Findings of the present study highlight the heterogeneity of pain patients treated at our psychosomatic in-patient unit. As the descriptive analysis shows, one third (28.5%) were given a primary diagnosis of somatic disorder. Primary diagnosis was equally distributed across both younger and older patient groups. It could be argued that for no other disorders the bio-psycho-social diagnostic model and the acknowledgement of the complexity involved in the physician-patient relationship is of such importance, which might explain the diversity of diagnoses given to patients suffering with pain [2, 3]. It is precisely in order to emphasise the phenomenon pain that we kept the undifferentiated term ”pain patients” despite the question raised as to whether or not it is appropriate to speak of such a group .
Chronic pain patients signify a considerable problem in the health care system [1, 12].
Our results have shown that the majority of these patients were women between 16 and 79 years of age. These results confirm findings from other studies, in that patients showed a high degree of somatisation, reported a considerable amount of physical as well as psychopathological symptoms, and scored high on depressive mood, anxiety and emotional liability. Nickel et al., 2010 , for example, reported that 75% of their sample (282 psychosomatic pain inpatients) were female and that 69% of those reported a co-morbid psychological disorder. Furthermore, they found that the intensity of the somatisation as well as reported lower quality of life increased with age. Here, we were, however, unable to confirm these findings as the present results were found to be statistically non-significant, but went in the same direction (somatisation scale, SCL-90). An interesting finding from our study is that younger patients were found to be single, in employment, and reported having a better standard of knowledge despite a lower number of graduations than older patients. 7% of the younger patients compared to 40.1% of the older patients in our sample dropped out. The previously mentioned Mainz-Study reported that only16.1% of their participants dropped out.
When we looked at the other dimensions of the age sub-groups, only marginal differences were found with regards to pain experience, psychological symptoms, and disease management.
Differences between the two age groups were found with regards to reported levels of aggression, the mean duration of stay in hospital, and the mean consumption of analgesic medication. The significance of aggression in pain patients was also emphasised in studies about fibromyalgia patients .
A newly-developed integrative, psychosomatic treatment model was piloted for a group of chronic pain patients that are routinely referred to our university clinic for in-patient treatment. 179 patients received the new treatment and preliminary results as to its effectiveness are outlined by a descriptive analysis of the patient group. Two-third of the patients with a primary diagnosis of somatoform pain disorder benefited from our treatment, as shown in a reduction of symptoms of depression and anxiety.
Opposed to our expectations, both age groups benefited from treatment. No significant differences were found, which was surprising  given that previous findings report older patients to benefit less from psychotherapeutic treatments [5, 15]. Neither diagnosis nor gender influenced treatment efficacy in terms of depressive symptom reduction. But, anxiety symptom reduction was influenced by age group as well as gender. The symptom severity of anxiety at the beginning of treatment was also found to have an effect. The 10.8% detection of variance is not so high, but the model was significant in demonstrating that these factors are important in the in-patient treatment of chronic somatoform pain patients. In line with previous findings , the higher symptom severity, the better they benefited from treatment.
An additional note concerning psychodynamic aspects of pain treatment: The role of attachment was only integrated in the kind of treatment of pain patients. In this study, we could not demonstrate data for this important aspect [17, 18]: It is often found that the physician-patient relationship reflects the early mother-infant attachment of the patient. According to the Bowlby (1969)  theory, children are born with a deep-seated, instinctive need for attachment to a primary care-giver. Attachment according to his theory means safety, warmth, availability and the provision of nourishment on the one hand, and appropriate stimulation for an exploration of the external environment without fear of losing the crucial bond on the other. Thus, an internal attachment organisation develops gradually in relation to the primary attachment figure. However, under certain circumstances, the failure of a secure attachment organisation may lead to an internal belief that reliance on a person for attachment is futile. In one where pseudo-autonomy is assumed, the individual tries to maintain an insurmountable distance. These infantile internal survival strategies are often reflected in the patient-physician relationship . Patients with an internal secure attachment organisation are generally able to describe their symptoms, and any affective component usually corresponds to its content (21). Both, as said above, provide good conditions for an accurate diagnosis and subsequent treatment approach. Patients with an insecure attachment organisation, on the other hand, have difficulties in relating to their physician. Descriptions and associated affects often do not correspond, thus leading to an inaccurate verbal as well as non-verbal account of the experienced pain (21). In our study, we did not find that younger pain patients were different compared to older ones according attachment styles. In both groups the physical pain of these individuals becomes the sole focus and means of communication. Given our knowledge of these highly complex processes, within the therapy we tried to facilitate a trusting and emotionally corrective physician-patient relationship, which offered care and security, and thus enabled the patient to become aware of the underlying conflicts and finally allow the emotional discharge of formally only verbally expressed affects.
Limitations of the study: The described therapy model is based on the long-standing experience of individual physicians working at our clinic, all of whom differ in their approach to diagnosing and treating pain patients. Within the scope of this preliminary evaluation, we were able to describe in detail only clinical patients. This does not allow generalization of the results to all pain patients. However, the diagnosis in accordance with the ICD-10 diagnostic criteria was carried out by an interdisciplinary workgroup. They consisted of psychosomatic clinicians, specialised pain therapists, internists, and neurologists. Given the large variability within the somatic disorders in general, we have only focused on ICD 10, F 45.4 as the main diagnosis. The heterogeneity with regards to somatic disorders as primary and psychiatric disorders as a secondary diagnosis in relation to gender and age effects  were thus considered only broadly. Given that we are reporting the results of our routine evaluation, we lack the inclusion of pain-specific psychological as well as other psychosomatic outcome measures. A further limitation is the lack of a comparison or control group. Finally it can be questioned whether the age groups in this study provide the whole picture of age effects.
Inspite of these limitations we were able to describe sociological, clinical and psychological differences between age groups in our examined pain patients. Additionally we could evaluate effects of the in-treatment in both groups, which were not different. This clinical evaluation study has shown there are differences between age groups, gender and somatics compared to somatoform pain diagnoses groups in chronic pain patients. This will need to be evaluated in the next few years to find out the best treatment strategies for “non responder patients”. In these cases we need a common view on the subjective experience and the individual disease management of patients. Pivotal is the exploration of the patient’s experience of pain. In order to gain a better understanding of the experience of pain patients, we have found it important not only to pay clinical attention to the individual but also to differentiate between different subgroups. This view, in conjunction with the basic understanding of pain management, is needed when dealing with with each patient as an individual suffering from pain.