Psychosomatics 42:29-34, February 2001
© 2001 The Academy of Psychosomatic Medicine
Correlation Between Somatic Sensation Inventory Scores and Hyperarousal Scale Scores
Mustafa A. Hammad, D.O.,
Arthur J. Barsky, M.D., and
Quentin R. Regestein, M.D.
Received December 17, 1999; revised March 3, 2000; accepted September 6, 2000. From Department of Medicine, State University of New York at Stonybrook; Department of Psychiatry, Brigham & Women's Hospital, Boston, Massachusetts. Address correspondence and reprint requests to Dr. Regestein, 75 Francis St., Boston, MA 02115; e-mail: qregestein{at}partners.org
|
ABSTRACT
|
Somatization mechanisms are poorly understood. The authors tested whether somatization might involve altered central nervous system information processing. They measured somatization using the Somatization Sensation Inventory (SSI) and information processing style using the Hyperarousal Scale, scores of which correlate with electroencephalogram(EEG) measures of cortical electrical responsiveness. SSI scores correlated highly with Hyperarousal scores. On logistic regression, two SSI items and two Hyperarousal items accounted for most of this correlation. These specific hyperarousal items had previously been found to covary with EEG activity and cortical evoked potential amplitudes. The authors concluded that somatization may involve altered CNS processing of somatic stimuli.
Key Words: Somatization • Hyperarousal • Information Processing
|
INTRODUCTION
|
Somatization results in a person persistently seeking help for medically unexplained bodily symptoms that interfere with social or economic life. The mechanisms of somatization are unknown. Somatization becomes somatization disorder (SMD) when the person continually complains of symptoms in several domains—nervous, cardiovascular, gastrointestinal and others. Somatization typically begins between the ages of 20 and 30. Lifetime prevalence ranges from 0.2% to 2% in women and less than 0.2% in men.
Somatization has a high social cost. About 50% of primary care visits, an estimated 20% of health care costs, and many of the 30%–80% of medical patients who have physical symptoms without organic cause are attributed to somatization.1 In one series, 42% of SMD patients were misdiagnosed in their first medical visit, with 24% of them diagnosed as depressed. Chronic somatizers disproportionately consume clinical appointments, diagnostic tests, and medical and surgical procedures up to 9 times the average American's health care cost.2–5 Such uncontrolled suffering and expense warrant further clarification and more specific treatment.
Barsky16 introduced a widely accepted hypothesis that somatization involves "misinterpretation" and "somatosensory amplification" of bodily symptoms. The distressed patient may use somatization to seek help. A presumed innate trait causes certain stimuli to be amplified. Somatization involves focusing on and perceiving bodily sensation as abnormal, intense, disturbing, and noxious. Barsky hypothesizes that amplification begins when patients misinterpret a transient medical problem or a benign sensation as serious and harmful. These benign symptoms become exaggerated enough to be disturbing. This process can transform physiological sensations into perceived pathology. Somatizers start to believe that they are ill, which leads to further symptom amplification.
Somatization becomes chronic because somatizers selectively look for information that reinforces their illness beliefs, such as media information or advertisements, but they block out information that contradicts these beliefs, such as negative laboratory tests or physician reassurances. This aggravates their anxiety and thus intensifies their symptoms.
Without a clear cause, there can be no specific treatment for somatization. Barsky recommends physicians try to understand patients' complaints and establish a trusting relationship by communicating with and educating them. Barsky's theory conceives of somatization as multiple reinforced exaggerations, but it does not clearly account for the initial misinterpretation itself.
We developed a hypothesis to explain the origin of somatosensory misinterpretations. In our hypothesis we take into account central nervous system (CNS) mechanisms for processing types of information that potentially predict somatization. To find whether somatization might be associated with altered information processing, we use a pilot study to test whether a measure of somatization correlates with a measure that predicts cortical arousal.
|
METHODS
|
We used the Somatic Sensation Inventory (SSI) developed by Barsky and Wyshak17 (see Table 1) to measure somatization. We used the Hyperarousal Scale, a correlate of electroencephalogram (EEG) arousal, to measure information processing. The Hyperarousal Scale is a self-report of symptoms initially described in insomnia patients (Table 2). Hyperarousal Scale scores correlate with several EEG measures of arousal, including frequency spectral and evoked potential measures. Hyperarousal scores in the high range signify increased general cerebral responsiveness but decreased selective attention.6 Hyperaroused subjects may thus be more open to stimuli but less likely to distinguish meaningful from random stimuli.
