This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kooiman, C. G. Articles by Trijsburg, R. W. Search for Related Content PubMed PubMed Citation Articles by Kooiman, C. G. Articles by Trijsburg, R. W. Related Collections Personality Somatoform

Alexithymia Does Not Predict the Persistence of Medically Unexplained Physical Symptoms

From the Department of Psychiatry (C.G.K., H.G.M.R.) and the Department of General Internal Medicine (J.H.B.) of the Leiden University Medical Center, and the Department of Medical Psychology and Psychotherapy, Erasmus University Rotterdam (R.W.T.), Rotterdam, the Netherlands.

Address correspondence and reprint requests to Cornelis G. Kooiman, Department of Psychiatry B1P, Leiden University Medical Center, 2300 RC Leiden, the Netherlands. E-mail: cgkooiman{at}lumc.nl

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
BACKGROUND: Alexithymia is thought to be associated with the development of medically unexplained physical symptoms (UPS). So far little research has been published on alexithymia as a risk factor for the persistence of UPS.

OBJECTIVE: To determine the clinical outcome in UPS patients and to study the relative importance of alexithymia in predicting that outcome.

METHODS: A follow-up study was conducted among general medical outpatients with UPS. Patients underwent extensive examinations at baseline and were reassessed after a mean 61-week interval. Outcome of the UPS and general health perception at follow-up were used as major outcome variables.

RESULTS: Outcome of the UPS and general health perception at follow-up were not strongly associated with each other. More than half (63%) of the patients reported improvement of their initial symptoms, but only 38% of the patients considered themselves at follow-up to be in good health. UPS outcome was predominantly predicted by the duration of the UPS and the number of additional physical symptoms at baseline. General health perception at follow-up was predominantly predicted by the general health perception at baseline and the number of additional physical symptoms and pain experience at baseline. The explained variance of the general health perception was three times as high as the explained variance for the UPS outcome. Alexithymia was not associated with any of the two outcome variables.

CONCLUSIONS: Outcome of the UPS and general health perception at follow-up are not strongly associated and are predicted by different variables. Alexithymia, however, is not an important predictor for the outcome in the majority of UPS patients.

Key Words: alexithymia, • somatization, • unexplained physical symptoms, • prognosis, • follow-up study,

Abbreviations: DES = Dissociative Experience Scale;; EPS = explained physical symptoms;; ESR = erythrocyte sedimentation rate;; HADS = Hospital Anxiety and Depression Scale;; HADSa = HADS Anxiety scale;; HADSd = HADS Depression scale;; PSQ = Physical Symptoms Questionnaire;; RCI = Reliable Change Index;; TAS-20 = Toronto Alexithymia Scale-20;; UPS = unexplained physical symptoms;; WI = Whiteley Index.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Somatization is defined in many ways, the hallmark of each definition being the presentation of physical symptoms that are not sufficiently explained by a medical condition. Lipowski (1) defines somatization as the tendency to experience and communicate medically unexplained physical symptoms (UPS) in response to psychological stress and to seek medical help for these symptoms. Kirmayer and Robbins (2) define somatization as the somatic presentation of a psychiatric disorder, such as depression or an anxiety disorder. Bridges and Goldberg (3) also define somatization as the somatic presentation of a psychiatric disorder, but in their definition they include the patient’s tendency to attribute these somatic symptoms to a physical problem and to search for a physical solution for them. Kirmayer and Robbins (2) use the term "true" somatization to describe somatization as defined by Bridges and Goldberg (3). They use the term "facultative" somatization for patients who readily consider emotional problems for their somatic symptoms when invited to by their physician. Kellner (4) uses the least restrictive definition of somatization as being the presentation of 1 or more physical complaints where, after appropriate medical evaluation, either one) the physical complaints are not at all explained by organic pathology or pathophysiological mechanisms or 2) the physical complaint or resulting social or occupational impairment is grossly in excess of what would be expected from the physical findings. Patients who somatize according to Kellner’s definition are relatively common in various medical settings (4–7), and it has been reported that their physical symptoms tend to be more chronic and refractory than those of patients with a defined medical illness (8).

Alexithymia is defined as a personality trait characterized by difficulties in describing and identifying feelings, a limited capacity for imagination and fantasy, and externally oriented thinking rather than reflection on inner experience. Because it is supposed that alexithymic patients insufficiently realize that physical sensations may be the somatic manifestations of emotions, these patients are considered to be vulnerable to incorrectly attributing innocent physical symptoms to physical disease and to seeking medical care for symptoms for which no medical explanation can be found. Alexithymia is therefore believed to be a predisposing and sustaining factor for the development of UPS, which eventually may result in one of the somatoform disorders (9). Various, generally nonclinical, studies have found an association between alexithymia and different measures of somatization (10).

Patients with UPS do not all have alexithymic characteristics to the same degree; one could therefore hypothesize that the prognosis of more alexithymic patients will be less positive than that of patients who are alexithymic to a lesser degree. However, we know of only 1 study (11) of the prognostic value of alexithymia. In this study, alexithymia predicted the existence of a somatoform disorder in patients who where clinically treated for this disorder or for a somatized anxiety disorder 2 years earlier.

