This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rutledge, T. Articles by Paul, D. Search for Related Content PubMed PubMed Citation Articles by Rutledge, T. Articles by Paul, D. Related Collections Parasympathetic Nervous System Sympathetic Nervous System Psychophysiology

Cardiovascular Recovery From Acute Laboratory Stress: Reliability and Concurrent Validity

From the University of Pittsburgh (T.R.), Pittsburgh, PA, and the University of British Columbia (W.L., D.P.), Vancouver, British Columbia, Canada.

Address reprint requests to: Thomas Rutledge, PhD, 1007 South Trenton Ave. 14, Pittsburgh PA 15221. Email: dr.tom{at}musclemail.com

	ABSTRACT

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
OBJECTIVE: We assessed the value of laboratory measures of cardiovascular recovery across four criteria: reliability across multiple tasks, reliability across a 3-year time interval, ability to predict daily ambulatory blood pressure, and interrelationships with coronary risk factors and psychosocial variables.

METHODS: Three hundred twenty-nine healthy adults (mean age = 27.1 years) completed a two-part protocol consisting of 1 day of laboratory testing and 1 day of ambulatory monitoring. The laboratory protocol included a 15-minute baseline assessment followed by three 5-minute laboratory challenges (mental arithmetic, speech, and handgrip). Five-minute recovery periods followed each exercise. One hundred twenty-five participants returned after 3 years to repeat the protocol.

RESULTS: When aggregated across tasks, cardiovascular recovery showed acceptable levels of internal consistency ( values = 0.7) and proved relatively stable across time (r values = 0.22–0.35). Recovery values statistically improved the prediction of daily ambulatory readings above baseline and stress reactivity laboratory values (p values < .001) but were largely unrelated to coronary risk factors or psychosocial measures.

CONCLUSION: These results suggest that cardiovascular recovery from acute laboratory stress can be treated as a stable individual difference variable that can improve standard laboratory-based predictor models of ambulatory readings.

Key Words: blood pressure, • reactivity • recovery, • stress • laboratory testing.

Abbreviations: DBP = diastolic blood pressure; HR = heart rate; SBP = systolic blood pressure.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Although individual differences in patterns of cardiovascular recovery from acute stress have been discussed as a potentially important factor in disease development for >60 years (1), empirical inquiry into the issue remains limited. Importantly, this does not seem to be a result of a general lack of enthusiasm for acute stress paradigms, as demonstrated by the rapidly growing literature concerning the disease-mediating effects of cardiovascular stress reactivity (2, 3) in which a disposition toward elevated autonomic and neuroendocrine responses to acute laboratory challenges has been prospectively linked to development of hypertension and atherogenesis (4–6).

Despite the relative lack of attention, existing investigations of cardiovascular recovery data have shown promise. For example, delayed recovery of blood pressure from acute stress has been associated with a strong parental history of hypertension (7) and with future hypertension status (8). Authors of a recent meta-analytic review of cardiovascular recovery and hypertension risk factors (9) also concluded that diastolic recovery was reliably improved among normotensive patients as compared with borderline hypertensive and hypertensive patients and among whites as compared with blacks. Physical fitness has also been linked to improved cardiovascular recovery (10, 11), whereas poor recovery after exercise has predicted development of hypertension in several studies (12, 13). This combination of outcomes favors the interpretation that differences in cardiovascular recovery patterns may be related to known coronary risk factors and to populations with a higher prevalence of hypertension.

Cardiovascular recovery may also be associated with psychosocial factors shown to predict the emergence and prognosis of heart disease. Mental stress protocols including anger provocation have consistently produced delayed recovery patterns in comparison with protocols not including elements of anger provocation (14–16), with results suggesting that recovery among angry or hostile men may be particularly impaired. Similar results were obtained in studies examining cardiovascular recovery among normal and chronic stress groups; in those studies delayed recovery rates were observed in subjects in the chronic stress group after exposure to an acute mental stress exercise (17, 18). Thus, impaired blood pressure recovery patterns may also be associated with disease-relevant psychosocial factors.