If SSI scores correlate with Hyperarousal scores, then distinctions between relevant and irrelevant stimuli would presumably be less evident to somatizers. Somatizers would be less likely to distinguish between physically harmless and threatening sensations. Impaired sensory filtering would increase hyperaroused persons' information load and make them less able to habituate to repeated stimuli. The decrease in selective attention potentially obscures the meaning of percepts. The resulting ambiguity may augment the recruitment of brain adversity management systems in the responsive person.
Our current study uses baseline data from a clinical trial of hypnotic drugs in primary insomnia subjects. Subjects were unaware of criteria for the drug study, which had occurred before the present study was planned.
This post-hoc study uses screening data not originally intended as outcome measures. Subjects fulfilled subjective criteria for primary insomnia, as defined in the DSM-IV. Primary insomnia has a somatization component because insomnia subjects most often have symptoms not adequately explained by objective tests. They show few or no abnormalities on polygraphic sleep recordings, or else generate records that insufficiently explain the magnitude of their insomnia symptoms.7–9
Subjects and Screening Criteria
Procedures. Subjects were respondents to advertisements who indicated that they usually took longer than 30 minutes to fall asleep after lights out, had a total night sleep of 4–6 hours at least 3 nights per week, and experienced sleep that was inadequate to prevent daytime fatigue or support daytime functioning. They denied having any of the following problems that potentially worsen sleep: a medical or psychological disorder, the use of more than 10 cigarettes daily, the use of hypnotic or experimental drugs within 30 days before screening, the use of psychotropic drugs, a history of seizures or traumatic head injury; drug or alcohol abuse within the previous year, or a shift-work schedule change within the past 6 months.
After signing the consent form approved by an Institutional Review Board, subjects completed the Hyperarousal and SSI scales. Each subject's subjective sleep times were verified by a diary they filled out on each of 7 consecutive placebo nights. Those who completed the entire study were paid $200.
At entrance to the study, all subjects had a body weight no more than 25% outside the ideal range, and good physical and mental health according to routine medical and psychiatric evaluations. There were 58 subjects (34 women and 24 men, median age=43, range 22–65) with complaints of chronic, primary insomnia. Psychiatric evaluation involved a 30-min interview by a psychiatrist (QR) but no standardized rating scales. This evaluation was meant to screen for a DSM-IV Axis I disorder, other than insomnia, and for inability to complete the protocol or communicate with staff. This procedure probably detects more depression than would routine clinical practice.10
Study Measures
The Somatic Sensation Inventory (SSI). The SSI is a 26-item self-report inventory of body sensations, symptoms and subjective overall health.11 Responses are scored from 1 to 5 where 1=Not At All, 2=A Little Bit, 3=Moderate, 4=Quite A Bit, and 5=A Great Deal. We compared SSI scores between completers and excluded insomnia subjects.
Hyperarousal Scale. The Hyperarousal Scale is a 26-item self-report that measures tendencies to introspect, think about feelings, respond intensely to unexpected stimuli, and other behaviors that putatively involve cortical arousal.12 Responses to each Hyperarousal Scale item were graded from 0 to 3, respectively, where 0=Not At All, 1=A Little, 2=Quite A Bit, and 3=Extremely.
Extreme Subscale. The responses marked Extremely were summed into a separate Extreme subscale score because we thought the response Extremely seemed categorically different from the alternative choices. We analyzed total Hyperarousal score and Extreme subscale score both before and after excluding from the analysis items that referred to patients' sleep, namely, "I have trouble falling asleep." Extreme scores correlate negatively with the sum of auditory evoked potential abnormalities (R= -0.31; P=0.015).13
Analysis
Total Hyperarousal, Extreme subscale, and SSI scores (summed item scores) were analyzed using the SAS statistical program (SAS Institute, Cady, NC).
