ORGANIZATIONAL POPULATION HEALTH MANAGEMENT PLATFORM AND PROGRAMS
Methods, systems, and computer-readable media are provided for organizational management of population health. A network management service builds and maintains data stores of organizational data associated with one or more healthcare organizations. A population health management service receives raw data from disparate sources, creates a reference record for each raw data source, and generates patient-centric longitudinal population records using the reference records. Programs for managing population health utilize both the organizational data and the population records to generate condition-specific or objective-specific solutions.
Latest CERNER INNOVATION, INC. Patents:
This application, having attorney docket number CRNI.176025, and entitled “Organizational Population Health Management Platform and Programs,” claims priority to U.S. Provisional Application No. 61/710,877, filed Oct. 8, 2012, and entitled “Organizational Management of Population Health.” The entirety of the aforementioned application is incorporated by reference herein.
This application, having attorney docket number CRNI.176025, and entitled “Organizational Population Health Management Platform and Programs,” is related by subject matter to concurrently filed U.S. patent application Ser. No. ______, having attorney docket number CRNI.196078, entitled “Provider and Patient Attribution Programs;” U.S. patent application Ser. No. ______, having attorney docket number CRNI.196079, entitled “Contracts and Organization Management Program;” U.S. patent application Ser. No. ______, having attorney docket number CRNI.196080, entitled “Outreach Program;” and U.S. patent application Ser. No. ______, having attorney docket number CRNI.196093, entitled “Score Cards.” The entireties of the aforementioned applications are incorporated by reference herein.
BRIEF SUMMARYThis Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.
In brief and at a high level, this disclosure describes, among other things, methods, systems, computer storage media, and graphical user interfaces for organizational management of population health. A system or platform for managing population health includes a network management service that builds, maintains, and updates data stores that include information about healthcare organizations, healthcare providers, and information concerning contractual provisions between healthcare organizations and payers (e.g., insurance companies). The network management service also includes a program builder that builds condition-specific and/or objective-specific program templates. The program templates are designed to, among other things, identify population segments based on condition or purpose, stratify the population segment using different factors, generate workflows, attribute patients within the segment to healthcare providers, manage healthcare organizations and contracts, measure intervention outcomes, and the like.
The system further includes a population management service that builds, maintains, and updates data stores that include longitudinal patent-centric information drawn from a variety of disparate data sources. Such information may include demographic information, claims information, pharmacy information, clinical care information, socioeconomic information, financial information, and the like. Additionally, the system includes a compiler that extracts a program template and customizes or localizes it based on a particular healthcare organization's organizational and provider information as well as the particular contractual provisions associated with the healthcare organization and its associated payers.
At run-time, a program engine in the system uses the customized program templates and the patient population information from the population management service to generate condition-specific or objective-specific patient population data for the healthcare organization. This data is used by the healthcare organization to, among other things, identify patient segments based on a condition or for a specific purpose, stratify patients within the segment by degree of risk, generate interventions, attribute patients in the segment to healthcare providers associated with the particular healthcare organization, measure intervention outcomes, manage payer/organization contracts, determine provider incentives, and the like.
Embodiments are described in detail below with reference to the attached drawing figures, wherein:
The subject matter of the present invention is described with specificity herein to meet statutory requirements. However, the description itself is not intended to limit the scope of this patent. Rather, the inventors have contemplated that the claimed subject matter might also be embodied in other ways, to include different steps or combinations of steps similar to the ones described in this document, in conjunction with other present or future technologies. Moreover, although the terms “step” and/or “block” may be used herein to connote different elements of methods employed, the terms should not be interpreted as implying any particular order among or between various steps herein disclosed unless and except when the order of individual steps is explicitly described.
In brief and at a high level, this disclosure describes, among other things, methods, systems, computer storage media, and graphical user interfaces for organizational management of population health. A system or platform for managing population health includes a network management service that builds, maintains, and updates data stores that include information about healthcare organizations, healthcare providers, and information concerning contractual provisions between healthcare organizations and payers (e.g., insurance companies). The network management service also includes a program builder that builds condition-specific and/or objective-specific program templates. The program templates are designed to, among other things, identify population segments based on condition or purpose, stratify the population segment using different factors, generate workflows, attribute patients within the segment to healthcare providers, manage organization and contracts, generate outreach events, measure intervention outcomes, and the like.
The system further includes a population management service that builds, maintains, and updates data stores that include longitudinal patent-centric information drawn from a variety of disparate data sources. Such information may include demographic information, financial information, socioeconomic information, claims information, pharmacy information, clinical care information, and the like. Additionally, the system includes a compiler that extracts a program template and customizes or localizes it based on a particular healthcare organization's organizational and provider information as well as the particular contractual provisions associated with the healthcare organization and its associated payers.
At run-time, a program engine in the system uses the customized program templates and the patient population information from the population management service to generate condition-specific or objective-specific patient population data for the healthcare organization. This data is used by the healthcare organization to, among other things, identify patient segments based on a condition or for a specific purpose, stratify patients within the segment by degree of risk, generate interventions, attribute patients in the segment to healthcare providers associated with the particular healthcare organization, measure intervention outcomes, manage payer/organization contracts, determine provider incentives, generate outreach events, and the like.
An exemplary computing environment suitable for use in implementing embodiments of the present invention is described below.
The present invention might be operational with numerous other purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that might be suitable for use with the present invention include personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above-mentioned systems or devices, and the like.
The present invention might be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Exemplary program modules comprise routines, programs, objects, components, and data structures that perform particular tasks or implement particular abstract data types. The present invention might be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules might be located in association with local and/or remote computer storage media (e.g., memory storage devices).
With continued reference to
The control server 102 typically includes therein, or has access to, a variety of non-transitory computer-readable media. Computer-readable media can be any available media that might be accessed by control server 102, and includes volatile and nonvolatile media, as well as, removable and nonremovable media. By way of example, and not limitation, computer-readable media may comprise computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by control server 102. Communication media typically embodies computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. Combinations of any of the above should also be included within the scope of computer-readable media.