	RATIONALE FOR THE CURRENT STUDY

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
In the current study, we assessed the value of cardiovascular recovery data collected from a large sample of young men and women. Participants completed a laboratory protocol consisting of mental arithmetic, speech, and handgrip exercises, in which blood pressure recovery was documented after each task. In addition to the laboratory component, all participants wore an ambulatory blood pressure monitor over the course of an 8- to 12-hour workday and provided information about major coronary risk factors and psychosocial variables. Finally, a subset of the original participants returned after a period of 3 years to repeat the laboratory protocol.

We contend that recovery data suffer from the same conceptual and methodological limitations once endured by cardiovascular stress reactivity paradigms and therefore bear the same burden of proof in demonstrating predictive value among medical and behavioral medicine researchers (19, 20). First, recovery patterns must be shown to be reliable across time and setting and should show relative stability across different forms of acute challenge. Second, to offer unique information, cardiovascular recovery data should be able to contribute to the assessment of blood pressure status, for example, by improving the prediction of ambulatory blood pressure levels above criteria based on baseline clinical readings and possibly cardiovascular reactivity values. Finally, we also need additional evidence relating cardiovascular recovery patterns to known coronary risk factors, such as family history and fitness, and to empirically supported psychosocial variables. The methodology of the current study provided the opportunity to address each of these criteria.

	METHODS

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Participants
Three hundred twenty-nine adults (157 men and 172 women, average age = 27.1 years) participated in the first component of a two-phase investigation associating laboratory and ambulatory blood pressure measurements with psychological predictors. Participants consisted of students from the University of British Columbia who volunteered in return for course credit and adults from the surrounding Vancouver community recruited through local advertising. In accordance with entrance standards, no participant had a concurrent diagnosis of coronary heart disease or hypertension. One hundred twenty-five participants (39% of sample, 63 men and 62 women, average age = 32.8 years) returned after 3 years to undergo the protocol a second time. The ethnic composition of both sample groups was approximately 70% white and 30% Asian.

Design
Participants completed an identical protocol on each testing date. The testing procedure involved 1 day of laboratory assessment, during which participants performed a counterbalanced set of 5-minute stress tasks (mental arithmetic, handgrip exercise, and a speech task in which they discussed an recent emotional event) while having their blood pressure monitored, and 1 day of ambulatory assessment, during which they wore an ambulatory blood pressure monitor over the course of an 8- to 12-hour workday. Recovery was assessed during a 5-minute period after each task. The ambulatory assessment always preceded laboratory testing and was conducted on a separate day. Participants completed a battery of psychological tests at the beginning of the laboratory session.

Instrumentation
Blood pressure information was collected in the natural environment using SpaceLabs model 90207 ambulatory monitors. These devices weigh approximately 0.7 kg and are worn in a protective pouch. Use and accuracy of the SpaceLabs monitor are supported by validation work (21). Participants were explicitly instructed to minimize physical activity during a measurement cycle and to avoid formal exercise while being monitored. Only approximately 10% of attempted measures were unusable. After the deletion of invalid readings, approximately 30 readings per individual were available for analysis. The primary reason for invalid measures was excessive movement. SpaceLabs monitors provide error codes for failed measurement attempts, thereby facilitating identification of the error . Importantly, our analyses of the ambulatory data did not control for the posture of the participant during readings, a factor known to influence the magnitude of the blood pressure response.

We collected blood pressure information from participants during the laboratory sessions using a Dinamap 845 vital signs monitor (Critikon Corporation, Tampa, FL). Validation work has shown that the Dinamap monitor provides blood pressure values that directly correspond with intraarterial measurements (22). We collected two readings during each task and recovery period at 1.5 and 3.5 minutes, respectively.

Measurement of Coronary Risk Factors and Psychological Variables
During the laboratory session, participants completed a battery of psychological instruments, including measures of depression (Beck Depression Inventory, Ref. 23), hostility (Cook-Medley Hostility Questionnaire, Ref. 24), daily stress (Daily Stress Inventory, Ref. 25), anger expression (Spielberger Anger Expression Scale, Ref. 26), and defensiveness (Balanced Inventory of Desirable Responding, Ref. 27). Several risk factors (ie, smoking status, exercise habits, family history, and age) were assessed by questionnaire. The family history items inquired whether participants had a family history of hypertension; we did not discriminate between those with one and those with two hypertensive parents (7). Lastly, we determined body fat levels by means of a six-site skin caliper test, which was performed twice (and averaged) by a trained research assistant for reliability.

Procedure
For the purposes of ambulatory monitoring, participants were asked to choose a typical day without specific stressors such as examinations. Monitors were fitted to participants, pretested on the spot, and returned 8 to 12 hours later for data analysis. Pretesting, which consisted of the first five ambulatory readings, was completed in the laboratory and was used to determine proper placement of the cuff. Whenever pretest values seemed questionable, the cuff was moved to another location, and the values were compared with Dinamap readings. Ambulatory readings were obtained every 20 minutes. Participants were directed to not ingest alcohol, caffeine, or nicotine or exercise strenuously 2 hours before the laboratory component of the study.

The laboratory protocol, which was always conducted on a separate day from the ambulatory monitoring protocol, consisted of a 15-minute baseline period, during which five blood pressure readings were obtained, followed by completion of the alternating 5-minute task and recovery periods. Two blood pressure readings were taken during each task and recovery period at 1.5 and 3.5 minutes, respectively.

Statistical Analyses
Recovery values
Cardiovascular recovery was defined as residualized change scores (ie, corrected for the pretask baseline mean) occurring between recovery periods and the baseline mean. We derived standardized residual scores by regressing baseline blood pressure and HR means on each task recovery mean (ie, calculating a separate standardized residual for each of the three tasks) and then averaging the resulting values; residual scores calculated in this manner, unlike simple change scores, are statistically unrelated to baseline values. Large residual values indicated delayed cardiovascular recovery. To maximize the reliability of our recovery index (28), we aggregated residual scores across the three stress tasks to create one residual for each of our DBP, SBP, and HR measures. Gender did not affect recovery indices in our sample; therefore, we combined across the male and female subgroups. Reactivity, when analyzed, was also defined in the form of standardized residual scores.

Alternate recovery definitions
Although the purpose of this investigation was not to examine various forms of recovery measures, it is important to acknowledge that in addition to simple or residual change scores, recovery can be defined in terms of percentages, slopes, absolute time to recovery, and area under the curve and curve-fitting techniques among other measures. Additional details about these methods can be found elsewhere (19). In our data calculations, the lack of moment-to-moment recovery recordings reduced the conceptual value of slope, time to recovery, and curve-fitting estimations. Among the measures that remained, we found residual scores to be the most reliable index.

Tests of stability
We assessed the reliability of recovery scores in our sample in two ways: 1) across tasks by generating intercorrelations and internal consistency coefficients (Cronbach’s ) among residual scores with year-1 laboratory data for the mental arithmetic, speech, and handgrip exercises; and 2) across time by computing test-retest correlations for participants completing both phases of the protocol.

Relationships with coronary risk factors and psychosocial variables
We tested relationships using Pearson r correlation coefficients between the aggregated residual change scores for blood pressure with age, body fat level, family history of hypertension, smoking status, and self-reported exercise habits with the year-1 data. Correlations were also calculated between recovery scores and our set of psychological scales (see above). Because of the number of correlations calculated (30 r values), we adjusted our criterion for significance to p < .005. Use of the year-1 data, with more than 300 participants, maximized the power of these tests.

Prediction of daily ambulatory readings
We tested the ability of cardiovascular recovery values to add to the prediction of ambulatory blood pressure levels sampled over an 8- to 12-hour workday through hierarchical regression equations in which mean laboratory baseline levels and reactivity residuals were entered at step 1, followed by the respective recovery index at step 2. This assessed the predictive validity of blood pressure recovery after controlling for baseline levels and reactivity residuals. Year-1 data were again used for these tests.

