Connection: Dankbaarheidsoefening

Diniz and colleagues (2023) searched MEDLINE, Embase and Cochrane Central and included 64 randomized clinical trials of gratitude interventions. The meta-analysis reported that participants assigned to gratitude practices experienced greater feelings of gratitude, better mental health, fewer symptoms of anxiety and depression, and more positive mood and emotions than controls. The authors frame gratitude as a usable therapeutic complement for anxiety and depression and as a way to raise positive feelings in the general population. As a pooled synthesis of trials with mixed comparison conditions and mixed quality, it establishes a consistent, modest benefit, not a large effect.

Diniz and colleagues (2023), pooling 64 randomized trials, reported fewer anxiety symptoms among participants doing gratitude practices. Davis and colleagues (2016), in a set of meta-analyzes that deliberately separated comparison types, found gratitude interventions did not outperform an alternative-activity condition on anxiety (d = 0.11, 95% CI -0.08 to 0.31, k = 5). The two are not straightforwardly reconcilable, and the most likely reason is the comparator: Davis showed across outcomes that the apparent benefit of gratitude shrinks toward zero once it is measured against another structured positive activity, not against a waiting list or measurement-only group. The anxiety signal is therefore graded emerging and its direction left mixed.

Davis and colleagues (2016) ran separate meta-analyzes by comparison group. Against a measurement-only control, gratitude interventions improved psychological wellbeing with d = 0.31 (95% CI 0.04 to 0.58, k = 5). Against an alternative-activity condition the effect was smaller, d = 0.17 (95% CI 0.09 to 0.24, k = 20), and against an activity-matched comparison d = 0.14 (95% CI 0.01 to 0.27, k = 18). Against a psychologically active comparison, one built to be as engaging as gratitude itself, the effect was d = -0.03 (95% CI -0.13 to 0.07, k = 9): as good as, but no better than, the alternative. This is the finding behind the calibrated framing on this page: gratitude gives a modest lift, and much of the apparent size in popular accounts comes from comparing it against nothing.

Emmons and McCullough (2003) ran three randomized experiments. In studies 1 and 2, participants were assigned to weekly or daily records of gratitude, hassles, or neutral or downward-comparison events, then logged mood, coping, health behaviors, physical symptoms and life appraisals. Study 3 assigned adults with neuromuscular disease to a gratitude or control condition. The gratitude-outlook groups showed higher wellbeing across several, though not all, outcome measures relative to comparisons, and the effect on positive affect was described as the most robust finding. These are the studies that launched the field; they are small and the benefit was uneven across outcomes, which is why the claim sits at emerging.

Wong and colleagues (2018) randomized 293 adults seeking university-based psychotherapy to therapy only, therapy plus expressive writing about stressful experiences, or therapy plus gratitude writing (letters expressing gratitude to others). At roughly four and twelve weeks after the intervention, the gratitude group reported significantly better mental health than both the expressive-writing and control groups, which did not differ from each other. A lower proportion of negative-emotion words in participants writing mediated the benefit. Because the active comparator here was another writing task, not nothing, this is a more demanding test than much of the literature, though the outcome was self-reported mental health and the setting was an adjunct to therapy.

Komase and colleagues (2021) screened 1,957 records and included nine randomized controlled trials of gratitude interventions among healthy workers. Eight used gratitude-list interventions, which significantly improved perceived stress and depression; effects on wellbeing were inconsistent across studies. Interventions involving four or fewer gratitude lists reported no significant change on any outcome, suggesting a dose element to the practice. No trial tested behavioral gratitude expression alone. The review points to gratitude as a plausible aid for occupational mental health while flagging that the wellbeing evidence and the small number of trials keep it at emerging.

Cheng, Tsui and Lam (2015) ran a double-blind randomized controlled trial in five public hospitals, assigning 102 practitioners to a gratitude diary, a hassle diary, or a no-diary control; the two diary groups wrote twice a week for four weeks. Depressive symptoms (primary outcome) and perceived stress were measured at baseline, post-treatment and three-month follow-up with intent-to-treat mixed-effects models. Relative to control at follow-up, the gratitude group reported lower depressive symptoms (-1.50 points, 95% CI -2.98 to -0.01, d = -0.49) and lower perceived stress (-2.65 points, 95% CI -4.00 to -1.30, d = -0.95); results versus the hassle diary were similar, and the hassle and control groups were essentially indistinguishable. The rate of improvement flattened over time. It is one trial in a single occupation, which is why it is graded emerging.

Wood and colleagues (2009) surveyed 401 community adults (186 men, 215 women, ages 18 to 68), 40% of whom scored above 5 on the Pittsburgh Sleep Quality Index, indicating clinically impaired sleep. Higher trait gratitude predicted greater subjective sleep quality and duration and less sleep latency and daytime dysfunction. The relationship between gratitude and each sleep measure was mediated by more positive and less negative pre-sleep cognitions, and all results were independent of the Big Five personality traits, including neuroticism, and of social desirability. Being cross-sectional, it identifies a mechanism-consistent association, not showing that cultivating gratitude changes sleep.

Boggiss and colleagues (2020) reviewed 19 randomized trials that assessed physical health and health-behavior outcomes of gratitude interventions. Subjective sleep quality improved in 5 of 8 studies that measured it, the most consistent physical finding. Other outcomes were understudied: single trials suggested improvements in blood pressure, glycemic control, asthma control and eating behavior (each 1 of 1), inflammation was mixed (1 of 2), and self-reported physical symptoms improved in 2 of 8. Most studies showed some risk-of-bias concerns, and the authors concluded that more research is needed to confirm efficacy, singling out the gratitude-sleep link as the priority.

Jackowska and colleagues (2016) randomized 119 young women to a two-week gratitude intervention, an active control (reporting everyday events), or no treatment. The gratitude group showed increases in hedonic wellbeing, optimism and sleep quality and a decrease in diastolic blood pressure. Improvements in wellbeing correlated with better sleep and lower blood pressure, but there was no relationship with cortisol. The authors suggest subjective wellbeing may support healthier biological function, while noting this is a brief experiment. Measured only in young women and on a single trial, the cardiovascular reading is graded preliminary.

Redwine and colleagues (2016) randomized 70 patients with stage B (asymptomatic) heart failure, mean age 66.2 years, to eight weeks of gratitude journaling or treatment as usual. The gratitude group showed improved trait gratitude scores, a reduced inflammatory biomarker index over time (F = 9.7, p = .004), and increased parasympathetic heart-rate variability during a gratitude journaling task (F = 4.2, p = .036). There were no resting pre- to post-intervention group differences in heart-rate variability (p values > .10). The authors conclude gratitude journaling may improve biomarkers related to heart-failure prognosis and that large-scale studies with active control conditions are needed, which is why the finding is graded preliminary.

Kini and colleagues (2016) recruited people entering psychotherapy for depression or anxiety; one group did a gratitude letter-writing intervention and a therapy-as-usual group did not. Three months later, participants performed a Pay It Forward task in an fMRI scanner in which they received monetary gifts and chose how much to pass on to charity in proportion to how grateful they felt. Brain activity correlating with self-reported gratitude, with guilt motivation and desire to help controlled for, fell largely in regions distinct from those for empathy or theory of mind. In a between-groups comparison, the gratitude-writing group showed greater and lasting neural modulation by gratitude in the medial prefrontal cortex. As a small imaging study, it offers a candidate mechanism, not a demonstrated causal pathway to symptom change.