 |
|
Personality Disorders Among Difficult Patients
Sean Schafer, MD;
David P. Nowlis, PhD
Arch Fam Med. 1998;7:126-129.
ABSTRACT
Objective To determine the association between "difficult" patient status and personality disorder.
Design A survey using the Diagnostic Interview for Personality Disorders.
Participants Twenty-one patients nominated by 9 family medicine providers who subjectively experienced their care as difficult and 22 control subjects systematically selected from the same practices.
Main Outcome Measure The presence of personality disorder measured by the Diagnostic Interview for Personality Disorders.
Results Personality disorders were more prevalent among the difficult patients: 7 of 21 difficult patients and 1 of 22 control subjects had at least 1 personality disorder (P=.02). Five of 7 difficult patients had dependent personality disorder. None of the providers realized that the difficult patients had personality disorders.
Conclusions Unrecognized personality disorder can make difficult provider-patient relationships more likely. Dependent personality disorder may be especially difficult. Improved physician awareness of personality disorders may lead to more effective understanding and treatment of some difficult patients.
INTRODUCTION
WHEN physicians experience negative reactions to patients, they often perceive the patients as "difficult."1-9 Such patients make up 15% to 30% of primary care practice populations.3, 5, 10-11 Physiologic or functional health outcomes may be worse for this group: they make more frequent medical visits, have lower Short Form Health Survey scores, receive more prescriptions, undergo more laboratory investigations, and receive more referrals1, 10, 12 than "nondifficult" control subjects. Physicians suffer significant stress from interactions with these patients.13
Negative feelings toward patients may be "important clinical data about the patient's psychology"4, 14 or even a "diagnostic tool"15 for recognition of underlying psychopathologic disorders. Groves4 described 4 stereotypical "hateful patients" with names evocative of specific personality disorders later described in the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III-R)16: dependent clingers, entitled demanders, manipulative help rejecters, and self-destructive deniers. Later, Hahn et al5 used a self-report questionnaire (the Personality Disorder Questionnaire–Revised) to demonstrate a greater prevalence of diagnosable personality disorder among their difficult patient sample. Greater Axis I psychopathologic disorder also has been demonstrated in the difficult patients, identified by the Difficult Doctor-Patient Relationship Questionnaire.5, 10 However, because self-report instruments have limited specificity for diagnosis of personality disorder,17-18 the initial finding of increased personality disorder among difficult patients remains to be replicated.
Difficulties in physician-patient relationships originate in the interplay of unique styles and personality characteristics of the physician and the patient and should not be ascribed solely to psychiatric disorder in the patient.19-20 However, some difficulties experienced by physicians within the physician-patient relationship may result from unrecognized psychopathologic disorders in patients, more specifically, personality disorder. The objective of the present study is to assess the relative prevalence of personality disorder among patients considered difficult by their physicians when compared with nondifficult control subjects.
PATIENTS AND METHODS
STUDY POPULATION
The study began in 1991 after approval from the Human Investigation Committee of Valley Medical Center in Fresno, Calif, and the California State University, Fresno, Committee on the Protection of Human Subjects. Primary care providers from 5 settings participated in the study: 3 family physicians in local private group practice; a family physician in local private solo practice; family nurse practitioners associated with the University of California at Davis/Merced Family Practice Residency Training Program; family physicians at the student health center at California State University, Fresno; and a faculty family practice physician associated with the University of California at San Francisco/Fresno Family Practice Residency. We intended this convenience sample of providers to represent a cross section of practice settings and the patient demographics of the local community. We also attempted to recruit patients from the practices of the family practice residents at the county hospital in Fresno but abandoned this approach because of low response rates.
Providers nominated 2 to 6 patients from their practice whose care they considered particularly difficult. Patients with Axis I or specific medical diagnoses were not excluded. Providers then described each patient they nominated using a list of 6 criteria derived from the medical literature on difficult patients. Providers cited 1 or more criteria for each patient: (1) feelings of frustration or aversion in the physician, (2) uncertainty of diagnosis, (3) a larger number of problems or complaints, (4) apparent lack of organic basis for complaints, (5) noncompliance with or nonresponse to standard medical therapy, and (6) inappropriate demands or dependence.
