Copyright (c) 1988 American Medical Association Arch Gen Psychiatry 1988; 45: 64-68 January, 1988 SECTION: ORIGINAL ARTICLE LENGTH: 3583 words TITLE: Prevalence of Major Depression, Simple Phobia, and Other Psychiatric Disorders in Patients With Long-standing Type I Diabetes Mellitus AUTHOR: Michael K. Popkin, MD; Allan L. Callies; Richard D. Lentz, MD; Eduardo A. Colon, MD; David E. Sutherland, MD, PhD ABSTRACT: To examine the prevalence of psychiatric disorders in patients with long-standing type I diabetes mellitus, we assessed a series of candidates for pancreas transplantation. Using the Diagnostic Interview Schedule, six-month and lifetime prevalences of psychiatric disorders were established for the candidates and their potential donors (first-degree relatives). Excluding tobacco use disorder and psychosexual dysfunction, 38 diabetic subjects (51%) received one or more psychiatric diagnoses. The lifetime prevalence of major depression was comparable for female (11 of 48 [22.9%]) and male (seven of 27 [25.9%]) diabetics; both rates were significantly higher than rates in first-degree relatives and the general population. Among female diabetics, the six-month and lifetime prevalences of simple phobia were increased vs donors and the general population; among male diabetics, the lifetime prevalence of antiscocial personality disorder was greater than that in the general population. None of these disorders was found to be related to the duration of diabetes or the presence of various complications. The data suggest that increased rates of psychiatric disorder in type I diabetics have both gender-independent and gender-related components. TEXT: Despite growing recognition that type I diabetes mellitus is a disease with profound emotional and psychosocial impact, there are few systematic studies in the literature addressing the nature and extent of psychiatric disorder (as defined per DSM-III) in type I diabetic populations. Some attention has been directed to the question of cognitive dysfunction emerging in adolescents with type I diabetes mellitus. [n1,n2] Perlmuter et al [n3] identified memory impairment in type II diabetics vs controls. In a 1983 critical appraisal of the problem of depression in diabetics, Lustman et al [n4] concluded, "the incidence and prevalence of depression in diabetics is unknown." Using the Diagnostic Interview Schedule (DIS), [n5] Lustman et al. [n6,n7] subsequently reported prevalences of various psychiatric disorders in samples of type I and type II diabetics. We recently reported the prevalence and distribution of psychiatric disorders in a series of early-onset type I diabetic patients hospitalized on medical-surgical services of a university hospital and referred for psychiatric consultation. [n8] In clinical examinations using the DSM-III criteria, psychiatric consultants diagnosed an affective disorder in 25%, adjustment disorder in 17%, organic mental disorder in 15%, and personality disorder in 11.5%. The overall distribution of the psychiatric diagnoses for the diabetic sample did not differ significantly from that which characterized a series of 1002 nondiabetic, medical-surgical patients referred for psychiatric consultation in the same periods and setting. However, the incidence of organic mental disorder in the diabetic subjects was comparable with that seen in a much older medical-surgical population. We suggested that this increment was most likely a function of vascular or hypoglycemic insults. To extend our investigations, we studied a consecutive series of patients with a course of ten years or more of type I diabetes mellitus presenting to the Clinical Research Center (CRC) of the University of Minnesota Hospital, Minneapolis, as candidates for pancreas transplantation. Pancreas transplantation is an emerging modality for the treatment of insulin-dependent diabetes mellitus. [n9] To date, the University of Minnesota program has performed over 130 such transplants. [n10] In contrast to the consultation sample, [n8] pancreas transplantation candidates were routinely assessed by our team with structured interviews independent of manifest psychiatric symptomatology or a perceived need for psychiatric intervention. Their admissions were elective and not associated with acute medicalsurgical illness. The data from this study (particularly the rates of major depression and phobic disorder) raise numerous concerns and underscore