Evaluation of a meaning in life intervention applied to work: A randomized clinical trial

Abstract

Meaning‐based interventions have demonstrated promising outcomes in enhancing meaning in life. However, understanding their efficacy in diverse contexts requires further research. This study aims to investigate the effectiveness of a meaning‐based intervention in a military sample and the impact of incorporating an additional emotion regulation module into the intervention. We conducted a randomized three‐group parallel trial with active‐duty military personnel: a waitlist control group (n = 21), a meaning‐based intervention group (n = 42), and a meaning‐based intervention with emotion regulation group (n = 43). Both intervention groups received six 2‐hour sessions and a follow‐up session 4 months later. Meaning in life and work‐related variables, as well as well‐being, depression, emotion regulation, and work‐related factors, were assessed using a mixed‐factorial analysis of variance (ANOVA). The results showed that only the meaning‐based intervention with an emotion regulation component significantly increased the presence of meaning in life. We also observed a trend toward improved well‐being scores over time that did not emerge for the other intervention. This study provides initial support for the efficacy of meaning‐based interventions within the military context when supplemented with emotion regulation skills.

INTRODUCTION

Meaning‐based interventions have garnered attention in scientific research due to their promising outcomes in addressing change, uncertainty, and stress in the general population (Pace et al., 2022). It is also worth exploring their effectiveness in the military context, where challenging and stressful situations are frequent.

Meaning in life, composed of the dimensions of coherence, purpose, and significance, allows individuals to perceive experiences with understanding, set a valuable course for living, and attribute meaning to daily life (Martela & Steger, 2016). Meaningful work is defined as the perception of being engaged in work aligned with one's strengths, values, and goals and the sense that the organisation's purpose connects with the life purpose of each worker (Allan et al., 2016).

Previous studies have demonstrated that the presence of the meaning in life is associated with multiple benefits for physical and mental health (Czekierda et al., 2017; Steger, 2018; Sutin et al., 2020). On the other hand, a decrease in meaning has been linked to the need to reduce tension in adverse circumstances, where understanding and constructing meaning can be complex (Cohen & Cairns, 2012). Although there is no consensus on this issue, some authors reported that seeking meaning in adverse circumstances is beneficial (Chu & Fung, 2020; Steger et al., 2008).

Meaning‐based interventions have been successfully applied in various populations. In clinical populations, they have proven effective in recovering from illnesses, addressing addictions, suicidal ideation, post‐traumatic stress, and fostering life purposes (Vos & Vitali, 2018). In nonclinical populations, interventions conducted with university samples, older adults, and the general population showed improvements in the presence of meaning, positive affect, life satisfaction, and reductions in negative affect, relative to control groups (Tsai et al., 2020; van Agteren et al., 2021). However, a recent study did not find any improvement in meaning (Olafsson & Kampman, 2022).

Several meaning‐based interventions applied in the general workforce have shown improvements in work engagement, reduction of burnout and depression, increased professional efficacy, and enhancement of meaning and satisfaction in work (Cantarero et al., 2022; Steger & Ekman, 2016; Zyl et al., 2020). In the military context, where challenging and stressful situations are inherent, meaning‐based interventions could be especially relevant (Morse et al., 2023). Researchers have reported that the presence of meaning is inversely related to the onset of post‐traumatic stress symptoms and insomnia and that it increases well‐being and organisational commitment (Fischer et al., 2020). Therefore, an intervention that combines elements of emotion regulation and meaning may help strengthen mental health in this population.

While meaning‐based interventions in nonclinical populations may have limited generalizability, it is essential to consider that the constructs of meaning in life and meaningful work tend to be stable over time (Steger & Kashdan., 2007). The most significant changes in these metrics are commonly observed after adverse events or therapeutic processes, perhaps because the person has a greater awareness of their emotions and needs (Vos & Vitali, 2018).

As a result, a possible innovation to improve the efficacy of meaning‐based interventions could be optimizing the emotional functioning of participants, thus fostering a stronger connection with their emotions and intrinsic values. This approach, supported by research such as that of Roth et al. (2019), highlights how integrative emotion regulation contributes to personal well‐being and affective relationships and suggests that accepting and addressing one's negative emotions can help build meaning, facilitate openness, and reduce defensive attitudes toward introspection. Similarly, recent studies have found connections between levels of mindfulness and emotion regulation and a greater sense of meaning in life (Baradaran et al., 2021; Chu & Mak, 2020; Crego et al., 2020), supporting the idea that an emotion regulation component could enhance the effects of meaning in life interventions.

THE PRESENT STUDY

The main objectives of this study were (a) to verify the effectiveness of a meaning‐based intervention addressing some previous methodological issues, such as the inclusion of a control group and the assessment of follow‐up results, and (b) to explore the potential beneficial effect of adding an emotion regulation component before the application of the meaning‐based intervention.

