CC BY
45
Assessing the impact of workplace stress on total cholesterol and low density lipoprotein levels: the example,,.
дятся «в разных измерениях»: тело — в реальном мире, а один из важнейших органов чувств, слух, — в виртуальном, созданном усилиями звукорежиссеров. Эти миры не имеют общих точек приложения, и поэтому мозг не может адекватно оценивать окружающую обстановку. В итоге подобного «раздвоения» можно получить травмы.
Конечно, наушники предназначены не только для развлечения. Многие слушают аудиокниги, занимаясь домашними делами, однообразной работой, в общественном транспорте. А слабослышащим людям наушники просто помогают услышать окружающие звуки. Но не стоит забывать о правилах их использования.
В творческой сфере даже профессиональные музыканты, в зрелом возрасте столкнувшиеся с ухудшением слуха, но при этом владеющие своим инструментом на интуитивном уровне, всё
равно вынуждены ограничить или даже полностью прекратить выступления и занятие музыкой. Например, Фил Коллинз, завивший о завершении своей концертной деятельности из-за прогрессирующего снижения слуха.
Обратимся к одному из авторитетных музыкальных журналов «Rolling Stone», в котором открыто обсуждается проблема глухоты от прослушивания музыки. Специалист-сурдолог сообщает, что без вреда для здоровья можно лишь в течение часа в день использовать обычные наушники с громкостью чуть выше половины от максимальной. Для минимизации неблагоприятных воздействий на слух, поведение и настроение человека необходимо внимательно отнестись к выбору наушников, следовать правилам их использования и обращаться к ним только по крайней необходимости.
Gaisenok Oleg Vladimirovich FSI « United Hospital with Outpatient Department» Managing Affairs of the President of the Russian Federation,
E-mail: [email protected]
Assessing the impact of workplace stress on total cholesterol and low density lipoprotein levels: the example of the application of log-linear analysis in medical studies with small samples
Abstract.
Objective: assess the impact of workplace stress on the lipid profile by using two-dimensional and log-linear analysis.
Methods: 118 patients on the basis of inclusion criteria selected from the PROFILE I registry’s database. They were divided into two groups, depending on the availability of nervous tension in the workplace: 1 — the presence of stress 2 — no stress.
Results: Significant differences in the level of total cholesterol (TC) and LDL-cholesterol (LDL-C) were observed between the two groups (5.1 [4.3, 5.9] vs. 4.5 [3.8, 5.1], p=0.01; [2.3; 3.9] vs. 2.5 [2.1, 3.1], p=0.04). It is worth noting the absence of statistically significant differences in the groups in statin intake (^2=2.48, p=0.1). Additional data analysis was done by using the log-linear analysis. The model which was created for the analysis included 5 signs (entrance signs: A-stress, B-physical activity, C-diet, D-statin therapy, and a sign-response: E-achievement of target lipid levels). If the effects of the interaction of pairs of signs regarded as equivalent to the effect of, the hierarchy of these factors contribute to the formation of this model is as follows: 1) sign B (physical
49
Section 3. Preventive medicine
activity) has a major influence on the formation of the model and its contribution averaged 41.65% (40.2-43.1); 2) sign C (diet) was second in importance in the formation of this model and its contribution averaged 30.85% (29.0-32.7); 3) sign D (statins intake) was the third of its importance in the formation of this model and its contribution averaged 19.85% (12.9-21.3); 4) the contribution of the sign A (workplace stress) on the formation of the model averaged 7.55% (5.7-9.4).
Conclusions: Stress is a known risk factor for CVD, both directly and indirectly through other factors. The impact of workplace stress on lipid factors was confirmed in this study. But the contribution of stress in the development of a common model was less important than other factors, and its influence revealed through sign’s interaction. Conducting a specially-designed randomized clinical trials is expedient to assess the role of stress on lipid factors. Log-linear analysis may be used to decision of tasks in medical studies with a small samples.
Keywords: workplace stress, lipid profile, registry, log-linear analysis.