the need for full examination of the nature and course of psychiatric disturbances in patients with type I diabetes mellitus. PATIENTS AND METHODS All candidates for pancreas transplantation admitted to the University of Minnesota Hospital's CRC between February 1984 and February 1986 were evaluated. Potential transplant recipients must meet a series of established entry criteria, including (1) at least a ten-year course of type I diabetes mellitus and (2) nephropathy with albuminuria greater than 30 mg/d and abnormal biopsy results (eg, mesangial thickening greater than 1+ on light microscopic examination). Potential recipients are excluded by any of the following: (1) advanced nephropathy with serum creatinine level greater than 180 mu mol/L (>2.0 mg/dL), 4+ mesangial change on biopsy, or creatinine clearance less than 0.50 mL/s (<30 mL/min); (2) serious and active large-vessel disease; or (3) major stroke or persisting transient ischemic attacks. During the initial CRC admission, each candidate was assessed by one of us (A.L.C.) using the DIS. [n5] The use of a single rater constitutes a methodologic limitation of the study. The DIS, developed under the auspices of the National Institute of Mental Health, provides a structured interview format for psychiatric assessment according to DSM-III criteria. Lustman et al [n6] have demonstrated the suitability of using the DIS in a diabetic population. Lifetime and six-month prevalences of psychiatric disorders identified by the DIS were tabulated. As a comparison group, all potential pancreas donors seen in the same interval were also assessed by the same examiner (A.L.C.) using the DIS. (Potential donors are first-degree relatives of the proposed recipient.) A second comparison was effected with prevalences estimated for the general population (using the DIS) from the Epidemiological Catchment Area (ECA) study. [n11,n12] The ECA data were presented separately for each of the three study sites. For comparison, overall prevalences were computed by summing the number of individuals with a given diagnosis for each site and dividing by the total number of study subjects. As the New Haven (Conn) study site employed a different version of the DIS phobia section, prevalences for phobias were computed using only data from the remaining two sites. The prevalence of psychiatric disorders in the group of potential recipients with respect to gender and ECA age group was examined for differences vs the two comparison groups using chi-square tests with Yates' correction. RESULTS In the two-year study interval (February 1984 through February 1986), a total of 75 consecutive candidates for pancreas transplantation were assessed with the DIS. The sample consisted of 48 (64%) women and 27 (36%) men with a median age of 31 years (range, 16 to 55 years) and a median duration of type I diabetes mellitus of 20 years (range, ten to 42 years). At the time of evaluation, 13 (17%) had already received renal transplantation, ten (13%) were blind, an additional eight (11%) had complete loss of vision in one eye, 67 (89%) had some degree of retinopathy, and 58 (77%) were suffering from varying degrees of neuropathy. In the same time interval, 34 potential donors were also evaluated with the DIS. (Potential recipients are not required to have a designated potential donor, as cadaveric transplantations are also performed.) Of the donor group, 19 (56%) were women and 15 (44%) were men; median age was 36 years (range, 21 to 66 years). In the second comparison group (ECA studies [n11,n12]), 60% of subjects were women and the minimum age was 18 years. Table 1 gives the diagnoses assigned to the potential pancreas transplantation recipients. For the 41 subjects assessed since 1985, an expanded DIS was used that included sections for generalized anxiety disorder, posttraumatic stress disorder, and bulimia. In all, 13 major diagnostic categories were identified in the potential recipient group. Excluding tobacco use disorder, 44 (59%) of the candidates received one or more psychiatric diagnoses. Of particular note are the prevalences for major depression, phobic disorder, generalized anxiety disorder, and psychosexual dysfunction. Lifetime rates for these groups ranged from 24% to 32%. Six-month prevalences ranged from 11% to 20%. Table 1. -- Prevalences of Psychiatric Disorders in 75 Potential Pancreas Transplant Recipients [SEE ORIGINAL SOURCE] Table 2 