The differential impact of duration of untreated psychosis on functioning and quality of life: A threshold analysis

Reduction of duration of untreated psychosis (DUP) remains a key goal of early intervention programs. While a significant body of literature suggests that a short DUP has a positive impact on outcome, little is known regarding the threshold above which various dimensions of outcome are impaired. In this study, we explore the DUP threshold that best discriminates subgroups with poorer outcome regarding global functioning and quality of life after 3 years of treatment.


| INTRODUCTION
Delay between psychosis onset and exposure to appropriate treatment or 'duration of untreated psychosis' (DUP), has been identified as a key target in specialized early intervention programs in the early phase of psychotic disorders (Joa et al., 2007;Melle et al., 2008;NHS, 2015). This is based on the assumption that its reduction may have a positive impact on the course of symptoms and functioning (Melle et al., 2008).
Several papers have shown DUP to be significantly associated with outcomes: these studies reported lower severity of symptoms, better functioning and quality of life, as well as higher remission rates in patients with shorter DUP (Kane et al., 2016;Malla et al., 2014;Marshall et al., 2005;Souaiby et al., 2016). In a sample of first-episode psychosis patients where the median DUP was 74 weeks, Kane et al. (2016) found that patients with DUP shorter than 74 weeks displayed ad greater improvement in quality of life and psychopathology than those with longer DUP.

12-week was recommended by the World Health Organization and
the International Early Psychosis Association. It was also empirically validated by Dama et al. (2019) who showed that a short DUP within this interval was critical for symptom remission in early intervention settings. Cost-effectiveness was also much greater for patients with a DUP shorter than 12 weeks (Groff et al., 2021). However, while a short DUP seems intuitively and empirically desirable, little is known about the precise DUP threshold above which functional recovery becomes unlikely. There is also a lack of knowledge regarding potential differential impact on different dimensions of outcomes such as symptom remission, quality of life or functioning level.
The objective of this study was to determine the DUP threshold that best discriminates subgroups with poorer outcome regarding global functioning and quality of life after 3 years of treatment. The decision to focus on functioning and quality of life rather than symptomatology is based on the observation that these two dimensions are considered by patients to be more important to achieve than symptom remission (Bonsack, 2019;Oberholzer, 2021;Solhdju & Hermant, 2015).

| Participants
The Treatment and early Intervention in Psychosis Programme (TIPP) is a specialized Early Intervention (EI) program run by the Psychiatry Department of Lausanne University Hospital's, in Switzerland (Baumann et al., 2013).
Patients aged between 18 and 35, living in the hospital's catchment area (population about 350 000) and meeting the criteria for 'psychosis threshold' subscale in the Comprehensive Assessment of At-Risk Mental States (CAARMS; Yung et al., 2005) are eligible to be included in this program. Exclusion criteria are drug-induced or organic psychosis, clinically assessed intellectual disability and antipsychotic medication for more than 6 months prior to their referral to the program. A multidisciplinary team (including psychiatrists and case management nurses) conduct an initial assessment to ensure the accuracy of inclusion criteria before admitting patients. This study is based on the data stemming from the prospective follow-up of 432 patients who had completed the 3 years treatment period by the end of 2020 and for which clinical data was available.
The principles of both case management interventions and assertive community treatment undertaken in outpatient settings are at the basis of the TIPP treatment. Over 3 years, case managers are available to patients up to twice a week. Patients are seen at least 100 times over the 3-year program, primarily by their case manager but also by a resident physician or an intern in psychiatry. Case managers and an experienced psychologist performed detailed evaluations of patients' past medical history, demographic characteristics, exposure to adverse life events, and their current symptoms and functioning using interviews and a structured questionnaire. All patients treated within the TIPP are fully assessed at baseline, after 2 months, 6 months and then prospectively every 6 months.
This study was carried out in accordance with the Declaration of Helsinki and was approved by the Human Research Ethics Committee of the Canton of Vaud (CER-VD; protocol #2020-00272). The data of all patients were used in the study if the latter did not explicitly object to the use of their data for research purposes. Only four patients refused the use of their clinical data for research.

