US20100100395A1 - Method for high-risk member identification - Google Patents

Abstract

A method for using claims data to identify high-risk members of a healthcare plan is disclosed. In one embodiment, the present invention includes searching the plurality of claims of members selected by a filtering criteria to identify the presence of an intervention flag, to identify factors influencing care intervention, and identifying a medical episode driving the member's cost. In one embodiment, the intervention flag the medical episode are displayed in association with an identification of the member. In another embodiment, the member selection criteria includes selection based on the member's predicted future health care utilization.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S)
• This application is a Continuation of U.S. patent application Ser. No. 09/733,215, filed Dec. 8, 2000, content of which is incorporated herein by reference.
BACKGROUND OF THE INVENTION
• The present invention relates to a system and method for identifying high-risk members in a healthcare plan. More particularly, it relates to a system and method for using information from physician claims, facility claims, and pharmacy claims to identify high-risk members in a healthcare plan and to provide salient information on those members to an intervention agent.
• For the past few decades, the predominant model of healthcare management used by most health care plans in the managed care industry has been one that focuses primarily on approving or denying coverage for medical procedures based upon specially developed criteria. This system has been subject to some criticism from doctors who feel that their treatment decisions should not be questioned, and from patients who feel that their health care plan places undue emphasis on financial consequences at the expense of sound medical care. Further, the current model employed by health care plans fails to place the appropriate amount of emphasis on proactive care. Studies have shown that an emphasis on proactive care can improve a health care plan member's overall health and well-being. Proactive care can also reduce the overall expense to a health care plan by replacing expensive medical procedures and treatments with less expensive proactive care activities.
• Finally, under the current model, plan members commonly have minimal or no positive contact with their health care plan. Members pay health care premiums individually or through their employer and hope that, when treatment is needed, the health care plan will approve coverage. There is a need in the art for a health care management model that addresses the above shortcomings of the predominant current model. To implement a proactive-care-based or interventional model that allows a healthcare plan to take a proactive approach to providing health care to its members, it is necessary to have an effective system of identifying high-risk plan members or identifying plan members that are amenable to intervention (i.e., those members who can be helped with proactive or interventional care) and compiling relevant information regarding those members.
• When a member of a health care plan receives care from health care providers, information regarding the care received is provided to plan administrators in documents commonly referred to as claims. Predominantly, this information is provided in the following three types of claims: physician claims, facility claims, and pharmacy claims. These claims are the documents that are submitted to the health care plan by physicians, hospitals, and pharmacies to receive reimbursement for care provided to the plan member. These documents generally contain coded data that provides information regarding the care received by the plan member. These claims are processed by the health care plan, and where appropriate, payment is transmitted to the health care provider.
• For purposes of this specification, the phrase “physician claim” is used to refer to any professional service claim submitted to a health plan, typically on an HCFA-1500 form or its equivalent, and the phrase “facility claim” is used to refer to any facility claim. The phrase “medical claim” is used to refer to both physician claims and facility claims. Finally, the phrase “pharmacy claim” is used to refer to any claim submitted by a pharmacy or durable medical goods provider. Medical claims generally include codes for diagnoses and procedures relating to the plan member. The reason for the visit is typically represented by an International Classification of Diseases (“ICD”) code, currently in its ninth revision and thus commonly referred to as “ICD-9.” The description of the service provided in a medical claim typically takes one of two formats, a Common Procedural Terminology (“CPT”) code (promulgated by the American Medical Association), or a Health Care Procedural Code (“HCPC”) (promulgated by the Health Care Financing Administration).
• The following materials serve as background for the present application and provide further information on some of the classification systems discussed:Physician's Current Procedure Terminology, CPT '94, published by the American Medical Association;HCPCS1994Medicare's National Level II Codes, published by Medicode, Inc.;Med-Index ICD9CM Fourth Edition1993, published by Med-Index, each of which is hereby incorporated by reference in its entirety.
BRIEF SUMMARY OF THE INVENTION
• In one embodiment, the present invention is a method for identifying intervention flags, from a plurality of claims, for a member of a healthcare plan. The method includes searching the plurality of claims to identify the presence of an intervention flag and generating a display showing the intervention flag in association with an identification of the member.
• Still other embodiments of the present invention will become apparent to those skilled in the art from the following detailed description, wherein is shown and described only the embodiments of the invention by way of illustration of the best modes contemplated for carrying out the invention. As will be realized, the invention is capable of modification in various obvious aspects, all without departing from the spirit and scope of the present invention. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not restrictive.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a block diagram illustrating the high-risk member identification method, according to one embodiment of the present invention.
• FIG. 2 shows an exemplary screen display of a member selection screen, according to one embodiment of the present invention.
• FIG. 3 shows an exemplary screen display of a member assignment screen, according to one embodiment of the present invention.
• FIG. 4 shows an exemplary screen display of a member information screen, according to one embodiment of the present invention.
• FIG. 5 shows an exemplary screen display of an intervenability factors screen for the current member, according to one embodiment of the present invention.
• FIG. 6 shows an exemplary screen display of an episode class intervention screen for the current member, according to one embodiment of the present invention.
• FIG. 7 shows an exemplary screen display of a non-compliant drug categories screen, according to one embodiment of the present invention.
• FIG. 8 shows an exemplary screen display of a behavioral risk factors screen, according to one embodiment of the present invention.
• FIG. 9 shows an exemplary screen display of a drug history report screen, according to one embodiment of the present invention.
DETAILED DESCRIPTION
