U.S. patent application number 12/889904 was filed with the patent office on 2011-03-31 for systems and methods of clinical tracking.
Invention is credited to Paul Bleicher, Agneta Breitenstein, Stanley Huang, Donald Pettini, Ryan Scharer, Hua Ye.
Application Number | 20110077972 12/889904 |
Document ID | / |
Family ID | 43781302 |
Filed Date | 2011-03-31 |
United States Patent
Application |
20110077972 |
Kind Code |
A1 |
Breitenstein; Agneta ; et
al. |
March 31, 2011 |
SYSTEMS AND METHODS OF CLINICAL TRACKING
Abstract
The clinical analytics platform automates the capture,
extraction, and reporting of data required for certain quality
measures, provides real-time clinical surveillance, clinical
dashboards, tracking lists, and alerts for specific, high-priority
conditions, and offers dynamic, ad-hoc quality reporting
capabilities. The clinical informatics platform may include a data
extraction facility that gathers clinical data from numerous
sources, a data mapping facility that identifies and maps key data
elements and links data over time, a data normalization facility to
normalize the clinical data and, optionally, de-identify the data,
a flexible data warehouse for storing raw clinical data or
longitudinal patient data, a clinical analytics facility for data
mining, analytic model building, patient risk identification,
benchmarking, performing quality assurance, and patient tracking,
and a graphical user interface for presenting clinical analytics in
an actionable format. The clinical informatics platform may enable
a method of clinical tracking that includes analyzing the
healthcare data to obtain at least one report, and presenting the
report in a graphical user interface, wherein the report can be
customized based on a criterion.
Inventors: |
Breitenstein; Agneta;
(Somerville, MA) ; Huang; Stanley; (Wellesley,
MA) ; Pettini; Donald; (Andover, MA) ;
Bleicher; Paul; (Boston, MA) ; Scharer; Ryan;
(Somerville, MA) ; Ye; Hua; (Belmont, MA) |
Family ID: |
43781302 |
Appl. No.: |
12/889904 |
Filed: |
September 24, 2010 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61245581 |
Sep 24, 2009 |
|
|
|
61249305 |
Oct 7, 2009 |
|
|
|
Current U.S.
Class: |
705/3 |
Current CPC
Class: |
G16H 10/60 20180101;
G16H 50/30 20180101; G06Q 50/22 20130101; G06F 3/0481 20130101;
G06Q 10/10 20130101; G06F 3/0484 20130101 |
Class at
Publication: |
705/3 |
International
Class: |
G06Q 50/00 20060101
G06Q050/00; G06Q 10/00 20060101 G06Q010/00 |
Claims
1-22. (canceled)
23. A method of clinical tracking, comprising: gathering healthcare
data from a plurality of sources; processing the healthcare data,
wherein processing comprises identifying, mapping and normalizing
healthcare data elements, wherein the mapping comprises assigning
data to a field of a database according to a hierarchically
organized lexicon of healthcare data elements, wherein multiple
data element entries in the lexicon are mapped to a single field
for at least one field; analyzing the healthcare data to obtain at
least one report; and presenting the report in a graphical user
interface, wherein the report can be customized based on a
criterion.
24. The method of claim 23, wherein the report identifies at least
one risk relevant to at least one patient based at least in part on
the gathered healthcare data.
25. The method of claim 23, wherein the report comprises an alert
relating to at least one risk associated with at least one patient
based at least in part on the gathered healthcare data, such alert
presented in at least one of an audible or visual manner.
26. The method of claim 23, wherein the report comprises an alert
identifying at least one patient care error and at least one
recommendation for correcting such at least one error.
27. The method of claim 23, wherein the report comprises
instructions for the manner in which one or more healthcare
providers are to provide care to one or more patients based at
least in part on the gathered healthcare data.
28. The method of claim 23, wherein the report identifies a
disparity between the available healthcare resources and the
patient needs identified based at least in part on the gathered
healthcare data.
29. The method of claim 23, wherein the report identifies a
high-cost patient based at least in part on the gathered healthcare
data.
30. The method of claim 23, wherein processing the healthcare data
also comprises validating the healthcare data elements.
31. The method of claim 23, wherein the data are gathered on a
periodic basis.
32. The method of claim 23, wherein the data are gathered on a
real-time basis, the report comprises instructions for the manner
in which one or more healthcare providers are to provide care to
one or more patients based at least in part on the gathered
healthcare data and the report is updated on a real-time basis.
33. The method of claim 32, wherein the real-time basis is at least
as frequent as every five minutes.
34. The method of claim 23, wherein the graphical user interface is
presented via a software-as-a-service architecture.
35. The method of claim 23, wherein the report relates to at least
one of a patient, a medical care protocol, an outcome, a
demographic, a behavioral risk factor, a disease risk factor, a
procedure, a therapeutic, a therapeutic over a given time period, a
risk level, a cost, an admission information, a utilization,
readmission information, mortality, and a complication.
36. The method of claim 23, wherein the criterion comprises at
least one of a patient name, an issue, a physician, a location, a
due by time for care or therapy, a risk level, a clinical measure,
a procedure completed and an image taken.
37. A method of optimizing a healthcare resource plan, comprising:
gathering healthcare data relating to a plurality of patients from
a plurality of sources, wherein the data are gathered on a periodic
basis; processing the healthcare data, wherein processing comprises
identifying, mapping and normalizing healthcare data elements,
wherein processing is repeated when new healthcare data are
gathered, wherein the mapping comprises assigning data to a field
of a database according to a hierarchically organized lexicon of
healthcare data elements, wherein multiple data element entries in
the lexicon are mapped to a single field for at least one field;
analyzing the healthcare data to obtain at least one patient risk
identification and patient tracking report, wherein analyzing is
repeated when new healthcare data are gathered and processed; and
preparing a healthcare resource plan for care of the plurality of
patients and optimizing the healthcare resource plan based on the
data contained in the at least one patient risk identification and
patient tracking report.
38. The method of claim 37, wherein the periodic basis is in
real-time.
39. The method of claim 38, wherein the real-time basis is at least
as frequent as every five minutes.
40. The method of claim 37, wherein processing the healthcare data
also comprises validating the healthcare data elements.
41. The method of claim 37, further comprising, re-optimizing the
healthcare resource plan when new healthcare data are gathered,
processed, and analyzed.
42. The method of claim 37, further comprising, re-optimizing the
healthcare resource plan when a manual change is made to an element
of the plan.
43. The method of claim 37, wherein the tracking report relates to
at least one of a patient, a medical care protocol, an outcome, a
demographic, a behavioral risk factor, a disease risk factor, a
procedure, a therapeutic, a therapeutic over a given time period, a
risk level, a cost, an admission information, a utilization,
readmission information, mortality, and a complication.
44. The method of claim 37, wherein patients at risk are
automatically detected by the analysis and an alert is generated
identifying such patients.
45. The method of claim 37, wherein high-cost patients are
automatically detected by the analysis and an alert is generated
identifying such patients.
46. The method of claim 37, wherein the healthcare resource plan is
presented in a graphical user interface via a software-as-a-service
architecture.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of the following
provisional applications, each of which is hereby incorporated by
reference in its entirety: U.S. Application Ser. No. 61/245,581,
filed Sep. 24, 2009; and U.S. Application Ser. No. 61/249,305,
filed Oct. 7, 2009.
BACKGROUND
[0002] 1. Field
[0003] The present invention relates to clinical informatics, a
clinical informatics platform and managing networks of health care
providers.
[0004] 2. Description of the Related Art
[0005] Health care systems are evolving at an unprecedented pace.
While much remains uncertain, one thing is clear: knowledge is one
key to success in times of change and uncertainty. The ability to
meaningfully capture, report, and use data to deliver better and
more cost-effective health care is critical. Value-based purchasing
is making clinical performance improvement more important than
ever. Success requires the ability to connect knowledge with action
to improve performance.
[0006] There remains a need for a clinical informatics platform
that automates the capture, extraction, and reporting of data
required for certain quality measures; provides real-time clinical
surveillance, clinical dashboards, tracking lists, and alerts for
specific, high-priority conditions; provides improved methods and
systems for analyzing and managing health care referral networks;
and offers dynamic, ad-hoc quality reporting capabilities.
SUMMARY
[0007] The clinical analytics platform automates the capture,
extraction, and reporting of data required for certain quality
measures; provides real-time clinical surveillance, clinical
dashboards, tracking lists, and alerts for specific, high-priority
conditions; provides improved methods and systems for analyzing and
managing health care referral networks; and offers dynamic, ad-hoc
quality reporting capabilities.