The control server 102 might operate in a computer network 106 using logical connections to one or more remote computers 108. Remote computers 108 might be located at a variety of locations in a medical or research environment, including clinical laboratories (e.g., molecular diagnostic laboratories), hospitals and other inpatient settings, veterinary environments, ambulatory settings, medical billing and financial offices, hospital administration settings, home healthcare environments, and clinicians' offices. Clinicians or healthcare providers may comprise a treating physician or physicians; specialists such as surgeons, radiologists, cardiologists, and oncologists; emergency medical technicians; physicians' assistants; nurse practitioners; health coaches; nurses; nurses' aides; pharmacists; dieticians; microbiologists; laboratory experts; laboratory technologists; genetic counselors; researchers; veterinarians; students; and the like. The remote computers 108 might also be physically located in nontraditional medical care environments so that the entire healthcare community might be capable of integration on the network. The remote computers 108 might be personal computers, servers, routers, network PCs, peer devices, other common network nodes, or the like and might comprise some or all of the elements described above in relation to the control server 102. The devices can be personal digital assistants or other like devices.
Computer networks 106 comprise local area networks (LANs) and/or wide area networks (WANs). Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets, and the Internet. When utilized in a WAN networking environment, the control server 102 might comprise a modem or other means for establishing communications over the WAN, such as the Internet. In a networking environment, program modules or portions thereof might be stored in association with the control server 102, the data store 104, or any of the remote computers 108. For example, various application programs may reside on the memory associated with any one or more of the remote computers 108. It will be appreciated by those of ordinary skill in the art that the network connections shown are exemplary and other means of establishing a communications link between the computers (e.g., control server 102 and remote computers 108) might be utilized.
In operation, an organization might enter commands and information into the control server 102 or convey the commands and information to the control server 102 via one or more of the remote computers 108 through input devices, such as a keyboard, a pointing device (commonly referred to as a mouse), a trackball, or a touch pad. Other input devices comprise microphones, satellite dishes, scanners, or the like. Commands and information might also be sent directly from a remote healthcare device to the control server 102. In addition to a monitor, the control server 102 and/or remote computers 108 might comprise other peripheral output devices, such as speakers and a printer.
Although many other internal components of the control server 102 and the remote computers 108 are not shown, such components and their interconnection are well known. Accordingly, additional details concerning the internal construction of the control server 102 and the remote computers 108 are not further disclosed herein.
Turning now to
The population health management system 200 includes a network management service 210, a population management service 212, a program engine 214 and a compiler 236 all in communication with each other through wired connections or through a network. The network may include, without limitation, one or more local area networks (LANs) and/or wide area networks (WANs). Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets and the Internet. Accordingly, the network is not further described herein.
The network management service 210, the population management service 212, the program engine 214 and the compiler 236 are merely exemplary. While each of these components/services/modules is illustrated as a single unit, it will be appreciated that each of these components/services/modules is scalable. For example, the network management service 210 may in actuality include a plurality of computing devices in communication with one another. Components of the network management service 210, the population management service 212, the program engine 214 and the compiler 236 may include a processing unit, internal system memory, and a suitable system bus for coupling various system components, including one or more data stores for storing information (e.g., files and metadata associated therewith). Each of these components/services/modules typically includes, or has access to, a variety of computer-readable media.
In some embodiments, one or more of the illustrated components/services/modules may be implemented as stand-alone applications. In other embodiments, one or more of the illustrated components/services/modules may be integrated directly into the operating system of the population health management system 200. The component/services/modules illustrated in
It should be understood that this and other arrangements described herein are set forth only as examples. Other arrangements and elements (e.g., machines, interfaces, functions, orders, and groupings of functions, etc.) can be used in addition to or instead of those shown, and some elements may be omitted altogether. Further, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in conjunction with other components/modules, and in any suitable combination and location. Various functions described herein as being performed by one or more entities may be carried out by hardware, firmware, and/or software. For instance, various functions may be carried out by a processor executing instructions stored in memory.
As shown in
The data stores 222, 224, 226, and 234 are configured to store information used by network management service 210 and the population management service 212. The content and volume of such information in the data stores 222, 224, 226, and 234 are not intended to limit the scope of embodiments of the present invention in any way. Further, although each data store 222, 224, 226, and 234 is illustrated as a single, independent component, the data stores 222, 224, 226, and 234 may, in fact, be a plurality of storage devices, for instance, a database cluster, portions of which may reside on the network management service 210, the population management service 212, the program engine 214, and/or any combination thereof.
Turning first to the network management service 210, the receiving component 216 is configured to receive organizational data from one or more healthcare organizations. The healthcare organizations may bear some logical relationship to each other, or, alternatively, the healthcare organizations may be disparate with no logical relationship to one another. As used throughout this application, a healthcare organization may comprise a single healthcare facility with an associated group of healthcare providers such as physicians, pharmacists, nurses, health coaches, therapists, and the like. A healthcare organization may also comprise a network of healthcare facilities, each healthcare facility having an associated group of healthcare providers. The healthcare organization may create such networks in order to achieve certain financial and/or clinical objectives. The healthcare facilities may be located geographically close to one another or may be spread across a large geographical area. Further, the healthcare facilities may comprise a network of hospitals, clinics, provider offices, nursing homes, pharmacies, home health services, and the like. As such, the term “healthcare organization” as used herein is meant to be construed broadly to cover a wide range of healthcare scenarios.
The organizational data may include data about each healthcare facility such as operating hours, type of care provided, unique characteristics associated with the health care facility, geographic location, accessibility, associated providers, and the like. The organizational data may also include healthcare provider data such as names, credentials, affiliations, areas of practice, geographic location, provider preferences, current patient load, demographic characteristics such as gender or age, provider outcome data, and the like.
