| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Institute of Psychology, J. W. Goethe-Universität, Frankfurt am Main, Germany.
Address correspondence and reprint requests to Ursula Voss, Institute of Psychology, J. W. Goethe-Universität, Mertonstr. 17, 60054 Frankfurt am Main, Germany. E-mail: Voss{at}psych.uni-frankfurt.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: The study compares 37 primary insomniacs (DSM IV criteria) and 47 good sleepers. Dependent measures included self-rating scales concerning sleep quality (Schlaffragebogen part B, Frankfurter Schlaffragebogen), coping styles (Frankfurt Monitoring Blunting Scales), and psychopathology (Neo Five Factors Inventory, Beck Anxiety Inventory). All measures had documented psychometric properties.
Results: Primary insomniacs were significantly more likely to rigidly resort to monitoring strategies in controllable as well as uncontrollable situations. Further, 73% of all rigid monitors identified in the sample were rated as primary insomniacs, whereas 86% of all rigid Blunters were good sleepers. Insomniacs showed higher levels of anxiety than good sleepers, coping style groups differed in anxiety and neuroticism scores.
Conclusion: The study further supports the hypothesis that a monitoring coping style is related to primary insomnia. Blunting appears to be a good predictor of high sleep quality. The results are discussed with regard to improving treatment for patients suffering from primary insomnia.
Key Words: sleep • monitoring • blunting • primary insomnia • cognitive models of insomnia • hyperarousal • coping styles
Abbreviations: DSM-IV = Diagnostic and Statistical Manual, Fourth Edition, Revised; SF-B = Schlaffragebogen part B; FSF = Frankfurter Schlaffragebogen; FMBS = Frankfurt Monitoring Blunting Scales; NEO FFI = Neo Five Factors Inventory; BAI = Beck Anxiety Inventory; TST = total sleep time; Mdn = Median; MANOVA = multivariate analysis of variance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Whereas stressful life events often precipitate and initiate a sleep disturbance, recent findings suggest that the appraisal of stress and the perceived lack of control over stressful events rather than the number of stressful events per se, enhance the vulnerability to insomnia (5). More specifically, the protective field model (6) defines sleep as an uncontrollable, potentially threatening, and thereby stressful situation. Taking an evolutionary approach to sleep behavior, the model is based on polysomnographic and behavioral data obtained from animal and human sleep research. According to the protective field model, insomniacs are characterized by the perception of heightened vulnerability and endangerment during the night. This feeling of vulnerability supposedly motivates insomniacs to engage in a continuous screening of their sleep environment for possible danger cues. Monitoring for danger cues represents a coping behavior that is incompatible with sleep and further contributes to the symptoms of insomnia. The article presented here tests the hypothesis of the protective field model concerning the relationship between excessive monitoring and primary insomnia by comparing preferred coping styles in primary insomniacs and good sleepers.
Lazarus and Folkman (7) define coping as cognitive and behavioral efforts used to manage external and internal demands that are appraised as taxing or exceeding one’s resources. According to their model, it is the perception of control over a stressful situation that determines an individual’s coping behavior in a given situation. Monitoring and blunting are defined as two of several coping modes that are predominantly utilized in situations implying threat or danger. Monitoring describes an information-seeking approach, whereas blunting refers to the reinterpretation of and distraction from the encumbering aspects of a situation. Focusing on these two coping modes, exclusively, Miller (8) developed a trait-state model of coping, proposing that (1) the beneficial effect of monitoring and blunting strategies on stress reduction depend on the control contingencies of a specific situation and (2) many individuals rigidly resort to either monitoring or blunting strategies, thus disregarding the controllability of the threatening event. For example, monitoring in controllable situations increases arousal but reduces stress because it alerts the person and minimizes the maximum possible danger to the individual (8). Blunting in controllable situations is inadequate because it increases the likelihood that important danger cues are ignored. By contrast, if the situation is uncontrollable, an individual can most effectively reduce stress by engaging in blunting techniques such as distraction or reevaluation of the event. Accordingly, employing monitoring strategies in controllable situations and blunting strategies in uncontrollable situations constitutes adaptive coping (9). However, many individuals will rigidly resort to either monitoring or blunting strategies. Rigid monitors, thus, will remain alert and try to seek information in controllable as well as in uncontrollable situations, supposedly in an attempt to increase the predictability of the events (10). Rigid blunters, on the other hand, will engage in distractive strategies, regardless of whether they have an opportunity to change the outcome of a situation. Approximately 13% of the population have been found to resort to rigid monitoring and 13% to rigid blunting, and only about 5% to adaptive coping (11,12). Monitoring and blunting have been found to be unrelated to sociodemographic variables such as age, occupational and educational status, and race. With regard to gender, significantly higher monitoring scores in women have been reported for Dutch and German samples (13,14) but not for Spanish (15) or North American samples (16), suggesting that cultural factors may play a role. Factors that might motivate an individual to rigidly or adaptively employ monitoring or blunting strategies have so far not been identified. Monitoring and blunting have been found to be unrelated to psychopathology such as anxiety, depression, neuroticism, and hypochondriasis (8,14,16). By contrast, elevated ratings for anxiety, neuroticism, and psychosomatic complaints have been reported for primary insomnia patients (17,18).