summarizes significant differences in lifetime prevalences by gender. Among female subjects, the lifetime prevalence of simple phobia in the recipient group was greater than that in the donor group (chi-square = 5.96, df = 1, P = .015) and greater than that in the general population (chi-square = 4.39, df = 1, P = .036). The lifetime prevalence of major depression in the recipient group was significantly greater than that in the general population for both female (chi-square = 15.44, df = 1, P<.001) and male (chi-square = 37.85, df = 1, P<.001) subjects. Comparable rates of major depression were found for female recipients (22.9%) and male recipients (25.9%). The lifetime prevalences of major depression in the donor groups appear closer to those in the general population but are not statistically different from recipients when stratified by gender, likely due to small sample sizes. (The prevalence for the combined group of female and male donors is lower than for the combined group of female and male recipients; chi-square = 4.51, df = 1, P=.034.) Among male subjects, the lifetime prevalence of antisocial personality disorder for the recipient group was greater than that in the general population (chi-square = 4.17, df = 1, P = .041). Table 2. -- Differences in Lifetime and Six-Month Prevalences of Psychiatric Disorders by Gender [SEE ORIGINAL SOURCE] Table 3 summarizes significant differences in lifetime prevalences for individuals within the 25- to 44-year age group. Seventy-six percent of potential pancreas transplant recipients, 62% of potential donors, and 39% of ECA subjects are included within this interval. The remaining small number of recipients precluded analysis of prevalences of other age groups. Lifetime prevalences for this age group (25 to 44 years) stratified by gender for the general population were unavailable. The lifetime prevalence of phobic disorder in the recipient group was greater than that in the donor group (chi-square = 6.35, df = 1, P = .012). Notably, the rate of simple phobia in the recipient group was greater than that in the donor group (chi-square = 5.79, df = 1, P = .016) and greater than that in the general population (chi-square = 6.58, df = 1, P = .010). The lifetime prevalence of major depression in the group of recipients was greater than that in the general population (chi-square = 13.71, df = 1, P<.001). Table 3. -- Differences in Lifetime Prevalence of Psychiatric Disorders for 25- to 44-Year Age Group [SEE ORIGINAL SOURCE] Table 2 gives significant differences in six-month prevalences by gender. Among female subjects, the rate of phobic disorder in the recipient group was greater than that in the general population (chi-square = 9.36, df = 1, P = .002). The rate of simple phobia exceeded that in the general population (chi-square = 9.68, df = 1, P = .002) and that in the donor group (chi-square = 4.77, df = 1, P = .029). The rate of major depression for the group of female recipients was greater than that in the general population (chi-square = 11.22, df =1, P<.001). Among male subjects, six-month prevalences of psychiatric disorders in the group of potential recipients were comparable with rates in the group of donors and in the general population. Table 4 gives significant differences in six-month prevalences for the 25- to 44-year age group. Among female subjects, the rate of phobic disorder in the recipient group was greater than that in the general population (chi-square = 16.41, df = 1, P<.001) and greater than that in the donor group (chi-square = 6.05, df = 1, P = .014). In particular, the rate of simple phobia was greater than that in the general population (chi-square = 16.73, df = 1, P<.001) and greater than that in the donor group (chi-square = 5.33, df = 1, P = .021). The rate of major depression in the female recipient group was greater than that in the general population (chi-square = 7.07, df = 1, P = .008). Six-month prevalences of psychiatric disorders in the male recipient group did not differ from those in the general population population and in the donor group. Table 4. -- Differences in Six-Month Prevalence of Psychiatric Disorders by Gender for 25-to 44-Year Age Group [SEE ORIGINAL SOURCE] Within the group of potential recipients, lifetime and six-month prevalences of major depression, simple phobia, and antisocial personality disorder were found to be unrelated to duration of diabetes, previous renal