Nondifficult control subjects for each provider were identified by consecutive sampling of every third to seventh patient from the most recent fully scheduled week of appointments for the provider. (The exact frequency depended on the number of control subjects needed to generate a 2:1 ratio of control subjects to difficult patients from each practice.) Patients younger than 18 years and non–English-speaking patients were excluded. We contacted potential participants by mail, explaining the nature of the study and the general psychological content of the survey instrument and offering a $10 stipend. We then invited participants by telephone to be interviewed. All interviews were conducted in person in a private office.
For each patient, the interviewer recorded gender, education, and current and lifetime marital status.
INSTRUMENT
Patients were interviewed using the Diagnostic Interview for Personality Disorders (DIPD),21-22 a semistructured interview divided into 11 sections. Each section explicitly inquires about the criteria for an Axis II personality disorder as delineated in the DSM-III-R16 (antisocial, avoidant, borderline, compulsive, dependent, histrionic, narcissistic, paranoid, passive-aggressive, schizoid, and schizotypal). The organization of the DIPD is similar to that of another semistructured interview, the Structured Clinical Interview for DSM-III-R Personality Disorders.23 Reliability testing for the DIPD was done with psychiatric inpatients without evidence of major psychotic disorder or organic brain disorder. Multiple interviewers agreed on the global presence or absence of personality disorder for 95% of the patients. Kappa values for specific personality disorders ranged from 0.46 to 0.84.21
In general, interrater reliability is easy to achieve with the DIPD. Before undertaking the present study, the principal investigator who developed the DIPD trained 1 of us (D.P.N.) in the use of the instrument. He established interrater reliability with the originators of the interview, with global agreement on the presence of personality disorder and kappa values for specific disorders consistently exceeding published values. The author who was trained (D.P.N.), in turn, trained additional interviewers for the present study: 3 graduate students in clinical psychology, a psychologist with a PhD employed by the Merced residency program, a second-year medical student, and a patient education specialist. Interrater reliability higher than published values was established before interviewers conducted study interviews. All interviewers were unaware of the status (difficult or control) of study patients.
OUTCOMES
We evaluated 2 outcomes: (1) the presence or absence of personality disorder, defined by meeting the required threshold score of 2 on any of the individual personality disorder scales of the DIPD, and (2) the total symptom score, developed to summarize and simplify the DIPD data by simply adding the presence of all positive symptoms together with equal weighting.
ANALYSIS
Association of difficult status with personality disorder was evaluated with the Fisher exact test. We used Student t tests for independent samples to evaluate association of personality disorder with socioeconomic and demographic variables and the Pearson correlation coefficient to analyze correlation of total symptom score with total number of reasons for difficulty and demographic and socioeconomic variables.
RESULTS
PATIENTS
Providers nominated 39 patients as difficult: 8 by 2 of the physicians from the private family practice group, 4 by the faculty practice physician, 6 by the solo practice family physician, 13 by 2 physicians in the university student health center, and 8 by the family nurse practitioners of the Merced residency program. Review of office schedules yielded a list of 62 control subjects: 17 from the private group practice, 7 from the faculty practice physician, 9 from the physician in solo private practice, 23 from the student health center, and 6 from the Merced family nurse practitioners. We completed interviews with 21 difficult patients and 22 control subjects, for an overall response rate of 43% (54% of difficult patients and 35% of control subjects).
Among those interviewed, control and difficult groups did not differ in years of education, current marital status, social status, gender, or age (Table 1).
|
|
|
|
Table 1. Subject Characteristics
|
|
|
TESTS OF THE HYPOTHESIS
Patients nominated as difficult by their regular provider were more likely than control subjects to have at least 1 personality disorder according to the DIPD (7 of 21 difficult patients and 1 of 22 control subjects; P=.02, 1-tailed test) (Table 2). Total symptom score (the sum of the interview) correlated with the total number of reasons for difficulty cited by the primary care provider (range, 1-6) among difficult patients (r =0.52; P<.01, 2-tailed test).
|
|
|
|
Table 2. Personality Disorder in Patients and Controls
|
|
|
DEMOGRAPHIC VARIABLES
Gender, education, social status, and current or lifetime marital status were not associated with presence of personality disorder or any difference in total symptom score.
TYPES OF PERSONALITY DISORDERS
Of the difficult patients diagnosed as having personality disorder, 3 had dependent disorder, 2 had paranoid disorder, 1 had dependent and obsessive disorders, and 1 met criteria for 6 distinct personality disorders, including dependent disorder. The 1 control subject with personality disorder had paranoid personality disorder (Table 2).