Introduction Thus, the study of the effect of stress on lipid
All of us in everyday life are often faced with factors has particular interest.
stressful situations. And as is known, stress is a risk factor for cardiovascular and other chronic noncommunicable diseases. The impact of stress factors on the development of the disease is probably related to the time duration and regularity of residence under stress — that is, “exposure time”.
Previous studies have shown that psychosocial factors account for 32.5% of the total contribution of the main factors in the risk of myocardial infarction (MI) in all regions of the world, regardless of age and sex of the patients studied [1]. Daily emotional stress increases the risk of MI more than 2.5 times (OR, 2.81; 95% CI, 2.07 to 3.82, p <0.0001) [2].
Czech scientist F. Blaga studied deceased prisoners of the Nazi camp of Dachau during World War II and studied the role of stress in the development of atherosclerosis. They showed signs of extreme exhaustion that many authors have classified as the term «alimentary dystrophy». At the same time they had a marked development of atherosclerosis with fresh cholesterol deposits in the vessel wall [3; 4].
The opposite results were obtained in the work of Volkova VG et al. Study of the dead residents of besieged Leningrad, were in a state of «alimentary dystrophy», found that they had evidence of resorption of cholesterol deposits and reducing the phenomena of hyalinosis of small arteries [5]. The appearance of differences in the data on the state of the vascular wall in patients in a state of «alimentary dystrophy» can presumably be attributed to the role of heavy painful stress tension.
Objective
The aim of this study was to assess the impact of workplace stress on the lipid profile by using twodimensional and log-linear analysis.
Methods
Database of Profile I registry was used for this analysis. PROFILE registry is a registry of patients of one of the divisions of the National Research Center of Preventive Medicine (NRCPM MoH) — department of preventive pharmacotherapy. All patients who applied to the Department for the purpose of consultation or participating in observational clinical trials sequentially turned to the PROFILE registry. The methodology of creating a registry has been described previously, and it can be found in previous publications [6; 7].
Patients who appealed to the department of preventive pharmacotherapy during the period from May 1 to December 31, 2011 were included in this study. Everyone who appealed during this period has been granted a specially designed questionnaire, which allows us to estimate the presence ofnervous tension in the workplace, qualifing by patient as everyday stress. The questionnaire also provided an opportunity to evaluate the use of cholesterol-lowering drugs — statins (regularly, no, irregularly). Registration card filled in for each patient applied for this period by the results of inspection and survey. Registration card included clinico-anamnestic and anthropometric data, the results of clinical, laboratory and instrumental studies, as well as data from a survey of patients.
50
Assessing the impact of workplace stress on total cholesterol and low density lipoprotein levels: the example,,.
Inclusion criteria were as follows: the
employment at the time of the study, the availability of data of presence of stress in the workplace, the availability of data of statin intake, the availability of data of investigation on lipid profile in the registration card.
Criteria for target lipid levels in this study were adopted for people with moderate cardiovascular risk (SCORE scale < 5%): — total cholesterol <5 mmol/L (<190 mg/dL) — LDL cholesterol <3 mmol/l (<115 mg/dL) [8].
All patients signed informed consent to participate in the study and the questionnaire was approved by the Independent Ethics Committee of NRCPM. Blood tests for cholesterol and its fractions were carried out in the laboratory of NRCPM in the day of patient’s appeal.
Statistical analysis
Statistical data processing was performed using software package Statistica 6.0 (Statsoft). Data are presented as median and interquartile range or the true number and% of total patients. X2 criterion was used to compare the groups on the basis of quality. U-Mann-Whitney test was used to compare two independent groups by quantity.
Differences were considered statistically significant at p <0,05. Log-linear analysis was applied for additional data analysis. Log-linear analysis allows to determine value of the impact of analyzed signs to sign-response (which was adopted for the achievement of target lipid levels).
Results
Data of 118 patients were valid for the final analysis. The scheme of selection of the patients is shown in Figure 1.
Patients were divided into two main groups based on the presence of nervous tension in the workplace, qualified by the patient as an everyday stress: 1 — presence of stress, 2 — absence of stress.