transplantation, retinopathy, impairment of vision, or neuropathy. COMMENT The data of this study suggest that patients with longstanding type I diabetes mellitus have high lifetime and six-month prevalences of psychiatric disorder when viewed in comparison with their first-degree relatives and the general population. [n11,n12] Excluding tobacco use disorder and psychosexual dysfunction, 38 (51%) of 75 of the diabetic subjects were assigned at least one psychiatric diagnosis, compared with 12 (35%) of 34 of potential donors and 33% of ECA subjects. Most striking is the 24% lifetime prevalence of major depression, which could not be attributed to gender, age group, duration of illness, or the presence of various diabetic complications. Notably, the criteria for pancreas transplantation exclude patients with identified vascular disease and stroke, a group at increased risk for mood disorder and for organic mental disorder. [n13,n14] This rate is comparable with our previous finding that 25% of medical-surgical patients with early-onset diabetes referred for psychiatric consultation were diagnosed as having affective disorder. [n8] Using the DIS, Lustman et al [n7] recently reported a 26% lifetime prevalence of major depression in a group of type I diabetics. The extent of Affective disorder in the group of potential pancreas transplant recipients can also be contrasted with findings in other studies examining medical patients. Hoeper et al, [n15] using the Schedule for Affective Disorders and Schizophrenia, reported a 5.8% lifetime prevalence of major depression in primary care practice patients. Helzer et al, [n16] using the Renard Research Interview, found a 36% lifetime prevalence of major depression in a group of patients with Crohn's disease. Leeper et al, [n17] using the DIS, found a 9.8% six-month prevalence of major depression in a study of ambulatory medical patients presenting for disability determinations. Schulberg et al, [n18] using the DIS, found a 6.1% one-month prevalence of major depression in a primary care population. Although the lifetime and six-month prevalences of major depression in our type I diabetic sample may not be remarkable when viewed against these medical populations, they are much higher than rates found in the ECA study and in the subjects' first-degree relatives. We have elsewhere commented on the pressing need for greater attention to the problem of depressive syndromes in the medically ill. [n19] Tolerance for mood disorder in general medical patients may be excessive. [n20] If physicians caring for diabetic patients view depression as an understandable concomitant (in a psychological, reactive sense) condition with the progressive course of type I diabetes mellitus, patients may be denied meaningful psychiatric intervention. This prospect is consistent with recent evidence that physicians caring for diabetic patients recognize and treat depression in as little as one third of cases. [n21] Adding to this concern is evidence that insulin requirements are affected by major depression. [n22,n23] The high lifetime prevalence of major depression in the study sample may be a function of subtle changes in the vasculature of these patients' central nervous systems. Postmortem studies show that the cerebral vasculature of the diabetic undergoes changes akin if not identical to those in the kidney. [n24,n25] The possible behavioral and cognitive sequelae of such changes have received little attention. Further study might clarify whether the rates of psychopathology reported herein are specific to diabetes. Roth [n26] and Post [n27] have suggested the prospect of a special relationship between cerebrovascular disease and affective illness. The prospect of "secondary depression" [n28] is heightened by the absence of a gender predominance in the depressed diabetic patients. The high six-month and lifetime prevalences of simple phobia in the female diabetic subjects is somewhat surprising. The rates appear similar to data from other studies of diabetics [n6,n7] that were not stratified by age and gender. Review of the specific nature of the phobias is not particularly illuminating; only one patient had a fear of needles and one was phobic regarding hospitals. Hypoglycemia, autonomic disturbances, and central nervous system changes must be considered as possible contributors to the emergence of phobias in this group. That the rate of simple phobia is