In this large, geographically and clinically diverse sample of US patients with schizophrenia, 23.9% met the criteria for symptom remission at baseline. This post-hoc analysis of a 3-year prospective, observational study demonstrated that failure to achieve remission at study enrollment was associated with increased symptoms of schizophrenia, increased healthcare costs, worse HRQOL and functional outcomes, and a greater likelihood of interacting with the criminal justice system. Even though remission status was defined at baseline, the differences for most measures appeared stable across all 3 years of the study.
Consistent with past research, this study found non-remitted patients had more severe symptoms of schizophrenia
[9–18, 30]. Given that remission is defined based on the symptom rating, this finding was expected at baseline. More informative was the finding that the difference in symptoms largely remained across the 3-year study. This finding replicates an earlier study
 which found non-remitted patients to continue to be more symptomatic than remitted patients 3 years later. The subset of patients who met remission criteria appeared to maintain lower levels of symptoms over time.
Non-remitters had significantly higher costs in every category but inpatient costs. Although the costs decreased over time, differences remained between the baseline non-remitters and remitters. After baseline, the total cost difference ranged between $1200 and $2800 greater for the non-remitted patients during every 6-month period. This finding appears to be a unique contribution to the literature. Prior cross-sectional research in Sweden found patients who obtain remission use fewer healthcare services, although this was not linked directly to costs
. Effective treatments that move patients into remission could potentially reduce the burden of schizophrenia on the healthcare system, but more research is needed.
The non-remitted patients reported worse medication adherence. Some of the increased costs could be due to reduced medication adherence resulting in more relapses
. Relapses have substantial effects on healthcare costs
 and medication adherence has been previously shown to be associated with remission
For multiple clinician and patient-rated measures of HRQOL and functioning, the non-remitted patients appeared significantly more impaired at baseline and across the 3-year study. This was found for all studied measures except the physical component score of the SF-12 and the percentage of patients living independently. Significance tests showed that some of the functional measures were changing over time for the non-remitted or remitted patients, but the time effects were small relative to the effect of symptom remission. Worse functioning and quality of life in non-remitted patients has been reported in past research
[9–11, 13, 19–23]. On the SF-12 summary scores, the remitted patients average score was below the population average score of 50. This highlights that meeting the criteria for symptom remission does not imply clinical recovery in schizophrenia.
Recovery is schizophrenia has been defined objectively as clinical recovery or subjectively as personal recovery
[33, 34]. Clinical recovery, which has been the focus in the scientific literature, defines recovery as the absence of symptoms and returning to levels of premorbid functioning including working, living independently, and carrying out activities of daily living
. Personal recovery focuses on the more subjective process of adaption to the illness and encompasses self-awareness, a sense of empowerment, and functioning at one’s best despite ongoing symptoms
[34, 35]. Important concepts in personal recovery include overcoming poverty, stigma, demoralization, hopelessness, and social isolation
. Recent research has found that the development of a personal narrative mediates the relationship between deficits in social cognition or social withdrawal and negative symptoms
 and that vocational rehabilitation is linked to reductions in self-stigma
. Future research is needed to examine the association between symptom remission and measures of personal recovery. Whether considered from the clinical or personal perspective, recovery in schizophrenia is the ultimate goal and goes beyond symptom remission
The current study contained a unique set of variables asking patients about past violence, victimization and arrests. Although the overall incidence for each was low, and appeared to decrease slightly over the 3-year study, non-remitted patients were significantly more likely to report violent behaviors as well as being victims of crimes than the remitted patients across the 3-year study. The difference in violent behaviors was more prominent in the first year of the study. Further research into the potential legal repercussions of failing to obtain remission is needed. Individuals with schizophrenia appear to be at an increased risk for repeat incarcerations
The findings of this study demonstrated that over a 3-year period, non-remitted patients have a substantially increased burden on the United States healthcare system compared to patients who have obtained baseline remission. Although reduced healthcare use has been shown previously in Sweden
, the current study extends this finding to costs over three years among a large representative sample of individuals with schizophrenia in the US. Potential healthcare savings of moving patients into remission could be as high as $1200 to $2800 per patient every six months. Perhaps, treating schizophrenia more aggressively with more efficacious agents
 or combination therapy
 could result in more patients reaching remission and reduce the economic burden on the healthcare system, but more research is clearly needed.
Alternatively, remission status in schizophrenia may represent a patient “trait” characteristic rather than a current “state.” Past research has identified certain patient characteristics that are predictive of obtaining symptom remission in schizophrenia: higher educational status
, lower symptoms severity
[12, 32], being married
, shorter duration of untreated psychosis
, no substance use diagnosis
, and higher levels of functioning (employed, living independently, and higher subjective well being under neuroleptics scores)
. Several of these same variables were significantly different between the remitted and non-remitted patients at baseline in the current study (see Table
1). Constructs from personal recovery in schizophrenia, such as a sense of personal agency, may have also differed between the remitted and non-remitted patients
, but these were not measured in our study. Achieving symptom remission may reflect characteristics of certain patients with schizophrenia who tend to have favorable outcomes rather than the effects of treatment. On the other hand, initial treatment with atypical instead of typical antipsychotics has been predictive of achieving symptom remission
 and treating first-episode patients with both antipsychotic and psychosocial treatment has been predictive of better long-term outcomes
. More research is needed to differentiate the patient selection effects from the treatment effects on symptom remission.
In this study, remission was defined at baseline only, but the published criteria also require the reduced symptoms to be maintained for a period of at least 6 months
. Had the longitudinal requirement been added, some patients classified as remitted may have been classified as non-remitted. However, the US-SCAP study only collected the PANSS annually and the consistent differences between the two cohorts on multiple measures over time suggest that most of those classified as remitted likely stayed in remission. In addition, the results do not provide information about gains from non-remitted patients who subsequently reached treatment remission. Instead, this study can only provide information about the differences between those who were classified as remitted or non-remitted at baseline. In this study, a substantial rate of missing data for the total costs might have led to unreliable estimates of cost differences between remitted and non-remitted patients. Nevertheless, sensitivity analysis using multiple imputation to impute the missing data confirmed the overall findings. Finally, the label of remission does not mean complete functional recovery. In this study, patients meeting the criteria for remission continued to display functional impairments and did not achieve functional levels of the general population.