Home

German Journal of Psychiatry

ISSN 1433-1055

Psychiatric Morbidity and Metabolic Control in Diabetic Patients

 

Jadala Mousleh2, Apostolos Iacovides1, John Yovos2, Konstantinos N. Fountoulakis1, Kyriakos Kazakos2, Magdalini Karagianni1, Anestis Vlachogiannis2, George Kaprinis1, Charalambos Ierodiakonou1

 

1Third Department of Psychiatry

2Diabetes Center of Aristotle University of Thessaloniki, AHEPA General Hospital, Thessaloniki, Greece

 

 

Corresponding address: Konstantinos N. Fountoulakis M.D. Ph.D., 53, Chrysostomou Smyrnis Street, 55132 Aretsou,  Thessaloniki, Greece, Tel: 30 31 435702, fax: 30 31 266570, e-mail: kfount@med.auth.gr

 

Abstract

 

Introduction: Psychiatric morbidity and especially depression are believed to be important aspects of diabetes mellitus (DM). The prevalence of depression in DM is reported to vary between 8.5% and 27%. The aim of the current study was to clarify the relationship between depression and DM. Material and Methods: Thirty-nine patients (20 males and 19 females) suffering from non-insulin dependent diabetes mellitus (NIDDM) entered the study. Their age was 54.41±15.7 years (range 41-70), Body Mass Index (BMI) values were 25.55±3.56 (range 17.91-32.46) and the duration of diabetes 8.58±7.56 years (range 1.5-33). Sixty-four percent of them were overweight and 68.57% were well controlled. Their glucosylated haemosphairine (HbA1c) was 7.7±1.72% (range 3.7-11%) and their glucose level was 204.31±88.55 mg% (range 81-454). The General Health Questionnaire (GHQ-28 whose A, B, C and D sub-scales, as well as total GHQ score, correlate highly with affective symptomatology and somatic manifestations of psychological origin) and Beck Depression Inventory (BDI) were used for psychometric evaluation. Forward Stepwise Multiple Linear Regression Analysis (FSMLRA) was used for the analysis of the data. The least squares method was used to draw the regression line in bivariate scatterplots, to search for non-linear relationships. Results: FSMLRA results showed that C dimension of GHQ correlates with HbA1c, D dimension of GHQ correlates with glucose levels and management of DM, and BDI correlates with HbA1c, management of DM and BMI. Bivariate scatterplots, however, showed a non-linear relationship between depression and DM. Depression, measured either by BDI or D dimension of GHQ seems to correlate weakly with management of DM when HbA1c values are under 10% or glucose levels under 250%. However, beyond these values, the increase in depression is impressive.

 

Key words: Psychiatric morbidity, depression, anxiety, diabetes mellitus, metabolic control.

Introduction

 

Depression is reported to occur frequently in patients with diabetes mellitus (DM). Few studies have been published on the epidemiology, brain chemistry and treatment of depression in diabetic patients.

In a review article[1] that searched MEDLINE for literature from January 1966 to July 1993 and cross-referenced the terms diabetes, glucose, hyperglycaemia, or hypoglycaemia, with antidepressants, 20 papers on epidemiology were found, 15 papers on neurochemicals and glucose control, and 28 papers on antidepressants and factors of importance to diabetics. These papers gave different estimates on the prevalence of depression in diabetics. Prevalence varied from 8.5% to 27.3%. The review article concluded that severity of depression correlates strongly with many symptoms of diabetes mellitus.

In another review article[2] which aimed to determine the prevalence of depression in adult diabetic populations through a comprehensive literature review and to critically evaluate the methods and findings of such studies from an epidemiological perspective, the systematic review of the scientific literature revealed a total of 20 studies, 14 of which had been conducted since 1988. Nine of the studies were controlled investigations, whereas the remaining 11 studies did not contain reference groups. The studies included both treatment and community samples. The range of the prevalence of current depression obtained from structured diagnostic interviews in diabetic samples was 8.5-27.3% (mean = 14.0%) in controlled studies and 11.0-19.9% (mean = 15.4%) in uncontrolled studies. These rates are at least three times the prevalence of major depressive disorder found in the general adult population of the U.S. The authors concluded that an increased prevalence of depression in diabetes relative to the general population is clearly suggested by the literature, although biases and methodological problems commonly encountered in prevalence studies may interfere with the strength of this conclusion. The increased prevalence of depression in diabetes relative to other somatic illnesses remained unproved.