To achieve these objectives, we compared a meaning‐based intervention (MLI) with a combined emotion regulation and meaning‐based intervention (ER + MLI) and a waitlist control group. The primary outcomes were meaning in life and meaningful work, and the secondary outcomes included well‐being, emotion regulation, depressive symptoms, burnout, and work engagement.

Our hypotheses were the following:

Hypothesis 1

The MLI and ER + MLI intervention groups will show higher meaning in life and meaningful work scores after the intervention and at follow‐up compared with the control group.

Hypothesis 2

The MLI and ER + MLI intervention groups will exhibit higher scores in the secondary indicator's subjective well‐being, emotion regulation, depressive symptoms, burnout, and work engagement after the intervention and at follow‐up compared with the control group.

Hypothesis 3

The ER + MLI group will show greater increases in well‐being, emotion regulation, and meaning in life after the intervention than the MLI group.

METHOD

Study design

This study used a randomized clinical trial with a longitudinal, three‐group parallel design (MLI, ER + MLI, CG), following the Consolidated Standards of Reporting Trials (CONSORT) guidelines (Boutron et al., 2008). The initial sample size was calculated using G*Power, which determined that 128 participants were needed to achieve 80 per cent statistical power with a moderate effect size of .37 (Park et al., 2019). Originally, the required sample size for testing Hypotheses 1 and 2 was three groups of 20 participants each. However, Hypothesis 3 required a larger sample of 44 participants per group to ensure 80 per cent power (i.e. comparison of the two active interventions with an expected effect size of .14). Due to recruitment limitations, we could not reach that number of participants for the three groups. Instead, the control group size was adjusted to meet the sample needs for Hypotheses 1 and 2 and the intervention group size for Hypothesis 3. As a result, randomization followed a 2:2:1 allocation ratio across the MLI, ER + MLI, and control (CG) groups to ensure larger experimental groups. Actual recruitment included an additional 20 per cent due to a possible nonresponse rate.

Participants

The final sample included 126 members of the Spanish Armed Forces from various units in Madrid. Eligibility criteria included (a) being 18–45 years old, (b) active service in the Army, and (c) volunteering to participate. Exclusion criteria included being on a medical or a psychiatric leave. The retirement policy at the study site set the age limit at 45. Participants did not receive any compensation. Group allocation was conducted using a random number generator in Excel, as shown in Figure 1.

FIGURE 1.

Consolidated Standards of Reporting Trials (CONSORT) participant flow diagram. MLI, meaning in life intervention; ER + MLI, meaning in life intervention and emotion regulation component; CG, control group.

Ethical aspects

All procedures were in accordance with the ethical standards set by the university research committee and in compliance with the principles outlined in the 1964 Helsinki Declaration and its subsequent amendments, as well as other comparable ethical standards. The ethics committee of the School of Psychology approved the study. All participants gave written informed consent. We preregistered the study in ClinicalTrials.gov (NCT05336292).

Procedure

Participants were recruited through informational campaigns, conferences, and email dissemination. Group randomization occurred prior to the informational session. During this session, participants provided informed consent and completed baseline measures before starting the intervention. The intervention consisted of six weekly group sessions, each lasting 2 hours, with a follow‐up booster session at 4 months, which began with additional measures. Due to the coronavirus disease 2019 (COVID‐19) pandemic, the intervention was conducted when the situation was stable, and participants were vaccinated. The group program took place in person at a Military Unit classroom from October 2021 to October 2022, adhering to recommended safety protocols. Despite pandemic‐related challenges, the in‐person format was successfully maintained across all groups.

The intervention

The interventions were led by a therapist with 10 years of experience and specialized training in meaning‐based interventions, supported by a co‐therapist who ensured adherence to the sessions and provided logistical assistance. Each session followed a structured format, including objectives, attention training, theoretical explanations, group discussions, practical exercises, and assigned tasks. To maintain consistency, two workbooks were created, containing the program content and worksheets. Table 1 outlines the content of the MLI and ER + MLI interventions. The meaning‐related content was adapted from Steger (2022), and the emotion regulation component was based on Hervas' (2012) manual for optimal emotional processing. The primary goal of the emotion regulation module was to enhance participants' knowledge and skills in managing emotions in both work and daily life. Specifically, we focus on six areas: awareness, attention, labeling, acceptance, analysis, and modulation (Hervas, 2011) and include knowledge‐focused activities, metaphors, and experiential exercises (e.g. emotion diaries).

TABLE 1.

Structure and contents of the meaning in life intervention.

