3.1 Study selection

Our initial search of five databases resulted in 1957 articles overall. After removing duplicates and adding four articles using a hand search, 1470 articles proceeded to the sifting phase. Among these, 1443 articles were excluded, and 27 articles proceeded to full-text review. Following this process, nine articles were included in the qualitative review (Figure 1).

PRISMA flow diagram

3.2 Study characteristics

The characteristics of the included studies are summarized in Table 2. Four gratitude intervention studies were conducted in the US35-38 and three in China.39-41 One study was conducted in Japan42 and one in Australia.27 The participants included in the study were mostly health care professionals (n = 2)39, 41 or teachers (n = 2).37, 40 Four studies conducted a follow-up survey after one week,35 one month,36, 42 and three months after the interventions,41 and the other five studies conducted a follow-up survey immediately after the intervention.27, 37-40 Regarding the response rate to the baseline survey (proportion of people who agreed to participate in the study out of the total number of people asked to participate), two studies had response rates above 80%,39, 41 and seven other studies did not report this. The follow-up rates ranged from 50% to 97%. The completion rate of the intervention was not summarized in any studies.

TABLE 2. Design and settings of the studies included in the systematic reviews: N = 9 Study ID Author, year Country Study design Recruit Dissemination for recruiting Participants information (population, age, and sample size) Response rate at baseline Follow-up period after intervention Follow-up /dropout rate 1 Ki, 2009.39 China, Hong Kong RCT A convenient sample. NR

Health care professionals (nurse, doctor, physical therapist, and occupational therapist) Age: 18–50 (range)

Sample size: 94 + 67 (men and women)

86% (180/210) Immediate post-intervention survey 89% (161/180) 2 Baker, 2011.35 The US RCT Conducted in a public university; only participants who had jobs and worked more than 10 h per week were recruited. Participants enrolled in introductory psychology or a similar course were granted participation credit upon completion of the study.

Employees who, for the most part, were also undergraduate students in a public university.

Age: 18–53 (range)

24 (mean for total)

Sample size: 65 + 98 (men and women)

NR 1 week 92% (165/180) 3 Otsuka et al., 2012.42 Japan RCT Conducted at the mental health seminar held in local government. NR

Daytime local-government employees

Age: 48.5 (mean for gratitude group)

48.4 (mean for the control group)

Sample size: 28 + 9 (men and women),

1 missing

NR 1 month 50% (38/76) 4 Chan et al., 2013.40 China, Hong Kong RCT Invitation notice was posted on the author's teaching webpage to recruit volunteers. To participate in an eight-week self-improvement project to enhance their well-being through self-reflection.

School teachers receiving in-service training for postgraduate degrees

Age: 22–58 (range)

33.7 (mean)

Sample size: 15 + 63 (men and women)

NR Immediate post-intervention survey 96% (78/81) 5 Kaplan et al., 2014.36 The US RCT Various departments in two large public universities were recruited as potential participants. The participants were told that the purpose of the study was to explore avenues to increase well-being at work. Employees who completed the study could receive a $10 gift certificate for participating.

Staff members from two large public universities.

Age: 43 (SD: 12.25)

Sample size: 112 (sex was NR)

NR 1 month 60% (67/112) 6 Cheng et al., 2015.41 China, Hong Kong Double-blind RCT The hospitals were chosen because of the availability of research assistants who were workers on-site. NR

Full-time professional workers (physicians, nurses, physiotherapists, and occupational therapists) in 5 hospitals

Age: NR

Sample size: 46 + 56 (men and women)

82% (102/125) 3 months 97% (99/102) 7 Neumeier et al., 2017.27 Australia RCT Advertisements via social media, local newspaper, and radio directed participants to a website with information about the study and eligibility. Aimed at testing two different online well-being programs consisting of seven brief exercises to be completed during the next 7 days at work.

Employees

Age: 19–73 (range),

41.7 (mean for PERMA group)

40.6 (mean for Gratitude group)

41.1 (mean for Wait list group)

Sample size: 303 (sex was NR)

NR Immediate post-intervention survey 70% (303/431) 8 Cook et al., 2017.37 The US RCT A flyer was distributed by central administrative staff to secondary teachers within the school district. The flyer offered access to free web-based training for staff interested in learning skills to manage job-related stress effectively and enhancing their overall well-being.

