| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Institute of Psychology (N.H., C.E., S.E.), Technical University of Braunschweig; Clinic of Neurology (M.M.M., C.M.), Municipal Hospital of Braunschweig; Research Institute of Cognitive Neurology (C.M.), Technical University of Braunschweig, Braunschweig, Germany.
Address correspondence and reprint requests to Nina Heinrichs, Institute of Psychology, Technical University of Braunschweig, Spielmannstr. 12a, 38106 Braunschweig, Germany. E-mail: n.heinrichs{at}tu-bs.de
| ABSTRACT |
|---|
|
|
|---|
Methods: In a prospective longitudinal study, we assessed fear of bodily sensations and cognitions related to anxiety at the time of hospital admission and 3 months later in 43 patients with an episode of vestibular neuritis (VN) or benign paroxysmal positional vertigo (BPPV). All participants were assessed for mental disorders using a structured clinical interview.
Results: Only the interaction between fear of bodily sensations within the first 2 weeks after admission and the type of vestibular disorder predicted the extent of dizzy complaints 3 months later; this accounts for 21% of the variance in a multiple regression analysis. Specifically, the prediction was valid only in patients with VN but not in patients with BPPV. Further analysis demonstrated that the interaction was not due to the peripheral vestibular disorder per se but rather determined by the initial severity of dizziness, which was significantly different in BPPV and VN patients.
Conclusions: The present study demonstrates that, for the development of persistent psychogenic dizziness after a peripheral vestibular disorder, the fear of bodily sensations is only relevant in interaction with the initial severity of dizziness experienced during the acute organic episode. To prevent development of persistent psychogenic dizziness, we feel that our results indicate the need to screen patients with vestibular disorders for at-risk status and offer them psychological support to deal with their symptoms.
Key Words: panic agoraphobia depression dizziness vertigo
Abbreviations: VN = vestibular neuritis; BPPV = benign paroxysmal positional vertigo; ACQ = Anxiety Cognitions Questionnaire; BSQ = Body Sensation Questionnaire; BAI = Beck Anxiety Inventory; MI = Mobility Inventory; BDI = Beck Depression Inventory; SCL-90-R GSI = Symptom Checklist Revised Global Severity Index; RC = (at least partially) recovered group; CC = continuously complaining group.
| INTRODUCTION |
|---|
|
|
|---|
Besides well-defined organic disorders, dizziness (more than vertigo) can be a symptom of mental disorders, especially anxiety disorders such as agoraphobia or panic disorder (2). Many patients with anxiety disorders specify dizziness as the most anxiety-provoking and uncomfortable sensation among the bodily sensations that may be caused by panic attacks (3,4). This observation has led researchers to investigate the co-occurrence of dizziness, vestibular disorders, and mental disorders (5). Simon and colleagues (6) provided an overview of this type of studies and concluded that, among patients presenting for the assessment and treatment of dizziness, a significant proportion suffered from panic disorder. Similarly, among patients presenting for the assessment and treatment of panic disorder, a significant proportion demonstrated evidence of vestibular dysfunction. This review as well as more recent studies led to a number of conclusions. a) Dizziness is prevalent in both types of disorders (somatic and mental). b) Dizziness due to a somatic disorder is not always accompanied by the characteristic medical assessment results. c) Dizziness due to a mental disorder is sometimes accompanied by the characteristic medical assessment results (i.e., there are indicators for functional impairments in the vestibular system in anxiety disorders). d) Patients with a vestibular disorder may develop fears that result in maintained sensations of dizziness despite a recovery of the primary somatic cause, or patients with a mental disorder may develop a vestibular disorder that results in maintained sensations of dizziness. Among the primary problems for interpreting the overlapping results (7) is that patients with anxiety disorder usually run through vestibular tests after they develop the disorder. Similarly, patients with vestibular disorders are psychologically assessed during the acute phase of their vestibular disorder. However, few studies are prospectively designed with long-term follow-up of either patient group, thus complicating the interpretation of correlational results.
