Dyadic coping, relationship quality, and psychosocial adaption between couples facing prostate cancer: A longitudinal cross-lagged study

Abstract

Objective

To explore the associations among dyadic coping, close relationship, and psychosocial adaption in prostate cancer couples during the first six months after diagnosis.

Methods

Newly diagnosed prostate cancer patients and their spouses were recruited from a tertiary uro-oncology center in China. Participants were investigated with three surveys including Dyadic Coping Inventory, Marriage Adjustment Test, and Psychosocial Adjustment to Illness Scale at diagnosis (T0), one month (T1), three months (T2), and six months (T3) after diagnosis.

Results

A total of 243 couples were involved. Both patients and their spouses reported the lowest levels of dyadic coping and psychosocial adaptation at T1, followed by a gradual improvement through T3. Relationship quality declined until T2 and then partially recovered. Better dyadic coping and relationship quality in both partners were associated with better psychosocial adaptation across all time points (r = −0.13 to −0.38, all P < 0.01). However, higher psychosocial adaptation in spouses was associated with poorer adaptation in patients. Cross-lagged analyses revealed time-dependent partner effects, with spouses' dyadic coping showing consistent prospective associations with patients' psychosocial adaptation from T1 to T3 (β = −0.208 to −0.185, all P < 0.01), whereas patients' dyadic coping had weaker and later effects on spouses’ adaptation (β = −0.046, P = 0.022). No significant longitudinal effects were observed between relationship quality and psychosocial adaptation.

Conclusions

Dyadic coping, relationship quality, and psychosocial adaptation are closely interrelated in couples coping with prostate cancer, with the first month after diagnosis representing a period of heightened psychological distress. Our findings underscore the complexity and time-sensitive nature of dyadic processes and underscore the importance of couple-level psychosocial support.

Introduction

Prostate cancer is one of the most prevalent malignancies affecting men worldwide. A cancer diagnosis imposes profound psychosocial burdens on both patients and their spouses beyond its physical impact.¹ How couples cope with physical, emotional, relational and social impacts of prostate cancer was identified as the process of psychosocial.² It was a dynamic process for couples to adjust to the diagnosis and manage disease-related events.

Challenges for prostate cancer like sexual dysfunction were inevitable, disrupting patients' normal life. As a result, many patients suffer from high levels of psychological distress, unmet psychological and social needs, and a reduced quality of life.³ Despite continuous treatment, patients’ psychological distress often increases, further compromising their overall well-being due to changes in body image and perceptions of masculinity. Spouses played an important role in decision making and cancer caring, and inevitably influenced by patients.^1,4 Compared to male patients, female spouses exhibit emotional sensitivity, and are more likely to experience a heavier and prolonged emotional and psychological burden.⁵ Female spouses suppress their own psychological distress to support patients, resulting in a lower quality of life compared to the patients themselves.^6,7

The interaction between prostate cancer couples conceptualized prostate cancer as ‘we disease’.⁸ From a systemic transactional perspective,⁹ stress like cancer experienced by one spouse spills over and crosses over to the other, triggering the possibility for dyadic coping. Dyadic coping refers to the collaborative strategies for couples to manage stress associated with a chronic illness like prostate cancer. It can be divided into positive (e.g., problem-focused or emotional-focused coping) and negative (e.g., ambivalent or hostile responses) or divided into 5 dimensions in accordance with dyadic coping inventory.¹⁰ Empirical evidence revealed the divergent psychological impacts of dyadic coping on both patients and their spouse, whereas the underlying mechanism is still unclear. For instance, Götze¹¹ and Osin¹² found that dyadic coping was associated with reduced depressive symptoms in both cancer patients and their spouses. However, previous studies have reported that dyadic coping was associated with improved psychosocial adaptation in patients, while exerting negative effects on their spouses.^13,14

Similarly, Sun Haiyan et al.¹⁵ identified interactive effects between dyadic coping and psychological resilience in both patients and spouses, suggesting that coping dynamics are shaped by individual and relational factors. Relationship in prostate cancer couples, indexed by marital satisfaction, plays a key role in psychosocial adaptation. The Relationship Intimacy Model¹⁶ suggested that behaviors or coping strategies in close relationship was associate with psychosocial adaption. In this model, intimacy referred not only sexual activities but also communication and non-verbal expression of concern and attention like hugging and physical closeness. Couples who confronted prostate cancer together other than individually predicted higher relationship satisfactory, which indicated the dyadic coping as the protective buffer against cancer diagnosis.¹⁷ Furthermore, one meta-analysis highlighted the close relationship between spouses as a critical supportive resource, assisting individuals to navigate the emotional and practical challenges of living with cancer.¹⁸

Regan et al.¹⁹ reported that regardless of the specific coping strategies, dyadic coping exerted a positive effect on relationship quality in prostate cancer couples; while another literature review by Professor Chen⁷ suggested that positive dyadic coping may be a double-edged sword, which meant it can enhance relationship satisfaction with additional increase of psychosocial burden on one or both spouses. Interestingly, delegated dyadic coping, where one spouse takes over certain responsibilities, was found to relieve the spouse's emotional burden, yet simultaneously undermine the patient's self-esteem, leading to increased depressive symptoms. Longitudinal studies have also highlighted these nuanced and sometimes contradictory outcomes. Two longitudinal studies published in 2014²⁰ and 2021²¹ examined psychosocial adaptation within and between couples coping with prostate cancer over time. The earlier study highlighted interactional effects from spouses on patients, whereas the later study showed that higher levels of dyadic communication were associated with better psychosocial adaptation in both patients with prostate cancer and their spouses.

These findings underscore the complex and nuanced role of dyadic coping, close relationship, and psychosocial adaption in the context of prostate cancer. No definite literature established the moderator and mediate role of dyadic coping and relationship satisfaction on psychosocial adaption in prostate cancer patients. Given the complex, bidirectional, and evolving nature of three elements, a cross-lagged panel model was employed in this study to exam the reciprocal influences between patients and spouses over time, exploring how one spouse's coping strategies or intimacy influenced the other's psychosocial outcomes at later time points. Considering inconsistent or even contradictory effects across individuals and time, this longitudinal approach examine how changes in one domain may precede or follow changes in psychosocial adaption for both patients and their spouses. Above outcomes might be essential for developing targeted psychosocial interventions that promote sustainable adaptation and well-being.

Methods

Design, study setting, and participants

A longitudinal study design was employed. A total of 271 couples were consecutively recruited between January and June 2024 at one of the largest uro-oncological centers in Shanghai, China.

The inclusion criteria were: i) adult patients with newly pathologically confirmed prostate cancer within one week of diagnosis; ii) a spouse who lived with the patient as a couple and undertook most daily caregiving responsibilities; and iii) neither the patient nor the spouse had a history of mental illness, accompanying malignancies, or severe cardiovascular or cerebrovascular diseases. Of the 271 eligible couples, 253 couples agreed to participate and provided written informed consent.

The exclusion criteria were: i) Secondary prostate involvement from non-prostatic primary malignancies (e.g., bladder cancer); ii) either spouse being in a remarried relationship; and iii) either spouse participating in other similar studies. Remarried couples were excluded to minimize heterogeneity in relationship history and family structure, as prior marital experiences may influence dyadic coping patterns, relationship quality, and caregiving dynamics. During the follow-up period, 10 couples withdrew successively due to refusal of telephone follow-up, inconvenience, or other health-related issues.

Participating couples were asked to complete three standardized questionnaires at four time points: baseline (T0, defined as within one week after diagnosis), one month after baseline (T1), three months after baseline (T2), and six months after baseline (T3). Baseline sociodemographic and disease-related information was collected at T0. Patients were approached during their first outpatient visit following prostate biopsy, provided with written information about the study, and enrolled after informed consent was obtained. Patients and their spouses were instructed to complete the questionnaires independently at each assessment to minimize mutual influence and better capture dyadic interaction effects. Completion of the questionnaires assessing dyadic coping, close relationship, and psychosocial adaptation required approximately 20–40 minutes.

Follow-up assessments were aligned with routine clinical care. T1 was defined as one month after diagnosis, a time point at which most patients were either hospitalized for surgical treatment or attending the hospital for the first course of neoadjuvant therapy; therefore, the majority of T1 data were collected in person during inpatient stays or outpatient visits. T2 and T3 assessments coincided with scheduled clinical follow-up visits for postoperative evaluation or ongoing treatment monitoring, and data were therefore predominantly collected face-to-face. Remote follow-up, including telephone interviews or completion of questionnaires via a secure online platform, was used only for participants who were unable to return to the hospital due to long travel distance or other practical constraints. All in-person or remote assessments were administered by trained research nurses using standardized instructions. Completed questionnaires were reviewed for completeness and internal consistency to ensure data quality.

