A striking claim about expressive writing circulates online: write for 15 minutes a day on four consecutive days about a difficult experience, and the immune system measurably improves months later.
The claim is close enough to famous research by psychologist James Pennebaker to sound precise. It is also a blend of two different experiments.
The original 1986 study used the 15-minute, four-day protocol and tracked visits to a student health centre over the following six months. It did not measure immune function. A 1988 replication collected blood and reported changes in a laboratory measure of T-lymphocyte response six weeks after a 20-minute, four-day protocol.
Both studies helped launch decades of research. Later trials and meta-analyses have produced a much less certain answer about whether expressive writing improves health.
We are writers, not clinicians. Writing about trauma can cause immediate distress and is not a replacement for professional care. This article explains the evidence rather than recommending a treatment.
The famous 15-minute experiment measured doctor visits
In 1983, Pennebaker and graduate student Sandra Beall recruited 46 introductory psychology students. Participants were assigned to write on four consecutive days for 15 minutes each day. Some wrote about personally upsetting experiences, while control groups wrote about trivial topics or separated the facts of an experience from their feelings.
The students who wrote about both the facts and emotions surrounding a traumatic experience became more upset during the sessions. Their blood pressure rose, and their immediate mood was more negative. The exercise did not produce instant relief.
Over the next six months, however, those students made fewer illness-related visits to the university health centre. Pennebaker and Beall described the results as preliminary in their 1986 paper in the Journal of Abnormal Psychology.
In a later review, Pennebaker recalled that students in the trauma-writing condition visited the health centre at about half the control group’s rate. He also called the original experiment “horribly underpowered”, an unusually direct acknowledgement of its small sample.
The result was still provocative. Random assignment made the comparison stronger than a survey asking habitual journal writers whether they felt healthier. Medical records offered a more objective outcome than asking participants to remember every illness. But 46 undergraduates at one university cannot establish a general health treatment.
The immune study used 20 minutes and a six-week follow-up
Two years later, Pennebaker worked with psychologist Janice Kiecolt-Glaser and immunologist Ronald Glaser on a replication that directly examined immune markers.
Fifty healthy undergraduates were randomly assigned to write about traumatic experiences or superficial topics for 20 minutes on each of four consecutive days. Blood was collected the day before writing began, after the final session, and again six weeks later.
The researchers tested how lymphocytes, a type of white blood cell, proliferated in response to two laboratory stimulants called phytohaemagglutinin and concanavalin A. These assays were intended as models of T-cell response to challenge. They were not tests of whether participants resisted an actual infection.
The trauma-writing group showed changes in the expected direction, with the strongest improvement among people whose essays contained more self-disclosure. The study, published in the Journal of Consulting and Clinical Psychology in 1988, also reported fewer health-centre visits during the six weeks after writing.
This is the source of the immune claim. It involved 20-minute sessions and a six-week follow-up, not 15 minutes and an outcome measured months later. The authors also stressed that there is no single general measure of “immune function”. Their result concerned specific responses of cultured lymphocytes to two mitogens.
A marker change is not the same as improved immunity
The phrase “improved immune function” compresses several inferential steps. A laboratory marker may reflect one part of immune activity without showing that a person gets fewer infections, recovers faster, or has better long-term health.
Immune responses are also not simply stronger-is-better. The immune system must mount an effective response to threats while limiting inappropriate inflammation and attacks on healthy tissue. Different markers can move in different directions, and their clinical meaning depends on context.
The 1988 experiment was small. Two participants missed the final blood draw, two sets of immune data were excluded, and several samples were lost during laboratory analysis. The finding was worthy of follow-up, but it was not a demonstration that a four-day writing exercise broadly strengthened immunity.
Later evidence did not deliver one verdict
Expressive writing studies multiplied, using different prompts, populations, session lengths and outcomes. Some reported benefits for healthcare use, symptoms or particular biological markers. Others found no effect.
Pennebaker’s own 2018 review of expressive writing research describes both replications and failures to replicate. He argued that effects depend on whom the exercise is given to, how it is framed, and what outcome is measured.
A 2009 meta-analysis of 30 randomised controlled trials reached a sceptical conclusion. When the researchers pooled long-term physical and psychological outcomes measured at least four weeks after writing, they found no significant overall effect. Exploratory analyses were also mostly null.
That was not the only synthesis. A separate meta-analysis of 146 studies reported a small overall effect, while a review focused on healthcare use found a small reduction among healthy samples but not among people selected for medical diagnoses or psychological criteria. Differences in inclusion rules and outcomes help explain why reviews disagree.
Individual trials show the same variability. In a randomised study of 116 healthy young adults, four 20-minute writing sessions produced no overall improvement in anxiety, depression or physical symptoms. Emotional expressiveness moderated anxiety results: people high in expressiveness improved, while those low in expressiveness became more anxious within the writing group.
The protocol is an experiment, not a universal prescription
The Pennebaker paradigm remains scientifically interesting because such a small intervention sometimes produced measurable differences weeks or months later. It opened questions about inhibition, language, stress, memory and the construction of a coherent account of difficult events.
Its simplicity may also invite overstatement. “Write for 15 minutes and boost immunity” sounds like a settled biological recipe. The research does not support that promise.
A more accurate summary keeps the two early experiments separate. Four 15-minute sessions were followed by fewer health-centre visits in a small 1986 study. Four 20-minute sessions were followed by changes in specific lymphocyte assays at six weeks in a small 1988 study. Larger bodies of research later found small, variable or null average health effects.
Expressive writing may help some people in some circumstances. The evidence does not turn a notebook into an immune treatment, and it gives no reason to judge someone who finds writing about a difficult experience unhelpful or overwhelming.