Correlation was sought between SSI scores and Hyperarousal scores. We used a stepwise regression for individual SSI items against the total SSI score. Logistic regression was used to analyze the nonparametric Hyperarousal Extreme score.
|
RESULTS
|
Mean SSI score per scale item for the current population was 1.55 (SE=0.047), significantly higher than the 1.41 (SE=0.035) reported for a normal population11(T=2.16; P=0.03). SSI scores correlated with Hyperarousal scores (Pearson correlation r=0.44; P=0.0005). Four Hyperarousal items jointly accounted for 36% of SSI score variance. These were the following: "I cannot take naps even if I try," "Alcohol makes me sleepy," "I keep thinking about some things long after they have happened," and "I take things personally." No major predictor was omitted from the model, as indicated by a Mallow's C (p), statistic of -8.9. Stepwise regression of SSI items on SSI Score revealed two items that each accounted for more than 10% of the total SSI variance: "Trouble catching your breath" (72%) and "Not feeling well most of the time in the past 5 years" (13%). These two SSI items combined would make an efficient index for the SSI score in this population. On stepwise logistic regression, two Hyperarousal items predicted this two-item SSI index (Model score=16.9; P=0.0007), namely, "I take things personally" (Wald  2=11.2; P=0.0008; odds ratio=0.38) and "I keep thinking of things long after they have happened" (Wald 2=4.2; P=0.04; odds ratio=0.47).
SSI score also correlated with the Hyperarousal Extreme score (Model 2=19.3; P=0.0002). This correlation was due to SSI items, "Not feeling well most of the time" (Wald 2=8.8; P=0.003) and "Feeling fatigued, weak, or tired all over" (Wald 2=4.4; P=0.04).
|
DISCUSSION
|
The results of our study suggest that SSI scores correlate significantly with Hyperarousal scores in this population. Because Hyperarousal scores have been shown to correlate with EEG arousal measures, our results were in accordance with our hypothesis that somatization involves abnormalities of cortical arousal.
Previously, we found evidence that high Hyperarousal scores may involve altered information processing mechanisms. In a pooled group of 20 normal subjects and 20 insomnia subjects, in the eyes-open condition, we observed that Hyperarousal scores correlate with total EEG activity (r=0.42; P=0.003).14 Because total EEG activity disproportionately represents slow-wave activity, subjects who have high Hyperarousal scores probably have increased EEG slow-wave activity during wakefulness, which potentially impairs their information processing.
The above observations provide indirect evidence that somatizing involves impaired information processing. This evidence can only foster hypotheses rather than support conclusions about somatization mechanisms.
As mentioned above, Barsky hypothesizes that chronic fatigue syndrome exemplifies somatization. Among 40 patients who met CDC criteria for chronic fatigue syndrome, Pavlova et al.15 observed that age-adjusted Hyperarousal scores correlate significantly with the P300 event-related EEG potential (R=-0.47; P=0.003). The P300 measures the updating of working memory, possibly by clearing information processing mechanisms to prepare for new information processing.
Online working memory updates the context of processed stimuli. People with impaired working memory, as indicated by high Hyperarousal scores, may be less able to differentiate between new and old information. They would be slower to update their view of the world; that is, they would be more likely to develop an inaccurate concept of the real world. Somatizers are therefore less able to readjust their worldview because of being less able to pick out relevant information from among incoming stimuli (i.e., somatizers are more likely to stick with an outdated worldview, regardless of what information is presented).
Total SSI scores were largely accounted for by just two items: "Trouble catching your breath" and "Not feeling well most of the time for the past 5 years." In turn, the combined score of these items was predicted by two Hyperarousal items: "I take things personally" and "I keep thinking of things long after they have happened." Previous observations showed that each of these items correlated with specific EEG variables. We can discern some hint of somatization mechanisms by considering these correlations.
In a pooled group of 20 normal and 20 insomnia patients, the "Personally" item correlated with EEG alpha activity (partial r2=0.25; F=10.2; P=0.003). This occurred in the eyes-open state, when visual processing would normally block alpha activity. The "I keep thinking" item contributed to several three-variable models of nonalpha activity in both eyes-open and eyes-closed conditions.14 Because nonalpha mostly represents slow-wave activity, subjects who have high scores on the "I keep thinking" item probably have a lower level of selective responsiveness. To summarize, the two SSI items that largely accounted for the total SSI score correlated with EEG variables to suggest that somatizers, when they process information, recruit relatively less perception-specific neural activity. Their background EEG rhythms are also slower, indicating a lower level of arousal.