Measures

Sociodemographic information

Baseline data included patients' sociodemographic characteristics and disease-related information. Sociodemographic variables covered the age, employment status, education level, and history of chronic diseases of both the patients’ and their spouses. Prostate cancer related variables included prostate-specific antigen (PSA) levels, tumor stage, history of prostate biopsy, family history of uro-oncological history, and current treatment modalities.

Dyadic coping inventory

The Dyadic coping inventory (DCI) developed by Professor Guy Bodenmann in 2008 explored how couples cope with stress together, focusing on the interactional dynamics within the close relationships.⁷ The DCI consists of 37 items with 6 dimensions including stress communication (8 items), supportive dyadic coping (10 items), delegated dyadic coping (4 items), negative dyadic coping (8 items), common dyadic coping (5 items), and evaluation of couple's dyadic coping (2 items).²² It is a self-reported assessment tool by a 5-point Likert scale (1 = never/very rarely, 5 = very often). The DCI yields a total score based on the sum of the first 35 items, with scores ranging from 35 to 175.²³ Higher scores reflects more effective dyadic coping within the couple. A total score exceeding 110 indicated a relatively high level of dyadic coping within the couples. The scale was culturally adapted for Chinese populations by Xu Feng et al.²⁴ in 2016 using a sample of 474 Chinese couples with acceptable internal consistency ranging from 0.51 to 0.80. It is mostly widely used instrument to assess dyadic coping in China now, even used in cancer patients and their spouses.⁷

Marriage Adjustment Test, MAT

The Marriage Adjustment Test (MAT), also known as the Locke-Wallace Marital Adjustment Test, developed by Harvey Locke and Karl Wallace in 1959, was designed to assess the quality of relationship.²⁵ The MAT consists of 15 items, with higher total scores indicating greater marital satisfaction and adjustment.²⁶ The MAT was applicable in Chinese population after cultural adaption with reliability and validity.²⁷

Psychosocial adjustment to Illness Scale, PAIS-SR

PAIS-SR developed by Professor Derogatis LR in 1986,²⁸ was used to assess how to adapt to the psychological and social demands imposed by chronic illnesses. This scale was introduced into China, and culturally adapted by Professor Yao Jingjing²⁹ in 2013 with adequate internal reliability (Cronbach's alpha 0.872) and validation. It is a 44 items with 7 dimensions including health orientation (7 items), vocational environment (6 items), domestic environment (7 items), sexual relationships (6 items), extended family relationships (5 items), social environmental (6 items), and psychological distress (7 items). Each item is rated on a 4-point scale (0 through 3) of adjustment, and higher PAIS-SR scores generally indicate poorer psychosocial adjustment to illness.

Statistical analysis

All data were independently entered by two trained researchers into Microsoft Excel to ensure data completeness and accuracy. Statistical analyses were conducted using Mplus (version 7) for cross-lagged panel modeling and GraphPad Prism 8.0 for data visualization.

Descriptive statistics were used to summarize sociodemographic and clinical characteristics. Continuous variables were assessed for normality using the Shapiro–Wilk test. Variables that met normality assumptions or showed only minor deviations from normality (i.e., acceptable skewness and kurtosis values) were summarized as means and standard deviations (SDs). Longitudinal changes in DCI, MAT, and PAIS-SR across time points were examined using one-way repeated-measures analysis of variance (ANOVA). When significant main effects were observed, Bonferroni-corrected pairwise comparisons were performed to assess differences between time points (T0, T1, T2, and T3). Pearson's correlation coefficient (r) was used to evaluate the associations among dyadic coping, psychosocial adaptation, and relationship quality at each time point.

For descriptive statistics, repeated-measures ANOVAs, and correlation analyses, analyses were conducted using complete cases available at the relevant time points. Given the small proportion of missing data, complete-case analysis was considered appropriate for these exploratory comparisons.

To explore the temporal and directional relationships among these three variables, cross-lagged panel models (CLPMs) was constructed using Mplus.³⁰ The CLPMs were specified to test main effects only; interaction (moderation) effects were not included, as the primary objective was to examine temporal directional associations rather than conditional effects. Each model included autoregressive paths for stability over time, cross-lagged actor effects, and cross-lagged partner effects to capture interpersonal influences within patient–spouse dyads. To account for dyadic interdependence, residual correlations between patients and spouses on the same construct at each time point were freely estimated.

Initial models specified all theoretically plausible autoregressive and cross-lagged paths. Model fit was evaluated using the chi-square statistic (χ²), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). Given acceptable model fit and theoretical coherence, no extensive post hoc modifications were undertaken, and the final models retained the hypothesized structure.

