Trajectories of patients with severe mental illness in two-year contact with Flexible Assertive Community Treatment teams using Routine Outcome Monitoring data: An observational study [2019]

Using outcome data collected routinely over a continuous two-year treatment period, we wished to distinguish homogeneous subgroups of patients with a severe mental illness whose psychosocial problems followed a similar pattern over time. By identifying the effectiveness of health services for different patient groups, this approach allowed us to identify patients at risk of deterioration and those recovering from their symptoms.

In total we included 2,660 patients who were in two-year continuous contact with a Flexible Assertive Community Treatment team (FACT). We collected outcome data on psychosocial functioning, needs for care and quality of life. We performed a latent class growth analysis (LCGA).

The LCGA identified six homogenous patient subgroups using trajectories of HoNOS scores. On the basis of the patterns of patients’ psychosocial problems over time, we labelled these as follows: 1) stable at a low problem-severity level (N = 709; 27%); 2) stable at a low medium problem-severity level (N = 1,208; 45%); 3) stable at a high medium problem-severity level (N = 528; 20%); 4) stable at a high problem-severity level (N = 116; 4%); 5) amelioration of problems (N = 42; 2%); and 6) deterioration of problems (N = 57; 2%). Patients with stable and a high severity of psychosocial problems had more practical and somatic unmet needs than those in other subgroups, and also had the fewest decrease in the number of unmet needs.

After linking patient subgroups with clinical features such as the need for care, we found that, over two years, most patients remained relatively stable in terms of psychosocial functioning, but that their unmet needs decreased over time. However, in terms of needs for treatment during two years of contact with a FACT team, patients in the subgroup with a stable and high problem-severity level tended to derive little or no benefit.

Hans Kortrijk, Barbara Schaefer, Jaap van Weeghel, Cornelis L. Mulder, Astrid Kamperman
PLOS ONE, January 9, 2019