Advertisement
Research Article

Social Relationships and Mortality Risk: A Meta-analytic Review

  • Julianne Holt-Lunstad equal contributor mail,

    equal contributor Contributed equally to this work with: Julianne Holt-Lunstad, Timothy B. Smith

    julianne_holt-lunstad@byu.edu

    Affiliation: Department of Psychology, Brigham Young University, Provo, Utah, United States of America

    X
  • Timothy B. Smith equal contributor,

    equal contributor Contributed equally to this work with: Julianne Holt-Lunstad, Timothy B. Smith

    Affiliation: Department of Counseling Psychology, Brigham Young University, Provo, Utah, United States of America

    X
  • J. Bradley Layton

    Affiliation: Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States of America

    X
  • Published: July 27, 2010
  • DOI: 10.1371/journal.pmed.1000316

Reader Comments (2)

Post a new comment on this article

Staying alive

Posted by gsilverm on 30 Jul 2010 at 17:20 GMT

Put another way, an OR of 1.5 means that by the time half of a hypothetical sample of 100 people has died, there will be five more people alive with stronger social relationships than people with weaker social relationships.
http://plosmedicine.org/article/info:doi/10.1371/journal.pmed.1000316#article1.front1.article-meta1.abstract3.sec3.p1

Rather than five more alive, I believe this should read, "there will be ten more people alive with stronger social relationships than people with weaker social relationships."

If this sample of 100 people is made up of 50 people with "stronger social relationships" and 50 people with "weaker social relationships," and half of the total sample stays alive, with a 50% higher survival rate in the strong relationships group, then we can conclude that 30 out of 50 people with stronger relationships stayed alive and 20 out of 50 people with weaker relationships stayed alive.

Calculations:
Strong relationships survival rate = 60%
Weak relationships survival rate = 40%
This satisfies the requirement that there be a 50% higher survival rate in the strong relationships group: 60%/40% = 1.5

Strong relationships survival = 60% * 50 = 30
Weak relationships survival = 40% * 50 = 20
Total surviving = 20 + 30 = 50 = 50%
This satisfies the requirement that half the total sample dies.

No competing interests declared.

RE: Staying alive

AlisonZ replied to gsilverm on 31 Jul 2010 at 09:13 GMT

Agree.
The maths is incorrect as pointed out by Staying Alive, but if it was correct the phasing would be nonsense.

Of the 50 surviving people it would be a challenge to have 5 more of one group than the other-

(22.5 people + 5 people) + 22.5 people =50 people,

oops half a person couldn't really be said to be surviving.

No competing interests declared.

RE: Staying alive

lennertveerman replied to gsilverm on 05 Nov 2010 at 06:02 GMT

If indeed the sample consisted of 50 persons in each group - that is an assumption the reader has to make; it's not given in the text.

No competing interests declared.

RE: Staying alive -- Response by the authors

TimSmith replied to gsilverm on 27 Oct 2012 at 14:26 GMT

Thank you for taking the time to review our manuscript and respond. We need to clarify that our calculation in the manuscript is an "odds ratio," which is the metric used throughout the manuscript. The calculation is accurate.

These online comments/posts (above) involved a "rate ratio," which is an different calculation. Either method is viable, but we used an odds ratio to be consistent with the rest of the manuscript. We agree that the wording in the text could have been clearer.

No competing interests declared.