Power
Because of the large sample for the year-1 testing, power levels to detect medium effect sizes (29) met or exceeded 0.9 for tests of internal consistency stability, interrelationships with risk factor and psychological variables, and incremental validity tests with year-1 ambulatory readings. Alternatively, because of the much smaller sample for the 3-year test and the relative infrequency of high blood pressure among the subjects, power levels approximated 0.7 for test-retest reliability values of moderate size.

	RESULTS

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Sample Characteristics
Table 1 provides means and standard deviations for the sample on relevant risk factor and psychosocial characteristics across the two protocol dates. Participants who returned for phase 2 were significantly older at the first test date (t = 3.2, p < .01) but did not differ on any measured risk factor, cardiovascular variable, or psychological variable.

As a second test of reliability, we computed test-retest coefficients for the subgroup completing both phases of the investigation. Three-year correlation values for the recovery scores were modest but statistically reliable even with the reduced sample size (r = 0.22 and p = .02 for recovery of DBP and 0.34 and p < .001 for recovery of SBP). HR residuals across time, however, were statistically unrelated (r = 0.11, p > .05). Thus, the magnitude of cardiovascular recovery from the set of laboratory challenges was marginally stable over a 3-year time frame, but only for blood pressure variables. Broken down across individual tasks, test-retest correlations were marginally lower (0.16–0.20 for blood pressure and 0.01–0.10 for HR) but did not differ from one another.

Intercorrelations With Risk Factors and Psychosocial Variables
We used Pearson r coefficients to estimate relationships between recovery values and known coronary risks factors and empirically supported psychological characteristics. Using a more stringent significance criterion to adjust for Type I error inflation from the number of tests, we observed largely nonsignificant findings (r values ranged from -0.10 to 0.17). Neither blood pressure nor HR recovery was related to age, body fat, family history, self-reported exercise, or smoking status. Similarly, cardiovascular recovery was not associated with depression, hostility, daily stress level, or anger expression. The sole robust finding was an association between defensiveness and delayed recovery of DBP (r = 0.17, p = .003). Subdivision by gender had no effect on these outcomes.

Incremental Validity
Using hierarchical regression methods, we next examined the potential for cardiovascular recovery data to independently improve the prediction of daily ambulatory readings based on laboratory baseline and reactivity levels. For the diastolic and systolic indices, regressions assessed recovery values after controlling for baseline means and reactivity levels at steps 1 and 2 of each equation. Results from these analyses, shown in Table 3, indicated that both blood pressure and HR recovery values added significantly to the prediction of respective ambulatory means after controlling for baseline and reactivity (DBP: F(3,311) = 15.0, p < .001; SBP: F(3,311) = 22.5, p < .001; HR: F(3,311) = 6.9, p = .009). R² changes were small, however (3.0%, 3.0%, and 1.5% for DBP, SBP, and HR, respectively), suggesting modest overall improvement to the predictor models.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
In response to a perceived lack of attention to the process of cardiovascular recovery from acute stress, we conducted a multifaceted investigation of recovery values collected using a combination of well-controlled laboratory and ecologically valid ambulatory measurement strategies. Our methodological aim was to assess cardiovascular recovery according to empirical criteria used in many previous studies to evaluate cardiovascular reactivity.

At the most fundamental level, we first argued that recovery must demonstrate reliability at the individual level. High reliability is a necessary ingredient for any predictor variable and is conceptually critical to making a case for the existence of stable individual differences in cardiovascular recovery processes. Our results indicate that when recovery values are aggregated across multiple tasks, they can achieve acceptable levels of reliability. The same outcome has previously been demonstrated with cardiovascular reactivity, which has also endured criticisms of poor consistency (3, 20). In addition, recovery values were shown to be somewhat stable across a 3-year test interval. To our knowledge, this finding is unique among published laboratory investigations, approaching the level of test-retest support in the larger cardiovascular reactivity literature (31).