Recently, Godemann and colleagues (8) reported results from a prospective study, where they repeatedly assessed cognition status in patients with an attack of vestibular neuritis over a 2-year period. They demonstrated that fearful thoughts arising on the first day of this vestibular disorder do not predict the subsequent development of panic disorder. However, the mean values of the fear of bodily sensations as measured with the Body Sensation Questionnaire (BSQ) and the cognitive interpretation (or misinterpretation) of the symptoms as measured with the Anxiety Cognition Questionnaire (ACQ) were predictive for panic disorder 2 years later if applied 6 weeks after the vertigo episode. In addition, the BSQ (but not the ACQ) was also predictive if applied 7 to 12 days after the admission. This may imply that some time to process the experience of vertigo psychologically is necessary to predict the adverse mental outcome of some of these patients 2 years later.
Patients with vestibular disorders have a high prevalence of anxiety disorders, which is often interpreted as a response to the experience of the vestibular episode. Thus, the experience of dizziness (or vertigo) in the context of a vestibular episode is considered a risk factor for the development of anxiety disorders (9,10). Among the few longitudinal studies, Godemann and colleagues (8) found that the tendency to interpret bodily sensations as dangerous symptoms make individuals more prone to developing panic disorder after an acute episode of vestibular neuritis. In contrast, Best and colleagues (11) concluded in their study with eight different diagnostic groups (among them healthy volunteers, patients with certain vestibular or mental disorders) that "high anxiety scores are not a result of vestibular deficits" (p. 658). However, Best and colleagues did not follow patients longitudinally and compared independent groups cross-sectionally.
In the present study, we collected data in a general community hospital with the aim of prospectively identifying individuals who will report dizziness 3 months after hospital admission, despite a good functional recovery of the initial vestibular disorder. Similar to the study by Godemann et al., we assumed that fear of bodily sensations as well as cognitions about these symptoms would contribute to the development of a secondary psychogenic dizziness. In contrast to their study, we were not interested in predicting the incidence of mental disorders (categorical approach) but in predicting the maintained complaint of dizziness after recovery of the vestibular disorder (dimensional approach). Our two-fold goal was a) to identify—at the time of the acute initial episode of vertigo—the subsample of patients with continuous complaints of dizziness and their request for further medical attention and b) to describe the patients psychologically at a follow-up assessment 3 months later. Usually, the initial peripheral vestibular dysfunction has been well compensated by then. We assumed that patients who may still complain about dizziness at the follow-up assessment time may already have differed psychologically (i.e., reported more psychological complaints such as anxiety or depression) at the initial assessment from those who recovered at follow-up and did not complain about dizziness.
We included patients admitted to the hospital with two distinct and common peripheral vestibular disorders: vestibular neuritis (VN) or benign paroxysmal positional vertigo (BPPV). Both disorders can be diagnosed based on their typical history and clinical findings, and both are characterized by an acute onset of vertigo (or dizziness) that is usually limited in time if treated properly. Nevertheless, a common clinical experience finds some of these patients complain about ongoing dizziness later on, although their original peripheral vestibular dysfunction cannot be reproduced on clinical examination.
We hypothesized that those who maintain their complaints about dizziness will be, in general, more psychologically impaired at the time of admission, as indicated in scores of standardized psychometric questionnaires reflecting broad psychological functioning (such as the revised Symptom Checklist-90). We also explored the prevalence of mental disorders at the time of admission as well as in the past (lifetime assessment). We had no specific hypothesis about the lifetime disorders but we expected an increase in anxiety disorders from the first to the second assessment. We only describe the frequency because 3 months incidence rates for mental disorders are usually low and require very large samples to be statistically valid. Furthermore, we hypothesized that the individual BSQ and the ACQ scores during the initial treatment period will predict the development of psychogenic dizziness 3 months after the acute vestibular episode in both disorders.
| METHODS |
|---|
|
|
|---|
A total of 51 patients were included in the study. Fifteen (23%) individuals were eligible but declined to participate in the study. Of the 51 individuals who participated in the first assessment, eight (16%) patients chose to not participate in the second assessment. leaving a total of 43 data sets with complete pre and post data. Table 1 provides sociodemographic data on both samples. Between those who completed the post assessment and those who did not, there were no significant differences in sociodemographic characteristics (age, gender, marital status) in any of the psychological dependent variables or in the composite index of impairment through the dizzy sensation—indicating that the drop-out did not lead to a biased sample.
