Good health from a good gossip?

Behind the Headlines

Friday June 12 2009

The office rumour mill in full effect

“Gossip is good for women's health,” according to The Daily Telegraph. “Men might argue that the last thing women need is another reason to gossip but scientists have discovered that it could be good for their health,” the newspaper added.

The research checked levels of the hormone progesterone in pairs of female students who performed a bonding exercise, in which they answered a number of preset questions designed to make them share personal information. These socialising women showed increases in progesterone compared to women who were given a group reading task.

It is important to note that the women talked about themselves in this study, rather than others, which is not what would be generally considered as gossiping. It is also unclear whether changes in progesterone levels would result in improvement in health. The study in 160 women is also missing some data.

Overall, the research may improve our understanding of the biological effects of bonding between women, but it does not prove that “gossip is good for women’s health”.


Where did the story come from?

Professor Stephanie Brown from the Health Services Research and Development Center of Excellence in Ann Arbor, Michigan, and international colleagues carried out this research. The study was supported by a grant from the National Institutes of Health in the US and was published in the peer-reviewed medical journal Hormones and Behaviour.


What kind of scientific study was this?

This was a randomised controlled trial in which the researchers tested their theory that pairs of women who performed a bonding exercise would have higher levels of the hormone progesterone in their saliva than women in a control group, who performed a reading and editing exercise. 

The authors of this study explained that progesterone is a female hormone produced by the ovaries, and that it has already been shown to be linked to an individual's motivation to bond with others. Women with higher levels of progesterone are apparently more satisfied by positive interpersonal relationships, and this motivation to bond is higher in women who take oral contraceptives (containing progestogens) than in women who don't take oral contraceptives or in men.

The researchers recruited 160 female college students and randomly grouped them into 80 pairs of women who already knew each other before the study. They randomly assigned half the pairs to a task designed to bring them closer together by verbally answering a number of preset questions.

The pairs assigned to the closeness task were told that the purpose of this task was to “get to know each other” and were offered 16 questions to ask each other. These included, “Given the choice of anyone in the world, whom would you want as a dinner guest?” and, “What is the greatest accomplishment of your life?” Partners took turns answering each question first.

The other 40 pairs were allocated to a control group and asked to proofread a research paper on botany together. One woman read aloud an edited version of the paper, which contained no errors, while their partner checked it against an unedited version and corrected as many errors as possible.

The volunteers had saliva samples taken to check their hormone levels before their task and 20 minutes after completion. The researchers measured both progesterone and another hormone, cortisol, which is known to increase with stress. All tests were done between midday and 7pm to allow for the natural variation of hormone levels throughout the day.

They also completed an assessment called the Inclusion of Others in Self (IOS) test, in which the participants defined their relationship to their test partner. This test required subjects to mark a chart, which featured a number of overlapping circles representing interpersonal relationships, to establish how individuals were connected to their partner in the study.

The researchers asked women to score on a five-point response scale how strongly they agreed with the statement, “I would risk my life for [my study partner].”

The participants were also randomly assigned to play a computerised card game with either their partner from their previous task or a new partner. They returned a few weeks later to play another session of the game. Their progesterone levels were measured before and after each game, and they were once again asked to rate their closeness to their partner and their willingness to risk their lives for them.


What were the results of the study?

Reliable hormone data was available in 141 out of 160 women. In the pairs of women who answered social questions, progesterone levels either stayed the same or increased. In the control group, progesterone levels decreased. There was no change in cortisol levels in either group.

Adjusting the results to account for progesterone levels at the start of the study (baseline), the researchers reported that the average post-task progesterone level was 47.62 picogrammes/ml in the women who carried out the closeness test, compared to a control group average of 37.68 picogrammes/ml. This was a statistically significant difference.

The IOS test results showed that those taking part in the closeness induction session felt closer to their partners than those participating in the editing task.

In their analysis, the researchers said that the change in progesterone during the first session (involving the closeness task or the editing task) was not related to the altruistic “willingness to sacrifice themselves for the partner”. However, changes in progesterone in the second session (the card game) a week later was linked to the altruistic "willingness to sacrifice themselves for the partner".


What interpretations did the researchers draw from these results?

The researchers say that their research is the first to show that hormonal changes (increased progesterone but not cortisol) are associated with an experimental manipulation of closeness. It also links progesterone to self-reported willingness to risk one's own life for another person.


What does the NHS Knowledge Service make of this study?

This study was relatively large, with 160 recruits, and it used validated scores and tests to measure perceptions and hormones. There are a few points to note:

  • The proportion of women with missing data was relatively high (12%) and differences between the proportions of women with missing data in the two groups could have caused inaccuracy in the results. The effect of missing data was not investigated or discussed in the report.
  • The baseline characteristics of the two groups are not reported, which means that, even though the groups were randomised, we cannot be sure that they were similar enough at the start of the study to provide a fair comparison.
  • The accuracy of the measurement of progesterone is not discussed in this study. Levels of the hormone might naturally vary throughout the day, even hour by hour or across the month, and it would be important to take these factors into account during analysis.
  • The study only measured levels of two hormones and not health outcomes or happiness. It is unclear whether the differences in hormone levels seen would result in any differences in health outcomes or happiness.

The lead author, Professor Brown, said "it's important to find the links between biological mechanisms and human social behaviour. These links help us understand why people in close relationships are happier, healthier and live longer than those who are socially isolated." This study furthers progress along this research path, but cannot be interpreted as showing that "gossip is good for health".

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

Gossip is good for women's health, scientists claim. Daily Telegraph, June 11 2009 

Gossiping is good for you. Daily Express, June 11 2009

Links to the science

Brown SL, Fredrickson BL, Wirth MM et al. Social closeness increases salivary progesterone in humans. Horm Behav, June 2009; Volume 56, Issue 1, p108-111


How helpful is this page?

Average rating

Based on 0 ratings

All ratings

0  ratings
0  ratings
0  ratings
0  ratings
0  ratings

Add your rating