A number of previous studies have attempted to link blood pressure and HR recovery differences with established coronary risk factors (9) with generally mixed results. We did not observe relationships between recovery and risk factors, including family history, body fat level, age, smoking status, or exercise habits, a pattern consistent with the majority of findings. In one recent study in which an association between family history and recovery was not observed (7), the relationship held only among participants with familial risk from both parents. We did not collect information on both parents in this study. Gender effects were also not demonstrated by our analyses. Finally, we were unable to provide information about the effects of black or white ethnicity that are sometimes reported (32) because our subjects were almost exclusively of white or Asian origin.

The absence of recovery relationships with coronary risk factors essentially mirrored findings with our psychosocial variables. Defensiveness, in the form of high self-deception scores, was the only identified psychological correlate; in this case high scores reflected delayed recovery. Coupled with the robust statistical association observed here, this result may be of future interest given the growing literature supporting relationships between defensiveness and blood pressure variables (33, 34).

Our final set of analyses served as the most stringent benchmark for recovery data. We posited a test of incremental validity by quantifying recovery effects on daily ambulatory readings after controlling for baseline and reactivity blood pressure and HR levels. Thus, only the residual variance unique to recovery could add to the regression equation. The results from these analyses supported the utility of recovery as an independent predictor variable, explaining small but statistically reliable variability within the ambulatory readings. Given the large sample tested in this study (and consequently the high levels of power for even small effects), however, and the relatively modest effects in the form of R² changes, the clinical implications of this demonstration may be limited.

Study Limitations
The men and women in our sample were young, relatively healthy (physically and psychologically), and from the university and surrounding community, characteristics limiting our ability to link cardiovascular recovery to disease development or psychosocial variables. Indeed, an existing diagnosis of hypertension or heart disease was a primary exclusion criterion for enrollment in the study. For this reason, our sample is unlikely to be representative of the general population in terms of coronary risk factors and psychosocial variables. Therefore, caution must be exercised when generalizing our findings.

Attrition, or more specifically the possibility of nonrandom attrition, is a second source of concern. Although we are unable to generate plausible scenarios to support how attrition could be related to differences in cardiovascular recovery, we cannot fully eliminate this prospect. What we can state with conviction is that after numerous comparisons of completers and dropouts, age seemed to be the only factor influencing the probability of completing the study (ie, older participants were more likely than younger university students to continue living in the community for the 3-year interval). Equally important is that attrition difficulties, regardless of their significance, threaten only the test-retest segment of our conclusions.

Our findings provide both psychometric and predictive support for a relatively straightforward definition of cardiovascular recovery. However, although we attempted to define and assess cardiovascular recovery in a manner consistent with previous studies and reviews (7, 9), our conceptualization of recovery deserves scrutiny for one key reason: It is not independent of reactivity. Other definitions and methods of analysis have been proposed (19), although each suffers from one or more shortcomings. Thus, although our results offer evidence in favor of using baseline-adjusted change scores to quantify recovery, they should not be misconstrued as offering evidence against other recovery formulations.

	CONCLUSION

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
We have offered what we believe is among the most exhaustive treatment of cardiovascular recovery processes to date. The large sample, inclusion of laboratory and ambulatory measurements, and the prospective design features permitted assessment of recovery from the basic level of reliability to the practical issue of predictive validity. The added measurement of a relatively comprehensive package of coronary risk factors and psychosocial variables allowed us to investigate relationships with other potential markers of disease and to integrate our findings into existing literature. Several clear themes emerged from our investigation: 1) Cardiovascular recovery, when aggregated across multiple laboratory tasks or repeated trials, can be conceptualized as a stable individual difference variable that is moderately reliable over even lengthy time periods; 2) recovery values seem to possess independent predictive potential with respect to ambulatory measures; and 3) even with a comparatively large sample, blood pressure and HR recovery displayed only weak interrelationships with more established coronary and psychosocial predictors of physical disease.

Received for publication November 2, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION RATIONALE FOR THE CURRENT... METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 

1. Hines EA Jr, Brown GE. A standard stimulus for measuring vasomotor reactions: its application in the study of hypertension. Procedural Staff Meeting of the Mayo Clinic 1932;7:332–5.
2. Lynch W, Everson SA, Kaplan GA, Salonen R, Salonen JT. Does low socioeconomic status potentiate the effects of heightened cardiovascular responses to stress on the progression of carotid atherosclerosis? Am J Public Health 1998; 88: 389–94.[Abstract/Free Full Text]
3. Pickering TG, Gerin W. Cardiovascular reactivity in the laboratory and the role of behavioral factors in hypertension: a critical review. Ann Behav Med 1990; 12: 3–16.
4. Barnett PA, Spence JD, Manuck SB, Jennings JR. Psychological stress and the progression of carotid artery disease. J Hypertens 1997; 15: 49–55.[Medline]
5. Matthews KA, Woodall KL, Allen MT. Cardiovascular reactivity to stress predicts future blood pressure status. Hypertension 1993; 22: 479–85.[Abstract/Free Full Text]
6. Kaplan JR, Pettersson K, Manuck SB, Olsson G. Role of sympathoadrenal medullary activation in the initiation and progression of atherosclerosis. Circulation 1991; 84 (Suppl 6): VI-23–32.
7. Gerin W, Pickering TG. Association between delayed recovery of blood pressure after acute mental stress and parental history of hypertension. J Hypertens 1995; 13: 603–10.[Medline]
8. Borghi C, Costa F, Boschi S, Mussi A, Ambrosioni E. Predictors of stable hypertension in young borderline subjects: a five-year follow-up study. J Cardiovasc Pharmacol 1986;8(Suppl 5):S138–41.
9. Hocking-Schuler JL, O’Brien WH. Cardiovascular recovery from stress and hypertension risk factors: a meta-analytic review. Psychophysiology 1997; 34: 649–59.[Medline]
10. Jamieson JL, Lavoie NF. Type A behavior, aerobic power, and cardiovascular recovery from a psychosocial stressor. Health Psychol 1987; 6: 361–71.[Medline]
11. McCubbin JA, Cheung R, Montgomery TB, Bulbulian R, Wilson JF. Aerobic fitness and opioidergic inhibition of cardiovascular stress reactivity. Psychophysiology 1992; 29: 687–97.[Medline]
12. Tanji JL, Champlin JJ, Wong GY, Lew EY, Brown TC, Amsterdam EA. Blood pressure recovery curves after submaximal exercise: a predictor of hypertension at ten-year follow-up. Am J Hypertens 1989; 2: 135–8.[Medline]
13. Wilson NV, Meyer BM. Early prediction of hypertension using exercise blood pressure. Prev Med 1981; 10: 61–8.
14. Linden W, Long B. Repression, hostility, and autonomic recovery from a laboratory stressor. J Clin Hypertens 1987; 3: 567–78.[Medline]
15. Lai JY, Linden W. Gender, anger expression preferences, and opportunity for anger release determine cardiovascular recovery from anger provocation. Psychosom Med 1992; 54: 297–310.[Abstract/Free Full Text]
16. Engebretson T, Matthews KA, Scheier M. Relationship between anger expression and cardiovascular reactivity: reconciling inconsistent findings through a matching hypothesis. J Pers Soc Psychol 1989; 57: 513–21.[Medline]
17. Fleming I, Baum A, Davidson LM, Rectanus E, McArdle S. Chronic stress as a factor in psychological reactivity to challenge. Health Psychol 1987; 6: 221–38.[Medline]
18. Lundberg G. Subjective well-being and temporal pattern of sympathetic-adrenal-medullary activity. Biol Psychol 1976; 4: 157–72.[Medline]
19. Linden W, Earle TL, Gerin W, Christenfeld N. Physiological stress reactivity and recovery: conceptual siblings separated at birth? J Psychosom Res 1997; 42: 117–35.[Medline]
20. Krantz DS, Manuck SB. Acute psychophysiological reactivity and the risk of cardiovascular disease: a review and methodological critique. Psychol Bull 1984; 96: 435–64.[Medline]