• FIG. 1 is a block diagram illustrating the high-risk member identification method of the present invention. As shown inFIG. 1, generally the high-riskmember identification method10 includes filtering members of a group for processing (block12), creating a database containing information for each plan member (block14), determining which intervenability factors are present for a member and determining the cost drivers for the member (block16), selecting members based on a selection criterion (block18), and transferring relevant plan member information to a member intervention application for display to an intervention agent (block20). The term “intervention agent” is used throughout this specification to refer to the individual that examines the information presented in the display for a given member and attempts to proactively intervene in that member's medical care. Each of these aspects of the present invention will be described in further detail below.
• As shown inbock12, in one embodiment, the high-riskmember identification method10 involves filtering members from a group or plan for processing. In other words the entire group is filtered according to a filter criterion to generate a subset of the group for processing by the present method. In one embodiment of the present invention, the members are filtered based on exceeding a threshold for actual or predicted future healthcare resource utilization. In another embodiment, the top five percent of the group members are chosen for further analysis. In another embodiment of the present invention, where a relative risk is calculated for each member of the group, the members are filtered based on exceeding a threshold for relative risk. In another embodiment, the top five percent of the group members by relative risk are chosen for further analysis. In another embodiment, any given member is chosen for any reason for further analysis. For example, a given group member may request that his or her claims history be examined for potential intervention flags. In another embodiment of the present invention, all members of the group are chosen for further analysis.
• As shown inblock14, information is then collected on each chosen member's medical diagnoses and health care utilization patterns over a specified time period. In one embodiment, this information is collected or compiled from claims data, including physician claims, facility claims, and pharmacy claims. The specified time period over which this information is collected for each member of a health care plan is referred to as the “focus period.” In one embodiment, the focus period is one year. In other embodiments, the focus period may be longer or shorter.
• As explained above, medical claims typically include information on medical diagnoses and treatments provided to the member. Typical fields included in physician claims, facility claims, and pharmacy claims are generally known to those of skill in the art. In one embodiment of the present invention, the important fields that are extracted from a physician claim, to generate the database, include the date of service, the physician provider identification, the reason for the visit, and a description of the service or services provided. In one embodiment of the present invention, the important fields in a facility claim include the date of service, the facility provider identification, the reason for the visit, and a description of the services provided. In one embodiment of the present invention, the important fields in a pharmacy claim include the date of fill, the pharmacy provider identification, the prescribing physician provider identification, and the description of the medication, generally in the form of a National Drug Code (“NDC”).
• One exemplary specification for extracting information from the claims data for each member of the health care plan during the focus period is described in U.S. patent application Ser. No. 09/635,911, now U.S. Pat. No. 7,444,291, entitled “System and Method for Modeling of Health Care Utilization,” by Gerald L. Lutgen et al., filed Aug. 10, 2000, which is hereby incorporated by reference in its entirety. In other embodiments of the present invention, other specifications known to those of skill in the art are used to extract relevant data or information from the medical and pharmacy claims for the focus period to compile the database.
• After the members are filtered and the database is compiled, the claims for each of the chosen members are analyzed to determine the number of intervenability factors or flags present for each plan member (block16). This is performed at this point in the method, because, in one embodiment, the number of intervenability factors is used as one of the selection criteria described below (block18). Also, performing this determination prior to selecting the intervention set reduces the time needed to generate the information display (block20), because the necessary processing is performed in advance. In another embodiment, the intervenability factors are not determined prior to selecting (block18) and thus are not used as one of the selection criteria.
• In one embodiment of the present invention, the top medical episodes driving risk for each plan member are also determined at this point in the method (block16). Again, this is done to reduce the time needed to generate the information display (block20). In another embodiment, the top medical episodes driving risk are calculated “on-the-fly” or during generation of the information display (block20). The technique for determining the top medical episodes driving risk is described in more detail below. In one embodiment, all other information generated for each plan member (as shown inFIG. 4) is generated on-the-fly, because it can be ascertained quickly and this approach eliminates the need to generate this information for members eliminated during the selection of the intervention set (block18).
• Intervenability factors are those areas or aspects of a member's care history or patterns that are amenable to intervention by an intervention agent. In other words, they are areas that can be readily identified using claims data, can be corrected or improved, and that are likely to result in a reduction of health care costs or improve the overall health and well-being of the member. In one embodiment of the present invention, the intervenable factors that are assessed for each plan member, to determine if they are present, include: (1) whether the member visited an emergency room during the focus period, (2) whether the member had any in-patient hospital admissions during the focus period, (3) whether the member incurred any out-of-network costs during the focus period, (4) whether the member visited more than three different provider specialists during the focus period, (5) whether the member was prescribed multiple pharmaceuticals during the focus period, (6) whether the member has no appropriate provider for a chronic episode during the focus period, (7) whether the member missed a targeted intervention during the focus period, and (8) whether the member is failing to comply with taking mediations prescribed during the focus period. The claims for a given plan member, in the database, are analyzed to ascertain the presence or absence of each of the intervenability factors. The total number of intervenability factors present for a member is recorded. Each of these intervention factors, and how their presence is determined, is described in more detail below. In other embodiments of the present invention, other intervenability factors are assessed for each plan member. In another embodiment, only one of the intervenability factors identified above is assessed.
