Back to contents

The best of friends

Gus Cairns
Published: 03 December 2009

Gus Cairns explores how friends can be good (and sometimes bad) for your health.

Recently we’ve been hearing a lot about swine flu, and how you should stay at home if you’ve got it. Makes sense: after all, these viruses are spread by contact. So if you keep in close proximity to others, you’re more likely to get sick, aren’t you?

Friends and colds

Not if a pioneering study conducted in the late 1990s in Pittsburgh, USA holds as true for flu as it does for that other air-borne misery, the common cold.1 It found that, on the contrary, the fewer social contacts people had, and the smaller the variety of their social contacts, the more likely they were to suffer from colds. Not to become infected – but to become ill once they were infected. The greater number and variety of friends people had, the less likely they were to develop cold symptoms.

It was a strong effect: people with the fewest social contacts were four times more likely than the most sociable people to get streaming colds if infected. The absolute number of contacts people had was not as important as the number of types of contacts people had.

In the study, 276 people aged 18 to 55 had a health screen, risk factors like smoking and stress were evaluated and a psychological screen gauged their personality type.

They were asked to number the people they’d had social contact with, in person or by phone, in the last two weeks and to divide these into categories including spouse, family, close friends, workmates, neighbours and members of social groups. They were exposed to cold viruses and then kept in quarantine while their cold symptoms were assessed by objective criteria, which included collecting and weighing their soggy tissues for ‘mucous production’.

Sociability was the strongest predictor of cold symptoms. Compared with the most sociable people, the least sociable people produced more mucous and their nasal passages were less efficient at getting rid of it.

The researchers found one chemical difference between the social and the non-social people: the latter had higher levels of adrenaline and noradrenaline - they were more stressed. Friends, in short, are good for your immune system.  

All very well, but is friendship able to influence potentially lethal illnesses to the same extent as it can a relatively trivial illness like a cold?

Very much so. Numerous studies have found that the number and variety of friends people have, has a directly positive effect on health outcomes. A study of women with breast cancer, for instance,2 found that a higher number of supportive friends, more contact with them and a larger social network independently predicted better survival rates, and the effect was nearly as strong a predictor of survival as the stage the cancer had reached when detected.

Another medical outcome strongly affected by social ties is the chance of having a second heart attack after you’ve had a first. A review of various studies3 found that people with a good social network had better heart functioning and a longer period between the first and any subsequent cardiovascular event. This wasn’t just because friends remind you to eat well, sleep well and stress less (though the influence of friends on behaviour is very important, as we’ll see below): having good social contacts appeared to have a directly beneficial effect on immune functioning and heart function.

Finally, it’s been known for 30 years that friendship, or lack of it, has a direct effect on mortality: a long-standing study of 6928 adults in California4 found that, regardless of age, men who lacked social and community ties were 2.3 times more likely, and women 2.8 times more likely, to die during the study period.

Loneliness and HIV

Being diagnosed, and living, with HIV can be a significant cause of social isolation. Loneliness can be caused by stigma and rejection, fear of stigma and self-isolation, and, especially in the earlier days of the epidemic and still in some settings, by having to deal with multiple bereavement, caring for the dying, and facing one’s own possible death.

Early studies certainly found a correlation between social isolation and factors like the speed of CD4 count decline. A 1993 study, for instance,5 found that people with HIV infection, largely symptomatic, had faster CD4 declines if they had poorer social support.

Loneliness is also predictive of poor adherence. A 2004 study of 90 people with HIV found that greater social support related to better adherence, whereas higher depression scores related to non-adherence.6 Having friends didn’t, it seemed, work directly to improve adherence in the sense of friends reminding each other to take their meds, but produced a positive state of mind in which people were more committed to their own health.

I found support groups and I found dating websites. But there was nothing in between.

Michael Patel,

Founder of PlusFriends

Friendship has also been studied in HIV prevention, where studies have experimented with training up influential people in existing social networks to act as peer-group safer-sex advocates. This ‘popular opinion leader’ concept has had its greatest successes among groups who are marginalised and where there is a corresponding sense of community solidarity, for instance young gay black men in the USA and young Roma (gypsy) men in eastern Europe. It works less well in situations where social bonds are not already strong.

An insight into how marginalisation can be overcome to create a healthier lifestyle comes from a survey of 154 young gay Asian men in New York.7 It found that, while experiences of homophobia and racism were associated with higher levels of sexual risk-taking, talking to gay friends about this discrimination was associated with lower levels of risk. It appears that men who didn’t have friends to talk to were more likely to blame themselves, both for being gay and for risky sexual encounters.

The Plus Friends group

Diagnosis can still feel like a lonely time even if you have a good circle of friends. Michael Patel was diagnosed at the age of 36 in 2005.

“I had a good social life and a lot of wonderful friends, but not one of them had HIV,” he says. “I didn’t feel I had to explain it to them, but they would say things like ‘I understand what you’re going through’ and I’d be thinking ‘You really don’t’.”

He went to a support group for newly diagnosed gay men, but didn’t enjoy the experience.

“I hated it. Support groups are very good and help a lot of people but they aren’t for me, and I suspect aren’t for a lot of people. It took me three weeks to say anything and I hated the way that in a support group you feel a kind of spotlight turning to you and you have to say something.

“Now, I’m the sort of person who’s learned to do things by myself, to ‘pick myself up, dust myself off, start all over again’. I’m too proud to break down in public, in a support group where initially nobody knows anybody. I did make friends via the support group, but in terms of what it actually did, I’d leave feeling really bad about myself – worse than when I came in.”

And yet, he says, he knew he badly needed support and the right kind of opportunity to talk. What he needed, in short, was an understanding friend who had HIV themselves and somewhere to meet one.

“I Googled, I searched, I couldn’t find a social group for HIV-positive people anywhere. I found support groups and I found dating websites. But there was nothing in between.”

So he found an existing internet structure – the Meetup website, through which people with a similar interest can arrange social events – and formed the London Gay HIV Meetup Group. Which soon became, when he realised the ‘HIV’ was putting people off, PlusFriends.

At one time, I went to a few PlusFriends meetings. Why? I’d had HIV for 18 years and if all the counselling and workshops I’d had to explore my issues about HIV hadn’t worked yet, they weren’t going to. I wanted to socialise, and maybe date, in an atmosphere that wasn’t all about sex. And I didn’t want to disclose every time or spend the evening educating someone about HIV. Evenings ranged from a boozy outing watching the Rugby World Cup to dinner in a fancy restaurant.

The PlusFriends group now has about 350 members, though some are just internet buddies and only 15 to 20 people may come along to any one event. A core group meets fairly regularly and has regular activities such as a Sunday lunch or Saturday brunch, which often turns into an afternoon of museums, shopping and a disco night. For fans of healthier activities, there are regular walks.

“If you meet in a pub, there’s not the pressure to talk,” says Michael, “and there’s the opportunity to do it your way, at your own pace, in a one-to-one setting. With guys who are very nervous, I encourage them to meet with one other group member first and then come along.”

People come along for a variety of different motives. “There are guys who have been HIV-positive for a long time and would ideally like to find a partner who also has HIV. There are some older members who come along and make it their job to welcome others. There’s newly diagnosed guys who need to talk but don’t like support groups. And there’s guys who probably should go to support groups – I’ve had to shepherd home a few who were so nervous they drank too much.” Michael, an accountant by trade, says that, ironically, running the group has got him interested in training as a counsellor.

The bathing beauties

Gay men, of course, are not the only people with HIV who may want to make HIV-positive friends. One of the incidents that prompted this article was an outing I had with a sociable female friend of mine, of equally long diagnosis, and her friends, none of whom had HIV. We had a nice dinner but afterwards, almost inevitably, ended up chatting to each other in a corner about pills, doctors and CD4s. “It’s such a relief at times to talk to someone and not have to explain all this,” she said.

“So how do HIV-positive people who have outgrown support groups socialise?” asks Angelina Namiba of Positively Women.

“In the beginning,” she says, “especially when someone is newly diagnosed, support groups are not only a source of much-needed support and information. They are a place for networking and socialising too, especially for many who have not disclosed their status. Support groups may be the only place that helps them overcome the isolation they may face and many friendships are born out of accessing support groups.

“Amongst the many I now consider my very good friends, are women I have met over the years in various support groups and through the course of my work.”

“But,” she adds, “many of us positive dinosaurs who have been diagnosed for a few years now have outgrown support groups.

“So what do we do for fun? Ours is an informal group of positive women with a combined total of over 100 positive years! We go salsa dancing...some of us go to the theatre now and again as well as go to lots of the free jazz events around London in the summer.

“Earlier on this summer, one of our group, Julia, came up with the idea of a picnic…at Hampstead Heath. A number of us even braved the cold water of the ladies’ pond! It was great. There were six of us altogether. Everyone brought a dish and a drink to share. It’s something we will certainly be doing again come next summer!”

United we fall – when friends are bad for you

So: friendship and socialising are good for your mental and physical health?

Not always. Like everything in life, the picture is more complex than that. Sometimes friends get each other into bad habits.

One study in the late 1990s8 found that, contrary to expectation, among 205 HIV-positive men, lower levels of loneliness predicted more rapid declines in CD4 counts. These CD4 declines did not seem to have anything to do with medication adherence or sexual practices. Significantly, they also were not correlated with negative emotions.

An exception to the rule? No. A review of other studies of friendship and social support9 by the same authors found that, quite consistently, people who perceived themselves as having stronger links to friendship networks actually progressed faster to AIDS – if they were asymptomatic. Once they became sick, however, the opposite happened: supportive friends produced better health outcomes and survival.

People don’t seek friends in order to have a healthier lifestyle: they seek friendship because it makes them happier. But if those friends are doing things that are bad for your health, this can have the paradoxical effect that social integration can make you ill. The authors found that strong friendship networks tended to expose people to more temptation: they were more likely to take part in high-risk sexual behaviour and “socially facilitated health-compromising behaviour” such as drink and drugs.

Social ties don’t always help in prevention either. This was noticed in a study of a gay men’s prevention intervention that reported in 2004.10 Peer-led workshops in San Francisco and New York were designed to get gay men talking about the sexual risks they’d been taking and think of ways of reducing the risk. Overall, the intervention did not produce a significant reduction in unsafe sex, to the researchers’ disappointment. When they investigated why, they found that the lower-risk gay men were tending to adopt the sexual practices of the higher-risk men. Peer group influences were working – but in the wrong way.

The way this works has been demonstrated in an extraordinary series of studies by Nicholas Christakis and James Fowler of the Harvard School of Public Health. The town of Framingham, Massachusetts, has given its name to a measure of heart attack risk – your Framingham score – because a huge scientific study has been ongoing since 1948, quantifying the risks associated with heart attacks in three generations of New Englanders.

The original records of the study included the details of study participants’ close friends, colleagues and family members, compiled so that people who dropped out of the study could be traced. This allowed Christakis and Fowler not only to see if social integration predicted health behaviours, but also to see how members of the social network influenced each other over time. If one person in a closely connected group changed their behaviour, would others follow? How fast did a health behaviour spread through a network?

They investigated a troubling behavioural epidemic that has taken hold in the last few decades, and threatens the health of people worldwide – obesity.11 They found that in 1948, 10% of the Framingham study group was obese. In 1985, 18% of the 12,067 people they investigated were obese; but the rate accelerated fiercely and by 2005 40% of the population was obese.

A 2004 study of 90 people with HIV found that greater social support related to better adherence, whereas higher depression scores related to non-adherence.

Christakis and Fowler found that obesity spread as if it was a virus, radiating out from individuals. Clusters of people would become obese at the same time, and the closer you were to someone who was obese, the more likely you were to become obese yourself.

But for the purposes of this article, the most striking finding is that the most ‘infectious’ people were close friends. Having an obese spouse increased your chances of becoming obese by 37%, but having an obese friend increased the risk by 60%.

This ‘infection rate’ ensured that thin people whose friends were obese soon became very rare, and that obese people tended to be the most well-connected; there were two large, closely connected networks of predominantly obese people in the centre of the social network, and a lot of thin loners on the outside. Doesn’t this sound like the studies of closely related genetic ‘clusters’ of HIV in high-risk communities?

This all seemed to be related to social norms of diet and physical appearance. Interviewed in Wired Magazine earlier this year, Nicholas Christakis said: “A bunch of people discovered fast food at the same time. Then the network took over.”

Not smoking and smiling – both are infectious

Luckily, this means that if a healthy behaviour becomes accepted as a social norm, then friendship networks can have a reinforcing effect. Christakis and Fowler found that giving up smoking spread like a virus too.12 In 1971, 65% of the Framingham cohort smoked – higher than average, even for the time. By 2001 that was down to the national average of 25%. The ‘infection rate’ of a friend giving up smoking was 36% above what you’d expect if stopping smoking was random. Here, however, the influence of a spouse was stronger at 67%. Smokers did not give up one by one – characteristically, a whole group would give up at once.

Christakis told the New York Times: “It’s not like one little star turning off at a time,” he said. “Whole constellations are blinking off at once.”

And yes, happiness is infectious too. The Framingham participants rated their mood in psychological questionnaires over decades.13 It found that unhappy people surrounded by happy people soon stopped being unhappy. Who were the most efficient transmitters of happiness? Not spouses, whose ‘happiness infection rate’ was only 8%, but friends. A mutual friend’s infection rate was 25%, and if they lived nearby, that rate skyrocketed to 65%.  

Friends may not always make us well: but they are amongst the strongest causes of happiness.


1. Cohen S et al. Social ties and susceptibility to the common cold. JAMA 277(24):1940-1944. 1997.

2. Waxler-Morrison N et al. Effects of social relationships on survival for women with breast cancer: a prospective study. Soc Sci Med. 33(2):177-83, 1991.

3. Seeman Teresa E Social ties and health: the benefits of social integration. Ann Epidemiol. 6(5):442-51, 1996.

4. Berkman LF et al. Social networks, host resistance, and mortality: a nine-year follow-up study of Alameda County residents. Am J Epidemiology 109(2):186-204, 1979.

5. Solano L et al. Psychosocial factors and clinical evolution in HIV-I infection: a longitudinal study. Journal of Psychosomatic Research.37:39-51,1993.

6. Gonzalez JS et al. Social support, positive states of mind, and HIV treatment adherence in men and women living with HIV/AIDS. Health Psychol. 23(4):413-8, 2004.

7. Yoshikawa H et al. Social discrimination, friendship network interactions, mental health and HIV risk among Asian gay men. National HIV Prevention Conference, Atlanta, Georgia, abstract M2-A0301, 2003.

8. Miller GE et al. Social relationships and immune processes in HIV seropositive gay and bisexual men. Ann Behav Med. 19(2):139-51, 1997.

9. Miller GE and Cole SW Social relationships and the progression of human immunodeficiency virus infection: a review of evidence and possible underlying mechanisms. Ann Behav Med 20(3):181-189, 1998.

10. Wolitski RJ et al. Prevention with HIV-seropositive men who have sex with men: lessons from the Seropositive Urban Men's Study (SUMS) and the Seropositive Urban Men's Intervention Trial (SUMIT). JAIDS 37 Suppl 2:S101-9, 2004.

11. Christakis NA and Fowler JH The spread of obesity in a large social network over 32 years. New England Journal of Medicine, 357(4):370-379, 2007.

12. Christakis NA and Fowler JH The collective dynamics of smoking in a large social network. NEJM 358(21):2249-2258. 2008.

13. Fowler JH and Christakis NA Dynamic spread of happiness in a large social network: longitudinal analysis over 20 years in the Framingham Heart Study. British Medical Journal 337: online publication, December 2008.

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.