A few studies have investigated the natural outcome of primary care patients with a somatic presentation of a psychiatric disorder. Fewer psychiatric symptoms at baseline (12) and improvement of mood disorder (13), but not attribution style (somatizers vs. psychologizers) (14), were associated with a better outcome.

Several outcome studies have been published on patients seeking professional help for UPS as defined by Kellner (4). Virtually all studies concerned homogeneous groups of patients with specific UPS such as dyspepsia (15), back pain (16,17), chest pain (18), cardiac palpitations (19,20), and chronic fatigue (21). Symptom severity (16,17), number of physical symptoms (17,21), subjective disability (16,17), psychiatric symptoms such as depressed mood (17,19), psychiatric disorder such as dysthymia (21), personality disorder (16), and bodily amplification (19,20) are reported as predictive of a poorer outcome in terms of physical symptoms. Reported recovery rates varied from 12% to 40% (15,18–21).

Speckens et al. (22) studied a mixed group of UPS patients. At follow-up, 9 to 17 months after initial assessment, most patients had recovered (30%) or improved (46%) with regard to their presenting symptoms. Female sex and a high number of symptoms predicted a poor outcome for improvement. In this study, the persistence of symptoms was not related to the duration of symptoms or to the presence of a psychiatric disorder as diagnosed by the Present State Examination. However, hypochondriasis, which might be seen as a personality trait, was associated with a poorer outcome (23).

The studies of the natural outcome of UPS have several limitations. Only 1 study relates to a mixed group of UPS patients (22,23), and various studies did not include the duration (14,18–20) or severity (18,20) of the symptoms as predictors of outcome. In a number of studies, the follow-up period is very short (6 months or less) (12,16,18,20), and nearly all studies were confined to the outcome of the presented UPS without additional measures for the general health perception. Apart from this, alexithymia was never studied as an independent variable.

In this study, we attempt to identify UPS patients with a high risk for an unfavorable clinical outcome, assuming that alexithymia may be one of the significant risk factors. If patients with a poor prognosis could be detected, this might result in cost-effective strategies with the restricted use of specialized interventions such as cognitive (24) or psychoanalytic psychotherapy (25) in only those patients with UPS that are not likely to resolve spontaneously. If alexithymia were found to be an important predictor for the clinical outcome, this might have practical consequences, because alexithymic patients are generally considered to fare better with structured therapy (9,26,27).

In addition to alexithymia, the effects of the following potential risk factors are investigated: sociodemographic characteristics; medical history; mental problems such as anxiety, depression, dissociation, hypochondriasis, and substance abuse; illness behavior as expressed by medical consumption; symptom characteristics such as severity and duration of symptoms; and the attribution of the symptoms (physical symptoms related/unrelated to psychiatric problems or difficulties) (28).

The outcome of the primary symptom (improved vs. not improved) was chosen in this study as the specific major outcome variable. However, it is possible that although a specific symptom does improve substantially, the patient’s physical condition does not improve because other symptoms have developed. Therefore, we chose the general health perception (good vs. poor) scale from a self-report questionnaire (29) as a second, more general, major outcome variable. The subjective, self-rated general health perception has been proven to be a powerful predictor of clinical outcome and a useful endpoint in clinical studies (30,31). Finally, we examined medical consumption and the prevalence of psychiatric pathology at follow-up as secondary outcome measures.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Patient Selection and Procedures
This study is the follow-up to a cross-sectional study among new patients of the general internal medicine outpatients’ clinic at Leiden University Medical Center. Patients gave their informed consent after receiving full verbal and written information about the nature and procedures of the study. The study was approved by the hospital’s research ethics committee. All new patients between 18 and 65 years of age were invited to participate in the cross-sectional study. Exclusion criteria included neurological disorder, organic mental disorder, mental retardation, psychotic disorder, use of neuroleptic agents, preoperative investigations, genetic counseling, sensory disorders, insufficient knowledge of the Dutch language, and serious medical conditions preventing participation in the study. All patients underwent a thorough physical examination including laboratory tests that were extended if necessary. On the basis of the internist’s final conclusions, two investigators (JHB and CGK) independently assigned the primary symptoms to one of two categories: 1) medically explained physical symptoms (EPS) and 2) medically unexplained physical symptoms (UPS) (the absence of objective pathological findings that satisfactorily explained the symptoms). The degree of agreement between the evaluations made by the two investigators was excellent (Cohen’s kappa = 0.94; p < .001). After the independent evaluations, the investigators held a consensus meeting, after which all included patients were definitively classified as having UPS or EPS. All patients participating in the cross-sectional study were asked whether they would be prepared to take part in a follow-up study. Twelve months after the first measurement, only the UPS patients were actually invited to take part in this study. At follow-up, all medical records were scrutinized to detect new diagnoses for the UPS since the initial assessment.

Instruments
At baseline, patients completed a structured interview regarding sociodemographic characteristics such as age, gender, education (32), and profession (33), their own medical and psychiatric history, and that of their parents and siblings during the respondent’s childhood until the age of 16 years; their current situation regarding work and family circumstances; their own medical and psychiatric history, and that of their partners and children after the respondent’s 16th birthday; the present use of alcohol and drugs; the duration and subjective severity of the presenting symptom and any additional symptoms; their medical consumption; and the patients’ judgment on the possibility of their physical symptoms being caused by emotional problems. Severe illnesses of participants, spouses, children, parents, or siblings were defined in terms of experiencing severe, protracted, or frequent medical or mental illnesses. Alcohol consumption was dichotomized into no (none or less than 3 U a day) and yes (sometimes or daily, more than 3 U a day).

