Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Quality-of-Life Measures and Methods Used to Study Antihypertensive Mecl ications So} Levine and Sydney H. Croog Severe objectives and concerns guided our selection of ~ns~uments to measure Me effects of antihypertensive medication on the quality of life of patients (Croog et al. 1986~. Our study was based on a randomized double-blind clinical trial of a relatively large population dispersed in 30 centers throughout me country. Hence, we wished to obtain measures mat were valid and objective and could be am insisted in a rigorous, stan- dardized manner In many different settings. We sought to obtain ins~u- ments that could be administered in a relatively brief time, and, as far as possible, had already demonstrated Weir usefulness In over studies. Other considerations involved our conception of the measurement of quality of life as a construct. Measures of quality of life necessarily must be modified by me seventy and course or trajectory of the illness or condition, as well as the social and demographic character~shcs of the individual and the social context in which he or she lives (Croog and Levine 1989~. Because we were studying hypertensive patients whose modes of life approximate those of o~en~rise hearty persons in most respects, we needed to obtain a comprehensive picture of me profile of me patient's life that would be very similar to Mat of a nonhypertensive, hearty person. If we were studying the quality of life of padents with chronic obstructive pulmonary disease, we would have modified our measures and selected a more constricted band of indicators. Or, if we Editors' Note: Abe editors inherited the authors to describe how they went about choosing quality-of- life measures in their msearch on antihypertensive medications. We also asked the authors to add any advice they cared to give others. 51
52 S. LEVINE ID So. CROOG were measuring Me quality of life of a terminally id padent, we would have focused on how wed the person could interact wad others, recognize others, denve some satisfaction from seeing friends or relatives, and the like. We obviously would not focus in a significant way upon the tenninaDy iU person's ability to carry out activities in me community. For our study of hypertensive patients, we selected measures In line with We conception that five major dimensions of quality of life must be assessed Levine and Croog 1984~. The first area for assessment was We performance of social roles, including those of spouse, parent, worker, friend, and community citizen. A second major dimension was the physiological state of the individual. The Gird was the emotional state of the individual; the fours, We inteUective or cognitive functioning status of He individual; and fifth, a general sense of weB-being and life satisfac- tion. Among possible measures, the RAND General WeD-Being Scale first met our requirements (Brook et al. 1979~. It consists of 22 self-admin- istered questions that compose six subscales assessing anxiety, depres- sion, general heath, positive weD-being, self-control, and vitality. This scale has a long history, has been used extensively In the large RAND Heath Insurance study, and has proven its usefulness. For the purposes of furler measuring emotional status, we used a series of subscales from the Brief Symptom Inventory (BSI), developed by Derogads and Spencer (1982~. The BST is a 53-item, self-report inventory designed to assess the psychological symptom patterns of respondents. In measuring cognitive or inteDective functioning, we used two tests that are among He most established and widely used In He field: the visual Reproduction Test of the Wechsler Memory Scale (Wechsler 1945, Wechsler and Stone 1973), and He Reitan Trail-Making Test (Re- itan 1958~. The Visual Reproduction Test assesses neuropsychologic~ function on the basis of diagram images, and He TrmI-Making Test measures visu~-motor speed and integration. Selecting instruments for use in He study of hypertensive padents was complicated because existing scales and measures were not directly perti- nent to this population for some dimensions of quality of life. Hence, it was necessary to adapt existing instruments for the special needs of this study and, in some instances, to construct new measures. To assess physical symptoms associated with andhypertensive medica- tions, we adapted questions used commonly in clinical practice, frmning these as a Physical Symptoms Distress Index (Hypertension Detection and Follow-up Program Cooperation Group 1982, Derogads and Spencer
STUDY OF A1~IlIYPERTENSIVE MEDICA77ONS 53 19821. Because sexual dysfi,nchon may be an ~mpormnt side effect In pharmacologic trea~anent of hypertensives, we developed a four-item index suited for the survey research approach In this study' the Sexual Symptoms Distress Index. It was adapted from previous work (Hogan et al. 1980, Derogatis and Spencer 19821. Measures of life satisfaction were based, in part, on scales from items employed by Campbell et al. (1976) In a study of quality of life, and in part on research by Haynes et al. (1978a,b) on stress and heart disease within me Fram~ngham Heart Study. To assess changes In work performance tart might be associated win an- tihypertensive medications, we employed a number of items conceding work performance, adapted In part from previous scale items by House (19811. If we were carrying out tills study again, we would probably follow a similar program in selecting instruments for assessing quality-of-life di- mensions. The measures would be adapted, of course, for the particular illness condition being studied because it is necessary to select a range or band of indicators specifically appropriate for the health condition under consideration. We would certainly select shorter, generic, or fewer scales when this could be done without sacnfic~ng validity and reliability. hnso- far as possible, we would employ widely used, standardized measures. We would again use He RAND General WeB-Being Scale. In smd~g cognitive function, we would employ a broader range of measures than we did in our previous study of hypertensive men' select- ~ng tests mat might be somewhat more sensitive to He effects of andhy- pertensive medications on cognitive function. We would select tests Hat would be less subject to ~e teaming eject unposed by repeats expen- ence, such as those employing digits or nonsense syllables. In short, we would employ a brief version of our tote ~ns~nent, although we have some reservations about how far we should go in shortening some of He scales. Finally, we would explore the possibilities of using computer- assisted melons In ca'Tyng out at least part of He dam collection, although Here are many advantages to having interviewers control the administration of the questionnaire as a whole. REFERENCES Brook, R.H., Rogers, W.H., Williams, K.N., Ware, J.E., Ir., Stewart, Am., Johnston, S.A., and Donald, C.A. Concepmalization and Meas- urement of Heath for Adults in He Heath Insurance Study. Vol. ITI. Mental Health R-1987/3-HEW. Santa Monica, California, The RAND Corporation, 1979.
54 5. LEVINE AND So. CROON Campbell, A., Converse, P.E., and Rodgers, W.L. The Quality of Amen- can Life: Perceptions, Evaluations, and Satisfactions. New Yolk, RusseB Sage Foundation, 1976. Croog, S.H., and Levine, S. Quality of life and health care ~nte~vendons. Freedman, H.E., and Levine, S., eds. The Handbook of Medical Sociology, Fourth Edition. Englewood Cliffs, New Jersey, Prentice Hall, 1989. Croog, S.H., Levine, S., Terra, M.A., Brown. B., BuIpia, C.J., Jenkins, C.D., Klem~an, G.~., and Williams, G.H. The effects of andhy- pertensive therapy on He quality of life. New England Journal of Medicine 314(26):1657-1664, 1986. Derogacis, L.R., and Spencer, P.M. The Brief Symptom Inventory (BSI), Administration, Scoring and Procedures Manual 1. Baltimore, Johns Hopkins University School of Medicine (privately punted), 1982. Haynes, S.G., Levine, S., Scotch, N., Feinleib, M., and Kannel, W.B. The relationship of psychosocial factors to coronary heart disease in He Frarningham study. I. Methods and risk factors. AmericarlJournal of Epidemiology 107~5):362-383, 1978a. Haynes, S.G., Feinleib, M., Levine, S., Scotch, N., and Kannel, W.B. The relationship of psychosocial factors to coronary heart disease in the Framingham study. II. Prevalence of coronary heart disease. Ameri- can Joumal of Epidemiology 107(5):3844029 1978b. Hogan, M.J., Wallin, J.D., and Baer, R.M. Antihypertensive therapy and male sexual dysfunction. Psychosomatics 21~3~:23~237, 1980. House, J.S. Work Stress and Social Support. Reading, Massachusetts, Addison-Wesley, 1981. Hypertension Detection and FoBow-up Program Cooperation Group. The effect of treatment on mortality in "mild" hypertension: Results of the Hypertension Detection and FoBow-up Program. New England Journal of Medicine 307~16~:976-980, 1982. Levine, S., and Croog, S.H. What constitutes quality of life? A conceptu- alization of He dimensions of life quality in healdly populations and patients with cardiovascular disease. In Wenger, N.K., Mattson, M.E., Furberg, C.D., and Elirlson, ]., eds. Assessment of Quality of Life in Clinical Tnals of Cardiovascular Therapies. New York, L`e Jacq Publishing, Inc., 46-58, 1984. Reitan, R.M. Tra~l-Malcing Manual for Administration, Sconng, and Interpretation Department of Neurology, Section of Neuropsychol- ogy, Indiana University Medical Center, Indianapolis, 1958. Wechsler, D. A standardized memory scale for clinical use. Journal of Psychology 19:87-95, 1945. Wechsler, D., and Stone, C.P. Instruction Manual for the Wechsler Memoir Scale. New York, The Psychological Corporation, 1973.
