Nine different EDS types that vary widely in prevalence have been defined by collagen biochemistry and structure, family history, and clinical presentation. EDS Type III (benign familial hypermobility) is the most common and is manifest by large and small joint mobility and joint dislocation. Joint hypermobility syndrome (JHS) is relatively common and clinically similar to EDS Type III; previously it was diagnosed as EDS Type XI. EDS Type I and the clinically similar but less severe EDS Type II, present with joint mobility and very soft, hyperextensible skin which is easily bruised, lacerated, and scarred. Rarer types include EDS Type IV, manifesting in weak arterial and hollow organ walls which may rupture and cause death, and the autosomal recessive EDS Type VI which includes ocular fragility in addition to soft hyperextensible skin.

The disease manifests in early childhood and is chronic. Ainsworth and Aulicino [1993] surveyed EDS patients, and found that pain and physical disability were common, as were obstetric and gynecological problems (miscarriage, premature membrane rupture). Given the chronic, painful, potentially disabling nature of EDS, one might expect some psychosocial impairment; no studies of psychological functioning in EDS have been conducted. The goal of this study was to examine psychosocially (via psychological testing and interviews) a large group of EDS patients to determine how having EDS affects them, to compare psychosocial functioning among different EDS subtypes, and to make recommendations to treating clinicians.

METHODS

Adults and children with EDS came to Detroit, Michigan in August 1992, to participate in the first multidisciplinary research clinic on EDS. The psychiatry research portion involved both psychological testing (Part 1) and interviews/discussions (Part II). Psychological testing was completed by 41 adults. Table I presents their sociodemographics overall and by EDS type. Five patients with JHS were retained in this sample because JHS is similar clinically to EDS Type III, suggesting that these patients would experience similar psychosocial issues. The adults were predominantly relatively highly educated white women having EDS Types I, II, or III, or JHS. Seven children with EDS (6 girls, one boy; all white; ages 6 to 12 years; three EDS Type I, one EDS Type III, one EDS Type IV, two unknown) also were evaluated via psychological tests (Table 1).

For part I, all adults completed a demographic, psychiatric history, and substance abuse questionnaire. They also completed the Symptom Checklist-90-Revised [SCL-90-R; Derogatis, 1983), which assesses 9 domains of psychiatric symptomatology (somatization, obsessions/compulsions, interpersonal sensitivity, depression, anxiety, hostility, phobias, paranoid thinking, and psychotic thinking). Patients rated each symptom on a 4-point scale for intensity, and symptom domain means and the total number of symptoms were calculated. EDS patients were compared to the published norms from a large sample of healthy young adults and a sample of psychiatric outpatient adult. Patients also completed the West Haven-Yale Multidimensional Pain Inventory [WHYMPI; Kerns et al,, 1985], which assesses a patient's pain experience using 52 statements rated on 7-point scales. We used 2 subscales: Pain Severity and Pain Interference; higher scores indicate more severe pain and more interference with daily activities. EDS patients were compared with published norms from a sample of chronic pain patients seen at a pain clinic [Kerns et al., 1985]. For the 7 EDS children, a parent completed the Child Behavior Checklist [CBCL; Achenbach and Edelbrock, 1983] about the child's social competence (social and athletic activities, friends and acquaintances, and school performance) and behavior problems (113 behavior problems rated on a 3-point intensity scale). The children's scores were compared with sex and age-matched norms.

In part II, groups of about 10 adult patients met for 2 to 3 hours to discuss psychosocial difficulties associated with EDS. Children were seen separately in a discussion group.

RESULTS

Part 1. Psychological Testing
Over half of the adults (56%) had been in outpatient psychotherapy some time during their lives, typically for depression/anxiety and relationship difficulties; 22 (54%) reported a history of at least one episode of significant depression, 19 (46%) had used either antidepressant or anxiolytic medication. Three adults had been hospitalized for psychiatric difficulties; 2 of whom had attempted suicide. Five patients reported a history of alcohol or illicit drug dependence and treatment. In all, 29 adults (71%) had received some psychiatric service (medication or therapy) at some
point in their lives.

Current psychiatric functioning was assessed with the SCL-90-R. For all psychiatric symptom domains. EDS patients scored approximately I standard deviation (SD) higher than healthy young adults, and 22% to 32% of the patients reported symptoms sufficiently severe to be scored in the "clinical range" (1.5 Sds above normative levels). Although all 9 symptom domains were increased among EDS patients, problems with anxiety, depression, and interpersonal sensitivity were dominant. Such things as phobias and excessive suspicion were not as problematic. When the EDS patients were compared with the psychiatric outpatient norms, EDS patients averaged about 0.5 SD lower on symptoms, suggesting that the sample experienced somewhat fewer or less intense symptoms than current psychiatric patients.

On the WHYMPI, the mean pain severity rating for the sample was 2.9 i 1.4 on the 0 to 6 scale, which is about 1.5 SDs lower than that of a sample of chronic pain patients seeking treatment, but a significant degree of pain, nonetheless. Pain at the midpoint or higher of the scale was reported by 63% of the sample; only 2 patients reported no pain. Pain was reported to interfere with daily living activities at 3.0 11.6 on the 0 to 6 scale, approximately I SD below the chronic pain sample mean; 50% of the EDS sample endorsed the midpoint or higher of the pain interference scale.

Table II presents psychiatric symptom and pain data by each EDS type for which there were at least 5 patients. To yield sufficient sample sizes for statistical analysis, types that have a similar clinical presentation were combined, because it was expected that the psychosocial effects would be similar for EDS Types I and II (combined n=15) and for EDS Type III and JHS (combined n=21). EDS Types IV and VI were too infrequent in this sample to include in analysis. Table II indicates that EDS Type 1/11 patients reported less symptomatology and less pain and pain interference than did the EDS Type III/JHS sample. These differences neared or reached significance for depression, anxiety, total number of psychiatric symptoms, and pain severity and interference. Examination of separate types suggests that the combined differences are attributable often to the relatively low levels of symptoms and pain for EDS
Type I (Table II).

Of the 7 EDS children, 4 had impaired social competence on the CBCL; the other 3 were within normal limits. Five children (including the 4 with social problems) were found to have clinically elevated behavior problems: 3 showed increased "internalizing" problems such as anxiety/depression, somatic complaints, and withdrawal; one showed excessive aggressive behavior, and a fifth had a range of behavior problems. Two children were normal psychiatrically.

TABLE I
Demographic Information by Type for Adults with EDS

(a) Type not reported for one patient
(b) Marital status not reported for one Type II patient

TABLE II
Comparison of EDS Type 1, 11, III, and JHS Patients on Psychiatric Symptoms and Pain

(a) Values presented are mean + standard deviation based on a 5 point scale (O = not at all 4 = extremely) or up to 90 total symptoms for SCL-90-R, or a 6-point scale (O = none; 6 = extreme) for WHYMPI
(b) t-test comparing the combined Types I and 11 with combined Types III and JHS, based on 34 degrees of freedom and using a 2-tailed P-value
(c) Obsess/Compul = Obsessions/Compulsions, Interper Sens = Interpersonal Sensitivity, Psychotic = Psychotic Thinking

PART II:
CLINICAL INFORMATION FROM DISCUSSIONS

Patients with EDS shared numerous concerns that we have grouped into 3 areas: difficulties with the medical system, problems in daily living, and concerns about pregnancy and reproduction. We present concerns of both adults and children, and we generalize across EDS types, except where noted. Although we do not present the number of patients with each concern, our statements reflect common rather than idiosyncratic issues.

Many patients consider EDS an "orphan" disease. Reportedly few physicians knew about it, especially primary care physicians, and referrals to a range of specialists were common prior to accurate diagnosis. Indeed, most patients were undiagnosed for many years, and many were accused of malingering, carelessness, or hypochondriasis. The lack of obvious disease and the relatively high functioning of many patients led some physicians to suggest that patients simply should quit worrying. Suggestions of psychiatric need were common. The lack of a medical diagnosis and their physicians' suspicions led many patients to have self-doubts and excessive frustration. Many patients suspected that their physicians were frustrated, embarrassed over their lack of knowledge about EDS, and defensive when the patients attempted to educate them about EDS.

In daily activities, most patients were cautious and some avoided many activities to prevent injury especially the relatively rare EDS Type IV patients who feared excessive exertion might lead to arterial rupture and death. In general, injury fears led many patients to be less socially active than they desired, ,especially during youth. School was difficult; some patients were considered "freaks" because of their mobility or hyperelasticity. Although a few patients turned this to their advantage by becoming a "class clown," others became depressed or angry. Parents of children with EDS often struggled with how much to protect their children's health versus teach them to engage fully in life without fearing their disease.

Sexual activity, pregnancy, and reproductive decisions were concerns for many patients. Intercourse was often unpleasant or difficult because of pain, fears of vaginal wall laceration, and the need to avoid certain sexual positions and reduce movements to avoid hip dislocation. Some women avoided intercourse because they feared pregnancy-induced rupture of the uterus. Finally, because EDS is inherited, some patients avoided having children, and others minimized risk by limiting the number of children. Some parents who had EDS children struggled with guilt.

DISCUSSION
Our findings from interviews and psychological testing of adults and children with EDS agree: EDS presents numerous psychosocial challenges, and many patients J~re affected greatly. In this study, patients suffered from mild to moderate levels of anxiety, worry/obsessionality, depression, anger, and interpersonal discomfort/avoidance; up to one-third had clinical levels of these emotional problems. Over two-thirds had had psychiatric services (medications or outpatient therapy).

Many of these psychosocial reactions are common to chronic illnesses generally (Moos, 1977), but EDS presents unique difficulties. Most EDS patients regularly experience pain, and although the pain is not as severe as that reported by patients with chronic pain problems, it significantly interferes with daily functioning. Further, having a chronic disease often is stigmatizing, and EDS, with its rare and unusual manifestations, elicits both casual curiosity and insensitive teasing from others, especially during childhood. Social and interpersonal difficulties occur when the potential for easy skin injury and dislocation or potentially fatal arterial rupture prompts patients to avoid or reduce activities, subsequently leading to depression and anger. Concerns about child-bearing and transmission of EDS and how best to raise children also leads to anxiety. Adding to patients' emotional problems is a medical system that has been unaware of this condition and too quick to attribute either the symptoms or the patient's concerns about them to psychological causes such as hypochondriasis. It should be noted that our psychiatric observations do not suggest any clinical association with autism, as has been suggested [Sieg, 1992; Tantam et al., 1990].

Different EDS types may have different psychosocial consequences. EDS Type I, with prominent skin hyperelasticity and easy bruising but relatively mild joint laxity, was associated with fewer psychosocial problems than EDS Type III and the clinically similar JHS (previously EDS Type XI), which are marked by severe joint hypermobility and dislocation but fewer skin problems. The severe joint problems in EDS Type III and JHS appear to lead to chronic, often debilitating pain, which may explain the increased psychosocial disturbances reported by these patients.

Psychiatric utilization was common in our sample, which appears appropriate, given their increased psychosocial symptomatology; yet many patients were disturbed by the referral process and reactions of mental health professionals. The primary clinician should assess mental status and daily coping to determine whether the patient needs a mental health referral. If so, the recommendation should be handled with care, to avoid the implication that the patient has a primary psychological rather than organic problem; in EDS, psychological problems typically are secondary. Rather, the physician should emphasize that the patient may need help coping with a very difficult disease, and the physician should make his/her own follow-up appointment regardless of the patient's decision whether or not to pursue psychiatric help. Mental health professionals are advised to bolster coping with pain and chronic illness and to alleviate anxiety/depression directly rather than address suspected primary psychological disturbances/conflicts.

The psychosocial difficulties experienced by EDS patients are very similar to those eloquently described by Cole [1993a] regarding osteogenesis imperfecta (OI), another group of defective collagen disorders. Observable physical abnormalities are more common in OI, however, and pain more common in EDS, potentially leading to differences in adaptation related to social versus physical stress. Yet both of these chronic conditions require patients to cope with demands that change over their lifespans, and professional attention to patient's psychosocial needs is a key aspect of their treatment

X ACKNOWLEDGEMENTS X
The authors thank Melanie Beutela and Birgit Katzei for assistance in data coding. The work was supported in part by Hutzel Hospital, the Ehlers-Danlos National Foundation, and the Coles Family Foundation.

Article References available upon request.