A structured interview at follow-up addressed the outcome of the primary symptom present at the beginning of the study and the occurrence of any other current physical symptoms leading to consultation of a physician, to taking medication, or to any adjustment of his/her life. The patient was also asked about the subjective severity of current physical symptoms and about medical consumption. The outcome of the primary symptom (UPS outcome) was used as the specific major outcome variable with the following possible answers: recovered, improved, unchanged, and worse.

The RAND-36 (33) is a questionnaire designed to investigate the perception of one’s general health and the influence of symptoms on daily life. The RAND-36 has 36 items and consists of 8 scales: physical functioning, social functioning inhibited by physical or emotional problems, role limitations due to physical symptoms, role limitations due to emotional problems, mental health, vitality, pain, and general health perception. It also includes a single item on the perceived improvement of health in the preceding year. The scales range from 0 to 100, with a higher score indicating a better state of health. English (34) and Dutch research (35–37) has demonstrated sufficient variability in scale scores and good construct validity of the RAND-36. The internal consistency of the individual scales is good ( = 0.91, 0.84, 0.86, 0.85, 0.82, 0.85, 0.83, 0.75, respectively). Patients completed the RAND-36 at baseline and at follow-up. The "general health perception" at follow-up was used as the general major outcome variable.

The Toronto Alexithymia Scale 20 (TAS-20) (38,39) is generally accepted as the standard instrument for determining the degree of alexithymia. The TAS-20 is a self-report questionnaire with 20 items on a 5-point Likert scale. The items in the TAS-20 relate to three alexithymic dimensions: difficulties in identifying feelings, difficulties in describing feelings, and externally oriented thinking. The results of the Dutch translation of the TAS-20 are best reflected in the total score (40). The theoretical score range is from 20 to 100 and a total score 60 is considered to indicate alexithymia (41). The internal consistency of the TAS-20 was excellent (Cronbach’s = 0.85).

Levels of anxiety and depression were assessed by means of the Hospital Anxiety and Depression Scale (HADS) (42), which has been translated into Dutch by Spinhoven et al. (43). The HADS is a 14-item self-report questionnaire especially relevant for use with patients with somatic symptoms, because the HADS does not represent somatic symptoms of anxiety and depression. The HADS differentiates between anxiety and depression; the scores on the anxiety and depression subscales (HADSa and HADSd, respectively) provide a reliable measure for the severity of the psychiatric condition. The score range is from 0 to 21 for the anxiety and depression subscales and from 0 to 42 for the total scale. A score of 11 on the subscales indicates that an anxiety disorder or depression is likely (43–45). The internal consistency of the subscales and the total scale (HADStot) in this study was good ( = 0.82 and 0.81, respectively; overall = 0.87).

The Dissociative Experience Scale (DES-II) (46,47) was used to measure dissociation. This scale has been translated into Dutch by Draijer and Boon (48). The DES is a self-report questionnaire, with 28 items measuring dissociative experiences to be rated on a percentage scale. The score range is from 0 to 100, and a total score of 25 or above on the DES implies that a dissociative disorder is likely (48). The reliability and validity of the Dutch version of the DES is good (49,50). The internal consistency of the questionnaire in this study was excellent ( = 0.87).

Hypochondriac preoccupation was measured using the Whiteley Index (WI) (51, 52), which has been translated into Dutch by Speckens et al. (53). The WI is a self-report questionnaire comprising 14 yes/no questions. The score range is from 0 to 14, and a total score of 7 or above indicates that hypochondriasis is likely. The internal consistency, the test–retest reliability, and the discriminating validity of this questionnaire are good (22,23). In this study, the internal consistency of the WI was reasonable ( = 0.74).

We used a list of questions (Physical Symptoms Questionnaire (PSQ)) pertaining to physical symptoms on a 4-point scale (never, sometimes, regularly, frequently). The items represent the physical symptoms in the somatoform disorders section of the SCAN (54), supplemented with a number of questions regarding fatigue. The total score represents the total number of symptoms that the patient reported as occurring either regularly or frequently, corrected for the difference in number of items between women (79 items) and men (73 items).

Statistical Analyses
To ease the interpretation of results and to focus on those associations that can be considered to be robust, we dichotomized the outcome variables and accepted some loss of measurement sensitivity. The UPS outcome measure was reduced to 2 categories: improved (recovered or improved) and not improved (unchanged or worse). The outcome of the general health perception (RAND-36) was divided into two categories: in good health or in poor health relative to the general health perception in the general population. Using published data from the Dutch general population (mean = 72.7; SD = 22.7) (36) and from our own sample (mean = 54.8; SD = 21.5), a cut-off score of 60 was determined. Scores above this cut-off are considered to be within the range of the general population (55). Using this cut-off score, general health perception has been dichotomized into "in good health" and "in poor health." When patients move from "in poor health" to "in good health" or vice versa, the change is considered to be clinically significant. Additionally, the Reliable Change Index (RCI) (55) was calculated. When RCI > 1.96, the change is also considered reliable.

The data were analyzed by means of the SPSS statistical package, version 10.0. In bivariate analyses, the chi-square test with Yates’ continuity correction (nominal data), Student’s t test (continuous data), Mann-Whitney U tests (nonparametric data) and Wilcoxon signed rank test (paired data) were used as appropriate. No adjustments for multiple comparisons were made because we wanted to explore all possible associations (56). We conducted a set of stepwise logistic regression analyses with age, gender, alexithymia, and those independent variables that were bivariately associated with any of the two major outcome variables. For all tests, statistical significance was set at the 5% level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
General Outcome Results
At baseline, 284 patients were classified as having medically unexplained physical symptoms (UPS). Fifteen patients fulfilled 1 or more exclusion criteria. Seven patients who refused and 2 patients who agreed to participate in the study proved to have been misclassified as having UPS. One example is a patient with fever of undetermined origin and an ESR of 110 that was initially overlooked by the physician. Furthermore, in the first 3 months after the initial diagnosis, a somatic disorder was found to explain the symptoms in 3 patients who refused to participate in the study and in 5 patients who agreed to participate. An example is a patient with abdominal complaints who eventually proved to have a malignancy of the pancreas.

In this study, further analysis was done on the 252 included patients whose symptoms were not reclassified. Of these patients, 127 (50.4%) agreed to participate and 125 (49.6%) declined (Figure 1). Participating patients were somewhat older than the patients who declined to participate (40.2 (12.7) years vs. 37.1 (11.8) years; p < .05). There were no differences in terms of gender, type of UPS, and number of additional diagnoses as formulated by the consulted physician.

View larger version (19K): [in this window] [in a new window]   Figure 1. Cohort profile and number of cases

Patients appeared to have had UPS for a considerable length of time (median, interquartile range: 12 months, 6–29 months) before they were referred to the general medical outpatient clinic. The alexithymia scores at baseline were not particularly high (TAS-20 total score mean (SD) = 48.0 (12.5), mode = 45) and only 22 (17%) UPS patients had TAS-20 scores above 60.

At follow-up, 26 (21%) patients reported complete remission of their initial symptoms and 53 (42%) patients reported improvement. In 37 (29%) patients the symptoms were unchanged, and 11 (9%) patients reported a deterioration of the symptoms. Thus, the UPS outcome resulted in N = 79 (62%) improved and N = 48 (38%) not improved patients.

Fifty-three (42%) of the 127 patients reported at follow-up the development of other symptoms during the previous year for which they had sought medical assistance, had to take medication, or had to adapt their daily activities. In 37 (70%) of these cases, a diagnosis had been established by a physician. Twenty-two (59%) of these 37 cases involved a chronic condition such as arthrosis or migraine. The other cases involved an acute condition such as a fracture or herpes zoster. At follow-up, 18 patients reported that the physical symptoms had not hindered them at all, and 82 (65%) patients reported that the presence of physical symptoms had not hindered them in their activities with family, friends, neighbors, or others.

At baseline, 92 (72%) patients judged themselves to be in poor health (RAND-36). At follow-up, this figure was 79 (62%) patients. Twenty-three (18%) patients changed from poor to good health, and 10 of these had an RCI > 1.96, indicating a change that was both clinically significant and reliable. Ten (8%) patients changed from good to poor health, and 2 of these had an RCI < -1.96.

Association of Major Outcome Variables and Mental Disorder at Follow-up
The associations regarding the major outcome variables and mental disorder at follow-up are presented in Table 1. At follow-up, 38 (30%) patients scored above the HADS and DES cut-off scores, and are considered to have a mental disorder. Patients whose UPS did not improve were more likely to have a mental disorder at follow-up. The general health perception was more strongly associated with the presence of a mental disorder at follow-up than with the outcome of the primary symptom (p = .08). Multiple logistic regression analysis showed mental disorder at follow-up to be associated with the HADS total score at baseline (B = 0.23 (0.04), p < .01), but not with the TAS total score or any of the other independent variables.

During the follow-up period, 69 (54%) patients sought referral to one or more somatic specialists of various disciplines (median = 1; interquartile range = 1–2). The number of specialties to which a person was referred was associated with no improvement of the UPS (MW-U test; p < .01), a poor general health perception at follow-up (MW-U test; p < .01), and the presence of a mental disorder at follow-up (MW-U test; p < .05).

Thirty-seven (29%) patients consulted 1 or more alternative healers (median = 0; interquartile range = 0–1) during the follow-up period. The number of forms of alternative medicine was associated with a poor general health perception (MW-U test; p < .01) and the presence of a mental disorder at follow-up (MW-U test; p < .05).

The TAS total score at baseline was not associated with the number of visits to general practitioners, the number of consulted specialties, or the number of alternative healers during the follow-up period.

Prognosis of the Unexplained Physical Symptoms (UPS Outcome)
The outcome of the UPS was not associated with demographic or social characteristics (Table 2) or with a history of severe illnesses in the patients, their parents, or siblings. Anxiety, depression, and hypochondriasis at baseline predicted the UPS outcome (Table 3). Patients whose symptoms did not improve had more symptoms of anxiety and depression at baseline and were more hypochondriac. Finally, the duration of the symptoms and the number of physical symptoms (PSQ) at baseline were associated with an absence of improvement of the UPS (Table 4). The UPS outcome was not associated with the degree of alexithymia (Table 3).

