R version 4

Comparison of MF, MB, and VAF across groups of individuals was performed by Mann–Whitney U test. Correlations were tested with Spearman rank test. Associations between categorical variables were tested with Fisher exact test. Two logistic regression models were constructed, one including the standard ovarian cancer risk variables (age and CA-125) and an exploratory model including age, CA-125, and TP53 MB. The models equated the relationship between variables with the occurrence of ovarian cancer to estimate beta coefficients with 95% confidence intervals. Because of a heavy right-tailed distribution, CA-125 was log transformed. Age and TP53 MB were presented as a per SD increase. All tests were two sided at an alpha level (type 1 error rate) of 0.05. Statistical analyses were performed with SPSS version 26 (31 ), R version 4.1.1 (28 ), and Stata 16 (32 ).

Baseline characteristics were summarized using measures of central tendency and variability. The association between diagnosis and time to relapse and time to discontinuation on PP3M and then PPLAI was examined with use of Cox proportional hazard models.
All regression models were adjusted for age, inpatient (Y/N), sex, polypharmacy, ethnicity, and dose, prior use of clozapine. Model assumptions were examined using Scholfeld’s residuals and plots for each variable in the model and overall. Adjusted cumulative hazard plots by F20 diagnosis were constructed at the mean of model covariates plots. Statistical analysis was conducted in Stata version 15.0 (StataCorp LLC, College Station, TX) and R version 4.0.2. (R Foundation for Statistical Computing, Vienna, Austria).

We conducted regression analyses to evaluate the association between disease status and COVID-19 behaviors or attitudes, as well as the change in attitudes by disease from survey 1 to survey 2 using R version 4.0.2 and STATA version 17 (18 ), controlling for the demographics listed above. For risk perceptions, pandemic fatigue, individual risk mitigation behaviors, approval of community-level NPIs, and vaccine/booster willingness we conducted OLS regressions. For vaccine attitudes, we conducted logistic regression. To test whether the slopes of the regression coefficients for “COVID-19 is a threat to me” and “COVID-19 is a threat to the public” were significantly different from one another we used seemingly unrelated regression to account for possible correlation of the equation errors (using the “systemfit” package) (19 ) and tested the linear hypothesis using an asymptotic Chi-square test (“car” package) (20 ). To assess the effect of chronic disease status on change in outcome from survey 1 to survey 2, we conducted OLS regression controlling for survey 1 response, disease status, and the factors above to predict survey 2 response among participants who completed both surveys. Results are presented as linear regression coefficients or odds ratios and 95% confidence intervals, and predicted values and standard error for the change in response between surveys.

First, the pre-treatment change in the RPR titer was illustrated using the ratio of RPR titer at treatment initiation/diagnosis. Second, we assessed whether the pre-treatment change in the RPR titer (i.e., the ratio of RPR measurements at treatment initiation/diagnosis) could affect the slope of the post-treatment relative change in the RPR titer. The ratio (i.e., relative change) of the RPR titer at post-treatment/treatment initiation was used for the analysis. The averaged monthly change in the RPR ratio (i.e., speed of the decrease in RPR) after treatment was calculated on the log2 scale for each patient by using the slope of linear approximation (i.e., log-linear was assumed). The mean of the monthly change in the RPR ratio was compared between the groups on the log2 scale using the Student’s t-test and antilog was used for describing the mean and 95% Confidence Intervals (CI). Univariable and multivariable linear regression analyses were conducted to assess the variables that affect the monthly change in RPR titer ratio. Variable selection was conducted by backward elimination using the P-values obtained with the Wald test. Statistical significance was defined as a 2-sided P-value <0.05. For highly correlated variables, one value was chosen based on clinical importance. Data analysis was performed using R version 4.0.2 and Stata MP 16.1 (StataCorp, TX, USA).