• In one embodiment of the present invention, a relative risk is calculated and assigned to each selected member. In this embodiment, the relative risk is used as one of the criteria for selecting the intervention set (block18). In another embodiment, relative risk is used as the basis for making the initial selection, as described inblock12 ofFIG. 1. The relative risk ranking is a function of the predicted future health care resource utilization for the member. The predicted future health care utilization is calculated from the claims data in the database. In one embodiment of the present invention, the cost of services provided to the member during the focus period is calculated. This number is then used as the marker for future health care utilization. In another embodiment of the present invention, the predicted future health care utilization is calculated by looking at only the claims information from the medical claims and computing a total cost of services provided during the focus period. In another embodiment of the present invention, future utilization is calculated by examining only data from the pharmacy claims and computing the total cost of pharmacy services provided during the focus period. In another embodiment of the present invention, future utilization is calculated based on chronic medical conditions only for the member during the focus period. In one embodiment of the present invention, the predicted future health care utilization is calculated using one of the techniques disclosed in U.S. patent application Ser. No. 09/635,911, now U.S. Pat. No. 7,444,291, entitled “System and Method for Modeling of Health Care Utilization,” identified above and hereby incorporated by reference in its entirety.
• In one embodiment, after the predicted future health care utilization cost is calculated, each member is assigned a relative risk, which is that particular member's predicted future health care utilization or cost, divided by the average for the entire group. In other words, the relative risk is basically a number representing a particular member's predicted future health care utilization in relation to the average for the group. In other words, if a given member's relative risk is two, his predicted future health care utilization is twice that of the average.
• In one embodiment of the present invention, once a relative risk is assigned to each plan member, the relative risk ranking can be assigned. The relative risk ranking is assigned to each of the selected plan members by assigning the member with the highest relative risk a ranking of “1,” the member with the next highest relative risk a ranking of “2,” and so on, until every selected member has been assigned a relative risk ranking. In one embodiment, the relative risk ranking serves as yet another criteria for filtering.
• As shown inblock18 ofFIG. 1, an intervention set of plan members is generated by selecting members according to one or more selection criteria.FIG. 2 is a screen display from a software-based application, for performing the method of the present invention, that allows an intervention agent to select a set of plan members from the group based on certain selection criteria. As shown inFIG. 2, themember selection screen30 includes boxes that allow the group members to be selected or extracted byrelative risk rank32,relative risk34,disease category36,zip code38, andcounty40. In the case of a member information database containing information for multiple employer groups, themember selection screen30 includes a selection to allow plan members to be extracted bygroup number42. Themember selection screen30 also allows members to be searched bymember name44 andmember ID46. As shown inFIG. 2, a user may search for all plan members having relative risk rank between specified numbers. The relative risk rank is assigned as described in detail above. As further shown, the user may also search byrelative risk34 between two specified numbers. The relative risk is determined as explained above.
• As also shown, the user may select plan members based ondisease category36. This will select all members having a claim within a specified disease category. In one embodiment of the present invention, the disease categories are ICD-9 codes. In another embodiment of the present invention, the disease categories are Clinical Care Group (“CCG”) codes. CCG's are an Ingenix innovation for classifying diagnosis codes and medical claims. The CCG system allows the over 14,000 ICD-9 codes to be placed into a more manageable number of CCG classes, namely, about 450 related disease or diagnosis categories. These approximately 450 CCG classes can be further reduced into about 120 CCG categories. Finally, these approximately 120 CCG categories can be placed into about 20 CCG specialties. Each of these levels provides various advantages in analyzing the claims data.
• In one embodiment of the present invention, CCG categories are assigned to each of the pharmacy claims. This process requires that the pharmacy claims be processed together with the medical claims in order to attach a CCG category to each pharmacy claim. In one embodiment, this is accomplished using the Ingenix Drug-Disease Matcher (“DDM”) application. This application is described in greater detail in U.S. patent application Ser. No. 09/571,648, filed on May 15, 2000, entitled “SYSTEM AND METHOD OF DRUG-DISEASE MATCHING,” by Gerald Lutgen, et al., which is hereby incorporated by reference in its entirety.
• As further shown inFIG. 2, in another embodiment of the present invention, themember selection screen30 allows for searching byzip code38. Searching byzip code38 allows the user to select a set of plan members in a certain geographic region. This may be beneficial if the user wishes to assign all of these members to a specific intervention agent (as further described below). In another embodiment of the present invention, themember selection screen30 allows the user to search the database bycounty name40. The search bycounty name40 is similar to a search byresidential zip code38 in that the user may select a set of plan members based on geographic location. In an embodiment where the member information database includes members from multiple groups, themember selection screen30 allows the members to be searched bygroup number42. This allows the user to select only those members within a specific group. In one embodiment, themember selection screen30 includes amember name44 field that allows the user to search for a specific plan member by name. In another embodiment, themember selection screen30 includes amember ID46 field that allows the user to select a plan member based on the member's numerical identification.
• In one embodiment of the present invention, the user may use themember selection screen30 to select a set of members from the group by using multiple selection criteria. In one embodiment of the present invention, each of the selection criteria shown inFIG. 2 is present in themember selection screen30, allowing the user to select an intervention set of members using any of the shown fields. The user may select an intervention set of members by selecting fields it wishes to search and entering appropriate selection information. In other embodiments, only one of the fields shown on themember selection screen30 is present.
• As shown inFIG. 3, the member selection listing50 includes asearch summary section52, amember listing section54, and a save-to-file button56. The member selection listing50 allows the user (e.g., a care management supervisor) to view the entered selection criteria, using thesection52, and assign an intervention agent to each member in themember listing section54. Themember listing section54 lists each of the members of the intervention set. Once an intervention agent has been assigned, the user can save the assignments to individual files using thesave button56. The user can then transmit each of these files to the appropriate intervention agent for intervention.
