P.508 COVID-19 impact on entry symptom severity in schizophrenia clinical trials – preliminary data

Introduction: The COVID-19 pandemic is expected to have wide ranging impact on schizophrenia clinical trials. Changes in symptom severity and expression might occur due to increased stress, fear of illness, diminution of contact with research site and other psychiatric personnel and enforced social distancing. In the current project we intend to analyze individual symptom severity as assessed by the PANSS [1] data collected in the screening period in ongoing clinical trials in schizophrenia that started before the COVID-19 pandemic and assesses whether COVID-19 has an impact on the severity of symptoms as well as the variability of the data.

Methods: Data are collected from currently ongoing schizophrenia clinical trials that started before the COVID-19 outbreak. To account for different protocol designs data collected in the screening period are analyzed separately by each clinical trial. We have arbitrarily set the date of the first confirmed COVID-19 case within each country as the reference date and treated all data collected after this date as data affected by the COVID-19 pandemic. On an item level we utilize Wilcoxon rank-sum test for equality of distributions, K-sample equality-of-medians test and Levene's robust test statistic for the equality of variances. Given the exploratory type of analysis we consider the data as potentially impacted by COVID-19 if at least one of the ‘distribution’ tests or the variance test with the median location estimator indicates statistical difference in the data collected after the reference date.

Results: Our dataset currently consists 1254 datapoints collected in the screening period, 1086(88%) collected before and 168(12%) collected after the reference date. The plan is to increase the dataset as additional data become available. In data coming from acute studies (399 data points, 114(29%) collected after reference date) significantly higher severity was observed in the data collected after the reference date for the following items: P3, G2, G4, and G16; significantly lower severity for items P2, P4, N7, G10 and G12. Variability was increased for items N4 and G16 and decreased for items P2, P3, G2 and G10. In data coming from studies in subjects with predominantly negative symptoms (855 data points, 54(6%) collected after reference date) significantly higher severity was observed in the data collected after the reference date for items N1 and N6, and significantly lower for items P3, P4, G1 and G14. Variability was increased for item G1 and decreased for item P4.

Discussion: Our preliminary analyses indicate that COVID-19 pandemic has a significant impact on a wide span of symptoms in schizophrenia clinical trial participants. Some effects, such as increase in anxiety, tension or active social avoidance, are within expectations, some, e.g. decrease in conceptual disorganization, are unexpected. The extent of possible impact on trial outcomes is yet to be determined, our limited dataset did not allow for analysis of change data at this point. We plan to increase the sample size and the spectrum of the analyses for the actual presentation as more data become available.

Disclosure statement: Both authors are full time employees of Signant Health. The abstract is financially supported by Signant Health.