Patient Complaints and Problems in Primary Care: Classification and Concordance, Schemes and Mind Maps of Medicine

Download Patient Complaints and Problems in Primary Care: Classification and Concordance and more Schemes and Mind Maps Medicine in PDF only on Docsity! The Predictive Value of the Presenting Complaint Robert C. Burack, MD, and Robert R. Carpenter, MD Detroit, Michigan, and Freeport, Texas The relationship between the presenting complaint and the principal problem identified during 103 new patient visits was assessed in an academic primary care setting. Complaints and problems were classified by content as somatic, psychosocial, or health maintenance and compared by category. The pre­ senting complaint correctly identified the category in 76 percent of somatic but only 6 percent of psychosocial principal problems (sensitivity of 76 percent and 6 percent, respective­ ly). The likelihood of a same-category principal problem (posi­ tive predictive value) ranged from 53 percent for somatic to 100 percent for psychosocial presenting complaints. A specific underlying motivation for the visit other than the presenting complaint was noted by the primary provider in 42 percent of the encounters and was most frequent in those encounters characterized by a lack of concordance between complaint and problem. The presenting complaint introduces the clinical en­ counter, but its value is limited in specifically identifying the principal problem. The identification of the patient’s principal problem during a clinical encounter represents the clinician’s synthesis of the historical, physical, and laboratory data obtained. The presenting com­ plaint is elicited early in the encounter with the expectation that it will serve to specifically direct the inquiry toward identification of the principal problem, a function that depends on the predictive value of the complaint. All complaints may not serve equally well in this regard, and the present- From the Division of Primary Care/Community Medicine, Department of Internal Medicine, University of Michigan, Ann Arbor, Ml. Requests for reprints should be addressed to Dr. Robert Burack, Department of Internal Medicine, Wayne State University, 540 E. Canfield, Detroit, Ml 48201. ing complaint may bear little resemblance to the principal problem ultimately defined.1'4 A disparity between complaint and problem implies the operation of a transition in the patient-physician interaction between the focus of the complaint and that of the problem. Such tran­ sitions may be initiated by the patient or by the physician in response to verbal or nonverbal pa­ tient cues, and a sensitivity to such cues may be an important clinical skill. In an effort to better define both the predictive value of the presenting com­ plaint and the clinician’s perception of such tran­ sition cues, a study was carried out to examine the clinical perception of concordance between the presenting complaint and the principal problem among new patients in an academic primary care setting. ® 1983 Appleton-Century-Crofts THE JOURNAL OF FAMILY PRACTICE, VOL. 16, NO. 4: 749-754, 1983 749 THE PRESENTING COMPLAINT Methods University Health Plan is a primary care inter­ nal medicine facility serving a population of both hospital employees, for whom care is provided as an employee benefit, and community patients. This facility is staffed by three faculty internists, three nurse clinicians, and 24 internal medicine residents, all of whom participated in the study. From April 1 through June 30, 1978, 131 newly en­ rolled patients were evaluated in a routine manner by a resident (85 percent of visits) or staff (15 per­ cent) primary care provider employing history and physical examination as well as office laboratory as appropriate (stool occult blood, urinalysis, or hematocrit determination). Informed consent was obtained prior to the encounter. Participating pa­ tients then completed a questionnaire that re­ quested demographic information and a statement of the presenting complaint. Following the en­ counter, the provider completed a questionnaire indicating the patient’s presenting complaint, an assessment of the patient’s principal problem, other identified problems, and any perceived un­ derlying motivation other than the presenting complaint that might have prompted the patient’s visit. Complete sets of questionnaires were avail­ able for the initial visits of the 103 patients who made up the study group. The complaints and problems were classified by the primary provider as somatic, psychosocial, or health maintenance and coded according to the In­ ternational Classification of Diseases (ICDA).5 Psychosocial problems were those classified in the ICDA as mental disorders and included traditional psychiatric diagnoses, situational disturbances, and disorders of presumed psychogenic origin. Health maintenance problems included preven­ tive, administrative, and well-care services (ICDA Y codes). Any underlying motivation noted by the provider' was classified by that provider as being a somatic, psychosocial, or health maintenance concern, and its specific content was classified by ICDA code. In those encounters in which more than one presenting complaint