Can Happiness Make Us Healthier? Evidence from a Randomized Controlled Trial

Being healthy is good for happiness, but does happiness also lead to better health? Correlational research shows that people with more positive emotional styles are three times less likely to develop symptoms of a cold or flu after deliberately being exposed to the rhino or influenza virus (Cohen, Alper, Doyle, Treanor, & Turner, 2006). And in a famous longitudinal study, nuns who expressed the most positive emotions in essays in their 20s lived about 10 years longer than nuns expressing the least positive emotions (Danner, Snowdon, & Friesen, 2001).

Correlation, of course, is not causation. Controlled lab experiments—the gold standard in establishing causality in research—would, however, seem a poor tool for establishing the long-term effects of happiness on health. For causal evidence, we must therefore look at randomized controlled trials (RCTs).

In five RCTs, for example, treatment for depression modestly improved physical health outcomes (O’Neil, Sanderson, Oldenburg, & Taylor, 2011). But the lack of negative emotions is not the same as the presence of positive emotions, and we already knew that negative emotions and stress are bad for our health (Cohen, Janicki-Deverts, & Miller, 2007).

To examine whether happiness has a causal impact on health, we thus conducted a six-month RCT with a nonclinical population of 155 community adults of average physical and mental health. We designed a 12-week positive psychological intervention—ENHANCE—focused explicitly on increasing happiness rather than treating mental illness.

In a paper published in the journal Psychological Science, we found that our positive psychological intervention had beneficial effects on self-reported physical health (Kushlev et al., 2020). Following the treatment phase of the trial, participants in our intervention reported fewer days in which they felt sick in the past month compared to a waiting-group control.

In a subsample of 100 participants, we were also able to explore the effect of treatment on the number of sick days participants reported each week over the course of treatment. Even at this more granular level, participants in our positive psychological intervention reported fewer sick days (Figure 1). Notably, participants in both conditions experienced an increase in the number of sick days during Weeks 4–6 of the study, coinciding with the start of the winter flu season in the northern hemisphere (October 2016). This pattern suggests that happiness does not somehow magically make people impervious to getting sick, but simply that happiness can be a protective factor in physical health.

ENHANCE included only evidence-based psychological tools and exercises. The first three weeks of the program focused on the Core Self, which helped individuals identify their values, strengths, and goals. The second part of the program, the Experiential Self, focused on emotion regulation and mindfulness and gave participants tools to identify maladaptive patterns of thinking. The final four weeks of the program focused on the Social Self and taught individuals’ techniques to cultivate gratitude, foster positive social interactions, and engage more with their community.

We deliberately did not include any modules on health behaviors that can directly improve health, such as sleep or exercise. In addition, we were able to statistically show that the improvements in health were preceded by increases in positive affect. Thus, we could be reasonably confident that positive emotions played a role in the observed improvements in self-reported health.

All participants in the trial completed weekly modules that included an hour-long lesson with information and exercises on target principles of happiness, including a weekly writing assignment and an active behavioral component like practicing guided meditation. Some participants (n = 55), however, attended in-person weekly group sessions led by a trained clinician, whereas the rest (n =100) completed the same modules individually online using a customized web platform. The magnitude of the observed effects was consistent across the two arms of the study. Our findings thus highlight the potential online interventions to reach a wider population without notable decrements to efficacy.

In short, we found that a targeted happiness intervention can make people feel healthier. Though consistent across analyses and measures, our findings are, nonetheless, limited, to self-reports of physical health. In fact, we did measure a few basic objective indicators of health, such as BMI and blood pressure. Though some may consider the drop in the BMI of the treatment group from 26.8 at baseline to 25.2 at post-test to be clinically significant, the effect was not statistically significant.

The broader literature on happiness and health by no means suggests that happiness is a panacea for all ailments. Evidence suggests that positive emotions can be particularly beneficial precisely in alleviating symptoms of feeling sick—from pain to cold and flu symptoms—and much less so for reducing weight (Diener, Pressman, Hunter, & Delgadillo-Chase, 2017).

In the context of the broader evidence, therefore, our pattern of findings might be better explained by the effects of happiness on specific aspects of physical health rather than by effects on subjective versus objective indicators of health. Thus, future RCTs need to examine whether happiness has causal effects on objective indicators of sickness, such as mucus production or body temperature.

Happiness is no cure-all, but converging evidence—including the evidence from our randomized controlled trial—consistently points to the conclusion that happiness does play a causal role in aspects of our physical health.

Reference Article

Kushlev, K., Heintzelman, S. J., Lutes, L. D., Wirtz, D., Kanippayoor, J. M., Leitner, D., & Diener, E. (2020). Does happiness improve health? Evidence from a randomized controlled trial. Psychological Science, 7, 807–821. https://doi.org/10.1177/0956797620919673

Discussion Questions

• What would be a good active control group in an RCT designed to examine the effect of happiness on health?
• Assuming limitless time and resources, what study design could provide the strongest evidence for a causal effect of happiness on health? Is there a study design that is superior to others, or does the best evidence come from triangulating on the strengths of different designs: longitudinal, lab experiments, and RCTs?
• Are objective indicators always superior to subjective reports when it comes to physical health? If one patient with the common cold reports feeling more tired, stuffed up, and sick than another patient with a cold who has a higher fever and produces more mucus, which of the two patients is sicker?
• There are well-established programs, such as Mindfulness-Based Stress Reductions, that have been shown to increase positive emotions and happiness. What might be the benefit of an eclectic program, such as ENHANCE, in improving well-being?

About the Authors