Psychological Stress and Cellular Aging in Cancer: A Meta-Analysis

Table 1

Characteristics of the most relevant epidemiological studies on psychosocial factors and risk for breast cancer.

First author/reference/location	Study design/sample	Follow-up years	Exposure	Effect size (95% CI)	Control for confounding
Kruk, 2012 [23] Poland	Hospital-based case-control study; 858 cases, 1085 controls	January 2003-May 2007	Death of a close family member	(1.70-3.64)	Age, lifetime recreational physical activity, place of residence, education, age at menarche, breastfeeding, family history of breast cancer, exposure to cigarette smoke, alcohol consumption, and intake of vegetables and fruits
			Personal injury, illness	(1.63-4.62)
			Imprisonment, trouble with law	(1.56-5.45)
			Retirement	(1.08-2.45)
			Lifetime	(3,41-8,50)
Wang et al. 2013 [24] Taiwan	Case-control study (157 cases, 314 controls)	June 2009-June 2011	High perceived PS	(1.10-2.47)	Adjusted for potential lifestyle factors
			Joint interactions high perceived PS with the following:
			Alcohol	(1.23-6.86)
			cigarette/day	(1.16-5.47)
			Low physical activity	(1.77-6.36)
			High fried and stir-fried food	(1.79-5.63)
			High meat and sea food intake	(1.09-3.27)
Li et al. 2016 [25] China	Comparative case-control study 582 cases, 540 controls	May 2013-May 2015	Frequent depression	(1.00-1.75)	Adjusted for main risk factors
			Negative emotional experiences	(1.03-1.29)
			Disharmonious marital status	(1.06-1.26)
Sawada et al. 2016 [26] Japan	Prospective study 29,098 women, 209 cases identified	21 years	Psychological traits:		Age, study area, education, family history of breast cancer, age at menarche, age at menopause, parity, use of exogenous female hormones, alcohol intake, daily walking, exercise, sedentary work, height, and BMI
			“Ikigai” agree strongly	(0.41-1.62)
			Decisiveness agree	(0.62-1.85)
			Ease of anger agree	(0.50-1.76)
			Perceived stress agrees strongly	(0.56-1.78)
Schoemaker et al., 2016 [27] UK	Prospective cohort study 106,000 women, 1783 cases	2003-July 2012	Breast		Age, age at menarche, age at first birth, parity breastfeeding, hormone use, BMI, smoking, alcohol intake, physical activity, family history of BC, and socioeconomic status
			Death of close relatives	(0.78-0.97)
			Death of husband/partner	(0.88-1.46)
			Divorce/separation	(0.96-1.38)
			Personal illness/injury	(0.87-1.22)
			Mather death	(1.02-1.67)
Yeh and Lee 2016 [28] Taiwan	Prospective cross-sectional study, 54 cases, 1106 controls	On day prior mammography	Borderline anxiety vs no anxiety	(1.685-5.698)	These risks were adjusted only for educational factors. Results of multiple logistic regression models presented goodness-of-fittest
			Anxiety vs no anxiety	(1.009-4.684)
			Depression vs no depression	(1.643-12.308)
			Stress	(1.062-1.190)
Özkan et al. 2017 [29] Turkey	Hospital based case-control study 491 cases, 512 controls	September 2013-September 2014	Experience of stressors in the last 5 years	(3.767-8.511)	Age, BMI, family history of cancer, marital status, employment status, and economic conditions
			Insufficient social support perception	(1.371-3.424)
			Avoidance coping strategy	(1.271-2.785)
			Child trauma	(1.14-1.91)
			Major life events	(1.22-2.53)
			Chronic stress	(1.52-2.67)
Butow et al. 2018 [30] Australia	Prospective cohort study (2,739 women, 103 cases)	May 2001-December 2010 (mean follow-up 7.2 years)	Experience of stressors in the last 3 years: severe (death of a spouse/child, handicapped child requiring full-time care) moderate (e.g., buying/selling a home)	Lack of statistical significance in all unadjusted and adjusted models	Family history of cancer, age at menarche, ovarian cancer, physical activity, HRT use, oral contraception use, BMI, anxiety, depression, parity, smoking, and number of live births
Fischer et al. 2018 [31] USA	Case-control study (664 cases, 203 population-based controls)	March 1st 1994	Cumulative adverse life events	(1.00-2.66)	Adjusted for age, age at first full-term pregnancy, menopausal status, family history of BC, HRT use, smoking, race/ethnicity, education level, and physical activity
		February 28 1995	Previous personal illness: Perceived as stressful	(1.96-4.11)
			Perceived as non-stressful	(1.34-8.94)
Yildirim et al. 2018 [32] Italy	Hospital-based case-control study (250 cases, 250 controls)	September 2013	Loss of father during childhood	(1.30-5.52)	Adjusted for age, education status, marital status, work status, economic status, social security, age at menarche, age at first pregnancy and birth, family history of cancer, and history of psychiatric disorder
		September 2014	Serious stressor within the last five years	OR = 4.72 (3.18-7.03)
			Inadequate social support	OR = 1.83 (1.23-2.73)

Abbreviations: HR: hazard ratio; OR: odds ratio; RR: relative risk; CI: confidence interval; BMI: body mass index; PS; psychological stress; HRT: hormone replacement therapy.