[0008] In an aspect of the invention, a method of clinical tracking
may include gathering healthcare data from a plurality of sources,
processing the healthcare data, wherein processing comprises
identifying, mapping and normalizing healthcare data elements,
wherein the mapping comprises assigning data to a field of a
database according to a hierarchically organized lexicon of
healthcare data elements, wherein multiple data element entries in
the lexicon are mapped to a single field for at least one field,
analyzing the healthcare data to obtain at least one report, and
presenting the report in a graphical user interface, wherein the
report can be customized based on a criterion. The report may
identify at least one risk relevant to at least one patient based
at least in part on the gathered healthcare data. The report may
include an alert relating to at least one risk associated with at
least one patient based at least in part on the gathered healthcare
data, such alert presented in at least one of an audible or visual
manner. The report may include an alert identifying at least one
patient care error and at least one recommendation for correcting
such at least one error. The report may include instructions for
the manner in which one or more healthcare providers are to provide
care to one or more patients based at least in part on the gathered
healthcare data. The report may identify a disparity between the
available healthcare resources and the patient needs identified
based at least in part on the gathered healthcare data. The report
may identify a high-cost patient based at least in part on the
gathered healthcare data. Processing the healthcare data also may
include validating the healthcare data elements. The data may be
gathered on a periodic basis. The data may be gathered on a
real-time basis, the report may include instructions for the manner
in which one or more healthcare providers are to provide care to
one or more patients based at least in part on the gathered
healthcare data and the report is updated on a real-time basis. The
real-time basis may be at least as frequent as every five minutes.
The graphical user interface may be presented via a
software-as-a-service architecture. The report may relate to at
least one of a patient, a medical care protocol, an outcome, a
demographic, a behavioral risk factor, a disease risk factor, a
procedure, a therapeutic, a therapeutic over a given time period, a
risk level, a cost, an admission information, a utilization,
readmission information, mortality, and a complication. The
criterion may include at least one of a patient name, an issue, a
physician, a location, a due by time for care or therapy, a risk
level, a clinical measure, a procedure completed and an image
taken.
[0009] In an aspect of the invention, a method of optimizing a
healthcare resource plan may include gathering healthcare data
relating to a plurality of patients from a plurality of sources,
wherein the data are gathered on a periodic basis, processing the
healthcare data, wherein processing may include identifying,
mapping and normalizing healthcare data elements, wherein
processing is repeated when new healthcare data may be gathered,
wherein the mapping may include assigning data to a field of a
database according to a hierarchically organized lexicon of
healthcare data elements, wherein multiple data element entries in
the lexicon are mapped to a single field for at least one field,
analyzing the healthcare data to obtain at least one patient risk
identification and patient tracking report, wherein analyzing is
repeated when new healthcare data may be gathered and processed,
and preparing a healthcare resource plan for care of the plurality
of patients and optimizing the healthcare resource plan based on
the data contained in the at least one patient risk identification
and patient tracking report. The periodic basis may be in
real-time. The real-time basis may be at least as frequent as every
five minutes. Processing the healthcare data also may include
validating the healthcare data elements. The method may further
include re-optimizing the healthcare resource plan when new
healthcare data may be gathered, processed, and analyzed. The
method may further include re-optimizing the healthcare resource
plan when a manual change is made to an element of the plan. The
tracking report may relate to at least one of a patient, a medical
care protocol, an outcome, a demographic, a behavioral risk factor,
a disease risk factor, a procedure, a therapeutic, a therapeutic
over a given time period, a risk level, a cost, an admission
information, a utilization, readmission information, mortality, and
a complication. Patients at risk may be automatically detected by
the analysis and an alert is generated identifying such patients.
High-cost patients may be automatically detected by the analysis
and an alert is generated identifying such patients. The healthcare
resource plan may be presented in a graphical user interface via a
software-as-a-service architecture.
[0010] In an aspect of the invention, a method of comparative
healthcare benchmarking may include gathering healthcare data from
a plurality of sources, processing the healthcare data, wherein
processing may include wherein processing may include identifying,
mapping and normalizing healthcare data elements, wherein the
mapping may include assigning data to a field of a database
according to a hierarchically organized lexicon of healthcare data
elements, wherein multiple data element entries in the lexicon are
mapped to a single field for at least one field, analyzing the
healthcare data to obtain at least one of a clinical, operational
and financial benchmark, repeating the steps of gathering,
processing, normalizing, and analyzing to obtain a data sample to
compare with the at least one clinical, operational, or financial
benchmark, wherein at least one change is made in at least one of
the repeated steps, and presenting the data sample with the
benchmark as a report in a graphical user interface, wherein the
report can be customized by at least one of changing at least one
criterion. The data may be gathered on a periodic basis. The data
may be gathered on a real-time basis, such as at least as frequent
as every five minutes. The plurality of sources may include sources
relating to different geographic regions. The plurality of sources
may include sources relating to different healthcare facilities.
The plurality of sources may include sources relating to a
specified geographic region. Processing the healthcare data also
may include validating the healthcare data elements. The method may
further include linking the healthcare data elements over time to
form a longitudinal data record. The graphical user interface may
be presented via a software-as-a-service architecture. The at least
one criterion may be a data source, a time period, a chart type, a
time interval for display, a time interval for analysis, a filter,
a hospital, a physician, a patient, a patient characteristic, a
cohort, a disease, a gender, an age group, a treatment, a payer
type and an insurance provider.
[0011] In an aspect of the invention, a benchmarking and
comparative analytics dashboard may include a clinical informatics
facility, including a data extraction facility that gathers
clinical data from numerous sources, a data mapping facility that
identifies and maps key data elements and links data over time,
wherein the mapping may include assigning data to a field of a
database according to a hierarchically organized lexicon of
healthcare data elements, wherein multiple data element entries in
the lexicon are mapped to a single field for at least one field, a
data normalization facility to normalize the clinical data, a
flexible data warehouse for storing raw clinical data or
longitudinal patient data, and a clinical analytics facility for
data mining and analytic model building, a user selectable
dashboard definer configured to provide user selectable options for
defining the clinical analytics to be presented in a report at a
dashboard, and a display definer configured to operate in
conjunction with the user selectable dashboard definer to define
the format in which the clinical analytics report from the clinical
informatics facility is to be presented at the dashboard. The data
may be gathered on a periodic basis. The data may be gathered on a
real-time basis. The real time basis may be at least as frequent as
every five minutes. The data normalization facility may de-identify
the data. The method may further include validating the clinical
data. The clinical analytics facility may enable patient risk
identification and patient tracking The selectable options may
include the addition of a comparative benchmark. The selectable
options may enable comparison to at least one of another patient,
healthcare provider, doctor, healthcare facility, hospital,
disease, condition, gender and age group. The selectable options
may include the addition of a patient risk identification and
patient tracking report relating to at least one of a patient,
medical care, an outcome, a demographic, a behavioral risk factor,
a disease risk factor, a procedure, a therapeutic, a utilization, a
readmission, mortality, and a complication. The format of the
report may include at least one of a table, a chart, text, and a
graph and the format may be customized based on at least one of a
data source, a time period, a chart type, a time interval for
display, a time interval for analysis, a filter, a hospital, a
physician, a patient, a patient characteristic, a cohort, a
disease, a gender, an age group, a treatment, a payer type and an
insurance provider. The dashboard may be presented via a
software-as-a-service architecture.
[0012] In an aspect of the invention, a method of ingesting and
analyzing healthcare data from a plurality of data sources in
real-time may include connecting to at least one data source,
retrieving data from the data source on a periodic basis to a
database, synchronizing data between the at least one data source
and the database, processing the data to identify data elements,
map data elements, and normalize data elements, wherein the data
elements are stored in a database, wherein the mapping may include
assigning data to a field of a database according to a
hierarchically organized lexicon of healthcare data elements,
wherein multiple data element entries in the lexicon are mapped to
a single field for at least one field, linking the data elements
over time to form a longitudinal data record, wherein the
longitudinal data records are stored in a longitudinal data
warehouse, and analyzing the at least one of the data elements and
data records to obtain at least one of actionable clinical
analytics, a patient risk identification, a disease-specific
analytic model, a predictive model, a benchmark and a quality
measure. The periodic basis on which data are retrieved may be
real-time. Real-time may be at least as frequent as every five
minutes. The at least one data source may include doctor's notes
from which data may be retrieved using natural processing language.
The at least one data source may include at least one of an
electronic medical record, an electronic health record, ambulatory
clinical data, claims data, paid claims data, adjudicated claims
data, inpatient clinical data, pharmacy data, doctor's notes,
self-reported data, census data, telemetry data, a networked
monitor, a home blood pressure device, a home health monitoring
device, a sensor device, mortality data, an internal management
system, a hospital inventory system, a clinical inventory system, a
clinical guideline, a specialty management system and an order set.
Processing the healthcare data also may include validating the
healthcare data elements. The at least one of actionable clinical
analytics, a patient risk identification, a disease-specific
analytic model, a predictive model, a benchmark and a quality
measure may be presented in a graphical user interface via a
software-as-a-service architecture.
[0013] In an aspect of the invention, a clinical informatics
platform may include a data extraction facility that gathers
clinical data from numerous sources on a periodic basis, a data
mapping facility that identifies and maps key data elements and
links data over time, wherein the mapping may include assigning
data to a field of a database according to a hierarchically
organized lexicon of healthcare data elements, wherein multiple
data element entries in the lexicon are mapped to a single field
for at least one field, a data normalization facility to normalize
the clinical data, a flexible data warehouse for storing at least
one of the raw clinical data and longitudinal patient data, a
clinical analytics facility for data mining, analytic model
building, patient risk identification, and patient tracking, and a
graphical user interface for presenting clinical analytics in an
actionable format. The periodic basis on which data are gathered
may be in real-time. Real-time may be at least as frequent as every
five minutes. The numerous sources may include doctor's notes from
which data may be retrieved using natural processing language. The
numerous sources may include at least one of an electronic medical
record, an electronic health record, ambulatory clinical data,
claims data, paid claims data, adjudicated claims data, inpatient
clinical data, pharmacy data, doctor's notes, self-reported data,
census data, telemetry data, a networked monitor, a home blood
pressure device, a home health monitoring device, a sensor device,
mortality data, an internal management system, a hospital inventory
system, a clinical inventory system, a clinical guideline, a
specialty management system and an order set. The data
normalization facility may de-identify the data. The method may
further include validating the clinical data. The graphical user
interface may be presented via a software-as-a-service
architecture.