increased only for female diabetics would suggest that such effects differ by gender. Perhaps female diabetics are more vulnerable to such variations and to behavioral consequences of them. Contributions and limitations of the learning theory model for anxiety and phobias have received growing attention in the literaturs. [n29] Sheehan and Sheehan [n30] have proposed a classic conditioning paradigm in the emergence of phobias. In the context of diabetes mellitus, acute symptoms secondary to hypoglycemia or chronic autonomic instability associated with neuropathy may be paired with environmental cues and lead to the emergence of phobias. Phobic disorders are encompassed under the broader rubric of anxiety disorders in DSM-III. It is important to recall that organic origins for anxiety are numerous. [n31] Persistent or permanent changes in brain structure and function due to hypoglycemia must be considered. Growing evidence points to metabolic and structural aberrations secondary to hypoglycemia, including transient changes in amino acids and damage to particular neuronal groups. [n32] The evolution of phobic disturbances under such circumstances perhaps represents potentiation or unmasking of earlier proclivities. Carefully recorded medical histories might clarify this point. Anxiety and anxiety disorders may be seen as part of the reactive or psychological response to the course and progression of diabetic illness. Though the 31.7% (13/41) rate of generalized anxiety disorder in the diabetic subjects appears high vs that in their first-degree relatives (9.5% [2/21]), the observed difference is not statistically significant, possibly owing to small sample size. Similarly, Lustman et al [n7] observed high rates of generalized anxiety disorder in both type I and type II diabetics, 44.4% and 37.5%, respectively. It seems probable that the high rates of anxiety disorders and mood disorders reflect a combination of organic and reactive factors. The sample of male diabetic subjects had a higher lifetime prevalence of antisocial personality disorder than the general population, but a rate comparable with the group of male potential donors. One can speculate that potential recipients and donors who characteristically reject social restrictions and limitations would behave accordingly as regards diabetes. Pancreas transplantation would be sought out as these individuals would be unwilling to accept the "inevitable" progression of diabetes and/or the limitations it imposes on the diabetic. A European study of diabetic outpatients [n33] found that among those with onset of diabetes mellitus by age 20 years and with 20 years' duration of their illness, approximately 65% had retinopathy, 50% had neuropathy, and 15% had nephropathy. Although the rates of complications among our study subjects are somewhat higher, their presence was found to be unrelated to the increased rates of major depression, simple phobia, and antisocial personality disorder. High rates of depression and anxiety disorders have been found in other studies of diabetic patients. [n6,n7] Thus, the prevalence and nature of psychiatric disorders among candidates for pancreas transplantation appear representative of patients with long-standing type I diabetes mellitus. The data presented herein suggest that increased rates of psychiatric disorder in type I diabetics have both genderindependent and gender-related components. They speak to the need for systematic, longitudinal studies of the nature and extent of psychopathology in patients with type I diabetes mellitus. Such knowledge is prerequisite to the provision of optimal care. SUPPLEMENTARY INFORMATION: Accepted for publication April 10, 1987. From the Departments of Psychiatry (Drs Popkin, Lentz, and Colon and Mr Callies), Medicine (Dr Popkin), and Surgery (Dr Sutherland), University of Minnesota Medical School, Minneapolis. Reprint requests to Box 345, Mayo Building University Hospitals, 420 Delaware St SE, Minneapolis, MN 55455 (Dr Popkin). REFERENCES: [n1.] Ryan C, Vega A, Drash A, Longstreet C: Neuropsychological changes in adolescents with insulin-dependent diabetes. J Consult Clin Psychol 1984;52:335-342. [n2.] Ryan C, Vega A, Drash A: Cognitive deficits in adolescents who developed diabetes early in life. Pediatrics 1985;75:921-927. [n3.] Perlmuter LC, Hakami MK, Hodgson-Harrington C, Ginsberg, J, Katz J, Singer DE, Nathan DM: Decreased cognitive function in