Main clinical, anamnestic and nosological characteristics of these groups are shown in Table
1. Comparison of these groups carried out on lipid profile (see Table 2).
Differences in the level of total cholesterol (TC), LDL- cholesterol (LDL-C) and triglycerides (TG) were marked between the two groups: all indicators of lipid profile in patients who had daily nervous strain in the workplace were higher than in patients who didn’t have it (see Table. 2).
Fig,1, Scheme of patient’s selection
51
Section 3. Preventive medicine
Table 1. - The main characteristics of groups
Sign Group 1: presence of stress (n=95) Group 2: absence of stress (n=23) p-level
Age 60,0 [53,5; 66,0] 66,0 [59,0; 70,0] 0,04
Male 53 (55,8%) 16 (69,5%) 0,14
High education 84 (88,4%) 21 (91,3%) 0,98
Smoking 13 (13,7%) 2 (8,7%) 0,76
Body mass index 27,9 [24,8; 30,5] 27,5 [25,2; 30,0] 0,87
Arterial hypertension 73 (83,1%) 18 (78,2%) 0,89
Diabetes mellitus 12 (12,6%) 2 (8,7%) 0,86
Coronary artery disease 19 (20,0%) 12 (52,2%) 0,01
Atrial fibrillation 15 (15,8%) 6 (26,1%) 0,39
Regular intake of statins 42 (44,2%) 15 (65,2%) 0,12
When using the statistical analysis the reliable worth noting the lack of statistically significant differences were registered only for the TC and differences in groups on the fact of statins intake (y2
LDL-C levels (see Table 2, Figure 1 and 2). It is = 2,485, df = 1, p = 0,12).
Table 2. - Comparative characteristics of groups by lipid profile
Sign* Group 1: presence of stress (n=95) Group 2: absence of stress (n=23) p-level
Mean (St. D.) Median [Lower; Upper] Mean (St. D.) Median [Lower; Upper]
TCH 5,17 (1,19) 5,10 [4,30; 5,90] 4,45 (0,90) 4,48 [3,80; 5,10] 0,01
LDL 3,20 (1,01) 3,06 [2,31; 3,92] 2,72 (0,82) 2,53 [2,13; 3,16] 0,04
VLDL 0,70 (0,38) 0,62 [0,46; 0,81] 0,63 (0,28) 0,62 [0,40; 0,78] 0,68
HDL 1,21 (0,31) 1,17 [1,01; 1,36] 1,21 (0,27) 1,22 [1,01; 1,39] 0,89
TG 1,44 (0,73) 1,31 [0,90; 1,69] 1,31 (0,63) 1,12 [0,84; 1,69] 0,42
* TCH — total cholesterol; LDL — low density lipoproteins cholesterol; VLDL — very low density lipoproteins cholesterol; HDL — high density lipoproteins cholesterol; TG — triglycerides
52
Fig. 2. Differences in groups by level of TC on the presence of nervous tension in the workplace, qualifying as stress
Assessing the impact of workplace stress on total cholesterol and low density lipoprotein levels: the example,,.
Log-linear analysis was applied for additional to sign-response (which was adopted for the data analysis. Log-linear analysis allows to achievement of target lipid levels). determine value of the impact of analyzed signs
6,0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5 1,0
□
□
12
□ Median I I 25%-75% Min-Max
workplace stress (1- yes, 2 - no)
Fig, 3, Differences in groups by level of LDL- C depending on the presence of nervous tension in the workplace, qualifying as stress
The model which was created for the analysis included 5 signs (entrance signs: A-stress, B-physical activity, C-diet, D-statin therapy, and a sign-response: E-achievement of target lipid levels).
Results of log-linear analysis to verify the significance order effects are shown in Table 3. The table shows that the effects of the first and second order are significant, i. e. the effects of all the factors separately and their pairwise interactions.
Table 3, - The results of log-linear analysis: testing the significance of effects of the K-th order
Number of interacting signs Max.Lik. Pearson Probab.