The ‘quality’ of the depressive symptomatology seen in diabetic patients has been the focus of research. In a study published in 1992[3], diabetic and psychiatric outpatients were studied to determine whether the symptom profile of depression was similar in medically ill and medically well subjects. Seventeen (81%) of 21 depressive symptoms were not statistically different in prevalence or severity between the depressed diabetic patients and the depressed psychiatric patients. Both of these depressed groups were significantly different from a non-depressed diabetic reference group in the prevalence and severity of every depressive symptom except weight loss. The authors of this study concluded that the symptom profile of depression in diabetic patients (in particular the cognitive symptoms) is similar to that in depressed psychiatric patients and is readily differentiated from the symptom profile in non-depressed diabetic patients. They also suggest that their observations support the diagnostic validity of the DSM-IIIR criteria for major depression in this medically ill outpatient sample.

Several research articles on the relationship of depressive symptomatology to other aspects of diabetes have been published. The number of other chronic conditions and age may be significant independent predictors of depressive symptoms in all diabetic men and women. Reports suggest that depressive symptoms in individuals with Type 2 diabetes may be related to awareness of diabetic condition in addition to poor health[4]. The patients with diabetic complications were found to score significantly higher on depression inventories than patients without complications and control subjects[5]. In a study from the United Kingdom, the presence of depression was unrelated to sex, ethnic group, duration or type (IDDM or NIDDM) of diabetes or social class, but was significantly related to type of accommodation, marital status, and amount of social contact.

Depression in diabetic patients is related to other factors that have an adverse effect on the general medical condition of the patient, like smoking or alcohol abuse. Among smokers, the number of cigarettes smoked per day was found to be independently associated with cognitive symptoms of depression[6].

On the other hand, alcohol abuse among patients with diabetes mellitus is dangerous and complicates therapy, but its prevalence and the factors that predict it are unknown. The need for family physicians to intensify alcohol screening efforts in this population is great[7].

From the above, it is clear that although a lot of studies report a positive correlation between DM and depression, the existence of many confounding variables in the design of these studies makes conclusions problematical.

The aim of the current study was to examine the relationship of DM, its severity and various biochemical indexes related to DM, with depressive symptomatology.

Material and Methods

Thirty-nine patients (20 males and 19 females) suffering from non-insulin dependent diabetes mellitus (NIDDM) entered the study. All provided written informed consent before entering the study. They were consecutive patients of the outpatient clinic of the Diabetes Center of Aristotle University of Thessaloniki, AHEPA General Hospital, Thessaloniki, Greece.

Their age was 54.41±15.7 years (range 41-70), Body Mass Index (BMI) values were 25.55±3.56 (range 17.91-32.46) and the duration of illness 8.58±7.56 years (range 1.5-33). Sixty-four percent of them were overweight and 68.57% were well controlled. Their glucosylated haemosphairine (HbA1c) was 7.7±1.72% (range 3.7-11%) and their glucose level was 204.31±88.55 mg% (range 81-454). The General Health Questionnaire-28 (GHQ-28) and Beck Depression Inventory (BDI) were used for psychometric evaluation.

GHQ-28[8] is one of the most widely used psychometric instruments to screen for mild forms of psychopathology in the general population. It is a self-report instrument. It consists of 28 questions scored on a Likert scale. The combination of these 28 items creates four sub-scales, named A, B, C and D. All sub-scales, as well as the total GHQ score, correlate highly with affective symptomatology and somatic manifestations of psychological origin.

The Beck Depression Inventory (BDI)[9] is a self-report questionnaire that measures depression. It was developed according to the cognitive theory on depression of Aaron T. Beck.

Forward Stepwise Multiple Linear Regression Analysis (FSMLRA) was used for analysis of the data. The least squares method was used to draw the regression line in bivariate scatterplots, to search for non-linear relationships.

Results

Descriptive statistics for all subscales of the GHQ, the total GHQ score and for the BDI score are shown in Table 1. From FSMLRA results (Table 2) it seems that the C dimension of GHQ correlates with HbA1c (but very weakly, explaining only 15% of variance), D dimension of GHQ correlates with glucose levels and management of DM, total GHQ score correlates with glucose levels, and BDI correlates with HbA1c, management of DM and BMI. A and B dimensions of GHQ do not correlate with any variable concerning DM. Bivariate scatterplots, however, showed a non-linear relationship between depression and DM. These scatterplots clearly showed that depression, measured either by BDI or D dimension (this dimension correlates more closely with depression than with anxiety) of GHQ seems to correlate weakly with management of DM when HbA1c values are under 10% or glucose levels are under 250% (Fig. 1). However, beyond these values, the increase in depression is impressive.

In fact, these scatterplots suggest the possibility that two distinct groups of NIDDM patients exist. One with good metabolic control and low depression level and a second one with poor metabolic control and high depression. The transition from the first to the second group does not seem to be continuous, with just a few ‘noise’ cases in between. These results point to a non-linear relationship between depression and diabetes, which suggests that depressive symptomatology may be non-reactive in nature.

 

 

Table 1: Descriptive statistics of GHQ dimensions, GHQ total score and BDI score in diabetic patients

 

Mean

Minimum

Maximum

Std.Dev.

A

6.54

0

21

4.25

B

6.97

0

16

4.16

C

7.69

1

17

3.20

D

3.36

0

20

4.83

GHQ

22.49

3

57

10.15

BDI

9.16

0

35

7.31


 

 

Table 2: Summary of stepwise multiple linear regression analysis

 

B

p

Regression Summary for Dependent Variable: C subscale of the GHQ R˛= 0.15 p<0.018

Intercept

1.77

0.47 (N.S)

HBA1c

0.76

0.02

Regression Summary for Dependent Variable: D subscale of the GHQ: R˛= 0.38 p<0.002

Intercept

-8.80

0.04

Glucose

0.03

0.00

Metabolic Control

-5.73

0.04

HBA1c

1.18

0.12 (N.S)

Regression Summary for Dependent Variable: BDI R˛= 0.55 p<0.019

Intercept

-43.43

0.01

HBA1c

4.94

0.00

Metabolic Control

-12.53

0.03

BMI

1.19

0.02

Age

-0.24

0.08 (N.S)

Duration

0.27

0.33 (N.S)

 

GHQ: General Health Questionnaire

BDI: Beck Depression Inventory

BMI: Body Mass Index

HBA1c: Glucosylated haemosphairine

 

 

 

 Discussion

The current study addresses the question of the possible relationship between psychiatric morbidity and especially anxiety, depression and metabolic control in diabetic patients. The results suggest that anxiety and depression correlate weakly with metabolic control in these patients. However, they also suggest that there is a specific group of patients with poor metabolic control and simultaneously high levels of depression. This could mean that, in a selected group of patients, either depression leads to non-compliance behaviours that worsen metabolic control, or diabetes has an impact on the psychological well being of the patients due to restrictions in diet, need for discipline or due to complications. A third approach might be that there are common biological pathways that cause diabetes and depression in these patients. We will address these questions by structuring the discussion in sections.

 

Is there a correlation between psychiatric morbidity and diabetes mellitus?

There are many studies in the international literature addressing this issue. In a study of psychiatric morbidity and social problems in a consecutive series of out-patients with insulin-dependent diabetes mellitus, the prevalence of psychiatric morbidity was found to be equal to 18%, and consisted of depression, anxiety, and attendant symptoms[10]. Patients reporting major social problems had significantly higher levels of psychiatric symptoms. In another study of 180 adult diabetic patients, a prevalence rate of 60.5% was reported. In that study, the relationship between diabetic control and depressive symptoms was found to be weak[11]. The effect of gender is quite complex and controversial results have been reported[12]. Generally, depression in diabetes is considered to be a prevalent and chronic condition of unknown aetiology[13].

In a follow-up study[14] of 37 diabetic adults with major depression, 18 (64%) had experienced an episode of major depression within the previous 12 months and 12 of these patients satisfied diagnostic criteria for depression at the time of re-evaluation. The authors of that study concluded that the occurrence of depressive episodes appeared to be independent of diabetes complications because both the depressed and reference groups had similar rates of neuropathy, retinopathy, and nephropathy, and suggested that the natural course of depression in diabetes is malevolent, possibly more so than depression in the medically well. According to another study[15], diabetics suffer from a similar amount of depression to non-diabetics, but psychiatric symptoms may be related to the frequency of diabetic complications or to social problems[16].

