Patent Application
20070219824
This invention relates, in general, to data processing techniques, and more specifically to analyzing healthcare data.
In the healthcare industry, payers of medical claims face the challenge of identifying and preventing healthcare fraud and abuse. One industry-wide effort relates to establishing rules engines that provide the capability to prevent payment to fraudulent claims. Several analytic tools are available, such as products offered by IBM (FAMS), Fair Isaac, and VIPS, but these products have minimal interoperability, operate very slowly in a traditional reporting fashion and/or require significant configuration, and require an intense effort to modify. Tools like SAS and SPSS provide statistical analysis capability to review very large data sets but they require either a predefined binary target or a test data set in order to derive relevant aberrancy patterns within the claims data. However, such tools may require significant upfront consulting fees, and the accuracy of such tools may be compromised during use.
Additionally, existing systems often focus on identification of aberrations or fraud in the context of claim data associated with an individual patient. However, some types of fraud are difficult to detect on an individual basis. Thus, there is a need for a system that can identify fraud and other claim data aberrations (e.g., billing irregularities) based upon analysis of groups of claim data.
As recognized by the present inventors, what is needed is a method and system for identifying and analyzing patterns of healthcare claims, so that fraud or other patterns of interest may be identified from the healthcare claims. It is against this background that various embodiments of the present invention were developed.
In light of the above and according to one broad aspect of one embodiment of the present invention, disclosed herein is a method for identifying aberrations in a patient's treatment pattern for a medical condition, aberrations in treatment patterns representing treatment of a number of patients for the same medical condition by one or more healthcare providers, or aberrations in patterns based upon other claim data groupings as desired by the user. Aberrations in treatment patterns identified by the system and method may be indicative, for example, of fraud, improper or poor quality of medical treatment, improper billing practices, and/or other undesirable aspects of treatment of the medical condition.
In one embodiment of the present invention, a method for identifying aberrations in treatment pattern for a medical condition includes the steps of: identifying a group of healthcare claim data to be analyzed, wherein the group of healthcare claim data to be analyzed includes claim data representing treatment of the medical condition; generating actual treatment pattern data representing the claim data in the group of healthcare claim data to be analyzed; retrieving defined treatment pattern data for the medical condition; comparing the actual treatment pattern data to the defined treatment pattern data to identify one or more discrepancies therebetween; and generating comparison data representing the results of the comparison. The method may further include the step of displaying the actual treatment pattern data, the defined treatment pattern data, and the comparison data and/or the step of generating the defined treatment pattern data by identifying and compiling claim data relating to treatment of the medical condition from a database of historical claim data.
The actual treatment pattern data may represent a patient's treatment for the medical condition, and the defined treatment pattern data may represent treatment of a group of other patients for the medical condition. Alternatively, the actual treatment pattern data may represent treatment of multiple patients for the medical condition by a healthcare provider, and the defined treatment pattern data may represent an established procedure for treatment of the medical condition. The actual treatment pattern data may include treatment dates, categories of treatment, and a list of treatments provided.
The established procedure may be derived from evidence based medicine and/or from historical claim data analysis.
A system for identifying aberrations in treatment pattern for a medical condition in accordance with the present invention includes a database for storing healthcare claim data and a processor. The processor identifies a group of healthcare claim data to be analyzed, wherein the group of healthcare claim data to be analyzed includes claim data representing treatment of the medical condition; generates actual treatment pattern data representing the claim data in the group of healthcare claim data to be analyzed; retrieves defined treatment pattern data for the medical condition; compares the actual treatment pattern data to the defined treatment pattern data to identify one or more discrepancies therebetween; and generates comparison data representing the results of the comparison.
The features, utilities and advantages of the various embodiments of the invention will be apparent from the following more particular description of embodiments of the invention as illustrated in the accompanying drawings.
Embodiments of the present invention provide for deriving healthcare treatment patterns for a patient from healthcare claims data and displaying such treatment patterns. These treatment patterns can be compared to group treatment patterns derived from healthcare claims data, so that aberrations in the patient's treatment pattern can be identified. These aberrations can help identify fraud, over utilization, or other problems. Moreover, using the patient's treatment patterns, if any combinations of events of interest occurred to the patient (i.e., such as an adverse drug reaction), such events of interest can be located to the extent they may have occurred in other patient's healthcare claims data.
In one example, graphical user interfaces or display screens allow users to quickly drill down on data associated with aberrant billing patterns, identifying claim ID, provider, specialty, regions, state, employer, diagnosis code, procedure code, etc., and any combination of one or more data points. The user interfaces can be used to identify types of trends that exist in the data and target the factors that may be related to such trends. Various embodiments of the present invention are described herein, and may be implemented as methods, systems, apparatus or in other forms.
