Stressors, Stress Response, and Cancer Recurrence

A Systematic Review

Briana L. Todd, MA, MS; Michal C. Moskowitz, MS; Alicia Ottati, MA; Michael Feuerstein, PhD, MPH


Cancer Nurs. 2014;37(2):114-125. 

In This Article


Systematic Review Search Strategy

A literature search was conducted using the electronic databases PubMed, Embase, Web of Science, and Cochrane (Figure 1). Limits on the search included studies published until April 2012 (ie, no limits were placed on the start date of the search, and the earliest date captured was August 1974), studies in the English language, and studies on humans. For PubMed, the MeSH search terms included (neoplasm [majr]) AND (neoplasm recurrence, local OR recurrence) AND (anxiety OR depression OR stress, psychological OR life change events). For Embase, the Emtree search terms included (cancer recurrence [exp/mj]) AND (stress [exp] OR anxiety [exp] OR depression [exp]). When searching Cochrane, the keyword search terms included (cancer) AND (recurrence) AND (stress OR anxiety OR depression OR stressful life event). The search terms for Web of Science included cancer, recurrence, stress, stressful life events, anxiety, and depression.

Figure 1.

Flowchart of the study selection process.

To be included, studies were required to measure stress in at least 1 of 3 ways: (1) environmental exposure to stressors such as cumulative life events that require adaptation (eg, death of a spouse), (2) psychological stress response such as measures of depression, anxiety, or mood in general, and (3) biological measures of either (a) direct stress response, for example, cortisol as an indicator of HPA axis activation, or (b) immune markers that are downregulated by activation of the stress response (eg, natural killer cell activity). These various modes of stressor exposure and stress response measurement are frequently used in the stress and health literature. All studies were required to include cancer recurrence as an outcome measure.

Exclusion criteria were as follows: cancer recurrence not an outcome variable (ie, dependent variable), stressor exposure and/or stress response not an independent variable (ie, exposure), stressor exposure and/or stress response did not proceed recurrence (ie, did not show prospective relationship), cancer was not examined, metastatic cancer was solely examined, not original articles, or cross-sectional, retrospective, or case studies. Two reviewers independently searched the titles and abstracts of the search results. When abstracts were ambiguous about meeting the inclusion/exclusion criteria for this review, the reviewers read the full text of the articles to determine inclusion. A third reviewer settled any disagreements in inclusion or exclusion between the first 2 reviewers. Reference lists were also examined to identify further articles for inclusion. Of 990 articles (77 duplicate articles were removed), lack of agreement occurred between the 2 initial reviewers on 18 articles. The interrater agreement was 98.18% (972/990).