Couples’ therapies can improve clinical outcomes of patients with post-traumatic stress disorder: meta-analysis of eighteen clinical studies

Abstract

Background

Partners play an important role in a couple's life, and the whole family may face various challenges when meeting post-traumatic stress disorder (PTSD). This study aimed to explore the therapeutic effect of couples’ therapies for PTSD.

Methods

This study obtained relevant data from databases including Ovid Medline(R) ALL, Ovid Embase, Cochrane Central Register of Controlled Trials, and American Psychological Association PsycInfo until July 03, 2025.

Results

The couples’ therapies showed an adherence rate (Proportion = 69%, 95%CI: 64% to 73%). Comparing with before couples’ therapies, there were significant improvement in for patients (SMD = -0.79, 95%CI: -1.13 to -0.46) and for partners (SMD = -0.78, 95%CI: -1.05 to -0.52) after therapy, and there were significant improvement in intimate relationship functioning for patients (SMD = 0.27, 95%CI: 0.12 to 0.42) and partners (SMD = 0.38, 95%CI: 0.20 to 0.56), and in depression level for patients (SMD = -0.61, 95%CI: -0.82 to -0.40) and partners (SMD = -0.29, 95%CI: -0.50 to -0.07) after couples’ therapies. Outcomes included comparisons with subgroup analysis which revealed that PTSD symptoms had better consequence for patients in age < 40 (SMD = -1.04, 95%CI: -1.93 to -0.16), male (SMD = -1.48, 95%CI: -1.95 to -1.01), and sessions ≥ 10 (SMD = -1.14, 95%CI: -1.53 to -0.74), trauma-focused (SMD = -1.17, 95%CI: -1.60 to -0.73), non-military personnel (SMD = -1.37, 95%CI: -1.94 to -0.79) and CBCT (SMD = -0.87, 95%CI: -1.24 to -0.49) had preferable outcomes for in PTSD symptoms.

Conclusions

This study demonstrated the improvement of couples’ therapies in PTSD symptoms, intimate relationship functioning and depression level. It was also found that the adherence rate of couples’ therapies for PTSD was higher in Males, and the PTSD symptoms showed more conspicuous improvement in patients younger than 40 years of age, males, longer sessions of therapy, trauma-focused and CBCT after couples’ therapies.

Trial registration

The protocol of meta-analysis was available at the Open Science Framework (https://osf.io/s39jn/).

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03464-8.

Background

Post-traumatic stress disorder (PTSD) was a long-term negative effect resulting from exposure to severe stressors [1], such as combat, natural disasters, or other events. The prevalence rate of PTSD in the general population was reported to be 1–8%, while psychiatric institutions reported a significantly higher prevalence of up to 50% [2]. This condition imposed a substantial mental health burden, characterized primarily by recurrent trauma, hypervigilance, negative cognition, avoidance behaviors, and even suicidal tendencies that not only caused significant distress to affected individuals but also disrupted family relationships due to emotional dysregulation and social withdrawal [3]. Moreover, parental PTSD was found to significantly impair parent–child bonding [4] and adversely affect children's psychosocial development. Studies indicated that children of affected parents exhibited higher susceptibility to anxiety and depression, along with an increased likelihood of displaying aggressive behaviors [5]. These cascading effects posed substantial challenges to family functioning.

Given the interpersonal toll of PTSD, couple-based therapies emerged as a promising treatment modality [6]. Cognitive behavioral combined therapy (CBCT) was developed as a targeted intervention for individuals with PTSD and their partners [7]. This approach integrated cognitive behavioral techniques with couple therapy, aiming not only to alleviate PTSD symptoms in affected individuals but also to enhance interpersonal relationships between partners [6]. The intervention employed a 15-session combined treatment protocol designed to achieve tripartite therapeutic objectives: (a) enhancing conflict resolution and communication competencies, (b) facilitating marital relationship reconstruction by promoting proactive trauma confrontation rather than avoidance behaviors, and (c) targeting the modification of maladaptive cognitions arising from traumatic event misinterpretations and dysfunctional belief systems concerning trust, autonomy, and emotional connection [8].

Hitherto, one study has been found to demonstrate the effectiveness of couples’ therapies for PTSD, but the results had some limitations in profundity [9]. It showed that in both couple and individual studies, there were moderate to large reductions in PTSD symptoms and small but significant improvements in close relationship functioning, but other outcomes resulting from couples’ therapies and possible influencing factors were not explored. As couple therapy is jointly influenced by patients and their families, it can also enhance the intimate relationship functioning while improving the PTSD symptoms of patients. During the treatment process, the psychological state of patients is often quite complex. Due to the profound impact of traumatic experiences, they may exhibit significant emotional fluctuations or defensive behaviors. This not only increases the difficulty of the treatment but may also affect the patient's cooperation with the treatment. If the patient lacks compliance with the treatment, it may limit the effectiveness of the intervention measures. Moreover, many patients with PTSD also experience depressive symptoms [10]. This comorbidity phenomenon may further weaken the patient's motivation and emotional regulation ability, making them more passive when facing the treatment. By introducing the support system of the partner, it helps patients establish a more positive emotional coping mechanism, thereby reducing the impact of depressive emotions. Therefore, this study aimed to explore more outcome indicators related to couples’ therapies for PTSD, such as therapy adherence rate, PTSD symptoms, intimate relationship functioning and depression level based on previous clinical studies, and further explored the influence of therapeutic effect according to the characteristics of the population and the characteristics of specific therapy methods.

Methods

Trial registration

The protocol of meta-analysis was available at the Open Science Framework (https://osf.io/s39jn/).

Literature search and screening process

A search of the four databases, including Ovid Medline(R) ALL, Ovid Embase, Cochrane Central Register of Controlled Trials, and American Psychological Association PsycInfo, initially identified all relevant articles until July 03, 2025. The search keywords mainly included “post-traumatic stress disorder” and “couples’ therapies”. The detailed search strategies were shown in the Supplementary Method 1. During the literature screening process, duplicates and entirely irrelevant records were first removed. Two reviewers independently screened the titles, abstracts, and full texts of the remaining records. After excluding obviously irrelevant studies, the final set of included literature was determined. Any discrepancies between the reviewers were resolved through discussion or, if necessary, arbitration by a third reviewer.