High & middle school teachers from a single educational service district

Age: NR

Sample size: 44 (sex was NR)

NR Immediate post-intervention survey 81% (44/54) 9 Ligon, 2019.38 The US RCT The study sample was recruited from one mental health call center located in Iowa and three customer service call centers in Ohio, Florida, and California. The purpose of the research was to examine whether three different work activities effectively reduce stress and increase hope, positive thinking, mental toughness, and confidence.

Employees of mental health and customer service call centers worked either a day shift or night shift full-time (i.e., 40 h per week).

Age:

18–29: 49 (sum of 3 groups)

30–39: 51 (sum of 3 groups)

40–49: 34 (sum of 3 groups)

50 or older: 14 (sum of 3 groups)

Sample size: 55 + 89 + 4

(men and women and transgender)

NR Immediate post-intervention survey 94.9% (148/156) Abbreviations: NR, Not reported; RCT, Randomized controlled trial. 3.3 Intervention strategies

Table 3 shows the summary of intervention methods and their effect on outcomes. Nine interventions reported in eight studies adopted gratitude list,27, 35, 36, 38-42 while one study conducted psychoeducational group sessions.37 No studies conducted incorporated only behavioral gratitude expression among workers. In studies using gratitude lists, six of eight studies asked participants to record "work-related gratitude".27, 35, 36, 39, 41, 42 Five studies were web-based,27, 35, 36, 39, 40 two studies were paper-based,41, 42 and in the remaining studies, the participants could choose one of the two.38 Ki incorporated a web-based gratitude list to 161 health care workers, including nurses, doctors, physical therapists, and occupational therapists, in China.39 Participants were asked to write down gratitude lists twice a week for 4 weeks, totaling eight lists. A negative activity was offered to the control group, which asked the participants to write down at least one hassle event at work.

TABLE 3. Interventions, outcomes, and results of the studies included in the systematic reviews: N = 9 Study ID Author/ Year Type Duration & Frequency/Program hours Gratitude conditions (plain/mixed) Control conditions (negative/neutral/positive) Outcomes Results on mental health (+, −, n.s.) Results on well-being (+, −, n.s.) 1 Ki, 2009.39 Gratitude list (web-based) Twice a week for 4 weeks in both conditions (8 times). Writing down at least one grateful event at work. (plain) Writing down at least one hassle event at work. (negative)

Perceived stress: PSS

PA and NA: CAS

Life satisfaction: SWLS

Depression: CES-D 10

Perceived stress: +

Depression: +

Positive affect: +

Negative affect: +

Life satisfaction: +

2 Baker, 2011.35 Gratitude list (web-based) Once a week, for 4 weeks (4 times). Reporting four positive events that occurred during the week while at work or related to their job. (plain) Measurement only(neutral)

Life satisfaction: SWLS

PA and NA: PANAS

Job satisfaction: JIG

NA

Life satisfaction: n.s.

Positive affect: n.s.

Negative affect: n.s.

Job satisfaction: n.s.

3 Otsuka et al., 2012.42 Gratitude list (paper-based) Once a week, for 4 weeks among both conditions (4 times). Writing down up to five people at work or in one's personal life to whom the participant was grateful during the past week in a journal. (plain) Writing down up to five events occurred at work or in one's personal life during the past week in a journal for 4 weeks. (neutral)

PA: PANAS

Life satisfaction: SWLS

Subjective happiness: SHS

NA

Positive affect: n.s.

Life satisfaction: n.s.

Happiness: n.s.

4 Chan et al., 2013.40 Gratitude list (web-based) Once a week for eight weeks among both conditions (8 times). Writing down up to three good things or events that happened to the participants. Setting at least 15 min at the end of the week to think about the meanings of these events. (plain) Writing down up to three bad things or events that happened to the participants. Setting at least 15 min at the end of the week to think about the meanings of these events. (negative)

Life satisfaction: SWLS

PA and NA: PANAS

NA

Life satisfaction: n.s.

Positive affect: n.s.

Negative affect: +

5 Kaplan et al., 2014.36 Gratitude list (web-based) Three times a week for 2 weeks (6 times). Logging in at least three times per week to record things they are grateful for related to their job. (plain)

Engaging in specific strategies to increase their social ties at work social and document those experiences on a secure Web site.

(positive)

PA and NA: JAWS NA

Positive affect: +

Negative affect: n.s.

