1. What is the rationale for centralizing assessments with expert clinical interviewers?

Assessments are the most important component of a clinical trial, establishing the foundation upon which the entire study rests. MedAvante's method improves signal detection for CNS trials by centralizing psychometric assessments in the hands of fewer, expert raters via high resolution video-conferencing. Centralizing the assessments increases the precision, accuracy, reliability and quality of the administration of psychiatric diagnostic and symptom assessments. This can:

1. Reduce variability by reducing the sheer number of raters;
2. Reduce inappropriate patient enrollment due to misaligned incentives because the raters are not associated with the investigative study site;
3. Reduce bias because remote assessors can be blinded to study protocols and visit order sequence;
4. Improve inter-rater reliability and interview quality by using highly trained and finely calibrated expert raters.

Overall, reducing inter-rater variability increases a study's power to detect a difference between the drug and the placebo. The net result: MedAvante's Expert Centralized Raters (comprehensive patents pending) provide pharmaceutical companies with faster, more certain decision-making regarding new CNS drugs.

top

2. What types of industry sponsored studies can MedAvante do?

MedAvante's Expert Centralized Raters are administering the primary efficacy measures in large industry (registration studies) and government sponsored clinical trails programs in depression, anxiety, and (acute in-patient) schizophrenia.

Therapeutic Area: Psychosis, Mood Disorders, Anxiety Disorders and ADHD
Patient Population: Acute in-patient, out-patient, adult and child
Study Objectives: Efficacy, Diagnostic and Instrument validation
Sponsors: Includes 5 of Top 10 CNS Pharmaceutical Companies

top

3. How large is MedAvante?

• Active network of 177 PI sites across North America and accelerating.
• 64 Professionals, 31 expert clinical interviewers with advanced graduate degrees in clinical psychology or social work.

4. How reliable is MedAvante's video-conferencing and assessment scheduling?

The video conferencing platform has proven itself a reliable and stable system. The system has been stress-tested not only by the overall volume of calls, but also by our ability to handle many calls simultaneously. For the most recent 2,000 assessments, greater than 99% were completed without any technical difficulties. Less than 1% of the assessments experienced a technical difficulty and resulted in postponement of the assessment.

Vis-a-vis customer service, we have always been able to accommodate a site's first preference of date and time for a patient interview when provided with the required lead-time.

top

5. Who are MedAvante's Expert Raters?

Includes 31 expert clinical interviewers and continues to grow. These clinical interviewers, all with Master's degrees or PhDs in clinical psychology or social work, have been carefully selected for their expertise in assessing a wide range of psychiatric disorders. They take part in on-going intensive training and are frequently calibrated to a gold standard.

top

6. What are the Intra-Class Coefficients for MedAvante's Expert Raters?

ICCs achieved using a stringent approach: independent interviews. Each MedAvante Expert Rater rated the patient independently; raters were blind to the other rater's interview and scores. Interviews were done on the same day, with the patient doing a distracting task in between interviews to minimize memory effects. The MADRS sample (conducted by MedAvante's head of Research, Ken Kobak, and Janet B. W. Williams) consisted of 120 interviews (60 pairs) utilizing 6 MedAvante Expert Raters, all of whom were paired in all possible combination of rater pairs. This compares of course with the standard rating of video-tapes which artificially inflates ICC due to lack of information variance.

top

7. Who is responsible for supervising your raters?

Janet B. W. Williams, DSW, Vice President of Clinical Development works in close collaboration with Dr. Ken Kobak, MedAvante's VP of Research, and MedAvante's senior scientists. This leadership team includes thought leaders in psychological assessment and clinical trials from academia and the leading global pharmaceuticals. Each MedAvante senior scientist has engaged in funded research examining improved assessment methodologies, has authored psychiatric scales, or has led efforts to update current scales.