Inclusion and exclusion criteria

The inclusion criteria were formulated for this study: 1) adults PTSD patients (≥ 18 years old); 2) Intervention was a concurrence therapy for patients and partners by psychotherapy, including CBCT, PTSD family education (PFE), structured approach therapy (SAT), behavioral family therapy (BFT), psycho-drug combination therapy, and abbreviated, intensive, multi-couple group version of CBCT (AIM-CBCT), etc.; 3) Comparison between pre- and post-couples’ therapies; 4) Outcomes: the outcomes included adherence rate (The adherence rate reflected the proportion of couples who completed the treatment regimen.), PTSD symptoms, intimate relationship functioning and depression level between before couples’ therapies and different follow-up for 1-mouth, 3-mouth, and 6-month. The primary outcomes were defined as adherence rate and PTSD symptoms, and the secondary outcomes were defined as intimate relationship functioning and depression level; and 5) Study design: randomized controlled trials (RCT) and uncontrolled trial.

Exclusion criteria were formulated for this study: 1) Sample size was less than five couples, 2) The standard of measurement results were not consistent with the scale scores, 3) Data were missing or did not have the required data.

Quality assessment

Two authors independently assessed the quality of each study. The six bias domains based on the Cochrane Risk of Bias 2 (RoB2) tool [11] were used to evaluate the quality of RCT studies, and the seven bias domains based on the Risk Of Bias In Non-randomised Studies of Interventions (ROBINS-I) tool [12] were used to evaluate the quality of uncontrolled trial. Each bias domain had several corresponding questions. Six bias domains in RoB2 were presented as “low risk, some concerns and high risk”, and 7 bias domains in ROBINS-I were presented as “low risk, moderate risk, serious risk and critical risk” by answering “Yes”, “Probably yes”, “Probably no”, “No” and “Not applicable or No information” to obtain the final quality of each domain. Disagreements between the two authors were resolved by the third author.

Data extraction

To ensure the accuracy of data extraction, two authors independently extracted the data, and then the data were carefully checked by one author. The underlying data extracted included the author of the research, year of publication, location, study design, total number of samples, number of therapies completed in the sample, mean age, gender of the patient, number of therapy periods, trauma-focused or not, military status or not, and specific interventions. There was also information extracted for subsequent meta-analysis on mean and standard deviation of PTSD symptoms, intimate relationship functioning, and depression level for patients and partners pre therapy, post therapy, and at different times of follow-up. If the mean or standard deviation between pre and post therapy or between pre therapy and 1-mouth, 3-mouth, and 6-month follow up was not directly provided in the research, it needed to be converted according to the corresponding sample size, standard error, and 95% confidence interval (CI) to obtain the required mean and standard deviation [13]. For example, Morland 2022 [9] calculated the mean and standard deviation after therapy and follow-up according to the mean and standard deviation of the pre therapy, the difference value and 95% CI between pre- and post- therapy or between pre-therapy and 1-mouth, 3-mouth, and 6-month follow up. Monson 2022 [14], Fredman 2020 [15], and Sautter 2015 [16] calculated the standard deviation according to the standard error and sample size, and Monson 2012 [17]] calculated the standard deviation with the use of 95% CI.

Statistical analysis

The single proportions rates [18] was calculated for the adherence rate, and other outcomes were analyzed between pre and post therapy, including PTSD symptoms, intimate relationship functioning and depression level. Due to the scales were different across studies in assessment of PTSD symptoms, intimate relationship functioning and depression level, effect sizes were standardized mean difference (SMD) and their 95% CI, as well as I² and P values. If I² ≤ 40%, the fixed effect model was set, and if I² > 40%, the random effect model was set. In addition, I² was able to quantify the statistics, with 0% to 25% as low heterogeneity, 25% to 50% as moderate heterogeneity, 50% to 75% as significant heterogeneity, and 75% to 100% as high heterogeneity. Heterogeneity was considered to be significant at a P value of less than 0.1. The R 4.1.1 was employed for all statistical analyses.

Subgroup analysis

Drawing on the shared characteristics across the included studies, it is evident that the majority of couples’ therapeutic interventions were centered around trauma-focused, military status and CBCT, all of which emerged as core focal points in the therapeutic frameworks examined. Therefore, the subgroups were set in patients’ and partners’ mean age (< 40 versus ≥ 40), patients’ gender (all-male versus mixed), number of sessions of therapy (< 10 versus ≥ 10), trauma-focused or not of patients’ therapy, military status or not of patients, and specific interventions in couples’ therapies (CBCT and non-CBCT) to further explore the influence of the characteristics of population and specific therapy methods between pre- and post- therapy, and determine the sources of heterogeneity in above outcomes.

Results

Study selection

The databases were preliminarily screened by search terms, and 16,771 articles were obtained. After removing repeated articles, 13,832 articles were obtained. Then, the titles and abstracts were simply screened according to the research design of this study, and 89 articles were obtained. After screening according to the full text information, 15 articles were finally confirmed. Finally, a total of 18 studies from 15 articles [9, 14–17, 19–28] were included (Fig. 1).

Fig. 1.

PRISMA flowchart

Basic characteristics of included studies

A total of 18 studies from 15 articles [9, 14–17, 19–28] were included in this research, all of which were RCT and uncontrolled trails. The sample size of couples involved in each study was a minimum of 6 and a Maximum of 113, and the sample size of those who adhered to completion was a minimum of 5 and a Maximum of 58. The mean age of participants in each study was roughly concentrated between 30 and 60 years. Most of the studies involved predominantly male patients, and most of the interventions used CBCT. Other subgroups including number of therapy sessions, trauma-focused or not and military sample or not were documented in Table 1.

Table 1.