Meaning in life intervention (MLI)
Session 1	Developing mindfulness skills for a meaning in life (Chu & Mak, 2020) Meaning in Life Model (Martela & Steger, 2016) Exploring the strengths of character (Park et al., 2004) and sources of meaning in life (Littman‐Ovadia & Niemiec, 2016)
Session 2	Mindfulness exercise (Linehan, 1993) Identifying ways to use strengths and enhance the meaning in life (Littman‐Ovadia & Niemiec, 2016) Developing an action plan to improve strengths
Session 3	Mindfulness exercise (Linehan, 1993) Exploring personal values (My 80th Birthday Speech; Harris, 2009) Recognizing the importance of values in personal relationships Giving coherence and purpose to day‐to‐day experiences
Session 4	Mindfulness exercise (Linehan, 1993) Learning and applying the Meaning Work Model—SPIRE (Steger, 2017) Job crafting (Berg et al., 2013) Improving relationship satisfaction (Chaves et al., 2017)
Session 5	Coherence, purpose, and importance in goals (Sheldon et al., 2002) Linking our strengths and values to our goals and purposes Visualization of our future best self (Burton & King, 2004)
Session 6	Mindfulness exercise (Linehan, 1993) Sharing action plan and goals with a partner and make it public through a narrative
Emotion regulation + meaning in life program (ER + MLI) a
Session 1	Emotional awareness + mindfulness Psychoeducation about emotion (believes and myths, emotional functioning, etc.) (Hervas, 2012) Emotion regulation as a path to meaning in life
Session 2	Mindfulness skills (Linehan, 1993) Emotional awareness + emotional labeling (Hervas, 2012) Emotional acceptance, analysis, and modulation (Hervas, 2012)
Session 3	Mindfulness exercise (Linehan, 1993) Detecting and using one's character strengths (Seligman et al., 2005) Character strengths from past to present: visualization Specifying values (My 80th Birthday Speech; Harris, 2009)

^a Sessions 4–6 are the same as those of the MLI‐only intervention.

Measures

Meaning in life

The Meaning in Life Questionnaire (MLQ; Steger et al., 2006; Góngora & Solano, 2011) consists of 10 items rated on a 7‐point Likert scale (ranging from “absolutely untrue” to “absolutely true”). In this study, the Spanish version of the questionnaire was employed. The MLQ is composed of two subscales: one assessing the presence of meaning in life and the other measuring the search for meaning in life. Scores on each subscale range from 5 to 35, with higher mean scores indicating higher levels of presence or search for meaning, respectively. The internal reliability of the subscales in this study was high, with Cronbach's alpha coefficients of .91 for presence and .96 for search. For the purpose of our study, we focused on the Presence of Meaning in Life subscale as a representation of increased meaning, consistent with other studies (Kwok et al., 2024) and aligned with our preregistered hypotheses.

Meaningful work

The Work and Meaning Inventory (WAMI; Steger et al., 2012, Duarte‐Lores et al., 2023) consists of 10 questions, rated on a 5‐point Likert scale (from “absolutely untrue” to “absolutely true”). We used the overall meaningful work score with scores ranging from 10 to 50. Higher mean scores reflect higher levels of meaningful work. The internal reliability was Cronbach's alpha of .93.

Burnout

The Maslach Burnout Inventory (MBI; Maslach & Jackson, 1981, Salanova et al., 2000) consists of 15 questions rated on a 7‐point Likert scale (from “never” to “every day”). This instrument measured burnout across three scales: five items assessed emotional exhaustion, four items evaluated cynicism, and six items measured professional efficacy. The internal reliability was Cronbach's alpha of .93, .95, and .90 for each subscale, respectively.

Engagement

The Utrecht Work Engagement Scale (UWES‐9; Schaufeli & Bakker, 2003, Schaufeli et al., 2006) consists of nine questions rated on a 7‐point Likert scale (from “never” to “always or every day”). We used in our study the global score including nine items reflecting the three dimensions of work engagement. The higher the average, the greater the work commitment. The internal reliability was Cronbach's alpha of .92, .94, and .91 for each subscale, respectively.

Depression

The Patient Health Questionnaire (PHQ‐9; Kroenke et al., 2001, Diez‐Quevedo et al., 2001) consists of nine questions rated on a 3‐point Likert scale (from “not at all” to “nearly every day”). The PHQ‐9 total score ranges from 0 to 27 (scores of 5–9 are classified as mild depression, 10–14 as moderate depression, 15–19 as moderately severe depression, and ≥20 as severe depression). The internal reliability was Cronbach's alpha of .94.

Emotion regulation

The Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004; Hervas & Jódar, 2008) assesses difficulties in the awareness, understanding, or modulation of emotion. It consists of 28 items with five subscales: Nonacceptance of Emotional Responses, Interference, Lack of Emotional Control, Inattention, and Confusion. Participants responded on a Likert scale from 1 (almost never) to 5 (almost always). Higher scores reflect greater ED. The Lack of Attention subscale was excluded from the DERS total score due to psychometric concerns. Research indicates that the Lack of Attention scale has weak correlations with the rest of the subscales and negatively impacts model fit (i.e. second‐order single factor including all subscales). Excluding it results in a more accurate and reliable measure of emotional dysregulation (Monell et al., 2022). The total score ranged from 24 to 120. The internal reliability was Cronbach's alpha of .97.

