| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
REACTIVITY SPECIAL SECTION |
From the University of British Columbia (W.L.), Vancouver, British Columbia, Canada; and Mount Sinai School of Medicine (W.G., K.D.), New York, New York.
Address reprint requests to: Wolfgang Linden, PhD, Psychology/UBC, 2136 West Mall, Vancouver BC V6T 1Z4, Canada. Email: wlinden{at}psych.ubc.ca
ABSTRACT
This article introduces a series of articles that assess the present status of the cardiovascular reactivity construct as well as the progress that has been made since a critical review of the reactivity literature by Pickering and Gerin was published in 1990.
Key Words: cardiovascular reactivity • hypertension • disease prediction • generalizability • stress recovery.
Abbreviations: CVD = cardiovascular disease disease; CVR = cardiovascular reactivity; SNS = sympathetic nervous system.
This issue of Psychosomatic Medicine is devoted to a critical review of the status of the cardiovascular reactivity (CVR) concept. The purpose of this first article in the series is to set the stage for the in-depth reviews that follow. We begin with a brief explanation about the history of this special issue. The editors of this issue arranged an informal workgroup session on the status of cardiovascular reactivity to accompany the Annual Meeting of the American Psychosomatic Society in Vancouver, Canada, in 1999. The timing of this study session was not accidental because the 10th anniversary of the frequently cited 1990 review by Pickering and Gerin (1) was coming up. That article had effectively spelled out an agenda for research on the reactivity concept, and we wanted to evaluate the current state of the reactivity hypothesis as well as to update this agenda for the coming decade. The members of the group were (in alphabetical order) Jos Brosschot, University of Leiden, Leiden, Netherlands; Karina Davidson, Mount Sinai School of Medicine, New York, New York; Harry Davis, Medical College of Georgia, Augusta; William Gerin, Mount Sinai School of Medicine, New York, New York; Thomas Kamarck, University of Pittsburgh, Pittsburgh, Pennsylvania; Wolfgang Linden, University of British Columbia, Vancouver, Canada; William Lovallo, University of Oklahoma Health Sciences Center, Oklahoma City; Neil Schneiderman, University of Miami, Miami, Florida; David Sheffield, University of Florida, Gainesville; Julian Thayer, National Institute on Aging, National Institutes of Health, Bethesda, Maryland; and Frank Treiber, Medical College of Georgia, Augusta. In hindsight, it should probably be taken as a sign of a successful meeting that the participants agreed to put pen to paper and share the conclusions reached at that meeting.
It is hard to imagine a reader of this special issue for whom the term cardiovascular reactivity is not already meaningful, and we shall not engage in a lengthy definition process. Briefly, CVR researchers hypothesize that the study of cardiovascular responses to controlled, short-term, physical, cognitive, and emotional challenges serves as a window into complex psychological and physiological processes that are involved in the development of disease. Probably the most overarching question for CVR researchers concerns whether "hyper" reactivity is causally implicated in the etiology of hypertension and/or coronary heart disease or whether it serves as only a marker of future risk without incurring a causal role. Nevertheless, even if CVR was only a reliable marker, it could still have potential clinical uses in studying disease processes. There is consensus that for CVR to be of use as a predictor or explanatory construct, its study must undergo the same rigorous assessment of its metric properties as any other psychological or physical test. Correspondingly, the second article in this series (2) describes in detail the concepts, definitions, and response topographies that are being studied and tackles the issue of the reliability and generalizability of the physiological measurements taken during CVR protocols. Issues of predictive validity are addressed in the third article (3), which reviews the evidence that CVR predicts preclinical indicators of the development of cardiovascular disease. The fourth article (4) discusses the biological plausibility of CVR as pathogenic influence in disease by examining a number of different pathways. The fifth and final article (5) tackles content and concurrent validity by linking variations in CVR in the laboratory to physiological variability in everyday life. This last article may also be the most provocative one in that the many limitations of the laboratory as a "window to the real world" are critically examined.
Why has CVR research so dominated the field as the primary biopsychological link to stress? A quick search of the literature using only the term "cardiovascular reactivity" resulted in a list of 652 references in Medline (starting at 1966) and 1729 references in PsychInfo. The ubiquity of this research strategy is probably due to several factors. For one, researchers can exercise a great deal of creativity in building challenging protocols, and stressful situations can be manipulated in infinitely different but generally controllable ways. A broad range of populations, human and animal, can be studied, and measurable change in the physiological outcome measures such as blood pressure, stroke volume, or heart rate can be triggered almost instantaneously. Although the recommendation of "further study" is generally a good idea, we believe that in such a proliferating field it is highly necessary to periodically assess the status quo to remind ourselves of the larger objectives and to make decisions about which questions are in most urgent need of answer and which methods may be the most promising. These considerations shaped the current series of articles, which in turn treat Pickering and Gerin’s (1) review and its criticisms as the "baseline" against which to judge the progress that has been made.
STATUS QUO IN 1990
Issue: Stability Over Time
If hyperreactivity is a contributing factor to disease, it needs to be a reasonably stable person characteristic. The challenge has always been to create protocols in the laboratory that indeed show good psychometrics. One such critical feature is test-retest reliability. Pickering and Gerin (1) reported numerous studies that suggested that reactivity is stable, but the reliability coefficients were typically much lower than what is generally regarded as adequate in psychometrics (ie, a target r > 0.8). Kamarck and Lovallo (2) describe in some detail how the question of test-retest stability in the lab has largely been settled by use of the aggregation principle, using multiple readings across multiple tasks as joint indicators of hyperreactivity. Application of this basic tenet of psychometrics has been shown to lead to stability coefficients that can indeed exceed the 0.8 criterion. Kamarck and Lovallo (2) and Schwartz et al. (5) also review the laboratory-to-life generalizability of CVR measures, and their reviews indicate that it is exceedingly difficult to study this correspondence and that the reported associations are modest at best. Kamarck and Lovallo (2) suggest that the study of within-person change over time is a more promising approach than is interindividual comparison and that aggregated scores may improve observed associations between laboratory change scores and measures taken using ambulatory blood pressure monitoring. In the final article of this series, Schwartz et al. (5) offer a clearly less optimistic view on this topic, suggesting that laboratory reactivity research has critical limitations that are difficult to overcome.
Issue: Is There Concurrent Validity Such That Individuals at Known Risk for Hypertension of CVD Show Greater Reactivity? Are the Relevant Etiological Pathways Understood?
If CVR does predict the development of hypertension, then it would make sense that individuals already at risk for the development of hypertension will show greater reactivity. Candidate populations worth studying are those having a positive family history of hypertension (FH+), those with blood pressure levels considered "borderline" hypertensive, and those with particular personality traits that have been linked to hypertension. In 1990 it was already apparent that FH+ individuals tend to be more reactive than FH- individuals and that hypertensives are more reactive than normotensives (6). In particular, FH+ individuals show more reliable diastolic blood pressure responses, and these changes are most evident in response to active stressors. In their critical review, Pickering and Gerin (1) suggested that arterial geometry and/or differences in membrane transport might represent the pathways underlying variation in blood pressure reactivity. The general assumption at the time was that chronically exaggerated SNS activity was the primary biological pathway that accounted for the greater blood pressure response. There also was mixed evidence that personality (in particular hostility) was linked to greater reactivity.
New evidence (see Ref. 2) argues against a simple SNS-hyperarousal model. Instead, these articles note that the balance of SNS and parasympathetic/vagal activity has received increasing attention. Furthermore, numerous researchers have added valuable discussions of neurohormonal influences (7, 8), and these dimensions may be viewed as an expansion of the reactivity model with a focus on several interacting physiological systems rather than a single one. This has led to the categorization of response patterns as being predominantly 
- vs. ß-adrenergic in nature, although it is not clear at this time whether one is more critical than the other for a disease predictor model. Also of use in furthering our understanding of pathways has been the study of baroreceptor activity in blood pressure regulation (9), and this has in turn been linked to opioid models (10, 11). A plethora of studies on distinct personality patterns and reactivity has appeared in the last decade, and these studies have produced seemingly confusing and often inconsistent results. A number of thorough reviews have facilitated critical analysis of this literature (eg, Ref. 12) and have concluded that exaggerated reactivity observed in the laboratory setting is particularly likely when personality-situation match models are being applied. In particular, hostility-prone individuals are more likely to react strongly to a laboratory stressor when this stressor is angering in nature (13). Progress in this area is briefly described in the final article by Schwartz et al. (5), who emphasize the need to focus on the interactions among the genetic factors, the environmental and social determinants, and the psychological and physiological responses.
Issue: Predictive Validity
In a sense, this is the most crucial issue involved. CVR is a useful and even exciting construct only to the extent that it adds unique predictive power for development of cardiovascular disease. If it is able to fulfill this role, then it contributes to our understanding of the biopsychological pathways by which environmental factors, including stress, coping, and social support, are linked to hypertension and coronary heart disease. Under these circumstances, understanding CVR would also suggest strategies for prevention and treatment of these diseases. Testing the ability of the concept to predict can be achieved by different strategies:
New Themes
Given that the use of the aggregation principle (ie, measuring across more than one measure or task) has improved reliability, there are now studies that in turn investigate which types of aggregation methodologies are most useful for this purpose (14). Ultimately there will need to be a balance between a high requirement for reliability that can be met with increasingly complex task protocols and the associated cost burden (ie, the burden to both participants and researchers). Not anticipated in the Pickering and Gerin (1) review was the reawakening of interest in cardiovascular recovery from stress (15, 16). As Schwartz et al. (5) discuss, the thorough study of arousal and recovery patterns together has provided promising opportunities for maximizing reactivity research. In some studies, analysis of recovery data served to reinforce what was already obvious in studies of baseline-to-task changes, but there are interesting additional findings that would have been missed without controlled and transparent recovery protocols (for reviews, see Refs. 15 and 16). Furthermore, recovery protocols provide an experimental strategy for studying the cognitive and affective underpinnings of sustained cardiovascular arousal, which in turn have promise as mediating variables in the process of acute stress leading to chronic maladaptation. Thus, in the final article of this series, Schwartz et al. (5) describe the role of rumination after an anger-provoking stimulus and present research on worry and perseverative thoughts and their effects on sustained parasympathetic activation.
In addition to the traditional emphasis on regulation within the cardiovascular system, Lovallo and Gerin (4) describe studies in which researchers have begun to integrate burgeoning knowledge in the endocrine and psychoneuroimmunology literature with cardiovascular pathway models. Recent research that suggests that CVD may, at least in part, be the result of infectious processes (17) could be perceived as a threat to the potential explanatory power of CVR in cardiac pathology; however, this is not an inevitable conclusion when one considers that psychological stress can affect immune function as much as cardiovascular function. Thus, this last decade has seen major advances in our theorizing about disease pathways, and there are more opportunities than ever to study the CVR concept in multiple pathway structures. How much of that can be done within laboratory protocols is an issue of ongoing debate (5).
There remains a great deal of controversy as well as ambiguity concerning the relevant theoretical constructs and the empirical support for them. This is inevitable, however, and, to the extent that it continues to stimulate intellectual debate and well-designed empirical tests, it is also greatly desirable. Failing to find support for a single pathway model is likely constructive in the long run even though early models may have looked promising at the time (17). The parameters of the models continue to grow more complex as multiple systems are integrated into them (see Ref. 5), and the focus has subtly shifted so that rather than looking solely at heart rate or blood pressure responses that occur while a stressor is physically present, we now think of a broad range of physiological adaptations to environmental challenges, which in turn are influenced by genetics, multiple interacting physiological systems, situational factors, and personality dimensions (4, 5, 18).
Received for publication June 5, 2001.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  P. H. Wirtz, L. S. Redwine, U. Ehlert, and R. von Kanel Independent Association Between Lower Level of Social Support and Higher Coagulation Activity Before and After Acute Psychosocial Stress Psychosom Med, January 1, 2009; 71(1): 30 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Wirtz, R. von Kanel, C. Mohiyeddini, L. Emini, K. Ruedisueli, S. Groessbauer, and U. Ehlert Low Social Support and Poor Emotional Regulation Are Associated with Increased Stress Hormone Reactivity to Mental Stress in Systemic Hypertension J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3857 - 3865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Masters, T. L. Lensegrav-Benson, J. C. Kircher, and R. D. Hill Effects of Religious Orientation and Gender on Cardiovascular Reactivity Among Older Adults Research on Aging, March 1, 2005; 27(2): 221 - 240. [Abstract] [PDF]
|
|
|
|
|
|
M. M. Llabre, S. Spitzer, S. Siegel, P. G. Saab, and N. Schneiderman Applying Latent Growth Curve Modeling to the Investigation of Individual Differences in Cardiovascular Recovery From Stress Psychosom Med, January 1, 2004; 66(1): 29 - 41. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