1 129,2276 239,5696 0,000000
2 24,2621 24,3319 0,006767
3 6,5373 5,5579 0,850942
4 2,5424 2,4861 0,778586
5 0,2808 0,2822 0,595253
Optimal model explaining the signs interaction in this study was automatically selected by the program. It was a model including the interaction of AC, BD and DE signs. When analyzing the Test of marginal and partial associations 7 effects were found out significant
and reliable. The effects listed in Prt.Ass. column of Table 4, and based on degree Km effect on the frequency of observation and calculated as Xx 2 = 146,1 to all effects. Significant effects of the interaction of factors explain the observed frequency in this study their complex influence.
Table 4, - The results of log-linear analysis: Tests of Marginal and Partial Association
Factor’s effect* Prt.Ass. K.,% Mrg.Ass.
1 2 3 4
A 8,44505 5,7 0,003660
B 58,72299 40,2 0,000000
53
Section 3. Preventive medicine
1 2 3 4
C 42,35201 29,0 0,000000
D 18,90443 12,9 0,000014
AC 5,55800 3,7 0,027543
BD 4,29919 2,9 0,015789
DE 7,99022 5,5 0,002549
*A — work stress, B — activity, C — diet, D — statin intake, E — target level
If the effects of the interaction of pairs of signs regarded as equivalent to the effect of, the hierarchy of these factors contribute to the formation of this model is as follows: 1) sign B (physical activity) has a major influence on the formation of the model and its contribution averaged 41.65% (40.2-43.1); 2) sign C (diet) was second in importance in the formation of this model and its contribution averaged 30.85% (29.0-32.7); 3) sign D
(statins intake) was the third of its importance in the formation of this model and its contribution averaged 19.85% (12.9-21.3); 4) the contribution of the sign A (workplace stress) on the formation of the model averaged 7.55% (5.7-9.4).
Discussion
Standard statistical analysis of the data based on the y2 test and Mann-Whitney test gave an affirmative answer to the question posed in the study. When carrying out of log-linear analysis were obtained the expected results, but they were ambiguous. One would expect that the D-sign (statins intake) will have a greater impact on the sign-response E.
Basic role in control over achievement of target lipid levels based on the results of this study in these patients belonged to the non-drug influencing factors, namely, regular physical activity and diet. Their joint contribution to achieving the goal was more than 70%. Despite the fact that doctors and patients often believe that statins play a major role in achieving target lipid levels, the contribution of statins therapy was more modest, but it was significant and amounted to a total of about 20%. The contribution of stress in the formation of a common model was even less essential (average about 7.55%) and its influence revealed through sign’s interaction.
The obtained interactions of signs AC (stress-diet) and BD (physical activity- statins intake) have also been logically explainable. Interaction of factors of stress and eating behavior is well known in
the medical communities. It has been studying for a long time and described in a number of recent publications [9-12]. This interaction appears reactive eating behavior — anorexia nervosa (typical for sympathicotonics) and bulimia nervosa (typical for vagotonics) are extreme manifestations of it.
The motivation of the patient to lead a healthy lifestyle is the explanation of the of sign’s interaction of physical activity and statin’s intake. If a person motivated to lead a healthy lifestyle, thus he more strictly observes the guidelines for treatment, including statin therapy [13-16].
Thus, if relations between signs in the model will be changed, they can redistribute a significant impact and have a totally different contribution to the final picture. This probable can be explained such a multidirectional process of atherosclerosis for the residents of besieged Leningrad and prisoners of Nazi camps.
At the same time it is worth noting limitations of this study: the small sample size and selectivity of the surveyed group. But it was devoted to the study of contemporary issues — the effects of stress on ordinary patients from real clinical practice.
If mainly males (especially the military and extreme professions) have been studied in previous studies associated with the effects of stress and the risk of CVD, but recently more and more attention paid to the presence of gender differences in the impact of stress in daily life.
Slopen N et al. showed in their study, that cardiovascular events during 10-year follow-up of almost 40% more likely to be registered in women experiencing a lot of stress at work than their colleagues who noted the minimum level of stress in the workplace [17].
An interesting fact was obtained in this study: that more than 70% contribution to the existence of the relationship between stress in the workplace and the
54
Assessing the impact of workplace stress on total cholesterol and low density lipoprotein levels: the example,,.
risk of cardiovascular events could not be explained by the presence of traditional risk factors for CVD.
This study was a part of a ten-year Women’s Health Study. In this study were included more than 22,000 women who were health care workers. They were interviewed for the presence of stress at work, including in relation to the intensity and workload requirements, skills and control over decision-making. 170 cases of myocardial infarction, 163 cases of ischemic stroke, 440 cases of coronary revascularization and 52 cases of death from cardiovascular disease have been reported in over 10 years of observation.
The following data were obtained from Cox proportional hazards models: women with high level of stress tension at workplace 38% more likely to have the chance of developing cardiovascular events than their colleagues who reported low level of of stress tension at workplace (rate ratio 1.38; 95% CI 1.08 -1.77).
Similar data are available regarding the influence of nervous tension in the workplace and the risk of hypertension and coronary heart disease in women. The aim of a recent meta-analysis of observational studies on hypertension has been explore whether the effect of emotional stress at work in the formation of arterial hypertension in the future.
The combined relative risk of developing of arterial hypertension, resulting from the analysis of data from all nine studies included in the analysis, was 1.3 (95% CI 1.14 to 1.48; p <0.001), from “case-control” studies — 3.17 (95% CI 1.79 to 5.60; p <0.001), from cohort studies — 1.24 (95% CI, 1.09 1.41; p <0.001) [18].
Preliminary results of the study carried out by Samad Z et al. have shown that there are psychological and physiological differences in how men and women with stable coronary artery disease respond to the impact of stress. The women had mental stress-induced myocardial ischemia symptoms more often than men (57% vs. 41%, p <0.04), but this was not observed no significant differences by gender in the identification of signs of myocardial ischemia after exercise test [19].
In this regard, it is worth noting the presence of another limitation of our study — in this study, we did not put the question ofwhether there are gender differences in the effects of stress in everyday life at risk for developing of cardiovascular disease. The study was not statistically significant differences between groups studied in relation to nervous tension in the workplace by gender.
Conclusion
Stress is a known risk factor for CVD, both directly and indirectly through others, in particular lipid factors. The impact of workplace stress on lipid factors was confirmed in this study. But the contribution of stress in the development of a common model was less important than other factors, and its influence revealed through sign’s interaction. Conducting a specially-designed randomized clinical trials is expedient to assess the role of stress on lipid factors. Log-linear analysis may be used to decision of tasks in medical studies with a small samples.
Conflict of interests: author declare no conflicts of interest.
References:
1. Yusuf S., Hawken S., Ounpuu S. et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the Interheart study): case-control study. Lancet. 2004 Sep 11-17;364 (9438):937-52.
2. Lanas F,. Avezum A., Bautista L.E., et al. Risk factors for acute myocardial infarction in Latin America: the Interheart Latin American study. Circulation. 2007 Mar 6;115 (9):1067-74.
3. Blaga F. On the pathogenesis of arteriosclerosis. Arkh Patol. 1963;25:13-21.
4. Typical pathological processes. Pathophysiology of metabolism. Textbook for students of higher medical schools, Samara, 2008, 100 p. UDC 612.02. Bibl. 6, Tab. 5. In Russian (Типовые патологические процессы. Патофизиология обмена веществ. Учебное пособие для студентов высших учебных медицинских заведений, Самара, 2008, 100 с. УДК 612.02. Библ. 6, Табл. 5).
55
Section 3. Preventive medicine
5. Volkova K. G. About aortic atherosclerosis in alimentary dystrophia. Proceedings of the Leningrad branch of the All-Union Institute of Experimental Medicine. L., 1946; p.105. In Russian (Волкова К. Г. Об атеросклерозе аорты при алиментарной дистрофии. Труды Ленинградского филиала Всесоюзного института экспериментальной медицины. Л., 1946; с. 105).
6. Gaisenok O., Martsevich S., Tripkosh S., Lukina Y. Analysis of lipid-lowering therapy and factors affecting regularity of statin intake in patients with cardiovascular disease enrolled in the PROFILE registry. Rev Port Cardiol. 2015 Feb;34 (2):111-6.