Study characteristics of included studies

Study	Year	Locations	Study Design	Total Participants (Couples)	Completed Participants (Couples)	Mean Age, Patient/Partner (SD)	Male Patient, Patient/Partner (%)	Number of Sessions	Trauma-Focused or Not	Military Sample or Not	Specific Interventions
Crenshaw (HOPES-CBCT)	2023	Canada,USA	UT	27	17	48.3(9.8)/ 48.3(9.8)	unknown	5	Yes	Yes	CBCT
Pukay-Martin	2022	USA	UT	113	58	46.62(14.72)/ NA	93.10	15	Yes	Yes	CBCT
Morland (OB-bCBCT)	2022	USA	RCT	46	34	39.3(12.5)/ 39.3(12.5)	73.90	8	NA	Yes	OB-CBCT
Morland (HB-bCBCT)	2022	USA	RCT	46	32	42.4(12.4)/ 42.4(12.4)	87.00	8	NA	Yes	HB-CBCT
Morland (PFE)	2022	USA	RCT	45	31	43.9(15.2)/ 43.9(15.2)	82.20	8	NA	Yes	PFE
Monson	2022	Canada	UT	18	11	NA	NA	5	Yes	Yes	CBCT
Wagner	2021	NA	UT	6	6	18–47	66.70	15	NA	NA	CBCT plus MDMA
Whealin	2017	Pacific Island	UT	40	30	46.2/43.0	96.00	9	NA	Yes	NA
Fredman	2020	USA	UT	24	24	40.49(7.12)/ 38.70(8.18)	95.83	15	Yes	Yes	AIM-CBCT
Weissman	2018	USA	UT	15	7	43/43	NA	26–36	No	Yes	CBCT plus SAT
Schumm	2015	USA	UT	12	9	42.22(16.4)/ 39.33(12.64)	100	15	Yes	Yes	CBCT
Sautter (PFE)	2015	USA	RCT	28	21	33.71(7.01)/ 32.25(7.89)	96.40	12	No	Yes	PFE
Sautter (SAT)	2015	USA	RCT	29	22	32.55 (6.16)/ 32.17(7.68)	100	13	Yes	Yes	SAT
Pukay-Martin	2015	Canada	UT	7	6	45.86(6.09)/ 44.86(7.78)	42.90	15	No	No	CBCT
Sautter	2014	USA	UT	7	7	38.7(10.8)/ 35.4(9.8)	100	12	Yes	Yes	CBCT
Monson	2012	Canada, USA	RCT	20	14	40.4(11.3)/ 40.7(12.5)	35/50	15	Yes	No	CBCT
Monson	2011	USA	UT	6	5	41.7(13.1)/ 40.3(12.8)	50/83.33	15	Yes	No	CBCT
Sautter	2009	USA	UT	6	6	59.2/53.1	100	10	Yes	Yes	BFT

HOPES-CBCT helping overcome PTSD and enhance satisfaction of CBCT, AIM-CBCT abbreviated intensive, multi-couple group version of CBCT, BFT behavioral Family Therapy, CBCT Cognitive-Behavioral Conjoint Therapy, HB-bCBCT home-based brief cognitive-behavioral conjoint therapy, MDMA Methylenedioxymethamphetamine, NA not applicable, OB-bCBCT office-based brief cognitive-behavioral conjoint therapy, PFE PTSD family education, RCT randomized controlled trial, SAT structured approach therapy, SD standard deviation, UT uncontrolled trial

Quality assessment

The quality of all studies was evaluated according to the study design, and the final risk was obtained in Supplementary Table 1–2. However, Morland 2022 [9] and Sautter 2015 [16] had some concerns, Monson 2022 [14] and Fredman 2020 [15] had serious risk of bias due to missing data. What’s more, Wanger 2021 [20] used a combination intervention of CBCT and drugs, and Weissman 2018 [26] used a combination intervention of CBCT and SAT, which both had moderate risk of bias in classification of interventions.

Primary outcomes

Adherence rate

A total of 18 studies with 495 couples from 15 articles [9, 14–17, 19–28] reported the primary outcome of adherence rate. The pooling result of meta-analysis showed the adherence rate of couples’ therapies (Proportion = 69%, 95%CI: 64% to 73%) for PTSD with low heterogeneity (I² = 18.1%) in Fig. 2.

Fig. 2.

Forest plot of adherence rates for couples’ therapies

PTSD symptoms

Compared with before couples’ therapies, couples’ therapies strikingly reduced the patient's PTSD symptoms, and as measured by the scale, PTSD symptoms significance decreased (SMD = 0.79, 95%CI: -1.13 to -0.46), which was significant heterogeneity (I² = 67%). The reduction in PTSD symptoms after 1-month (SMD = -0.71, 95%CI: -1.1 to -0.24) was statistically significant improvement. Additionally, 3-mouth (SMD = -0.73, 95%CI: -1.65 to 0.19) and 6-month (SMD = -0.09, 95%CI: -0.39 to 0.21) of follow-up compared with before couples’ therapies was not statistically significant, and no data for partners at 6-month follow up (Table 2 and Fig. 3).

Table 2.

Comparisons between pre-therapy and different follow-up times in primary and secondary outcomes

Outcomes	Different Follow-Up Time	Patients					Partners
		N	Sample	ES, 95%CI	I2	P	N	Sample	ES, 95%CI	I2	P
Adherence rate	Pre-Post	18	340/495	69% (64%, 73%)	18.1%	0.24	0	0	NA	NA	NA
PTSD Symptoms	Pre-Post	14	250/399	− 0.79 (− 1.13, − 0.46)	67%	< 0.01	14	111/183	− 0.78 (− 1.05, − 0.52)	58%	0.03
	Pre-1 month FU	2	35/42	− 0.71 (− 1.18, − 0.24)	0%	0.58	2	35/42	− 0.36 (− 0.81, 0.1)	0%	0.50
	Pre-3 month FU	4	71/87	− 0.73 (− 1.65, 0.19)	84%	< 0.01	2	30/30	0.37 (− 2.05, 2.78)	90%	< 0.01
	Pre-6 month FU	3	64/137	− 0.09 (− 0.39, 0.21)	0%	0.99	0	0	NA	NA	NA
Intimate Relationship Functioning	Pre-Post	15	280/439	0.27 (0.12, 0.42)	0%	0.79	12	198/302	0.38 (0.20, 0.56)	0%	0.85
	Pre-1 month FU	2	35/42	− 0.02 (− 0.47, 0.43)	0%	0.63	2	35/42	0.27 (− 0.19, 0.72)	0%	0.88
	Pre-3 month FU	4	71/87	0.28 (− 0.04, 0.60)	14%	0.32	4	71/87	0.27 (− 0.05, 0.58)	0%	0.70
	Pre-6 month FU	4	70/143	0.05 (− 0.24, 0.34)	0%	0.64	1	6/6	0.42 (− 0.73, 1.57)	NA	NA
Depression Level	Pre-Post	8	147/241	− 0.61 (− 0.82, − 0.40)	0%	0.66	7	139/222	− 0.29 (− 0.50, − 0.07)	0%	0.80
	Pre-1 month FU	2	35/42	− 0.58 (− 1.05, − 0.12)	0%	0.53	1	24/24	− 0.01 (− 0.58, 0.55)	NA	NA
	Pre-3 month FU	3	65/81	− 0.62 (− 1.09, − 0.14)	50%	0.13	3	65/81	− 0.16 (− 0.49, 0.16)	0%	0.59

ES effect size, CI confidence interval, FU follow-up, N the number of included study, NA not applicable, PTSD post-traumatic stress disorder

Fig. 3.

Forest plot of PTSD symptoms for pre- and post-couples’ therapies. Note: A indicated comparing of PTSD symptoms pre- and post-couples’ therapies for patients, B indicated comparing of PTSD symptoms pre- and post-couples’ therapies for partners

For partners, couples’ therapies reduced the PTSD symptoms compared with before couples’ therapies, with a significance reduction (SMD = -0.78, 95%CI: -1.05 to -0.52) and significant heterogeneity (I² = 58%). After 1-month follow-up, the decrease of PTSD symptoms had not statistical significance (SMD = -0.36, 95%CI: -0.81 to 0.10) for partners compared with before couples’ therapies, and after 3-month follow-up, the increase of PTSD symptoms for partners also had not statistical significance (SMD = 0.37, 95%CI: -2.05 to 2.78) with high heterogeneity (I² = 90%) (Table 2 and Fig. 3).

Secondary outcomes

Intimate relationship functioning

There was statistically significant in improvement of intimate relationship functioning for both patients and partners (Fig. 4). Compared with before couples’ therapies, the intimate relationship functioning for patients had improvement after therapy (SMD = 0.27, 95%CI: 0.12 to 0.42), and 1-mouth (SMD = -0.02, 95%CI: -0.47 to 0.43), 3-mouth (SMD = 0.28, 95%CI: -0.04 to 0.60), and 6-month (SMD = 0.05, 95%CI: -0.24 to 0.34). For partners, the intimate relationship functioning had improvement after therapy (SMD = 0.38, 95%CI: 0.20 to 0.56) and 1-mouth (SMD = 0.27, 95%CI: -0.19 to 0.72), 3-mouth (SMD = 0.27, 95%CI: -0.05 to 0.58), and 6-month (SMD = 0.42, 95%CI: -0.73 to 1.57) (Table 2 and Fig. 4).

Fig. 4.

Forest plot of intimate relationship functioning for pre- and post-couples’ therapies. Note: A indicated comparing of intimate relationship functioning pre- and post-couples’ therapies for patients, B indicated comparing of intimate relationship functioning pre- and post-couples’ therapies for partners

Depression level

There was statistically significant in decrease of depression level for both patients and partners (Fig. 5). Compared with before couples’ therapies, the depression level decreased after therapy (SMD = -0.61, 95%CI: -0.82 to -0.40), 1-mouth (SMD = -0.58, 95%CI: -1.05 to -0.12) and 3-mouth (SMD = -0.62, 95%CI: -1.09 to -0.14); and for partners, after therapy (SMD = -0.29, 95%CI: -0.50 to -0.07), 1-mouth (SMD = -0.01, 95%CI: -0.58 to 0.55), 3-mouth (SMD = -0.16, 95%CI: -0.49 to 0.16) (Table 2 and Fig. 5).

Fig. 5.

Forest plot of depression level for pre- and post-couples’ therapies. Note: An indicated comparing of depression level pre- and post-couples’ therapies for patients, B indicated comparing of depression level pre- and post-couples’ therapies for partners

Subgroup analyses

According to the characteristics of the included studies, subgroup analyses by patients’ and partners’ age (< 40 and ≥ 40), patients’ sex (all-male and mixed), number of sessions of therapy (< 10 and ≥ 10), trauma-focused or not of patients’ therapy, military status or not of patients, and specific interventions in couples’ therapies (CBCT and non-CBCT) were performed for therapy adherence rate (Table 3), PTSD symptoms, intimate relationship functioning and depression level (Table 4).

Table 3.

Subgroup analyses of the adherence rates between pre- and post-couples’ therapies

Subgroup Analysis		N	Sample	Proportion, 95%CI	I2	P
Patients’ Mean Age	< 40	4	84/110	78% (69%, 85%)	59%	0.06
	≥ 40	12	239/361	66% (61%, 71%)	29%	0.16
Patients’ Sex	Male	4	44/54	92% (70%, 100%)	72%	0.01
	Mix	12	278/408	68% (63%, 72%)	34%	0.12
Number of Sessions	< 10	4	127/177	72% (65%, 78%)	0%	0.89
	≥ 10	11	178/258	87% (73%, 96%)	87%	< 0.01
Trauma-focused or Not	Yes	10	173/262	66% (60%, 72%)	7%	0.38
	No	3	34/50	69% (46%, 88%)	58%	0.09
Military Status or Not	Yes	14	309/456	68% (63%, 72%)	30%	0.14
	No	3	25/33	76% (60%, 89%)	0%	0.61
Specific Interventions	CBCT	15	282/416	68% (63%, 72%)	27%	0.16
	non-CBCT	3	58/79	83% (55%, 99%)	73%	0.02

CBCT cognitive-behavioral conjoint therapy, CI confidence interval, N the number of included study

Table 4.