Well‐being

The Pemberton Happiness Index (PHI; Hervas & Vázquez, 2013) consists of 11 items related to remembered well‐being, each with an 11‐point Likert scale, and 10 items related to experienced well‐being (positive and negative events that occurred the day before), with dichotomous response options (yes/no). The sum of these scales produced an integrative well‐being index. The internal reliability was Cronbach's alpha of .94.

Satisfaction

The Client Satisfaction Questionnaire (CSQ‐8; Roberts et al., 1984) is a structured survey used to assess level of satisfaction with care. Items are scored on a Likert scale from 1 (low satisfaction) to 4 (high satisfaction) with different descriptors for each response point. Total scores range from 8 to 32, with higher scores indicating greater satisfaction. The questionnaire was adapted for the military context, tailored for individuals without specific health concerns. Relevant treatment‐related terms were replaced with program‐related language (see Table S2a,b). The internal reliability was Cronbach's alpha of .86.

Additional measures

A questionnaire assessed the perceived effect of the program on meaning in life, meaningful work, and personal well‐being with seven Likert scale questions evaluating the perceived effect of the program (see Table S1). There were four open‐ended questions for participants (see Tables S3–S6). The internal reliability was Cronbach's alpha of .84.

Data analysis

Data were analyzed using IBM SPSS Statistics for Windows, version 18.0 (IBM Corp., 2009). A mixed‐factorial analysis of variance (ANOVA) was conducted, with three groups (MLI, ER + MLI, and CG) and three time points (pre‐intervention, post‐intervention, and follow‐up) for each primary and secondary outcome variable. Main effects were analyzed to assess differences due to group and time independently, while interaction effects evaluated how these factors combined to influence outcomes over time. Sphericity was tested, and when violated, the Greenhouse–Geisser correction was applied. The Holm–Bonferroni method was used to adjust for multiple comparisons across hypotheses (ANOVAs), and Bonferroni adjustments were applied for all post hoc analyses.

RESULTS

The sample consisted of 69 per cent men (n = 86) and 31 per cent women (n = 37), with 61.2 per cent having a medium to upper educational level. No significant differences were found between groups in demographic terms. No differences were found in primary variables and well‐being between groups at the beginning of the study in MLQ‐P (F(2, 126) = 2.46, p = .89), MLQ‐S (F(2, 126) = 0.87, p = .41), WAMI (F(2, 126) = 0.75, p = .47), and PHI (F(2, 123) = 2.42, p = .09). See Table 2 for demographics and Table 3 for the means and standard deviations of each condition.

TABLE 2.

Baseline characteristics and group differences.

Demographic characteristics	MLI (n = 42)	ER + MLI (n = 43)	CG (n = 21)	Group differences
Age (%)
18–25	10.3	6.4	7.2	$x^2\left(6,123\right)=6.69$, $p=.35$ Cramer's V = .16, p = .35
26–35	29.2	36.2	32.1
36–40	29.2	29.8	50.0
41–45	31.3	27.6	10.7
Levels of education (%)
Basic‐general education	16.7	17	25.0	$x^2\left(6,123\right)=4.40$, $p=.62$ Cramer's V = .13, p = .62
Bachelor/professional medium	64.6	63.8	57.1
University/higher professional training	18.8	14.9	17.9
Master's/PhD	0	4.3	0
Gender (%)
Man	75	61.7	75	$x^2\left(2,123\right)=2.44$, $p=.29$ Cramer's V = .14, p = .29
Woman	25	38.3	25
Mean no. of sessions attended, M (SD)	5.98 (0.15)	5.98 (0.15)		t(−0.017) = 83, p = .98

Abbreviations: CG, control group; MLI, meaning in life intervention; ER +MLI, emotion regulation + meaning in life intervention.

TABLE 3.

Descriptive statistics and mixed‐factorial analysis of variance (ANOVA) results for all variables by groups and times.