| Measures
In order to be in line with most research in the domain, we defined DUP as the time between the onset of psychosis defined by the Comprehensive Assessment of At-Risk Mental States (CAARMS) instrument (Yung et al., 2005) and the admission to the TIPP. The psychosis threshold and its time are determined prior admission based on an expert consensus between the TIPP psychiatrists and case managers using information from medical or hospitalization reports from treating psychiatrists if available, as well as from the detailed report of the clinician who addressed the patient to the program. If the psychosis threshold cannot be determined clearly based on these reports, further specialized clinical assessments are conducted based on the structured interview for psychosis-risk syndromes (SIPS;McGlashan et al., 2001). Following this process, the clinical director of the TIPP completes the CAARMS. The CAARMS defines this psychotic disorder threshold as frank psychotic symptoms such as delusions, hallucinations and thought disorder persisting for longer than 1 week and with a frequency of at least 3-6 times a week for longer than 1 h each time or daily for less than 1 h each time. This is a standard and widely used criteria for first episode psychosis threshold (Nelson et al., 2014;Polari et al., 2011). We subdivided Patients' socioeconomic statuses into low, intermediate and high (Chandola & Jenkinson, 2000). The Case-managers used the Global Assessment of Functioning (GAF) instrument to estimate functional levels during the 3-year follow-up.
It assesses a combination of symptoms and functioning. The Casemanagers also used the Social and Occupational Functioning Assessment (SOFAS) in order to provide a measure that only takes the social and occupational functioning into account, regardless of the intensity of symptoms. In this study, we used the endpoint GAF and SOFAS (36 months). The Case-managers assessed quality of life at discharge with the World Health Organization Quality Of Life assessment scale (WHOQOL GROUP, 1995, 1998, which is a 26-item self-rated scale measuring satisfaction with life and self-esteem based on 5-point Likert scales ranging from 1 (low satisfaction) to 5 (high satisfaction).
The WHOQOL assesses self-perceived quality of life considering environment, social relationships, psychological and physical health.
This instrument showed high internal consistency and coherent pattern of correlations with other validated psychiatric measures in French language patients in our region (Golay et al., 2019;Golay, Martinez, et al., 2021;Golay, Moga, et al., 2021). If quality of life after 36 months was not available, we used the 30 months value instead.

| Statistical analysis
In order of determine the threshold for DUP regarding impact on quality of life and general functioning, patients were distributed into short and long DUP subgroups with varying allocation rules. We then compared groups for differences in quality of life and general functioning using Cohen's d effect sizes. Several cut-off points to define high/low DUP were used, including (1)

| RESULTS
The sample characteristics of the 432 patients are shown in Table 1.
A DUP threshold of 86 weeks yielded the most significant difference in the quality of life (600 days; d = 0.80; Figure 1). In our sample, only 21.1% (91) of the patients had a DUP equal or greater than 86 weeks. Examination of the curve also revealed that very short cutoffs were not associated with large differences in quality of life.
When the DUP cut-off was set at 74 weeks, the difference between patients above and below this threshold regarding quality of life at the end of the program was smaller although still large (d = 0.72). A total of 332 (76.9%) patients had a DUP shorter than 74 weeks.
When the DUP cut-off was set 12.5 weeks (our median DUP value), difference between patients above and below this threshold regarding quality of life at the end of treatment was smaller (d = 0.36).
A DUP threshold of 3 weeks yielded the most significant difference in GAF at the end of the program (20 days; d = 0.45; Figure 2). In our sample, 80.3% (347) of the patients had a DUP greater than 3 weeks. Difference in GAF between patients with a low or high DUP was small (d = 0.18) according to the 74 weeks cut-off. The difference according to the 12.5 weeks cut-off was larger (d = 0.34).
Examination of the SOFAS revealed a pattern of results very similar to the GAF. The maximum difference in SOFAS was found when the cut-off was set to a value greater than 3 weeks; (20 days; d = 0.42; Figure 3). Difference between patients with a low or high DUP was small (d = 0.12) according to the 74 weeks cut-off. The difference according to the 12.5 weeks cut-off was larger (d = 0.27).
Because Post Hoc analysis revealed that patients with a diagnosis of schizophrenia were under-represented within patients with very short DUP (<3 weeks; 38.8% vs. 62.0%), analysis were repeated on data of patients with schizophrenia diagnosis only. The cuff-off value for quality of life remained identical (86 weeks) but the effect size was slightly lower (d = 0.66). For GAF respectively the SOFAS, the cut-off associated with the greatest difference was slightly longer (8.6 weeks for both measures) and the effect size was lower (d = 0.24 and d = 0.10).

