U.S. patent application number 13/925262 was filed with the patent office on 2013-12-26 for systems and methods for predictive analytics for site initiation and patient enrollment.
This patent application is currently assigned to Quintiles Transnational Corp.. The applicant listed for this patent is Wade Kenneth Brant, Joseph William Charles Goodgame, Gavin Thomas Nichols. Invention is credited to Wade Kenneth Brant, Joseph William Charles Goodgame, Gavin Thomas Nichols.
Application Number | 20130346094 13/925262 |
Document ID | / |
Family ID | 49774051 |
Filed Date | 2013-12-26 |
United States Patent
Application |
20130346094 |
Kind Code |
A1 |
Goodgame; Joseph William Charles ;
et al. |
December 26, 2013 |
Systems and Methods For Predictive Analytics for Site Initiation
and Patient Enrollment
Abstract
Methods and systems for predictive analytics for site initiation
and patient enrollment are disclosed. One method may include:
receiving a user's selection of one or more parameters associated
with a clinical trial; accessing a database of data associated with
a plurality of previous clinical trials; comparing the one or more
parameters to the previous clinical trials; determining one or more
factors associated with the clinical trial based on the comparison;
and displaying the parameters and the factors on a display.
Inventors: |
Goodgame; Joseph William
Charles; (Zionsville, IN) ; Nichols; Gavin
Thomas; (Raleigh, NC) ; Brant; Wade Kenneth;
(Greenwood, IN) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Goodgame; Joseph William Charles
Nichols; Gavin Thomas
Brant; Wade Kenneth |
Zionsville
Raleigh
Greenwood |
IN
NC
IN |
US
US
US |
|
|
Assignee: |
Quintiles Transnational
Corp.
Durham
NC
|
Family ID: |
49774051 |
Appl. No.: |
13/925262 |
Filed: |
June 24, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61663292 |
Jun 22, 2012 |
|
|
|
61663057 |
Jun 22, 2012 |
|
|
|
61663299 |
Jun 22, 2012 |
|
|
|
61663398 |
Jun 22, 2012 |
|
|
|
61663219 |
Jun 22, 2012 |
|
|
|
61663357 |
Jun 22, 2012 |
|
|
|
61663216 |
Jun 22, 2012 |
|
|
|
Current U.S.
Class: |
705/2 |
Current CPC
Class: |
G06F 16/211 20190101;
G16H 40/63 20180101; G06F 40/166 20200101; G16H 10/20 20180101;
G06F 16/248 20190101; G06F 16/345 20190101; G06F 17/10 20130101;
G16H 10/60 20180101; G06T 11/206 20130101; G06F 3/0484
20130101 |
Class at
Publication: |
705/2 |
International
Class: |
G06F 19/00 20060101
G06F019/00 |
Claims
1. A method for predictive analytics, the method comprising:
receiving a user's selection of one or more parameters associated
with a clinical trial; accessing a database of data associated with
a plurality of previous clinical trials; comparing the one or more
parameters to the previous clinical trials; determining one or more
factors associated with the clinical trial based on the comparison;
and displaying the parameters and the factors on a display.
2. The method of claim 1, wherein receiving the user's selection of
one or more parameters comprises determining that the user has
adjusted one or more sliders associated with the parameters.
3. The method of claim 1, wherein the factors comprises one or more
of: an estimated cost of the clinical trial, an estimated required
number of patients, and a predicted length of time of the clinical
trial.
4. The method of claim 1, wherein the previous clinical trials are
associated with the clinical trial.
5. The method of claim 1, wherein at least one of the one or more
parameters comprises a country in which to perform the clinical
trial.
6. The method of claim 5, wherein the one or more parameters
further comprises one or more of: the country's population, a
prevalence of patients in the country, the country's regulatory
environment, a number of clinical trials performed in the country,
a length of time of the clinical trial, a length of time to start
the clinical trial, historical data associated with the country, or
a level of risk associated with conducting the clinical trial in
the country.
7. The method of claim 1, further comprising: receiving the user's
selection of one or more new parameters associated with the
clinical trial; comparing the one or more new parameters to the
previous clinical trials; determining one or more additional
factors associated with the clinical trial based on the comparison;
and updating the display to show the one or more additional
factors.
