The World Statistics Day celebration continues here in the Community. We all need reliable data for sound decision making. Do you have a data source that you trust most? Head over to Discussions to tell us about it.
Choose Language Hide Translation Bar
Staff (Retired)
7 things to love about JMP Clinical 5.0

New versions of JMP Clinical and Genomics are available starting today, so I wanted to take the opportunity to give a brief overview of some of the new features you’ll come to enjoy with the new release of JMP Clinical 5.0. Below are seven things to love!

1. Risk-Based Monitoring (RBM). If you’ve been following my posts, features for RBM should come as no surprise. If you need to catch up, you can do so here. This new functionality for RBM was developed using the recommendations of TransCelerate BioPharma, so you can be assured that the current implementation should meet the needs of your company.

2. Fraud Detection. Who doesn’t relish the opportunity to play Sherlock Holmes? New and updated features include:

  • Clustering Subjects Across Study Sites. This analysis can help identify patients who have enrolled at two or more study sites within the same trial, or in multiple studies within the same development program. If basic subject information like initials or birth date is unavailable, this analysis allows you to identify subjects who are overly similar based on pre-dosing data. Zero-in on interesting pairs of subjects with matching gender and race, while allowing for minor differences in age, height and weight.
  • Weekdays and Holidays. Previous functionality identified holidays common to the U.S. and Canada. Though some holidays are celebrated globally, new features allow the user to define custom holidays or events (such as severe weather events) that may interrupt normal business activity at clinical sites, taking into account the country of the sites.
  • Perfect Scheduled Attendance. A new screening feature helps identify the particular site-visit combinations that appear unusual. You can identify sites where the patient attendance appears too good to be true or sites with severe scheduling delays.
  • Figure 1. Digit Preference Volcano Plot

    • Digit Preference. This analysis helps identify any differences in the distribution of the trailing digit between sites for all Findings domains (Figure 1). A screening feature helps identify the particular site-visit-test combinations that appear unusual. This can help identify sites that may tend to round analysis values, improperly conducted procedures (e.g., taking blood pressure manually in lieu of using an automated blood pressure cuff), improperly calibrated equipment, or important differences in subjective measurements (such as reporting clinical signs using a Likert scale), which could suggest that additional training is needed.
    • Figure 2. Hierarchy of Billiary Disorders SMQs

      3. Standardised MedDRA Queries (SMQs). Using your current version of the MedDRA dictionary, JMP Clinical identifies occurrences of SMQs using broad, narrow or algorithm criteria. It will summarize findings in histograms and diagrams (Figure 2), and conduct incidence analyses. Further functionality allows the user to identify which preferred or lower-level terms contributed to the SMQs.

      4. Predictive Modeling. A Predictive Modeling Review (Figure 3) feature enables the analyst to quickly drag and drop models (with the ability to tune various options) to define a set of predictive models for testing. With cross-validation and learning curve techniques, users can easily identify the most useful form of the predictive model, identify important covariates and limit problems due to overfitting.

      Figure 3. Predictive Modeling Review Builder

      5. Subgroup Identification. A new subgroup analysis menu enables users to identify subgroups with enhanced treatment response (or excess safety risk) using either the prune-as-you-go interaction tree or Virtual Twins algorithms.

      6. Review Builder. Similar to the Predictive Modeling Review, the Review Builder (Figure 4) enables the analyst to quickly drag and drop reports to define a set of analyses (with the ability to tune various options) that can be run in rapid succession each time the study database is updated. You can easily apply these reviews to other studies or modify them to address any changes required due to design, endpoints or options.

      Figure 4. Clinical Review Builder

      7. Patient Profiles. Due to popular demand, we have added a tabular display to our patient profiles. Users can customize which columns are summarized, the sort order of the rows, and save these tables to PDF or RTF reports.

      As you can see, we’ve been busy here at JMP Life Sciences! We've been working on ways to help you understand and reduce safety and quality risks in your clinical trials, more easily predict important safety and efficacy outcomes and generate clinical reviews, and identify subgroups that may potentially be of greater interest. You can expect a deeper dive for many of these features in the weeks to come.