Subgroup analyses of PTSD symptoms, intimate relationship functioning and depression level between pre- and post-couples’ therapies

Subgroup Analysis			Patients						Partners
			N	Sample	SMD, 95%CI	I2	P		N	Sample	SMD, 95%CI	I2	P
Patients’ and Partners’ Mean Age	PTSD Symptoms	< 40	4	77/110	-1.04 (-1.93, -0.16)	85%	< 0.01		1	9/12	-1.28 (-2.24, -0.31)	NA	NA
		≥ 40	9	162/271	-0.71 (-1.08, -0.35)	58%	0.01		4	33/40	-1.09 (-1.60, -0.57)	0%	0.47
	Intimate Relationship Functioning	< 40	4	77/110	0.21 (-0.08, 0.51)	19%	0.30		4	59/76	0.18 (-0.17, 0.52)	0%	0.65
		≥ 40	9	186/305	0.28 (0.10, 0.47)	0%	0.74		5	64/89	0.56 (0.23, 0.89)	0%	0.77
	Depression Level	< 40	2	43/57	-0.33 (-0.73, 0.07)	0%	0.36		4	59/76	-0.28 (-0.62, 0.07)	0%	0.78
		≥ 40	5	93/166	-0.75 (-1.02, -0.49)	0%	0.93		2	22/33	-0.16 (-1.08, 0.75)	49.9%	0.16
Patients’ Sex	PTSD Symptoms	Male	4	44/54	-1.48 (-1.95, -1.01)	34%	0.21		2	15/18	-1.33 (-2.11, -0.55)	0%	0.85
		Mix	8	188/312	-0.52 (-0.85, -0.19)	62%	0.01		4	85/147	-0.76 (-1.44, -0.08)	67%	0.03
	Intimate Relationship Functioning	Male	3	38/48	0.50 (0.07, 0.94)	0%	0.70		3	38/48	0.33 (-0.11, 0.76)	0%	0.81
		Mix	10	224/358	0.23 (0.06, 0.40)	0%	0.58		7	142/221	0.37 (0.16, 0.59)	0%	0.55
	Depression Level	Male	2	31/41	-0.56 (-1.03, -0.08)	0%	0.76		3	38/48	-0.34 (-0.77, 0.09)	0%	0.66
		Mix	4	98/167	-0.64 (-0.90, -0.39)	35%	0.20		4	101/174	-0.22 (-0.49, 0.05)	0%	0.48
Number of Sessions	PTSD Symptoms	< 10	3	73/120	-0.22 (-0.51, 0.08)	0%	0.41		0	0	NA	NA	NA
		≥ 10	9	150/229	-1.14 (-1.53, -0.74)	54%	0.03		6	100/165	-0.93 (-1.51, -0.34)	64%	0.02
	Intimate Relationship Functioning	< 10	4	112/177	0.13 (-0.10, 0.37)	10%	0.35		1	30/40	0.65 (0.16, 1.13)	NA	NA
		≥ 10	9	150/229	0.37 (0.16, 0.58)	0%	0.84		9	150/229	0.31 (0.10, 0.52)	0%	0.87
	Depression Level	< 10	1	21/28	-0.14 (-0.71, 0.43)	NA	NA		0	0	NA	NA	NA
		≥ 10	6	129/208	-0.62 (-0.85, -0.40)	0%	0.44		6	122/195	-0.26 (-0.49, -0.03)	0%	0.77
Trauma-focused or Not	PTSD Symptoms	Yes	8	134/212	-1.17 (-1.60, -0.73)	57%	0.02		6	105/176	-0.64 (-1.07, -0.21)	48%	0.09
		No	3	34/50	-0.71 (-1.17, -0.26)	0%	0.60		1	6/7	-2.03 (-3.47, -0.59)	NA	NA
	Intimate Relationship Functioning	Yes	7	128/206	0.41 (0.19, 0.64)	0%	0.92		7	128/206	0.35 (0.13, 0.58)	0%	0.98
		No	3	34/50	0.15 (-0.29, 0.59)	0%	0.72		3	34/50	0.23 (-0.21, 0.68)	36%	0.21
	Depression Level	Yes	6	119/198	-0.69 (-0.93, -0.45)	0%	0.86		6	118/194	-0.31 (-0.54, -0.08)	0%	0.72
		No	2	28/43	-0.27 (-0.75, 0.21)	0%	0.38		1	21/28	-0.16 (-0.73, 0.40)	NA	NA
Military Status or Not	PTSD Symptoms	Yes	11	223/365	-0.66 (-1.00, -0.32)	68%	< 0.01		4	84/149	-0.60 (-1.18, -0.01)	59%	0.06
		No	3	27/34	-1.37 (-1.94, -0.79)	0%	0.93		3	27/34	-1.02 (-1.58, -0.47)	11%	0.33
	Intimate Relationship Functioning	Yes	11	247/399	0.23 (0.07, 0.39)	0%	0.71		8	165/262	0.37 (0.17, 0.57)	0%	0.65
		No	3	27/34	0.59 (0.07, 1.11)	0%	0.91		3	27/34	0.38 (-0.13, 0.90)	0%	0.63
	Depression Level	Yes	6	128/215	-0.56 (-0.79, -0.34)	0%	0.66		6	134/216	-0.31 (-0.53, -0.09)	0%	0.91
		No	2	19/26	-1.00 (-1.63, -0.36)	0%	0.70		1	5/6	0.47 (-0.74, 1.68)	NA	NA
Specific Interventions	PTSD Symptoms	CBCT	11	193/320	-0.87 (-1.24, -0.49)	68%	< 0.01		6	105/177	-0.72 (-1.21, -0.22)	59%	0.03
		non-CBCT	3	57/79	-0.50 (-1.15, 0.16)	62%	0.07		1	6/6	-1.44 (-2.77, -0.10)	NA	NA
	Intimate Relationship Functioning	CBCT	13	229/366	0.33 (0.16, 0.50)	0%	0.85		11	177/274	0.44 (0.24, 0.63)	0%	0.97
		non-CBCT	2	51/73	0.01 (-0.35, 0.36)	0%	0.97		1	21/28	-0.08 (-0.64, 0.49)	NA	NA
	Depression Level	CBCT	7	126/213	-0.69 (-0.91, -0.46)	0%	0.93		6	118/194	-0.31 (-0.54, -0.08)	0%	0.72
		non-CBCT	1	21/28	-0.14 (-0.71, 0.43)	NA	NA		1	21/28	-0.16 (-0.73, 0.40)	NA	NA

CBCT cognitive-behavioral conjoint therapy, CI confidence interval, NA not applicable, N the number of included study, PTSD post-traumatic stress disorder, SMD standardized mean difference

Patients’ and partners’ mean age

The PTSD patients younger than 40 years old showed an higher adherence rate (Proportion = 78%, 95%CI: 69% to 85%) compared to those older than 40 years old (Proportion = 66%, 95%CI: 61% to 71%), as presented in Table 3. The PTSD symptoms for age < 40 patients (SMD = -1.04, 95%CI: -1.93 to 0.16) and age ≥ 40 patients (SMD = -0.71, 95%CI: -1.08 to -0.35) had statistical significance improvement. The intimate relationship functioning for age ≥ 40 patients (SMD = 0.28, 95%CI: 0.10 to 0.47) had statistical significance improvement but age < 40 patients no statistical significant. The depressive level for age ≥ 40 patients (SMD = -0.75, 95%CI: -1.02 to 0.49) had statistical significance improvement but age < 40 patients no statistical significant (Table 4).