• In one embodiment of the present invention, the members of the intervention set are listed according to relative risk rank, with the member having the lowest rank listed first. In one embodiment of the present invention, the members of the intervention set are listed by first sorting the members by relative risk and then further sorting by the number of intervenability factors present for that member, such that within a given relative risk number, the plan members with the highest number of intervenability factors are ranked first, and those with a lower number of intervenability factors are ranked lower. In one embodiment of the present invention, the first step, sorting the members of the intervention set by relative risk, involves placing plan members into a relative risk category based on whole numbers only. In other words, for example, all members having a relative risk of “3” and a fraction are placed into the “3” relative risk category.
• In another embodiment of the present invention, the members of the intervention set are assigned a relative risk ranking by first sorting the members according to the number of intervenability factors present for each member. In other words, all members having eight intervenability factors present are placed at the top of the list, all members having seven intervenability factors are placed second, and so on. Next, within each number of intervenability factors, members are ranked according to relative risk, with the member having the highest relative risk placed first. In other embodiments of the present invention, some other formula is used to order the members of the intervention set based on some combination of the member's relative risk and the member's intervenability factors. The specific formula used to order members is not important, but the idea is to rank the members of the intervention set according to those who most need intervention and those who are most amenable to intervention.
• As shown inblock20 ofFIG. 1, a display is generated containing relevant information regarding the plan member. This display is used by the intervention agent to intervene in the member's care. The display is generated when the intervention agent selects a particular member.FIG. 4 shows themember information screen60, according to one embodiment of the present invention. Themember information screen60 provides the intervention agent with information regarding a particular plan member that assists the agent in determining where the plan member would benefit from intervention in care. In one embodiment of the present invention the display only includes that information previously discussed. In another embodiment of the present information, further information is now extracted from the claims for the particular member, as discussed below.
• As shown inFIG. 4, in one embodiment, themember information screen60 includes ademographic information section62, autilization summary64, and a “risk factors”section66. In one embodiment, the demographic information includes the member's name, address, age, gender, phone number, and identification number. In the embodiment shown inFIG. 4, theutilization summary64 includes an indication of the amount of out-of-network costs incurred by the member, a total health care cost for the member in the past year (or some other historic period), a predicted total cost for an upcoming time period, and the number of emergency room visits in the past year (or some other time period).
• The “risk factors”section66 of themember information screen60 provides information used by the intervention agent to intervene in the member's health care treatment strategy. As shown inFIG. 4 on the left hand side, the “risk factors”section66 includes information regardingrelative risk68 andintervenability70. The section also includes information on other factors influencing care intervention, including mental health diagnoses72, self-care characteristics74, equipment or monitors76, anddrug history78. On the right hand side, the “risk factors”section66 includes a top medicalepisode categories section79. The top medicalepisode categories section79 identifies the top medical episode categories that are driving risk and future health care resource utilization for the current plan member. In one embodiment of the present invention, only intervenability70 is present in the “risk factors”section66. In another embodiment, only intervenability70 is present.
• The memberdemographic information62 is extracted from the information in the database extracted from the claims data. The information in theutilization summary section64 is likewise extracted from the database including claims information, with the exception of the predicted total cost. The predicted total cost is calculated as described above with reference to predicted future health care utilization. The predicted total cost may be computed using those methods described herein or any methods known to those of skill in the art.
• Relative risk68 is computed as described above and displayed in themember information screen60 to give the intervention agent a quick summary of the current plan member's level of risk. The entry forintervenability70 is the number of intervenability factors present for the current member. The entry for mental health diagnoses72, in the embodiment of the invention shown inFIG. 4, is a “yes” or “no” entry. A “yes” is entered formental health diagnoses72 if an examination of the physician claims for the current member include an ICD-9 code corresponding to a mental health diagnosis. If an appropriate ICD-9 code is found in the claims data, a “yes” is entered in this field; if not, a “no” is entered. In another embodiment of the present invention, the entry for mental health diagnoses72 is determined by examining the facility claims for the current plan member for an appropriate ICD-9 code. In another embodiment of the present invention, all medical claims for the current plan member are examined.
• The entry for self-care characteristics74 is also either a “yes” or a “no” entry. In one embodiment of the present invention, the appropriate entry for self-care characteristics74 is determined by examining the present member's claims for an ICD-9 code corresponding to a diagnosis that represents some type of disability. Self-care characteristics74 is used by the intervention agent to determine factors effecting provision of care. In one embodiment of the present invention, the self-care characteristics74 searched for in the claims data includes hearing impairment, visual impairment, ambulatory impairment, and diabetes. In other embodiments of the present invention, more or fewer self-care characteristics are searched for in the claims data for the current plan member. In one embodiment of the present invention, the pharmacy claims are also searched for any self-care characteristics.
• The entry for equipment or monitors76 is also either a “yes” or a “no” entry. The proper entry is determined by examining the medical claims for the current member for a HCPC or CPT code for medical equipment or a monitoring device. If such a code is found, this entry is recorded as a “yes.” Likewise, the pharmacy claims are examined for the presence of a code corresponding to appropriate durable medical goods. Again, if such a code is found, this entry is recorded as a “yes.” Like the entry for self-care characteristics74, the entry for equipment or monitor76 is used to provide the intervention agent with information on the current plan member and with an entry point for discussion regarding whether the member is properly using the equipment or monitor. The entry fordrug history78 is also either a “yes” or “no” entry. The proper entry fordrug history78 is ascertained by examining the pharmacy claims for the current plan member to determine whether the member has been prescribed any medications. If the member has been prescribed medications, a “yes” is entered fordrug history78.