was recorded, the complaint most closely concordant with the prin­ cipal problem was selected for analysis to avoid underestimating concordance levels. Complaint-problem concordance was assessed using a system similar to that of Freidin et al.4 The presenting complaint and principal problem, as identified by the provider, were said to be com­ pletely concordant if both represented a somatic problem of the same organ system, a psychosocial issue, or an identical health maintenance service. The complaint and problem were partially con­ cordant if they identified somatic problems of dif­ fering organ systems, a somatic manifestation of a psychosocial problem, or differing health mainte­ nance services. Completely discordant complaint problem pairs included those differing in their somatic, psychosocial, or health maintenance cat­ egorization with the exception of potential somatic manifestations of psychosocial problems (eg, headache as a manifestation of depression), which were classified as partially concordant. The sensitivity, specificity, and predictive value of the presenting complaint were assessed using the category of the complaint (somatic, psycho­ social, or health maintenance) as a “ test” for a principal problem of the same category.6 Thus, for a somatic problem, sensitivity is the proportion of all those with a somatic problem who present a somatic complaint (positive test). Specificity is the proportion of all those with a nonsomatic problem who present a nonsomatic complaint (negative test). Positive predictive value is the proportion of all those with a somatic complaint (positive test) who have a somatic problem, and negative predic­ tive value is the proportion of all those with a non­ somatic complaint (negative test) who have a non­ somatic problem. Groups were compared using chi-square analysis. Results A description of the 103 patients is presented in Table 1. These patients introduced 118 presenting complaints from which their clinicians generated 110 principal problems. The relationship between the category of the presenting complaint and that of the identified principal problem is indicated in Table 2. Overall, 55 (50 percent) of the complaint-problem pairs were completely con­ cordant, and 72 (65 percent) were either partially or completely concordant. Psychosocial principal problems were significantly less likely to have been introduced by a concordant presenting com­ plaint than either somatic or health maintenance problems (P < .01). Of the 46 somatic principal problems, 35 had presented as concordant somatic 750 THE JOURNAL OF FAMILY PRACTICE, VOL. 16, NO. 4, 1983 THE PRESENTING COMPLAINT This potential for complaint-problem disparity limits the ability of the presenting complaint to specifically focus clinical inquiry. The extent of this limitation can be assessed through considera­ tion of the performance of the presenting com­ plaint as a diagnostic test for the presence of a problem of matching content. While technological procedures are routinely subjected to such an analysis, clinical tools such as the presenting complaint are rarely, if ever, so evaluated. As­ sessed as such a “ diagnostic tool,” the presenting complaint was of limited value in predicting a principal problem of the same category. Somatic and health maintenance complaints provided sub­ stantially lower positive predictive values than did psychosocial complaints. Psychosocial com­ plaints, however, were distinctly uncommon and thus insensitive in identifying only 2 of the 35 psychosocial principal problems. The positive predictive value of a test for a given condition var­ ies directly with the prevalence of that condition. The positive predictive value of a somatic com­ plaint would therefore be higher in a population characterized by a higher prevalence of somatic disease (or less frequent psychosocial problems) than that observed in this study population. Thus the observation by Freidin et al4 of physician- patient concordance on the biological nature of the principal problem in 83 percent of return visit encounters is consistent with the high prevalence of chronic disease in their study population and, although not reported, would have produced a higher positive predictive value for somatic com­ plaints than that observed in the present study. In addition, higher concordance rates might be anticipated for return visits, to the extent that they represent mutually agreed upon agenda for follow-up, than for new patient visits.11 Sixty percent of the observed complaint- problem discordance (33 cases) occurred in en­ counters characterized by a psychosocial major problem, reflecting in part the prevalence of psy­ chosocial problems in a medical population that is reluctant to present them directly.1215 Patients experiencing psychosocial distress may be more sensitive to, and threatened by, minor somatic deviations and may seek medical “ caring” when other sources of social support prove inadequate. Tessler et al16 have prospectively demonstrated increased medical utilization among such dis­ tressed patients. THE JOURNAL OF FAMILY PRACTICE, VOL. 16, NO. 4, 1983 Another source of discordance observed in this population was the appropriate medical