[0014] In another aspect of the invention, a computer readable
medium having code which implements a method for describing,
evaluating, understanding, or managing a network of health care
providers, the method may include constructing a referral network
database of physicians and health care providers from at least one
of a private and a public data source, extracting data pertaining
to shared patients or referrals between the physicians and health
care providers from a database, and generating a graphical
representation of referral patterns in the referral network of
physicians and health care providers, wherein at least one element
of the graphical representation depicts a measure of an extent of a
type of activity within the referral network. The element of the
graphical representation may use at least one of size, thickness,
color and pattern to depict a type of activity. The element of the
graphical representation may depict how many patients are shared
among at least two health care providers. The medium may further
comprise analyzing the referral patterns in the graphical
representation to examine characteristics of the practice of the
network and to enable managing the network of health care
providers. The step of constructing a referral network of
physicians and health care providers may use data mining techniques
to find relationship data between physicians and health care
providers. The step of constructing a referral network of
physicians and health care providers may identify physicians and
health care providers as nodes with linkages in a referral network.
The data sources may include automated collection and
user-generated data sources for referral network construction. The
user-generated data may be from a survey. The data pertaining to
shared patients or referrals may be extracted from a claims or
electronic health record database. The graphical representation may
be an x-y coordinate system, an xyz coordinate, a pie chart, a
radar display, a GIS map, and other non-xy plots. Groups of
physicians and health care providers may be differentiated in the
graphical representation by at least one of a color, a shape, a
shading, and a size. The size of the object representing the
physicians or health care providers in the graphical representation
may correlate with a metric. The metric may be at least one of
cost, quality of care, compliance, or other measure of medical
care, cost, resource use, quality or patient outcome.
[0015] These and other systems, methods, objects, features, and
advantages of the present invention will be apparent to those
skilled in the art from the following detailed description of the
preferred embodiment and the drawings.
[0016] All documents mentioned herein are hereby incorporated in
their entirety by reference. References to items in the singular
should be understood to include items in the plural, and vice
versa, unless explicitly stated otherwise or clear from the text.
Grammatical conjunctions are intended to express any and all
disjunctive and conjunctive combinations of conjoined clauses,
sentences, words, and the like, unless otherwise stated or clear
from the context.
BRIEF DESCRIPTION OF THE FIGURES
[0017] The invention and the following detailed description of
certain embodiments thereof may be understood by reference to the
following figures:
[0018] FIG. 1a depicts a block diagram of the clinical analytics
platform.
[0019] FIG. 1b depicts a workflow of the clinical analytics
platform.
[0020] FIGS. 2a-2b depict a benchmarking and analytics tool of the
clinical informatics platform.
[0021] FIGS. 3a-3b depict a data processing and clinical
surveillance tool of the clinical informatics platform.
[0022] FIG. 4 depicts a heat map of a daily encounter volume.
[0023] FIG. 5 depicts a heat map of diabetes co-morbidity.
[0024] FIG. 6 depicts a heat map for diabetes prescribing
patterns.
[0025] FIG. 7 depicts a parallel coordinate plot of patients with a
decrease of >1% in Hemoglobin A1c.
[0026] FIG. 8 depicts a parallel coordinate plot of patients with
an increase of >1% in Hemoglobin A1c.
[0027] FIG. 9 depicts a parallel coordinate plot profiling change
in Hemoglobin A1c.
[0028] FIG. 10 depicts only those patients who had greater than or
equal to five endocrinology encounters on a parallel coordinate
plot.
[0029] FIG. 11 depicts those patients who had an endocrinology
encounter on a parallel coordinate plot.
[0030] FIG. 12 depicts a plot of physicians treating diabetes by
outcome and resource utilization.
[0031] FIG. 13 depicts a heat map of doctors with 10+ actively
managed diabetes patients.
[0032] FIG. 14 shows a visual representation of interactions in a
primary care physician network.
[0033] FIG. 15 shows a visual representation of interactions among
primary care physicians and endocrine specialists in a referral
network.
[0034] FIG. 16 shows another visual representation of interactions
among primary care physicians and endocrine specialists in a
referral network.
[0035] FIG. 17 shows a visual representation of primary care and
endocrine care providers in a referral network.
[0036] FIG. 18 depicts a logical flow for a computer-implemented
method of managing health care providers in a referral network.
[0037] FIG. 19 depicts the output of an AMI detection
algorithm.
[0038] FIG. 20 depicts a COAG Risk group tracking dashboard.
[0039] FIG. 21 depicts a network topology.
[0040] FIG. 22 depicts a block diagram of the data life cycle.
DETAILED DESCRIPTION
[0041] The clinical analytics platform automates the capture,
extraction, and reporting of data required for certain quality
measures, provides real-time clinical surveillance, clinical
dashboards, tracking lists, and alerts for specific, high-priority
conditions, provides improved methods and systems for analyzing and
managing health care referral networks, and offers dynamic, ad-hoc
quality reporting capabilities. Throughout this specification,
real-time indicates that an action is taken in an interval of time
such that the data that are available to the platform 100 are data
that are current as of an interval of time not far from the current
time. The interval can vary from a few hours or minutes, such as
five minutes, all the way to instantaneous.
[0042] The clinical informatics platform may empower health care,
pharmaceutical and biotechnology firms, medical device
manufacturers, government agencies, and financial services firms
with insight into how to manage provider networks and provider
network shared patients or referrals, how patient populations are
treated, which treatments and procedures are prescribed, and
importantly, the quality, efficacy, and cost of this care. The
clinical informatics platform may assemble, standardize, and
analyze clinical, operational, social network, referral, insurance
and financial data across varied treatment settings and time
periods to generate a longitudinal, comprehensive view of patient
care. The clinical informatics platform may address the specific
needs of inpatient and outpatient health care providers,
pharmaceutical and biotechnology firms, medical device
manufacturers, government agencies, and financial services firms by
combining deep, retrospective capabilities with powerful real-time
predictive tools that connect knowledge with action.
[0043] The clinical informatics platform may enable organizations
to transform an immense reservoir of data into valuable, actionable
knowledge using a comprehensive suite of software-as-a-service
(SaaS) solutions that unlock the clinical information needed to
improve patient care while improving financial performance. The
SaaS-based clinical informatics platform applies sophisticated
techniques to mine, standardize, validate and/or aggregate health
care data from disparate IT systems, all within a state-of-the-art,
HIPAA-compliant, and highly secure environment. The clinical
informatics platform analyzes clinical, operational, social
network, referral, insurance and financial data and delivers
powerful analytic insights and comparative benchmarks with
cost-effective, retrospective, and real-time SaaS-based tools. The
SaaS-based tools enable delivering real time comparative analytics
without the end user having to purchase or maintain any additional
hardware or human resources and includes rapid, scalable, data
extraction, mapping, and ontological normalization systems. The
clinical informatics platform may combine both retrospective,
deep-dive analytic systems, such as the benchmarking and analytics
tool 202 or the data processing and clinical surveillance tool 302,
with real-time data processing capabilities. In embodiments, the
platform may be modular and contain all available tools or only
certain tools. The clinical informatics platform may include
disease-specific analytic tools, predictive models, and modules.
The clinical informatics platform may include or enable the
generation of detailed, customizable clinical, operational, social
network, referral, insurance and financial benchmarks. The clinical
informatics platform may support collaborative development and
testing of performance and operational improvement strategies
within and among organizations.
[0044] The historic barriers to leveraging health care, referral,
insurance and social networking data are many: data reside in many
different systems, data are trapped in local terminologies and free
text, robust clinical models require costly tools and large
samples, and real-time clinical analytics are costly and
difficult-to-use, to name a few. The needs are clear: extract all
the data, normalize all the data, provide robust clinical and
networking analytics, and deliver powerful and timely insights. The
clinical analytics platform meets these needs: it has flexible,
platform-agnostic data extraction capabilities, provides scalable
data normalization and next-generation natural language processing
(NLP), a singular, relational longitudinal patient data warehouse,
real-time predictive analytics, modeling, patient tracking tools,
social networking and referral analysis tools, and clinical
checklists. With the tools available in the clinical informatics
platform, users may: gain valuable insight into the clinical and
operational performance of an organization; conduct real-time and
retrospective analytics and benchmark clinical performance; and
employ disease-specific clinical analytics and evidence-based data
to intervene in a timely manner to identify patients at risk,
reduce morbidity, mortality, and complications in real-time, ensure
that opportunities for improvement are identified before the
patient has left the care setting, and, and manage provider
networks and provider network shared patients or referrals, and
otherwise effect positive change.