Prognosis Regarding General Health Perception
Patients with a poor general health perception at follow-up were more likely to receive social benefit (Table 2), and were more likely to have experienced prolonged and/or serious illnesses during childhood (32% vs. 8%, p < .01). At baseline, they were more anxious and depressed and more likely to have hypochondriasis (Table 3). The general health perception at follow-up was not associated with the degree of alexithymia at baseline (Table 3).

General health perception at follow-up was strongly associated with symptom characteristics and health perception at baseline (Table 4). At baseline, patients with a poor general health perception at follow-up had had the UPS longer, they were more likely to have it daily or constantly, subjectively they suffered more from their symptoms, they were more likely to feel limited in their daily life by their symptoms, and they felt less fit and healthy. They were also more likely to believe that their symptoms might be due to mental problems, but most of these appeared to be related to patients who actually had a mental disorder.

In the logistic regression model, general health perception at baseline (B = -0.04 (0.01), p < .01), PSQ score (B = 0.07 (0.03), p < .05), and pain (B = -0.03 (0.01), p < .05) were associated with a poor general health perception at follow-up with 46% of the total variance explained (Nagelkerke R² = 0.46).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
This is the first study on the prognostic significance of alexithymia in a nonpsychiatric sample of outpatients with heterogeneous UPS. Furthermore, in addition to assessing the outcome of the UPS, we used the general health perception as a general outcome measure. This enabled us to gain an impression not only of the outcome of the presenting symptom, but also of the patient’s perceived general state of health.

We studied the impact of alexithymia, primary or secondary (57–58), as a predictor for the prognosis of patients with UPS. Most patients did not have a rather high alexithymia score at baseline, and only a minority of patients had TAS-20 scores indicating clinically significant alexithymia. In the baseline study (6), we had ascertained that the total alexithymia score was not associated with the notion (or the lack thereof) that the physical symptoms might be the result of emotional problems; as in the study of Kirmayer and Robbins (2), the majority of patients with a mental disorder adequately considered the possibility that their symptoms are the result of psychiatric problems. Accordingly, alexithymia was not associated with any of the two major outcome variables. Re-analyses of the data with outcome of UPS defined as recovered vs. not recovered resulted in practically the same results.

An important secondary finding of this study is that the outcome of the UPS is only very weakly associated with the general health perception. Both outcome measures also appear to differ in terms of their main predictive variables. UPS outcome was determined primarily by the duration of the symptoms and the total number of reported physical symptoms. The general health perception was primarily predicted by the general health perception at baseline, the total number of physical symptoms, and pain experience at baseline. The logistic regression analysis with outcome UPS as dependent variable resulted in an explained variance of only 14% indicating that, in contrast with the general health perception, the outcome UPS is poorly predicted by the available independent variables. These results suggest that future research should be focused on the general health experience and its determinants and not merely toward the initially presented UPS. Likewise, the same is probably true in clinical practice when a therapy is indicated.

Although Speckens et al. (22) conducted their study among general medical outpatients in the same hospital with the same criteria for the outcome of UPS, they did not find the duration of symptoms to be predictive of the outcome of UPS. This discrepancy may be caused by a longer duration of symptoms at baseline (median 12.0 months vs. median 7.8 months) and a less favorable outcome of the UPS (63% improvement vs. 76%) in our study, but probably also to a shortage in statistical power in the study of Speckens et al. (22) with only 81 participating patients.

It would be reasonable to expect that an association between alexithymia and both major outcome measures might have been found with a different definition of UPS—instead of UPS being defined as the absence of objective pathological findings that satisfactorily explain the symptoms (4) but instead as the tendency to experience and communicate medically unexplained physical symptoms in response to psychological stress (1). The latter definition is clinically appealing and fits more closely with the postulated relationship between alexithymia and UPS, but it is difficult to apply without ambiguity.

The patients with UPS appear to have some common characteristics. The medical consumption by patients with UPS during the follow-up period did not appear to be excessive. A quarter of the patients did not even consult their general practitioner during the follow-up period, and most of the patients did not visit the general practitioner more than twice. This rate is lower than the average of four general practitioner visits per year by respondents of the same age from the general population (59). However, 1 year after medical examination, 38% of the patients had still not experienced any improvement in the UPS, and 62% of the patients rated their general health as poor, the latter being more markedly associated with psychiatric symptoms. In short, despite 60% of the patients reporting an improvement in the UPS, an equal percentage of patients perceived their health as poor at follow-up, a phenomenon that is associated with an increased prevalence of psychiatric symptoms both at baseline and at follow-up. These data suggest that a substantial number of UPS patients experience chronic feelings of discomfort with "mild" but possibly long-standing psychiatric symptoms.

The strengths of this study are that the symptoms of the patients were categorized as unexplained only after complete medical examination, and that at follow-up all medical files were scrutinized for possible medical explanations for the symptoms during the follow-up period. We therefore focused a great deal of attention on correct classification of the patients’ symptoms, both at the time of the baseline study and at the time of follow-up. Despite extensive examination by the internist and systematic monitoring, the symptoms of nine patients were retrospectively found to have been incorrectly classified as UPS at baseline. In these cases, the internist usually reached a final conclusion prematurely on the basis of clinical impression, and in anticipation of the ordered laboratory examinations revealing no positive findings. Eight patients were also found during the course of the follow-up period to have a (sometimes serious) somatic condition to which the original UPS could be ascribed. In all, there were 17 (17/269 = 6%) patients for whom the symptoms were initially incorrectly classified as UPS. Misdiagnoses will always occur in studies and in clinical practice, and a misdiagnosis rate of 6% should perhaps not be considered excessively high. However, in general this figure will be higher than in this study, where the research conditions optimized clinical practice.