COMMENT
Are physicians' most difficult patients more likely to have DSM-III-R –based personality disorders? The present study suggests that the answer is yes. In addition to association of personality disorder with difficult patients, total personality disorder symptom score (an experimental index summing all pathologic personality traits measured by the DIPD with equal weighting) correlated with the number of reasons cited by a clinician for difficulty with a patient.
Interpretation of our findings is hindered by several limitations. We encountered a low response rate; only 35% of control subjects were interviewed, which may have led to significant response bias. Clearly, there are personality disorders, such as schizoid, paranoid, antisocial, and avoidant, with features that might lead a patient to avoid unnecessary contact with health care providers and to avoid our interview. Also, including criteria for nomination may have biased our sample toward specific personality disorders.
Although 2 screening instruments exist for the identification of difficult patients, the Difficult Doctor-Patient Relationship Questionnaire5 and the Patient Description Questionnaire,12 we chose to acquire our sample by asking physicians who were unaware of the study hypothesis to designate their most conspicuous difficult patients. We chose this design because we were most interested in the diagnostic sensitivity of negative physician affect, reasoning that physicians would be most likely to be distressed by those patients to whom they had the strongest emotional reactions. Moreover, the Patient Description Questionnaire was reliability tested on a preselected population of difficult patients nominated by their physicians according to criteria similar to those used in this study: (1) "presentation with multiple symptoms across physical systems," (2) "excessive demands for care and attention," or (3) "sickness prone with a low threshold for complaints." Similarly, many of the Difficult Doctor-Patient Relationship Questionnaire questions were written to identify patients who invoked strong subjective responses in their physicians. Cutoff thresholds were chosen such that physicians viewed nearly all problem patients as "demanding and irritating." It is possible, however, that our less systematic selection biased the sample toward patients with personality disorder. Further research is needed with patients screened via various difficult patient criteria from different primary care practice populations to verify and extend our finding of more personality disorder among difficult patients.
We did not control for Axis I or specific short- or long-term medical diagnoses in this study. When we interviewed the primary care providers on completion of the study, most difficult patients identified with personality disorder had specific or suspected Axis I diagnoses. Subsequent research will need to control for concurrent Axis I disorders.
The nature and validity of personality disorders as defined by DSM-III-R has been a subject of prolonged and complicated discussion and research.24 Diagnosis has traditionally been problematic, with poor interrater agreement being common regardless of the method used.23 The DIPD is one of several semistructured and self-report instruments designed for the diagnosis of DSM personality disorders. Others include the Structured Interview for DSM-III Personality Disorders,25 the Structured Clinical Interview for DSM-III-R Personality Disorders,26 and the Personality Diagnostic Questionnaire.18 The DIPD has not been validated with primary care, nonpsychiatric patients or compared with other methods for diagnosis of personality disorder. It is similar in design to the Structured Clinical Interview for DSM-III-R Personality Disorders, which also has been compared in published reports17, 27-31 with other methods for diagnosis of Axis II disorders. Unfortunately, diagnostic concordance between methods generally has been poor, with median chance corrected agreement approximately 0.25 between compared methods.23
Although the study was not designed to establish prevalence of any single disorder among the study groups, dependent personality disorder was present as an isolated disorder or as one of multiple personality disorders among 5 of 7 patients identified as difficult by their regular providers. Dependent personality disorder is characterized by excessive need to be taken care of, and the trend toward inclusion of these patients in this sample is consistent with some of the anecdotal literature4, 32 suggesting that patients with "clinging" features have always been difficult for primary care providers. Surprisingly, given the frequency of reports of aversion to patients with the disorder,33 only 1 individual from the difficult group was diagnosed as having borderline disorder (a patient meeting criteria for several distinct personality disorders). However, our sample is small; although the general finding of personality disorder among the difficult group did achieve statistical significance, the sample is inadequate to address the question of relative frequency of specific disorders among this group.