Continuing, the organizational data may also include contractual data. By way of background, when a healthcare organization enters into a payment agreement with a payer, a contract is generated that includes a variety of information or provisions. A payer refers to an entity that intends to pay or is responsible for payment of healthcare services. Payers include without limitation employers, government entities (such as Centers for Medicare and Medicaid Services), charities, patients, insurers and secondary insurers. The contract between a payer and a healthcare organization may include financial objectives (e.g., payment terms), clinical objectives, warranties, terms and conditions, and the like. For example, a payer may agree to pay a healthcare organization a certain amount if the healthcare organization meets certain quality measures or clinical objectives (e.g., a fee-sharing arrangement). The quality measures may include keeping readmission rates below a certain level, keeping costs of care low, meeting standards-of-care for certain disease conditions, and the like. As used in this application, the term “organizational data” is meant to be construed broadly to cover a wide range of information regarding healthcare organizations, healthcare providers, and contractual or agreement terms.
The data store builder 218 is configured to build, maintain, and update the data stores 222, 224, and 226 using information received by the receiving component 216. The contract data store 222 stores contractual information between one or more healthcare organizations and their associated payers; the contractual information, as mentioned, includes clinical objectives as well as financial objectives. The organization directory data store 224 stores organizational information pertaining to one or more healthcare organizations, and the provider directory data store 226 stores healthcare provider information. The information stored in the data stores 222, 224, and 226 may arise from disparate organizations or sources that bear no relationship to each other. The data stores 222, 224, and 226 may be networked storage or distributed storage including storage on servers located in the cloud. Thus, it is contemplated that for some embodiments, the information stored in the data stores 222, 224, and 226 is not stored in the same physical location. As new organizational information is received by the receiving component 216, the data store builder 218 updates the existing data stores 222, 224, and/or 226.
The program builder component 220 is configured to, in one aspect, automatically build one or more condition-specific or objective-specific program templates to be used by, for example, a healthcare organization to manage the health of a population segment or to achieve one or more objectives. In another aspect, a user, such as a programmer associated with a healthcare organization, can use the program builder component 220 to build a condition-specific or objective-specific program template that is customized to the particular healthcare organization. In general terms, a condition-specific or objective-specific program template may be defined as a systematic approach to identify, predict, and/or manage an objective or condition at a population, provider, or patient level. The program builder component 220 may have access to knowledge repositories that include algorithms, outcome-related goals, reference materials, standards-of-care, recommendation protocols, and the like. This information may be specific to a healthcare organization or facility, or the information may be promulgated by, for example, nationally-recognized medical organizations or governing bodies and be applicable to multiple different healthcare organizations.
As mentioned, in one aspect, the program templates may be condition specific and designed to better help healthcare organizations manage, for example, a population segment with a certain disease condition. The number and type of conditions are numerous and examples include wellness, heart disease, pre-diabetes, diabetes, stroke, and the like. The program templates are configured, in one aspect, to identify patients within a population segment who have a defined condition or have risk factors that may lead to the development of the condition. The program templates are also configured to stratify patients within the identified segment by degree of severity or risk or by disease type, and to generate interventions or action workflows designed to reduce disease severity or risk and to improve outcome. Additional uses for the condition-specific program templates are to measure outcomes related to treatment interventions, predict outcomes based on the implementation of a proposed action, and also to attribute patients within the identified segment to appropriate care providers (e.g., primary care physicians, health coaches, specialists such as endocrinologists, podiatrists, and the like) best suited to treat the condition in a cost-effective manner.
Program templates may also be objective-specific and designed to assist a healthcare organization reach a defined objective. Objective-specific templates include public health surveillance, contract management, organization management, provider incentives, research solutions, interoperability solutions, creation of research data marts, patient outreach, and the like.
As mentioned, the condition-specific and/or objective-specific program templates may be automatically generated, or may be automatically generated and modified by a user, or may be built entirely by the user. The user in this case, may include a programmer or administrative personnel associated with a healthcare organization. This user may customize the template so that it more directly reflects the characteristics of the organization including any contractual provisions that the organization adheres to. For instance, the user may modify or customize the template by modifying population qualifiers, reference ranges, measurement thresholds, intervention strategies, algorithms, nomenclature concepts, and the like.
Turning back to
Clinical data, as used herein, refers to any healthcare or medical data particular to a patient. In embodiments, clinical data is medical care or healthcare data resulting from or associated with a health or medical service performed in association with a provider in a healthcare environment (e.g., lab test, clinical encounter, ecare, evisit, etc.). Clinical data may include, but is not limited to, a health history of a patient, patient information, patient identifiers, patient demographics (e.g., age, gender, race, etc.), diagnoses, treatments, a family history, genomic history, immunization records, medications, test results, allergies, adverse reactions, procedures performed, social history, advanced directives, alerts, claims data, vital sign information, data traditionally captured at the point of care or during the care process, a combination thereof, and the like. The clinical data may be derived from claims data, data supplied by the patient's insurer, electronic medical record data, device data, health information exchange (HIE) data, personal health records (PHRs), continuity-of-care documents (CCD), patient-inputted data (via, for example, a patient portal), provider-inputted data, pharmacy data, and the like.
Financial data refers to any financial information relevant to a patient, such as insurance data, claims data, payer data, patient-provided data, and the like. For example, a patient may input via a patient portal financial information such as income, purchasing habits, credit card information, etc. Activity data refers to data that may impact a patient's health but is derived from sources outside of or remote from the patient's healthcare environment. For example, activity data may comprise data supplied by county health agencies, state health agencies, communities, research data marts, federal public health agencies, and the like. Representative types of activity data include disease statistics or crime statistics associated with a community or geographic area, socioeconomic characteristics of population segments located in a particular geographic area as indicated by zip code (e.g., age, income level, education level, race, gender, and the like), support services provided by a community such as elder care programs, health education programs, health clinics, church programs, nutrition-assist programs, exercise programs, and the like.