An important limitation of this study is the relatively high number of patients who decided not to participate. This will certainly have resulted in a loss of statistical power, possibly reflected in the nonsignificant tendencies of alexithymia to be associated with a poorer prognosis on both major outcome measures (Table 3). Selection bias may be less probable because the mean (SD) alexithymia scores of the participants did not differ significantly from those of the patients who decided not to participate in the follow-up study (48.1 (12.6) respectively 50.9 (12.9); p = .24). Of course, we do not know the alexithymia scores of the patients who already had declined to participate in the baseline study.

A second limitation of this study is that follow-up assessments were primarily based on the results of a standardized structured interview that focused on the patients’ evaluation of his/her symptoms without conducting an independent medical examination at follow-up. Theoretically, this means that for a few patients a somatic explanation for the symptoms may have become apparent in the interim, without this being known to the patient or the investigators. However, in the Dutch health care system, regular contact between family physician and specialist are customary, and an exhaustive study of the medical records was carried out for the patients who took part in the follow-up. The number of patients in whom a somatic diagnosis may have been missed is therefore likely to be negligible.

Finally, one may put forward that people who are alexithymic may not always be able to report this adequately on a self-report questionnaire or that the TAS-20 preferably should be used in combination with other research instruments to assess the ability to identify and express feelings. However, so far the TAS-20 has been the standard instrument to assess alexithymia, and in several studies the TAS-20 has been shown to have sufficient construct and discriminant validity (9,40).

We conclude from our study that although about 60% of the UPS patients experience recovery or improvement of their initial symptoms, an almost equal proportion of patients have chronic feelings of discomfort or malaise, although not of great severity. The majority of the UPS patients are not alexithymic, and alexithymia therefore does not play an important role in the clinical outcome for the majority of the patients.

Although UPS patients do appear to represent a special group with chronic feelings of discomfort or malaise, in the majority of cases there is no obvious psychiatric disorder. In accordance with the treatment recommendations made by Barsky and Borus (8), a physical examination combined with reassurance will therefore be sufficient in most cases. However, in view of the number of misclassifications and the number of diagnoses that required reconsideration over the course of a few months, we wish to stress that the physical examination must be carried out properly and we want to warn against jumping to the conclusion that there is no somatic explanation for the symptoms in the case of (initial) negative findings, or worse: that, in the absence of a somatic explanation, the symptoms must have a psychiatric origin. For a small group of patients, psychotherapy and/or psychopharmacotherapy are indicated. These are patients with an anxiety disorder or depression, and those with UPS who are unable to adequately deal with their symptoms and/or in whom the symptoms are related to mental stress. In the latter group of patients, both cognitive (24) and psychodynamic psychotherapy (25) may be indicated.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
We thank the participating patients and physicians for their contribution to the data collection.

Received for publication January 16, 2003.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 