|
Materials
Medical Assessment
The medical assessment included a careful medical history, a general clinical examination, and a standardized bedside neuro-ophthalmological and neuro-otological test battery including electroystagmography and caloric testing (1). On hospital admission, a cranial magnetic resonance imaging (MRI) scan and acoustic evoked potentials were additionally performed in all patients. Participants were asked to complete a standardized questionnaire to assess duration, frequency, and intensity of their sensation of dizziness (or vertigo), and nausea. For each item, we asked for a ranking on a visual scale (range = 0–5 or 0–10). We subsequently composed an index of subjective impairment due to dizziness/vertigo consisting of the sum of the three items. For this index, all items were transformed onto a 0 to 10 scale to maintain equity across duration, frequency, and intensity estimations. The nausea ratings were not considered in the present study.
The diagnosis of VN were mainly based on the history of acute onset rotational vertigo lasting for at least 2 to 3 days, and, on examination, characteristic contraversive horizontal-rotational spontaneous nystagmus, postural imbalance with falls to the side of the affected labyrinth in balance tests, ipsilateral pathological VOR response in the Halmagyi-Corthoys head impulse test, and a significantly impaired VOR response of the affected labyrinth in caloric testing.
BPPV was diagnosed based on a history of brief attacks of positional vertigo and the characteristic provocation of vertigo and a crescendo-decrescendo nystgmus in the lateral head-trunk tilt maneuver. The linear component of the nystagmus had to be directed toward the undermost ear occurring after a latency period of a few seconds (12). Patients were excluded if they did not meet these criteria or showed abnormalities of hearing or other nonvestibular function or lesions of the central nervous system on MRI scans.
Psychological Assessment
The psychological assessment consisted of a structured interview for mental disorders (Diagnostic Interview for Mental Disorders, DIPS) (13) and a questionnaire battery. The DIPS, based on the Anxiety Disorders Interview Schedule— Revised (ADIS-III-R) (14), was slightly modified to adjust it to the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) and to assess lifetime mental disorders (as done in the ADIS-IV which was not available in German at the time of the study). DIPS interviews were conducted by MA students (C.E., S.E.), both of whom received intensive training in the use and scoring of the instrument. Each case was reviewed by the study director (N.H.), representing the psychological expertise. In addition to the interview, all patients completed a questionnaire package consisting of the questionnaires indicated below.
Beck Anxiety Inventory (BAI)
The BAI is a 21-item scale that measures the severity of anxiety symptoms (e.g., numbness or tingling, heart pounding or racing, shaky, feelings of choking) (15,16). The BAI has a high internal consistency of 0.92 and sufficient validity data.
Anxiety Cognition Questionnaire (ACQ)
The ACQ is a 14-item questionnaire used to assess anxiety cognitions, e.g., fear of dying by myocardial infarction or loss of control (17,18). The internal consistency of the German version is 0.75.
Body Sensations Questionnaire (BSQ)
The BSQ is a 17-item questionnaire used to assess the anxiety with regard to bodily symptoms and has an internal consistency of 0.85 (German version) (17,18).
Mobility Inventory (MI)
The MI, consisting of 27 items, assesses the patient's avoidance behavior with regard to the most common agoraphobic situations (18,19). Patients rate the avoidance of the situations twice, while they were alone (MIA) and while accompanied by another person (MIB). Internal consistencies for the German version are 0.97 and 0.96, respectively.
Beck Depression Inventory (BDI)
The BDI is a 21-item self-report questionnaire used to assess the severity of depression (20,21). Common depressive symptoms and attitudes are assessed. The BDI is the most frequently used measure of patient improvement in psychotherapy outcome studies. Internal consistency for the German version is 0.88 and sufficient validity data are provided.
Symptom Checklist-90-R (SCL-90-R)
The SCL-90-R is a 90-item questionnaire assessing nine primary symptom dimensions and a Global Severity Index (GSI), based on all 90 items (22,23). The GSI is used to measure the intensity of perceived distress. Internal consistency for the German version's GSI is 0.97.
Design and Procedure
All patients were psychologically and medically assessed twice—initially after admission within 10 days after the onset of vertigo/dizziness and on follow-up about 3 months later.