It has been suggested that coping is a better predictor for metabolic control than emotional adaptation and life events[17]. Longitudinal studies[18] show that only depression decreased over the period studied, while coping and denial remained stable. The destructive effect of denial was indicated only by delayed requests for assistance, while no correlation could be shown for phase-specific internal restructuring of the psychological function of denial to compliance and metabolic control.

Conclusively, there are some data in the international literature supporting the possibility of increased psychiatric morbidity in diabetic patients. However, a variety of adjustment disorders can be detected and these disorders make the picture unclear.

The results of the current study support the existence of a subgroup of diabetic patients with a high level of depression and poor metabolic control, but not the theory of increased psychiatric morbidity in these patients.

Is there evidence supporting the theory that diabetes has an impact on the psychological well being of the patients due to restrictions in diet, need for discipline or due to complications?

Psychiatric morbidity was not associated with the presence of complications of diabetes[19]. Psychosocial factors have been studied as well[20] and results are inconclusive. Psychosocial differences may exist according to both the number and the type of diabetic complications present[21]. Because poorer quality of life and symptoms of depression may both result from complications, prospective follow-up is needed to clarify their temporal interrelationships, and to determine whether type A personality affords any protection against complications or is diminished as a result of developing complications. Relatives of patients suffer from psychological distress as well, and the family environment may suffer as a whole.

Wing et al, in 1991[22], reported that very-low-calorie diets, used in the context of a behavioural weight-control programme, resulted in reductions in hunger and improvement in mood state comparable to those observed on more moderate weight-loss regimens.

The comparison of standard intensive treatment and patient education with a specific distress reduction programme to standard intensive treatment and patient education without this programme, showed less anxious coping behaviour, less depression and less denial at the 9-month follow-up and less denial at the 15-month follow-up in the group with the specific distress reduction programme, but no differences in metabolic control between the two groups at any time[23]. According to Connell et al[24], diabetes-specific social support, self-efficacy, and outcome expectancies were not significant predictors of depression.

The findings of another study are impressive. It reported that, in a limited group of patients, a multiple injection regimen by pen treatment did not lead to an improved metabolic control. However, subjective psychological tests showed that some aspects of well-being tended to improve[25].

Conclusion:

Dietary and other restrictions in lifestyle that therapy for diabetes demands, do not cause depression, nor any severe form of distress by themselves.

Aggressive treatment improves the emotional status of the diabetic patient, and this improvement is unrelated to the true therapeutic effect of the treatment on metabolic control

The specific treatment and specific management of diabetes does not have an effect on depression. A large effect on emotional state comes from the patients' subjective feeling about their environmental support and treatment.

The results of the current study suggest a non-linear relationship between depression and diabetes. This may mean that depressive symptomatology is not reactive in nature. The findings of the international literature may be true concerning the majority of diabetics but the current study poses questions regarding the relevance of the above conclusions in depressed diabetic patients.

Does depression lead to non-compliance behaviours that worsen metabolic factors?

There are several studies in the international literature that address this issue, often in an inconclusive way. Weekly group psychotherapy focused on topics chosen spontaneously by the patients revealed that most often the topic for discussion included complications and fears regarding complications, diet, and relationships with physicians, and improvement with compliance was reported[26]. The results of another study[27], showed that both anxiety and depression have a weak positive correlation with blood sugar.

It has been reported[28] that a history of Major Depression among type II diabetic patients seeking obesity treatment was not related to pre-treatment glycolic control, but was associated with higher rates of attrition from treatment. Diabetes education programmes and support groups can have a major impact on the psychosocial condition of the patients and can have long term benefits in knowledge, psychosocial functioning, and glycolic control for older diabetic patients[29],[30].

Treatment adherence in adolescents with insulin-dependent diabetes mellitus is reported to be associated with behavioural and psychological variables[31].

On the other hand, it seems that depression may be the real source of many reported symptoms often attributed to diabetes[32]. Diabetes symptoms may be unreliable indicators of poor metabolic control when features suggestive of depression are present.

