Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.
Helping Organizations Improve Study Planning andExecution Using Advanced Analytics for Data-Driven Decision MakingJuergen ...
Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients ...
The National Children’s Study – an Enterprise ofunprecedented size and complexity• Congress authorized the National Instit...
Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients ...
Modeling and Simulation as Analytic Techniques that formthe Foundation for Data-Driven Decision Making• Booz Allen conduct...
Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients ...
Select Analytic Techniques are combined into a decisionsupport framework• We are using a Markov models, implemented as    ...
Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients ...
Key planning questions could be answered• Sample planning questions studied and answered   • What boost strategies can be ...
Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients ...
Helping Booz Allen’s Clients be Ready for What’s Next• The novel data-driven, evidence-based approach to study planning ha...
Learn More about our Advanced Analytic Capabilities              www.boozallen.com/analytics                Juergen A. Kle...
Prochain SlideShare
Chargement dans…5
×

Using Advanced Analytics for Data-Driven Decision Making

3 836 vues

Publié le

Helping Organizations Improve Study Planning and Execution

Publié dans : Business, Technologie
  • Soyez le premier à commenter

Using Advanced Analytics for Data-Driven Decision Making

  1. 1. Helping Organizations Improve Study Planning andExecution Using Advanced Analytics for Data-Driven Decision MakingJuergen A. Klenk, PhDPrincipal, Health AnalyticsMarch 2012
  2. 2. Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients be Ready for What’s Next 2
  3. 3. The National Children’s Study – an Enterprise ofunprecedented size and complexity• Congress authorized the National Institutes of Health (NIH) with the Children’s Health Act of 2000 to undertake the National Children’s Study (NCS)• Focus of the NCS is to examine the effects of the environment on growth, development, and health of children• The NCS attempts to enroll and follow at least 100,000 and up to 250,000 women and their children, from before birth to age 21, at a minimum of 100 and up to 175 study locations across the United States• The NCS is expected to generate data that form the basis of child health guidance, interventions, and policy for generations to come• NIH saw an opportunity to pioneer novel techniques for data- driven, evidence-based study planning, to manage the daunting task of planning the execution of the NCS and maximize chances of success
  4. 4. Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients be Ready for What’s Next 4
  5. 5. Modeling and Simulation as Analytic Techniques that formthe Foundation for Data-Driven Decision Making• Booz Allen conducted research to identify the best analytical approach to developing a data-driven planning and decision making framework• Key analytic techniques employed include • Markov Models • Stochastic Models • Predictive Modeling • Discrete Event Simulation• The resulting model was calibrated against real-world data and used to evaluate alternative scenarios of study operations• Data needs could be met by the Vanguard pre-study to the NCS, and specific attention was given to data understanding, preparation, and QA/QC
  6. 6. Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients be Ready for What’s Next 6
  7. 7. Select Analytic Techniques are combined into a decisionsupport framework• We are using a Markov models, implemented as Resource Independent Transition stochastic models using a discrete event simulation Resource Dependent Transition Active State tool Final State• Potential participants are represented Sampled Enumeration as populations in various states• Dynamics are determined Not Screen Screen Eligible by the rates of transition Eligible from one state to another Follow up Not Pregnancy• Transition rates are (Not Preg.) Eligible affected by pregnancy Follow up rates, consent (Preg.) rates, allocation of staff Consent Eligible resources, etc., and are implemented as predictive Not Consenting, Not Consenting, models Consenting Soft Hard
  8. 8. Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients be Ready for What’s Next 8
  9. 9. Key planning questions could be answered• Sample planning questions studied and answered • What boost strategies can be employed to shorten the recruitment period, and how does this affect overall cost • What data are required to effectively monitor study operations, and how can these data be collected in sufficient quantity and quality • What sampling unit sizes for study locations are optimal to achieve recruitment targets as quickly as possible, and at the lowest cost• One of the most important questions answered was to calculate the number of participants required to successfully conduct the Study: 250,000• The NCS Main Study is scheduled to launch in 2013• The same innovative approach will now be applied to plan beyond the recruitment phase, to include planning factors such as retention, compliance, study operations, logistics, and cost
  10. 10. Table of contents• The Challenge• Our Solution• Analytic Techniques• Results & Applications• Helping Booz Allen’s Clients be Ready for What’s Next 10
  11. 11. Helping Booz Allen’s Clients be Ready for What’s Next• The novel data-driven, evidence-based approach to study planning has helped NIH to successfully plan for and get ready to launch one of its most important enterprises, the NCS• The novel approach of data-driven planning is viewed as a way to plan and conduct studies in the 21st century• The approach, developed for study operations planning, could be extended to scientific planning• This work helps NIH be ready to leverage critical information contained in large quantities of data for study planning, and thus accomplish a transformation to a data-driven decision making paradigm
  12. 12. Learn More about our Advanced Analytic Capabilities www.boozallen.com/analytics Juergen A. Klenk, PhD Principal, Health Analytics klenk_juergen@bah.com Phone (703/377-7205)

×