Even before media statements or insurance payments reinforce the somatizers' specific bodily sensations, they may not distinguish well what somatic information is generally relevant or irrelevant to their interest. This lower capacity to distinguish relevant from irrelevant information accords with a previously observed inverse correlation between Hyperarousal scores and EEG event-related potential amplitude.15
The current data suggest that a substantial part of the variance in a measurement of somatization (about 20%) may be accounted for by hyperarousal. The Hyperarousal Scale items that independently predicted SSI score, (i.e., taking things personally, thinking of things long after they have happened) seem consistent with somatizers' tendency to connect too many stimuli with issues of personal survival and to tenaciously maintain rather than update their worldview.
In our study, we used retrospective data to test our specific hypothesis. Several weaknesses of the study make the reported observations preliminary. First, the Hyperarousal questionnaire measure is a proxy for the EEG measures against which it was validated and not the measures themselves. Second, we used a population of experimental subjects who had some features of somatization but were not specifically identified as manifesting somatization. However, they did have abnormally higher SSI scores, as might be predicted in chronic insomnia.
The current results may not generalize to a population of individuals who meet formal criteria for somatization and who would likely have higher SSI scores. Nevertheless, the results suggest that somatization is associated with altered information processing and that the mechanisms of somatization may be further clarified by the study of information processing in somatizers. The current results justify subsequent study to directly test whether EEG measures in somatizers to show evidence of altered information processing.
|
REFERENCES
|
-
Kellner R: Somatization: theories of research. J Nerv Ment Dis 1990; 178:150–160[CrossRef][Medline]
-
Coryell WH, Norton SG: Briquet's Syndrome (somatization disorder) and primary care depression: comparison of background and outcome. Comprehens Psychiat 1981; 22:249–256
-
Smith GR, Monson RA, Ray DC: Patients with multiple unexplained symptoms: their characteristics functional health, and health care utilization. Arch Intern Med 1986; 146:69–72[Abstract/Free Full Text]
-
Zoccolillo MS, Cloninger CR: Excess medical care of women with somatization disorder. South Med J 1986; 79:532–535[CrossRef][Medline]
-
Swantz M, Blazer D, George L: Somatization disorder in a community population. Am J Psychiat 1986; 143:1403–1408
-
Regestein QR, Dambrosia J, Hallett M: Daytime alertness in patients with primary hypersomnia. Am J Psychiat 1993; 150:1529– 1534
-
Mendelson WB, James SP, Garnett D: A psychophysiological study of insomnia. Psychiat Res 1986; 19:267–284[Medline]
-
Hauri PJ: Cognitive deficits in insomnia patients. Acta Neurol Belg 1997; 97:113–117[Medline]
-
Sugerman JL, Stern JA, Walsh JK: Daytime alertness in the subjective and objective insomnia: some preliminary findings. Biol Psychiat 1985; 20:741–750[CrossRef][Medline]
-
Wells KB, Hays RD, Burnham A: Detection of depressive disorder for patients receiving prepaid or fee for service care. JAMA 1989; 262:3298–3302
-
Barsky AJ, Wyshak G, Herman GL: Hypochondriasis: an evaluation of the DSM-III criteria in medical outpatients. Arch Gen Psychiat. 1986; 43:493–500.
-
Regestein Q, Pavlova M, Casares F: Validation of the hyperarousal scale in primary insomnia subjects. Sleep Res 1996; 25:344
-
Pavlova M, MacAnulty G, Duffy F: Auditory evoked abnormalities associated with hyperarousal in patients with insomnia or chronic fatigue. Sleep Res 1997; 26:33
-
Pavlova M, Regestein Q: Self-reported behavioral correlations of total EEG power and cortical evoked potentials in insomnia and control subjects. Sleep Res 1997; 26:465
-
Pavlova M, MacAnulty G, Duffy F: P300 evoked potential and hyperarousal behavioral traits in chronic fatigue syndrome patients (abstract). Sleep Res 1997; 26:34
-
Barsky AJ, Klerman GL: Overview: hypochondriasis, bodily complaints, and somatic styles. Am J Psychiatry 1983; 140:273–283[Abstract/Free Full Text]
-
Barsky AJ, Wyshak G: Hypochondriasis and somatosensory amplification. Br J Psychiatry 1990; 157:404–409[Abstract/Free Full Text]
Get information about faster international access.
Privacy Policy
Copyright © 2001
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