Patients with poor metabolic control reported more positive and neutral life events during the past year, suggesting that even those life changes which individuals view benignly may be associated with metabolic control difficulties. These patients also reported more recent symptoms of depression, anxiety and hostility than did patients in moderate or very good control-symptomatology which may further impair their ability to adhere to a complex self-care regimen. Patients with long disease duration reported more positive and negative recent life experiences than did subjects with short disease duration, but did not provide evidence of concomitant disruptions in metabolic control. The role that experience with a chronic disease may play in this finding was unclear, but these findings suggest that clinically relevant subgroup parameters, subjects' perceptions of life change, and demographic variables may be important factors to assess[33].

Conclusion: Depression and diabetes control are correlated, but the impact of this correlation and the role of the central nervous system chemistry, personality factors and life events remain to be established further.

The results of the current study, as noted above, suggest a non-linear relationship between depression and diabetes. This may mean that depression is a direct syndrome in the context of diabetic mellitus disease. Also, reversibly, depression may affect glycaemic control both through neuroendocrinological pathways and behavioural disorders, leading to a vicious circle.

Is there any evidence supporting the existence of common biological pathways that cause both diabetes and depression?

Genetic studies have attempted to link (bipolar) affective disorder to the short arm of chromosome 11 (where the loci for insulin, insulin growth factor (IGF), tyrosine hydroxylase (TH) and h-ras oncogene are located), but they have failed to do so[34]. Since TH and the insulin receptor require phosphorylation by protein kinases, a defect of the h-ras oncogene or its products (p21) could disturb both these systems and compromise catecholaminergic transmission in neurones and energy flow in glial cells and lead not only to a predisposition to depression ('trait markers') but to neurotoxic damage.

During a stress interview, catecholamines and plasma cortisol levels showed no significant increase, whereas there was a significant increase of GH over time with a trend in diabetics to have a more marked GH response than controls[35]. Blood glucose levels remained unaffected by the interview. Depressed diabetics showed significantly higher cortisol increases than non-depressed diabetics, whereas there was no difference among depressed and non-depressed controls. Depression was not associated with an increase of other hormones or blood glucose levels in either group. The authors concluded that their results confirm specific pathways in which individual emotional arousal and depression may lead to chronic metabolic disturbances as a result of GH and cortisol hypersecretion.

Similar studies reported that protracted, war-induced stress in persons with Type II diabetes mellitus need not worsen glycaemic control in Type II diabetic patients[36].

The study of muscarinic cholinergic receptors[37] and their contribution in the anomalous GH response to GnRH in men with insulin-dependent diabetes mellitus and in male patients with major depression, indicated that a muscarinic cholinergic mechanism is involved in the anomalous responses of GH to GnRH and TRH in diabetic men and in male patients affected by major depression.

There is evidence that in all cases there is an increased activity of the regulatory hormones in the diabetic patients, pointing to an increased glucose utilisation in these patients. This glucose consumption may be stress-related or reflects an increase in autonomic activity in these patients.

However, the most impressive finding comes from the study of Baekkeskov et al[38] who reported the presence of antibodies to glutamic acid decarboxylase (GAD) in almost all patients with insulin-dependent diabetes mellitus. The presence of these antibodies precedes the onset of diabetes. GAD is the enzyme involved in the synthesis of γ- aminobutyric acid (GABA) from glutamate (Glu). It is known that GABA and Glu manifest opposite functions in the central nervous system. Glu is an excitatory amino acid, while GABA is an inhibitory one. Glu hyperfunction leads to neuronal death and this mechanism is considered responsible for the development of depression and schizophrenia. GAD isoenzymes are present in β cells of the pancreatic isles where GABA is suspected to play a role in pancreatic endocrine function. Autoantibodies to GAD may therefore underlie the development of insulin-dependent diabetes as well as that of a mental disorder.

Conclusion: There are some data in the international literature supporting the possibility of the existence of common biological pathways in the aetiopathogenesis of diabetes and depression.

The results of the current study provide no data on this issue.

Conclusion

The results of the current study support the existence of a subgroup of diabetic patients with high levels of depression and poor metabolic control, and are, in general, in accord with the international literature.

These results point to a non-linear relationship between depression and diabetes. This may mean that depressive symptomatology is not reactive in nature. If this is true, it may mean that depression, at least in a large percentage of depressive diabetics, is a direct syndrome in the context of diabetic mellitus disease. Also, since neurotransmitter systems may be severely affected, depression may affect glycaemic control both through neuroendocrinological pathways and behavioural disorders in a true vicious circle.

 

 

References

 



[1] Goodnick-PJ; Henry-JH; Buki-VM: Treatment of depression in patients with diabetes mellitus. J-Clin-Psychiatry. 1995; 56(4): 128-36

[2] Gavard-JA; Lustman-PJ; Clouse-RE: Prevalence of depression in adults with diabetes. An epidemiological evaluation.  Diabetes-Care. 1993; 16(8): 1167-78

[3] Lustman-PJ; Freedland-KE; Carney-RM; Hong-BA; Clouse-RE:  Similarity of depression in diabetic and psychiatric patients. Psychosom-Med. 1992; 54(5): 602-11

[4] Palinkas-LA; Barrett-Connor-E; Wingard-DL: Type 2 diabetes and depressive symptoms in older adults: a population-based study.  Diabet-Med. 1991 Jul; 8(6): 532-9

[5] Leedom-L; Meehan-WP; Procci-W; Zeidler-A: Symptoms of depression in patients with type II diabetes mellitus. Psychosomatics. 1991; 32(3): 280-6

[6] Haire-Joshu-D; Heady-S; Thomas-L; Schechtman-K; Fisher-EB Jr: Depressive symptomatology and smoking among persons with diabetes. Res-Nurs-Health. 1994 ; 17(4): 273-82

[7] Spangler-JG; Konen-JC; McGann-KP:  Prevalence and predictors of problem drinking among primary care diabetic patients. J-Fam-Pract. 1993 ; 37(4): 370-5

[8] Goldberg D, Williams P: A User’s Guide to the General Health Qustionnaire. Psychol Med, 1979;9:139-145

[9]Beck AT, Ward CH et al: An Inventory for Measurintg Depression. Arch Gen Psychiatry, 1961;4:561-571

[10] Wilkinson G; Borsey DQ; Leslie P; Newton RW; Lind C; Ballinger CB:  Psychiatric morbidity and social problems in patients with insulin-dependent diabetes mellitus. Br J Psychiatry. 1988; 153: 38-43

[11] Padgett-DK:  Sociodemographic and disease-related correlates of depressive morbidity among diabetic patients in Zagreb, Croatia. J-Nerv-Ment-Dis. 1993 Feb; 181(2): 123-9

[12] Popkin-MK; Callies-AL; Lentz-RD; Colon-EA; Sutherland-DE: Prevalence of major depression, simple phobia, and other psychiatric disorders in patients with long-standing type I diabetes mellitus. Arch-Gen-Psychiatry. 1988; 45(1): 64-8

[13] Lustman-PJ; Griffith-LS; Gavard-JA; Clouse-RE Depression in adults with diabetes. Diabetes-Care. 1992; 15(11): 1631-9

[14] Lustman-PJ; Griffith-LS; Clouse-RE: Depression in adults with diabetes. Results of 5-yr follow-up study. Diabetes-Care. 1988; 11(8): 605-12

[15] Robinson-N; Fuller-JH; Edmeades-SP:  Depression and diabetes. Diabet-Med. 1988; 5(3): 268-74

[16] Roy-M; Collier-B; Roy-A: Excess of depressive symptoms and life events among diabetics. Compr-Psychiatry. 1994 ; 35(2): 129-31

[17] Spiess-K; Sachs-G; Moser-G; Pietschmann-P; Schernthaner-G; Prager-R: Psychological moderator variables and metabolic control in recent onset type 1 diabetic patients--a two year longitudinal study. J-Psychosom-Res. 1994 ; 38(3): 249-58

[18] Spiess-K; Sachs-G; Frischenschlager-O; Moser-G; Prager-R: Zur phasenspezifischen Funktion der Verleugnung beim Typ 1 Diabetes Patienten nach Krankheitsausbruch. Z-Psychosom-Med-Psychoanal. 1994; 40(1): 52-67

[19] Wilkinson-G; Borsey-DQ; Leslie-P; Newton-RW; Lind-C; Ballinger-CB: Psychiatric morbidity and social problems in patients with insulin-dependent diabetes mellitus. Br-J-Psychiatry. 1988; 153: 38-43

[20] Mayou-R; Peveler-R; Davies-B; Mann-J; Fairburn-C:  Psychiatric morbidity in young adults with insulin-dependent diabetes mellitus. Psychol-Med. 1991; 21(3): 639-45