1. Lipowski ZJ. Somatization: the concept and its clinical application. Am J Psychiatry 1988; 145: 1358–68.[Abstract/Free Full Text]
2. Kirmayer LJ, Robbins JM. Three forms of somatization in primary care: prevalence, co-occurrence, and sociodemographic characteristics. J Nerv Ment Dis 1991; 179: 647–55.[Medline]
3. Bridges KW, Goldberg DP. Somatic presentation of DSM-III psychiatric disorders in primary care. J Psychosom Res 1985; 29: 563–9.[CrossRef][Medline]
4. Kellner R. Somatization. Theories and research. J Nerv Ment Dis 1990; 178: 150–60.[CrossRef][Medline]
5. Hemert van AM, Hengeveld MW, Bolk JH, Rooijmans HGM, Vandenbroucke JP. Psychiatric disorders in relation to medical illness among patients of a general medical out-patient clinic. Psychol Med 1993; 23: 167–73.[Medline]
6. Kooiman CG, Bolk JH, Brand R, Trijsburg RW, Rooijmans HGM. Is alexithymia a risk factor for unexplained physical symptoms in general medical outpatients? Psychosom Med 2000; 62: 768–79.[Abstract/Free Full Text]
7. Nimnuan C, Hotopf M, Wessely S. Medically unexplained symptoms. An epidemiological study in seven specialities. J Psychosom Res 2001; 51: 361–7.[CrossRef][Medline]
8. Barsky AJ, Borus JF. Functional somatic syndromes. Ann Intern Med 1999; 130: 910–21.[Abstract/Free Full Text]
9. Taylor G J, Bagby RM, Parker JDA. Disorders of affect regulation: alexithymia in medical and psychiatric illness. Cambridge: Cambridge University Press; 1997.
10. Kooiman CG. The status of alexithymia as a risk factor in medically unexplained physical symptoms. Compr Psychiatry 1998; 39: 152–9.[CrossRef][Medline]
11. Bach M, Bach D. Predictive value of alexithymia: a prospective study in somatizing patients. Psychother Psychosom 1995; 64: 43–8.[Medline]
12. Downes-Grainger E, Morris R, Gask L, Faragher B. Clinical factors associated with short-term changes in outcome of patients with somatized mental disorder in primary care. Psychol Med 1998; 28: 703–11.[CrossRef][Medline]
13. Craig TKJ, Boardman AP, Mills K, Daly-Jones O, Drake H. The south London somatisation study I: longitudinal course and the influence of early life experiences. Br J Psychiatry 1993; 163: 579–88.[Abstract/Free Full Text]
14. Garcia-Campayo J, Larrubia J, Lobo A, Pérez-Echeverría MJ, Campos R. Attribution in somatizers: stability and relationship to outcome at 1-year follow-up. Acta Psychiatr Scand 1997; 95: 433–8.[Medline]
15. Morris C, Chapman R, Mayou R. The outcome of unexplained dyspepsia. A questionnaire follow-up study of patients after endoscopy. J Psychosom Res 1992; 36: 751–7.[CrossRef][Medline]
16. Gatchel RJ, Polatin PB, Kinney RK. Predicting outcome of chronic back pain using clinical predictors of psychopathology: a prospective analysis. Health Psychol 1995; 14: 415–20.[CrossRef][Medline]
17. Dionne CE, Koepsell ThD, Korff M von, Deyo RA, Barlow WE, Checkoway H. Predicting long-term functional limitations among back pain patients in primary care settings. J Clin Epidemiol 1997; 50: 31–43.[CrossRef][Medline]
18. Potts SG, Bass CM. Psychological morbidity in patients with chest pain and normal or near normal coronary arteries: a long-term follow-up study. Psychol Med 1995; 25: 339–47.[Medline]
19. Barsky AJ, Cleary PD, Coeytaux RR, Ruskin JN. The clinical course of palpitations in medical outpatients. Arch Intern Med 1995; 155: 1782–8.[Abstract/Free Full Text]
20. Barsky AJ, Ahern DK, Bailey ED, Delamater BA. Predictors of persistent palpitations and continued medical utilization. J Fam Pract 1996; 42: 465–72.[Medline]
21. Clark MR, Katon W, Russo J, Kith P, Sintay M, Buchwald D. Chronic fatigue: risk factors for symptom persistence in a 2-year follow-up study. Am J Med 1995; 98: 187–95.[CrossRef][Medline]
22. Speckens AEM, Hemert AM van, Bolk JH, Rooijmans HGM, Hengeveld MW. Unexplained physical symptoms: outcome, utilisation of medical care and associated factors. Psychol Med 1996a; 26: 745–52.[Medline]
23. Speckens AEM, Hemert AM van, Spinhoven P, Bolk JH. The diagnostic and prognostic significance of the Whitely Index, the Illness Attitude Scales and the Somatosensory Amplification Scale. Psychol Med 1996b; 26: 1085–90.[Medline]
24. Kroenke, K, Swindle R. Cognitive-behavioral therapy for somatization and symptom syndromes: a critical review of controlled clinical trials. Psychother Psychosom 2000; 69: 205–15.[CrossRef][Medline]
25. Guthrie E. Treatment of functional somatic symptoms: psychodynamic treatment. In: Mayou R, Bass C, Sharpe M, eds. Treatment of functional somatic symptoms, chapter 8. Oxford: Oxford University Press; 1995.
26. Krystal H. Alexithymia and the effectiveness of psychoanalytic treatment. Int J Psychoanal Psychother 1982–1983 9: 353–78.
27. Sifneos PE. Psychotherapies for psychosomatic and alexithymic patients. Psychother Psychosom 1983; 40: 66–73.[Medline]
28. Mayou R, Bass C, Sharpe M. Overview of epidemiology, classification and aetiology. In: Mayou R, Bass C, Sharpe M, ed. Treatment of functional somatic symptoms. Oxford: Oxford University Press; 1995.
29. Hays RD, Sherbourne CD, Mazel RM. The RAND 36-item health survey 1.0. Health Econ 1993; 2: 217–27.[Medline]
30. Eriksson I, Undén A -L, Elofson S. Self-rated health. Comparisons between three different measures. Results from a population study. Int Epidemiol Assoc 2001; 30: 326–33.
31. Fayers PM, Sprangers MAG. Understanding self-rated health. Lancet 2002; 359: 187–8.[CrossRef][Medline]