	Group CG, n = 21; MLI, n = 42; ER + MLI, n = 43	MT1 (SD)	MT2 (SD)	MT3 (SD)	F
WAMI	CG	36.81 (8.49)	37.38 (7.59)	36.86 (7.77)	(T) F(1.87, 193) = 4.26 p = .02, ƞ 2 = .04 Corrected α‐value = .007 (IT) F(3.75, 193) = 1.16 p = .33, ƞ 2 = .02 Corrected α‐value = .01 (I) F(2, 103) = 0.09 p = .90, ƞ 2 = .002 Corrected α‐value = .013
	MLI	35.81 (6.47)	37.10 (5.30)	36.98 (6.17)
	ER + MLI	34.58 (7.99)	36.81 (6.95)	37.44 (8.18)
MLQ‐P	CG	26.29 (5.18)	25.95 (5.12)	26.14 (5.66)	(T) F(1.85, 191) = 3.43 p = .03, ƞ 2 = .03 Corrected α‐value = .008 (IT) F(3.71, 191) = 2.79 † p = .03, ƞ 2 = .05 Corrected α‐value = .005 (I) F(2, 103) = 0.46 p = .64, ƞ 2 = .01 Corrected α‐value = .005
	MLI	26.67 (4.70)	27.07 (4.24)	27.24 (3.92)
	ER + MLI	26.02 (5.51)	28.05 (5.75)	46.63 (5.29)
MLQ‐S	CG	17.52 (7.90)	12.57 (5.58)	13.24 (6.30)	(T) F(2, 206) = 2.92 p = .06, ƞ 2 = .05 Corrected α‐value = .013 (IT) F(4, 206) = 5.90* p < .001, ƞ 2 = .10 Corrected α‐value = .004 (I) F(2, 103) = 3.00 p = .05, ƞ 2 = .06 Corrected α‐value = .004
	MLI	18.29 (7.56)	19.62 (8.94)	18.74 (8.66)
	ER + MLI	18.40 (8.05)	19 (8.94)	18.74 (8.66)
PHI	CG	8.00 (0.94)	8.02 (1.20)	8.06 (1.17)	(T) F(2, 206) = 10.08* p < .001, ƞ 2 = .09 Corrected α‐value = .005 (IT) F(4, 206) = 3.03 p = .02, ƞ 2 = .06 Corrected α‐value = .005 (I) F(2, 103) = 0.17 p = .85, ƞ 2 = .003 Corrected α‐value = .01
	MLI	7.78 (1.23)	8.01 (1.08)	8.04 (1.16)
	ER + MLI	7.45 (1.72)	8.00 (1.41)	8.08 (1.35)
UWES_VI	CG	4.27 (1.32)	4.62 (1.00)	4.51 (1.13)	(T) F(1.82, 187) = 3.80 p = .03, ƞ 2 = .04 Corrected α‐value = .01 (IT) F(3.64, 187) = 1.95 p = .11, ƞ 2 = .04 Corrected α‐value = .006 (I) F(2, 103) = 0.08 p = .92, ƞ 2 = .002 Corrected α‐value = .03
	MLI	4.40 (1.11)	4.33 (1.11)	4.38 (1.10)
	ER + MLI	4.26 (1.16)	4.56 (0.93)	4.51 (1.01)
UWES_DE	CG	4.30 (1.50)	4.62 (1.12)	4.51 (1.32)	(T) F(1.84, 190) = 2.27 p = .11, ƞ 2 = .02 Corrected α‐value = .03 (IT) F(3.69, 190) = 0.317 p = .85, ƞ 2 = .05 Corrected α‐value = .05 (I) F(2, 103) = 0.236 p = .79, ƞ 2 = .005 Corrected α‐value = .007
	MLI	4.21 (1.20)	4.38 (1.18)	4.40 (1.13)
	ER + MLI	4.21 (1.48)	4.27 (1.30)	4.29 (1.49)
UWES_AB	CG	4.02 (1.36)	4.33 (1.08)	4.25 (1.15)	(T) F(2, 206) = 0.90 p = .41, ƞ 2 = .009 Corrected α‐value = .05 (IT) F(4, 206) = 0.45 p = .77, ƞ 2 = .009 Corrected α‐value = .03 (I) F(2, 103) = 0.265 p = .77, ƞ 2 = .005 Corrected α‐value = .006
	MLI	4.09 (1.20)	4.12 (1.10)	4.05 (1.21)
	ER + MLI	3.94 (1.48)	3.98 (1.46)	4.01 (1.48)
MBI‐1	CG	1.84 (1.09)	1.44 (0.67)	1.72 (1.13)	(T) F(1.86, 191) = 7.96* p = .001, ƞ 2 = .07 Corrected α‐value = .004 (IT) F(3.72, 191) = 1.79 p = .14, ƞ 2 = .03 Corrected α‐value = .006 (I) F(2, 103) = 0.19 p = .83, ƞ 2 = .004 Corrected α‐value = .008
	MLI	1.92 (1.20)	1.81 (1.06)	1.66 (0.90)
	ER + MLI	2.01 (1.22)	1.58 (0.96)	1.51 (1.05)
MBI‐2	CG	1.76 (1.48)	1.38 (1.04)	1.34 (1.16)	(T) F(2, 206) = 11.56* p < .001, ƞ 2 = .10 Corrected α‐value = .005 (IT) F(4, 206) = 1.41 p = .23, ƞ 2 = .27 Corrected α‐value = .008 (I) F(2, 103) = 0.11 p = .90, ƞ 2 = .002 Corrected α‐value = .02
	MLI	1.73 (1.52)	1.77 (1.34)	1.36 (0.98)
	ER + MLI	1.77 (1.54)	1.84 (1.50)	1.33 (1.14)
MBI‐3	CG	4.63 (0.94)	4.81 (0.79)	4.75 (0.90)	(T) F(1.79, 184) = 2.43 p = .10, ƞ 2 = .023 Corrected α‐value = .02 (IT) F(3.58, 184) = 0.65 p = .61, ƞ 2 = .012 Corrected α‐value = .02 (I) F(2, 103) = 0.05 p = .95, ƞ 2 = .001 Corrected α‐value = .016
	MLI	4.65 (0.94)	4.73 (0.87)	4.67 (0.98)
	ER + MLI	4.64 (0.90)	4.76 (0.88)	4.84 (0.90)
PHQ‐9	CG	4.05 (3.81)	3.43 (3.10)	4.38 (3.81)	(T) F(1.80, 185) = 4.63* p = .01, ƞ 2 = .04 Corrected α‐value = .01 (IT) F(3.61, 185) = 1.54 p = .20, ƞ 2 = .029 Corrected α‐value = .007 (I) F(2, 103) = 0.36 p = .70, ƞ 2 = .007 Corrected α‐value = .006
	MLI	5.43 (4.66)	4.57 (3.64)	3.95 (3.94)
	ER + MLI	5.58 (4.77)	4.49 (4.11)	4.19 (3.95)
DERS_TOT	CG	42.95 (11.01)	41.05 (9.86)	42.19 (10.69)	(T) F(2, 206) = 6.32* p = .002, ƞ 2 = .06 Corrected α‐value = .006 (IT) F(4, 206) = 1.27 p = .36, ƞ 2 = .021 Corrected α‐value = .013 (I) F(2, 103) = 1.55 p = .22, ƞ 2 = .029 Corrected α‐value = .005
	MLI	49.38 (15.37)	47.26 (11.39)	45.10 (15.11)
	ER + MLI	50.23 (17.03)	48.33 (17.05)	44.07 (11.76)