[21] Lloyd-CE; Matthews-KA; Wing-RR; Orchard-TJ: Psychosocial factors and complications of IDDM. The Pittsburgh Epidemiology of Diabetes Complications Study. VIII. Diabetes-Care. 1992 Feb; 15(2): 166-72

[22] Wing-RR; Marcus-MD; Blair-EH; Burton-LR: Psychological responses of obese type II diabetic subjects to very-low-calorie diet. Diabetes-Care. 1991; 14(7): 596-9

[23] Spiess-K; Sachs-G; Pietschmann-P; Prager-R: A program to reduce onset distress in unselected type I diabetic patients: effects on psychological variables and metabolic control. Eur-J-Endocrinol. 1995; 132(5): 580-6

[24] Connell-CM; Davis-WK; Gallant-MP; Sharpe-PA: Impact of social support, social cognitive variables, and perceived threat on depression among adults with diabetes. Health-Psychol. 1994 ; 13(3): 263-73

[25] Buysschaert-M; Janne-P; Mpoy-M; Reynaert-C; Pirson-F; Cassiers-L; Lambert-AE:  Metabolic and psychological evolution of insulin dependent diabetic patients during a 6 months insulin-pen treatment. Diabete-Metab. 1988;14(2): 75-9

[26] Toth-EL; James-I: Description of a diabetes support group: lessons for diabetes caregivers. Diabet-Med. 1992; 9(8): 773-8

[27] Eaton-WW; Mengel-M; Mengel-L; Larson-D; Campbell-R; Montague-RB: Psychosocial and psychopathologic influences on management and control of insulin-dependent diabetes.  Int-J-Psychiatry-Med. 1992; 22(2): 105-17

[28] Marcus-MD; Wing-RR; Guare-J; Blair-EH; Jawad-A:  Lifetime prevalence of major depression and its effect on treatment outcome in obese type II diabetic patients.  Diabetes-Care. 1992; 15(2): 253-5

[29] Gilden-JL; Hendryx-MS; Clar-S; Casia-C; Singh-SP: Diabetes support groups improve health care of older diabetic patients. J-Am-Geriatr-Soc. 1992; 40(2): 147-50

[30] Rubin-RR; Peyrot-M; Saudek-CD: Effect of diabetes education on self-care, metabolic control, and emotional well-being Diabetes-Care. 1989 Nov-Dec; 12(10): 673-9

[31] Littlefield-CH; Craven-JL; Rodin-GM; Daneman-D; Murray-MA; Rydall-AC: Relationship of self-efficacy and binging to adherence to diabetes regimen among adolescents. Diabetes-Care. 1992; 15(1): 90-4

[32] Lustman-PJ; Clouse-RE; Carney-RM: Depression and the reporting of diabetes symptoms. Int-J-Psychiatry-Med. 1988; 18(4): 295-303

[33] Karlsson-JA; Holmes-CS; Lang-R:  Psychosocial aspects of disease duration and control in young adults with type I diabetes.  J-Clin-Epidemiol. 1988; 41(5): 435-40

[34] Newman-JC; Holden-RJ:  The 'cerebral diabetes' paradigm for unipolar depression. Med-Hypotheses. 1993 ; 41(5): 391-408

[35] Sachs-G; Spiess-K; Moser-G; Kautzky-A; Luger-A; Pietschmann-P; Schernthaner-GS; Prager-R:  Hormonal and blood glucose responsiveness as an indicator of specific emotional arousal in type 1 diabetics. J-Psychosom-Res. 1993 ; 37(8): 831-41

[36] Roglic-G; Pibernik-Okanovic-M; Prasek-M; Metelko-Z:  Effect of war-induced prolonged stress on cortisol of persons with type II diabetes mellitus. Behav-Med. 1993; 19(2): 53-9

[37] Coiro-V; Volpi-R; Capretti-L; Speroni-G; Castelli-A; Mosti-A; Marchesi-C; Gardini-E; Rossi-G; Chiodera-P: Simultaneous inhibition by pirenzepine of the GH responses to GnRH and TRH in insulin-dependent diabetics and in patients with major depression. Acta-Endocrinol-Copenh. 1989; 120(2): 143-8

[38] Baekkeskov S, Landin M, Kristensen J, et al: Antibodies to a 64,000 Mr Human Islet Cell Antigen Precede the Clinical Oncet of Insulin Dependent Diabetes. J Clin Invest 1987;79:926-934