Family Involvement in the Clinical Care of Clients With First-Episode Psychosis in the RAISE Connection Program

Sixty-five participants were enrolled in the RAISE IES study in Baltimore and New York City. Eligible persons were age 15–35 years; met DSM-IV criteria for schizophrenia, schizoaffective disorder, schizophreniform disorders, delusional disorder, or psychosis not otherwise specified; and had experienced psychotic symptoms for at least a week with onset in the past two years. Exclusion criteria included having a nonpsychiatric medical condition that impaired functioning, another diagnosis causing the psychosis, or mental retardation. Participants provided informed consent; for minors, the parent or guardian provided informed consent and the minor provided assent. The institutional review boards of the New York State Psychiatric Institute and the University of Maryland approved all study procedures.

Starting on July 1, 2011, trained clinical research interviewers completed standardized assessments at baseline and at six, 12, 18, and 24 months or until data collection ended on June 30, 2013. Participants who were enrolled after February 7, 2013, or who declined further participation had shorter follow-up periods. Clinical assessment and study participation details have been previously reported (7). The Positive and Negative Syndrome Scale (PANSS) (8) was used to assess symptoms; the Mental Illness Research, Education and Clinical Center Global Assessment of Functioning (MIRECC GAF) social and occupational functioning scales adapted for individuals with early psychosis were used to assess social and occupational functioning (9); and two items from the Addiction Severity Index (10) were used to assess alcohol or drug use. Chart reviews provided information on family participation in visits with each team member. Three items from the six-month assessment were used to assess the client’s experience of the treatment model focused on family involvement: In the previous month, “How often did providers talk with you about how family could help support you in reaching your goals?” “How often did providers talk with you about how you would like family/supportive others to be involved in your treatment?” and “How often did providers help you talk with family about your thoughts and feelings?” Possible responses were 1, not at all; 2, a little; 3, a moderate amount; and 4, a lot.

Percentages and means were calculated to describe family visit attendance with each team member and overall and the percentage of clients whose family attended at least one visit for each three-month quarter of program participation (for example, quarter 1, months 1–3 of the participant’s involvement; quarter 2, months 4–6 of the participant’s involvement) was determined. A multiple regression that included age, gender, baseline clinical characteristics, and illness severity (20-item PANSS total score) (11), substance use (presence or absence), inpatient hospitalization within six months of baseline (illness severity marker), and functioning (MIRECC GAF scores) as independent variables was conducted to assess predictors of days of family visits during the first six months of program participation. Analysis was limited to the first six months because of drop-off in the number of participants (due to study termination). A separate regression model was used to assess whether clients’ report of client-provider discussions regarding family and family involvement was related to the number of days of family visits during the first six months of program participation.

A total of 41 (63%) participants were male; 24 (37%) were female. Most had a diagnosis of schizophrenia (N=43, 66%). The mean±SD age of participants was 22.2±4.2 years. Eight participants were younger than age 18 years. A total of 28 (43%) identified themselves as black, 25 (39%) as white, 4 (6%) as Asian/Pacific Islander, and 8 (13%) as a member of other groups or did not specify a group. Most participants were living with family (N=49, 76%). Almost all participants (N=62, 95%) had a family member who had contact with the Connection team during program involvement; a majority had multiple contacts. Table 1 shows data on attendance by family members during visits with various team members and the visit pattern over time.

Multivariate analyses showed that age (β=.66, t=−4.97, df=1, p<.001), symptom severity (β=.21, t=2.15, df=1, p=.036), and substance use (β=−.28, t=−3.02, df=1, p=.004) predicted number of family visit days during the first six months of the program. [Further details about the analyses are available in the online supplement to this report.] Participants younger than age 18 years, with greater symptom severity, and with no substance use had more family visit days. This model explained 59% of the variance (F=11.12, df=7 and 54, p<.001).

The mean response for frequency of client-provider conversations around how family could support goals, how clients would like family to be involved in treatment, and helping clients talk with family about thoughts/feelings was between “a little” and “a moderate amount” for all three questions (2.59±1.14; 2.63±1.07; 2.91±1.07, respectively). Increased self-reported discussions of family involvement did not predict the number of days family members were present at visits with a team member during the first six months (F=.92, df=3 and 52, p=.44).

Almost all of the study participants with FEP had a family member involved in the program at some point during their care, with many having family contact with multiple team members. Although it is not surprising that most contact was with the team leader and the psychiatrist, given their roles, families had substantial contact with other team members, suggesting that families were interested in an array of care components. This may be due to the program’s shared decision-making approach to services, which allows providers to educate clients and family members concerning ways family can be involved and how involvement can support recovery and which ensures access and opportunities to work with a variety of team members.

Family member involvement occurred more frequently during the first quarter of the program. Greater involvement in instances of acute distress or during service transitions has been shown to be common, because of the greater need for information, guidance, and support (12). Declines in family presence at visits after the first quarter may reflect less perceived need for services later in care. Despite the decline, family involvement remained substantial over the course of the program. One key element of the program was creating a family friendly environment, with ongoing family outreach via phone, home visits, and mailings; inclusion efforts; and collaborative decision making around ways to address client and family needs. Qualitative interviews conducted with clients and family members indicated that ongoing, active outreach and support and tailoring of the program and services to client and family needs helped both clients and families engage with and stay connected to the program, likely contributing to ongoing family involvement (3,13).

Consistent with previous research (4,5), we found that clients’ being younger (younger than age 18 years) and having more psychiatric symptoms predicted greater family involvement during the first six months of the program. Because youths younger than age 18 years require a family member or guardian for care decisions, it is not surprising their families had more contact. However, the benefits of the outreach and engagement efforts, as well as working collaboratively with clients and families to identify recovery goals and ways family can support those goals, are just as important when working with younger clients. Less family involvement among young adults age 18 years and older may reflect less perceived need for family involvement. Alternatively, concerns about privacy, family burden, or loss of autonomy and independence may lead young people to struggle more with decisions around involving family (12). Similarly, family members may have reservations about involvement, due to limited knowledge, respect for the client’s privacy and autonomy, and time constraints (3). Potential barriers such as these highlight the importance of engaging clients with FEP and their families in discussions about the benefits of involvement and addressing any concerns that might prevent family participation.

More frequent family involvement among clients with greater symptoms may reflect greater perceived family need. Greater psychiatric symptoms exhibited by a client with FEP can lead to family distress and perceived burden (14), which, in turn, may lead to greater family involvement. Less family participation among those with substance use may reflect less interest on the part of the client in including family, especially if substance use is a source of disagreement, or less client participation in services, limiting opportunities for family involvement.

Client-provider conversations around family involvement and support were common, with most participants indicating these occurred to some extent during the previous month. However, client-provider conversations about family or family involvement did not predict family attendance at visits. Information on client-provider conversations was collected only for the month prior to the assessment, rather than over the full six months of initial program participation. Although responses suggest that conversations frequently occurred, whether they occurred consistently throughout participation and their potential impact on family involvement are unknown. Participation in family meetings was shown to mediate improvements in social functioning among individuals in the Connection Program (15). Future work is needed to determine whether regular discussions about family involvement are associated with family participation over time.

Family involvement in the ongoing care of clients with FEP is common, especially when active, ongoing efforts are made to engage clients with FEP and their families. Younger age, greater symptoms, and lack of substance use prior to entering the program predicted family attendance at program visits; client-provider discussions regarding family or family involvement in the month prior to the six-month assessment did not predict family attendance. Additional research is needed to determine whether regular, ongoing discussions are associated with family participation over time.