The data store builder 230 is configured to build, maintain, and/or update the population data store 234 using the patient data received by the receiving component 228. The data store builder 230 receives the incoming raw data in various formats or nomenclature including proprietary formats. The raw data may include data that is structured, unstructured or semi-structured. A copy of all the raw data received by the data store builder 230 is stored in its raw form. This enables the population health management system 200 to derive new knowledge using the raw data at a later point in time.
For each source of data, the data store builder 230 transforms the raw data into industry-standard nomenclatures such as Logical Observation Identifiers Names and Codes (LOINC), Unified Medical Language System (UMLS), or Systemized Nomenclature of Medicine (SNOMED), and compiles the data by source into a reference record. Reference records are cleansed, standardized, and machine-readable. The reference records support all industry-standard terminologies, which enables the fidelity of the data sources to be retained. When the raw data is in a proprietary format, the data store builder 230 leverages machine learning techniques to map codes to industry-standard terminologies. Machine learning techniques recommend appropriate standard terminology codes thus speeding up the mapping process. The data store builder 230 is also configured to apply natural language processing techniques to data contained in clinical notes, reports, and documents. The data store builder 230 is configured to reconcile conflicting pieces of data or synthesize pieces of data that should be viewed together and to learn from each data source thereby enabling the mapping and transformation processing times to be reduced for new data sources.
As mentioned, the data store builder 230 is configured to track the source of the data and associate the source with the source's reference record. The reference record for a particular data source contains consent, access, and/or privacy rules associated with the data source. A particular reference record may be utilized by multiple different organizations. This is especially useful in situations where the organizations that utilize a particular reference record change over time. By way of illustrative example, a provider group may provide data to the population health management system 200. The provider group may initially be affiliated with a first healthcare organization and grant the first healthcare organization access rights to the provider group's reference record. At some later point in time, the provider group may cancel its contract with the first healthcare organization and establish a contract with a second healthcare organization. In this case, instead of the provider group having to re-send all of its data to the second healthcare organization, it can simply grant the second healthcare organization access rights to its reference record and withdraw access rights from the first healthcare organization.
The data store builder 230 combines the information contained in the different reference records to form person-centric longitudinal health records known as population records. Each record is longitudinal in that it contains information on all of the patient's health encounters even though the encounters may have occurred at disparate locations and with disparate providers. Each population record is stored in the population data store 234. Thus, the population data store 324 includes multiple population records relevant to a population of patients. The data store builder 230 utilizes the matching component 232 to match patients with their data and create the person-centric longitudinal population records.
The matching component 232 utilizes one or more probabilistic patient matching algorithms to match patients to their clinical, financial, and activity data. Patients may be matched to their data based on for example, patient name, patient identifiers, and the like. For more general data such as socioeconomic conditions by zip code, a patient may be matched to an appropriate set of socioeconomic data using, for example, the patient's zip code. The matching component 232 is also configured to reconcile all of the patient's records even if the records are associated with different patient identifiers. For instance, a pharmacy may use a first patient identifier when dispensing medications to a patient while a care clinic may use a second patient identifier when treating the patient. The matching component 232 is configured to reconcile these patient identifiers and match all of the pertinent medical information associated with the patient. Each patient's data is also normalized via grouping codes from multiple terminologies that mean the same thing thereby reducing redundant data.
Different patient matching algorithms may be utilized by the matching component 232 to generate different population records suitable for different purposes or use cases. For example, one algorithm may be used to create a population record that includes identified, detailed clinical information about the patient. This type of population record is suitable for use by clinical-type programs that need access to granular data about the patient. A different matching algorithm may be used to create a population record that is de-identified (e.g., stripped of patient-identifying data). A de-identified population record may contain less granular information about the patient and be suitable for use by research-type programs after obtaining the patient's consent. Other patient-centric population records may be created for, for instance, analytics purposes.
The population data store 234 may be networked storage or distributed storage including storage on servers located in the cloud. Thus, it is contemplated that for some embodiments, the information stored in the population data store 324 is not stored in the same physical location. The information within the population data store 234 may exist in a variety of standardized formats including, for example, narratives and other data.
At a step 718, the raw data is transformed into industry-standard nomenclatures by, for example, techniques such as cleaning, standardizing, natural language processing, machine learning, and the like. After transformation at step 718, the transformed data is stored in reference records 720, 722, and 725. Each reference record 720, 722, or 724 stores transformed data from a single data source. The reference records 720, 722, and 722 may be networked storage or distributed storage including storage on servers located in the cloud.
At a step 726, the data from the reference records 720, 722, and 724 is combined and undergoes a second transformation process before being stored in patient-centric population records 728, 730, and 732. The transformation process at step 726 includes application of probabilistic patient matching algorithms, normalization, and reconciliation as explained above. The population records 728, 730, and 732 may be networked storage or distributed storage including storage on servers located in the cloud.
Different programs 734, 736, and 738 can access the population records 728, 730, and 732 or sub-portions of the population records 728, 730, and 732. The programs 734, 736, and 738 may be condition-specific such as decision support, patient registries, care management, predictive risk, and the like. Likewise, the programs 734, 736, and 738 may be objective-specific such as public health surveillance, interoperability solutions, attribution, outreach, intelligent learning systems, contract management, provider incentive, physician experience/outcome, network management, and the like. A particular population record may be accessed by more than one program. Likewise, a particular program may access a plurality of population records.
At a step 740, different end-users are able to utilize the output of the programs 734, 736, and/or 738 via one or more computer applications. End-users are numerous but representative examples include population managers, primary care physicians, care managers or health coaches, specialists, patients, healthcare organization administrators, payers, and the like.
Turning back to
The program engine 214 is configured, at run-time, to access patient population data stored in population records in the population data store 234 and execute the customized condition-specific or objective-specific programs against one or more of the population records, or sub-portions of the population records to generate patient population data for the healthcare organization.