The first psychological assessment occurred on average 6 days after admission. The medical assessment was always conducted on the first 2 days after admission (except caloric tests and MRI). The second assessment was conducted in the context of an outpatient clinic appointment to which all patients were invited. After completing the assessment, all participants were informed about their diagnosis, including the underlying organic pathophysiology, and possible treatments by one of the treating neurologists (C.M. or M.M.M.). We stressed the relatively benign course of the disease with a good chance of full recovery within a few weeks. Patients with VN were initially treated with a 3- to 5-day course of corticosteroids (24), antiemetics if required, and instructed physical exercises (25). BPPV patients were trained by a physiotherapist to perform a liberation maneuver (either the Epley or Semont maneuver) at least three times a day until cessation of vertigo (26).
Data Analysis
We established the frequency of the following four groups at time point 2, i.e., 3 months after the initial admission: a) no vestibular functional impairment on medical examination/no sensation of dizziness (RC for recovered group); b) no vestibular functional impairment/continued complaints of dizziness (CC for continuously complaining group); c) vestibular functional impairment/no sensation of dizziness; and d) vestibular functional impairment/continued complaints of dizziness.
For the first hypothesis, descriptive statistics for all questionnaires were calculated for the first and second assessment dependent on the presence or absence of continuous complaints about dizziness at the second assessment. In addition, for lifetime and current diagnoses, cross-tables and
2 tests were used to compare the CC group with the RC group.
For the second hypothesis, multiple regressions were conducted with the BSQ, and the ACQ, respectively, and the vestibular disorder at time 1 as well as their interaction as predictors and the subjective impairment index due to dizziness at time 2 as the criterion.
| RESULTS |
|---|
|
|
|---|
|
Hypothesis 1: Are those who Continued to Complain About Dizziness in General More Psychologically Impaired at the Time of Admission?
Questionnaire Data
A multivariate F test including general psychological distress (SCL-90-R), depressive symptoms (BDI), anxiety symptoms (ACQ, BSQ, BAI) and avoidance (MI) yielded a nonsignificant result, F(7, 35) = 1.4; p > .21;
2 = 0.23. Contrary to our expectations, neither the ACQ nor the BSQ significantly differed between groups at the time of admission (Table 3).
|
Three months later, both groups differed significantly across all dependent variables, F(7, 35) = 4.4; p < .001;
2 = 0.47. Univariate follow-up tests supported the significant difference for each dependent variable. These results remained similar when the analyses were repeated with nonparametric tests due to violations of assumptions of variance analyses (e.g., homogeneity of variances).
Mental Disorders
The CC group had a higher prevalence of mental disorders (lifetime). Both the proportion of past as well as current diagnoses were slightly higher in this group, although the small sample size limits the interpretation of results. The lifetime prevalence (current and past history, comorbidity not considered, i.e., the presence of one disorder counts once, subsequent disorders are not counted anymore) was 43% in the RC group and 77% in the CC group, which was significantly different,
2 (1) = 4.1; p < .05. A separation by type of disorder demonstrates that current disorders are primarily related to anxiety although past disorders were primarily related to depression, independent of group membership (RC versus CC; Table 3). During the 3 months until the second assessment, no one with a current mental diagnosis recovered. Six individuals developed new mental disorders (incidence rate of 22%). No difference between the CC and RC groups occurred in incidence rates.
Hypothesis 2: Do the Fear of Bodily Sensation or the Cognition About Anxiety Symptoms at Time of Admission Predict the Extent of Phobic Dizziness at Time 2 in VN and in BPPV?
Table 4 shows the correlation between the different dependent variables. Two multiple regression analyses tested the second hypothesis. Each time, the psychological variable (BSQ or ACQ) and the medical variable (type of vestibular disorder) as well as the interaction of the two variables were entered as predictors and the degree of continued dizziness 3 months later was the criterion. The other dependent variables were not included here because no specific hypotheses were formulated and the sample size did not recommend a larger number of predictors.
|
The BSQ did not significantly predict the persistence of dizziness; however, the type of vestibular disorder (VN versus BPPV) as well as the interaction yielded significant results, F(3, 42) = 3.5; p < .03; R2 = .21 (Table 5). To shed light on this interaction, the regression was independently repeated for each vestibular disorder. Interestingly, the BSQ was now a significant predictor in patients with VN, F(1, 23) = 9.7; p < .01; R2 = .31, but not in patients with BPPV, F(1, 18) = 0.001; p > .95, R2 = .00.
|
Similarly, the ACQ did not significantly predict the persistence of dizzy feelings but, in contrast to the BSQ, neither the type of vestibular disorder nor the interaction between both variables demonstrated predictive power, F(3, 42) = 1.7; p > .18, R2 = .12.