• In one embodiment of the present invention, more information can be obtained on each of the entries shown in the “risk factors”section66, shown on the left hand side of this section, by selecting the entry (e.g., by “clicking” on the entry). In one embodiment of the present invention, all of the risk factors66, shown on the left hand side can be selected to obtain further information.
• In one embodiment of the present invention, further information on the intervenability factors present for the current member can be obtained by selectingintervenability70.FIG. 5 is a screen display showing an intervenability factors screen80, according to one embodiment of the present invention. As shown inFIG. 5, the intervenability factors screen80 provides a brief description of each of the intervenability factors present for the current member.
• In one embodiment of the present invention, as shown inFIG. 5, possible intervenability factors that may be displayed on the intervenability factors screen80 include whether the current plan member had any emergency room visits in the focus period (numeral82), whether the current member had any in-patient hospital admissions within the focus period (numeral84), whether the current member incurred any out-of-network costs (numeral86), whether the current member had more than three different provider specialists within the focus period (numeral;88), whether the current member had more than a specified number of pharmacy prescriptions (numeral90), whether the member is seeing an appropriate health care provider for a chronic episode (numeral92), whether the current member is missing any targeted interventions for medical diagnoses (numeral94), and whether the current member is non-compliant with any prescriptions (numeral96). Further information can be obtained on each of the intervenability factors shown on the intervenability factors screen80 inFIG. 5, by selecting the desired intervenability factor. In other embodiments of the present invention, more or fewer intervenability factors are employed.
• Whether the current plan member had any emergency room visits within the focus period (intervenability factor82) is determined by examining the medical claims for the member to search for any codes in the claims indicating an emergency room visit. Further information can be obtained onintervenability factor82 by selecting it, as described above. In one embodiment of the present invention, whenintervenability factor82 is selected, the medical claims indicating an emergency room visit by the plan member are displayed for further review by the intervention agent. WhileFIG. 5, describes this factor as “ER visits in past year,” any time period (e.g., the focus period) may be examined in accordance with the present invention.
• Whether the current plan member had any hospital admissions during the focus period (intervenability factor84) is likewise determined by examining the medical claims for the plan member. This factor is identified as “inpatient admits in past year,” inFIG. 5. If any of the medical claims include a code indicating that the member was admitted to a hospital or other appropriate health care facility, then intervenabilityfactor84 is present for the current member. The intervention agent may obtain further information onintervenability factor84 by selecting it on the intervenability factors screen80. In one embodiment, selectingintervenability factor84 causes a display of the medical claims evidencing an in-patient hospital admission of the current plan member.
• Whether the current plan member incurred any out-of-network costs (intervenability factor86) is again determined by an examination of the member's medical and pharmacy claims. If any of the claims were submitted by providers that are not within the health care plan's network, the total costs are calculated andintervenability factor86 is counted as present for the current member. In one embodiment of the present invention, ifintervenability factor86 is selected, the claims evidencing out-of-network costs for the member are displayed.
• Whether the current plan member saw more than three different specialists in the focus period (intervenability factor88) is again determined by examining the medical claims for the member. If the medical claims indicate that the member visited more than three providers during the focus period,intervenability factor88 is counted as present for the current member. In one embodiment of the present invention, if the intervention agent selectsintervenability factor88, the claims showing the specialists seen by the member are displayed.
• Whether the current plan member has been prescribed more than a specified number of medications (intervenability factor90) is determined by an examination of the pharmacy claims for the member. This factor is identified as “Poly Rx,” inFIG. 5. In one embodiment of the present invention, theintervenability factor90 is counted as present for the current member if eight or more distinct medications are present in the pharmacy claims for that member. In another embodiment, theintervenability factor90 is counted as present if more than one distinct medication is present for the member. By “distinct medications,” it is meant that if a member has multiple prescriptions filled for the same medication, it will count as only one prescription for purposes of determining the presence of theintervenability factor90. In other embodiments of the present invention, a greater or fewer number of distinct medications trigger the presence of theintervenability factor90. In one embodiment of the present invention, when the intervention agent selects theintervenability factor90, the pharmacy claims showing the distinct medication prescribed to the member are displayed.
• Whether the current plan member is seeing an appropriate provider for a medical episode is the subject of theintervenability factor92. In one embodiment of the present invention,intervenability factor92 only examines chronic episodes, as chronic episodes provide a significant contribution to health care resource utilization. In another embodiment, all medical episodes are considered. In one embodiment of the present invention, the presence of theintervenability factor92 is determined by searching the member's medical claims to determine whether a particular medical episode is present for the member. In one embodiment of the present invention, the medical episodes that are investigated underintervenability factor92 are stored in a look-up table that can be easily used manually or by a software program. In one embodiment of the present invention, the particular medical episodes that are investigated are specified by CCG classes or categories. A skilled healthcare practitioner can readily identify medical episodes and corresponding healthcare providers that are amenable to examination.
• In one embodiment of the present invention, for example, the medical episodes investigated include otitis media (e.g., middle ear infection) and pregnancy. The look-up table also includes a list of one or more providers corresponding to each of these medical episodes. For example, the corresponding providers for pregnancy include an obstetrician, a family practice physician, or an internist. If the medical claims for the current member indicate a pregnancy, the physician claims are searched to verify that an obstetrician, a family practice physician, or an internist is managing the member. If one of the specified medical episodes is present for the current member, and the member has failed to see an appropriate physician, theintervenability factor92 is counted as present for the current member. In one embodiment of the present invention, the intervention agent can obtain further information on theintervenability factor92 by selecting it on theintervenability factor screen80. If theintervenability factor92 is selected, the medical claims corresponding to the medical episode for which the plan member has failed to visit an appropriate provider are displayed.