32. UNESCO. International Standard Classification of Education (ISCED). Paris, UNESCO 1976.
33. Jonsson JO. Class origin, cultural origin and educational attainment: the case of Sweden. Eur Sociol Rev 1987; 3: 229–42.[Abstract/Free Full Text]
34. Brazier JE, Harper R, Jones NMB, O’Cathain A, Thomas KJ, Usherwood T, Westlake L. Validating the SF-36 health care questionnaire: new outcome measure for primary care. BMJ 1992; 305: 160–4.
35. Zee van der KI, Sanderman R. Het meten van de algemene gezondheidstoestand met de RAND-36. Een handleiding. Noordelijk Centrum voor Gezondheidsvraagstukken/RUG, Groningen, 1993[Measuring general health status with the RAND-36. A guide. Northern Centre for Health Issues/RUG, Groningen, the Netherlands, 1993].
36. Zee van der KI, Sanderman R, Heyink, JW, Haes de H. Psychometric qualities of the RAND 36-item Health Survey 1.0: a multidimensional measure of general health status. Int J Behav Med 1996a; 3: 104–22.[CrossRef][Medline]
37. Zee van der KI, Sanderman R, Heyink J. A comparison of two multidimensional measures of health status: the Nottingham Health Profile and the RAND 36-Item Health Survey 1.0. Qual Life Res 1996b; 5: 165–74.[CrossRef][Medline]
38. Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale-I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994a; 38: 23–32.[CrossRef][Medline]
39. Bagby RM, Taylor GJ, Parker JDA. The twenty-item Toronto Alexithymia Scale-II. Convergent, discriminant, and concurrent validity. J Psychosom Res 1994b; 38: 33–40.[CrossRef][Medline]
40. Kooiman CG, Spinhoven P, Trijsburg RW. The assessment of alexithymia: A critical review of the literature and a psychometric study of the Toronto Alexithymia Scale-20. J Psychosom Res 2002; 53: 1083–90.[CrossRef][Medline]
41. Parker J, Taylor G Bagby M. Alexithymia and the processing of emotional stimuli: an experimental study. New Trends Exp Clin Psychiatry 1993; 9: 9–14.
42. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67: 361–70.[Medline]
43. Spinhoven P, Sloekers PPA, Kempen GIJM, Speckens AE, van Hemert AM. A validation study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch subjects. Psychol Med 1997; 27: 363–70.[CrossRef][Medline]
44. Aylard PR, Gooding JH, McKenna PJ, Snaith RP. A validation study of three anxiety and depression self-assessment scales. J Psychosom Res 1987; 31: 261–8.[CrossRef][Medline]
45. Lewis G, Wessely S. Comparison of the General Health Questionnaire and the Hospital Anxiety and Depression Scale. Br J Psychiatry 1990; 157: 860–4.[Abstract/Free Full Text]
46. Bernstein EM, Putnam FW. Development, reliability, and validity of a dissociation scale. J Nerv Ment Dis 1986; 174: 727–35.[Medline]
47. Carlson EB, Putnam FW. Manual for the Dissociative Experiences Scale. Beloit, 1992.
48. Draijer N, Boon S. The validation of the Dissociative Experience Scale against the criterion of the SCID-D, using Receiver Operating Characteristics (ROC) analysis. Dissociation 1993; 6: 28–37.
49. Boon S, Draijer N. Screening en diagnostiek van dissociatieve stoornissen (Screening and diagnostics for dissociative disorders). Swets & Zeitlinger, Lisse, 1995.
50. IJzendoorn van MH, Schuengel C. The measurement of dissociation in normal and clinical populations: meta-analytic validation of the dissociative experiences scale (DES). Clin Psychol Rev 1996; 16: 365–82.[CrossRef]
51. Pilowski I. Dimensions of hypochondriasis. Br J Psychiatry 1967; 113: 89–93.[Abstract/Free Full Text]
52. Pilowski I. Abnormal illness behaviour (dysnosognosia). Psychother Psychosom 1986; 46: 76–84.[CrossRef][Medline]
53. Speckens AEM, Spinhoven P, Sloekers PPA, Bolk JH, Hemert AM van. A validation study of the Whitely index, the Illness Attitude Scales and the Somatosensory Amplification Scale in general medical and general practice patients. J Psychosom Res 1996c; 40: 95–104.[CrossRef][Medline]
54. Wing JK, Babor T, Brugha T, Burke J, Cooper JE, Giel R, Jablenski A, Regier D, Sartorius N. SCAN; Schedules for clinical assessment in neuropsychiatry. Arch Gen Psychiatry 1990; 47: 589–93.[Abstract/Free Full Text]
55. Jacobson NS, Truax P. Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. J Consult Clin Psychol 1991; 59: 12–19.[CrossRef][Medline]
56. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990; 1: 43–6.[Medline]
57. Freyberger H. Supportive psychotherapy techniques in primary and secondary alexithymia. Psychother Psychosom 1997; 28: 337–42.
58. Wise TN, Mann LS, Mitchell JD, Hryvniak M, Hill B. Secondary alexithymia: an empirical validation. Compr Psychiatry 1990; 31: 284–8.[CrossRef][Medline]
59. CBS, Centraal Bureau voor de Statistiek. Vademecum Gezondheidsstatistiek Nederland (Vademecum Health Statistics Netherlands). Den Haag: SDU 2000,pg. 146.

This article has been cited by other articles:

				A. K. Mattila, E. Kronholm, A. Jula, J. K. Salminen, A.-M. Koivisto, R.-L. Mielonen, and M. Joukamaa Alexithymia and Somatization in General Population Psychosom Med, July 1, 2008; 70(6): 716 - 722. [Abstract] [Full Text] [PDF]

				F. Friedberg and J. Quick Alexithymia in Chronic Fatigue Syndrome: Associations With Momentary, Recall, and Retrospective Measures of Somatic Complaints and Emotions Psychosom Med, January 1, 2007; 69(1): 54 - 60. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kooiman, C. G. Articles by Trijsburg, R. W. Search for Related Content PubMed PubMed Citation Articles by Kooiman, C. G. Articles by Trijsburg, R. W. Related Collections Personality Somatoform