Understanding and engaging families: An education, skills and support program for relatives impacted by borderline personality disorder

Abstract

Background: Few methodologically sound studies have been conducted to facilitate the understanding of relationships in families with a member suffering from severe personality disorders. Because of severity, symptomatology, and high rates of co-occurring disorders, borderline personality disorder (BPD) particularly affects family members and others in their social environment. However, there are few interventions available to address the needs of these family members.

Aim: To report on a replication and extension study of ©Family Connections (FC), a 12-week community-based BPD education program for family members.

Method: FC participants (N = 55) were assessed pre, post and at 3-month post program follow-up on: (i) levels of change in family well-being outcomes: burden, grief and empowerment; and (ii) differences between male and female participants in these outcomes.

Results: Findings from the original FC study were replicated and extended: Participants showed significant improvements on all well-being variables, including significant reductions in depression. Outcomes for male vs. female participants were comparable at program completion except for grief, on which women remained higher than men despite significant improvements for both.

Conclusion: BPD family members experience significant distress but benefited from this semi-structured group program led by family members. Findings support the use of the FC program.

Declaration of interest: None.

Keywords:

• Personality disorders
• borderline personality disorder
• family members
• community-based program

Notes

1 When testing post-test to follow-up change for all outcomes, a problem was encountered with the variance/covariance matrix. Setting the level one residual parameter variance to a near-zero value did not alleviate this problem, so post to follow-up change was tested as a fixed rather than a random effect. As a result, change from post-test to follow-up in these two models was tested with degrees of freedom according to the number of level-one units rather than the level two degrees of freedom used in the testing of pre to post change. Because of this change in the model, the following results should be interpreted with caution, as they were tested with more degrees of freedom than the other outcomes in this study, and than the outcomes in the initial study (Hoffman et al., 2005).