The clinical relevance of the monitoring/blunting construct has been widely demonstrated, showing that both rigid monitoring and rigid blunting are often maladaptive in terms of promoting health or accelerating convalescence (19–21). The attractiveness of the monitoring and blunting concept for sleep research is supported by the circumstance that sleep presents an uncontrollable situation bearing potential danger and necessitating precautionary protective behaviors such as seeking a safe shelter in which to sleep (monitoring). The initiation and maintenance of sleep itself, however, requires that intrusive thoughts, as well as environmental, mostly acoustic, stimuli should be ignored (blunting). As a consequence, it is proposed here that continuous monitoring should be related to primary insomnia (monitoring hypothesis). This hypothesis is in agreement with most newer models of primary insomnia (22–24), claiming that presleep processes most likely associated with good sleep include dearousal of cognitive processes. Whereas these models are phenomenologically oriented, however, the current study is aimed at the identification of factors that predispose an individual to engage in intrusive thinking and high cognitive arousal. The monitoring hypothesis is supported by earlier research on healthy subjects with good sleep quality showing that rigid monitors were not able to ignore even irrelevant acoustic stimuli and experienced major difficulties falling asleep. Blunters, on the other hand, fell asleep easily and were not even disturbed by the presentation of salient acoustic stimuli, e.g., the subject’s own name (25). A survey on sleep quality of healthy subjects in common and uncommon sleep environments showed that the sleep of monitors was more easily interrupted by external events than the sleep of blunters (26). These studies suggest that individuals preferring a monitoring coping style are more susceptible to symptoms of insomnia than blunters or adaptive copers. Since the sleep quality survey on healthy subjects (26) found no group differences in sleep ratings pertaining to familiar situations, however, these findings do not necessarily permit the conclusion that monitoring is linked to primary insomnia. The present study was conducted to determine the role of a monitoring coping style in primary insomnia.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
The group of good sleepers was composed of 47 participants (23 female, 24 male, ages 20–68, Mdn = 38) who reported satisfactory sleep quality, sleep latencies <15 minutes, a maximum of one to two short awakenings/night (<15 minutes) and TST > 6 hours/night (mean = 7.53, s.e. = 0.10). Groups differed in age (t = 4.72, df = 66.63, p < .01) and gender (
2 = 0.57, df = 1, p < .05). Good sleepers did not report use of sleep-affecting medication. Individuals were not included in the study if they reported an acute illness, chronic mental or somatic illness not mentioned in the telephone interview, or if they did not fully meet the criteria for assignment to either diagnostic group.
A cover letter was sent along with the questionnaires informing participants about the aim of the study and insuring confidentiality. Informed consent was obtained, and a self-addressed envelope was enclosed in all mailings.
Dependent Measures
Assessment of relevant variables relied on a number of self-report questionnaires with sufficient to good psychometric properties (12,26–30).
Sleep Quality and Sleep Hygiene
Self-ratings of sleep-related behaviors were obtained using a standardized German sleep inventory (Schlaffragebogen part SF-B) (27). The SF-B comprises several factor-analytically validated and reliable composite scores such as sleep quality, feeling of being refreshed in the morning, emotional balance in the evening, tiredness in the evening, sleep-related psychosomatic symptoms, several waking stressors, and dream recall over the past 2 weeks (Cronbach’s 
> 0.79) (28).