7. Martsevich S.Yu., Gaisenok O. V., Tripkosh S. G., Zagrebelnyy A. V., Lukina Yu. V. Medical supervision in specialized center and the quality of lipid-lowering therapy in patients with cardiovascular diseases (according to the PROFILE register). Ration Pharmacother Cardiol 2013;9 (2):133-137. In Russian (Марцевич C. Ю., Гайсёнок О. В., Трипкош С. Г., Загребельный А. В., Лукина Ю. В. Наблюдение в специализированном медицинском центре и качество гиполипидемической терапии у пациентов с сердечно-сосудистыми заболеваниями (по данным регистра ПРОФИЛЬ). РФК 2013;9 (2):133-137).
8. National Guidelines on Cardiovascular Prevention. First edition 2011. Cardiovascular therapy and prevention 2011; 10 (6) Appendix 2. In Russian (Национальные рекомендации по кардиоваскулярной профилактике. Кардиоваскулярная терапия и профилактика 2011; 10 (6) Приложение 2).
9. Ulrich-Lai Y.M., Fulton S., Wilson M., et al. Stress exposure, food intake and emotional state. Stress. 2015 Aug 13:1-19.
10. Tajik E., Latiffah A.L., Awang H., et al. Unhealthy diet practice and symptoms of stress and depression among adolescents in Pasir Gudang, Malaysia. Obes Res Clin Pract. 2015 Jul 20. pii: S1871-403X (15)00099-X. doi: 10.1016/j.orcp.2015.06.001.
11. Pelletier J.E., Lytle L.A., Laska M.N. Stress, Health Risk Behaviors, and Weight Status Among Community College Students. Health Educ Behav. 2015 Aug 13. pii: 1090198115598983.
12. El Ansari W, Suominen S, Berg-Beckhoff G. Mood and food at the University of Turku in Finland: nutritional correlates of perceived stress are most pronounced among overweight students. Int J Public Health. 2015 Jul 23. [Epub ahead of print]
13. Urbinati S., Olivari Z., Gonzini L., et al. Secondary prevention after acute myocardial infarction: Drug adherence, treatment goals, and predictors of health lifestyle habits. The BLITZ-4 Registry. Eur J Prev Cardiol. 2014 Dec 1. pii: 2047487314561876. [Epub ahead of print]
14. Fuster V An alarming threat to secondary prevention: low compliance (lifestyle) and poor adherence (drugs). Rev Esp Cardiol (Engl Ed). 2012 Jul;65 Suppl 2:10-6.
15. Patrick A.R., Shrank W.H., Glynn R.J., et al. The association between statin use and outcomes potentially attributable to an unhealthy lifestyle in older adults. Value Health. 2011 Jun;14 (4):513-20. doi: 10.1016/j.jval.2010.10.033. Epub 2011 Apr 22.
16. Pitkala K.H., Strandberg T.E., Tilvis R.S. Interest in healthy lifestyle and adherence to medications: Impact on mortality among elderly cardiovascular patients in the DEBATE Study. Patient Educ Couns. 2007 Jul;67 (1-2):44-9.
17. Slopen N., Glynn R.J., BuringJ.E., et al. Job strain, job insecurity, and incident cardiovascular disease in the Women’s Health Study: Results from a 10-year prospective study. PLoS One 2012; DOI:10.1371/jour-nal.pone.00450512. Available at: http://www.plosone.org.
18. Babu G.R., Jotheeswaran A.T., Mahapatra T, et al. Is Hypertension Associated With Job Strain? A Metaanalysis of Observational Studies. Occup Environ Med. 2014;71 (3):220-227.
19. Samad Z., Boyle S., Ersboll M., et al. Sex differences in platelet reactivity and cardiovascular and psychological response to mental stress in patients with stable ischemic heart disease: Insights from the REMIT study. J Am Coll Cardiol 2014; 64:1669-1678. Available at: http://content.onlinejacc.org/journal.aspx.
56