Note: Groups: CG, control group; MLI, meaning in life intervention; ER + MLI, emotion regulation + meaning in life intervention. M_T1, mean pretest; M_T2, mean posttest; M_T3, mean at follow‐up; (T), time; (IT), interaction; (I), intersubject; ƞ ², effect size. Corrected α‐values were calculated using the Holm–Bonferroni method.

Abbreviations: DERS_TOT, Emotion Regulation scale without Lack of Attention subscale; MBI, Maslach Burnout Inventory; MBI‐1, MBI Exhaustion scale; MBI‐2, MBI Cynicism scale; MBI‐3, MBI Professional Efficacy scale; MLQ, Meaning in Life Questionnaire; MLQ‐P, MLQ Presence of Meaning in Life scale; MLQ‐S, MLQ Search for Meaning in Life scale; PHI, Pemberton Happiness Index; PHQ‐9, Patient Health Questionnaire‐9; UWES, Utrecht Work Engagement Scale; UWES_AB, UWES Absorption scale; UWES_DE, UWES Dedication scale; UWES_VI, UWES Vigor scale; WAMI, Work and Meaning Inventory.

^† Nonsignificant for Hypotheses 1 and 2 but significant for Hypothesis 3.

^*Significant after corrected α‐values.

Participant flow and program adherence

Out of the 126 participants, 123 were randomly assigned to each intervention arm (see Figure 1). Three participants were excluded as they were on leave at the start of the intervention. The dropout rate during the program was 12.5 per cent in the MLI group, 8.5 per cent in the ER + MLI group, and 25 per cent in the control group (CG). In the MLI group, six participants enrolled but did not start the program due to work requirements. In the ER + MLI group, four participants did not initiate the program for the same reason. In CG, five participants did not complete post‐intervention measures for unspecified reasons. Data were missing for only two individuals at follow‐up because they relocated. Ninety‐eight per cent of participants attended all six program sessions, with the remaining 2 per cent attending at least five group sessions and making up the remaining session individually.

Program satisfaction, rated on a scale of 0–32, was high in both interventions (M = 28.18, SD = 3.37, see Table S1). Both interventions showed perceived improvement in meaningful life, work meaning, and well‐being due to the program (see Table S1). Participants' comments on the program evaluation are available in the Supporting Information (see Tables S3–S6).

Result analysis

Differences between active and control groups for primary variables (Hypothesis 1)

Meaningful work (WAMI)

As observed in Table 3, there were no significant main effects of time, between‐subject effects, or interaction effects. Since the interaction was not significant, there is no evidence that the MLI and ER + MLI intervention groups showed higher meaningful work scores after the intervention and at follow‐up compared with the control group.

Presence of Meaning in Life (MLQ‐P)

A trend‐level significant interaction effect was found for the MLQ‐P variable, with a small effect size (see Table 3). Post hoc comparisons revealed a significant improvement in the ER + MLI intervention group between pre‐ and post‐intervention (p < .001) and a significance effect between pre‐intervention and follow‐up (p = .003). However, no significant differences were found in the MLI intervention group or the control groups. A graphical representation of these results can be seen in Figure 2.