Personality disorder is not the only reason for difficult patients in the primary care setting. Physician-patient difficulties also can be a problem of the relationship.9, 19 No proven treatment is available for personality disorders other than reducing morbidity and mortality with appropriate treatment for concurrent Axis I and medical diagnoses.24 Patients with personality disorders have trouble adapting to the trials and misfortunes of life. Without help from their health care professionals, morbidity and mortality are high among patients with associated medical and psychological conditions.24, 33 Realistic prognosis for a difficult patient is more likely when personality disorders are detected. Other potential benefits of diagnosis include improved understanding of the patient and of the provider's feelings, more rapid and informed requests for help, and decreased cost to the medical system from unnecessary medical intervention. Recent research results also suggest that understanding the nature of the patient's problem can alleviate some of the "heart sink" these patients typically cause in the physician.33-34
Among physicians, difficult patients are apparently a common phenomenon. If unprocessed or not understood further, the feelings of aversion and discomfort that such patients evoke can compromise the therapeutic interaction. Although personality disorders may not be responsible for the behavior of most difficult patients, information such as that reported here could help the reflective clinician recognize a difficult relationship constructively, moving to potential explanations and beyond, to interdisciplinary strategies for more effective management and satisfactory outcomes.
AUTHOR INFORMATION
Accepted for publication December 30, 1996.
This study was funded by a grant from the California Academy of Family Physicians Foundation/Center for Studies in Family Medicine, San Francisco.
We thank Nancy Bleile, MD, and Jean Crane, MD, who nominated participants and helped recruit other participating clinicians, and Lynn Lester, PhD, Jim McDiarmid, PhD, Karen Siscoe, PhD, and Liz Wolheim, PhD, who conducted structured interviews.
Reprints not available from the authors.
From the Department of Family Practice, University of California at San Francisco/Fresno Medical Education Program, Fresno.
REFERENCES
1. John C, Schwenk TL, Roi LD, Cohen M. Medical care and demographic characteristics of "difficult" patients. J Fam Pract. 1987;24:606-610.
2. Carson J, Norris G, Haworth R. The difficult patient. Practitioner. 1993;237:313-317.
WEB OF SCIENCE
| PUBMED
3. Crutcher JE, Bass MJ. The difficult patient and the troubled physician. J Fam Pract. 1980;6:933-938.
4. Groves JE. Taking care of the hateful patient. N Engl J Med. 1978;298:883.
WEB OF SCIENCE
| PUBMED
5. Hahn SR, Thompson KS, Wills TA, Stern V, Budner NS. The difficult doctor-patient relationship: somatization, personality and psychopathology. J Clin Epidemiol. 1994;47:647-657.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
6. Lechky O. There are easy ways to deal with difficult patients, MDs say. Can Med Assoc J. 1992;146:1793-1795.
PUBMED
7. Malcom R, Foster HK, Smith C. The problem patient as perceived by family physicians. J Fam Pract. 1977;5:361-364.
WEB OF SCIENCE
| PUBMED
8. Schuller AB. About the problem patient. J Fam Pract. 1977;4:653-654.
WEB OF SCIENCE
| PUBMED
9. Schwenk TL, Marquez JT, Lefever D, Cohen M. Physician and patient determinants of difficult physician-patient relationships. J Fam Pract. 1989;28:59-63.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
10. Hahn SR, Kroenke K, Spitzer RL, et al. The difficult patient: prevalence, psychopathology, and functional impairment. J Gen Intern Med. 1996;11:1-8.
WEB OF SCIENCE
| PUBMED
11. Lin EH, Katon W, Von Koff M, et al. Frustrating patients: physician and patient perspectives among distressed high users of medical services. J Gen Intern Med. 1991;6:241-246.
WEB OF SCIENCE
| PUBMED
12. McGahie WC, Whitenack RH. A scale of measurement of the problem patient labeling process. J Nerv Ment Dis. 1982;170:598-604.
WEB OF SCIENCE
| PUBMED
13. McCue JD. The effects of stress on physicians and their medical practice. N Engl J Med. 1982:306;458-463.
14. Smith S. Dealing with the difficult patient. Postgrad Med J. 1995;71:653-657.
FREE FULL TEXT
15. Goodwin JM, Goodwin JS, Kellner R. Psychiatric symptoms in disliked medical patients. JAMA. 1979;241:1117-1120.
FREE FULL TEXT
16. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised. Washington, DC: American Psychiatric Association; 1987.
17. Zimmerman M, Coryell WH. Diagnosing personality disorders in the community: a comparison of self-report and interview measures. Arch Gen Psychiatry. 1990;47:527-531.