Dr. Williams joined MedAvante full-time in July 2007 as Vice President of Clinical Development. In her previous 32 year tenure at Columbia University College of Physicians and Surgeons, and at the New York State Psychiatric Institute, Dr.  Williams was Professor of Clinical Psychiatric Social Work in the Departments of Psychiatry and Neurology (Columbia), and Research Scientist and Deputy Chief of the Biometrics Research Department (Psychiatric Institute).  Originally focused on schizophrenia and related psychotic disorders, her later work concentrated on the development of psychiatric classifications and instruments to measure psychopathology. She is well-known for her interview guides for the Hamilton Rating Scales (SIGH-D, SIGH-A, SIGH-AD, SIGH-ADS) and the Montgomery-Asberg Depression Rating Scale (SIGMA,  co-authored with Ken Kobak, PhD).  She was made an Honorary Fellow of the American Psychiatric Association for her contributions to the development of the third revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III), and she had leading roles in the development of DSM-III-R and DSM-IV we all.  She also co-authored the Structured Clinical Interview for DSM-IV (SCID), a standard in the field for diagnostic evaluations. She collaborated on the development and testing of the PRIME-MD, an interview guide designed to help primary care physicians make mental disorder diagnoses, and its self-report version, the Patient Health Questionnaire (PHQ).  The PHQ and one of its derivatives for measuring anxiety (GAD-7) were cited by Medscape as two of the Top Ten most popular news stories of 2006. 

Dr. Williams is the author of many rating instruments and interview guides, and over 230 scholarly publications. She serves on the editorial boards of several psychiatric and social work journals, and is an active consultant to clinical trials. Dr. Williams holds a BS in biology from Tufts University, an MS in marine biology from the University of Massachusetts Dartmouth, and an MS and DSW in social welfare from the Columbia University School of Social Work. In 1994 Dr. Williams founded the Society for Social Work and Research (SSWR; now over 1300 members) and served as its President for two years. In 1999 she was inducted into the Columbia University School of Social Work Alumni Association Hall of Fame, and in 2000 she received the Lifetime Achievement Award from SSWR.  Since 2002 Dr. Williams has been listed as an ISI Highly-Cited Researcher in Psychology/Psychiatry, a designation awarded to the top 250 preeminent individual researchers in each of 21 subject categories in life sciences, medicine, physical sciences, engineering and social sciences, who have demonstrated great influence in their field as measured by citations to their work.   In 2005, Dr. Williams was awarded the Knee/Wittman Lifetime Achievement Award in Health & Mental Health Policy and Practice by the National Association of Social Workers Foundation.

top

8. What scales can you conduct?

Our expert raters are experienced in administering the most commonly used rating scales in clinical trials, including but not limited to the HAMD, MADRS, HAMA, LSAS, Y-BOCS, PANSS, PSP, SAPS, Adult ADHD Clinician Diagnostic Scale (ACDS), The Child Depression Rating Scale (CDRS), CGI-S, QOLS, CAARS, CDRS-R, ADAS, as well as diagnostic interviews, such as the SCID and MINI.

Our expert raters are also administering semi-structured interviews including the SIGH-D (HAMD), SIGH-A (HAMA), SIGH-ADS (combined HAMD and HAMA), SIGMA (MADRS), SCI-PANSS (PANSS), etc.

top

9. What is the FDA's position?

According to Tom Laughren speaking on behalf of himself, Russ Katz, and Paul Andreason:

“Regarding the general matter of remote centralized ratings, we have no objection to this practice, and agree that this should increase the precision of assessments in a trial.”

August 04, 2004

top

10. What is the impact of centralized ratings on the power of the study?

Centralized ratings increase study power in four ways: by decreasing the sheer number of raters; by improving the reliability between the smaller group of raters used; by improving the quality of the interviews; and reducing bias by enabling blinding to study visit and use of different raters at baseline and endpoint.

• Reducing the sheer number of raters alone should decrease variability and increase power (a typical 100 site CNS study may have as many as 200-300 raters, while the same trial could hypothetically be conducted with 20-30 centralized raters).
• Reliability training currently done at startup meetings is typically inadequate. The rigorous methodology utilized by the centralized raters insures high ongoing levels of calibration.
• The high quality of the ratings performed should increase effect size. Recent data has found that higher interview quality is associated with larger effect sizes.
• Recent studies have also shown that using different raters at baseline and endpoint increases signal detection. Using raters who are also blinded to study visit should increase this effect.

Additionally, poor reliability decreases the statistical power of the clinical trial, resulting in the need for larger sample sizes to detect significant differences between the active drug and a placebo or between any active treatments. For example, a study having an assessment reliability that falls from 1.0 to .80 drops in power from .80 to .71 and requires 25% more subjects to detect a significant difference (Muller & Szegidi, 2002).

top