Although a fully integrated model including all three constructs for both patients and spouses across four time points (i.e., 24 observed variables) would be theoretically ideal, such a model would require a substantially larger sample size to ensure stable parameter estimation. Considering the modest sample size and model complexity, a stepwise modeling strategy was adopted, with separate CLPMs constructed for theoretically relevant variable pairs to balance interpretability, statistical power, and estimation stability.

Missing data in the CLPM analyses were handled using full information maximum likelihood (FIML) estimation, which allows inclusion of all available data and provides unbiased parameter estimates under the assumption of missing at random. A two-tailed P-value of < 0.05 was considered statistically significant.

Although an a priori power analysis was not performed due to the exploratory nature of this longitudinal study, the adequacy of the sample size was evaluated post hoc. With a final sample of 243 couples (486 individuals), the study exceeded commonly recommended sample sizes for CLPM with four time points, which typically require at least 200 participants to achieve stable parameter estimation and acceptable statistical power.³¹

Simulation and methodological studies have shown that sample sizes in the range of 200–400 are generally sufficient to detect small-to-moderate standardized path coefficients (β≈0.20-0.30) in CLPM with acceptable bias and solution stability.31, 32, 33 Based on the observed effect sizes and the number of estimated parameters in the present models, the sample size was considered adequate to support the primary longitudinal analyses.

Results

Participants characteristics

Of the 271 couples initially screened, 253 couples (253 patients and 253 spouses) met the inclusion criteria and consented to participate at baseline (T0). 10 couples withdrew due to refusal of follow-up, inconvenience, or health-related reasons during the follow-up period. Consequently, 243 couples (243 patients and 243 spouses) completed all follow-up assessments and were included in the final statistical analyses. The mean age of patients and their spouses was 70.08 ± 0.50 and 68.43 ± 0.50 years respectively. At the time of diagnosis, 84.77% of patients were retired, and 81.07% of couples lived together without co-residing children. Detailed sociodemographic and clinical characteristics are presented in Table 1.

Table 1.

Couples’ general information (N = 243).

Variables	Patients, n (%)	Spouses, n (%)
Age (years, Mean ± SD)	70.08 ± 0.50	68.43 ± 0.50
Area of birth
Shanghai	156 (64.20)	124 (51.03)
Other cities	87 (35.80)	119 (48.97)
Employment
At work	37 (15.23)	33 (13.58)
Out of work	206 (84.77)	210 (86.42)
Chronic diseases
Hypertension	101 (41.56)	75 (30.86)
Diabetes mellitus	74 (30.45)	36 (14.81)
Heart disease	45 (18.52)	33 (13.58)
Other tumors	38 (15.64)
Comorbidity of 3 diseases	36 (14.81)	15 (6.17)
None	21 (8.64)	28 (11.52)
Medical insurance
Medical insurance	157 (64.61)
Self-pay	86 (35.39)
Number of child birth
None	13 (5.35)
1	146 (60.08)
2	75 (30.86)
3 and more	9 (3.70)
Living with children
Yes	46 (18.93)
No	197 (81.07)

Trajectory of dyadic coping, close relationship, and psychosocial adaption

Dyadic coping scores decreased from T0 (81.95 ± 18.95, 82.25 ± 18.07) to T1 (79.08 ± 16.62, 78.32 ± 16.29), and then increased slowly from T1 to T3 (87.33 ± 16.68, 85.90 ± 16.59) in both patients and their spouses with significantly difference over time (P < 0.001). A similar temporal trend was observed in psychosocial adaption, characterized by a decline from T0 (77.21 ± 7.60, 75.26 ± 7.36) to T1 (81.90 ± 6.59, 80.26 ± 6.64) followed by a gradual improvement from T1 to T3 (75.94 ± 6.60, 76.09 ± 5.89). However, as higher scores on the psychosocial adaptation scale indicate poorer adaptation, these results suggest that psychosocial adaptation initially worsened after diagnosis and subsequently improved, reaching a level lower than baseline by T3. The close relationship declined from T0 (103.40 ± 13.91, 102.58 ± 12.89) to T2 (98.58 ± 14.98, 98.98 ± 13.33) among couples, and demonstrated a gradual recovery thereafter (P < 0.001). Given the relatively large sample size and repeated-measures design, small P-values were expected for longitudinal comparisons. Therefore, results were interpreted with attention to effect sizes and clinical relevance rather than statistical significance alone. More details are presented in Table 2.

Table 2.

The trajectory of dyadic coping, close relationship, and psychosocial adaption.

	Dimensions	T0	T1	T2	T3	F	Pvalue
Patients	Dyadic coping inventory	81.95 ± 18.95	79.08 ± 16.62a	83.73 ± 16.23ab	87.33 ± 16.68abc	43.99	< 0.001
Spouses		82.25 ± 18.07	78.32 ± 16.29a	81.53 ± 16.61b	85.90 ± 16.59abc	35.69	< 0.001
Patients	Close relationship	103.40 ± 13.91	101.80 ± 13.22a	98.58 ± 14.98ab	100.86 ± 14.12ac	13.259	< 0.001
Spouses		102.58 ± 12.89	100.84 ± 13.44a	98.98 ± 13.33ab	102.14 ± 14.23c	7.719	< 0.001
Patients	Psychosocial adaption	77.21 ± 7.60	81.90 ± 6.59a	79.65 ± 6.64ab	75.94 ± 6.60abc	75.755	< 0.001
Spouses		75.26 ± 7.36	80.26 ± 6.64a	79.35 ± 6.27ab	76.09 ± 5.89bc	72.544	< 0.001

T0, baseline; T1, 1 month; T2, 3 months; T3, 6 months. ^a indicates that the difference compared with T0 is statistically significant. ^b indicates that the difference compared with T1 is statistically significant. ^c indicates that the difference compared with T2 is statistically significant.

P < 0.05 was considered statistically significant. ∗∗P < 0.01; ∗P < 0.05.

Time-point correlation analysis of dyadic coping, close relationship, and psychosocial adaption in prostate cancer couples

At T0 (Supplementary Table S1), patients' total psychosocial adaption scores were negatively correlated with their own dyadic coping (r = −0.287, P < 0.01) and close relationship (r = −0.130, P < 0.01), as well as with their spouses' dyadic coping (r = −0.236, P < 0.01). That was, better dyadic coping and close relationship in patients and dyadic coping in their spouses led to a higher level of psychosocial adaption in patients. The close relationship in spouses didn't relate to patients' adaption (P > 0.05). However, the better the spouses' psychosocial adaption was, the worse the patients' adaption was (r = 0.423, P < 0.01). In addition, the better of the couples' dyadic coping (r = −0.383 and r = −0.326 respectively, P < 0.01)and close relationship (r = −0.185 and r = −0.157 respectively, P < 0.01), the better psychosocial adaption was found in spouses. Similar correlation were observed across T1 to T3 with no substantial changes in the direction or strength of associations. Full correlation coefficients for all time points are presented in Supplementary Table S1.

A cross-lagged model of patients' psychosocial adaption and spouses’ dyadic coping

The final modified model showed χ²/df = 145.793, RMSEA = 0.197, CFI = 0.856, indicating a good model fit. Results demonstrated that patients' scores of psychosocial adaption at T0 didn't significantly predict their own psychosocial adaption at T1 (β = −0.006, P = 0.911). However, patients' psychosocial adaption at T1 was significantly associated with their scores at T2 (β = 0.482, P = 0.001) and a similar association was between T2 and T3 (β = 0.528, P = 0.001). Spouses' dyadic coping at T0 predicted a higher score of patients' psychosocial adaption (β = 0.458, P = 0.001), which meant spouses' dyadic coping at T0 predicted a better level of patients' psychosocial adaption. However, spouses' effects of dyadic coping on psychosocial adaption showed an opposite direction from T1 to T2 (β = −0.185, P = 0.001) and from T2 to T3 (β = −0.208, P = 0.001), which meant a better dyadic coping in spouses indicated a worse patients' psychosocial adaption. Patients' psychosocial adaptation scores showed negatively significant continuity from T0 to T1, T1 to T2 and from T2 to T3 on spouses' dyadic coping, which meant that a worse level of patients' psychosocial adaption led to a worse dyadic coping in spouses. More details were illustrated in Fig. 1.

Fig. 1.

A cross-lagged model of spouses' dyadic coping and patients' psychosocial adaption. P, patient; S, spouse; T0, baseline; T1, 1 month; T2, 3 months; T3, 6 months. P < 0.05 was considered statistically significant (red arrow indicates an interactional effect on patients' psychosocial adaption). ∗∗P < 0.01; ∗P < 0.05.

A cross-lagged model of spouses' close relationship and patients’ psychosocial adaption

The final modified model showed χ²/df = 153.801, RMSEA = 0.203, CFI = 0.806, indicating a good model fit. Results showed that a higher score of spouses' MAT at T0 related to a higher score of patients' psychosocial adaption (β = 0.458, P = 0.001) at T1, which meant a better close relationship in spouses indicated a worse patients' psychosocial adaption. However, no significant difference were observed from T1 to T2 (β = −0.016, P = 0.774) and T2 to T3 (β = −0.018, P = 0.734). On the other hand, patients' psychosocial adaption at T0 indicated spouses’ close relationship (β = −0.343, P = 0.001), with no significant difference at other 2 time points (β = −0.057, P = 0.196 and β = −0.028, P = 0.576 respectively). More details were illustrated in Fig. 2.

Fig. 2.

A cross-lagged model of spouses' close relationship and patients' psychosocial adaption. P, patient; S, spouse; T0, baseline; T1, 1 month; T2, 3 months; T3, 6 months. P < 0.05 was considered statistically significant (red arrow indicates an interactional effect on patients' psychosocial adaption). ∗∗P < 0.01; ∗P < 0.05.

A cross-lagged model of patients' dyadic coping and spouses’ psychosocial adaption

The final modified model showed χ²/df = 230.561, RMSEA = 0.252, CFI = 0.811, indicating a good model fit. No significant association were observed between patients' dyadic coping and spouses' psychosocial adaption from T0 to T1 and T1 to T2 (β = 0.010, P = 0.133 and β = 0.004, P = 0.857 respectively), while patients' dyadic coping at T2 predicted better spouses' psychosocial adaption at T3 (β = −0.046, P = 0.022). On the other hand, the higher score of spouses' psychosocial adaption at T0 indicated a higher score of dyadic coping of patients at T1, which meant a worse level of spouses’ psychosocial adaption at T0 was associated with a better dyadic coping for patients at T1 (β = 0.820, P = 0.001). No significant association were observed at other time points. More details were illustrated in Fig. 3.

Fig. 3.

A cross-lagged model of patients' dyadic coping and spouses' psychosocial adaption. P, patient; S, spouse; T0, baseline; T1, 1 month; T2, 3 months; T3, 6 months. P < 0.05 was considered statistically significant (red arrow indicates an interactional effect on spouses' psychosocial adaption). ∗∗P < 0.01; ∗P < 0.05.

A cross-lagged model of patients' close relationship and spouses’ psychosocial adaption

The final modified model showed χ²/df = 166.376, RMSEA = 0.212, CFI = 0.803, indicating a good model fit. Patients' close relationship showed no significant interactional effects on spouses' psychosocial adaption at any time points. On the other hand, the higher the spouses' psychosocial adaption score at T0, the lower the patients’ close relationship score at T1 (β = −0.343, P = 0.001). None similar or opposite association were found at other time points. More details were illustrated in Fig. 4.

Fig. 4.

A cross-lagged model of patients' close relationship and spouses' psychosocial adaption. P, patient; S, spouse; T0, baseline; T1, 1 month; T2, 3 months; T3, 6 months. P < 0.05 was considered statistically significant. ∗∗P < 0.01; ∗P < 0.05.

Discussion

To the best of our knowledge, this is one of the first longitudinal studies to explore the interactional effects of dyadic coping and close relationship on psychosocial adaptation within prostate cancer couples during the first six months after diagnosis. Our finding highlights that spouse effects of spouses' dyadic coping on patients psychosocial adaption are more obvious than vice verse. Rather than implying a uniformly beneficial role, this result contributes to a clearer understanding of interpersonal processes within couples coping with prostate cancer and underscores the spouses' roles in patients’ psychosocial trajectories.

Our results indicate that the first month following a prostate cancer diagnosis represents a period of markedly heightened psychological and physical burden for both patients and their spouses. With the exception of patients’ psychosocial adaptation, which did not show a significant change from T0 to T1, significant actor effects were observed for dyadic coping, close relationship, and psychosocial adaptation across time points. These findings are consistent with a longitudinal study³⁴ conducted in Japan, which also identified the first month after diagnosis as the most distressing period for prostate cancer couples.

This early phase involves confronting the emotional shock of cancer diagnosis, making treatment-related decisions, and coping with distressing symptoms such as urinary incontinence and sexual dysfunction resulting from neoadjuvant hormonal therapy or radical prostatectomy.³⁵ Unlike breast or gastrointestinal cancers, prostate cancer triggers a more internalized and stigmatized responses in men, characterized by emotional suppression or avoidance,³⁶ which hamper effective communication and mutual support within couples.^21,37 Above physical and emotional stressors contribute to a disruption of psychosocial adaptation during the initial month. In addition, previous studies^38,39 reported heightened distress at both one and six months after diagnosis, further advancing the necessarily to explore time-dependent and reciprocal influences within couples, as modeled in the present study.

Although the first month after diagnosis was associated with obvious psychosocial distress and rapid changes in dyadic processes, the longitudinal trajectories observed in this study primarily reflect a natural adaptation process. Importantly, the CLPM did not proved that early measurements robustly predict long-term psychosocial outcomes. Therefore, rather than defining the first month as a uniquely actionable intervention window, our findings suggest that it represents a period of heightened vulnerability during which couples may be particularly sensitive to psychosocial stressors. Future studies are needed to determine whether interventions delivered during this phase can meaningfully alter long-term adaptation trajectories.

Notably, the interdependent effects of spouses' dyadic coping on patients' psychosocial adaptation were evident shortly after diagnosis. Apart from a negative association between spouses' dyadic coping at T0 and patients' psychosocial adaptation at T1, our findings suggest that coping strategies adopted by spouses at T0 exert a delayed and context-dependent effects on patients' psychological outcomes. One possible explanation is that couples may prioritize crisis management and family resource adjustment rather than mutual emotional support during the initial phase after diagnosis. In this context, elevated levels of spousal dyadic coping may reflect reactive or compensatory efforts in response to patients' distress, which fail to translate into immediate improvements in patients’ adaptation. This interpretation remains speculative and warrants further empirical examination.

Our results are partially consistent with findings from Bodschwinna et al.¹³ and Rottmann et al.,¹⁴ which demonstrated positive spouse effects in other cancer populations, although the study populations differed from prostate cancer couples. A review by Falconier et al.⁴⁰ also reported heterogeneous and sometimes contradictory effects of dyadic coping across studies. In addition, a study conducted among Chinese cancer couples¹⁵ identified both actor and spouse effects without focusing specifically on prostate cancer. Interestingly, patients' positive effects on their spouses emerged only from T2 to T3, which may reflect patients' gradual adaptation to the disease and more active engagement in coping processes, thereby alleviating part of the spouses’ psychological burden. Overall, these findings highlight the complexity of dyadic coping processes observed in this study, particularly the time-dependent reversal of spouse effects. It is also important to note that dyadic coping is a multidimensional construct, and the present study did not distinguish between potentially supportive versus overprotective or mismatched coping behaviors, which may partly explain the heterogeneous spouse effects observed across time points.

In contrast, no clear and consistent interactional effects of close relationship on psychosocial adaptation were identified across time points. Treatment-related side effects, especially sexual related changes, undermine intimacy in prostate cancer couples.⁴¹ Previous research by Kamen et al.⁴² demonstrated that higher spouse support was associated with lower distress (16.20–19.19 points lower, P < 0.001), suggesting that emotional support enhances feelings of being understood and valued. Moreover, open communication has been recognized as facilitating mutual understanding and intimacy, thereby alleviating psychosocial burden.⁴³ These dynamics are well described in the Relationship Intimacy Model¹⁶ and were partially validated in our findings. However, our longitudinal analyses did not reveal robust time-dependent causal effects between close relationship and psychosocial adaptation. Similarly, a recent study²² identified indirect associations between psychological distress and relationship satisfaction. Taken together, these findings suggest that while close relationship and psychosocial adaptation are interrelated, their temporal and causal relationships within prostate cancer couples remain complex and inconclusive.

The influence of dyadic coping and close relationship from spouses to patients was generally stronger than the reverse direction, indicating an asymmetric nature of psychosocial interdependence within prostate cancer couples. This observation is consistent with recently published research showing that female spouses often experience greater emotional fluctuations than male patients during the early months after diagnosis, partly due to caregiving responsibilities and family roles.^7,18 Patients and their spouses may face distinct physical and psychological burdens, particularly when confronted with a sudden cancer diagnosis and limited disease-related knowledge.⁴⁴ For example, treatment-related sexual dysfunction is difficult for patients to disclose, whereas spouses’ involvement in decision-making may result in support flowing predominantly from spouse to patient.⁴⁵ At the same time, patients experience increased burden when spouses adopt overprotective or overly controlling coping behaviors, which may partly explain the negative spouse effects observed in the early post-diagnosis period.⁴⁶ These interpretations should be viewed as hypothesis-generating rather than definitive explanations.

In light of the time-dependent, asymmetric, and occasionally negative spouse effects observed in this study, the clinical implications should be interpreted with caution. First, spouses of prostate cancer patients also experience substantial psychosocial distress, which may equal or greater than that of patients, yet their needs are frequently overlooked by both patients and health care providers. Second, although spouses' dyadic coping was longitudinally associated with patients' psychosocial adaptation, our findings do not support promoting dyadic coping as a universal or uniformly beneficial intervention target. Early increases in spouse coping may lead to heightened distress or overprotective dynamics, which could inadvertently place additional psychological burden on patients. Third, consistent with the longitudinal patterns observed in this study, the first month after diagnosis may be better understood as a period of heightened psychosocial vulnerability rather than a definitive intervention window. During this phase, couples may be particularly sensitive to psychosocial stressors, underscoring the importance of careful clinical attention without assuming that early intervention alone will determine long-term adaptation. In sum, psychosocial care may benefit from a careful assessment of how and when coping efforts are enacted within couples, with particular attention to potential mismatches between coping intentions and recipients’ needs. Future intervention studies are warranted to determine whether modifying specific dyadic coping patterns, rather than simply increasing coping efforts, can lead to improvements in psychosocial outcomes for both patients and their spouses. Although cross-lagged panel models allow examination of temporal precedence, causal inferences should be made cautiously, and unmeasured confounding factors may still influence the observed associations.

This study has several strengths. First, it was a longitudinal dyadic design with multiple assessment points within the first six months after prostate cancer diagnosis, allowing exploration of temporal and directional associations between patients and spouses rather than just cross-sectional correlations. Second, the use of a cross-lagged panel model enabled simultaneous estimation of actor and partner effects, accounting for interdependence within couples and providing a more comprehensive understanding of dyadic processes over time. Third, the relatively high follow-up rate and minimal missing data enhanced the credibility of the longitudinal analyses. Together, these methodological strengths contribute to the growing evidence on couple-based psychosocial adaptation following prostate cancer diagnosis.

Our study has several limitations that should be acknowledged. First, although the study was conducted at one of the largest tertiary uro-oncology centers in China, the single-center design limit the generalizability of the findings to other clinical settings, geographic regions, or cultural contexts. Second, while the longitudinal CLPM allowed exploration of temporal associations among dyadic coping, relationship quality, and psychosocial adaptation, the observational nature of the data precludes definitive causal inferences. It is also possible that alternative longitudinal analytic approaches could reveal different patterns of association. Third, the study sample consisted exclusively of heterosexual couples, which limits the applicability of the findings to couples with other gender compositions or relationship structures. Finally, as this was a non-interventional study, it cannot determine whether modifying dyadic coping or relationship processes would directly improve psychosocial adaptation in prostate cancer couples. Future randomized or theory-driven intervention studies are needed to translate these relational findings into clinically actionable strategies.

Conclusions

Psychosocial adaptation represents an important health-related outcome for couples coping with major stressors such as prostate cancer. The present findings suggest that spouses' dyadic coping is longitudinally associated with patients' psychosocial adaptation during the first six months after diagnosis, whereas patients' dyadic coping appears to be related to spouses’ psychosocial adaptation primarily at the later stage of follow-up. No consistent longitudinal associations were observed between relationship quality and psychosocial adaptation within prostate cancer couples.

Our results highlight the complexity and time-dependent nature of dyadic processes following a prostate cancer diagnosis. The findings underscore the importance of cautiously interpreting dyadic coping dynamics and suggest that future psychosocial interventions should be informed by a comprehensive assessment of spouse roles, timing, and potential unintended effects, rather than uniformly targeting dyadic coping strategies after diagnosis.

CRediT authorship contribution statement

Xiuqun Yuan: Data curation, investigation, Writing - Original Draft, Writing - Review and Editing. Min Chen: Data curation, project administration. Ting Zhang: Data curation, investigation. Zhongyang Sun: Statistical modeling, Formal analysis, Data analysis and result verification; Hongfan Yin: Methodology, Writing - Review and Editing. Zhihong Yang: Data curation, investigation. Yan Yang: Writing - Review and Editing, Supervision. All authors have read and approved the final manuscript.

Ethics statement

The study was approved by the Institutional Review Board of the School of Public Health and Nursing, Shanghai Jiao Tong University School of Medicine (Approval No. SJUPN-HY-202312-23-KS1) and was conducted in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All participants provided written informed consent.

Data availability statement

All data and analysis code are available from the corresponding author, YY, upon reasonable request.

Declaration of generative AI and AI-assisted technologies in the writing process

AI-assisted tools (e.g., ChatGPT) were used only for language editing. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Funding

This study was supported by the National Natural Science Foundation of China (Grant No. 72174120). The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication.

Declaration of competing interest

The authors declare no conflict of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: