مدیریت استرس مداخله برای پیشگیری اولیه از فشار خون بالا: نمایش نتایج: از شماره تفصیلی مرحله اول محاکمه پیشگیری از فشار خون بالا (TOHP-I)
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|7032||2000||14 صفحه PDF||سفارش دهید||9740 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Annals of Epidemiology, Volume 10, Issue 1, January 2000, Pages 45–58
PURPOSE: Stress Management Intervention (SMI) was one of seven nonpharmacologic approaches evaluated in Phase I Trials of Hypertension Prevention (TOHP-I) for efficacy in lowering diastolic blood pressure (BP) in healthy men and women aged 30 to 54 years with diastolic BP 80–89 mm Hg. METHODS: A total of 242 and 320 participants were randomized to SMI or an “assessment only” SMI Control, respectively, at four clinical centers. The SMI consisted of 37 contact hours in 21 group and two individual meetings over 18 months and included: training in four relaxation methods, techniques to reduce stress reactions, cognitive approaches, communication skills, time management, and anger management within a general problem-solving format. Standardized protocols detailed methods and timing for collecting BP, psychosocial measures, and urinary samples from both SMI and SMI Control participants. RESULTS: In intention-to-treat analyses, although significant baseline to termination BP reductions were observed in both groups, net differences between the SMI and SMI Control groups' BP changes (mean (95% CI)) were not significant: −0.82 (−1.86, 0.22) for diastolic BP, and −0.47 (−1.96, 1.01) for systolic BP. Extensive adherence sub-group analyses found one effect: a significant 1.36 mm Hg (p = 0.01) reduction in diastolic BP relative to SMI Controls at the end of the trial for SMI participants who completed 61% or more of intervention sessions. CONCLUSIONS: While the TOHP-I SMI was acceptable to participants as evident from high levels of session completion, the absence of demonstrated BP lowering efficacy in intention-to-treat analyses suggests that the TOHP-I SMI is an unlikely candidate for primary prevention of hypertension in a general population sample similar to study participants. The isolated finding of significant diastolic BP lowering in SMI participants with higher adherence provides very weak evidence of SMI BP lowering efficacy and may be a chance finding. Whether similar or other stress management interventions can produce significant BP lowering in populations selected for higher levels of BP, stress, or intervention adherence remains to be demonstrated.
Although mortality rates from stroke and coronary heart disease (CHD) have declined for the last two decades as high blood pressure (BP) has been treated in larger segments of the population, progress has slowed substantially in recent years (1). The identification of nonpharmacologic methods to prevent, or significantly delay the onset of hypertension would represent an important advance in the primary prevention of cardiovascular disease (1). The objective of Phase I of the Trials of Hypertension Prevention (TOHP-I) was to assess the efficacy of seven nonpharmacologic interventions in lowering or preventing an increase in diastolic blood pressure (BP) over 18 months of follow-up (2). These nonpharmacologic interventions included three lifestyle change programs: stress management intervention (SMI), dietary sodium reduction, and weight reduction. Four nutritional supplement regimens utilizing calcium, magnesium, fish oil, and potassium were also evaluated. Primary results from the trial have been presented elsewhere (3). This report presents detailed findings from the SMI arm of the trial. The inclusion of SMI in Phase I of TOHP was supported by three types of evidence. First, some treatment outcome studies 4, 5 and 6 had reported modest BP reductions associated with SMI. Second, a plausible sympathetic nervous system mechanism for the development of stress-related hypertension had been proposed (7) and was supported by laboratory studies in animal models 8 and 9 and humans (10). Third, observational studies of human populations suggested an association between a variety of stressful events and autonomic nervous system responses that included both immediate and delayed cardiovascular reactions such as elevated BP 11 and 12.
