1. Consensus and Relationship Distress before and after a Brief Relationship Intervention for Low-Income Couples
LUCIA MIRANDA, M.S.
KATIE LENGER, M.A.
AMY RAUER, PH.D.
PATRICIA ROBERSON, PH.D.
KRISTINA COOP GORDON, PH.D.

2. Consensus in Couple Relationships
Lack of consensus is associated with greater relationship distress (Avivi, Launrenceau, & Carver, 2009; Gaunt, 2006; Gonzaga, Campos & Bradbury, 2007)
Partners show little consensus on reasons for seeking couples therapy (Doss, Simpson, & Christensen, 2004), which may diminish their benefits from brief interventions (Biesen & Doss, 2012)
Consensus on strengths may be less important than consensus on concerns
Marriage has been described as “constructing a shared reality.”
We choose partners similar to ourselves, because shared values foster positive interactions and mutual attraction (assortative mating), AND/OR, converge to become more similar with our partners over time, because it validates us and helps us coordinate responses to environment (emotional convergence).
In contrast, lack of consensus is associated with relationship distress.
When partners seek couples therapy, they often do not not agree on the reasons for doing so.
And when they do not agree on presenting concerns, they have been shown to benefit less from brief problem-focused approaches.
On the other hand, less is known about the role of consensus on relationship strengths (or the things that are going well in the relationship), but some evidence suggest that it is less important for partners to show agreement about these (as satisfied partners may derive pleasure from different areas of the relationship)

3. Why focus on brief interventions?
Low income couples are at greater risk for relationship distress and dissolution, (Amato, Johnson, Booth, & Rogers, 2003) despite highly valuing marriage (Trail & Karney, 2012)
Yet they are underrepresented in the literature (Bradbury & Lavner, 2012)
Couples’ pre-treatment characteristics may impact intervention efficacy (Rauer et al., 2014)
There has been increased focus on brief couples interventions, because these are less costly and more accessible.
However, we know they don’t work for everyone. Instead, pretreatment characteristics need to be examined, and recent research has focused on figuring out ”what works for whom?”
Brief interventions may be particularly important for reaching low income couples, who are more at risk, despite valuing marriage just as much as other couples.

4. Current Study
Is couple consensus on presenting relationship concerns associated with presenting distress and changes in distress from pre- to post-intervention?
Is couple consensus on presenting relationship strengths associated with presenting distress and changes in distress from pre- to post-intervention?
Thus, this study focused on exploring consensus as a pre-treatment characteristic of low-income couples, and whether it moderated distress and changes in distress post intervention H1: Couples with greater relationship concerns consensus should be less likely to be clinically distressed pre- and post-intervention.
H2: Couples change in clinical distress pre- to post-test will be similar regardless of their consensus on presenting relationship concerns.
Due to the lack of prior research, did not make a priori hypotheses, but explored associations

5. Method: Participants Relationship Rx Program
1480 individuals (740 couples)
Married (n=867) or cohabiting (n=613)
Relationship length: M = 9.14 years (SD = 9.55)
Ages: 18 to over 64 years
Race/ethnicity: 79.2% White, 15.8% Black, 5% other
Median household income: $10,000 to $19,000
Median education: HS Diploma
We used data from the Relationship Rx program, a brief intervention designed for low income couples

6. Method: Procedure & Measures
Baseline Couple Satisfaction Index (CSI-16; Funk & Rogge, 2007)| Relationship Checkup Questionnaire (Cordova, 2014)
Intervention Session 1: assessment Session 2: feedback
One-month follow-up Post-intervention RCQ & CSI
Data were collected before and one month after the intervention
Each partner self-reported via the RCQ (pictured)
Consensus was coded based on number of agreements on the RCQ

7. Method: Procedure & Measures
Baseline Couple Satisfaction Index (CSI-16; Funk & Rogge, 2007)| Relationship Checkup Questionnaire (Cordova, 2014)
Intervention Session 1: assessment Session 2: feedback
One-month follow-up Post-intervention RCQ & CSI
Data were collected before and one month after the intervention
Each partner self-reported via the RCQ (pictured)
Consensus was coded based on number of agreements on the RCQ

8. Method: Procedure & Measures
Baseline Couple Satisfaction Index (CSI-16; Funk & Rogge, 2007)| Relationship Checkup Questionnaire (Cordova, 2014)
Intervention Session 1: assessment Session 2: feedback
One-month follow-up Post-intervention RCQ & CSI
Data were collected before and one month after the intervention
Each partner self-reported via the RCQ (pictured)
Consensus was coded based on number of agreements on the RCQ