Because VN patients reported significantly more severe initial dizziness at time of admission (VN: 27.2 ± 1.8; BPPV: 20.6 ± 1.6; t(41) = 12.5; p < .001), we repeated the regression analyses for the total sample (BPPV + VN) and included the severity of initial dizziness as a predictor (as well as the interaction between initial dizziness and the BSQ). The only significant predictor for the persistence of dizziness was the interaction between initial subjective severity of dizziness and the BSQ, F(5,42) = 71.1; p < .001; R2 = .91 with the interaction yielding ß = 1.0; t = 15.7; p < .01.
| DISCUSSION |
|---|
|
|
|---|
Hypothesis 1: Psychological Impairment at Time of Admission
In our analysis, we compared patients who fully recovered from their initial dizziness (RC) and patients with continuous complaints of dizziness (CC). In both groups, patients were psychologically distressed to some extent at the time of hospital admission. No difference in employed psychological questionnaires occurred at this time between these two groups although the effect size indicated a large effect (
2 = 0.23). The nonsignificant finding on statistical examination might therefore also be due to low power for detecting this effect. Nevertheless, these findings are in line with the recent results reported from Best and colleagues (11), who also failed to find clinically elevated scores on psychological dimensions in patients from these two types of vestibular disorders.
The clinical evaluation with a structured interview for mental disorders, however, demonstrated a difference between these groups in their mental health. The CC group had a higher percentage of mental disorders in their past. All CC patients had a past experience of a depressive disorder. Thus, the CC group represents a recovered depressed group, which explains the BDI score at the time of hospital admission—it represents a typical score for a not clinically depressed group (27). One might conclude that depressive episodes in the past may possibly make individuals more prone to processing the experience of a vestibular disorder in a threatening way. Caution in overinterpretating the results are necessary though, due to the small sample size, and the fact that part of the RC group also reported depressive episodes in the past (as would be expected because depression is among the most frequent mental disorders in the general population). In general, the lifetime prevalence rates (43%) in the RC group are equivalent to lifetime prevalence rates reported in the literature for the general population in Germany which vary across epidemiological studies from 31% to 48% (28). In contrast, the CC group yielded higher prevalence rate than the general population (77%). Again, these results support the assumption recently expressed (11) that "preexisting psychopathological personality should be considered pathogenic factors in any linkage between organic and psychometric vertigo syndromes" (p. 658).
Hypothesis 2: Prediction of Continued Dizziness 3 Months Later
For the second hypothesis, we found only partial evidence. Whereas the BSQ had—as expected—significant predictive power for identifying persistent (phobic) dizziness, the ACQ lacked this power. However, this pattern was only convincingly established for patients with VN but not for patients with BPPV. This is in line with the results of the study by Godemann and co-workers (8), who also demonstrated that the BSQ is a better predictor 10 days after admission than the ACQ in a sample of VN patients.
In clarifying this type of vestibular disorder and its interaction with the BSQ, we discovered that VN patients reported on average a more severe vertigo/dizziness at the time of hospital admission than the BPPV patients. Thus, we attempted to elucidate if this predictive value of the interaction is based on the type of vestibular disorder or on the initial severity of the dizziness. It became evident that it is the initial subjective severity of dizziness, and not the diagnosis, which is more relevant for the prediction. Another interpretation that may explain the lack of prediction of persistent dizziness through fear of bodily sensations in BPPV patients is the amount of control they may have on their symptoms. In general, patients with BPPV quickly realize that they can prevent acute vertigo and nausea by avoiding head movements in the functional plane of the affected semicircular canal whereas in the VN patients, vertigo/dizziness initially may continue even when they keep their head still. Control is an important psychological construct for the explanation of anxiety and its disorders (29). Thus, future studies should assess this variable as well to exclude (or include) this dimension as important for these patients.
We conclude that there are around 30% of individuals who will continue to complain about dizziness even after 3 months. At that time, the underlying organic disorder recovered functionally in almost every patient. This group therefore presents with persistent dizziness best described as a secondary psychogenic (or phobic) dizziness. Medical settings in which these patients with peripheral vestibular vertigo are initially seen and treated should therefore consider the patients' mental health state, and specifically their fear of bodily sensations. Patients with a secondary psychogenic dizziness pose a significant challenge in clinical practice. In specialized dizziness clinics, patients with psychogenic dizziness including phobic dizziness/vertigo often present with a history suggesting an earlier episode of an organic vestibular vertigo. Furthermore, in these clinics, psychogenic dizziness/vertigo represents the second most common diagnosis (30), and the most common cause in the age range between 20 and 50 years (31).
The clinical implication of the present findings could be to identify those patients at risk to develop secondary psychogenic dizziness at the very early stage when they seek help with an acute peripheral vestibular disorder and to offer them a brief psychological intervention to prevent the adverse course. Yardley and colleagues have worked on developing and applying a vestibular rehabilitation training which might address this issue (7,32). Based on the present results, the screening procedure could be fairly easy: 1) apply the BSQ within the first week of the vestibular episode (but not on the first day (8) and 2) assess the subjective severity of dizziness on the first days after admission by requiring the patient to complete a rating of the intensity, frequency, and duration of dizziness/vertigo. Extending the results of Godmann and colleagues (7), it would not be necessary to limit such a preventive intervention to patients with vestibular neuritis. The same results may apply for BPPV, and probably also for other causes of acute vestibular vertigo. We did not find a qualitative difference between these two types of vestibular disorders, which we analyzed in our study with respect to their impact on psychological variables. The impact is rather determined by the interaction of the individual's fear of bodily sensation and the subjective severity of dizziness at the time of hospital admission.
The present study is limited in several ways. The sample size is relatively small, which complicates the inclusion of other potentially relevant predictors, such as the BDI.
A power analysis conducted with G*Power 3 (33) demonstrated that the power for the BSQ to be identified as a significant predictor was 0.78 compared with only 0.47 for the ACQ. This is due to the larger effect size found for the BSQ (large) compared with the ACQ (small to medium). It again emphasizes, however, that there may be more predictors relevant than only the BSQ. Related to this issue, the decision to include the 9% of individuals who still met partial functional impairment organically to the phobic dizziness group is critical. The present sample size did not allow a separate analysis of this group. Future studies, however, should be aware that there may be a small number of patients belonging to this type of group.
Attrition also causes a threat to the interpretation of the results although the initial sample and the finally included sample did not differ significantly from each other in sociodemographic or psychological measures.
In conclusion, the present study suggests three important points. First, persistent complaints about dizziness after an acute vestibular episode are more often determined by psychogenic factors rather than by persistence of the original somatic-functional deficit. Second, the stronger an individual who suffers from dizziness originally organically triggered perceives the sensation at the time of hospital admission and the higher the fear of body sensations within the first 10 days after admission, the higher the risk for developing a secondary persistent dizziness. It would be interesting to determine in a future study if the subjects' heightened fear of bodily sensations predated the onset of the peripheral vestibular disorder, or became elevated in reaction to the acute disorder. Third, lifetime mental disorders are prevalent in these patient groups but they do not predict the onset of persistent dizziness. Moreover, patients with vestibular disorders primarily developed anxiety-related disorders within 3 months after the onset of an acute vestibular disorder. This is in line with the interpretation that this type of processing is a response to the experience of the vestibular episode (6). Thus, the experience of dizziness in the context of a vestibular episode may be considered a risk factor for the development of anxiety disorders for all individuals. Moreover, the BSQ turned out to be of primary significance because the instrument is able to predict the adverse psychological development (in contrast to the ACQ) but only in interaction with the initial severity of the dizziness. Thus, both parameters must be evaluated to identify patients at-risk for psychogenic dizziness. Godemann et al. (7) found a similar significance of the BSQ in their study with VN patients. They have not considered the initial severity of dizziness, however. Their study as well as the present study demonstrate that the tendency to interpret bodily sensations as dangerous symptoms make individuals more prone to continuously complain about dizziness despite recovery after 3 months of experiencing a vestibular episode. Developing and testing appropriate brief behavioral interventions for these individuals are an important next step.
| NOTES |
|---|
|
|
|---|
DOI:10.1097/PSY.0b013e318151a4dd
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |