Can Reading Support Well-Being? What the Evidence Can and Cannot Tell Us

The short answer is careful: reading can be a meaningful leisure experience, and some studies associate reading or reading-based activities with aspects of well-being. But the evidence does not show that ordinary, self-selected reading reliably improves mental health for everyone. It does not establish that reading prevents or treats anxiety, depression, stress, loneliness, or any other condition.

The research is a mixture of observational associations, small or context-specific interventions, null findings, and reviews that suggest possible mechanisms rather than a single average effect. What a person reads, why they read, whether they read alone or with other people, what outcome is measured, and who is included in a study all matter. A useful conclusion is therefore modest: if reading suits you, it may be worth making room for as a leisure practice. It is not a treatment plan.

Research note

The best-supported public claim is not “reading makes you well.” It is that reading may matter differently in different contexts, while the causal evidence for ordinary leisure reading remains limited and mixed.

Why the question is harder than it sounds

“Reading” is not one research exposure, and “well-being” is not one outcome. A survey might ask how often someone reads books and compare that answer with a distress or life-satisfaction score. A trial might assign older adults to six weeks of fiction or nonfiction audiobooks. A shared-reading program might bring a small group together weekly to read aloud and discuss literary texts. Bibliotherapy might use structured self-help material with clinical screening and limited professional support.

Those are meaningfully different activities. Evidence from one should not be silently transferred to another. In particular, a study of a facilitated reading group does not prove that a person reading a novel alone will experience the same social benefits. A bibliotherapy trial is not evidence that any novel functions as mental-health care. And a correlation between reading frequency and a questionnaire score does not tell us whether reading changed the score, whether well-being made reading easier, or whether both were influenced by time, resources, health, education, personality, or social circumstances.

It also helps to name the outcomes. Psychological distress, positive and negative affect, life satisfaction, meaning in life, loneliness, anxiety, depressive symptoms, and general mental well-being overlap, but they are not interchangeable. A small change on one measure should not be advertised as a universal improvement in “wellness.”

What observational and longitudinal studies can show

One frequently cited study followed 231 students at a large Canadian university from the beginning to the end of an academic year. Levine and colleagues (2022) found that the number of books students reported reading recreationally was associated with lower reported psychological distress over the year after baseline distress was taken into account. The study is useful because it followed readers over time rather than taking a single snapshot.

It is not a randomized test of reading. Students chose whether and what to read; the measure was a self-reported number of recreational books; the setting was one university; and the analytic sample was smaller than the number initially enrolled. The authors also examined autonomous motivation and need frustration, but their mediation analysis does not turn the result into proof of causation. The most defensible summary is: in this college sample, more recreational reading was associated with a more favorable change in reported distress. That is narrower than “reading reduces distress.”

Large surveys can be informative while still leaving the same problem. Reading frequency is a blunt measure: it does not say whether the material was chosen freely, whether it was enjoyable, whether the reader was absorbed, whether reading was solitary or social, or whether it displaced sleep, exercise, conversation, or other activities. A critical review of reading-for-pleasure research by Cremin and Scholes (2024) notes that much of this literature relies on large self-report surveys and statistical modeling. Those studies can identify patterns worth investigating, but they cannot capture every part of the reading experience or by themselves establish that reading caused an outcome.

What stronger designs add—and what they do not

The most useful counterweight to an appealing association is a study that makes the comparison more difficult to explain away. In a six-week randomized study of 94 older adults, Poerio and Totterdell (2020) assigned participants to fiction, short stories, or nonfiction audiobooks. Participants chose a book within their assigned condition. The study did not find a reliable fiction advantage over nonfiction, nor a reliable average improvement over time in positive or negative affective well-being, life satisfaction, meaning in life, loneliness, or mental well-being.

Some secondary analyses found that people who were more absorbed in or appreciative of their audiobook later reported more favorable well-being or meaning scores. That is interesting, but those engagement measures were not randomly assigned. They may reflect personality, prior interest, or other differences between participants. The intervention used audiobooks, not print reading; the sample was small and older; and six weeks is a short follow-up. The result is best read as mixed evidence, not as proof that fiction does nothing and not as proof that being absorbed in a story improves health.

A newer longitudinal analysis asks a different question: whether more frequent reading predicts later adolescent mental health after stronger adjustment for confounding. Murray and colleagues (2025) used counterfactual weighting and random-intercept cross-lagged panel models in a large Swiss adolescent cohort. After adjustment, neither approach found significant effects of reading frequency on later anxiety, depression, or psychosis-like symptoms. The study does not settle the question for adults or for every form of reading. Its frequency measure was coarse, and the authors note possible unmeasured confounding, time-varying confounding, and limited power for very small effects. It does, however, make a universal protective-factor story harder to defend.

Other experimental work also points to context. In a set of five studies, Carney and Robertson (2022) found no consistent mood result from direct fiction exposure, while recalling or discussing fiction showed more supportive patterns in some studies. Those activities are not interchangeable: remembering a story and talking with another person may involve reflection or social connection that is absent from simply being assigned a text. The authors explicitly caution against treating fiction like a dose of a pharmaceutical substance.

Taken together, stronger designs do not give us a single “reading effect.” They show why the question has to include the activity’s form, motivation, content, social setting, and outcome. A null result does not prove that no individual ever finds a book absorbing or comforting. It does weaken claims that reading frequency alone is a dependable mental-health intervention.

Shared reading can be a social intervention

Shared reading deserves its own category. In the studies summarized by Järvholm and colleagues (2025), a trained facilitator typically led a group that read literary fiction or poetry aloud and discussed it. Their scoping review found 15 studies, mostly from the United Kingdom and Scandinavia, often involving small groups and people living with dementia, mental illness, chronic pain, or other vulnerabilities. The review reported positive participant accounts, but quantitative designs were commonly small or lacked comparison groups, terminology varied, and long-term follow-up was uncommon.

A systematic review focused on older adults by Milani and colleagues (2025) included 11 heterogeneous studies of shared reading. It found promising signals for social interaction, quality of life, loneliness, and depressive symptoms, while calling for stronger studies. “Promising” is the right level of confidence here. A regular group, a facilitator, discussion, and a shared text may create opportunities for connection that a person reading alone does not have. These findings cannot be used to promise that solitary leisure reading, or a book-tracking app, will produce the same result.

Bibliotherapy is a different category again

Bibliotherapy generally refers to structured self-help or therapeutic reading used with people who have a diagnosed condition or clinically meaningful symptoms, sometimes with guidance from a professional. For example, Li and colleagues (2018) reviewed randomized trials of bibliotherapy for depression and anxiety disorders in children and adolescents. That evidence belongs to a defined clinical or near-clinical intervention: selected material, a specified population, measured symptoms, and a treatment context.

It should not be used as evidence that any self-selected book treats depression or anxiety. Nook & Spine is not a bibliotherapy platform, and this article is not making a treatment recommendation. Leisure reading is not a substitute for professional care. If distress, low mood, anxiety, loneliness, or another concern is affecting your life, seek support from a qualified health professional or an appropriate local service rather than asking a book—or a reading record—to do a clinician’s job.

A practical “fit, not fix” reading experiment

The following is a low-pressure planning framework, not a tested health intervention and not a prescription for how often to read. It is a way to notice whether reading fits your life without turning the experience into a pass/fail wellness project.

1. Choose for fit. Pick a book that matches your interest, available time, energy, format needs, and current tolerance for difficult material. There is no evidence-based requirement to choose a particular genre or number of pages.
2. Lower the stakes. Try one session or one chapter. A book is allowed to be interesting, demanding, neutral, or simply not right for you.
3. Notice the experience, not a promised outcome. You might ask: Was I engaged? Did the pace suit me? Did the subject feel nourishing, effortful, or emotionally heavy? These observations are personal signals, not proof that reading changed your health.
4. Keep a small record. Record the title and edition, mark whether it is on your TBR list, in progress, completed, or a DNF, and add a short note about what you want to remember or return to.
5. Review without judgment. Continue, pause, switch books, or stop. Finishing is not a health outcome, and abandoning a book is not a personal failure.
6. Add connection only if it suits you. A friend, book club, or facilitated shared-reading group may make the activity social. That is a different context from reading alone, and its benefits should not be assumed in advance.

This framework is an editorial inference from the evidence’s emphasis on context, motivation, engagement, and social setting. It has not been tested as a package, so it should stay modest: make reading easier to try, and learn what fits, rather than promising a result.

How Nook & Spine fits—without making a health claim

Nook & Spine can support the logistics of an intentional reading routine or record. Its library entries can track TBR, Reading, Completed, and DNF status, along with notes and started or finished dates. The TBR queue supports genre filtering, reordering, and a saved next-up picker. Those are organization features: they can help you remember what you chose and what you want to return to.

The app does not measure mood, stress, loneliness, or mental-health symptoms. It does not deliver therapy, diagnose a condition, or make reading more effective. Keeping a reading record may be useful if you like seeing your choices in one place; it is not a wellness metric.

Sources and further reading

• Levine, Cherrier, Holding, and Koestner (2022), “For the love of reading: Recreational reading reduces psychological distress in college students and autonomous motivation is the key,” Journal of American College Health. DOI. A longitudinal college study that found an association between recreational reading and reported distress, but did not randomize reading.
• Poerio and Totterdell (2020), “The effect of fiction on the well-being of older adults: a longitudinal RCT intervention study using audiobooks,” Psychosocial Intervention. DOI. A six-week randomized audiobook study with null or mixed average well-being findings.
• Murray and colleagues (2025), “A counterfactual and random intercept cross-lagged panel analysis of the effects of reading frequency on adolescent mental health in a large longitudinal study,” JCPP Advances. DOI. A longitudinal adolescent analysis that found no significant adjusted effects of reading frequency on later mental-health outcomes.
• Sirisena and colleagues (2025), “Reading for wellbeing: a realist review of evidence,” Perspectives in Public Health. DOI. A realist review of 43 papers that proposes context-dependent mechanisms rather than a pooled causal effect.
• Cremin and Scholes (2024), “Reading for pleasure: scrutinising the evidence base—benefits, tensions and recommendations,” Language and Education. DOI. A critical review useful for understanding the limits of survey-heavy, correlational reading research, especially among younger readers.
• Järvholm and colleagues (2025), “Shared reading as an intervention to improve health and well-being in adults: a scoping review,” Frontiers in Psychology. DOI. A review of facilitated shared-reading programs, kept separate here from solitary leisure reading.
• Milani and colleagues (2025), “Shared reading interventions to promote psychosocial well-being in older adults: a systematic review,” Health Promotion International. DOI. A systematic review reporting promising but heterogeneous shared-reading evidence in older adults.
• Carney and Robertson (2022), “Five studies evaluating the impact on mental health and mood of recalling, reading, and discussing fiction,” PLOS ONE. DOI. Mixed experimental and social-data evidence that separates direct exposure from recalling or discussing fiction.
• Li and colleagues (2018), “Comparative efficacy and acceptability of bibliotherapy for depression and anxiety disorders in children and adolescents: a meta-analysis of randomized clinical trials,” Psychiatry Research. Open-access record. Included only to define why clinical bibliotherapy should not be confused with ordinary leisure reading.