FIGURE 2.

Interaction effects between the main effect of time (pretest–posttest and follow‐up) and condition (MLI, ER + MLI, CG) on the Presence of Meaning in Life variable. *Significant within‐group comparison using Bonferroni correction (p < .05). MLQ‐P, Meaning in Life Questionnaire's Presence of Meaning in Life variable; MLI, meaning in life intervention; ER + MLI, meaning in life intervention and emotion regulation component; CG, control group.

Search for Meaning in Life (MLQ‐S)

A significant interaction effect was found for the MLQ‐S variable, with a small effect size (see Table 3). Post hoc comparisons revealed a significant decrease in search for meaning within the control group (CG) between pre‐ and post‐intervention (p < .001), as well as between pre‐intervention and follow‐up (p = .002).

Differences between active and control groups for secondary variables (Hypothesis 2)

Well‐being (PHI)

For the PHI variable, a trend was found in the interaction result, with a moderate effect size (see Table 3). Post hoc comparisons revealed significant improvement in the ER + MLI intervention group between pre‐ and post‐intervention and between pre‐intervention and follow‐up (p < .001) but not in the MLI intervention nor control groups.

Emotion regulation (DERS)

No significant interaction effects were found for this variable. However, a significant main effect of time was found, with a moderate effect size. Post hoc comparisons revealed a significant improvement between pre‐intervention and follow‐up (p = .002) in both intervention groups but not in the control group.

Depression (PHQ‐9)

No significant interaction effects were found for this variable. However, a significant main effect of time was observed, with a moderate effect size. Post hoc comparisons indicated a significant improvement in depression scores between pre‐intervention and follow‐up (p < .05) in both intervention groups, but not in the control group. Regarding the work‐related variables, the Vigor (UWES_VI), Dedication (UWES_DE), and Absorption (UWES_AB) scales did not show significant main effects or interaction effects. For the Exhaustion (MBI‐1) variable, a significant main effect of time was found with a moderately large effect size. Post hoc comparisons revealed a significant improvement between pre‐intervention and post‐intervention (p < .01) and between pre‐intervention and follow‐up (p < .01) only in the ER + MLI intervention group. For the Cynicism (MBI‐2) variable, a significant main effect of time was found, with a moderate effect size. Post hoc comparisons revealed a significant improvement between pre‐intervention and follow‐up (p < .001) only in the intervention groups (i.e. ER + MLI and MLI groups) and not in the control groups. No significant differences were found in other work‐related variables regarding main effects or interactions (see Table 3).

Differences between ER + MLI and MLI groups for meaning in life, well‐being, and emotional regulation (Hypothesis 3)

Presence of Meaning in Life (MLQ‐P)

As reported above, a mixed ANOVA was conducted and interaction was calculated (see Table 3). Using Holm–Bonferroni correction for 12 variables, the p‐value does not reach significance (i.e. corrected α = .005). However, this may be excessively conservative since in the preregistered protocol, we included only three variables for Hypothesis 3. Using the Holm–Bonferroni correction for three variables, the interaction of this ANOVA reaches significance (i.e. corrected α = .03). Thus, following this argument, we concluded that considering this result significant was more adequate for preventing a type 2 error. Post hoc comparisons revealed an improvement in the ER + MLI intervention group between pre‐ and post‐intervention (p < .001) and between pre‐intervention and follow‐up (p = .003). No significant results were found for the MLI intervention. See Figure 2 for a graphical representation of this result.

Well‐being (PHI)

As previously mentioned, for the PHI variable, a trend toward significance in the interaction effect was observed (see Table 3). Post hoc comparisons indicated a significant improvement in the ER + MLI intervention group between pre‐ and post‐intervention, as well as between pre‐intervention and follow‐up (p < .001). No significant effects were found for the MLI intervention group.

Emotion regulation (DERS)

No significant interaction effects were found for this variable.

DISCUSSION

This study is the first experimental investigation of a meaning in life intervention within a military population. The findings offer valuable insights into how a meaning in life intervention, combined with an emotion regulation component, impacts meaning in life in this specific context. Through comprehensive data analysis, we derived important conclusions regarding the intervention's effectiveness, potential applications, and areas for further research.

The primary objective of this study was to assess the differential efficacy of a meaning in life intervention, both with and without an emotion regulation component. The results indicate improvements in life meaning, particularly in the group that received the intervention with emotion regulation. Regarding well‐being, the trend toward significance in the interaction between the variables does not allow for definitive conclusions about the combined effect of the interventions on well‐being. However, post hoc comparisons suggest that the integration of emotion regulation and meaning in life interventions may generate more consistent and lasting effects on well‐being compared with the meaning‐based intervention alone.