To create groupings of claim data to be analyzed for patterns and aberrations in accordance with the present invention, claim data may be grouped using any grouping methodology or definition desired by the user. For example, groupings of claim data to be analyzed by the analysis engine may be episode treatment groups (ETGs), which define a grouping of medical episodes of related etiology. A detailed description of the grouping of medical claim data into ETGs is described in U.S. Pat. No. 6,223,164 B1, issued Apr. 24, 2001 to Seare et al., and U.S. Pat. No. 6,370,511 B1, issued Apr. 9, 2002 to Dang, both of which are incorporated herein by reference. Groupings of data may be performed by the medical service provider or by the facility where the service is provided.
Various interfaces may be provided to the analysis engine, including interfaces for customers, interfaces for payers, interfaces for physicians, interfaces for insurance providers, and interfaces for employers, as examples. Each of these interfaces may be implemented over a network, for instance within a browser program of the user for providing remote access to the analysis engine and the data stored in the claims database.
Generally, the analysis engine allows for identification of patterns present in healthcare treatments of individual patients or groups of patients. These patterns have numerous uses as described herein, for instance for comparing a pattern of treatment received by an individual patient against the general pattern of treatment of a group of patients having the same medical condition as the individual patients.
Using the patterns of interest identified in operation 20, subgroups of patients with the same or similar pattern of interest may be identified. In one example operation 30 includes mining the claims database for additional patients which may also have experienced the same or similar pattern as identified by operation 20. In this regard, where a pattern of interest identified in operation 20 includes, for example, an adverse reaction to a medical treatment or a combination of drugs, operation 30 can mine the claims database to determine if other patients have experienced similar adverse reactions.
In operation 40, the data of the subgroups identified by operation 30 can be studied and variables within the data set can be isolated to identified other trends. For instance, where a subgroup of patients has experienced and adverse reaction to a medical treatment or combination of prescribed drugs, operation 40 may examine whether the patients in the subgroups have other traits in common or other statistical analysis may be performed. Hence operations 20-40 can be used to examine whether a pattern of interest that the individual patient experienced was also experienced by other patients whose records are in the claims database.
In another embodiment, in place of operations 20-40, a set of operations can be performed to identify aberrations in the treatment pattern of an individual patient when compared against a group of patients have the same medically diagnosed condition. In one example, a group treatment pattern as received by a group of patients for the medical condition is derived from the healthcare claims data and this group treatment pattern is compared to the first display to one or more discrepancies therebetween. A second display of the group treatment pattern may be provided, and the first display and the second display may be combined to visually display the one or more discrepancies between patterns of treatment.
In step 220, the analysis engine creates pattern data illustrating all treatments given to patients of provider A relating to the treatment of condition X.
In step 230, the analysis engine retrieves from an associated database the applicable predefined pattern of treatment for condition X. The predefined pattern of treatment may be, for example, defined based upon evidence based medicine standards or defined through historical claim analysis. The predefined pattern of treatment for condition X may be defined to take into account factors such as the difficulty of the treatments provided by provider A and/or provider A's casemix. By considering the applicable casemix, provider A's treatment patterns are compared with other healthcare providers that have a similar casemix. By considering the applicable staging indicators (e.g., risk of mortality or severity of illness), provider A's treatment patterns are compared against treatment of condition A at similar levels of difficulty or complications.
In step 240, provider A's treatment patterns are compared against the predefined treatment pattern retrieved by the analysis engine and displayed to the user. Using this method, the user may identify, for example, unusual, unnecessary, and/or excessive treatments routinely provided by provider A in treatment of condition X; unusual, unnecessary, and/or excessive use of facilities by provider A in treatment of condition X; and/or unusual, unnecessary, and/or excessive billing practices by provider A. Identification of such aberrations may be indicative of fraud, poor quality of care, billing irregularities, etc., that may be difficult to detect when examining claim data relating only to one patient being treated by provider A for condition X. The results of this comparison may be used, for example, to generate reports concerning treatment patterns of certain healthcare providers to enable assessment of the performance of the healthcare provider for various purposes. Similar analyses may be performed for multiple medical conditions and/or multiple healthcare providers.
The display layout may also include a list of services received, or representation thereof, which may be arranged to be displayed within each subcategory to which the particular service corresponds. It can be seen that for each data point based on actual healthcare claim data, the data point is plotted to correspond with the date of the treatment, and is also plotted to be displayed within the appropriate category of services provided. Moreover, in one example, for each data point, the diagnosis codes relating to that data point may be displayed, as well as a description or representation of the actual services performed.
In this example, the patient has been categorized in a group of patients having the medical condition of leukemia. By grouping the patients into various groups based on the medical conditions they suffer from, treatment patterns for all such patients in a particular group can be identified, and variations of treatments received by individual patients can be readily identified when compared against the pattern of treatment that other patients in the group receive.
Further, if desired, each data point may include the actual cost associated with the medical service or event that occurred and is represented by that data point.
Using a display layout such as illustrated in
The other services received by this patient during the course of their healthcare treatment are shown in this example
Accordingly, the display layout of
In the example of
The patterns of treatment such as shown in
In the example of
Using the display layout of
Alternatively, a pattern can be introduced into the display layout, and that pattern can be compared to entries within the claims data base. For instance, a pattern of best practices or best treatment practices can be introduced and then used for comparison purposes against other patterns derived from the claims data or compared against treatments received by individuals or by groups of individuals represented within the claims database. Other patterns may include patterns by specialty group, patterns of quality in which actual treatment patterns are compared to predefined treatment patterns to assess the quality of care provided to one or more patients and/or provided by one or more healthcare providers, patterns of physician's treating habits to see how often a physician sees patients, patterns of over utilization and under utilization, and patterns of wellness programs.
Embodiments of the present invention may be used to notify insurance providers and their members of fraud because the display layouts may be used to examine the ways in which the patients are receiving care. For instance, if the diagnosis codes don't support the services or prescriptions being received by a patient, such may indicate fraud.
Embodiments of the invention may also be used as case management tools, because particular patterns of best treatment practices can be matched up with physicians who empirically treat their patients according to such best practices, and hence new patients with such conditions can be referred to those physicians.
For example, employers may use embodiments of the inventions to help design or adjust their medical benefit programs. Employers can perform sensitivity analysis to study potential costs or effects of changes in health care policies.
For instance, if the employer or a payer is proposing to change co-payments for office visits from $15 to $20, the employer or payer can retrieve and study the general access pattern for the groups of employees and determine whether the change in the co-payment amount will result in any cost savings.
By crafting a problem statement and a solution statement that conforms with the taxonomy, embodiments of the present invention can provide statistical analysis of healthcare claims data as well as relevant displays of results related to the particular problem statement formed.
In one example and as shown in
The quantitative elements category may include costs/efficiency, quality/effectiveness, value, and risk. The segment category may include group health, P&C, pharmaceuticals, government, employer. The sub segments may include HMO, PPO, TPA, indemnity, CMS, POS, consumer, auto, DOD, VA, in one example. The entity may include physician, hospital, lab, other professional, other facility, patient, pharmacy, in one example. The component category may include segment, entity type, specialty, geographic, demographic, data element, and derived, in one example. Inherent components may be included in the source data and may form a reference point between the entity category and the view category towards the development of a solution. Derived components may be developed from the source data to create a new level of grouping that is not inherently present in the source data, such as aggregation.
A view category may be provided which may include elements of pharmacy, research, subrogation, fraud, actuarial, claims administration, disease management, pricing and underwriting, network management, clinical operations, customer service administration, claims administration, claims administration, net work and provider management, utilization management, and employer, and one example. The views can represent the prospective of the organization utilizing the software as well as the characteristics of organizations within the industry.
A format category may also be included which may include elements of scorecard, summary, profile, trend, pattern, benchmark, and detail, and one example. The formats address the needs of a specific customer solution. The scorecard format can provide an overview which compares multiple entities within levels. Scorecards may also be used to compare individual or multiple, quantitative segment and entity summary levels in a numeric format. Comparative formats include the summary, profile, trend, pattern, and benchmark formats. The detail format provides the values (source data or derived data) directly.
The taxonomy may also include a market prospective category including elements of payer, patient, and provider in one example. The market perspective elements represent the prospective of the individual viewing the data, and can provide context in terms of the use and display of the data.
As shown in
Hence, it can be seen that embodiments of the present invention provide for identifying aberrations or important patterns in a patient's treatment for a medical condition.
Embodiments of the invention can be embodied in a computer program product. It will be understood that a computer program product including features of the present invention may be created in a computer usable medium (such as a CD-ROM or other medium) having computer readable code embodied therein. The computer usable medium preferably contains a number of computer readable program code devices configured to cause a computer to affect the various functions required to carry out the invention, as herein described.
While the methods disclosed herein have been described and shown with reference to particular operations performed in a particular order, it will be understood that these operations may be combined, sub-divided, or re-ordered to form equivalent methods without departing from the teachings of the present invention. Accordingly, unless specifically indicated herein, the order and grouping of the operations is not a limitation of the present invention.
It should be appreciated that reference throughout this specification to “one embodiment” or “an embodiment” or “one example” or “an example” means that a particular feature, structure or characteristic described in connection with the embodiment may be included, if desired, in at least one embodiment of the present invention. Therefore, it should be appreciated that two or more references to “an embodiment” or “one embodiment” or “an alternative embodiment” or “one example” or “an example” in various portions of this specification are not necessarily all referring to the same embodiment.
Furthermore, the particular features, structures or characteristics may be combined as desired in one or more embodiments of the invention.
It should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed inventions require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment, and each embodiment described herein may contain more than one inventive feature.
While the invention has been particularly shown and described with reference to embodiments thereof, it will be understood by those skilled in the art that various other changes in the form and details may be made without departing from the spirit and scope of the invention.