The program templates may include identification programs 238 that identify different types of population segments. For example, population segments may be identified based on a condition such as, for example, wellness, diabetes, stroke, hypertension, and the like. Population segments may be identified based on affiliation with a particular healthcare organization or a particular provider. Additionally, population segments may be identified that have a condition and are eligible for different outreach programs. The type of identification is dependent upon the underlying purpose of the program. The identified segment along with associated patient data is stored in a registry 248 and is available for use by various applications utilized by the healthcare organization.
Another program is a prediction program 240 that takes a population segment identified by the identification program 238 and stratifies patients within the segment based on disease type, and/or severity of the disease condition, and/or severity of risk for contracting the disease condition. For example, a diabetes population may be identified and then stratified based on whether the patient has Type 1 Diabetes versus Type 2 Diabetes, whether the diabetes is controlled versus uncontrolled, and/or whether the patient(s) is compliant with medication versus non-compliant. Other stratification schemes may be employed such as stratifying by cost of care, demographic features, and the like. The stratification information is stored in a stratification data store 250 that is available for use by various applications utilized by the healthcare organization.
Yet another program is an attribution program 244 that attributes patients to various healthcare providers within a healthcare organization. Attribution is geared towards pairing a patient with an optimal set of healthcare providers to effectively manage the patient while keeping costs down. As explained above, the attribution program may be customized to the healthcare organization or may be a standard attribution program. Turning now to
At a step 812, organizational data associated with the healthcare organization is accessed. The organizational data may be stored in association with a data store such as the contracts directory 222, the organization directory 224, and the provider directory 226 of
The organizational data also includes objectives specified in one or more payer contracts that the healthcare organization has entered into. The objectives specified in the payer contracts may indicate that providers meet quality measures for a predefined number of patients (these patients are known as “scorable patients”) before the provider and/or the organization is reimbursed and/or receives incentive fees from the payer. If the attribution program is a standard program that is currently not customized to the healthcare organization, organization-specific parameters may be accessed. These parameters specify attribution rules for the healthcare organization. For example, the healthcare organization may have a rule specifying that patients with low levels of complexity be assigned solely to a primary care physician, while patients with high levels of complexity be assigned to a primary care physician as well as a specialist.
At a step 814, the program engine again accesses the population records of those patients identified as being affiliated with the healthcare organization to determine, among other things, previous and current provider relationships associated with the patients. This is possible because each population record is longitudinal in nature and contains a history of all healthcare encounters associated with the patient, even if the encounters occurred at disparate locations and disparate times. Other examples of patient information that is accessed include geographic location of the patient, disease conditions, personal preferences regarding the type of provider the patient wishes to see (e.g., female versus male, board certified, older versus younger, specialist versus primary care, and the like), socioeconomic information as determined by, for example, the zip code at which the patient resides, demographic information (e.g., gender, age, and the like), financial information as determined by claims data or patient-provided data, genomic information associated with the patient (e.g., whether the patient has been identified as having a gene mutation leading to an increased risk of cancer or disease), and whether the patient is disabled and requires specialized access to provider offices. Patient data may also include social information such as the presence of family care givers, alcohol and tobacco use, church affiliations, use of support services, and the like. Some of this information may be supplied directly by the patient through a patient portal while other information is gleaned from different data sources as explained above.
At a step 816, the program engine attributes each of the patients associated with the healthcare organization to one or more providers within the healthcare organization based on both the organizational data and the patient data. The program engine is designed to take into account each piece of organizational data and patient data when defining the attribution scheme with the result that the attribution scheme is not only customized to the healthcare organization but is personalized to the patient. For example, a first healthcare organization may utilize health coaches while a second healthcare organization does not. An attribution scheme for the first healthcare organization would include health coaches while an attribution scheme for the second healthcare organization would not include health coaches.
In one aspect, a tiered attribution scheme may be utilized wherein a patient is first attributed to a certain type of provider based on the organizational data, and then the actual identity of the provider is determined based on patient data. For example, a diabetic patient may first be attributed to a primary care physician and an endocrinologist based on organization-specific parameters or rules. Patient data is then used to determine, for example, a primary care physician and an endocrinologist within geographic proximity to the patient and who meet patient preference factors. The output of the attribution program is stored in association with an attributed relationships data store such as the attributed relationships data store 254 of
In one aspect, an existing attribution scheme can be modified by providers. In this aspect, the provider may be presented with a list of patients that have been attributed to him or her. The provider has the option to accept and/or reject the attribution scheme. If accepted, the attribution scheme is implemented. If the provider chooses to reject the attribution scheme, the provider is given the option of proposing a new attribution scheme. For example, in an intra-office setting, the provider may suggest that a patient be attributed to another provider within the group. Based on the proposal, a segment of the attribution program may be re-executed to determine if the proposed change is feasible based on organizational and patient data. In another aspect, a healthcare organization is able to run an attribution scenario using organizational data to see if the attribution scenario is feasible before actually implementing the attribution scenario.
At a step 914, the population records of those patients who are currently affiliated with the healthcare organization are again accessed to determine patient information such as previous provider relationships, current provider relationships, patient preferences, social history, and the like. At a step 916, each of the patients is attributed to one or more providers based on the organizational data and the patient data.
At a step 918, updates to one or both of the organizational data and/or the patient data are received. The types of updates may include any change to the organizational data and/or the patient data such as the addition of a new patient to the network, changes in patient or provider location, updated patient preferences, change in patient disease state that may necessitate the addition of another provider to care for the patient or the deletion of a currently-attributed provider, the addition or deletion of providers associated with the healthcare organization, a modification to a payer contract that changes incentive or reimbursement objectives, new organization attribution rules, and the like.