[0045] An understanding of all aspects of clinical and operational
performance, such as the quality, safety, and cost of healthcare
provider care, may be enabled by the clinical informatics platform.
The health care provider may be enabled to act in real-time to
ensure delivery of the best and most efficient care. Health care
providers may be enabled to compare, analyze, and identify best
practices, and then collaborate with peers around the development
and dissemination of best practices and to optimize
performance-based reimbursement. The clinical informatics platform
may connect cost and outcomes to maximize cost-effective care,
identify and track the care of high-cost and high-risk patients in
real-time to reduce preventable complications and/or never events,
and improve performance on Joint Commission (JCAHO) and
pay-for-performance measures, and prove the value of the care
delivered to payers with customizable, comparative performance
benchmarks. Health care providers may be empowered to qualify for
certain government and private programs, such as the American
Recovery and Reinvestment Act (ARRA) funding.
[0046] Referring to FIG. 1a, a block diagram of the clinical
analytics platform 100 is shown. A data extraction facility 104 can
extract data from a plurality of disparate, healthcare and claims
data sources 118 to enable the real-time collection, processing and
centralized storage of health records in a database. Data ingestion
techniques may be applied to a heterogynous system of EMRs from
various vendors and systems to obtain data, normalize them and
store the newly processed data in a homogenous database where a
single set of applications can be used to interface with and
analyze the data. Real-time, continuous data ingestion may come
from various data sources 118 which may include ambulatory clinical
data, pharmacy data, doctor's notes, EHRs, EMRs, inpatient clinical
data, biographical data, hospital billing data, claims data, census
data, self-reported data, networked devices and monitors (e.g.
blood pressure device, glucose meter, etc.), mortality data,
telemetry, inventory systems, clinical guidelines, management
systems, order sets, and the like. For example, NLP techniques can
be used to gather data from doctor's notes or other transcriptions,
both numeric and text data. Data may be extracted, optionally
encrypted, and ingested by the system in a running load process.
Data may be obtained through an RSS feed or a transmission or
extraction of data in a format such as XML, HL7, SCRIPT, X12, CSV,
HL7v2, HL7v3, Dicom, X12N, NCPDP, and the like. Extract, Transform,
and Load (ETL) tools may be used to connect remote databases to the
platform 100 and pull data out of the remote databases so that the
data goes from database to database. This method speeds things up
and requires less hardware since a copy of the data does not have
to be written for transmission or extraction.
[0047] The platform 100 enables ingestion and semantic
normalization of the healthcare data by converting the data in the
healthcare records to standardized data elements using a data
normalization facility 110 and mapping the converted data with
standard terminologies, such as Federal Health Architecture (FHA)
terminologies, billing codes, IDC codes, CPT codes and the like
using a mapping application of the data processor 108. The data
processor 108 may transform data from the various formats in which
it exists. Data may be mapped iteratively against divergent source
systems. Mapping data may take advantage of standard and custom
terminologies and combinations thereof. For example, the
terminologies may enable identifying data elements by the various
ways they may be described in different data sources and mapping
all of the disparate elements to a single terminology used by the
platform 100. In another embodiment, mapping may be ontological,
that is, the terminologies may have a hierarchy. For example, 5
different variables may be found in a single or a plurality of data
sources. In choosing which target variable of the platform to map
the 5 variables to, a terminology may be consulted. Multiple
possibilities may exist in the terminology, but a hierarchy of the
terminologies may facilitate choosing which target variable of the
platform to use. A rules database 112 may be used for storing
terminologies, codes, hierarchies, rules for data
de-identification, and the like. The rules database 112 may be
updated periodically as new terminology becomes available or
updated. The rules database 112 may provide rules to the data
processor 108 for mapping. The rules database 112 may also store
rules, attributes, characteristics, and criteria that are used in
each analytic model.
[0048] Data may be linked over time to create longitudinal patient
records. Data may also be linked along the lines of cohorts,
practice groups, geographic areas, and the like. The data may be
subject to validation. Validation may include identifying and
omitting outlier values from the data, removing unreliable data and
the like.
[0049] The data may be stored in a flexible data warehouse, such as
a raw data store 118, data mart 120 or a longitudinal patient data
warehouse 114.
[0050] The data may be analyzed by the data processor 108. Since
the data may be real-time or near real-time, the analysis can
enable providing care instructions, flagging medication and/or care
errors, flagging events for follow-up or treatment, making
recommendations, supporting disease management, cost containment,
generate epidemiological/bioterrorism alerts, and the like.
Real-time data ingestion, processing, and analysis enables
automating processes and generating and updating care plans in near
real time. The data may be certified. Interfaces to the platform
100, such as a user interface 122, report facility 124 audit
facility 128, and other interfaces 130, may be used to search and
view data, initiate analyses, visualize data, generate reports,
generate a tracking page, and the like.
[0051] In a workflow of the clinical informatics platform as shown
in FIG. 1b, data, such as ambulatory clinical data, financial data,
inpatient clinical data, pharmacy data, and the like, may be
gathered and/or extracted from source systems. The extracted data
may undergo manipulations, such as mapping and normalization prior
to storage in a database. The data may then be analyzed, tracked,
manipulated and the like by any number of clinical analytics tools.
The analytics may be modular, such as by disease, condition,
cohort, patient area, geographic area, therapeutic protocol,
practice group, hospital, and the like. The analytics may generate
granular comparative data. The analytics may enable predictive
modeling and understanding of the cost and efficiency of care. The
analytics may enable quality measures, such as PQRI and HEDIS
registry reporting. For example, a clinical analytics tool may
enable analytic data grouping. FIG. 22 depicts a block diagram of
the data life cycle including the following steps: pre-extraction
where inventory is taken of systems and the best extraction
approach is identified, data extraction, processing, mapping,
ingestion, normalization, validation, analytics, and data
certification, such as medical validation, QA, analytics and
general validation.
[0052] In an embodiment, the platform 100 may comprise tools for
analysis and data presentation and reporting. Certain tools may
enable near real-time quality/risk identification and workflow.
Tools may enable disease-specific analytic models. Tools may enable
data mining, such as to identify patients at risk. In any of these
tools, the analytics may be presented as an actionable
visualization that may highlight variance. The presentation may
include patient, physician, group views, and the like. The data
presentation may be a collaboration platform. The data presentation
may include real-time alerts, such as alerts relating to at least
one risk associated with at least one patient based at least in
part on the gathered healthcare data. Alerts may be presented in at
least one of an audible or visual manner. In an embodiment, data
presentation may be flash-based or involve some other dynamic media
and/or animation.
[0053] Referring to FIG. 2, the clinical informatics platform may
comprise a tool for enabling robust clinical, operational, and
financial benchmarking and comparative analytics across the
continuum of health care. The benchmarking and analytics tool 202
may be a dashboard for presenting comparative analytics. Data from
disparate data sources 118 are extracted as described herein,
normalized and mapped as described herein, then analyzed to obtain
a benchmark and a data sample to compare to the benchmark. In some
embodiments, the benchmark is known and does not have to obtained
through analysis.
[0054] The analytics may be presented in a number of report
formats, such as tables and graphs. The graphs may be of any
format, such as bar graph, pie chart, scatter plot, line graph, and
the like. The graphs may be customized using a number of built-in
features of the tools, such as by changing the data source, the
time period for analysis, the chart type, time intervals for
display, a custom or built-in filter, a comparison to another
subject (such as hospital, physician, disease, gender, age group,
and the like), and the like. A graphical user interface to the
platform 100 may be used to present the comparative data and the
benchmark as a report.
[0055] For example, the chart 202 in FIG. 2a, shown expanded in
FIG. 2b, shows medical care analytics in the field of utilization.
FIG. 2b depicts a chart that compares the mean length of stay (LOS)
of patients on regular human insulin with pressure ulcer stages III
and IV, with the mean hospital LOS on the y-axis and the time
period on the x-axis. In this graph, the mean LOS for one hospital
is compared to a regional aggregate hospital benchmark. A filter
may be applied to this chart, such as by using a filter wizard. For
example, the data may be filtered by insurance provider so that LOS
is displayed for particular insurance providers, such as MEDICARE,
MEDICAID, MEDICARE and MEDICAID, private insurance,
government-sponsored insurance, and the like. To enable a different
comparison, a comparison wizard may be employed. For example, a
Charlson co-morbidity index comparison may be requested for the
data. Instead of having a single data point, the data may be
presented for each time interval by Charlson index or by range of
Charlson indices. In another example, the data may be compared to
another hospital including to or instead of the benchmark. In yet
another example, the data may be presented at a more granular
level, such as by attending physician, hospital floor, hospital
unit, hospital bed, procedures, and the like. By providing
different ways to present and manipulate data, patterns and
outliers may be more readily identified. The visual dashboard
provides a number of benefits. It enables real-time clinical
intervention and reduces cost, morbidity, and mortality. It
gathers, maps, and normalizes data in near real-time to predict and
track which patients are likely to be high-risk and/or high cost
and to alert for compliance to Joint Commission Core Measure
metrics.