8. The method of claim 7, wherein the new parameters comprise one
or more of: a number of clinical trials in a therapeutic area, a
length of time to start a clinical trial, investigator specific
parameters, or data associated with previous clinical trials.
9. The method of claim 8, wherein the investigator specific
parameters comprise one or more of: a number of trials that an
investigator has run or a time it typically takes the investigator
to enroll patients.
10. The method of claim 8, wherein the data associated with
previous clinical trials comprises one or more of: a number of
patients, a number of failures, a dropout rate, or a screen failure
rate.
11. A non-transitory computer readable medium comprising program
code, which when executed is configured to cause a processor to:
receive a user's selection of one or more parameters associated
with a clinical trial; access a database of data associated with a
plurality of previous clinical trials; compare the one or more
parameters to the previous clinical trials; determine one or more
factors associated with the clinical trial based on the comparison;
and display the parameters and the factors on a display.
12. The non-transitory computer readable medium of claim 11,
wherein receiving the user's selection of one or more parameters
comprises determining that the user has adjusted one or more
sliders associated with the parameters.
13. The non-transitory computer readable medium of claim 11,
wherein the previous clinical trials are associated with the
clinical trial.
14. The non-transitory computer readable medium of claim 11,
wherein the factors comprise one or more of: an estimated cost of
the clinical trial, an estimated required number of patients, and a
predicted length of time of the clinical trial.
15. The non-transitory computer readable medium of claim 11,
wherein at least one of the one or more parameters comprises a
country in which to perform the clinical trial.
16. The non-transitory computer readable medium of claim 15,
wherein the one or more parameters further comprises one or more
of: the country's population, a prevalence of patients in the
country, the country's regulatory environment, a number of clinical
trials performed in the country, a length of time of the clinical
trial, a length of time to start the clinical trial, historical
data associated with the country, or a level of risk associated
with conducting the clinical trial in the country.
17. The non-transitory computer readable medium of claim 11,
further comprising program code, which when executed by the
processor is configured to cause the processor to: receive the
user's selection of one or more new parameters associated with the
clinical trial; comparing the one or more new parameters to the
previous clinical trials; determining one or more additional
factors associated with the clinical trial based on the comparison;
and updating the display to show the one or more additional
factors.
18. The non-transitory computer readable medium of claim 17,
wherein the new parameters comprise one or more of: a number of
clinical trials in a therapeutic area, a length of time to start a
clinical trial, investigator specific parameters, or data
associated with previous clinical trials.
19. The non-transitory computer readable medium of claim 18,
wherein the investigator specific parameters comprise one or more
of: a number of trials that an investigator has run or a time it
typically takes the investigator to enroll patients.
20. The non-transitory computer readable medium of claim 18,
wherein the data associated with previous clinical trials comprises
one or more of: a number of patients, a number of failures, a
dropout rate, or a screen failure rate.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional
Application No. 61/663,292, filed on Jun. 22, 2012, entitled
"Method and System to Manipulate Multiple Selections Against a
Population of Elements;" U.S. Provisional Application No.
61/663,057, filed on Jun. 22, 2012, entitled "Systems and Methods
For Predictive Analytics For Site Initiation and Patient
Enrollment;" U.S. Provisional Application No. 61/663,299, filed on
Jun. 22, 2012, entitled "Methods and Systems for Predictive
Clinical Planning and Design and Integrated Execution Services;"
U.S. Provisional Application No. 61/663,398, filed on Jun. 22,
2012, entitled "Systems and Methods for Subject Identification (ID)
Modeling;" U.S. Provisional Application No. 61/663,219, filed Jun.
22, 2012, entitled "Systems and Methods for Analytics on Viable
Patient Populations;" U.S. Provisional Application No. 61/663,357,
filed Jun. 22, 2012; entitled "Methods and Systems for a Clinical
Trial Development Platform;" U.S. Provisional Application No.
61/663,216, filed Jun. 22, 2012; entitled "Systems and Methods for
Data Visualization." The entirety of all of which is hereby
incorporated by reference herein.
FIELD OF THE INVENTION
[0002] The present invention relates generally to systems and
methods for the creation and analysis of clinical trials. The
present invention relates more specifically to systems and methods
for predictive analytics for site initiation and patient
enrollment.
BACKGROUND
[0003] Clinical trials for molecules that may become pharmaceutical
products often last for years. The core cost of the trial is
affected primarily by the length of the trial. And a delay of even
a single day can cost hundreds or thousands and even millions of
dollars.
[0004] One of the variables most likely to affect the length of the
trial has historically been patient enrollment, i.e., how quickly
investigators are able to bring patients into a trial. This is due
in part to the fact that a typical trial has a very limited amount
of start time at the beginning and analysis time at the end.
Further, the trial will require a fixed duration of treatment time
in order to provide sufficient data for a submission. Because of
these factors, the trials cannot be arbitrarily shortened. Further,
a trial requires a certain minimum number of patients in order to
prove efficacy and safety of a molecule. Thus, patient enrollment
becomes the primary variable factor.
[0005] Since the amount of time that it takes to complete patient
enrollment is critical to the length of a trial, this aspect is
often the focus of companies' efforts to shorten the trials.
Various companies have attempted to address the need to shorten the
trial. These companies include drug companies, software companies,
and consulting groups.
[0006] The conventional methods that these companies have
implemented for shortening the length of a trial have focused on
statistical models. A company might, for example, look at a hundred
prior trials and attempt to extrapolate what would happen in a new
trial based on the data culled from the earlier trial. Such models
provide a relatively static view of the potential trial. Further,
they often do not provide details regarding the impact of variables
that go into creating the trial, such as the impact of conducting
the trial in particular countries and at particular investigation
sites. Thus the conventional systems lack flexibility and provide
data that may or may not accurately reflect the planned trial.
SUMMARY
[0007] Embodiments of the present disclosure provide systems and
methods for Predictive Analytics for Site Initiation and Patient
Enrollment. In one embodiment, a three-tier system provides a
client application to view and set parameters related to planning a
clinical trial, a patient modeling engine for determining likely
outcomes based on the parameters supplied by the user, and a
database for storing historical data related to
previously-completed and ongoing trials. The application in such an
embodiment receives the user's parameter selections and attempts to
determine in which countries and at which investigation sites the
clinical trial should be conducted to meet the user's goals. The
application is able to provide the user with the impact of the
user's selections on the amount of time the trial is likely to take
and the likely cost of the trial. The user may also be provided
with best and worst case scenarios so that the user can balance
risks and costs associated with a planned clinical trial.
[0008] In one embodiment, the user is also provided with a clinical
plan perspective--a view that encompasses many different molecules
and clinical trials for those molecules. Such a view allows the
user to determine if all the various planned clinical trials can be
performed simultaneously. And if not, i.e., if saturation exits,
the application allows the user to vary parameters so that the user
can prioritize which trials should be done first.
[0009] This embodiment is mentioned not to limit or define the
invention, but to provide an example of an embodiment of the
invention to aid understanding thereof. Embodiments are discussed
in the Detailed Description, and further description of the
invention is provided there. Advantages offered by the various
embodiments of the present invention may be further understood by
examining this specification.
BRIEF DESCRIPTION OF THE FIGURES
[0010] These and other features, aspects, and advantages of the
present disclosure are better understood when the following
Detailed Description is read with reference to the accompanying
drawings, wherein:
[0011] FIG. 1 is a block diagram illustrating an exemplary
environment for implementation of one embodiment of the present
disclosure;
[0012] FIG. 2 is a screen shot of a country editor view, showing
color-coded countries and a popup categories table;
[0013] FIG. 3 is a screen shot of an investigator site editor view
in one such embodiment;
[0014] FIG. 4 is a flow chart illustrating the general steps
involved in one such process;
[0015] FIG. 5 is a flowchart illustrating one method for performing
tuning on an enrollment model according to one embodiment of the
present disclosure;
[0016] FIG. 6 is a flowchart illustrating method of adjusting
enrollment for geographic distribution in one embodiment of the
present disclosure;
[0017] FIG. 7 is a screen shot of an enrollment country view in one
embodiment of the present disclosure;
[0018] FIG. 8 is a screenshot of a graph, illustrating the risk
associated with various scenarios in one embodiment of the present
disclosure;
[0019] FIG. 9 is a screen shot of enrollment views in embodiments
of the present disclosure;
[0020] FIG. 10 is a screen shot of enrollment views in embodiments
of the present disclosure; and
[0021] FIG. 11 is a screen shot of enrollment views in embodiments
of the present disclosure.
