Sprehn et al1 investigated the correlation between marital status at cancer diagnosis and relative survival (RS) using the Surveillance, Epidemiology, and End Results (SEER) database. They reported that men who were separated from their spouses had lower RS compared with all other groups; whereas, among women, RS was decreased similarly among the widowed and the separated. Sprehn et al emphasized stress immunology as a pivotal mechanism but presented no relevant analyses to support that hypothesis, relying instead on unproven lay beliefs about the role of stress and immune function in the progression of cancer. Although SEER data are limited by a lack of information concerning comorbidity and many relevant treatment variables, the analyses that were presented failed to consider even the basic demographic, biomedical, and treatment covariates available to help explain their exploratory findings.

Sprehn et al cited dated literature concerning possible links between stress and cancer. A recent systematic review2 investigating the contribution of stress to cancer incidence and mortality produced strong evidence of positive publication bias. Even so, confidence intervals for stress-cancer associations barely excluded 1.0, and poor social support was not associated significantly with cancer outcomes. Evidence for a stress-immune functioning-illness pathway is much less convincing for cancer than for other diseases,3 and any effects are likely to vary markedly across cancer sites. Unfortunately, the analyses presented by Sprehn et al were stratified only by sex, combined cancer sites in the analyses, and failed to examine potentially important and available confounders.

Despite producing no relevant evidence themselves, Sprehn et al encouraged cancer patients to seek psychological interventions to improve their survival by influencing immunity. To support this, the authors cited psychoneuroimmunology literature that relied on an arbitrary sampling of immune parameters from complex, highly redundant systems. Changes in specific immune parameters related to interventions do not signal necessarily improved immunity.4 In addition, the authors ignored findings that no psychological intervention study in which survival was an a priori outcome and in which there was no confounding of psychological with medical intervention has ever identified an effect on survival.5 By staking their claims on limited and irrelevant data from their study and ignoring a mass of contradictory data from other studies, Sprehn et al are not responsive to the weight of evidence and mislead any vulnerable patients who might be paying attention.


  1. Top of page
  • 1
    Sprehn GC, Chambers JE, Sayhkin AJ, Konski A, Johnstone PA. Decreased cancer survival in individuals separated at time of diagnosis: critical period for cancer pathophysiology? Cancer. 2009; 115: 5108-5116.
  • 2
    Chida Y, Hamer M, Wardle J, Steptoe A. Do stress-related psychosocial factors contribute to cancer incidence and survival? Nat Clin Pract Oncol. 2008; 5: 466-475.
  • 3
    Cohen S, Janicki-Deverts D, Miller GE. Psychological stress and disease. JAMA. 2007; 298: 1685-1687.
  • 4
    Miller GE, Cohen S. Psychological interventions and the immune system: a meta-analytic review and critique. Health Psychol. 2001; 20: 47-63.
  • 5
    Coyne JC, Stefanek M, Palmer SC. Psychotherapy and survival in cancer: the conflict between hope and evidence. Psychological Bull. 2007; 133: 367-394.

Laura J. Hanisch PsyD*, James C. Coyne PhD*, Steven C. Palmer PhD*, * Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania.