Your Immune System Is Listening: Stress, Sleep, Loneliness and What Actually Helps

The grades in this post: Experimental means people (or animals) were deliberately exposed or randomised. Observational means researchers measured what people already did, so other differences between them can’t be fully ruled out. Mixed means trials point different ways or didn’t reach their own threshold. Untested means a reasonable idea that hasn’t been tried as an intervention.

The rat that taught its immune system a taste

In 1975 a psychologist, Robert Ader, and an immunologist, Nicholas Cohen, gave rats saccharin-sweetened water alongside a drug that suppresses the immune system. Later, the sweet water on its own, with no drug, suppressed the rats’ antibody response. The immune system had learned an association the way a nervous system does (Ader & Cohen 1975).

Fifty years on, the human evidence is better than most people realise. It’s also weaker than wellness marketing implies. Both halves matter.

What 300 studies say, in one paragraph

In 2004 two researchers pooled more than 300 human studies of stress and immune measures (Segerstrom & Miller 2004).

• Short stressors (minutes, like public speaking) tend to mobilise parts of the front-line immune system. That makes sense: it’s preparation for injury.
• Brief real-life stressors (exams) tend to dampen cell-based immunity.
• Chronic stressors (long-term caring, unemployment, ongoing threat) were linked with suppression of both cell-based and antibody responses.
• How stressed people said they felt usually didn’t track the immune changes. The circumstances did.

That last point is the one I’d underline. Your sense of “I’m coping fine” is not a reliable read-out of what your body is doing.

Four places it shows up in people

1. Whether you catch a cold

This is the strongest human evidence in the field, because the researchers didn’t wait for colds to happen. They gave volunteers a measured dose of a cold virus, quarantined them, and confirmed infection in the lab.

• Stress. In 394 volunteers, colds went from about 27% to 47% across the range of psychological stress, rising step by step. The difference was in getting infected, not in how bad the symptoms were. Smoking, alcohol, exercise, diet and sleep quality didn’t explain it (Cohen 1991).
• Why stress might do it. People who’d recently been through a long-term threatening event had immune cells that responded less to cortisol’s “calm down” signal (glucocorticoid receptor resistance). They were more likely to develop a cold and produced more inflammatory signalling once infected (Cohen 2012).
• Sleep. In 164 adults whose sleep was measured with a wrist device for a week beforehand, those sleeping under 5 hours and those sleeping 5 to 6 hours were each roughly four times as likely to develop a cold as those sleeping over 7 (Prather 2015). The confidence intervals were wide; the direction was clear. Six to seven hours didn’t differ significantly from seven-plus.
• People. Volunteers with the fewest types of social tie (spouse, friend, workmate, club and so on) had about 4 times the risk of a cold compared with those with the most, in a dose-response pattern (Cohen 1997). The number of types mattered, not just the number of people.
• Mood style. People who habitually felt happy, relaxed and lively were less likely to develop a verified cold: those in the least positive third had about 2.9 times the risk of those in the most positive third (Cohen 2003). Habitually negative mood didn’t predict colds, but it did predict reporting symptoms with no infection behind them.
• Hugs. Over two weeks, more conflict at home or work went with higher infection risk after exposure. Feeling supported, and being hugged more often, blunted that rise (Cohen 2015).

2. How fast a wound heals

• Thirteen women caring for a relative with dementia took 48.7 days to heal a small standard skin wound, against 39.3 days in matched controls (Kiecolt-Glaser 1995).
• Eleven dental students each had a small wound made in the mouth twice: once in the summer holiday and once three days before an exam. During exams they healed on average 3 days, or 40%, slower, and one healing signal (IL-1β) fell by about two-thirds. No student healed as fast at exam time as in the holiday (Marucha 1998).

3. How well the immune system learns

Researchers often use vaccine responses as a standard test of how well the immune system learns something new, because the dose and timing are exactly known. This section is about what that tells us regarding stress and sleep, not an argument about any vaccine.

• Caring for a spouse with dementia: 32 carers made weaker antibody responses to the flu vaccine than 32 matched controls (Kiecolt-Glaser 1996).
• Sleep: in 125 adults aged 40–60, shorter sleep measured by a wrist device predicted a weaker hepatitis B antibody response and a lower chance of reaching the protective level six months after the course (Prather 2012). How well people said they slept didn’t predict it.
• Loneliness: in 83 first-year university students, loneliness and small social networks each went with a weaker response to one component of the flu vaccine (Pressman 2005).

4. What’s happening underneath

• One short night. Forty-two healthy men kept awake from 10pm to 3am showed lower natural killer (NK) cell activity the next day. It recovered after a normal night’s sleep; another immune signal (IL-2) hadn’t (Irwin 1996).
• Loneliness in the genes. In older adults followed over time, loneliness went with immune cells switching on inflammation genes and switching down antiviral ones. In monkeys, the same team showed the mechanism and a weaker response to a viral infection (Cole 2015). I covered this in Loneliness Is a Clinical Variable.

What doesn’t follow

• “Boost your immune system.” Nothing above shows that a calmer, better-slept person has a “stronger” immune system across the board. It shows better-regulated responses: fewer infections after exposure, faster repair, a cleaner response to a new challenge, less inflammation that doesn’t switch off.
• The effects are real but moderate. Going from 27% to 47% matters. It also means most highly stressed people exposed to a cold virus in the original study didn’t get a cold, and plenty of calm people did. Exposure still sets most of the risk.
• A single test won’t tell you your “immune status”. I don’t know of a validated lab test that does. Be wary of anyone selling one.
• Feeling is not a measure. Segerstrom and Miller found self-reported stress usually didn’t match immune change, and Prather found measured sleep mattered while reported sleep didn’t. That’s why I’d rather measure what you do than ask how you feel.

What’s been tested to help

This is where the evidence thins out, and I’d rather say so.

• Mindfulness training or exercise. In a trial of adults aged 50 and over (154 randomised, 149 completed), eight weeks of mindfulness training cut the overall severity of colds and flu that winter, and both meditation and exercise cut days off work (Barrett 2012). A larger repeat in 390 adults pointed the same way (fewer infections with mindfulness, milder illness with exercise) but missed the trial’s own threshold for significance, and the authors called it likely underpowered (Barrett 2018). Grade: mixed.
• Tai chi. In 112 adults aged 59–86, 16 weeks of tai chi raised immune responses to the shingles virus by about as much as the vaccine did, and combined with the vaccine it added to it. Physical function and vitality scores improved too (Irwin 2007). Grade: one randomised trial.
• Regular aerobic exercise. In 1,002 adults followed through autumn and winter, people exercising aerobically on five or more days a week had 43% fewer days with cold symptoms than largely sedentary people (Nieman 2011). Grade: observational; fitter people differ in other ways.
• Sleeping more. The link between short sleep and infection is consistent. I didn’t find a trial that extended people’s sleep and then counted infections, so “sleep more and you’ll catch fewer colds” is a reasonable inference. Grade: untested.
• More social connection. Same position: consistent observational evidence, few trials testing whether changing it changes infection. Grade: untested.

What I do with clients

None of this needs a special protocol. It needs taking seriously the things people treat as optional.

1. Sleep is measured, not guessed. If you wear a tracker, we look at total sleep across weeks, not one bad night. The research threshold that keeps coming up is about seven hours. Six is where risk climbs.
2. Movement most days. Moderate and regular is the pattern linked with fewer infections. I adjust intensity when someone is already depleted, because a hard session on four hours’ sleep is a stressor, not a treatment.
3. People. I ask about this the way I ask about bowel habits. The range of connections mattered in the research, not just one close relationship.
4. Name the stressor. Long-term threat was what changed immune cells’ response to cortisol. Finding what’s actually long-term, rather than “stress” in general, is half the work (Stress Is Personal).
5. Test what can be tested. A DUTCH test (the HPA axis) shows the shape of your cortisol day. It measures hormones, not immune function, and no study has shown that DUTCH results predict infection risk. It tells us whether the system that’s meant to switch inflammation off is running high, flat or low. Blood tests check iron, B12 and vitamin D, which matter for immunity and are worth correcting when they’re actually low.
6. Food first, supplements last and targeted. Food that keeps blood sugar steady and the gut fed supports all of the above. Supplements come in only for something tested and low, or as a time-limited trial with a clear stop point.

Where this sits

• My own story of getting properly ill after a long stretch of strain is in Can You Catch a Cold? and Why I Got Properly Ill. Those posts are one person’s experience. This one is the research behind why I took it seriously.
• The hormone side is in Stress Is a Biochemistry Problem and The HPA Axis.
• Social connection is in Loneliness Is a Clinical Variable; long-held strain in The Body Keeps the Score.

Evidence summary

Sources

1. Ader R, Cohen N. Behaviorally conditioned immunosuppression. Psychosom Med 1975;37:333–40. PMID 1162023. doi:10.1097/00006842-197507000-00007
2. Segerstrom SC, Miller GE. Psychological stress and the human immune system: a meta-analytic study of 30 years of inquiry. Psychol Bull 2004;130:601–30. PMID 15250815. doi:10.1037/0033-2909.130.4.601
3. Cohen S, Tyrrell DA, Smith AP. Psychological stress and susceptibility to the common cold. N Engl J Med 1991;325:606–12. PMID 1713648. doi:10.1056/NEJM199108293250903
4. Cohen S et al. Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk. PNAS 2012;109:5995–9. PMID 22474371. doi:10.1073/pnas.1118355109
5. Prather AA et al. Behaviorally assessed sleep and susceptibility to the common cold. Sleep 2015;38:1353–9. PMID 26118561. doi:10.5665/sleep.4968
6. Cohen S et al. Social ties and susceptibility to the common cold. JAMA 1997;277:1940–4. PMID 9200634.
7. Cohen S et al. Emotional style and susceptibility to the common cold. Psychosom Med 2003;65:652–7. PMID 12883117. doi:10.1097/01.psy.0000077508.57784.da
8. Cohen S et al. Does hugging provide stress-buffering social support? Psychol Sci 2015;26:135–47. PMID 25526910. doi:10.1177/0956797614559284
9. Kiecolt-Glaser JK et al. Slowing of wound healing by psychological stress. Lancet 1995;346:1194–6. PMID 7475659. doi:10.1016/s0140-6736(95)92899-5
10. Marucha PT, Kiecolt-Glaser JK, Favagehi M. Mucosal wound healing is impaired by examination stress. Psychosom Med 1998;60:362–5. PMID 9625226. doi:10.1097/00006842-199805000-00025
11. Kiecolt-Glaser JK et al. Chronic stress alters the immune response to influenza virus vaccine in older adults. PNAS 1996;93:3043–7. PMID 8610165. doi:10.1073/pnas.93.7.3043
12. Prather AA et al. Sleep and antibody response to hepatitis B vaccination. Sleep 2012;35:1063–9. PMID 22851802. doi:10.5665/sleep.1990
13. Pressman SD et al. Loneliness, social network size, and immune response to influenza vaccination in college freshmen. Health Psychol 2005;24:297–306. PMID 15898866. doi:10.1037/0278-6133.24.3.297
14. Irwin M et al. Partial night sleep deprivation reduces natural killer and cellular immune responses in humans. FASEB J 1996;10:643–53. PMID 8621064. doi:10.1096/fasebj.10.5.8621064
15. Cole SW et al. Myeloid differentiation architecture of leukocyte transcriptome dynamics in perceived social isolation. PNAS 2015;112:15142–7. PMID 26598672. doi:10.1073/pnas.1514249112
16. Barrett B et al. Meditation or exercise for preventing acute respiratory infection: a randomized controlled trial. Ann Fam Med 2012;10:337–46. PMID 22778122. doi:10.1370/afm.1376
17. Barrett B et al. Meditation or exercise for preventing acute respiratory infection (MEPARI-2). PLoS One 2018;13:e0197778. PMID 29933369. doi:10.1371/journal.pone.0197778
18. Irwin MR, Olmstead R, Oxman MN. Augmenting immune responses to varicella zoster virus in older adults: a randomized, controlled trial of Tai Chi. J Am Geriatr Soc 2007;55:511–7. PMID 17397428. doi:10.1111/j.1532-5415.2007.01109.x
19. Nieman DC et al. Upper respiratory tract infection is reduced in physically fit and active adults. Br J Sports Med 2011;45:987–92 (online 2010). PMID 21041243. doi:10.1136/bjsm.2010.077875