aging non-insulin dependent diabetic patients. Am J Med 1984;77:1043-1048. [n4.] Lustman PJ, Amado H, Wetzel RD: Depression in diabetes: A critical appraisal. Compr Psychiatry 1983;1:65-74. [n5.] Robins LN, Helzer, JE, Croughan J, Rutcliff KS: NIMH Diagnostic Interview Schedule: Its history, characteristics, validity. Arch Gen Psychiatry 1981;38:381-389. [n6.] Lustman PJ, Harper GW, Griffith LS, Clouse RE: Use of the Diagnostic Interview Schedule in patients with diabetes mellitus. J Nerv Ment Dis 1986;174:743-746. [n7.] Lustman PJ, Griffith LS, Clouse ER, Cryer PE: Psychiatric illness in diabetes mellitus: Relationship to symptoms and glucose control. J Nerv Ment Dis 1986;174:736-742. [n8.] Popkin MK, Callies AL: Psychiatric consultation to inpatients with early onset type I diabetes mellitus in a university hospital. Arch Gen Psychiatry 1987;44:169-171. [n9.] Sutherland DER, Kendall DM: Pancreas transplantation: Registry Report and a commentary. West J Med 1985;143:845-852. [n10.] Sutherland DER, Goetz FC, Kendall DM, Najarian JS: One institution's experience with pancreas transplantation. West J Med 1985;143:838-844. [n11.] Robins LN, Helzer JE, Weissman MM, Orvaschel H, Gruenberg E, Burke JD, Regier DA: Lifetime prevalence of specific psychiatric disorders in three sites. Arch Gen Psychiatry 1984;41:949-958. [n12.] Myers JK, Weissman MM, Tischler GL, Holzer CE, Leaf PJ, Orvaschel H, Anthony JC, Boyd JH, Burke JD, Kramer H, Stoltzman R: Six-month prevalence of psychiatric disorders in three communities. Arch Gen Psychiatry 1984;41:959-967. [n13.] Robinson RG, Lipsey JR: Cerebral localization of emotion based on clinical-neuropathological correlations: Methodological issues. Psychiatr Dev 1985;4:335-347. [n14.] Robinson RG, Kubos KL, Starr LB, Rao K, Price TR: Mood disorders in stroke patients: Importance of location and lesion. Brain 1984;107:81-93. [n15.] Hoeper E, Nycz G, Cleary P, Regier D, Goldberg I: Estimated prevalence of RDC mental disorder in primary medical care. Int J Ment Health 1979;8:6-15. [n16.] Helzer JE, Chammas S, Norland CC, Stillings WA, Alpers DH: A study of the association between Crohn's disease and psychiatric illness. Gastroenterology 1984;86:324-330. [n17.] Leeper J, Badger L, Milo T: Mental disorders among physical disability determination patients. Am J Public Health 1985;75:78-79. [n18.] Schulberg HC, Saul M, McClelland M, Ganguli M, Christy W, Frank R: Assessing depression in primary medical and psychiatric practices. Arch Gen Psychiatry 1985;42:1164-1170. [n19.] Popkin MK, Callies AL, Colon E: A framework for the study of medical depression. Psychosomatics 1987;28:27-33. [n20.] Keller MB, Klerman GL, Lavori PW, Fawcett JA, Coryell W, Endicott J: Treatment received by depressed patients. JAMA 1982;248:1848-1855. [n21.] Lustman PJ, Harper GW: Nonpsychiatric physicians' identification and treatment of depression in patients with diabetes. Compr Psychiatry 1987;28:22-27. [n22.] Normand PS, Jenike MA: Lowered insulin requirements after ECT. Psychosomatics 1984;25:418-419. [n23.] Kronfol Z, Greden J, Farroll B: Psychiatric aspects of diabetes mellitus: Diabetes and depression. Br J Psychiatry 1981;139:172. [n24.] Alex M, Baron EK, Goldenberg S, Blumenthal HT: An autopsy study of cerebrovascular accident in diabetes mellitus. Circulation 1962;25:663-673. [n25.] Kannell WB, McGee DL: Diabetes and cardiovascular disease: The Framingham Study. JAMA 1979;241:2035-2038. [n26.] Roth M: The natural history of mental disorder in old age. J Ment Sci 1955;101:281-301. [n27.] Post F: Significance of Affective Symptoms in Old Age: A Follow-up Study of 100 Patients, Maudsley Monograph No. 10. New York, Oxford University Press Inc, 1962. [n28.] Clayton PJ, Lewis CE: The significance of secondary depression. J Affective Disord 1981;3:25-35. [n29.] Mavissakalian M, Barlow DH: Phobia: Psychological and Pharmacological Treatment. New York, Guilford Press, 1981, pp. 1-34. [n30.] Sheehan DV, Sheehan KH: The classification of anxiety and hysterical states: II. Toward a more heuristic classification. J Clin Psychopharmacol 1982;2:386-393. [n31.] Mackenzie TB, Popkin MK: Organic anxiety syndrome. Am J Psychiatry 1983;140:342-344. [n32.] Auer RN: Progress review: Hypoglycemic brain damage. Stroke 1986;17:699-708. [n33.] Pinart J: Diabetes mellitus and its complications: A prospective study of 4400 patients observed between 1947 and 1973. Diabetes Care 1978;1:168-188.