For partners, the PTSD symptoms for age < 40 patients (SMD = -1.28, 95%CI: -2.24 to -0.31) and age ≥ 40 patients (SMD = -1.09, 95%CI: -1.60 to -0.57) had statistical significance improvement. The intimate relationship functioning for age ≥ 40 (SMD = 0.56, 95%CI: 0.23 to 0.89) had statistical significance improvement but age < 40 had no statistical significance. There was not statistical significance for partners in depression level (Table 4).

Patients’ sex

The subgroup analysis of all males’ patients showed an high adherence rate (Proportion = 92%, 95%CI: 70% to 100%), with significant heterogeneity (I² = 72%). In comparison, mixed-sex group demonstrated a lower adherence rate (Proportion = 68%, 95%CI: 63% to 72%), with moderate heterogeneity (I² = 34%) and indicating that sex was not a source of heterogeneity in Table 3. The PTSD symptoms for all males’ patients (SMD = -1.48, 95%CI: -1.95 to -1.01) and mixed-sex group (SMD = -0.52, 95%CI: -0.85 to -0.19) had statistical significance improvement. The intimate relationship functioning for all males’ patients (SMD = 0.50, 95%CI: 0.07 to 0.94) and mixed-sex group (SMD = 0.23, 95%CI: 0.06 to 0.40) had statistical significance improvement. The depression level for all males’ patients (SMD = -0.56, 95%CI: -1.03 to -0.08) and mixed-sex group (SMD = -0.64, 95%CI: -0.90 to -0.39) had statistical significance improvement (Table 4).

For partners, the PTSD symptoms for all males’ patients (SMD = -1.33, 95%CI: -2.11 to -0.55) and mixed-sex group (SMD = -0.76, 95%CI: -1.44 to -0.08) had statistical significance improvement. The intimate relationship functioning for mixed-sex group (SMD = 0.37, 95%CI: 0.16 to 0.59) had statistical significance improvement. There was not statistical significance for mixed-sex group in intimate relationship functioning and all males’ patients and mixed-sex group in depression level (Table 4).

Number of sessions of couples’ therapy

The couples with sessions ≥ 10 had a high adherence rate (Proportion = 87%, 95%CI: 73% to 96%), with high heterogeneity (I² = 87%). In contrast, the adherence rate of sessions < 10 was significantly lower (Proportion = 72%, 95%CI: 65% to 78%), suggesting that the difference in session count was not a source of heterogeneity, as indicated in Table 3. PTSD symptoms of patients with sessions ≥ 10 (SMD = -1.14, 95%CI: -1.53 to -0.74) had a prominent reduction, with significant heterogeneity (I² = 54%). It was not statistical significance in group of sessions < 10 (SMD = -0.22, 95%CI: -0.51 to 0.08), and number of sessions was not a source of heterogeneity for PTSD symptoms. The intimate relationship functioning for sessions ≥ 10 (SMD = 0.37, 95%CI: 0.16 to 0.58) had statistical significance improvement but sessions < 10 had no statistical significance. The depression level for sessions ≥ 10 (SMD = -0.62, 95%CI: -0.85 to -0.40) had statistical significance improvement but sessions < 10 had no statistical significance (Table 4).

For partners, the PTSD symptoms for sessions ≥ 10 (SMD = -0.93, 95%CI: -1.51 to -0.34) had statistical significance improvement. The intimate relationship functioning for sessions < 10 (SMD = 0.65, 95%CI: 0.16 to 1.13) and sessions ≥ 10 (SMD = 0.31, 95%CI: 0.10 to 0.52) had statistical significance improvement. The depression level for sessions ≥ 10 (SMD = -0.26, 95%CI: -0.49 to -0.03) had statistical significance improvement (Table 4).

Trauma-focused or not of patients’ therapies

Trauma-focused therapy had a high adherence rate (Proportion = 66%, 95%CI: 60% to 72%) with low heterogeneity (I² = 7%) for couples’ therapies, whereas non-trauma-focused therapy exhibited a comparable adherence rate (Proportion = 69%, 95%CI: 46% to 88%) but with significant heterogeneity (I² = 58%), which suggests that non-trauma-focused therapy was a notable source of heterogeneity, as presented in Table 3. The PTSD symptoms for trauma-focused therapy (SMD = -1.17, 95%CI: -1.60 to -0.73) and non-trauma-focused therapy (SMD = -0.71, 95%CI: -1.17 to -0.26) had statistical significance improvement. The intimate relationship functioning for trauma-focused therapy (SMD = 0.41, 95%CI: 0.19 to 0.64) had statistical significance improvement but non-trauma-focused therapy had no statistical significance. The depression level for trauma-focused therapy (SMD = -0.69, 95%CI: -0.93 to -0.45) had statistical significance improvement but non-trauma-focused therapy had no statistical significance.

For partners, the PTSD symptoms for trauma-focused therapy (SMD = -0.64, 95%CI: -1.07 to -0.21) and non-trauma-focused therapy (SMD = -2.03, 95%CI: -3.47 to -0.59) had statistical significance improvement. The intimate relationship functioning for trauma-focused therapy (SMD = 0.35, 95%CI: 0.13 to 0.58) had statistical significance improvement but non-trauma-focused therapy had no statistical significance. The depression level for trauma-focused therapy (SMD = -0.31, 95%CI: -0.54 to -0.08) had statistical significance improvement but non-trauma-focused therapy had no statistical significance (Table 4).

Military status or not of patients

The couples’ therapies demonstrated a high adherence rate among military patients (Proportion = 68%, 95%CI: 63% to 72%) with moderate heterogeneity (I² = 30%), as well as among non-military patients (Proportion = 76%, 95%CI: 60% to 89%) in Table 3. The PTSD symptoms for military patients (SMD = -0.66, 95%CI: -1.00 to -0.32) and non-military patients (SMD = -1.37, 95%CI: -1.94 to -0.79) had statistical significance improvement. The intimate relationship functioning for military patients (SMD = 0.23, 95%CI: 0.07 to 0.39) and non-military patients (SMD = 0.59, 95%CI: 0.07 to 1.11) had statistical significance improvement. The depression level for military patients (SMD = -0.56, 95%CI: -0.79 to -0.34) and non-military patients (SMD = -1.00, 95%CI: -1.63 to 0.36) had statistical significance improvement (Table 4).

For partners, the PTSD symptoms for military patients (SMD = -0.60, 95%CI: -1.18 to -0.01) had statistical significance improvement and non-military patients (SMD = -1.02, 95%CI: -1.58 to -0.47) had statistical significance improvement. The intimate relationship functioning for military patients (SMD = 0.37, 95%CI: 0.17 to 0.57) had statistical significance improvement but non-military patients had no statistical significance. The depression level for military patients (SMD = -0.31, 95%CI: -0.53 to -0.09) had statistical significance improvement but non-military patients had no statistical significance (Table 4).

Specific interventions in couples’ therapies

The couples with CBCT had a high adherence rate (Proportion = 68%, 95%CI: 63% to 72%) with moderate heterogeneity (I² = 27%), whereas non-CBCT showed a high adherence rate (Proportion = 83%, 95%CI: 55% to 99%) with significant heterogeneity (I² = 73%), indicating that it was not a source of heterogeneity in Table 3. The PTSD symptoms for CBCT (SMD = -0.87, 95%CI: -1.24 to -0.49) had statistical significance improvement but non-CBCT had no statistical significance. The intimate relationship functioning for CBCT (SMD = 0.33, 95%CI: 0.16 to 0.50) had statistical significance improvement but non-CBCT had no statistical significance. The depression level for CBCT (SMD = -0.69, 95%CI: -0.91 to -0.46) had statistical significance improvement but non-CBCT had no statistical significance (Table 4).

For partners, The PTSD symptoms CBCT (SMD = -0.72, 95%CI: -1.21 to -0.22) and non-military patients (SMD = -1.44, 95%CI: -2.77 to -0.10) had statistical significance improvement. The intimate relationship functioning for CBCT (SMD = 0.44, 95%CI: 0.24 to 0.63) had statistical significance improvement but non-CBCT had no statistical significance. The depression level for CBCT (SMD = -0.31, 95%CI: -0.73 to -0.40) had statistical significance improvement but non-CBCT had no statistical significance (Table 4).

Discussion

With advancements in both clinical and psychological research, the treatment of PTSD had significantly progressed. Previously, therapeutic approaches had been largely confined to either pharmacotherapy or psychological counseling alone. However, contemporary meta-analyses demonstrated that numerous evidence-based treatment options existed, supported by higher-quality research [29]. Empirical evidence strongly supported the efficacy of couples-based interventions in treating PTSD [30]. Patients demonstrated high adherence to couples-based therapy; however, comparative studies revealed no significant efficacy differences between individual and couples-based interventions for PTSD [29]. Consequently, treatment selection can be optimized for higher efficacy by considering population characteristics and individual needs. This meta-analysis of 18 clinical studies demonstrated that couples-based interventions for PTSD yielded multidimensional improvements for both patients and their partners.

Post-therapy assessments revealed significant reductions in PTSD symptoms among patients, potentially alleviating fear-related distress and facilitating the restoration of daily functioning and interpersonal trust. Notably, non-PTSD partners also exhibited symptom attenuation, though this did not indicate pre-existing PTSD pathology. Standardized measures demonstrated that couples-based interventions conferred therapeutic benefits for both parties, with particularly robust improvements in dyadic relationship functioning [31]. Longitudinal follow-up assessments revealed that both patients and partners generally maintained therapeutic gains throughout the post-intervention period. However, standardized scale scores revealed a progressive worsening of PTSD symptoms among partners during extended follow-up periods. This deterioration may reflect either secondary traumatization resulting from exposure to patients' severe symptom, or limited statistical power due to the small sample size [32].

The intervention significantly improved dyadic relationship functioning. Shared participation in PTSD treatment enhanced cognitive alignment between partners, facilitating the development of more stable and supportive communication patterns. These improvements strengthened emotional connectivity and interpersonal trust, ultimately fostering more cooperative marital dynamics and positive family functioning. Furthermore, comorbid depressive symptoms in PTSD patients showed significant improvement post-intervention [33]. The dyadic treatment format facilitated mutual emotional disclosure and reciprocal support, enabling partners to collaboratively address PTSD-related challenges. The enhanced emotional communication demonstrated significant improvements in both dyadic relationship functioning and depressive symptoms, while also promoting more positive affect regulation [34]. However, longitudinal assessments revealed partial recurrence of symptoms, with relationship functioning and depression scores showing clinically meaningful rebound at 6-month follow-up.

This study was designed to account for objective factors that might influence therapeutic outcomes [35]. Some of the results showed high heterogeneity. Therefore, this study also conducted subgroup analyses to explore the sources of heterogeneity. Population characteristics and specific therapeutic interventions exhibited variations in adherence rates, PTSD symptoms, intimate relationship functioning, and depression levels, as measured before and after treatment. The results of some subgroup analyses still showed high heterogeneity, which implies that the specific subgroup in question may not constitute the source of such heterogeneity. In this study, the possible reasons might be that there were too few studies included in this subgroup. On the other hand, in clinical treatment, there exist discrepancies among physicians in the selection of assessment scales for PTSD symptoms, intimate relationship functioning, and depression levels. Nevertheless, the majority of the results remain statistically significant, thereby holding considerable reference value for clinical interventions.

Age was a significant factor [36], as younger adults exhibited greater receptiveness to novel therapies and higher adherence rates during recovery. In contrast, older individuals experienced increased psychological distress and physical limitations, which May have reduced therapeutic effectiveness and resulted in less pronounced reductions in PTSD symptoms compared to those under 40 years of age [37]. However, older patients and their partners demonstrated greater improvements in intimate relationship functioning. This advantage likely stemmed from their accumulated shared experiences [38], which fostered greater mutual comfort and support. Gender also emerged as a significant factor, as males and females exhibited inherent differences in trauma responses and therapeutic outcomes. These differences in psychological processing and coping mechanisms suggested distinct post-trauma needs and treatment responses between genders. For instance, male soldiers developed PTSD from combat trauma and required targeted interventions to facilitate reintegration into civilian life, while females often experienced psychological distress from different trauma types, such as interpersonal violence [39]. Consequently, male patients demonstrated higher therapy adherence and greater improvements in marital relationships compared to their female counterparts. However, the differential improvement in PTSD symptoms and depression levels had indicated that therapy for unaffected male partners was less effective in addressing core psychopathology than direct patient treatment. Additionally, military status significantly influenced outcomes. Given their unique occupational demands, soldiers frequently experience extreme stress and trauma during duty [40], increasing PTSD susceptibility and necessitating specialized interventions tailored to their professional context. Military personnel demonstrated higher engagement and completion rates in couples' therapy [41], along with greater treatment motivation, resulting in superior therapeutic outcomes.

A longer duration of therapeutic intervention enabled more comprehensive and profound patient recovery. The results demonstrated that interventions exceeding 10 sessions yielded significantly greater improvements in PTSD symptoms, relationship functioning, and depression levels for both patients and their partners. Furthermore, trauma-focused interventions directly addressed the underlying pathology, facilitating gradual resolution of negative emotional states and traumatic memories. Consequently, this approach demonstrated superior efficacy for PTSD treatment compared to non-trauma-focused interventions [42]. Various treatment modalities existed for PTSD, including pharmacotherapy and cognitive behavioral therapy. However, most studies in this analysis employed CBCT. This integrative approach combined behavioral and cognitive therapeutic principles to modify maladaptive thought patterns and behaviors associated with PTSD. Specifically, CBCT facilitated cognitive restructuring of traumatic experiences while promoting healthier emotional regulation and coping strategies [43]. Although CBCT implementation presented greater complexity than alternative interventions (e.g., PTSD family education or behavioral family therapy), potentially affecting treatment adherence, it demonstrated superior clinical efficacy.

Clinical significance

As psychotherapy advanced, PTSD patients gained access to increasingly tailored treatment options [44]. This study demonstrated that couples-based interventions could significantly enhance both psychological well-being and relationship functioning for patients and their partners. Furthermore, we systematically analyzed moderating factors of treatment efficacy, with detailed results presented above. These findings offered novel insights for developing future interventions that incorporated family system dynamics and spousal support mechanisms. Specifically, the research has revealed the significant impact of family internal interaction patterns on an individual's mental health, laying a solid theoretical foundation for family-centered intervention strategies. By systematically analyzing the emotional connections in couples’ therapies, the researchers further clarified the crucial role of spouse support in improving the condition and enhancing the intimacy of the relationship. For example, for different gender, occupation groups, the researchers can adopt different treatment models to intervene in PTSD and strengthen mutual support between spouses.

Advantages and limitations

This study offered several methodological and substantive contributions to the literature. On the one hand, it employed meta analysis considered the highest level of evidence—to evaluate couples-based therapies for PTSD, yielding robust and reliable findings. On the other hand, through novel subgroup analyses, it examined treatment effectiveness in relation to patient characteristics and therapeutic approaches, an aspect previously unexplored. Notably, our analysis revealed significant reductions in PTSD symptoms among patients under 40 years old, a finding not previously documented in the literature.

This study had several limitations in addition to its significant findings. First, the included studies exhibited varying degrees of missing partner data and differing follow-up periods, potentially compromising result reliability due to insufficient sample sizes. Second, required data transformations in some studies, along with one study's combined psychotherapy-pharmacotherapy intervention, may have influenced data accuracy. Finally, reliance on clinician-administered scale assessments introduced potential variability based on regional differences and clinicians' evaluation competencies.

Conclusion

Couples' therapies demonstrated significant efficacy in ameliorating PTSD symptoms, enhancing relationship functioning, and reducing depression levels. Furthermore, analysis revealed higher treatment adherence among Male participants and more pronounced symptom improvement in patients under 40 years old, male participants, those receiving extended therapy sessions, and those undergoing trauma-focused or CBCT interventions. This study not only addressed a gap in PTSD couples therapy research but also provided comprehensive empirical support for treatment strategies while offering novel insights for developing population-specific therapeutic approaches based on demographic and methodological characteristics.

Abbreviations

AIM-CBCT

Abbreviated, intensive, multi-couple group version of CBCT

CBCT

Cognitive-behavioral conjoint therapy

CI

Confidence interval

PTSD

Post-traumatic stress disorder

RCT

Randomized controlled trials

ROBINS-I

Risk Of Bias In Non-randomised Studies of Interventions

SMD

Standardized mean difference

Authors’ contributions

Chao Zhang and Ling Yang conceptualized and coordinated the study. Yu-Ting Tao and Chao Zhang developed the search strategy. Yu-Ting Tao, Jin Xie, Hua-Ling Jiang, and Teng-Yu Gao screened citations and assessed studies for eligibility. Jin Xie and Yu-Ting Tao extracted data. Yu-Ting Tao and Jin Xie performed quality assessments. Yu-Ting Tao, Hua-Ling Jiang, Ling Yang and Xin Liu performed statistical analyses. Yu-Ting Tao, Ling Yang and Chao Zhang provided methodologic expertise in knowledge synthesis and resolved disagreements regarding study eligibility or quality assessments. Yu-Ting Tao, Jin Xie, Chao Zhang and Ling Yang critically reviewed the manuscript for important intellectual content. All o the authors gave final approval of the version to be published and agreed to be accountable for all aspects of the work.

Funding

This study was supported by the Natural Science Foundation project of Hubei Province in 2021 (No. 2024AFD284).

Data availability

The data are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.