DETAILED DESCRIPTION
[0022] Embodiments of the present disclosure provide systems and
methods for predictive analytics for site initiation and patient
enrollment.
Illustrative Embodiment of the Present Disclosure
[0023] One illustrative embodiment of the present disclosure
comprises an application for selecting sites and analyzing patient
enrollment for clinical trials. The embodiment allows a user to
access an application that presents a variety of clinical
trial-related parameters for various countries and investigators.
These parameters may include, for example, the population of a
country, the regulatory environment, and the level of risk
associated with conducting a trial in a particular country.
[0024] Once the user has set the parameters that are of most
interest for the countries, the user is presented with a graphical
representation of the countries that reflects the appropriateness
of a particular country in view of the parameters specified by the
user. For example, the countries may be color-coded. The user is
then able to drill down into particular countries to identify
particular investigators.
[0025] In the illustrative embodiment, for each investigator, the
user is again able to specify investigator-specific parameters,
such as the number of trials that an investigator has run, the time
it typically takes an investigator to enroll patients and other
relevant parameters. Once the user selects these parameters, the
user may then be presented with a graphical representation of the
investigators or investigator sites, illustrating the relative
value of using particular site for a particular set of
parameters.
[0026] The process is iterative; the user is able to change the
parameters for countries and investigators to determine the most
appropriate sites to utilize for a clinical trial. The results of
the user's selections can then be used as part of a larger clinical
trial analysis application.
[0027] This illustrative embodiment neither limits nor defines the
disclosure. Rather, the illustrative embodiment is meant to provide
an example of how the present disclosure may be implemented.
Illustrative Environment
[0028] Referring now to the drawings, in which like numerals
indicate like elements throughout the several figures, FIG. 1 is a
block diagram illustrating an exemplary environment for
implementation of one embodiment of the present disclosure. The
embodiment shown in FIG. 1 includes a client 100 that allows a user
to interface with an application server 200, web server 300, and/or
database 400 via a network 500.
[0029] The client 100 may be, for example, a personal computer
(PC), such as a laptop or desktop computer, which includes a
processor and a computer-readable media. The client 100 also
includes user input devices, such as a keyboard and mouse or touch
screen, and one or more output devices, such as a display. In some
embodiments of the disclosure, the user of client 100 accesses an
application or applications specific to one embodiment of the
disclosure. In other embodiments, the user accesses a standard
application, such as a web browser on client 100, to access
applications running on a server such as application server 200,
web server 300, or database 400. For example, in one embodiment,
the memory of client 100 stores applications including a design
studio application for planning and designing clinical trials. The
client 100 may also be referred to as a terminal in some
embodiments of the present disclosure.
[0030] Such applications may be resident in any suitable
computer-readable medium and executable on any suitable processor.
Such processors may comprise, for example, a microprocessor, an
ASIC, a state machine, or other processor, and can be any of a
number of computer processors, such as processors from Intel
Corporation, Advanced Micro Devices Incorporated, and Motorola
Corporation. The computer-readable media stores instructions that,
when executed by the processor, cause the processor to perform the
steps described herein.
[0031] The client 100 provides a software layer, which is the
interface through which the user interacts with the system by
receiving and displaying data to and from the user. In one
embodiment, the software layer is implemented in the programming
language C# (also referred to as C Sharp). In other embodiments,
the software layer can be implemented in other languages such as
Java or C++. The software layer may be graphical in nature, using
visual representations of data to communicate said data to one or
more users. The visual representations of data may also be used to
receive additional data from one or more users. In one embodiment,
the visual representation appears as a spider-like layout of nodes
and connectors extending from each node to a central node.
[0032] Embodiments of computer-readable media comprise, but are not
limited to, an electronic, optical, magnetic, or other storage
device, transmission device, or other device that comprises some
type of storage and that is capable of providing a processor with
computer-readable instructions. Other examples of suitable media
comprise, but are not limited to, a floppy disk, CD-ROM, DVD,
magnetic disk, memory chip, ROM, RAM, PROM, EPROM, EEPROM, an ASIC,
a configured processor, all optical media, all magnetic tape or
other magnetic media, or any other medium from which a computer
processor can read instructions. Also, various other forms of
computer-readable media may be embedded in devices that may
transmit or carry instructions to a computer, including a router,
private or public network, or other transmission device or channel,
both wired and wireless. The instructions may comprise code from
any suitable computer-programming language, including, for example,
C, C++, C#, Visual Basic, Java, Python, Perl, and JavaScript.
[0033] The application server 200 also comprises a processor and a
memory. The application server may execute business logic or other
shared processes. The application server may be, for example, a
Microsoft Windows Server operating in a .NET framework, an IBM
Weblogic server, or a Java Enterprise Edition (J2E) server. While
the application server 200 is shown as a single server, the
application server 200, and the other servers 300, 400 shown may be
combined or may include multiple servers operating together to
perform various processes. In such embodiments, techniques such as
clustering or high availability clustering may be used. Benefits to
architectures such as these include redundancy and performance,
among others.
[0034] In the embodiment shown in FIG. 1, the application server
200 is in communication with a web server 300 via a network
connection 250. The web server 300 also comprises a processor and a
memory. In the memory are stored applications including web server
software. Examples of web server software include Microsoft
Internet Information Services (IIS), Apache Web Server, and Sun
Java System Web Server from Oracle, among others.
[0035] In the embodiment shown in FIG. 1, the web server 300 is in
communication with a database 400 via a network connection 350 and
a network connection 450. The web server 300 provides a web service
layer that, together or separate from application server 200, acts
as middleware between a database 400 and the software layer,
represented by the client 100. The web server 300 communicates with
the database 400 to send and retrieve data to and from the database
400.
[0036] The network 500 may be any of a number of public or private
networks, including, for example, the Internet, a local area
network ("LAN"), or a wide area network ("WAN"). The network
connections 150, 250, 350, and 450 may be wired or wireless
networks and may use any known protocol or standard, including
TCP/IP, UDP, multicast, 802.11b, 802.11g, 802.11n, or any other
known protocol or standard. Further, the network 100 may represent
a single network or different networks. As would be clear to one of
skill in the art, the client 100, servers 200, 300, and database
400 may be in communication with each other over the network or
directly with one another.
[0037] The database 400 may be one or a plurality of databases that
store electronically encoded information comprising the data
required to plan, design, and execute a clinical trial. In one
embodiment, the data comprises one or more design elements
corresponding to the various elements related to one or more
clinical trials. The database 400 may be implemented as any known
database, including a SQL database or an object database. Further,
the database software may be any known database software, such as
Microsoft SQL Server, Oracle Database, MySQL, Sybase, or
others.
Country Editor
[0038] One embodiment of the present disclosure comprises a country
editor for selecting the countries for the application to consider
in determining the sites in which to conduct a clinical trial. The
country editor allows the user to identify, based on historical
information, which countries have investigator sites likely to
enroll the necessary patients in a timely manner. The country
editor in one such embodiment presents choices to the user in the
form of slider input tools ("sliders"). The user then adjusts the
sliders to set limits for data categories associated with each
slider. For example, limits may comprise minimum, ideal levels, and
maximum levels. In one embodiment, the data categories include:
[0039] Patient Prevalence (per 100 k population); [0040]
Extrapolated Prevalence (# of patients); [0041] Total Population;
[0042] Trial Saturation (# of active trials); [0043] Regulatory
Approval (cycle time in days); [0044] Cycle Time (CT) Materials
(cycle time in days); [0045] Historical Recruitment (patients per
site per month (PSM)); [0046] Clinical Research Associates (RA's)
(total #); and [0047] Site Start-Up (cycle time in days).
[0048] The application executing on the application server 200
executes an algorithm to compare the limits of a category to a
category value obtained from data stored in the database 400 for
each country. The application then generates a value for each
category and country combination and generates a view of that data
for presentation on the client 100. The embodiments described
herein may operate in a similar way within this architecture or may
be implemented in other ways. For example, in some embodiments, the
client 100 may perform much of the processing in addition to
providing the display and receiving the user's input.
[0049] FIG. 2 is a screen shot 200 of a country editor view,
showing color-coded countries and a popup categories table 202. The
countries and categories are assigned colors based on the slider
values and the proprietary data. In the embodiment shown, upon
selection of a country, a popup table 202 presents the categories
and category colors for the selected country. The category and
country values and/or colors may be saved as a country plan for use
by the enrollment editor or by other elements of a larger clinical
trial system.
[0050] In the embodiment shown in FIG. 2, the sliders 204 divide
the data range of each category into three sub-ranges. Data falling
within the first sub-range is assigned the number 0, data in the
middle sub-range is assigned the number 1, and data in the last
sub-range is assigned the number 2. Categories are colored
according to their assigned values, where 0, 1 and 2 correspond,
respectively, to the colors red, yellow and green. In the
embodiment shown, a user may change the category colors by changing
the limits on the sliders.
[0051] In one embodiment, the values for all the categories for a
particular country are averaged to determine a value for the
country. In such an embodiment, the country may, for example, be
colored red if the value falls below 0.8, yellow if the value falls
between 0.8-1.2, and green if the value exceeds 1.2. A country can
also be colored to indicate that the country cannot or should not
be selected. Indicators such as shape, cross-hatching, symbols or
other methods may be used instead of or addition to the colors
shown.
Investigator Site Editor
[0052] One embodiment of the present disclosure comprises a site
editor for selecting the investigation sites for the application to
consider in determining the sites in which to conduct a clinical
trial. Preferably, the investigator site editor is functionally
similar to the country editor and relies on the results of
selections made in the country editor and/or other parts of the
system. FIG. 3 is a screen shot of an investigator site editor view
300 in one such embodiment. The investigator site editor view 300
shown comprises color-coded circles 302 representing the likelihood
and timing of enrolling patients within geographical boundaries,
such as regions, based on proprietary investigator data. Once
displayed, the user can select other geographical groupings for
displaying grouped data such as countries, states or cities. As
shown in FIG. 3, circles 302 can be colored red, yellow, green, and
partially one color and partially another. In the embodiment shown,
the data categories include: [0053] Trials in Therapeutic Area (#
of trials); [0054] Site Start-Up (SSU) (cycle time in days); [0055]
Patient Randomization (average #/trial); [0056] Patients Screened
(average #/trial); [0057] Screen Failure Rate (% of screen
patients); [0058] Drop Out Rate (% of randomized patients); [0059]
Queries (#/100 pages); and [0060] Performance Index (average across
trials).
[0061] The embodiment shown uses the three colors, red, green, and
yellow as indicia of the acceptability or applicability of
investigators. However, while the various countries and
investigators are described in terms of the color coding, other
indicia may be used to identify the various described
categories.
[0062] In the embodiment shown in FIG. 3, the yellow and
green-coded Investigators make up the available population. The
following data rules are followed for determining the measures set
out above: [0063] If the Investigator SSU is null or 0, then the
Country SSU is used; [0064] If the Investigator SSU StDev is null,
then use the Country SSU StDev; [0065] If the Investigator PSM is
null or 0, then the Country PSM is used; [0066] If the Investigator
PSM StDev is null, then use the Country PSM StDev;
[0067] Once the population of potential payments is established,
then if it is less than needed, the user will expand the selection.
If the population of potential patients is greater than is needed,
the user may narrow the population by making various selections,
including, for example, selecting an option called "Sites like
These." Running the model yields a potentially narrower population
of "Modeled" selections.
Illustrative Enrollment Process
[0068] In embodiments of the present disclosure, the user is
attempting to determine how most successfully to enroll patients in
a clinical trial given the relative importance of the various
parameters of the trial. One embodiment deals with high level
metrics related to the country, investigators and patients. These
high level metrics drive the time it will take to enroll patients
and the amount of money it will cost for the modeled patients.
[0069] In such an embodiment, the application tracks the number of
countries that are available for the trial, the number of countries
in the set created based on the values of the parameters supplied
by the user, and the number of countries selected (i.e., predicted)
by the model. The application tracks these same three metrics
(available, selected, and modeled) for the investigators.
[0070] Based on these metrics and on historical data, the
application is able to determine the potential number of patients.
Then, based on the number of desired patients to be enrolled in the
study, the application can calculate the predicted number of
patients to enroll. Based on these predictions, the model in such
an embodiment can determine the modeled (i.e., predicted) time to
enroll the desired number of patients as well as the cost of the
modeled patients. A user can then use this information for planning
purposes. Embodiments of the present disclosure utilize an
iterative process or enrollment modeling.
[0071] FIG. 4 is a flow chart 400 illustrating the general steps
involved in one such process. In the process shown, the user begins
enrollment modeling. As a first step, the investigator uses various
views, such as those illustrated in FIGS. 2 and 3 to model patients
402. The user then adjusts the model based on geographic
distribution 404, time 406, and the total number of investigators
used 408. For example, the user may adjust for high-performing to
low-performing sites based on cost even though that could increase
the risk. In some embodiments, the user is able to use a feature
called a "quick rule" that automatically selects sites with, for
example, the lowest startup time or with a startup time that is
lower than a number of days selected by the user. FIG. 4 is merely
illustrative. Various other types of adjustments may be made in
embodiments of the present disclosure.
[0072] FIG. 5 is a flowchart illustrating one method for performing
tuning on an enrollment model according to one embodiment of the
present disclosure. In the embodiment shown, the application
determines the number of modeled patients. The user then determines
if the number of modeled patients is greater than or equal to the
number of desired patients 502. If so, the required number of
patients has been identified, and the process ends. If not, the
process continues in an iterative fashion until a sufficient number
of patients have been modeled.
[0073] Next, the application determines whether the number of
potential patients is greater than or equal to the number of
desired patients 504. If so, then the time limit may be increased
to increase the number of modeled patients 506. This may be done
manually or automatically. In the embodiment shown, the time limit
is increased to 36 months.
[0074] If the number of potential patients is less than the desired
number of patients, the application highlights the unelected
investigators, if any. If unselected investigators exist 508, the
user is provided with a list of those investigators to select from
510, and the application again evaluates whether the potential
number of patients is greater than the desired number 504.
[0075] If no unselected investigators exist, the application
determines whether any "red" investigators exist 512. These are
investigators that the application has identified as not satisfying
the criteria selected by the user initially. If these investigators
exist, the user is provided with an opportunity to revise the
enrollment criteria to be more inclusive 514, and the application
again evaluates whether the potential number of patients is greater
than the desired number 504.
[0076] In the embodiments shown, if no investigators are indicated
as "red" investigators in the model, then the application
determines that all investigators have been exhausted 516. The
application provides the user with the capability of adjusting
parameters or adding custom sites, i.e., sites that do not meet the
criteria but might be appropriate for the trial in any event. The
application again evaluates whether the potential number of
patients is greater than the desired number and repeats the
remaining evaluations until the number of modeled patients meets or
exceeds the number of desired patients.
[0077] FIG. 6 is a flowchart illustrating a method of adjusting
enrollment for geographic distribution 600 according to one
embodiment of the present disclosure. In the process shown, the
user reviews the country summary tab of a view provided by the
application 602. For example, in one embodiment, the user might
utilize the view shown in FIG. 2; in another embodiment, the user
might utilize the view shown in FIG. 7, which is described
below.
[0078] The user determines if the countries that the user desires
to be included are included 604. If so the application determines
if the country's modeled (i.e., predicted) number of patients meets
the desired number 606. If so, the process shown ends. If the
desired countries are not listed, the user selects investigators in
the additional desired countries 608, and the application displays
the updated list of countries.
[0079] If the countries' modeled patients do not meet the desired
number, the user can use a Required Patients section to set minimum
and/or maximum number of patients by country 610. If the number of
patients is now meeting the user's goals 612, the user can continue
the process for additional countries. If not, the user returns to
the process illustrated in FIG. 6
[0080] FIG. 7 is a screen shot of an enrollment country view 700 in
one embodiment of the present disclosure. This view provides the
user with the distribution of sites 702, patients 704 and costs 706
across countries. Such a view facilitates the tuning process
described herein and highlights risk (a.k.a. buffer) for particular
trials.
[0081] In the embodiment shown, the left hand column lists the
countries selected out of those available--of 89 countries, 9 were
selected, 7 contributed patients (non-red highlight). The view also
illustrates the number of available investigators--of 1274
investigators, 46 were selected, and 34 contributed patients. The
view also illustrates the potential patients and modeled
patients--of 539 potential patients (Historical Max), 201 patients
were modeled/predicted. In this case, the desired number of
patients is 200. The view also provides the user with the predicted
time for the trial--8 months.
[0082] The risk or buffer is determined by comparing the available
number of countries or investigators to the number selected and the
number that are eventually modeled. The larger the difference, the
lower the risk. If the modeled number is very nearly the same as
the number selected, then a higher likelihood exists that the
modeled plan will not be successfully implemented. The patient
buffer is determined by comparing the number of potential patients
to the number of modeled patients.
[0083] FIG. 8 is a screenshot of a graph 800, illustrating the risk
associated with various scenarios in one embodiment of the present
disclosure. In the embodiment shown, the best 804, median 806, and
worst 808 scenarios are shown to highlight buffer or risk in the
model. Time is shown on the x-axis, and the number of patients is
illustrated on the y-axis. The number of patients desired is 200,
which is illustrated by the solid line 802 from left to right
across the graph. In this example, all the illustrated scenarios
meet the desired goal for patients, but in the worst case scenario
808 (the farthest right line), time is significantly impacted.
[0084] FIGS. 9, 10, and 11 are screen shots of enrollment views in
embodiments of the present disclosure. These screenshots illustrate
how data may be brought together and displayed to the user in a way
that allows the user to vary the parameters of the clinical trial
to balance the risk of various scenarios with the goals of the
trial.
Advantages
[0085] Embodiments of the present disclosure provide many
advantages over conventional methods of predicting the enrollment
for clinical trials. For example, embodiments of the present
disclosure allow the enrollment process to be wholly data driven.
The performance data from previous trials is collected and
organized for use in the model. The team is then able to set
expectations for the types of investigators best-suited to conduct
the trial. For example, investigators with many patients can often
start up very quickly.
[0086] Once the types of investigators are chosen, embodiments of
the present disclosure are able to take a mathematical approach to
analyzing and presenting data regarding the actual investigators
and investigation sites. The embodiments can then create graphical
representations, e.g., line graphs that display information, such
as predictions for likely scenarios based on average performance as
well as best and worst-case scenarios based on outlier data.
[0087] Conventional systems are statistical in nature; they use
real-world data but treat all data as the same. In other words,
these conventional systems use a pool of performance data that
includes all of the studies ever done to create an average. For
example, the system may run data through a model 100 times to
derive variability. This is both less accurate and can take many
minutes or even hours to execute. In contrast, embodiments of the
present disclosure build the models dynamically. They are able to
illustrate the effect of a particular variable or set of variables
on a planned clinical trial and present that data in real-time.
[0088] Further, embodiments of the present disclosure provide a
user with the ability to understand the variability inherent in
planning clinical trials. For example, if a user is contemplating
using physical investigation sites in China versus the U.S., the
risks and costs associated with each of those countries is
illustrated to the user so that the user can make an informed
decision regarding the risks. Some factors that drive variability
and which are available to the user in various embodiments include:
the patient community, investigators, countries
General
[0089] The foregoing description of the embodiments of the
disclosure has been presented only for the purpose of illustration
and description and is not intended to be exhaustive or to limit
the invention to the precise forms disclosed. Numerous
modifications and adaptations are apparent to those skilled in the
art without departing from the spirit and scope of the
invention.
* * * * *