In response to the updated organizational and/or patient data, the attribution scheme may be automatically modified at a step 920. Like above, the modification is made based on both the organizational data and the patient data. By way of illustrative example, updated organizational data may indicate that a provider has moved his/her office location. A patient currently attributed to the provider may initially be attributed to another provider that is located closer to the patient. However, this patient's preferences may indicate that the patient wishes to maintain the relationship with the current provider regardless of the provider's location. In another case, a patient may indicate that geographic location of the provider is his/her primary concern and, thus, this patient would be attributed to another provider that is located closer to the patient.
With this as a background, and turning to
At a step 1014, population records of those patients currently associated with the healthcare organization are accessed and pertinent patient data related to attribution is identified. At a step 1016, each of the patients is attributed to one or more providers based on both the organizational data and the patient data as outlined above.
At a step 1018, and with respect to at least one of the providers, patients attributed to the provider who are scorable are identified. Again, this may be based on information contained in the population records and on contract data. A patient is identified as scorable if he or she meets the parameters spelled out in the payer contract. For instance, a payer contract may specify that a patient is scorable for diabetes management if the patient is between a certain age range and suffers from Type 1 diabetes. Patients meeting these criteria who have been attributed to the provider are thus identified as scorable.
At a step 1020, a subset of the scorable patients is attributed to the provider as a scorable patient group. Again, this may be based on terms in the payer contract that specify how many patients should be in a scorable group in order to measure achievement of quality measures. Thus, for example, although a provider may be attributed 1000 patients, only 200 of these may be considered scorable, and of these, only 60 are placed in the measured scorable group.
Yet another program is a contract and organization management program (not shown in
With this as a background, and turning to
At a step 1112, patient data of patients in the scorable patient groups is accessed from population records stored in association with, for example, the population data store 234. The patient data accessed includes data related to the quality measure contract objectives. The type of data accessed is dependent on the nature of the quality measure but may include lab results, procedure results, diagnoses, indications of whether actions have been taken or attempted to be taken, readmission rates, and the like. After being accessed the patient data may also be stratified based on, for example, scorable patient group, payer, attribution to a particular provider, and the like.
At a step 1114, for each identified scorable patient group, it is determined the number of or the percentage of patients in the scorable patient group who meet the quality measure(s) as specified by the quality measure contract objectives. At a step 1116, a determination is made for each scorable patient group whether the percentage of patients in the scorable group that meet the respective quality measure is equal to or greater than the percentage or number set forth in the quality measure contract objective. Additionally, a determination may be made of the percentage or number of providers associated with the healthcare organization whose patients meet the quality measure contract objectives.
If, at a step 1118, it is determined that the percentage is equal to or greater than that set forth in the quality measure contract objectives, then an amount of financial incentive is determined. Again, the amount of the incentive is dictated by the contract terms and may be scaled based on the percentage met. The amount of financial incentive may be determined at the provider level and the organization level.
If, however, at a step 1120, it is determined that the percentage is less than that specified in the quality measure contract objectives, then one or more recommendations are automatically generated that will enable the healthcare organization to better meet the contract objectives. The recommendations may include changing the current attribution scheme to re-align poor performing patients with providers that have historically better provider outcomes, changing the current attribution scheme to address the over- or under-utilization of providers or departments (a concept known as venue shifting), making recommendations on types of providers that should be hired to better assist patients in meeting quality measures, and the like. By way of illustrative example, outcome data may indicate that providers in the organization's emergency room department are not meeting quality measure contract objectives because the department is being over utilized. Based on the type of quality measures not being met, a recommendation may be generated suggesting that, for example, a health coach and a primary care physician be hired by the organization. Recommendations may also be made as to which quality measures need most improvement so that the healthcare organization may make appropriate changes including, for example, increasing efforts at patient outreach. Besides generating recommendations at step 1120 if contract objectives are not met, the financial impact of not meeting those contract objectives may also be determined. This may be broken down by provider, provider groups, provider type, payers, type of quality measures, and the like.
Besides the method outlined in
The information generated by the contract and organization management program may be presented to the organization on one or more user interfaces (UIs) as shown in
Area 1316 presents selectable recommendations automatically generated by the contract and organization management program. The recommendations address areas of concern as indicated by the organizational and patient data. In one aspect, the recommendations comprise venue shifting or provider shifting recommendations where a different care scheme other than the existing care scheme is recommended. The venue shifting or provider shifting recommendations are based on organizational data for the specific healthcare organization. For example, as shown at numeral 1318, the system has made a recommendation that 2% of primary care physician visits be shifted to eVisit (e.g., contacting a patient electronically to determine how the patient is doing) based on current care distribution information. This recommendation leverages organization information indicating that the healthcare organization has capabilities to implement eVisits. The organization can select which recommendations it would like to implement and additional graphical representations showing the impact of the selected recommendations are generated and presented as shown in
Continuing, another program is an outreach program (not shown in
Turning to
At a step 1612, organizational data is accessed from, for example, the organization directory 224 and the provider directory 226 of
A record of attempts to contact the patient is maintained. This information is useful to establish an audit trail that may allow the provider and/or the organization to be excused from meeting a certain quality measure contract objective. For example, contract terms may provide that performance of a quality measure contract objective is excused or considered met if the provider or organization makes a certain number of attempts to contact the patient. Keeping a record of attempted patient contact is important for meeting this type of contract term.
For patients who qualify for multiple outreach events related to different quality measures, the method 1600 is adapted to synthesize the multiple different quality measures and generate a single outreach event or communication related to the multiple different quality measures using the organizational data and the patient data. For example, a person may need to have his or her blood drawn for hemoglobin A1C testing as well as receive a mammogram. Organizational data and patient data is used to generate a single communication that addresses both of these quality measures and includes, for example, a single time frame and/or location that enables the patient to accomplish both of these quality measures and that further meets the patient's preferences.
Returning to
An additional program may include a measurement program 242 that measures various outcomes associated with the identified population segment. Outcomes may include, for example, readmission rates, medication compliance rates, re-infection rates, patient satisfaction results, compliance with standards-of-care, and the like. The output of the measurement program 242 is stored in association with a measures data store 252 that is available for use by the healthcare organization to improve quality control measures, etc. The information may also be utilized by payers to determine rates of reimbursement. For instance, a payer may have incentive reimbursements if readmission rates are kept below a predefined threshold. In another example, reimbursement rates may be reduced if the patient satisfaction results are poor. Any and all such aspects are contemplated as being within the scope of the invention.
As mentioned the output of the programs 238, 240, 242, 244, and 246 and the other programs set forth above may be used by a healthcare organization via one or more computer applications in a variety of ways. Some examples include using the data to initiate alerts, provide access to patient records, create patient lists or registries, generate summary pages, initiate quality control measures, and the like. Another example is a score card application that utilizes a builder program to enable the organization to build a scoring plan for itself, its providers, its payers, and/or its patients and present the results of the score plan on one or more score card user interfaces. Execution of the score card application leverages the organizational data, the patient-centric longitudinal population records, and the output of some or all of the programs outlined above.
With respect to the builder program for the score card application,
Turning to
With respect to the organization 2016, options are provided for scoring the organization by providers or patients at area 2030. Like with providers 2014, operators and targets can be selected or inputted. Thus, as shown in
Once the score plan has been built and implemented, the results are presented on score card UIs. The score card UIs may be configured to present outcome data that is relevant to a provider, a provider group, and/or the organization as a whole. One example of a provider score card is depicted in
The UI 2400 depicts the top five quality measures 2422, 2424, 2426, 2428, and 2430 for which the provider 2410 is demonstrating the least progress towards completion, thus enabling the provider 2420 to quickly assess areas for improvement. The quality measures 2422, 2424, 2426, 2428, and 2430 are presented in the largest rectangles to help draw the user's attention. For instance, with respect to the quality measure 2422 “Diabetes Care,” the provider 2410 is 0% of the way towards achieving the total score for this quality measure 2422. Additional information may be presented upon hovering over the various rectangles. For instance, the number of patients in the scorable patient group for the particular quality measure may be presented, along with the number of patients in the scorable patient group who have achieved the particular quality measure. The determination of which top five quality measures to present on the UI 2400 is determined by the system. The system may base its determination of the top five quality measures 2422, 2424, 2426, 2428, and 2430 based on not only the percent progress towards completion, but also the number of patients in the scorable group, the impact of achievement of the quality measure on the provider's overall total score with the least amount of effort, and the like.
The UI 2500 contains some of the same elements as the UI 2400 and like numbering is used to label these elements. The UI 2500 includes a treemap 2520 displaying all of the quality measures against which the provider 2410 is being scored. Instead of percentage of completion information being presented however, opportunity index information is presented in association with the quality measures. The information in the UI 2500 is useful at the beginning of the provider's scoring period and enables the provider 2410 to quickly see what areas offer the most opportunity to improve the provider's overall score. The top five opportunity index quality measures as determined by the system are shown in as 2522, 2524, 2526, 2528, and 2530. The quality measures 2522, 2524, 2526, 2528, and 2530 are presented in the largest rectangles to draw the user's attention. Using the quality measure “Pediatric Wellness” 2522 as an example, the provider 2410 has an opportunity to improve his overall score by 0.6% if he meets this quality measure. The provider scorecard UI 2500 may dynamically shift to the provider scorecard UI 2400 as the scoring year progresses allowing the provider to see where he or she can improve with the least amount of effort.
Score cards can also be generated at a provider group level and/or organization level.
The UI 2600 includes the provider group name 2610 as well as a bar graph 2612 that illustrates the percentage towards completion of the total number of scorable points for the provider group 2610. For example, the provider group 2610 is 42% towards completion of the total number of scorable points (indicated by element 2614). The UI 2600 also indicates the total number of patients 2618 attributed to the group 2610 and the total number of patients in scorable patient groups 2616 attributed to the group 2610.
A treemap 2620 is presented that includes all of the quality measures against which the provider group 2610 is being scored. Like the UI 2500, the top five opportunity index quality measures 2622, 2624, 2626, 2628, and 2630 are presented along with opportunity index percentages. The treemap 2620 may be presented at the beginning of the provider group's scorable year and dynamically transition to a treemap similar to the UI 2400 that displays percentage toward completion data as the scoring year progresses. The UI 2600 also includes a graph 2632 displaying trending information regarding the completion of quality measures by the provider group.
Although not shown, it is contemplated that score cards could be generated that display quality measure information about individual patients. This may be useful when a patient attributed to a provider suffers from multiple medical conditions that place him or her into multiple quality measure scorable patient groups. Patients of this type may have a large impact on the provider's total score, and the provider would be interested in knowing what quality measures associated with the patient may be most difficult to meet (e.g., at the beginning of the provider's scoring year), and/or which of the patient's quality measures could be improved with the least amount of effort by the provider (e.g., later in the provider's scorable year). Information on the patient's quality measures could be presented in the form of a treemap as shown in UIs 2400, 2500, and 2600.
In one aspect, a healthcare organization can run the programs outlined above in a demonstration mode and generate a projected set of data for analysis.
The population health management system 200 is an closed-loop system meaning that as a healthcare organization utilizes the output of the programs outlined above, more organizational and patient data is generated which is fed back into the system 200 and used to update the various data stores. Further, the system 200 operates over and is compatible with existing electronic medical record systems associated with healthcare organizations, even if these electronic medical record systems are disparate in nature. Thus, a healthcare organization can utilize the system to manage population health without having to incur significant financial costs to reconfigure its already-existing electronic medical record system.
Turning now to
Once the condition-specific program template is extracted at the step 310, it is customized or localized to a particular healthcare organization at a step 312. Customization includes taking into account the organizational data associated with the healthcare organization. This data may be stored in data stores such as the contracts data store 222, the organization directory data store 224, and the provider directory data store 226 of
At a step 414, the program engine executes the customized condition-specific program templates against the patient data. The output of the program engine comprises population health data that can be used by the healthcare organization to manage the health of a population segment. For instance, the program engine may generate a list of patients associated with the healthcare organization who are suffering from a disease condition. The patients may be stratified by disease severity, and the patients may be attributed to one or more providers associated with the healthcare organization who can efficiently manage the disease condition. Workflows may be generated to improve outcome and efficiently utilize the attributed healthcare providers. Intervention outcomes are measured; these outcomes may be used by payers to determine an amount of reimbursement to the healthcare organization.
The present invention has been described in relation to particular embodiments, which are intended in all respects to be illustrative rather than restrictive. Further, the present invention is not limited to these embodiments, but variations and modifications may be made without departing from the scope of the present invention.
Claims
1. A system for organizational management of patient population health, the system comprising:
- a computing device associated with a population health management service having one or more processors and one or more computer-storage media; and
- one or more data stores coupled with the population health management service,
- wherein the population health management service: receives a plurality of sets of organizational data associated with one or more healthcare organizations; stores the plurality of sets of organizational data in association with the one or more data stores; receives raw data associated with a population of patients from one or more raw data sources; transforms the raw data into one or more standard medical nomenclatures; and generates a plurality of population records using the transformed data, wherein each of the plurality of population records comprises a longitudinal record of a particular patient's data.
2. The system of claim 1, wherein the one or more healthcare organizations are disparate.
3. The system of claim 1, wherein the plurality of sets of organizational data comprises at least data about the one or more healthcare organizations including at least operating hours, type of care provided, and geographic location.
4. The system of claim 1, wherein the plurality of sets of organizational data comprises at least data about one or more providers associated with at least one of the one or more healthcare organizations including at least provider names, areas of provider practice, and geographic location of providers.
5. The system of claim 1, wherein the plurality of sets of organizational data comprises at least data about one or more contracts between the one or more healthcare organizations and respective payers associated with the one or more healthcare organizations including at least quality measure contract objectives contained in the one or more contracts.
6. The system of claim 5, wherein the at least quality measure contract objectives comprise clinical quality measures against which a healthcare organization or a provider is scored.
7. The system of claim 1, wherein the one or more raw data sources from which the raw data associated with the population of patients is received are disparate raw data sources.
8. The system of claim 7, wherein the raw data associated with the population of patients is received in different formats or nomenclatures.
9. The system of claim 1, wherein the one or more raw data sources from which the raw data associated with the population of patients is received comprise at least clinical data sources, financial data sources, and federal and state data sources.
10. A computerized method carried out by a population health management service having at least one processor for creating patient-centric longitudinal population records, the method comprising:
- receiving from a plurality of raw data sources raw data associated with a population of patients;
- for each raw data source of the plurality of raw data sources, transforming, using the at least one processor, the each raw data source's raw data into standard medical nomenclature and compiling the transformed data into a reference record;
- for each patient in the population of patients, extracting the each patient's data from a set of reference records; and
- for the each patient in the population of patients, generating a longitudinal record of the each patient's data.
11. The computerized method of claim 10, wherein subsequent to receiving the raw data associated with the population of patients from the plurality of raw data sources, storing a copy of the raw data.
12. The computerized method of claim 10, wherein the raw data is transformed into standard medical nomenclature using at least machine learning techniques and natural language processing.
13. The computerized method of claim 10, wherein the standard medical nomenclature comprises at least one of Logical Observation Identifiers Names and Codes (LOINC), Unified Medical Language System (UMLS), or Systemized Nomenclature of Medicine (SNOMED).
14. The computerized method of claim 10, wherein each reference record includes consent, access, and privacy rules associated with the each reference record's raw data source.
15. The computerized method of claim 10, wherein the longitudinal record of the each patient's data is generated using one or more probabilistic patient matching algorithms.
16. The computerized method of claim 10, wherein the longitudinal record of the each patient's data is identified.
17. The computerized method of claim 10, wherein the longitudinal record of the each patient's data is de-identified.
18. One or more non-transitory computer-readable media having computer-executable instructions embodied thereon that, when executed, perform a method of enabling one or more healthcare organizations to manage health of one or more patient population segments, the method comprising:
- receiving a plurality of sets of organizational data associated with the one or more healthcare organizations, the plurality of sets of organizational data comprising contract data, provider data, and healthcare organization-specific data;
- storing the plurality of sets of organizational data in one or more data stores;
- receiving from a plurality of raw data sources raw data associated with a population of patients;
- for each raw data source of the plurality of raw data sources, transforming the each raw data source's raw data into standard medical nomenclature and compiling the transformed data into a reference record;
- for each patient in the population of patients, extracting the each patient's data from a set of reference records; and
- for the each patient in the population of patients, generating a longitudinal population record of the each patient's data; and
- executing at least one of a disease condition-specific program or objective-specific program against at least a portion of the plurality of sets of organizational data stored in association with the one or more data stores and longitudinal population records of at least a portion of the population of patients to generate one or more results used by at least one of the one or more healthcare organizations to manage the health of a patient population segment.
19. The media of claim 18, wherein the at least one disease condition-specific program or the objective-specific program is customized to the at least one of the one or more healthcare organizations based on parameters specified by the at least one of the one or more healthcare organizations.
20. The media of claim 19, wherein the at least one disease condition-specific program or the objective-specific program is customized by personnel associated with the at least one of the one or more healthcare organizations using a logic editor.
Type: Application
Filed: Oct 8, 2013
Publication Date: Apr 10, 2014
Applicant: CERNER INNOVATION, INC. (Kansas City, KS)
Inventors: NEAL PATTERSON (CAPTIVA, FL), RYAN HAMILTON (KANSAS CITY, MO), BHARAT SUTARIYA (PARKVILLE, MO), TEHSIN SYED (KANSAS CITY, MO)
Application Number: 14/048,349
International Classification: G06F 19/00 (20060101);