نتیجه گیری انگلیسی
For the primary study outcome, BP, the intention-to-treat analyses found the TOHP-I SMI did not significantly lower either diastolic or systolic BP at any point in time compared to the reductions observed in the SMI Controls. These results were not influenced by differences between the SMI and SMI Controls at baseline, nor were imbalanced changes from baseline in potential BP confounders observed. Furthermore, no baseline variable, including clinic location, was associated with differential BP outcome. While a subgroup analysis found that participants completing 61% or more of sessions had a significant 1.36 mm Hg diastolic BP reduction at termination compared to the SMI Controls, there was no evidence of significant reductions in systolic BP with this higher adherence group. There was also no evidence of a dose-response relationship between any measure of reported level of relaxation usage and either diastolic or systolic BP. The isolated finding of diastolic BP lowering in participants with higher attendance provides weak evidence of greater BP lowering with higher SMI adherence and may represent a chance finding. The absence of an SMI effect on BP for the high normal blood pressure population of TOHP-I is in agreement with earlier, comprehensive reviews 31 and 32 of similar behavioral methods employed for reduction of blood pressure in hypertensives. Generally, larger, well-controlled studies with extended follow-up like TOHP-I have not replicated the results of smaller, less well- controlled and less well-designed studies with shorter-follow-up. Hence, increased methodologic rigor may account for differences between the negative outcomes from TOHP-I and the more positive findings from other studies. For example, Blanchard and coworkers reported being unable to replicate their own earlier BP lowering effects and attribute this failure to increased methodologic sophistication (33). Considering intermediate, non-BP outcomes, it was hypothesized that the SMI would produce decreases in Hassles frequency and intensity, and that these changes would be associated with BP reductions. Hassles frequency and intensity were not significantly lowered in the SMI versus SMI Controls, and Hassles frequency and intensity were unrelated to any BP change in any period. Hassles frequency and the Self-control subscale of the Psychological General Well-Being Schedule both increased significantly in the SMI compared to the SMI Controls. However, the significance of both of these effects was the result of decreases in the SMI Controls that were equivalent in magnitude to the increases in the SMI. Offsetting changes of plus or minus one item in the 131-item Hassles Scale was statistically significant because of the large number of participants, but this change is clinically insignificant in a group of 242 participants especially considering the absence of any group differences in Hassles intensity. Furthermore, since the level of change observed in these two measures was not related to any measure of SMI adherence, these observed increases most likely represent chance findings rather than intervention effects. It was also hypothesized that, compared to SMI Controls, the SMI would affect urinary catecholamines and cortisol. Dopamine increases were expected to be associated with BP reductions (34), but no significant differences between the groups were observed. The conceptual model for the TOHP-I SMI assumed that participants appropriately applying acquired SMI knowledge and skills would experience lower levels of stress. It was expected that these reduced levels of stress would be associated with lowered levels of the intermediate measures: Hassles, the Psychological General Well Being Schedule, and the urinary biochemical measures. Furthermore, it was expected that lower levels in the intermediate measures would be associated with BP reductions. The results of this study do not support this conceptual model of TOHP-I SMI action: the SMI did not lower the measures of stress and no reductions in BP were observed relative to SMI Controls. Aside from the conclusion that the TOHP-I SMI did not significantly lower BP in this population, four alternative explanatory hypotheses for these findings involve the following: 1) the stress levels of the study population; 2) the initial BP levels of the study population; 3) the formulation of the SMI; and 4) the implementation of the SMI. These are discussed in order, below. One hypothesis is that recruiting participants with higher stress levels might have allowed the SMI to reduce stress enough for a BP effect to be observed. TOHP-I recruitment communications described a study investigating approaches to preventing hypertension using methods such as weight loss, sodium reduction, stress management, and dietary supplements. As a result, TOHP-I participants were not recruited for levels of stress or specific interest in relaxation methods or stress management. While it is possible that participants selected for specific interest in relaxation or stress management or for relatively