Advanced Healthcare Analytics
Data analytics services in healthcare involve comprehensive solutions that integrate healthcare data and the use of large data sets to drive actionable insight.
Over the past 20 years, our analytics team has been involved in thousands of projects with medical practices geared towards improving efficiency, profitability and compliance. Clients have ranged from solo docs to practices with over 1,000 physicians, academic medical centers (AMC’s), oncology clinics and national associations such as the Medical Group Management Association (MGMA) and the American Medical Association (AMA).
What We Do
DoctorsManagement Business Analytics division has worked with physicians and practices in nearly 60 different specialties and within every state in the U.S. Our Advanced Healthcare Analytics division specializes in litigation support and post-audit analyses, working with healthcare organizations, state’s attorneys general and private legal firms to aggressively defend physician’s rights against payers and regulating agencies.
Recovery audits of all types have increased to such a degree that they are now more the rule than the exception. Our experts focus their time and effort on the quantitative component to ensure audits were conducted in accordance with accepted and standard statistical and analytical techniques.
DM Analytics Project Examples
Process improvement projects have included such areas as scheduling, wait time, check-in, patient throughput, revenue cycle and many other related areas. Our analytical work has included a wide range of diverse projects including:
- Advanced Healthcare Analytics
- Litigation Support
- Post-Audit Analysis
- Fee Scheduling
- Compliance Risk Analysis
- Provider Performance
- Payer Analysis
- Revenue Cycle
Defending an Audit
Offensively, you want to be able to analyze the potential risk for the audit, seeing yourself the way that the auditor does. Defensively, you want to be able to analyze the results in such a way as to minimize the damage from overpayment estimates. Defending an audit falls into two distinct categories; qualitative and quantitative.
- The Qualitative Review looks at the reason for the overpayment determination and when appropriate, challenges those reasons based on accepted coding techniques and standards of medical practice. This normally requires the services of a coding and/or billing specialist.
- The Quantitative Component looks at the methodologies, calculations and statistical applications of the data and this is where we come in, particularly when extrapolation is involved.
According to CMS Publication 100-08, Chapter 3, Section 10.2, there are six distinct components that are used to conduct a valid extrapolation analysis. They are:
- Defining the universe
- Defining the sample frame
- Defining the sampling units
- Using proper randomization
- Accurately measuring the variables
- Using the correct formula for estimation
As part of our engagement, we will critically analyze each of the components above as a deficiency in any one would likely result in either setting aside the extrapolation, a reduction in the amount demanded or, in some cases, disqualifying the audit altogether.
STATISTICAL VALIDATION OF THE RANDOM SAMPLE
A statistically valid random sample (SVRS) is one in which every unit within the universe has an equal opportunity to be selected. Quite often, we do, in fact, see that, by this definition, the audit does consist of a SVRS. Often, however, a problem arises whereby the universe itself is defined poorly.
Random sampling involves a lot more than a set of random numbers and simple testing of paid claim amounts. In order for the sample to be truly random, the universe from which the sample is pulled needs to exhibit a high degree of homogeneity, without which, it is nearly impossible to get a true random sample. Using advanced statistical techniques, we analyze the data in several vectors in order to determine whether the sample is truly random, going well beyond what most auditors would even consider.
DoctorsManagement assists organizations in addressing complex issues using data-driven actions in order to provide insight to making decisions. We offer custom solutions to our clients and work with your data, not simply benchmarking data.