| DISCUSSION
In the present study, we tried to identify the DUP value that best separated two populations of patients with the most significant outcome differences regarding quality of life and GAF and SOFAS scores.
Because only very few patients refused the use of their clinical data for research, we consider the study sample to be highly representative of the population of patients with first-episode psychosis who need specialized psychiatric treatment. Our results show that the length of DUP has a different impact on WHOQOL and GAF/SOFAS scores.
While the cut-off for DUP must be set at close to 1.5 years in order to separate two groups with clearly distinct outcomes regarding quality of life at the end of treatment, the threshold occurs already after only 3 weeks for the GAF and the SOFAS scores. Although attenuated, the same pattern of results is found when restricting analysis to patients with schizophrenia.
Taken together, these results suggest that the impact of DUP on outcome depends on the dimensions that are explored. Indeed, while GAF and the SOFAS are expert rated tools assessing a mix of function and symptoms on an objective way (American Psychiatric Association, 1994), WHOQOL is a self-assessment (WHOQOL GROUP, 1995, 1998 measuring subjective satisfaction of patients regarding various aspects of their life: these two domains seem to be very differently impacted by DUP. Based on this, we can draw two main conclusions. First, when treatment aims are defined, according to the GAF or SOFAS, in terms of reduction of symptoms and return to activity, the impact of treatment delay is very high. If such are the treatment targets, programs should aim at DUP that do not exceed 3 weeks, which is in line with current recommendations for the treatment of first episode psychosis such as NICE for example (NHS, 2015). Second, when it comes to self-assessed quality of life, DUP seems to have a more limited impact. Indeed, while a shorter DUP seems preferable in this domain as well, the outcome difference for a cut-off of up to more than a year remains modest. This observation that only patients with very long DUP have a significantly poorer quality of life than the others at the end of the treatment period needs to be explored in more details. In addition to DUP, premorbid adjustment also seems to be associated to poorer quality of life (MacBeth & Gumley, 2008). We hypothesize that long DUP is linked to factors such as long-lasting social isolation and exposure to various forms of trauma for example, which in turn lead to global and profound deterioration of patients' situation and to conditions where integrated treatment remains ineffective to improve quality of life.
Taken together, these observations suggest that aiming at very short DUP is relevant in order to foster symptoms remission and return to activity (Golay et al., 2016;NHS, 2015). However, assuming that achieving a maximum DUP of 3 weeks could be a major challenge However, general functioning as measured in the present study also captures the impact of symptomatology and provides a broad picture of social and professional functioning which has the secondary advantage of being easier to administer by clinicians (Golay, Ramain, et al., 2021). Third, our measure of quality of life differed from the Kane et al. study (2016) which relied upon the Heinrichs' Quality of Life Scale, a 21-item scale based on a semi-structured interview designed to assess deficit symptoms (1984). Therefore, our quality of life measure could be considered as more subjective. It is likely affected by many non-clinical variables, some of these may be entirely external to the person or their illness. It is also worth noting that quality of life at baseline was not taken into account and is also likely substantially linked to long term outcome. Changes in theses scores should also be a target for measuring the effect of DUP.
Finally, our DUP measure was based on an expert consensus for the psychosis threshold between the clinicians of the specialized early intervention in psychosis program rather than on tools specifically designed to measure DUP. The measurement of DUP could also be dependent of memory of past events and be less reliable when the onset of psychosis is distant in time. Despite reliability of the DUP measure cannot be estimated, these clinicians are experts in the field and have received standardized training allowing us to believe that this measure is robust.

| CONCLUSIONS
DUP seems to have a differential impact on the various aspects of outcome after a first episode of psychosis, such as for example on quality of life and global functional level. Our results suggest in particular that very long DUP has an impact on quality of life while its effect on general functioning 3 years later is already important when it exceeds 3 weeks. While efforts must be pursued to shorten very long DUP in order to guaranty an acceptable quality of life, aiming at the maximum of 3 weeks proposed in current guidelines seems justified in order to allow return to good functional levels.