[0056] Referring to FIG. 3a, the clinical informatics platform may
also comprise a dashboard for a near real-time data processing and
predictive clinical surveillance system that identifies high-risk,
high-cost patients, tracks necessary care, and supports clinicians
to intervene to improve care. The data processing and clinical
surveillance tool 302 may be a dashboard for presenting data
processing and clinical surveillance and enabling real-time
predictive risk tracking Data from disparate data sources 118 are
extracted as described herein, normalized and mapped as described
herein, then analyzed to obtain clinical tracking data that can be
presented by a tool 302 in a graphical user interface of the
platform 100. An embodiment of a report is shown in an expanded
version in FIG. 3b. A reports tab may enable a user to generate
reports related to a number of topics, such as the patient, medical
care, and outcomes, and sub-topics, such as demographics,
behavioral risk factors, disease risk factors, procedures,
therapeutics, utilization, ICU, readmission, mortality,
complications, and the like. Since the data are real-time,
real-time clinical intervention is enabled as patients who are
high-risk and/or high-cost are more readily identifiable and easy
to track. The screen in FIG. 3b displays a `Tracking` tab of the
tool 302 that shows active tracked patients in a hospital unit who
are in the acute myocardial infarction (AMI) risk group.
Demographic information is available for each patient as well as
therapeutics over a given time period, risk level, physician,
location, arrival date/time, and the like. The tool 302 provides a
way to manually include or exclude patients from tracking A reason
may need to be recorded for exclusion or inclusion. The reason may
be selected from a list of standard reasons or manually entered or
entered in some other way. Filters may be applied to the data
presentation. In an `Issues` tab, issues, such as `all open issues`
may be shown for patients being tracked. The `Issues` tab may show
Patient name, Issue, Physician, Location, Due By timer, and the
like. Filters may be applied to this display. For example, patients
in only certain locations may be included in the listing, or in
other embodiments, patients in all locations may be included. In
another example, patients in only certain risk groups may be
included in the listing, or in other embodiments, patients in all
risk groups may be included. When a patient is selected, their
profile may be displayed. For example, patient information may be
shown, their AMI risk profile may be shown, or some other risk
profile may be shown. In an example, the AMI risk profile may
include statuses over time, such as location, blood pressure,
temperature, anticoagulants, beta blockers, thrombolytics, CK-MB,
triglyceride levels, CBC, glucose level, troponin levels, images
taken, procedures done, and the like. A patient may be manually
excluded or included in a risk group, but a reason may need to be
recorded. The reason may be selected from a list of standard
reasons or the reason may be manually entered. Given a patient's
risk profile which is known in real-time, predictive analytics may
be applied to identify patients at risk. Patients at risk may be
automatically detected by the tool 302, such as by using the AMI
detection algorithm tool shown in FIG. 19. The visual dashboard of
the tool 302 provides a number of benefits. The platform 100
enables "at a glance" status checks on the floor, automates and
optimizes identification of patients to be tracked, reduces the
number of patients to be tracked, connects knowledge to action,
provides clinical data for better understanding, compresses "time
to intervention" for better outcomes, compresses "time to action"
for core measure compliance, reduces cost, morbidity, and
mortality, and the like. FIG. 20 depicts another example of a
clinical surveillance dashboard for a risk group of patients on
anticoagulation medications.
[0057] In an embodiment, the clinical surveillance dashboard may
enable a health care provider or health practitioner to see each
patient's countdown to events that need to be done within a certain
period of time, such as within an hour of admission, day of
admission, and the like. The health practitioner's plan for care
can be viewed by doctor, patient, floor, clinic, disease, and the
like, along with all of the relevant data and measures that went
into establishing the plan. The plan for care itself may be
automatically customized based on an indication, therapeutic
protocol, and the like. The plan and/or its timeline for action may
be updated in real-time, such as when new data become available to
the platform. The plan for care may be for a particular patient and
may be adjusted based on real-time data regarding that patient. For
example, if the real-time data indicates that the patient is
recovering more slowly than expected, the plan may be revised to
include higher doses of painkillers and more frequent testing and
monitoring.
[0058] The clinical informatics platform 100 may also comprise a
tool for a near real-time data processing and predictive clinical
surveillance system that identifies diabetic patients. Data from
various sources, such as laboratory data and pharmacy data, may be
analyzed using an algorithm to determine if a patient may be
diabetic, based on some known combination of laboratory and
pharmacy data that indicates a high likelihood of the
pathology.
[0059] The clinical informatics platform 100 may also comprise a
tool for a near real-time data processing and predictive clinical
surveillance system that identifies cohorts of patients that fit
the JCAHO guidelines.
[0060] The clinical informatics platform 100 may also comprise
tools for analytic model building. For example, to build a disease
model, aspects of the disease that might be of interest in
determining the quality, cost, and/or outcome of care may be
obtained from the literature, textbooks or know how. These aspects
may be defined as inputs to the model in terms of rules,
attributes, characteristics, criteria, or the like. These inputs
may be defined in a rules database 112 and updated periodically or
as needed. Data may be analyzed according to the model by the
platform 100 to enable determining a disease state. For example, a
diabetes model may be consulted to determine or predict if a
patient has diabetes. The model may require that certain data be
available, such as diagnoses codes, glucose test results, HgbA1C
levels, outpatient prescriptions, and the like. These data may be
analyzed according to rules of the model. For example, the model
may indicate that a patient is diabetic if a glucose level is over
a prescribed amount and if an HgbA1C level is over a prescribed
amount. If the actual glucose level is below the prescribed amount
and the HgbA1C level is above the prescribed amount, when these
data are input to the model, it may be determined that there is a
moderate likelihood that the data corresponds to a patient with
diabetes. Other disease specific models may be enabled by the
platform 100, such as models for congestive heart failure,
hypertension, COPD, dyslipidemia, coronary artery disease,
peripheral vascular disease, acute myocardial infarction,
cerebrovascular disease, stroke, renal failure, osteoarthritis,
rheumatoid arthritis, ulcer, depression, heart failure, pneumonia,
septicemia, adult preventative screening, CAD, adult asthma,
pediatric asthma, chronic kidney disease, anti-coagulation/VTE,
fibromyalgia, back pain, obesity, osteoporosis, estrogen-related
disorders, inflammatory bowel syndrome, dementia, BPH, pain
management, immune disorders, HIV, colon cancer, prostate cancer,
breast cancer, pneumonia, TB, anemia, lupus, gout, thyroid
disorders, hepatitis, atrial fibrillation, arrhythmias, and the
like.
[0061] The clinical analytics platform 100 may support application
programming interfaces (APIs) integrated with the data warehouse
114 to allow the development of applications that can leverage the
normalized data, such as for applications directed to regional
healthcare issues, individual health providers, research or
clinical studies, pharmaceutical and biotechnology companies, and
the like.
[0062] The clinical analytics needs of ambulatory providers may be
significantly different from those of acute care providers. Care in
ambulatory environments may occur over a longer time period with
multiple discrete events. These events may occur in different
locations, under the care of multiple providers, and prescribed
treatments may only show results over an extended period of time.
In addition, documentation, coding practices, disease specificity
and multiplicity all combine to make clinical analytics far
different from similar efforts in hospital settings. The clinical
informatics platform is uniquely designed to handle the specific
needs of ambulatory care providers. The clinical informatics
platform integrates clinical, claims, lab, and other data to
generate a complete and longitudinal view of member organizations'
patient populations, provides disease-specific clinical
classification and advanced analytics to define appropriate patient
cohorts, treatment pathways, outcomes and associated costs,
provides treatment effectiveness and outcomes analysis to support
evidence-based process improvements, thus allowing physician
practices to enhance cost-effectiveness while providing the highest
quality patient care, facilitates comparative analytics and
benchmarking by aggregating data from participating medical groups,
facilities, practices, and physicians into a single, standard
clinical ontology, supports organizations to develop, compare, and
share best practices and performance improvement strategies through
our unique collaborative programs, and the like. The clinical
informatics platform may enable users to access comprehensive
patient data and the data necessary for quality reporting,
effectively manage chronic disease patients with protocol tracking
tools and task lists, automate real-time surveillance and employ
clinical decision support at the point of care with real-time
reminders and alerts, and the like. The clinical informatics
platform may help organizations demonstrate the value of the care
they deliver, receive appropriate compensation for the quality of
the care they deliver, get clinical detail on best practices that
lead to improved outcomes, improve patient care in a timely manner
by combining real-time clinical surveillance with robust
retrospective clinical analytics, attract patients by enhancing
their organization's reputation for quality, and the like.
[0063] The clinical informatics platform may also enable the
activities of life science firms. Using the clinical informatics
platform, life sciences firms may be able to quantify patient
populations, market share, and market opportunities, all by disease
severity and co-morbidities. The clinical informatics platform may
profile patient segments with detailed clinical specificity (e.g.
lab results and radiology reports), identify treatment decisions by
physician specialty and practice setting, and elucidate the
associated costs and outcomes--information vital to generating
appropriate and tailored marketing strategies and tactics.
[0064] The clinical informatics platform may provide life sciences
companies with the tools necessary to understand the clinical
drivers of treatment decision-making, to quantify their brand's
unique benefits, and to accurately assess a myriad of market
opportunities. The clinical informatics platform may offer the
timeliest and most complete clinical data needed to manage and
succeed in today's challenging marketplace. The clinical
informatics platform's longitudinal clinical data offer
unprecedented insight into brand choices and the associated
clinical outcomes and cost effectiveness. With the clinical
informatics platform, life sciences firms can more accurately and
expediently quantify patient populations, market share, and market
opportunities. The clinical informatics platform enables profiling
patient segments with detailed clinical specificity (e.g., lab
results, radiology, co-morbidities), identifying the clinical
drivers of treatment decisions by physician specialty, and
elucidating the costs and outcomes associated with treatments. The
clinical informatics platform may provides longitudinal clinical
data needed to gain a more accurate picture of specific patient
sub-populations, brand-specific clinical profiles, including lab
and radiology results, for improved marketing message development
and effectiveness tracking, clinical evidence for the development
of refined segmentation strategies, clinical data highlighting
which treatments occurred when, and by which specialty, to
accurately define the sequence of care and associated outcomes, and
the like. The clinical informatics platform clinical data enables
life sciences companies to: measure clinical outcomes within
specific patient segments, more accurately identify unmet needs and
associated market opportunities, tailor marketing messages to
accurately address the specific needs of provider groups and
patient cohorts, and the like. The clinical informatics platform
enables users to quickly and accurately answer numerous questions,
such as the following: Am I attracting the right patient segments
based on my product's unique clinical profile?; How does the
clinical profile of patients on my brand compare to those of my
competitors, both branded and generic?; In which patient segments
does my brand outperform the competition in terms of clinical
outcomes and cost effectiveness?; How do I maximize the pricing and
reimbursement for my brand?; and the like.
[0065] The clinical analytics platform 100 may be deployed in many
different environments to provide for data extraction, processing,
storage, analysis, and presentation. For example, the platform 100
may be deployed in ambulatory care facilities, life science firms,
acute care facilities, hospice, clinical trial facilities,
insurance companies, senior living facilities, veterinary
facilities, epidemiological centers, triage centers, emergency
rooms, and the like.
[0066] The clinical informatics platform may further be used for
social network analysis of health care providers, provider network
shared patients, referrals and the like. By way of one example, in
instances where such analysis is desired, the clinical informatics
platform may extract various data from numerous health care
providers that relate to patient movement and treatment within the
network. The clinical informatics platform can then normalize the
data and apply analytics to the data that will display patient
movement, course of treatment and progress, physician referral, or
other results based on the data provided and information desired.
The network in question and its associated data may be visualized
and integrated into the clinical informatics platform user
interface through the use of a network analysis visualization
tool.
[0067] Further, the clinical informatics platform may enable social
network analysis of health care provider, such as primary care
physicians and specialists, interactions which may enable managing
provider networks and provider network shared patients or
referrals. The social network analysis results may be visualized on
a coordinate system, such as an x-y coordinate system, an xyz
coordinate, a pie chart, a radar display, a GIS map, other non-xy
plots, and the like. For example, the Y component of the coordinate
system may be the physician and the X component may be key care
variables around the way that care is delivered in a particular
disease. Another coordinate may identify the physicians by their
clinic. The visualization may be examined for physicians who
cluster together by using an algorithm. The clusters may be
indicative of patterns of care that are characteristic of the
cluster, and may be suggestive of a pattern of care, cost, or
outcome that is either positive or negative.
[0068] Social network analysis for managing provider networks and
provider network shared patients or referrals may be described with
reference to an example involving an internist or specialist
network. A first step in social network analysis in this example
may involve identifying all of the encounters between a physician,
a patient and a medical center to create a bi-partite network. For
example, connections between physicians, patient and physician, and
another physician may be represented in the network. The bi-partite
network may then be condensed into a doctor-to-doctor network. The
doctor-to-doctor network may be made bi-partite again in that all
of the same doctor type to same doctor type connections may be
eliminated. The now condensed network may be an internist to
specialist network.
[0069] A social network analysis visualization tool may enable
visualizing a network in many ways, such as by using a coordinate
system. For example, groups of providers may be differentiated by a
specific color, a specific shape, or a size. The thickness of the
connections, represented as lines, between members of the group may
be a measure of how many patients they share between them. This
measurement may be utilized as a weight in the analysis. From this
visualization, significant patterns may become apparent that may
enable examination of characteristics of the practice of medicine.
For example, the social network analysis visualization may be used
to show patients who are on MEDICARE versus commercial insurance
versus government insurance. The analysis visualization may
identify patient encounters for patients that are above average.
The analysis visualization may show a practice cluster that stands
out in their utilization of imaging, patient outcomes, cost, and
the like by being able to correlate the clusters and connections
with various metrics. For example, the analysis may enable a
mapping of the quality and cost of a network of doctors. In some
embodiments, the size of the objects representing a group may be a
visual indicator of some kind of measurement, such as how much was
spent per visit on a patient. In this embodiment, the social
network visualization may show that some groups practice the same
medicine and get different outcomes at the same cost.
[0070] The social network analysis visualization tool may enable
determining who are popular providers and influencers of other
providers and care outcomes, not by examining communication or
information flow, but rather by analyzing actual care
characteristics.
[0071] Referring to FIG. 14, a visual representation of
interactions in a primary care physician network is shown. Each
diamond represents a primary care physician in the network, or a
vertex, and each connecting line between each primary care
physician, or the edge between two vertices, represents the number
of shared patients by the thickness of the line. The visualization
shows distinct clusters of physicians who have many shared patients
among them with a smattering of smaller clusters and physicians who
do not share patients with any other physicians or have very few
shared patients. In this example, primary care physicians may be
identified as part of an institution or practice by a unique
shading or coloring of the diamond.
[0072] Referring to FIGS. 15 and 16, an example of a social network
analysis of interactions among primary care physicians and
specialists in a referral network is shown. In this example, a
visual representation of a social network analysis of interactions
among primary care physicians and endocrine specialists for
diabetes mellitus type 2 in a referral network is shown. FIG. 16 is
a close-up view of the referral network in FIG. 15 where the
isolated physicians and endocrinology specialists have been removed
for clarity. Each diamond represents a primary care physician in
the network, or a first vertex, each hexagon represents an
endocrinology specialist, or a second vertex, and each connecting
line between each primary care physician and endocrinology
specialist, or the edge between two vertices, represents the number
of shared patients by the thickness of the line. In this example,
primary care physicians may be identified as part of an institution
or practice by a unique shading or coloring of the diamond. The
visualization shows a distinct cluster of physicians and an
endocrinology specialist, where a single endocrinology specialist
receives many shared patients or referrals. The visualization also
shows physician or endocrinology specialist clusters where multiple
specialists are referred to by the physicians. Finally, there are
also physicians displayed who make no referrals, as well as
endocrinology specialists who receive few or no shared patients or
referrals. The visualization highlights where certain line
thicknesses could be increased, that is, where more shared patients
or referrals can be made between physician and endocrinology
specialist. The visualization also enables correlating metrics,
such as expense, care outcomes, compliance, and the like, with a
thickness of connection. By identifying endocrinology specialists
by referral, the visualization may suggest which specialists to go
to and which ones to avoid. In embodiments, the social network
analysis may be very sensitive for a particular institution. For
example, the visualization shows that certain endocrinology
specialists are members of distinct clusters, where the distinct
clusters are representative of referral networks arising from
distinct institutions. In this example, the three endocrinology
specialists within the rectangle on FIG. 16 are all part of the
same clinic but the top-most specialist seems to be a major part of
two clusters in the referral network, while the two other
endocrinology specialists get few shared patients or referrals from
physicians in the other cluster. Thus, the visualization may
identify certain practitioners who are key to the interactions in
the referral network or certain practitioners who are draining
referral patients from a certain clinic. The visualization may be
examined over time to determine the ebb and flow of the referral
network. As clusters and connections become apparent, they may be
actionable, and modifications may be made to practices.
[0073] Referring to FIG. 17, a visual representation of primary
care and endocrine care providers in a referral network in diabetes
mellitus type 1 is shown. This visualization is different from that
shown in FIGS. 15 and 16 in that the disease is different, diabetes
mellitus type 1 is an autoimmune disease while diabetes mellitus
type 2 is a disorder that is characterized by high blood glucose in
the context of insulin resistance and relative insulin deficiency.
By examining the differences in the visualizations, a comparison
can be made with respect to how care is delivered in a particular
disease. For example, in this visualization, there are no isolated
primary care physicians, suggesting that primary care physicians
will usually refer their diabetes mellitus type 1 patients to a
specialist for care.
[0074] Referring to FIG. 18, a method for describing, evaluating,
understanding, or managing a network of health care providers may
include constructing a referral network database of physicians and
health care providers from at least one of a private and a public
data source 1802, extracting data pertaining to shared patients or
referrals between the physicians and health care providers from a
database 1804, and generating a graphical representation of
referral patterns in the referral network of physicians and health
care providers 1808, wherein at least one element of the graphical
representation depicts a measure of an extent of a type of activity
within the referral network. The element of the graphical
representation may use at least one of size, thickness, color and
pattern to depict a type of activity. The element of the graphical
representation may depict how many patients are shared among at
least two health care providers. The medium may further comprise
analyzing the referral patterns in the graphical representation to
examine characteristics of the practice of the network and to
enable describing, evaluating, understanding, or managing the
network of health care providers 1810. The step of constructing a
referral network of physicians and health care providers may use
data mining techniques to find relationship data between physicians
and health care providers. The step of constructing a referral
network of physicians and health care providers may identify
physicians and health care providers as nodes with linkages in a
referral network. The data sources may include automated collection
and user-generated data sources for referral network construction.
The user-generated data may be from a survey. The data pertaining
to shared patients or referrals may be extracted from a claims or
electronic health record database. The graphical representation may
be an x-y coordinate system. Groups of physicians and health care
providers may be differentiated in the graphical representation by
at least one of a color, a shape, a shading, a size and the like.
The size of the object representing the physicians or health care
providers in the graphical representation may correlate with a
metric. The metric may be at least one of cost, quality of care,
compliance, or other measure of medical care, cost, resource use,
quality, patient outcome and the like. In another embodiment,
analytical and visual tools may be used to examine the process of
care. One such tool may be based on heat maps, which may be a
graphical representation of data where the values taken by a
variable in a two-dimensional map are represented as colors.
Generating a heat map may include taking a sequence of numeric
values and representing them with color.
[0075] Heat maps may enable identifying similarities in clinics or
other groups of doctors in how they manage disease and generally
provide care. Heat maps may enable visualizing the organization of
healthcare providers into groups based on a similarity in providing
care. In this way, healthcare providers may be identified as
outliers or who may fit into similar groups. The heat map enables
understanding the nature of a group of healthcare providers and
enables exploring the characteristics of that group.
[0076] Heat maps may be used to look to examine various elements of
care, such as co-morbidity, prescription use, and the like. For
example, referring to FIG. 4, a heat map for Daily Encounter Volume
may include taking a sequence of numeric values associated with
daily encounters indexed by date, and representing them as a
calendar with the days filled with colors representing the values.
In another example, referring to FIG. 5, co-morbidities with
diabetes are represented graphically as a heat map. One axis of the
map relates to diabetic status and the other axis of the map
relates to other diseases or conditions, such as ophthalmic
disorders, PVD, Charlson co-morbidity score, stroke, CVD, renal
disease, lipids, HTN, and the like. The colors of the plot may
indicate the presence of a co-morbidity, and in some embodiments,
quantify the co-morbidity. In another example, referring to FIG. 6,
a heat map for diabetes prescribing patterns is shown. One axis of
the map relates to diabetic status and the other axis of the map
relates to prescribing patterns, such as percentage of patients
treated with alpha-glucosidase inhibitors, percent treated with
insulin, percent treated with insulin secretagogues, percent
treated with insulin sensitizers, average treatment values thereof,
combinations thereof, and the like. The colors of the plot visually
indicate the quantifiable differences in prescribing for different
diabetic statuses.
[0077] Another analytical and visual tool that may be used to
examine the process of care may be parallel coordinate plots.
Parallel coordinate plots are a unique way to look at patterns over
time, such as by week, month, year, and the like. Parallel
coordinate plots may be used to examine the process of care for
individuals in a patient-by-patient way. Each line of the plot may
represent an individual patient. For example, referring to FIG. 7,
patients, segmented into patients who are pre-diabetic, type I,
type II, and type unknown, with a decrease of >1% in Hemoglobin
A1c are shown in a parallel coordinate plot where each line
represents an individual patient. Levels of hemoglobin A1c are
typically used as indicators of diabetes disease management. HbA1c
levels depend on the blood glucose concentration. That is, the
higher the glucose concentration in blood, the higher the level of
HbA1c. Levels of HbA1c are not influenced by daily fluctuations in
the blood glucose concentration but reflect the average glucose
levels over the prior six to eight weeks. Therefore, HbA1c is a
useful indicator of how well the blood glucose level has been
controlled in the recent past and may be used to monitor the
effects of diet, exercise, and drug therapy on blood glucose in
diabetic patients. For example, hemoglobin A1c (HbA1c) levels may
be measured at the diagnosis and then measured again after a period
of treatment time to determine a change in hemoglobin A1c with
treatment. Along with these data points, other elements of care can
also be examined, such as number of endocrine visits, number of
therapies used, co-morbidity of ages, and the like. In the parallel
coordinate plot, a pattern may emerge of people who do well and
people who don't do well with respect to their care map. Referring
to FIG. 8, patients with an increase of >1% in Hemoglobin A1c
are shown in a parallel coordinate plot where each line represents
an individual patient. Referring to FIG. 9, a parallel coordinate
plot profiling change in HbA1c is shown. Referring to FIG. 10, only
those patients who had greater than or equal to five endocrinology
encounters are shown on the plot. Referring to FIG. 11, those
patients who had an endocrinology encounter are shown on the
plot.
[0078] Referring to FIG. 12, a corrgram plot of physicians treating
diabetes by outcome and resource utilization is shown. Each column
and row represents a particular characteristic of clinical practice
in the care of diabetes such as percent patients with renal
failure, on insulin, with high LDL, and having type 1 diabetes, so
that each characteristic is represented once vertically, and once
horizontally. The ordering of the placement of each characteristic
on the chart is determined by a statistical determination of the
correlation of each characteristic with the others in an analysis
of many different physician practices. Highly correlated
characteristics will be clustered together horizontally and
vertically using a defined algorithm. The intensity of the
correlation is indicated by the color of the boxes on the lower
left half of the corrgram at the intersection of each horizontal
and vertical characteristic, with dark blue being most highly
positively correlated, and dark red being most negatively
correlated. The white lines within the boxes are a visual aid to
demonstrate the direction of correlation, and help identify outlier
boxes. The matrix in the upper right adds additional information
about the correlation, showing an ellipse encompassing 68% of the
most concentrated data points from each practice, and a line
indicating a loess smoothed curve of the data points from each
practice. The corrgram can help to identify which care parameters
are correlated with each other in the practices of physicians
treating patients with diabetes.
[0079] For example, this corrgram demonstrates that the percentage
of patients in a practice who have a hospital admission is highly
correlated with the average amount of inpatient charges for this
practice. Also, the percentage of a practice over the age of 65 is
negatively correlated with the percentage who have high LDL
laboratory values.
[0080] Referring to FIG. 13, a heat map of doctors with 10+
actively managed diabetes patients is shown. Each column represents
a particular characteristic of a physician clinical practice in the
care of diabetes such as percent patients with renal failure, on
insulin, with high LDL, having type 1 diabetes, with highest values
in dark blue, medium values neutral, and lowest values in dark red
for each measurement. An individual physician is represented by a
row. The heat map is constructed using a self-organizing clustering
algorithm which clusters physicians with similar characteristics
clustered together vertically, and similar care characteristics
clustered together horizontally. The tree structure on the far left
and top of the heat map indicate the degree of similarity. The
clinic to which each physician belongs is color coded as either
blue or gray on the vertical column on the left of the heat map.
Review of the heat map allows an administrator, physician, or nurse
to identify similarities in the care of diabetes patients, and to
identify key differences between individuals on particular care
parameters. The heat map can identify similarities within and
between clinics as well, and can identify physicians who fall
outside the care characteristics of their particular clinic.
Finally, by examining the clustering of care characteristics, the
viewer can identify groups of care parameters that tend to be used
in similar frequency by all physicians. For example, the yellow/red
cluster in the lower right are all indicators of charges, with
these physicians submitting charges on the lower end of their
peers. An administrator could study other parameters of care to see
where they differ from their peers to determine the reasons why
their practices generate lower charges, on average, than other
practices.
[0081] In an embodiment, the clinical analytics platform 100 may be
embodied in the network topology depicted in FIG. 21. This network
topology includes a distributed, layered architecture comprising
intrusion protection and intrusion detection systems. Such a
topology provides the security and manageability needed to deploy
the platform 100 as a SaaS solution.
[0082] The methods and systems described herein may be deployed in
part or in whole through a machine that executes computer software,
program codes, and/or instructions on a processor. The processor
may be part of a server, client, network infrastructure, mobile
computing platform, stationary computing platform, or other
computing platform. A processor may be any kind of computational or
processing device capable of executing program instructions, codes,
binary instructions and the like. The processor may be or include a
signal processor, digital processor, embedded processor,
microprocessor or any variant such as a co-processor (math
co-processor, graphic co-processor, communication co-processor and
the like) and the like that may directly or indirectly facilitate
execution of program code or program instructions stored thereon.
In addition, the processor may enable execution of multiple
programs, threads, and codes. The threads may be executed
simultaneously to enhance the performance of the processor and to
facilitate simultaneous operations of the application. By way of
implementation, methods, program codes, program instructions and
the like described herein may be implemented in one or more thread.
The thread may spawn other threads that may have assigned
priorities associated with them; the processor may execute these
threads based on priority or any other order based on instructions
provided in the program code. The processor may include memory that
stores methods, codes, instructions and programs as described
herein and elsewhere. The processor may access a storage medium
through an interface that may store methods, codes, and
instructions as described herein and elsewhere. The storage medium
associated with the processor for storing methods, programs, codes,
program instructions or other type of instructions capable of being
executed by the computing or processing device may include but may
not be limited to one or more of a CD-ROM, DVD, memory, hard disk,
flash drive, RAM, ROM, cache and the like.
[0083] A processor may include one or more cores that may enhance
speed and performance of a multiprocessor. In embodiments, the
process may be a dual core processor, quad core processors, other
chip-level multiprocessor and the like that combine two or more
independent cores (called a die).
[0084] The methods and systems described herein may be deployed in
part or in whole through a machine that executes computer software
on a server, client, firewall, gateway, hub, router, or other such
computer and/or networking hardware. The software program may be
associated with a server that may include a file server, print
server, domain server, internet server, intranet server and other
variants such as secondary server, host server, distributed server
and the like. The server may include one or more of memories,
processors, computer readable media, storage media, ports (physical
and virtual), communication devices, and interfaces capable of
accessing other servers, clients, machines, and devices through a
wired or a wireless medium, and the like. The methods, programs or
codes as described herein and elsewhere may be executed by the
server. In addition, other devices required for execution of
methods as described in this application may be considered as a
part of the infrastructure associated with the server.
[0085] The server may provide an interface to other devices
including, without limitation, clients, other servers, printers,
database servers, print servers, file servers, communication
servers, distributed servers and the like. Additionally, this
coupling and/or connection may facilitate remote execution of
program across the network. The networking of some or all of these
devices may facilitate parallel processing of a program or method
at one or more location without deviating from the scope of the
invention. In addition, any of the devices attached to the server
through an interface may include at least one storage medium
capable of storing methods, programs, code and/or instructions. A
central repository may provide program instructions to be executed
on different devices. In this implementation, the remote repository
may act as a storage medium for program code, instructions, and
programs.
[0086] The software program may be associated with a client that
may include a file client, print client, domain client, internet
client, intranet client and other variants such as secondary
client, host client, distributed client and the like. The client
may include one or more of memories, processors, computer readable
media, storage media, ports (physical and virtual), communication
devices, and interfaces capable of accessing other clients,
servers, machines, and devices through a wired or a wireless
medium, and the like. The methods, programs or codes as described
herein and elsewhere may be executed by the client. In addition,
other devices required for execution of methods as described in
this application may be considered as a part of the infrastructure
associated with the client.
[0087] The client may provide an interface to other devices
including, without limitation, servers, other clients, printers,
database servers, print servers, file servers, communication
servers, distributed servers and the like. Additionally, this
coupling and/or connection may facilitate remote execution of
program across the network. The networking of some or all of these
devices may facilitate parallel processing of a program or method
at one or more location without deviating from the scope of the
invention. In addition, any of the devices attached to the client
through an interface may include at least one storage medium
capable of storing methods, programs, applications, code and/or
instructions. A central repository may provide program instructions
to be executed on different devices. In this implementation, the
remote repository may act as a storage medium for program code,
instructions, and programs.
[0088] The methods and systems described herein may be deployed in
part or in whole through network infrastructures. The network
infrastructure may include elements such as computing devices,
servers, routers, hubs, firewalls, clients, personal computers,
communication devices, routing devices and other active and passive
devices, modules and/or components as known in the art. The
computing and/or non-computing device(s) associated with the
network infrastructure may include, apart from other components, a
storage medium such as flash memory, buffer, stack, RAM, ROM and
the like. The processes, methods, program codes, instructions
described herein and elsewhere may be executed by one or more of
the network infrastructural elements.
[0089] The methods, program codes, and instructions described
herein and elsewhere may be implemented on a cellular network
having multiple cells. The cellular network may either be frequency
division multiple access (FDMA) network or code division multiple
access (CDMA) network. The cellular network may include mobile
devices, cell sites, base stations, repeaters, antennas, towers,
and the like. The cell network may be a GSM, GPRS, 3G, EVDO, mesh,
or other networks types.
[0090] The methods, programs codes, and instructions described
herein and elsewhere may be implemented on or through mobile
devices. The mobile devices may include navigation devices, cell
phones, mobile phones, mobile personal digital assistants, laptops,
palmtops, netbooks, pagers, electronic books readers, music players
and the like. These devices may include, apart from other
components, a storage medium such as a flash memory, buffer, RAM,
ROM and one or more computing devices. The computing devices
associated with mobile devices may be enabled to execute program
codes, methods, and instructions stored thereon. Alternatively, the
mobile devices may be configured to execute instructions in
collaboration with other devices. The mobile devices may
communicate with base stations interfaced with servers and
configured to execute program codes. The mobile devices may
communicate on a peer to peer network, mesh network, or other
communications network. The program code may be stored on the
storage medium associated with the server and executed by a
computing device embedded within the server. The base station may
include a computing device and a storage medium. The storage device
may store program codes and instructions executed by the computing
devices associated with the base station.
[0091] The computer software, program codes, and/or instructions
may be stored and/or accessed on machine readable media that may
include: computer components, devices, and recording media that
retain digital data used for computing for some interval of time;
semiconductor storage known as random access memory (RAM); mass
storage typically for more permanent storage, such as optical
discs, forms of magnetic storage like hard disks, tapes, drums,
cards and other types; processor registers, cache memory, volatile
memory, non-volatile memory; optical storage such as CD, DVD;
removable media such as flash memory (e.g. USB sticks or keys),
floppy disks, magnetic tape, paper tape, punch cards, standalone
RAM disks, Zip drives, removable mass storage, off-line, and the
like; other computer memory such as dynamic memory, static memory,
read/write storage, mutable storage, read only, random access,
sequential access, location addressable, file addressable, content
addressable, network attached storage, storage area network, bar
codes, magnetic ink, and the like.
[0092] The methods and systems described herein may transform
physical and/or or intangible items from one state to another. The
methods and systems described herein may also transform data
representing physical and/or intangible items from one state to
another.
[0093] The elements described and depicted herein, including in
flow charts and block diagrams throughout the figures, imply
logical boundaries between the elements. However, according to
software or hardware engineering practices, the depicted elements
and the functions thereof may be implemented on machines through
computer executable media having a processor capable of executing
program instructions stored thereon as a monolithic software
structure, as standalone software modules, or as modules that
employ external routines, code, services, and so forth, or any
combination of these, and all such implementations may be within
the scope of the present disclosure. Examples of such machines may
include, but may not be limited to, personal digital assistants,
laptops, personal computers, mobile phones, other handheld
computing devices, medical equipment, wired or wireless
communication devices, transducers, chips, calculators, satellites,
tablet PCs, electronic books, gadgets, electronic devices, devices
having artificial intelligence, computing devices, networking
equipments, servers, routers and the like. Furthermore, the
elements depicted in the flow chart and block diagrams or any other
logical component may be implemented on a machine capable of
executing program instructions. Thus, while the foregoing drawings
and descriptions set forth functional aspects of the disclosed
systems, no particular arrangement of software for implementing
these functional aspects should be inferred from these descriptions
unless explicitly stated or otherwise clear from the context.
Similarly, it will be appreciated that the various steps identified
and described above may be varied, and that the order of steps may
be adapted to particular applications of the techniques disclosed
herein. All such variations and modifications are intended to fall
within the scope of this disclosure. As such, the depiction and/or
description of an order for various steps should not be understood
to require a particular order of execution for those steps, unless
required by a particular application, or explicitly stated or
otherwise clear from the context.
[0094] The methods and/or processes described above, and steps
thereof, may be realized in hardware, software or any combination
of hardware and software suitable for a particular application. The
hardware may include a general purpose computer and/or dedicated
computing device or specific computing device or particular aspect
or component of a specific computing device. The processes may be
realized in one or more microprocessors, microcontrollers, embedded
microcontrollers, programmable digital signal processors or other
programmable device, along with internal and/or external memory.
The processes may also, or instead, be embodied in an application
specific integrated circuit, a programmable gate array,
programmable array logic, or any other device or combination of
devices that may be configured to process electronic signals. It
will further be appreciated that one or more of the processes may
be realized as a computer executable code capable of being executed
on a machine readable medium.
[0095] The computer executable code may be created using a
structured programming language such as C, an object oriented
programming language such as C++, or any other high-level or
low-level programming language (including assembly languages,
hardware description languages, and database programming languages
and technologies) that may be stored, compiled or interpreted to
run on one of the above devices, as well as heterogeneous
combinations of processors, processor architectures, or
combinations of different hardware and software, or any other
machine capable of executing program instructions.
[0096] Thus, in one aspect, each method described above and
combinations thereof may be embodied in computer executable code
that, when executing on one or more computing devices, performs the
steps thereof. In another aspect, the methods may be embodied in
systems that perform the steps thereof, and may be distributed
across devices in a number of ways, or all of the functionality may
be integrated into a dedicated, standalone device or other
hardware. In another aspect, the means for performing the steps
associated with the processes described above may include any of
the hardware and/or software described above. All such permutations
and combinations are intended to fall within the scope of the
present disclosure.
[0097] While the invention has been disclosed in connection with
the preferred embodiments shown and described in detail, various
modifications and improvements thereon will become readily apparent
to those skilled in the art. Accordingly, the spirit and scope of
the present invention is not to be limited by the foregoing
examples, but is to be understood in the broadest sense allowable
by law.
[0098] All documents referenced herein are hereby incorporated by
reference.
* * * * *