Lifetime Trauma, Praying for Others, and C-Reactive Protein

Abstract

Research indicates that praying for others may offset the effects of stress on self-rated health and psychological well-being. The purpose of the current study is to extend this literature by seeing whether praying for others moderates the effects of exposure to lifetime trauma on a key marker of inflammation: C-reactive protein. The data come from a recent nationwide survey of adults of all ages (N = 1,589). Levels of C-reactive protein were obtained from assays of blood spots drawn from a capillary fingerstick. The findings suggest that the magnitude of the relationship between lifetime trauma and C-reactive protein is completely offset for study participants who frequently pray for others. The theoretical implications of this research are discussed.

Funding

Funding for this study was provided by a grant from the John Templeton Foundation.

Supplementary

Supplemental data for this article can be accessed on the publisher’s website.

Notes

1. Subtracting the sum of the cases associated with each reason for exclusion from the original sample (N = 3,010) does not equal the number of cases in the subsample that was used in the analyses (N = 1,589). This occurred, for example, because some individuals who do not pray also refused to give a sample of their blood (i.e., the reasons for exclusion are not mutually exclusive).

2. Some investigators exclude study participants with CRP values in excess of 8.6. This approach is taken because values in excess of 8.6 may reflect the influence of a physical injury (Herd, Karraker, and Friedman 2012). The analyses for the current study were repeated after excluding study participants with CRP values greater than 8.6 (N = 78). The findings (not shown in Table 2) indicate that the proposed statistical interaction effect between praying for others and lifetime trauma is still present in the data (b = −.034; p < .01). The fact that the effect size is somewhat smaller is understandable given that the upper tail of distribution has been eliminated. The fact that the level of significance is lower may be explained, in part, by the loss of cases.

3. One might wonder why the overall frequency of private prayer was not included in the model, as well. This issue was examined in an additional set of analyses. As shown in Table 1, the interaction involving praying for others and trauma (b = −.078) is significant at the .001 level. Including an interaction between the frequency of private prayer and trauma produced two key findings. First, the interaction between the frequency of private prayer and trauma was not statistically significant (b = −.002; ns). Second, the interaction between praying for others and trauma remained virtually unchanged (b = −.077). However, the level of statistical significance dropped from .001 to .05. This suggests that multicollinearity might be a problem. This is not surprising because the multiplicative terms associated with the two measures of prayer both contain trauma. Consistent with this view, the bivariate correlation between the two multiplicative terms is: r = .801; p < .001. Moreover, the frequency of private prayer subsumes praying specifically for others. Evidence of this may be found in the correlation between the two measures of prayer (r = .783; p < .001). Due to these issues, the frequency of private prayer is not included in the main study model.

4. Our discussion of why it is important to focus on praying for others may create the impression that people who have experienced more lifetime trauma are also more likely to pray for others in the first place. Even so, the bivariate correlation between lifetime trauma and the frequency of praying for others (r = .041; ns) suggests this is unlikely.

Additional information

Funding

Funding for this study was provided by a grant from the John Templeton Foundation.