Additionally, participants filled out a questionnaire assessing sleep quality in familiar and unusual sleep environments (Frankfurter Schlaffragebogen FSF). The FSF contains several items regarding sleep hygiene, regular bedtimes, alcohol consumption, or frequency of afternoon naps. Further, the FSF assesses sleep quality when at home in the absence of unusual distractions (familiar sleep environment) and when either away from home or at home being exposed to unfamiliar acoustic distractors (unusual sleep environment). Participants are asked to imagine a given scenario (dripping water tap, job interview next morning, car trouble necessitating an overnight stay at a friend’s house, noisy hotel room, double bedroom in hotel, unusual noise) and rate the sleep quality they would expect in such a situation on a 4-point rating scale (Cronbach’s = 0.93) (26).
Psychological Measures
The German version of the NEO Five Factors Inventory (NEO FFI) (29) and the Beck Anxiety Inventory (BAI) (30) were selected as psychological assessment instruments. The NEO-FFI consists of 60 items and provides composite scores for neuroticism, extraversion, openness to experience, conscientiousness, and agreeableness (Cronbach’s > 0.71) (29). The BAI contains 21 items, which are summed to yield a composite score for BAI-anxiety (Cronbach’s = 0.88) (30).
Coping Style
Coping style was assessed by the Frankfurt Monitoring Blunting Scales (FMBS, 12,14). The FMBS comprises four uncontrollable and threatening situations (e.g., in-flight disturbances) and four controllable yet encumbering situations (e.g., important job interview). Each situation is followed by eight behavioral choices. Among these behavioral responses, four items pertain to a monitoring and four to a blunting style of coping with aversive events. Participants are instructed to respond to each item on a four-point rating scale. The item scores are summed within each situational type, resulting in four composite scores for monitoring and blunting in uncontrollable and in controllable situations, each. Individuals are classified as rigid "monitors" or "blunters" or as "adaptive copers" on the basis of their scores according to the scheme outlined in Table 1. Due to the small sample size, classification of coping types was based on individual scores lying above or below the respective scale mean of a normative German sample (Cronbach’s >0.70) (12).
|
Individuals not fulfilling either the criteria for rigid monitoring, blunting, or adaptive coping are referred to as mixed types.
Data Analysis
Unless stated otherwise, participants rated as primary insomniacs and healthy controls were compared by multivariate analyses of variance (MANOVA), conducted separately for each set of related variables. Since the two diagnostic groups differed in gender and age, gender was treated as fixed factor and age (continuous variable) was included as covariate. All multivariate statistics report Pillai-Spur values. Except for the a priori hypotheses concerning coping style and insomnia, all univariate levels of statistical significance were Bonferroni corrected to account for multiple testing.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
With regard to the six scales of the SF-B, MANOVA results showed a significant effect for group (F = 9.46, df = 6, 64, p < .01) and gender (F = 2.38 df = 6, 64, p < .05). As expected, insomniacs provided lower ratings of sleep quality than good sleepers (F = 54.03, df = 1, 69, p < .01). Furthermore, insomniacs felt less refreshed after sleep (F = 9.78, df = 1, 69, p < .05). Females reported greater psychosomatic symptoms than males (F = 11.09, df = 1, 69, p < .05).
Psychological Measures (BAI, NEO-FFI)
Albeit the individual scales of the NEO FFI are considered independent, in this sample, some of the correlation coefficients between individual NEO FFI scales were substantial (e.g., extraversion and neuroticism: r = –0.58, df = 74, p < .01). Since, furthermore, BAI scores correlated significantly with several NEO FFI scales (e.g., BAI and neuroticism: r = 0.58, df = 74, p < .01), data were analyzed by MANOVAs conducted separately for diagnostic groups and for coping style groups. Results on the analyses of psychological measures in diagnostic groups yielded a significant group x gender interaction (F = 2.29, df = 6, 74, p < .05) as well as significant effects for group (F = 3.74, df = 6, 74, p < .01) and gender (F = 4.45, df = 6, 74, p < .01). Tukey-HSD post hoc procedures revealed higher ratings for conscientiousness in male insomniacs compared with either female insomniacs (p < .05) or male (p < .01) and female good sleepers (p < .05). As expected, insomniacs scored higher in BAI-anxiety (F = 7.88, df = 1, 79, p < .05) than good sleepers.
With regard to the relatedness of psychological variables and coping style, results showed a main effect for group (F = 2.65, df = 18, 216, p < .01), with significant between-subject effects for neuroticism (F = 5.22, df = 3, 75, p < .05) and BAI-anxiety (F = 4.58, df = 3, 75, p < .05). Tukey-HSD post hoc procedures revealed higher neuroticism ratings for monitors than blunters (p < .01) or adaptive copers (p < .05). Blunters had lower scores than mixed types (p < .01). Concerning BAI-anxiety scores, monitors had higher scores than either blunters (p < .01), adaptive copers (p < .05), or mixed types (p < .01).
Coping Style (FMBS) and Sleep
These analyses concern the major hypothesis of the study. It was expected that primary insomniacs would have significantly higher monitoring and lower blunting scores than good sleepers. Since the two monitoring scales, as well as the two blunting scales, of the FMBS were known to be interrelated, data were analyzed using MANOVA procedures. Results showed a significant effect for group (F = 4.12, df = 4, 79, p < .01). Table 2 shows that, conforming to the study’s hypothesis, primary insomniacs had significantly higher monitoring scores than good sleepers in both uncontrollable and controllable situations. Mean differences yielded a medium effect size (d = 0.46) for monitoring in controllable situations and a large effect size for monitoring in uncontrollable situations (d = 0.73). Blunting scores were lower for primary insomniacs only in uncontrollable situations.
|
Insomniacs and good sleepers differed significantly in their coping style preferences, yielding a higher proportion of monitors in the primary insomniac group and a higher proportion of blunters among the good sleepers (see Figure 1). A total of 73% of all monitors identified in the sample were rated as primary insomniacs, whereas 86% of all blunters were good sleepers. As for adaptive copers, all five participants (6%) in this group reported having good sleep quality. Comparable to the previously reported distribution in the literature, about 18% of participants were rated as monitors, 17% as blunters, and 6% as adaptive copers. About one third of primary insomniacs in this sample were monitors.
|
Additional Analyses
As mentioned in the Methods section, seven patients (six females, one male) from the primary insomnia group reported having been in psychotherapy to alleviate their sleep problems before this study. Of these, 2 patients were rigid monitors and five were mixed types. The exclusion of these patients did not alter significant results in the MANOVAs conducted. Exploratory t tests yielded significant differences in several variables, however, showing higher ratings for former psychotherapy patients in monitoring in uncontrollable situations (t = –3.00, df = 35, p < .01), neuroticism (t = –3.50, df = 35, p < .01) and BAI anxiety (t = –2.94, df = 35, p < .05) compared with untreated primary insomniacs. Also, former psychotherapy-treated primary insomniacs scored significantly lower on extraversion (t = 2.44, df = 35, p < .05) than untreated primary insomniacs.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
With regard to psychological measures, male insomniacs scored higher in conscientiousness than either female insomniacs or male and female good sleepers. This finding is interesting since it points at a possible trait-related difference between male and female insomniacs. However, methodological issues, most importantly the different distributions of age and gender in the two diagnostic groups, also need to be kept in mind. Albeit care was taken to incorporate both factors in the statistical analyses, the number of men in the primary insomnia group was very small, restricting a generalization of the observed gender-related differences. Future studies on gender differences in primary insomnia should further address this issue. As was to be expected, both male and female primary insomniacs had significantly higher scores for anxiety than good sleepers. This result is in agreement with a wealth of findings showing that patients suffering from insomnia are likely to also experience anxiety and other psychopathological symptoms (17,18). Surprisingly, neuroticism ratings did not differ significantly between the two diagnostic groups. However, this finding might be attributable to a loss of power due to the application of Bonferroni corrections to control the probability of a Type I error.
Seven primary insomniacs had received psychotherapy to alleviate their sleep problems in the past. The exclusion of these patients within the scope of explorative analyses did not affect group differences reported above. However, comparisons of psychotherapy-treated insomniacs with nontreated insomniacs showed former psychotherapy patients to have higher scores for monitoring in uncontrollable situations, neuroticism, and anxiety, as well as lower ratings for extraversion. However, due to the small number of therapeutically treated participants, the influence of age and gender or the relatedness to sleep data could not be assessed. Further studies should try to clarify whether these effects might be due to more severe symptoms in the former psychotherapy patients or whether psychotherapy might have detrimental effects on sleep-related variables (e.g., by putting strong emphasis on monitoring behavior independent of the controllability of the situation).
Analyzing coping style by looking at monitoring and blunting as continuous variables revealed that primary insomniacs had significantly higher monitoring scores in both controllable and uncontrollable situations than good sleepers. In accordance with previous studies (12,14) and with theory (8), effect size was greatest for uncontrollable situations, indicating that participants with primary insomnia are unable to refrain from information-seeking behaviors even in situations where this behavior is to be considered dysfunctional. Using a categorical approach, primary insomniacs were much more likely to be classified as rigid monitors than good sleepers, confirming the hypothesis that monitoring is related to insomnia, whereas blunting seems to be positively associated with good sleep. With regard to adaptive coping, results suggest that it is as conducive to sleep as blunting. However, the small count of adaptive copers in this sample restricts a more elaborate interpretation of these data. Concerning psychological measures, results concerning a monitoring coping style were similar to those found for primary insomniacs in terms of elevated ratings in BAI anxiety. However, in addition to higher anxiety levels, monitors scored higher in neuroticism than blunters or adaptive copers. This finding is in disagreement with previous studies which reported primary insomnia (17,18) but not coping style (in healthy subjects) (12,14,26) to be associated with neuroticism. Albeit sample size did not permit a MANOVA to explore a possible interdependence of group and coping style, these data are suggestive of a third variable that might moderate both primary insomnia and coping style. Possibly, primary insomniacs who are monitors suffer greater psychological distress than primary insomniacs who are not monitors. Future research should address this question.
Concerning the study’s hypothesis about the role of monitoring in the initiation and maintenance of primary insomnia, the study established a close link between insomnia and a monitoring coping style. Further, blunting, as well as adaptive coping, seems to be conducive to satisfactory sleep. However, the study’s correlational design does not allow one to draw inferences about the direction of causality. It is theoretically possible, for example, that the development of primary insomnia might be a precipitating factor for the development of a monitoring coping style. The psychophysiological data on evoked potentials and monitoring and blunting in healthy, non–sleep-disturbed subjects (25) make this interpretation seem unlikely. Further, laboratory data on the effects of sleep disruptions on non–sleep-deprived monitors, blunters, and mixed types (6), showing that the sleep of monitors is more easily interrupted than the sleep of blunters and mixed types, rather support the opposite mode of operation.
Thus, the results of the current study suggest that a monitoring coping style represents a predisposing factor in the genesis of primary insomnia. Moreover, it seems that whereas not all insomniacs are monitors (30%), most monitors (73%) experience sleep problems. With reference to the protective field model, this interpretation implies that rigid monitors have difficulty initiating and maintaining sleep because they excessively screen their sleep environment in search for information related to possible threats to their perceived safety. This specific attribution, however, has yet to be tested.
As for clinical implications, it might be helpful to include psychoeducational elements concerning coping with controllable and uncontrollable situations and monitoring versus blunting coping styles into the treatment of primary insomnia. Of special interest is the fact that the adaptive coping mode is very similar to the "serenity prayer" advocated by 12-step groups for alcoholism: "God grant me the serenity to accept the things I cannot change; courage to change the things I can; and wisdom to know the difference (33)."
From a psychological point of view, it would be very interesting to find out whether insomnia patients resorting to a rigid monitoring coping style have difficulties in recognizing uncontrollable situations or whether their monitoring behavior is motivated by dysfunctional attitudes such as "monitoring every situation closely is helpful." Therapeutic implications could be a discrimination training between controllable and uncontrollable situations and the use of behavioral experiments to challenge these dysfunctional beliefs.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Received for publication March 23, 2005; revision received October 7, 2005.
DOI:10.1097/01.psy.0000195881.01317.d9
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  W. M. Troxel, J. M. Cyranowski, M. Hall, E. Frank, and D. J. Buysse Attachment Anxiety, Relationship Context, and Sleep in Women With Recurrent Major Depression Psychosom Med, September 1, 2007; 69(7): 692 - 699. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