function of uncovering somatic disease in patients either requesting health maintenance (II cases) or pre­ senting with unrelated somatic symptoms (4 cases). Conversely, there were five encounters in which patients had evidently presented a somatic complaint as justification for a “ checkup.” In ad­ dition, the misattribution of symptoms by patient or clinician may produce complaint-problem dis­ parity.17,18 Furthermore, adherence by clinicians to a model focusing upon a solitary “ chief’ com­ plaint may preclude elicitation of other potentially congruent patient concerns.19 It should also be noted that the principal problems with which pre­ senting complaints are being compared are those diagnoses established at the conclusion of a single visit. It thus remains possible that an alternative diagnosis could be established over time that might more closely relate to the original presenting complaint. The presence of complaint-problem discord­ ance as perceived by the clinician does not imply that the complaint is irrelevant or that the clini­ cian’s assessment of the principal problem is cor­ rect. Just as patients may misattribute symptoms, so too may clinicians misinterpret complaints and apply priorities distinct from those of the patient in assessing the importance of problems. Thus, there is not a single optimum level of concordance. Rather, the appropriate level of concordance may vary among encounters; complete concordance implies that the physician never uncovers unsus­ pected illness, while complete discordance sug­ gests disregard of patient concerns. Appropriate concordance may be reflected in both measures of patient satisfaction and health outcomes, as sug­ gested by Starfield et al.11,20 The perception by the clinicians of the presence of an underlying patient motivation for the visit was an important element in the encounter, con­ tributing directly to the identification of 35 percent of the principal problems. The process by which clinicians identified this underlying motivation remains undefined. After identifying a principal problem seemingly unrelated to the presenting complaint, the clinicians may have re-examined the process by which they had arrived at the diag­ nosis and only in retrospect noted any underlying motivation. In certain encounters, however, there may have been cues suggesting the presence of a 753 THE PRESENTING COMPLAINT problem distinct from the presenting complaint, and these cues may have prompted the transition to consideration of a new problem. If the cues that prompt such transitions can be identified, and sensitivity to them acquired, they may prove to be a valuable addition to the presenting complaint in guiding clinical inquiry. References 1. Bain S, Spaulding W: The importance of coding presenting symptoms. Can Med Assoc J 97:953, 1967 2. McWhinney J : Beyond diagnosis: An approach to the integration of behavioral science and clinical medicine. N Engl J Med 287:384, 1972 3. Morrell DC: Symptom interpretation in general practice. J R Coll Gen Pract 22:297, 1972 4. Freidin RB, Goldman L, Cecil RR: Patient-physician concordance in problem identification in the primary care setting. Ann Intern Med 93:490, 1980 5. International Classification of Diseases, rev 8, adapted for use in the United States. National Center for Health Statistics (Hyattsville, Md). PHS publication No. 1693. Government Printing Office, 1968 6. Vecchio T J : Predictive value of a single diagnostic test in unselected populations. N Engl J Med 274:1171, 1966 7. Stevenson I: The Diagnostic Interview, ed 2. New York, Harper & Row, 1971, pp 41-42 8. Balint M: The Doctor, His Patient and the Illness. New York, International Universities Press, 1957 9. Jarsky A: Hidden reasons some patients visit doc­ tors. Ann Intern Med 94:492, 1981 10. Meyer E, Derogatis L, Miller M, et al: Medical clinic patients with emotional disorders. Psychosomatics 19 611 1978 11. Starfield B, Steinwachs D, Morris I, et al: Patient- doctor agreement about problems needing follow-up visit JAMA 242:344, 1979 12. Stoeckle J , Zola I, Davidson G: The quantity and significance of psychological distress in medical patients. J Chronic Dis 17:959, 1964 13. Locke B, Krantz G, Kramer M: Psychiatric need and demand in a prepaid group practice program. Am J Public Health 56:895, 1966 14. Locke B, Gardner E: Psychiatric disorders among the patients of general practitioners and internists. Public Health Rep 84:167, 1969 15. Gardner E: Emotional disorders in medical practice. Ann Intern Med 73:651, 1970 16. Tessler R, Mechanic D, Dimond M: The effect of psychological distress on physician utilization: A prospec­ tive study. J Health Soc Behav 17:353, 1976 17. Mechanic D: Social psychological factors affecting the presentation of bodily complaints. N Engl J Med 286: 1132, 1972 18. Imboden JB , Canter A, Cluff L: Symptomatic recov­ ery from medical disorders. JAMA 178:1182, 1961 19. Enelow A, Swisher S : Interviewing and Patient Care, ed 2. New York, Oxford University Press, 1979, pp 42,62 20. Starfield B, Wray C, Hess K, et al: The influence of patient-practitioner agreement on outcome of care. Am J Public Health 71:127, 1981 754 THE JOURNAL OF FAMILY PRACTICE, VOL. 16, NO. 4, 1983