an antidote to our self-deceptions.