• Whether the current plan member has aspects of care for a specified disease meriting intervention (intervenability factor94) is addressed in a similar manner as that described with respect tointervenability factor92. This factor is identified as “missing targeted interventions,” inFIG. 5. Medical episodes or conditions of interest are specified in conjunction with corresponding interventions. A healthcare provider (e.g., a physician) having ordinary skill in the art, based on his or her knowledge and experience, can readily perform this specification. In one embodiment of the present invention, for example, the conditions that are investigated include diabetes, heart failure, asthma, and chronic obstructive pulmonary disease (“COPD”). In other embodiments of the present invention, other medical conditions may also be analyzed under this intervenability factor. Again, a skilled medical practitioner can specify corresponding interventions. In one embodiment, the medical episodes are specified using CCG class or categories.
• If it is determined that the current plan member has diabetes, for example, the medical and pharmacy claims for the member are examined to determine whether that member is seeking certain types of treatment that should be associated with diabetes, as identified by procedural and pharmacy codes in the claims. In one embodiment of the present invention, there are separate treatment protocols, one for insulin-dependent diabetes and one for non-insulin-dependent diabetes. In the case of insulin-dependent diabetes, the treatments that are searched for include whether the member is receiving insulin, whether the member is receiving an annual eye exam, whether the member is receiving a hemoglobin test and whether the member is visiting an endocrinologist. If any of these conditions corresponding to insulin-dependent diabetes is missing from the member's claims, theintervenability factor94 is counted as present for the current member. This same process is applied for all specified medical conditions. Also, for example, if the member is diagnosed with heart failure, pharmacy claims are examined to verify that the member is filling prescriptions for ACE-inhibitors. If the member's pharmacy claims do not indicate a fill of this prescription, theintervenability factor94 is indicated as present for the member. In one embodiment of the present invention, theintervenability factor94 is examined using a table including a list of medical conditions of interest in one column, and including a list of corresponding medications in another column.
• In one embodiment of the present invention as shown inFIG. 6, the intervention agent can obtain further information on theintervenability factor94 by selecting it on the intervenability factors screen80.FIG. 6 shows a display of an episode class interventions screen100, according to one embodiment of the present invention. The episodeclass intervention screen100 provides the intervention agent with additional information regarding the treatments or interventions for a member having a diagnosis of COPD in the claims. As shown inFIG. 6, an episode of COPD causes the method of the present invention to search for whether the member has a nebulizer (found in the member's pharmacy claims), whether the member has had an oxygen DME (found in the member's medical claims), and whether the member has had a pulmonary function test (found in the member's medical claims). The episodeclass intervention screen100 provides a “yes” or “no” indication for each to the intervention agent for each of these interventions. As shown inFIG. 6, the episodeclass intervention screen100 includes anepisode102 column and anintervention104 column. In one embodiment of the present invention, the episodes are specified in terms of CCG classes, categories, or specialties, using the Ingenix CCG system described above. In another embodiment of the present invention, the episodes are specified using ICD-9 codes. Use of CCG's or ICD-9 codes facilitates the method of the present invention in searching the claims for the member.
• Whether the current plan member is compliant with taking prescribed medications (intervenability factor96) is determined by examining the pharmacy claims for the member. This factor is identified as “Rx compliance issues,” inFIG. 5. In one embodiment, the pharmacy claims are examined for the presence of medications that are generally prescribed for chronic conditions or are generally taken over a long term. The pharmacy claims are also examined for any medication for which treatment duration is known or is typical. If such a medication is found in the member's pharmacy claims, they are further examined to verify that the member has continued to fill the prescription over the entire time of the focus period, or at least over the normal required time associated with the medication. Examples of medications for which if they are filled once should normally be filled throughout the focus period include insulin, asthma inhalers, and ACE-inhibitors. If the current member has a pharmacy claim for one of these medications during the focus period and fails to continue to fill the prescription throughout the period,intervenability factor96 is indicated as present for the member. If theintervenability factor96 is shown on the intervenability factors screen80 inFIG. 5, the intervention agent can obtain further information by selecting it.
• In one embodiment of the present invention, medication compliance is analyzed for the current member by examining all medications having at least two fills indicated by the member's pharmacy claims. For each medication appearing at least twice, a formula is applied using the fill dates and the fill durations (e.g., 30 days, 60 days, 90 days, etc.) to determine whether the member is compliant with that particular medication. For example, if the current member's pharmacy claims show that Medication A was filled three times, each time spaced 90 days from the previous fill, and each fill was for a duration of 30 days, the formula will indicate thatintervenability factor90 is present for the member.
• FIG. 7 shows a display of a non-compliant drug categories screen110, according to one embodiment of the present invention. As shown inFIG. 7, the non-compliantdrug category screen110 includes a column forcompliance rate112 and a column fordrug category114. In one embodiment of the present invention, the compliance rate is a percentage indicating the percent of times the member filled a prescription for a particular medication. Thedrug category114 indicates the particular medication that the member is not filling. In one embodiment of the present invention, the drug categories are determined using National Drug Code (“NDC”) categories. In another embodiment of the present invention, more general drug categories including multiple NDC's are used.
• While eight intervenability flags are identified above, a healthcare practitioner having ordinary skill in the art may readily identify other appropriate intervention flags signaling that a member is amenable to intervention in care by an intervention agent. For example, a member's claims can readily be examined for compliance with specified physician practice patterns relating to treatment of a specified disease or medical condition.
• The next factor, shown in the “risk factors”section66 on themember information screen60 inFIG. 4 is mental health diagnoses72. As described above, the intervention agent can obtain additional information by selecting this entry.FIG. 8 shows a display of additional information regarding this category in a behavioral risk factors screen120. In the embodiment shown inFIG. 8, the behavioral risk factors screen120 includes a list of each of the behavioral or mental health diagnoses present in the claims of the current plan member. The presence of behavioral or mental health diagnoses is ascertained by examining both the medical claims and the pharmacy claims for the member for ICD-9, CPT, HCPC, or NDC codes corresponding to mental health diagnoses. In one embodiment of the present invention, only one of the categories of medical claims and pharmacy claims is examined. In one embodiment of the present invention, the method involves examining the claims for the presence of a specified code or set of codes corresponding to behavioral or mental health diagnoses. In one embodiment of the present invention, the intervention agent, may access further information on one of the behavioral risk factors by selecting it on the behavioral risk factors screen120. If the intervention agent selects the behavioral risk factor, the corresponding medical and pharmacy claims will be displayed.
• The next risk factor, shown in the “risk factors”section66 on themember information screen60 is self-care characteristics74. The intervention agent may obtain further information on this category by selecting it. If the intervention agent selects self-care characteristics74, the corresponding claims for the member from the claims database will be displayed. Thenext risk factor66 shown is equipment or monitors76. By selecting this category, the intervention agent can access the claims corresponding to equipment or monitors for the current member. Thefinal risk factor66 shown isdrug history78. By selecting this category, the intervention agent can obtain a link to each of the pharmacy claims for the member.
• FIG. 9 shows a drughistory report screen130 according to one embodiment of the present invention. As shown inFIG. 9, the drug history report includes a column fordrug category132,fill date134,NDC code136, anddrug name138. An intervention agent may use thedrug history report130 shown inFIG. 9 to further assess the current member's health care situation and any necessary interventions.
• The right hand side of the “risk factors”section66 is a listing of topmedical episode categories79. This listing (briefly described above) is intended to provide the intervention agent with summary information regarding the current member's primary medical conditions driving the member's risk and health care resource utilization. In one embodiment, this listing is generated by examining the current member's medical claims to determine which medical episode has the highest associated cost. In one embodiment, the medical episodes are organized according to CCG classes or categories, as described above. In another embodiment, the episode categories are organized by specified groups of ICD-9 codes, or by individual ICD-9 codes. In another embodiment of the present invention, this listing is generated by searching the member's medical claims to determine the presence of various CCG categories for the member. Next, a predetermined ranking of CCG categories is used to identify those considered to be high risk categories. In this embodiment, each of the CCG categories is assigned a ranking by associated risk. The top risk CCG categories for a given plan member are then listed in the topmedical episode categories79. In this embodiment, the ranking of the CCG categories is done according to historical experience. A person skilled in the art can assign a risk or health care resource utilization ranking to a CCG category based on their knowledge and experience with that particular medical episode.
• In another embodiment of the present invention, each CCG category is assigned an average cost by examining a base set of claims for an historical time period, and calculating the average cost associated with each CCG category. This average cost is then used as the ranking for each CCG category. In another embodiment of the present invention, the listing incategories79 is determined by using a combination of the costs associated with the particular CCG category for the current member, and the average historical costs associated with the CCG category, using an appropriate formula. In this embodiment, a first ranking is assigned to each CCG category present based on the member's particular cost and a second ranking is assigned to each CCG category present based on the average historical costs. The first and second rankings are then summed together to reach an overall ranking for each CCG category. In one embodiment, the top five CCG categories for the member are displayed in the topmedical episode categories79.
• In one embodiment of the present invention, the intervention agent can obtain further information on any of the listed medical episodes by selecting it. If the intervention agent selects an episode, the member's episodes for that CCG category will be displayed by CCG class. In one embodiment, if the intervention agent then selects one of the displayed CCG classes, a breakdown of that CCG class is displayed by summarizing claims in three categories: (1) emergency room visits, (2) in-patient admissions, (3) specialist care, and (4) pharmacy cost. If one of these categories is selected, the claims corresponding to that category will be displayed. In another embodiment, the intervention agent may cause the corresponding claims to be displayed by selecting the displayed medical episode.
• In one embodiment, the method of the present invention operates using the following sequence of steps. First, a predicted total cost and a relative risk are calculated for each member of a group. Next, the members of the group are ranked by each member's relative risk and a high-risk group of members is chosen by choosing member's from the group having the highest relative risk. In one embodiment, the top three to five percent of the member's are chosen as the high-risk members. Next, the corresponding claims for each of the high-risk members are examined to identify intervention flags, top medical episodes driving risk, and care influencing factors. Next, a high-risk database is compiled containing information on each of the high-risk members. This information includes the items specified above and a complete claim detail history.
• A set of intervention members is then selected from the high-risk members. The selection is done on a member intervention application using one or a combination of the following criteria: relative risk, relative risk rank, presence of a disease category, and demographic information. Next, the member intervention application displays a list of the intervention members to an intervention agent. Finally, the intervention agent selects one of the intervention members for display, and the member intervention application displays the above-identified information corresponding to the selected intervention member.
• Using themember information screen60, the intervention agent can quickly obtain relevant and useful information regarding a member's medical treatment history and use that information to contact the member and proactively intervene in the member's care by providing guidance and suggestions for improving that care.
• Although the present invention has been described with reference to preferred embodiments, persons skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention.

