Introduction 1 2 3 4 5 6 7 8 9 10 12 13 14 15 17 Methods GPs and patients 1 Between February 2002 and December 2003, participating GPs included adult patients, presenting with a complaint that the GP designated as “unexplained” at the end of the first consultation. Patients who had presented with the same complaint in the previous 6 months were not included. Patients were eligible if they presented with fatigue, abdominal complaints, musculoskeletal complaints, weight changes or itch as their main complaint. Individual GPs decided whether a complaint was “unexplained”. There was no standardization of this process other than that implied by the Dutch College of General Practitioners’ definition. GPs invited patients to participate in the study at the end of the consultation. Informed consent procedure and medical ethics committees All participating patients gave written informed consent after having read information provided by the GP directly after the consultation. The medical ethics committees of the University of Amsterdam and Maastricht University approved the study protocol. Data collection: study population 18 19 20 Data collection: reference populations Reference population for comparison of demographic characteristics 21 Reference populations for comparison of quality of life The results section of this manuscript shows that the quality of life of patients with early stage unexplained complaints is poor. Our initial objective was purely descriptive. However, when such low quality of life figures were found, we felt the need to compare these to other patient groups. First, we wondered whether maybe all patients consulting their GP would have such low quality of life figures and therefore compared the unexplained complaints group with a Dutch general practice population. We also thought that maybe our patient population included a lot of depressed patients which could explain the poor quality of life and therefore also compared our quality of life figures with those of a population of depressed patients. 22 n 23 Statistical analysis First, we compared the demographic characteristics of the five unexplained complaint categories to each other. Those of the total UC study population were compared to those of a Dutch general practice population. Differences were tested using the Fisher’s exact test. 24 t Results General characteristics of the study group Of the 91 GPs who intended to participate, 18 dropped out before including any patients. They predominantly reported lack of time as the reason for drop out. Ten GPs did not include any patients during the inclusion period, although they did not formally drop out. Thus, 63 GPs (69%) included 513 patients with unexplained complaints (range: 1–36 patients per GP). Questionnaires from 466 (91%) patients were available for analysis. Forty-seven patients stopped their participation to the study or did not fill out this particular questionnaire. Demographic characteristics of the study group 1 Table 1 Demographic characteristics of the total UC study population, per complaint group and of a Dutch general practice population Demographic characteristic Categories n n n n n n  Sex Male 26.4 26.1 25 24.6 33.3 49.5 Female a 73.9 75 75.4 66.7 50.5 Age Mean (years) 44.2 41.8 42.3  b 50.3 Unknown 0–19 4.9 6.1 6.7 1.1 0 23.4 20–39 37.1 42 45 13 30.9 30.4 40–64 a 41.7 40 66.7 50 33 65+ 12 10.2 8.3 18.8 19.1 13.2 Educational level None 1 0 0  b 2.4 16 Elementary 9 7.5 3.3  b 14.3 18.9 Secondary a 68.1 80 b 64.3 48.1 Higher a 24.4 16.7 18.8 19 17 Marital state Single 28.5 29.8 20 30.4 28.6 Unknown Married/cohabiting 71.5 70.2 80 69.5 71.4 Unknown Insurance Type Public 65.2 65.8 68.3 69.6 50 67.4 Private 34.8 34.2 31.7 30.4 50 32.6 Nationality Dutch 95.3 96.3 96.7 91.3 93 86.2 Not Dutch 4.7 3.7 3.3 8.7 7 13.8 Region of the Netherlands West 56.4 57.6 46.7 62.3 52.4 Unknown South 43.6 42.4 53.3 37.7 47.6 Unknown a  P b  P The main reason for encounter was fatigue (63.3%) and only few patients with weight changes or itch were included. In all further analyses, these latter two groups were combined into an “other complaints” group. P P Quality of life of the total UC study population and per unexplained complaint group 2 Table 2 Quality of life of the total UC study population and per complaint group  All Fatigue Abdominal Musculoskeletal Other N 466 295 60 69 42 Domains RAND-36 Mean Mean Mean Δ (95% CI) Mean Δ (95% CI) Mean Δ (95% CI) Physical functioning 73.4 73.3 81.3 a 62.2 a 76.3 3.0 (−4.3 to 10.4) Social functioning 66.5 62.5 70.8 8.3 (−0.2 to 16.9) 76.4 a 72 9.5 (−0.7to 19.7) Role functioning physical 37.2 31.5 56.3 a 39 7.5 (−3.5 to 18.5) 47 15.4 (−6 to 31.4) Role functioning emotional 60.2 54.1 74.4 a 74.2 a 59.3 5.2 (−10.4 to 20.8) Mental health 63.8 61.8 69.4 a 67.9 a 62.9 1.0 (−6.8 to 8.8) Vitality 40.5 33.6 53.7 a 54.1 a 48.5 a Bodily pain 67.7 72.7 61.7 a 55.5 a 61.3 a General health 55.7 55.2 61.1 a 53.8 −1.4 (−3.6 to 0.8) 54.3 −0.9 (−5 to 3.3) Differences (Δ) indicate differences with the fatigue subgroup, where the mean of the fatigue subgroup was subtracted from the other subgroup’s mean Confidence intervals (95% CI) were calculated using the Huber-White sandwich variance estimator which accounts for within physician correlation (for details, see main text) a P P P Comparison of quality of life with other patient groups 3 P Table 3 Comparison of RAND-36 scores of different patient groups Domains RAND-36 Total UC study population Dutch GP population Depression Age 18–87 18–80+ 18–64 N 466 4,024 204  Mean Mean Δ (95% CI) Mean Δ (95% CI) Physical functioning 73.4 78.5 a 81.2 a Social functioning 66.5 74.5 a 62.5 a Role functioning physical 37.2 62.4 a 63.4 a Role functioning emotional 60.2 75 a 42.6 a Mental health 63.8 69.6 a 46.6 a Vitality 40.5 58.5 a 40.3 −0.2 (−3.6 to 3.2) Bodily pain 67.7 68.4 0.7 (−1.7 to 3.1) 68 0.3 (−4.3 to 4.9) General health 55.7 65.7 a 55.7 0 (−3.2 to 3.2) Differences (Δ) indicate differences with the total UC study population, where the mean of the total UC study population was subtracted from the other patient group’s mean 95% CI: 95% confidence interval a P t P Discussion Our findings indicate that patients with early stage unexplained complaints are mainly women in their forties, with secondary or higher education levels and with an overall remarkably poor quality of life. Their quality of life in all but one domain of the RAND-36 is significantly worse than that of patients from a general practice population, even taking into account that such a population also includes (around 13% of) unexplained complaint patients. Patients with unexplained complaints predominantly score badly on physically oriented domains, compared to depressed patients, who predominantly score badly on mental/emotional oriented domains. For the remaining domains they score on a comparably low level. Therefore, practitioners may consider to pay attention to quality of life aspects of patients with early stage unexplained complaints, even though this may not always lead to a firmer diagnosis or instant improvement in treatment options for most patients. A more intense monitoring of these patients could, however, be advised. 25 early stage unexplained complaints Second, we cannot fully exclude the possibility that some degree of selection bias is present. Participating GPs may have selectively included older patients or those with poorer quality of life. However, in a non-inclusion study in the participating general practices, we searched the electronic medical files by means of text words for eligible but not included patients with unexplained complaints. This non-inclusion study did not show major sex and age differences between included and not-included patients. However, differences may exist on other characteristics. A third limitation of our study is that no specific depression or anxiety questionnaire was used. At the start of the study, such a poor quality of life was not anticipated and, therefore, only a more general questionnaire was considered sufficient. The RAND-36 mentally/emotionally oriented domain figures and the differences we found when comparing the quality of life profile of unexplained complaint patients with depressed patients however, are not pointing towards the presence of depression or anxiety. Furthermore, the GPs did not consider the included patients to be depressed or suffering from an anxiety disorder, otherwise they would not have labelled the patient as having unexplained complaints by definition. It is not impossible however, that depression or anxiety disorders might play a role in the poor quality of life of patients with unexplained complaints. Maybe these diagnoses are established only by GPs over time, and do not become clear in (one of) the first consultations. Finally, the patient groups used for comparison of quality of life are perhaps not totally comparable to our study population. For example, there are older patients included in our study population than in the depressed patient groups. Since age has its influence on quality of life this can have influenced the contrast. Also, a depression is a treatable condition, whereas early stage unexplained complaints are not treated yet. The better quality of life in this patient group on some domains can therefore be a treatment effect. We did, however, not intend to study a fully comparable contrast in this sense beforehand, it was a result driven comparison. Although much research has been performed in patients with more chronic consequences of unexplained complaints, to our knowledge, no other study on demographic characteristics and quality of life of patients with early stage unexplained complaints in general practice has been published. Patients with unexplained complaints appear to be mainly highly educated women in their forties. They report remarkably poor levels of quality of life. Future research should explore whether and how quality of life scores and other characteristics could help in early identification of patients at risk of developing chronicity. Until then practitioners should at least be aware that early stage unexplained complaints may not always be as mild as is frequently assumed. Early stage unexplained complaints may be associated with considerable suffering on a daily basis. Awareness of potential poor quality of life may help physicians realise that they are dealing with a patient, at least, in need of more intense monitoring but maybe also of more intense treatment approaches.