Young adults who drop out of psychosis treatment programs are more likely to have lower baseline negative symptom scores, a shorter duration of untreated psychosis, and a lack of an involved family member.
"With the increased attention to early intervention, understanding predictors of disengagement may have important implications for the development of specialty programs for this young population," wrote Jacqueline Stowkowy in the April issue of Schizophrenia Research.
Indeed, "in the design of services or programs for first-episode psychosis patients, it may be useful from the beginning to determine strategies that may prevent disengagement," for example, family outreach (Schiz. Res. 2012;136:7-12).
Ms. Stowkowy of the University of Calgary (Alta.) and her colleagues studied 266 patients admitted to the 36-month Calgary Early Psychosis Treatment Service, "a well-established comprehensive program for individuals who are experiencing their first episode of psychosis."
According to the authors, the service involves psychiatric and case management, a range of group programs, individual therapy, and family intervention.
The majority of patients were white males, with a mean age of 24.5 years. In addition, most of the participants were single (60%), lived at home (77.1%), and had completed the 12th grade (60%).
"Disengagement" from the program was defined as dropping out after fewer than 30 months, since the final 6 months were given to discharge planning. Individuals were considered to be dropouts when they did not return calls, could not be reached, or would not attend meetings for 3 months.
Overall, Ms. Stowkowy and her colleagues found a dropout rate of 31% for the program, "a number broadly consistent with previous research."
The authors then compared dropouts with patients who stuck with the program until reaching the 3-year anniversary of their admission. They found that the patients who left had lower negative symptom scale scores at baseline, compared with patients who stayed with the program (hazard ratio = 0.946; 95% confidence interval, 0.909-0.985).
Dropouts also had a shorter duration of untreated psychosis, compared with their counterparts (HR = 0.997; 95% CI, 0.994-0.999).
However, the most important overall predictor was not having a family member involved in the program, the researchers found (HR = 0.310; 95% CI, 0.196-0.490). "The two groups did not differ demographically on sex, marital status, living arrangements, or age," wrote the researchers.
Nor did they differ on many other scales, including the presence of positive symptoms, general psychopathology scores, insight, Calgary Depression Scale for Schizophrenia ratings, Global Assessment of Functioning scores, premorbid functioning scale scores, Quality of Life Scale scores, alcohol use, drug use, or cognition.
The researchers then divided the dropouts into four groups: those who dropped out prior to the 6-month assessment, those who remained in treatment for 6-17 months, those who held on for 18-29 months, and those who were still in treatment after 30 months.
In this model, using a one-way analysis of variance, "baseline cannabis and other drug use significantly differentiated those who dropped out of treatment within the first 6 months from both those who remained in treatment or did not leave before 6 months," the authors said.
The authors offered several possible explanations for their findings.
In the case of disengagement associated with fewer negative psychotic symptoms, "young, first-episode individuals whose positive symptoms have resolved and who have lower levels of negative symptoms may think that they are well enough to no longer need treatment," Ms. Stowkowy said.
And in relation to the finding that a shorter duration of untreated psychosis was predictive of dropout, she wrote, "it may be that those who sought treatment earlier were quickest to leave should there be a remission of symptoms."
However, the findings on associated drug use and family involvement offer the most utility in tailoring programs that minimize dropout rates.
"Those who enter with a substance problem need to be considered at entry as potential dropouts, and every effort should be made to engage them in the treatment," the investigators said.
"Another group to identify for specific interventions is those whose families are not engaged in treatment either due to their own or their family’s choice. Efforts to engage the family may serve to maintain the young person longer in the treatment," including active outreach to families who do not attend therapy with the patient.
Two authors disclosed personal funding from noncommercial sources, including the National Institute of Mental Health and the Alberta Heritage Foundation for Medical Research. The study also was supported partly by Eli Lilly.