higher stress levels would have experienced different results, the TOHP-I SMI results can only be interpreted with regard to the population recruited. Although not recruited for high stress levels, it was assumed when TOHP-I was designed that the general population had stress levels sufficient for the SMI to significantly reduce stress. Hassles frequency values at baseline were (mean ± SD) 17.4 ± 12.9 and 18.2 ± 11.8, respectively, for the TOHP SMI and SMI Control groups and 24.75 ± 18.41 for the community sample reported by Zarski (35). Hassles intensity scores for the SMI and SMI Control groups were 1.3 ± 0.4 and 1.4 ± 0.4, respectively, compared with 1.6 ± 0.37 for the same community sample. In contrast to Hassles, on the Psychological General Well-Being Anxiety subscale, SMI and SMI Controls (16.7 ± 4.0 and 16.8 ± 3.8, respectively) reported slightly higher anxiety (lower score = higher anxiety) than the 17.9 + 4.7 reported from the RAND community sample (29). Overall, these measures suggest that the TOHP population was comparable to other community samples on Hassles and the PGWBS Anxiety subscale. The possibility of selecting SMI participants for elevated levels of stress was considered during the study design phase. Although randomization to the TOHP-I weight loss intervention required weight above 110% of ideal based on the 1983 Metropolitan Life Tables (13), this was the sole intervention-specific selection criterion in the study. We know of no stress management study involving blood pressure which has included results of stress or anxiety assessments as eligibility criteria. Nonetheless, selecting participants for elevated levels of stress is conceptually appealing and might be a fruitful area for future research. However, intervention-specific selection criteria can reduce the public health applicability of study results and are likely to reduce recruitment yields. A second hypothesis is that recruiting participants with higher BP levels might have allowed the SMI to reduce BP enough for an effect to be observed. TOHP-I was designed to recruit a population with high-normal BP (2). However, since the effects of both pharmacologic and nonpharmacologic interventions are often greater among those with higher initial BP (36), future studies might address this population. Participants with uncontrolled hypertension, demonstrated BP-stress associations, elevated cardiovascular reactivity, or highly variable ambulatory monitoring results might be more likely to show BP reductions with stress management. A third hypothesis is that a different SMI might have reduced BP enough for an effect to be observed. After considering numerous alternatives for the design of the TOHP-I SMI, we decided to deliver a broadly conceived intervention with a number of different elements thought by investigators, consultants, and SMI Program Leaders to be important for a stress management program. In early 1987 when the TOHP-I SMI was being developed, Patel and colleagues' (4) work with a multi-component intervention seemed the most persuasive and consistent with more of the studies reporting blood pressure reduction effects. It is not clear how selection of a single component like Schneider's focus on Transcendental Meditation (37) would have influenced the results of the TOHP-I SMI. A fourth hypothesis is that a more effectively implemented SMI might have reduced BP enough for an effect to be observed. The level of adherence achieved in the TOHP-I SMI is a key issue to consider in evaluating the lack of significant BP lowering observed. If adherence levels were adequate, then this specific SMI applied to the TOHP-I population was not effective in lowering BP. If adherence was not adequate, then the SMI was not fairly tested and its efficacy in lowering BP would remain unknown. A comprehensive assessment of adherence is arguably not feasible when the intervention is clinically sophisticated, involves multiple components, and encourages individualization of stress management knowledge, skills, and activities. The feasible alternative was to choose session completion and reported relaxation frequency and duration as objectively measurable and consistent representatives of overall adherence. However, by themselves, session completion and relaxation were not assumed to represent more than a small fraction of the overall knowledge, skills, and activities that could contribute to comprehensive SMI adherence. The actual SMI adherence results were mixed. Session attendance and session completion exceeded the intervention goals while adherence with daily relaxation usage was about half of goal. Still, we believe that the level of adherence observed was sufficient to fairly evaluate the impact of the TOHP-I SMI on BP in this population. While levels of adherence declined through time, this was expected. Furthermore, we believe that the levels observed represent a realistic estimate of SMI adherence expected from a general population sample in a trial of 18 months duration. That the level of adherence accomplished was sufficient for a fair evaluation of the impact of the TOHP-I SMI on BP is based on four observations. First, all SMI skills were easy to learn and often were enjoyable to use and/or their use provided immediate feedback on their effectiveness. Second, all SMI Program Leaders had significant experience with behaviorally oriented group interventions, and had prior experience with many of the methods used in the TOHP SMI. Third, the SMI's goals explicitly targeted practical knowledge and skill acquisition, practice, and application to real-life situations away from intervention meetings. The SMI structure of 23 sessions with most of the 100 minutes allocated to learning and applying skills within a generic problem solving framework with the guidance of experienced Program Leaders was directly derived from the strongly applied emphasis of the SMI goals. As a part of each session, Program Leaders observed, assessed, and facilitated improvements in participants' use of SMI skills, including relaxation methods. Fourth, the overall proportion of session completion was very high with very small numbers of treatment dropouts through 23 sessions lasting at least 18 months. Most participants reported continued relaxation activity after 18 months. We believe participants persisted in coming to meetings and continued SMI activities because of the SMI's relevance and benefit. Ultimately, if higher levels of long-term adherence with SMI components are required for consistent, persistent, and clinically significant effects to be observed, and if these higher levels of adherence are not attainable for a significant proportion of the general population, then SMI has limited public health applicability. Further progress in evaluating stress management program impact on BP would benefit from the development of validated measures of stress and adherence that also have demonstrated relationships with BP. At the time the TOHP-I SMI was developed, there were no validated instruments to assess stress directly and no stress inventories associated with blood pressure. While Hassles and the PGWBS had been associated with some biological outcomes and had some face validity as measures of stress, neither had a demonstrated relationship with BP change. Similarly, while attendance and frequency/duration of relaxation were thought to be good representatives of adherence, it was not clear what specific SMI participant activities or changes would be associated with blood pressure change. To this end, the finding that changes in PGWBS and dopamine were associated with BP lowering may be useful. The TOHP-I SMI had a number of significant strengths. TOHP-I was a large, randomized, controlled, multicenter trial with at least 86% power to detect a 2 mm Hg change in diastolic BP in all interventions (2). BP was well characterized with multiple measurements by blinded technicians who underwent periodic recertification in standardized BP measurement. SMI participants were instructed to avoid use of any SMI methods during BP measurement. The protocol for the 37 contact-hour, 23 session intervention was conducted from extensively documented intervention and participant manuals by behaviorally oriented, experienced Program Leaders who had ongoing training and supervision throughout the intervention. All of these elements assured consistent delivery of the intervention to participants, and rigorous assessment procedures and standardized measurement protocols assured high quality data. Finally, high levels of attendance through the 18 months of intervention indicated continuing participant interest and effort. In summary, TOHP-I provided a compelling test of the effects of this particular intensive, multifaceted SMI on BP because of the trial's large and diverse participant sample, its rigorous randomized design, and multiple high-quality BP assessments over an 18-month follow-up period. In the “intention-to-treat” analysis at the end of the trial, the 23-session, 18-month effort produced an insignificant reduction of less than one mm Hg in diastolic BP relative to the Control participants. We are confident that the failure to significantly lower blood pressure in this population is attributable to the ineffectiveness of a comprehensive, intensive, and well-implemented SMI with acceptable adherence. As a result, the SMI tested in TOHP-I is not a good candidate for primary prevention of hypertension in a similar population with high normal blood pressure. The results of this study cannot rule out the possibility that BP lowering effects might be observed for other types of SMI or for other populations such as individuals with hypertension, high stress levels, or those who are willing to comply at higher levels with the SMI. In contrast, other findings from Phase I of TOHP (3) demonstrated that weight loss and sodium reduction interventions can be effective in reducing BP and consequently may be promising approaches for preventing hypertension among those at the high end of the normotensive range. The long-term efficacy of these approaches has been further evaluated in Phase II of TOHP (38).