Clinical Correlates Of Initial Treatment Disengagement In First-Episode Psychosis

The following is a summary of Myers N, Bhatty S, Broussard B, Compton M. Clinical correlates of initial treatment disengagement in first-episode psychosis. Clin Schizophr Relat Psychoses. 2014 Nov 3:1-21. [Epub ahead of print], which was developed independently of the article authors.

• Suggested to be indicative of high-quality relationships between mental health professionals, patients, and caregivers, early engagement in care is thought to reduce disabling social losses (eg, homelessness, incarceration, and violence) related to the DUP.
  • Studies of adults with serious mental illness suggest that more than half (median 58%) disengage from treatment after hospitalization.
  • There is little known about initial disengagement among patients with FEP, or disengagement that occurs within the first 3 follow-up appointments after initial hospitalization for psychosis.
• Psychotic symptoms may continue if left unchecked, and DUP may be prolonged in patients who drop out of treatment and disengage. Preventing longer DUP is important as a longer DUP is not only associated with greater psychiatric symptoms but also contributes to repeated incarcerations, hospitalizations, homelessness, interpersonal violence, and suicide, as well as poor recovery outcomes more generally.
• Treatment engagement is complex, and there are currently no “engagement best practices”.
  • A better understanding of early treatment drop-out may facilitate the development of appropriate interventions to prevent disengagement.

This study compared key early-course clinical variables between patients with FEP who engaged in treatment after hospitalization and patients who did not engage in treatment.

• Clinical variables included medication-adherence attitudes and behaviors, knowledge about schizophrenia, insight, symptom severity, and persistence of alcohol and drug use.

• This study used longitudinal data from an observational study of predictors of DUP.
• Patients were recruited from 3 hospital settings (an inpatient psychiatric unit and 2 crisis stabilization units) within the same geographic and economic area.
  • Within these settings, patients with FEP received usual treatment (ie, the public mental health system).
  • No specialized early psychosis services were available.
  • Upon hospital discharge, patients received standard follow-up appointments for a local community mental health clinic outside the immediate vicinity of the hospital, where care would have centered on medication management.
• To be included in the study, patients aged 18–40 years were required to have been hospitalized for a first episode of a primary psychotic disorder and to have attended a follow-up research assessment 6 months after discharge from initial hospitalization.
• Structured questions were used to assess follow-up adherence and medication adherence.
  • Information was collected regarding the number of outpatient appointments attended by the patient in the 6 months since hospital discharge, as well as the average number of days per week the patient was adherent to medication in the past month and the past 6 months.
  • Patients who attended ≥3 follow-up appointments in the 6 months after hospital discharge were considered engaged, and patients who attended <3 outpatient appointments in the 6 months after hospitalization were considered disengaged.
• Various rating scales were used to assess other clinical variables.
  • The 10-item Medication Adherence Rating Scale (MARS) was used to determine patients’ behaviors and attitudes towards their medications.
  • The 18-item Knowledge about Schizophrenia Test (KAST) was used to assess patients’ knowledge about schizophrenia.
  • A modified version of the 8-item Birchwood Insight Scale (BIS) was used to assess patients’ level of insight and need for care.
  • The Positive and Negative Syndrome Scale (PANSS) was used to assess current/past-month symptom severity.
  • The Alcohol Use Scale was used to determine the level of alcohol misuse, and the Drug Use Scale was used to determine the level of drug misuse.
    • Categories were dichotomized into “abstinent” (score of 1) versus “use” (all scores ≥2).

Regarding study sample

• Of the overall sample of 109 patients included in a baseline assessment, 33 patients were assessed at a follow-up appointment 6 months after hospitalization and were included in this study.
  • In terms of sociodemographic characteristics, baseline alcohol use, drug use, symptom severity, and insight, these 33 patients were similar to the 76 patients who completed a baseline research assessment but were not reassessed at 6 months (all p>0.34).

Regarding sociodemographic characteristics

• The mean ± standard deviation (SD) age of the study sample (n=33) at hospitalization was 22.9 ± 4.5 years, and patients had completed an average of 11.3 ± 2.1 years of education.
• Most patients were male (81.8%), African American (90.9%), single and never married (90.9%), living with family members (63.6%), unemployed (66.7%), and living below the federal poverty level (65.6%).
• There were 6 (18.2%) patients with a Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, fourth edition, Axis I Disorders—diagnosed alcohol-use disorder (abuse or dependence), and 21 (63.6%) patients had a cannabis-use disorder.

Regarding prevalence of treatment disengagement

• In the 6 months after hospitalization, 18 (54.5%) patients attended <3 outpatient appointments (disengaged), and 15 (45.5%) patients attended ≥3 outpatient visits (engaged).
• Among the 18 patients who had disengaged, 15 (83.3%) had attended no outpatient treatment visits during the 6 months after hospitalization.

Regarding comparison of disengaged and engaged patients

• Sociodemographic characteristics
  • The 18 patients who were disengaged differed from the 15 who were engaged in terms of educational attainment and living below the federal poverty line.
  • Patients who were disengaged had completed an average of 12.0 ± 2.3 years of education compared with 10.5 ± 2.3 years for those who were engaged (p=0.047).
  • Compared with 85.7% of patients who were engaged, 46.7% of those who were disengaged lived below the federal poverty line (p=0.05).
• Key clinical variables
  • Compared with patients who were engaged, those who were disengaged reported significantly less adherence to medications in the past month (4.6 ± 3.2 vs 0.6 ± 1.8 days/week; p=0.001) and 6 months (5.7 ± 2.1 vs 1.5 ± 2.2 days/week; p<0.001).
  • Compared with patients who were engaged, those who were disengaged had significantly lower scores on the MARS (5.9 ± 1.7 vs 4.5 ± 1.5; p=0.041), KAST (12.0 ± 2.9 vs 8.5 ± 3.2; p=0.004) and BIS (7.9 ± 2.4 vs 5.4 ± 3.4; p=0.023) rating scales.
  • PANSS positive symptom severity score was 22.8 ± 8.8 for patients who were disengaged versus 17.2 ± 7.7 for those who were engaged (p=0.069), with no significant differences between the groups with regard to negative symptom severity and general psychopathology symptom scores.
  • Patients who were disengaged were significantly more likely than those who were engaged to have drug use at 6 months (61.1% vs 26.7%; p=0.05), and were also more likely to use alcohol (72.2% vs 46.7%), but this difference did not reach statistical significance (p value not reported).
  • At baseline, BIS scores, PANSS positive symptom severity scores, and levels of drug and alcohol use were not significantly different between patients who would later engage or disengage.

• In this sample, disengagement with treatment was common.
• Compared with patients who were engaged, those who were disengaged had worse medication attitudes and adherence, were not as knowledgeable about schizophrenia, and had poorer insight, more severe positive symptoms, and more persistent drug use.
  • Such differences were not apparent at baseline.
  • The only demographic differences at baseline were lower educational attainment and a lower likelihood of living in poverty among patients who later engaged in care, which the authors noted was counterintuitive.
• The authors suggested that further studies may be needed to identify the factors that affect young people with early psychosis who do not view care as a pathway to a better future, never engage, or are lost to follow-up after first contact with care.
• In this study, disengagement from treatment appeared to be a common problem in young patients with first-episode psychosis, and was associated with relatively poor clinical status.
• The authors suggested that age-appropriate interventions to improve outpatient treatment engagement may help to limit the burden of illness and associated psychosocial impairment that can occur when young patients are “lost to follow-up” or disengage from treatment after hospitalization for a first episode of psychosis.

This summation has been developed independently of the authors. There were no disclosures reported in the original article.