These results highlight the value of incorporating emotion regulation skills (e.g. labeling, acceptance, and understanding), as enhancing participants' ability to manage emotions may reduce barriers to the introspective work necessary for cultivating meaning in life. Previous research supports the idea that emotions are closely linked to the development of meaning in life (Roth et al., 2019). However, future studies should explore these relationships in more detail and consider other influencing factors. Our results align with prior research demonstrating the positive effects of meaning in life interventions on meaning in life in general populations (Tsai et al., 2020; van Agteren et al., 2021). The perception of a meaningful life, with purpose and coherence, directly influences overall well‐being. This supports the integration of emotion regulation with meaning‐based interventions to enhance subjective well‐being. However, our findings contrast with Olafsson and Kampman (2022), who did not observe significant improvements in presence of meaning in life or well‐being in their intervention. Differences in intervention design and the absence of follow‐up measures in their study may account for the discrepancy. Their intervention focused on motivational and cognitive aspects but neglected elements like belonging and self‐transcendence. Additionally, the lack of follow‐up assessments may have limited the detection of longer term effects.

COVID‐19 was an ongoing factor during the data collection period (October 2021 to October 2022). Although we could not directly analyze its impact on the results, its potential moderating effects should be discussed. The results showed that participants' presence of meaning in life increased in the intervention groups, whereas search of meaning declined in the control group. This could be attributed to the supportive space provided by the interventions, which helped participants reflect on their meaning in life, particularly during the challenging pandemic period. These findings are consistent with other studies that emphasize the importance of reflection and support when addressing fluctuations for meaning (Steger et al., 2008, 2023). Thus, it is possible that the pandemic has increased the need for meaning and fostered the efficacy of the active interventions.

Regarding work‐related meaning, only marginal changes were found in the group that received the emotion regulation component. This contrasts with Cantarero et al. (2022), who observed significant improvements after their online program, possibly due to the privacy afforded by the online format. It appears that meaning in life and work‐related meaning may evolve along separate trajectories, with work‐related meaning perceived as more influenced by external factors (Bailey et al., 2019).

Significant changes were also observed in work‐related variables such as exhaustion and cynicism, as well as in personal domains like depression and emotion regulation for both intervention groups. However, since the time × group interaction did not reach statistical significance, the relevance of these findings is uncertain. The small to moderate effect sizes we found suggest that larger sample sizes may be necessary to detect these differences (i.e. to increase statistical power since the effect size was smaller than expected).

Feedback from participants indicated high acceptability and satisfaction with the interventions in concordance with a previous pilot study using the same meaning‐based intervention (Ríos & Hervas, 2024). The low dropout rate and high attendance underscore the program's appeal. While the military context may encourage discipline, voluntary participation was emphasized, and no additional incentives were offered.

Several limitations must be acknowledged. The relatively short intervention duration may have limited the magnitude of change in some variables. Future studies could extend the intervention period and include longer follow‐up assessments. Additionally, the tools used to measure meaning in life and work‐related meaning may not have been sensitive enough to capture subtle changes. Incorporating more refined instruments, such as the recently published three‐dimension meaning scale (Martela & Steger, 2022), could provide deeper insights on what specific variables may change to a higher degree after the intervention. Lastly, potential social desirability bias in measuring work‐related variables may have influenced responses. Future research should broaden the focus beyond individual workers to include organisational factors like leadership and working conditions (Blustein et al., 2023).

In conclusion, this study makes a significant contribution to research on meaning in life interventions and their role in enhancing well‐being, particularly in military contexts. The findings suggest that integrating emotion regulation into these interventions is valuable for improving mental health. Further research with larger, more diverse samples is needed to validate and expand these results. These insights have important implications for developing preventive mental health programs and fostering resilience in military settings.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

The study was conducted following the Declaration of Helsinki and approved by the Ethics Committee of the School of Psychology of the Complutense University of Madrid (protocol code 2019/20‐049, date of approval: July 31, 2020). Informed consent was obtained from all participants involved in this study.

Supporting information

Table S1. Description of the perception of increase in meaning in life, work, and well‐being

Table S2a Spanish Version of the Adaptation of CSQ‐8

Table S2b English Version of the Adaptation of CSQ‐8

Table S3a Preferred Session in MLI Intervention

Table S3b Preferred Session in Emotional Regulation and MLI Intervention

Table S4a The Best of the MLI Intervention

Table S4b The Best of the Emotional Regulation and MLI Intervention

Table S5a The Negative Aspects of the MLI Intervention

Table S5b The Negative aspect of Emotional Regulation and MLI Intervention

Table S6a Aspects to Improve in MLI Intervention

Table S6b Aspects to Improve in Emotional Regulation and MLI Intervention

Ríos, D. C. , & Hervas, G. (2025). Evaluation of a meaning in life intervention applied to work: A randomized clinical trial. Applied Psychology: Health and Well‐Being, 17(1), e12622. 10.1111/aphw.12622

DATA AVAILABILITY STATEMENT

Data will be made available upon reasonable request.