9. Method: Procedure & Measures
Baseline Couple Satisfaction Index (CSI-16; Funk & Rogge, 2007)| Relationship Checkup Questionnaire (Cordova, 2014)
Intervention Session 1: assessment Session 2: feedback
One-month follow-up Post-intervention RCQ & CSI
Data were collected before and one month after the intervention
Each partner self-reported via the RCQ (pictured)
Consensus was coded based on number of agreements on the RCQ

10. Analyses
Multi-level logistic regressions in Mplus, with interaction term
Control variables: gender, marital status, poverty, race, and parenting status
To examine whether agreement on strengths and concerns moderated change
in satisfaction pre- and post-intervention, we examined two three-level multi-level models.
Specifically, we modeled time points (Level 1) nested in individuals (Level 2) nested within
couples (Level 3). This statistical method controls for the observational dependence of both the
repeated measures and the dyadic data. Control variables were included at the appropriate level
of the model: gender (Level 2), marital status (Level 3), poverty (Level 3), race (Level 2), and
parental status (Level 3).

11. Results
For Concerns
Main effect of time (i.e. the intervention; 67% decrease in odds)
Main effect of consensus (57% decrease in odds)
Significant effect of Consensus x Time interaction
Table 2. Consensus on Relationship Concerns Predicting Relationship Distress over Time
Overall, all couples appeared to be less distressed after the intervention
Overall, the couples with more consensus appeared to be distressed pre-intervention, but there was an interaction of consensus and consensus
For Strengths
No statistically significant effect of strengths consensus
Note. *p < .05, ** p <.001

12. Results – Concerns Agreement
Figure 1. Percentage of Distressed Couples in Each Level of Consensus for Concerns
Only couples with 0 (B = -1.30; OR = .27, t = -4.25, p = .00) or 1 concern agreement (B = -1.09; OR = .34, t = -2.00, p = .05) had statistically significant decrease in odds of being clinically distressed

13. Results – Concerns Agreement
Figure 1. Percentage of Distressed Couples in Each Level of Consensus for Concerns
agreed least
Only couples with 0 (B = -1.30; OR = .27, t = -4.25, p = .00) or 1 concern agreement (B = -1.09; OR = .34, t = -2.00, p = .05) had statistically significant decrease in odds of being clinically distressed

14. Results – Concerns Agreement
Figure 1. Percentage of Distressed Couples in Each Level of Consensus for Concerns
agreed least
agreed most
Only couples with 0 (B = -1.30; OR = .27, t = -4.25, p = .00) or 1 concern agreement (B = -1.09; OR = .34, t = -2.00, p = .05) had statistically significant decrease in odds of being clinically distressed

15. Results – Concerns Agreement
Figure 1. Percentage of Distressed Couples in Each Level of Consensus for Concerns
agreed least
greatest change
agreed most
Only couples with 0 (B = -1.30; OR = .27, t = -4.25, p = .00) or 1 concern agreement (B = -1.09; OR = .34, t = -2.00, p = .05) had statistically significant decrease in odds of being clinically distressed

16. Considerations and Conclusions
Consensus on concerns appears to be more indicative of relationship satisfaction than consensus on strengths
This brief integrative approach appears to be most beneficial for couples with lower levels of initial consensus
Future studies that use random-assignment, include indicators of change beyond distress, and assess long-term change are necessary
Agreement on concerns was more important than agreement on satisfaction. Maybe it does not matter as much whether we agree on the things we’re doing well as a couple, because they are not threatening. Maybe it is actually good that we see ourselves having different strengths, because it gives us a greater repertoire to draw from.
This intervention appeared to be most beneficial for the couples that agreed least on concerns, in contrast to previous studies that found brief problem-focused interventions were less effective for those who disagreed
Limitations of this study

17. Implications
Pre-treatment assessment of consensus on presenting concerns can help therapists to select a treatment modality
Consensus might indicate better relationship functioning, but it is not a prerequisite for improved satisfaction over time
Non-traditional brief interventions that integrate emotion-focused, acceptance, motivation, and behavioral techniques can reduce relationship distress
Even for couples with contextual barriers

18. Acknowledgments
The Relationship Rx Project was funded by the U.S. Department of Health and Human Services, Administration of Children and Families, Office of Family Assistance   Grant #90FM0022
We thank the Relationship Rx Team
and
all the couples who participated!