6 Cheng et al., 2015.41 Gratitude list (paper-based) Twice a week for 4 weeks among both conditions (8 times). Writing diaries about work-related thankful events. (plain)

There are two control conditions.

Control A: Writing diaries about work-related hassle events. (negative)

Control B: Measurement only (neutral)

Depression: CES-D 10

Perceived stress: PSS

Compared with Control A,

Depression +

Perceived stress: +

Compared with Control B: Depression: +

Perceived stress: +

NA 7A Neumeier et al., 2017.27 Gratitude list (web-based) During the following 7 days at work (7 times). Writing down three things about work or workplace for which the participants genuinely feel grateful and reflect on them. (plain) Measurement only (neutral)

Happiness: Eight well-being indicatorsa

Happiness at work: eight well-being indicators at workb

NA

Happiness: +

Happiness at work: +

7B Neumeier et al., 2017.27 Gratitude list (web-based) During the following 7 days at work (7 times). Conducting a positive intervention program containing seven exercises (practicing gratitude, savoring the moment, you at your best, random acts of kindness, visualizing your best possible self, wearing a smile, and brainstorming meaningfulness). (mixed) Measurement only (neutral)

Happiness: Eight well-being indicatorsa

Happiness at work: eight well-being indicators at workb

NA

Happiness: +

Happiness at work: +

8 Cook et al., 2017.37 Others 2.5 h × 5 times in both conditions Compounded program based on positive psychology consisting of eight practices: practicing gratitude (writing and delivering a gratitude letter, gratitude journaling, and making thank you notes), mindfulness-based practice, helping and doing good deeds for others. (mixed) Participate in the controlled program to discuss topics related to their daily work. (neutral)

Perceived stress: PSS

Job satisfaction: SWWS

Perceived stress: + Job satisfaction: + 9 Ligon, 2019.38 Gratitude list (web-based and paper-based) Each group completed the assigned task for 10 min once per week on the day of their choice for 2 weeks at work (twice). Participants in the gratitude group were asked to recount and then write about three lifetime events for which they were grateful. (plain)

Control A: The optimism intervention involved writing about one's best possible future self (positive).

Control B: The control condition involved writing about typical daily work activities (neutral).

Burnout: MBI-GS

Burnout: n.s.

(Result of comparing Gratitude condition and Control B)

NA Note: +, significant favorable effects; −, significant adverse effects; n.s., non-significant effect; NA, not applicable. The difference between gratitude interventions between study ID 5A and 5B in general/Work-related subjective well-being was non-significant. Summary of the scales and other information. CAS, Chinese Affect Scale65; CES-D 10, Center for Epidemiologic Studies Depression Scale67, 68; JAWS, Job-Related Affective Well-being Scale72; JIG, Job In General scale70; MBI-GS, Maslach Burnout Inventory74; NA, Negative affect; PA, Positive affect; PANAS, Positive and negative affect schedule69; PSS, Perceived Stress Scale64; SHS, Subjective Happiness Scale71; SWLS, Satisfaction with Life Scale66; SWWS, Satisfaction with Work Scale.73 a Calculated by happiness, satisfaction with life and positive and negative affect.27 b Calculated by happiness at work, job satisfaction and positive and negative affect at work.27

Baker incorporated a web-based gratitude list to 163 employees recruited from a public university in the US.35 Participants were asked to wire four gratitude lists related to their job once a week for 4 weeks. The study adopted a neutral control group that measured the outcomes only.

Otsuka et al. incorporated a paper-based gratitude list to 38 employees in a local government in Japan.42 Participants were asked to write four gratitude lists once a week for 4 weeks. They listed five people at work or in one's personal life to whom the participant was grateful during the past week. The study set a neutral comparison that asked the participants to write up to five events at work or in one's personal life during the past week in a journal for 4 weeks.

Chan et al. incorporated a web-based gratitude lists to 78 schoolteachers receiving in-service training for postgraduate degrees in China.40 The author disseminated that this is an eight-week self-improvement project to enhance their well-being through self-reflection. Participants were asked to write three good things or events that happened to the participants once a week for 8 weeks. They also had at least 15 min to think about the meanings of these events at the end of the week. The control group was offered a negative activity that asked the participants to write down three bad things or events that happened to the participants and think about these events' meanings.

Kaplan et al. incorporated a web-based gratitude list to 112 staff members from two large public universities in the US.36 The participants were told that the purpose of the study was to explore avenues to increase well-being at work. Participants were asked to create six gratitude lists, recording grateful things related to their job three times a week for 2 weeks. The control group completed a positive activity that asked the participants to engage in specific strategies to increase their social ties at work three times per week and document those experiences on a secure Web site.

Cheng et al. incorporated a paper-based gratitude list to 102 health care workers, including physicians, nurses, physiotherapists, and occupational therapists recruited in five hospitals in China.41 Participants were asked to create eight lists by writing diaries about work-related thankful events twice a week for 4 weeks. Two control groups were included. One group completed a negative activity, writing diaries about work-related hassle events (Control A), while the neutral comparison group completed only the measurements (Control B).

Neumeier et al. incorporated a web-based gratitude list to 303 participants recruited via social media, newspapers, and radio in Australia27 and assigned to three groups (plan gratitude group, mixed gratitude group, and neutral control group). In the plan gratitude group, participants were asked to create seven lists by writing down grateful things about work or the workplace. Subsequently, they reflected on the grateful events during the following 7 days at work. In the mixed gratitude group, participants were asked to complete seven positive exercises during the consecutive 7 days at work: "practicing gratitude” (the same exercise the plain gratitude group), "savoring the moment” (mindfully savoring a pleasurable activity by paying attention to your immediate experience), “you at your best” (writing a story about a time when you were at your best at work and reflecting on your personal strengths displayed in the story), “random acts of kindness” (performing three acts of kindness in your workplace to benefit others or make others happy), “visualizing your best possible self” (thinking and writing about your best possible professional self/working life and defining specific goal that would help you to attain this best possible future scenario), “wearing a smile” (relaxing, finding something that makes you laugh, and frequently wearing a smile over the day), and “brainstorming meaningfulness” (brainstorming about tasks or elements in your work that you find meaningful or that are significant to you, and creating a mind map about sources of meaningful experiences in your job). The participants in the control group completed only the measurements.

Cook et al. conducted psychoeducation that included gratitude activities in group sessions.37 The participants were 44 high and middle school teachers from a single educational service district in the US. They completed five 2.5-h group sessions. The program was developed as an intervention promoting the well-being of teachers, helping them become resilient educators by focusing on eight practice areas: (1) increasing awareness and empowerment through mindfulness-based practices, (2) paying attention to the positive and practicing gratitude, (3) helping and doing good deeds for others, (4) identifying unhelpful thoughts and altering them to be more helpful, (5) developing good sleep habits, exercising regularly, and eating well, (6) clarifying values and committing to them, (7) establishing good social support, role models, and a mentor (relationships), and (8) rewarding oneself through relaxation and recreation. Practicing gratitude requires three specific activities: (1) writing and delivering a gratitude letter, (2) weekly gratitude journaling that identifies 3–5 things one is grateful for and why, in addition to imagining what the week would be like if the things did not happen, and (3) paying attention to the small things and writing thank you notes to life (e.g., thankful for being able to take a warm shower and getting clean because it makes me feel better; thankful for being able to take walks and having time to think and getting healthy all at once). The control group underwent five neutral 2.5-h group sessions to discuss topics related to their daily work.

In a more recent study, Ligon incorporated gratitude list to 148 mental health and customer service call centers in the US.38 It was both allowed to make gratitude list by paper-based or web-based. The participants were told that the purpose of the study was to examine whether three different work activities effectively reduce stress and increase hope, positive thinking, mental toughness, and confidence. The study lasted for 2 weeks, and the participants were asked to create two gratitude lists, one for each week, spending 10 min each week on the day of their choice to write about three lifetime events for which they were grateful. The comparison groups completed two other activities. In the positive activity group, the participants wrote about one's best possible future self once a week for 2 weeks. In the neutral group, they wrote about typical daily work activities once a week for 2 weeks.

3.4 Effects of the intervention programs on the outcomes

The included study adopted various mental health and well-being outcomes. Mental health included perceived stress (n = 3),37, 39, 41 depression (n = 2)39, 41 and burnout (n = 1).38 Well-being encompassed positive affect (n = 5),35, 36, 39, 40, 42 negative affect (n = 4),35, 36, 39, 40 life satisfaction (n = 4),35, 39, 40, 42 job satisfaction (n = 2),35, 37 happiness (n = 2),27,