21. O’Brien E, Mee F, Atkins N, O’Mally K. Accuracy of the SpaceLabs 90207 determined by the British Hypertension Society protocol. J Hypertens 1991; 9: 573–4.[Medline]
22. Borow KM, Newburger JW. Noninvasive estimation of central aortic pressure using the oscillometric method for analyzing systemic artery pulsatile blood flow: comparative study of indirect systolic, diastolic, and mean brachial artery pressure with simultaneous direct ascending aortic pressure measurements. Am Heart J 1982; 103: 879–86.[Medline]
23. Beck AT. Depression inventory. Philadelphia (PA): Center for Cognitive Therapy; 1978.
24. Cook WW, Medley DM. Proposed hostility and pharisaic-virtue scales for the MMPI. J Appl Psychol 1954; 38: 414–8.
25. Brantley PJ, Waggoner CD, Jones GN, Rappaport NB. A daily stress inventory: development, reliability, and validity. J Behav Med 1987; 10: 61–74.[Medline]
26. Spielberger C, Johnson EH, Russell SF, Crane RJ, Jacobs GA, Worden TJ. The expression and experience of anger: construction and validation of an anger expression scale. In: Chesney MA, Rosenman R, editors. Anger and hostility in cardiovascular and behavioral disorders. Washington: Hemisphere; 1985. p. 5–30.
27. Paulhus DL. Measurement and control of response bias. In: JP Robinson, PR Shaver, Wrightsman LS, editors. Measures of personality and social psychological attitudes. San Diego (CA): Academic Press; 1991 p. 17–59.
28. Kamarck TW, Jennings JR, Debski TT, Glickman-Weiss E, Johnson PS, Eddy MJ, Manuck SB. Reliable measures of behaviorally-evoked cardiovascular reactivity from a PC-based test battery: results from student and community samples. Psychophysiology 1992; 29: 17–28.[Medline]
29. Cohen J. Statistical power for the behavioral sciences. 2nd ed. Hillsdale (NJ): Erlbaum; 1988.
30. Anastasi A. Psychological testing. New York: MacMillan; 1961.
31. Swain A, Suls J. Reproducibility of blood pressure and heart rate reactivity: a meta-analysis. Psychophysiology 1996; 33: 162–74.[Medline]
32. Gillin JL, Mills PJ, Nelesen RA, Dillon E, Ziegler MG, Dimsdale JE. Race and sex differences in cardiovascular recovery from acute stress. Int J Psychophysiol 1996;23:83–90.
33. Shapiro D, Goldstein IB, Jamner LD. Effects of cynical hostility, anger out, anxiety, and defensiveness on ambulatory blood pressure levels in black and white college students. Psychosom Med 1996; 58: 354–64.[Abstract/Free Full Text]
34. Grossman P, Watkins LL, Ristuccia H, Wilhelm FH. Blood pressure responses to mental stress in emotionally defensive patients with stable coronary artery disease. Am J Cardiol 1997; 80: 343–6.[Medline]

This article has been cited by other articles:

				J. V. Moseley and W. Linden Predicting Blood Pressure and Heart Rate Change With Cardiovascular Reactivity and Recovery: Results From 3-Year and 10-Year Follow Up Psychosom Med, November 1, 2006; 68(6): 833 - 843. [Abstract] [Full Text] [PDF]

				T. W. Kamarck and W. R. Lovallo Cardiovascular Reactivity to Psychological Challenge: Conceptual and Measurement Considerations Psychosom Med, January 1, 2003; 65(1): 9 - 21. [Abstract] [Full Text] [PDF]

				A. R. Schwartz, W. Gerin, K. W. Davidson, T. G. Pickering, J. F. Brosschot, J. F. Thayer, N. Christenfeld, and W. Linden Toward a Causal Model of Cardiovascular Responses to Stress and the Development of Cardiovascular Disease Psychosom Med, January 1, 2003; 65(1): 22 - 35. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rutledge, T. Articles by Paul, D. Search for Related Content PubMed PubMed Citation Articles by Rutledge, T. Articles by Paul, D. Related Collections Parasympathetic Nervous System Sympathetic Nervous System Psychophysiology