FREE FULL TEXT
18. Hyler SE, Rieder RO, Willimas JBW, Spitzer RL, Lyons M, Hendler J. A comparison of clinical and self-report diagnoses of DSM-III personality disorders in 552 patients. Compr Psychiatry. 1989;30:170-178.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
19. Anstett R. The difficult patient and the physician-patient relationship. J Fam Pract. 1980;11:281-286.
WEB OF SCIENCE
| PUBMED
20. Schwenk TL, Romano SE. Managing the difficult physician-patient relationship. Am Fam Physician. 1992;11:1503-1509.
21. Zanarini MC, Frankenburg FR, Chauncer DL, Gunderson JG. The Diagnostic Interview for Personality Disorders: interrater and test-retest reliability. Compr Psychiatry. 1987;28:467-480.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
22. Zanarini MC, Gunderson JG, Frankenberg FR, Chauncer DL. Discriminating borderline personality disorder from other Axis II disorders. Am J Psychiatry. 1990;147:161-167.
WEB OF SCIENCE
| PUBMED
23. Perry JC. Problems and considerations in the valid assessment of personality disorders. Am J Psychiatry. 1992;149:1645-1653.
WEB OF SCIENCE
| PUBMED
24. Gunderson JG. Personality disorders. In: Nichol AM Jr, ed. The New Harvard Guide to Psychiatry. Cambridge, Mass: Harvard University Press; 1988:337-357.
25. Pfohl B, Stangl DA, Zimmerman M, Bowers W, Corenthal C. Structured Interview for DSM-III Personality Disorders (SIDP). Iowa City: University of Iowa; 1982.
26. Spitzer RL, Williams JBW, Gibbon M. Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II). New York: New York State Psychiatric Institute, Biometrics Research; 1987.
27. Hyler SE, Skodol AE, Kellman HD, Oldham JM, Rosnick L. Validity of the Personality Diagnostic Questionnaire–Revised: comparison with two structured interviews. Am J Psychiatry. 1990;147:1043-1048.
WEB OF SCIENCE
| PUBMED
28. Hogg B, Jackson HJ, Rudd RP, Edwards J. Diagnosing personality disorders in recent-onset schizophrenia. J Nerv Ment Dis. 1990;178:194-199.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
29. O'Boyle M, Self D. A comparison of two interviews for DSM-III-R personality disorders. Psychiatry Res. 1990;32:85-92.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
30. Pilkonis PA, Heape CL, Ruddy J, Serrao P. Validity in the diagnosis of personality disorders: the use of the LEAD standard. Psychol Assess. 1991;3:1-9.
31. Skodol AE, Oldham JM, Rosnick L, Kellman HD, Hyler S. Diagnosis of DSM-III-R personality disorders: a comparison of two structured interviews. Int J Methods Psychiatr Res. 1991;1:13-26.
32. Finkel JB, Hageman E. Working with patients who present special problems. In: Kornfield DS, Finkel JB, eds. Psychiatric Management for Medical Practitioners. New York, NY: Grune & Stratton Inc; 1982:19-46.
33. Nowlis D. Borderline personality disorder in primary care. J Fam Pract. 1990;30:329-335.
WEB OF SCIENCE
| PUBMED
34. O'Dowd TC. Five years of heartsink patients in general practice. BMJ. 1988;197:528-530.
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
The Potential Role of Dependency in the Weaning Process: The Case of a 57-Year-Old Woman Connected to a Mechanical Ventilator
Ayalon
Clinical Case Studies 2007;6:459-467.
ABSTRACT
Understanding and Engaging the Hostile Patient
Kahn
Mayo Clin Proc. 2007;82:1532-1534.
FULL TEXT
How Respected Family Physicians Manage Difficult Patient Encounters
Elder et al.
J Am Board Fam Med 2006;19:533-541.
ABSTRACT
| FULL TEXT
Borderline Personality Disorder in Primary Care
Gross et al.
Arch Intern Med 2002;162:53-60.
ABSTRACT
| FULL TEXT
The difficult patient' as perceived by family physicians
Steinmetz and Tabenkin
Fam Pract 2001;18:495-500.
ABSTRACT
| FULL TEXT
The Physician-Patient Relationship: Three Psychodynamic Concepts That Can Be Applied to Primary Care
Goldberg
Arch Fam Med 2000;9:1164-1168.
ABSTRACT
| FULL TEXT
|