Claims (7)

1-30. (canceled)

31. A computer-implemented high risk member identification method, comprising a computer performing the following:

identifying a group of members to be analyzed using a computer software application, each group member having an associated relative risk value, wherein the relative risk value for each member is a function of predicted future healthcare resource utilization for the member;

filtering the group members using the computer software application to identify members having an associated relative risk value that exceeds a threshold value of relative risk, thereby identifying a subset of the group members for potential intervention;

analyzing healthcare claim data associated with each subset member using the computer software application to ascertain the presence or absence of each of a plurality of intervenability factors present to the subset member, wherein the intervenability factors for each subset member are identified based upon aspects of each subset member's healthcare history that are amenable to intervention by an intervention agent;

using the computer software application to assign each subset member a number of intervenability factors representing a total number of the intervenability factors present in the subset member's healthcare claim data;

using the computer software application to assign a relative risk ranking to each subset member based upon the subset member's associated relative risk value and the number of intervenability factors assigned to the subset member; and

displaying (a) a list of the subset members ordered by respective relative risk rankings and (b) the relative risk value for each subset member on an electronic display.

32. The method ofclaim 31, wherein the intervenability factors include at least one of the following: (a) whether the member visited the emergency room during the focus period, (b) whether the member had any in-patient hospital admissions during the focus period, (c) whether the member incurred any out-of-network costs during the focus period, (d) whether the member visited more than three different provider specialists during the focus period, (e) whether the member was prescribed multiple pharmaceuticals during the focus period, (f) whether the member has no appropriate provider for a chronic episode during the focus period, (g) whether the member missed a target intervention during the focus period, or (h) whether the member fails to obtain fills of prescribed medication during the focus period.

33. The method ofclaim 31, further comprising creating a database of the healthcare claim data for the subset members, wherein the database includes all medical diagnoses and healthcare utilization patterns for each subset member during a focus period, including any physician claims, facility claims and pharmacy claims associated with each subset member during the focus period.

34. The method ofclaim 31, further comprising using the computer software application to determine one or more top medical episodes driving risk of each subset member, wherein the top medical episodes are determined by examining the subset member's claim data by diagnosis code or medical episode to determine which of the subset member's medical conditions has the highest associated cost.

35. The method ofclaim 31, further comprising:

receiving a selection of one or more displayed subset members input into the computer software application by an intervention agent; and

displaying demographic information, a utilization summary, risk factors including behavioral risk factors and self-care characteristics, and the intervenability factors for each subset member selected by the intervention agent on the electronic display.

36. The method ofclaim 35, wherein the intervention agent may filter the displayed subset members by zipcode, county, group numbers, products, member ID or member names.

Images