2.
1
Executive Summary
The forprofit start up company NextUp Care is conducting pilot studies in order to test and
confirm the effectiveness of its product to potential customers. The product in question is a
subscription to a scheduling system based on an optimization algorithm intended to reduce wait
times for subscribed hospitals. The implementation of the scheduling system is planned to start
off with scheduling MRI and CT scans as the length of such appointments are less variable than
more hands on health services. Currently, pilot studies are estimated to take anywhere between
46 months to complete depending on the hospital. It is in the client’s interest for such pilot
studies to both take less time and less input on part of the client.
The pilot studies require more time to implement than desired, the root causes including the lack
of dedicated labour and poor data quality. Due to the size of the company and its external
relation to the hospitals pilot studies are being conducted at, there are various constraints that
plausible solutions must adhere to. Financial and personnel cost should be minimized. Proposed
solutions cannot interfere with the normal operation of the hospitals.
Solutions considered included Volunteer Aid with incentives, Data Assessment and Endorsement
of Electronic Health Records (EHR) via partnership with Telus Health.
Pilot studies currently make use of volunteers from a pool of hospital staff and associated doctors
who are normally preoccupied with their primary occupation. Currently, there is an untapped
labour pool found in populations of both high school and undergraduate students who are on the
waiting lists for unpaid internship opportunities.
A Data Assessment is a process which examines the health care systems current data quality
within its multiple data collection systems in order to evaluate the possible data cleaning that
may be required, or if the data quality is not up to standard.
EHR systems make it possible for patient data to be available to various stakeholders outside the
confines of individual hospitals which can minimize the need for volunteers to collect data.
Endorsing Telus Health’s EHR system, which is the most prevalent EHR system in Ontario, can
aid in pushing Ontario’s health care system towards a standard, thus making the pilot process
faster for NextUp Care.
Taking into account the the cost versus the benefits during a pilot study of each solution, our
recommendation is the Data Assessment. There are things to consider if any of the proposed
solutions are implemented. If volunteers are pulled from schools in order to aid the study,
resources may need to be allocated to organize a sub group. If Data Assessments will be used to
select for ideal candidate hospitals for pilot studies, there must be plans to be made when such
hospitals are exhausted. If Telus Health’s EHR becomes widespread among hospitals further
integration between Telus Health’s EHR and NextUp Care’s scheduling system should be
pursued to eliminate the human intermediary between the two systems for greater efficiency.
3.
2
1. Introduction
NextUp Care is a forprofit software company looking to sell a subscription for a scheduling
system based on an optimization algorithm advertised to reduce wait times for participating
hospitals. The scheduling system is still in its infancy and has yet to be proven to the company’s 1
prospective customers, thus the first pilot study is being conducted at Toronto East General.
Participation in pilot studies is open to any interested hospital and such hospitals are promised a
free 1 year subscription to the system as compensation if they choose to implement. There is no
specific methodology given to the volunteer staff and collection is driven entirely on collecting
the required inputs for the model such as booking data, patient data, cancellation data and
noshow data along with reason codes. The focus of the company’s system at this stage is the
scheduling of MRI and CT scans which involves less variability than things such as surgery.
Currently, pilot study is estimated to take anywhere between 4 to 6 months to complete which is
longer than desired.
2. Problem Description
NextUp care is currently conducting a pilot study. It’s goal is to centralize MRI scheduling for
most, if not all, of the hospitals in Ontario, and use that to help alleviate the high wait times some
patients face. The client, Wayne Li, the CEO of NextUp Care has presented a case with the
intention of reducing the time required to conduct data from pilot studies as well as reduce the
amount of input required on behalf of the client to implement them. The intent is to maximize
data collection per unit time, thus saving time if data is collected faster. The client has stressed
that pilot studies require more time and input than desired. Current pilot studies are estimated to
require anywhere between 4 to 6 months to complete. The client has stated the slowdown caused
by the dependency of the working group, and the entire data collection process.
2.1. Stakeholders
The client representative evaluating the case solutions is Wayne Li, the CEO of NextUp Care.
Excluding himself, his team responsible for the current pilot project at South Lake consists of Dr.
Patrick and 2 strategic advisors with 12 spots for data analysis as well as process analysis open.
NextUp Care is targeting at various hospitals in Canada. Several hospitals are undergoing pilot
studies in order to discern the feasibility of NextUp Care’s commercial solution. Those who
participate in pilot studies outside NextUp Care’s employ are all volunteers. Volunteers consist
of staff at various levels in the healthcare system.
2.3. Problem Definition and Scope
NextUp Care is targeting Canadian hospitals to help test its pilot. Due to the differences in which
individual hospitals conduct themselves, not all pilot studies can be implemented in the same
way which leads to fluctuating longer than expected delays in the process.1
The founder of
1
Li, Wayne. (2015). NextUp Care. Retrieved March 15, 2016, from http://www.nextupcare.com/.
4.
3
NextUp Care has requested the consulting team to come up with solutions that can decrease the
time taken to complete pilot studies.
Currently, NextUp Care is running its first pilot at Toronto East General. In this pilot, NextUp
Care collaborates with many departmental heads and staff under a working group that help
gather different sets of data to test in NextUp Care’s product. Much of the time spent on the data
collection process is the wait time between request for data from the working group, and delivery
of requested data.1
For the scope of this case, since NextUp Care does not have influence over a
hospitals healthcare system, which mandates its own set of priorities for the working group, the
consulting team will focus on areas, processes and best practices in which NextUp Care can
improve on its end to help expedite the process, thus helping decrease the total time taken during
the pilot phase.
3. Problem Analysis
3.1. Lack of Dedicated Labour
The first main cause identified is a lack of time dedicated towards data collection. Many
participants in the working from for NextUp care are physicians, as Wayne pointed out in the
interview, and a large reason as to why physicians cannot commit to deadlines or consistent
hours is because most of them have demanding work schedules.1
Secondly, physicians are a part
of a unique population that is more frequently approached for surveys and research opportunities,
thus making them more hesitant to accept or on the opposite end spreading too thin by accepting
many side projects , which may be the case with NextUp Care. Lastly, most of the working 2
group has a set of priorities that supersede the data collection task requested by NextUp Care.
These three underlying reasons may be why there is a longer than expected delay in NextUp
Care’s pilot study and may occur in future pilot studies. Since most of these reasons are out of
the control of NextUp Care, the proposed solutions will focus on helping alleviate the task from
physicians in the working group, and creating less dependency on them in the data collection
process.
3.2. Poor Data Quality
From the interview with the client, the team identified poor data quality as one of the root causes
that is creating delay in the current pilot project. Poor data quality can be caused by a lack of
adequate data infrastructure, which can lead to mismanagement of data and create bad data . The 3
hospital NextUp Care is running its first pilot on, uses a fragmented data management
methodology in which each clinic manages its own data in a paper based system. Not only is the 4
paper based system time consuming, but also creates duplicate entries, conflicts with business
rules of data points and its semantics depending on healthcare assistant's memory and
2
Flanigan, Timothy S, Emily McFarlane, and Sarah Cook. "Conducting survey research among physicians and other
medical professionals: a review of current literature." Proceedings of the Survey Research Methods Section,
American Statistical Association 15 May. 2008: 413647.
3
N/A. (2014). Top 5 Reasons to Implement a Data Quality Solution. Retrieved March 22, 2016, from
http://sapinsider.wispubs.com//media/Alloy/Whitepapers/BackOfficeEB_Top5ReasonsUS.pdf.
4
N/A. (2009). Waiting from the Hospital Perspective. Retrieved March 19, 2016, from
http://www.cprn.org/documents/28492_fr.pdf.
5.
4
convenience at the time of recording.3
Since there is a lack of structure in a paper based
management system, business rules on data cannot be enforced.3
For example, recording booking
times on paper, a hospital administrator can omit arrival times of patients to save time if current
ward queue is busy, thus creating confusion and bad data for future research purposes.
Furthermore, in an investigation that measured data quality of a paper based system, it was
observed that the paper based system had over 30% of records as incomplete caused by bad data
compared to an electronic one, which only had 6%. Thus with poor quality of data, it requires 5
more cleaning and validating data, which in turn takes more time and resources.
4. Possible Solutions
The constraints a startup company like NextUp Care faces must be taken into account. Both the
amount of funds and number of personnel the company is willing to spend to improve the
efficiency of data collection is low. As NextUp Care is merely an external organization in respect
to the various hospitals undergoing a pilot study, solutions cannot be overly intrusive to the
hospital’s normal function.
4.1. Volunteer Aid from Community Outside Hospital and Incentives
One common way is to use high school or undergraduate volunteers for dedicated data
collection. Since the doctors and other staff have busy schedules, NextUp can hire volunteers as
associates and provide incentives for volunteers. For students who seek experience before
entering medical school, opportunities such as data collection at a hospital are fairly attractive.
According to UHN, approximately 2000 students are volunteering for their 4 direct hospitals. 6
Many more hospital such as St. Michael's, NYGH and Sick Kids are getting thousands of
volunteers and tens of thousands of working hours. If NextUp can negotiate the terms with pilot
hospitals and come up with appropriate training schedules and supervision, students can be very
helpful to this volunteer shortage.
One concern about hiring more volunteers is the cost. Even through volunteers sound free, there
is hidden cost in training and supervision. The training of volunteers requires experienced staff
and designated training sessions. The process of finding people with expertise in training and the
time it takes to train the volunteers are both costing the work hours of the hospital staff. In order
to build a sustainable staffvolunteer relationship, coaching hospital staff to work closely with
volunteers is also necessary. Furthermore, there should be designated supervisors to provide
feedback and information to volunteers. As a startup, NextUp is soon expecting a fundraising 7
which might provide enough incentives for hospitals to hire volunteers.
5
Sarkies, M., Bowles, K., Skinner, E., Mitchell, D., Haas, R., Ho, M., et al. (2015). Data collection methods in
health services research: Hospital length of stay and discharge destination. Applied clinical informatics, 6, 96109.
6
N/A. (2015). Volunteer Information Session (UHN). Retrieved March 18, 2016, from
http://www.uhn.ca/corporate/wayshelp/Volunteering_UHN/Documents/Volunteer_Info_Session.pdf.
7
Handy, Femida. (2004). Promising practices for volunteer administration in hospitals ... Retrieved March 18, 2016,
from http://sectorsource.ca/resource/book/promisingpracticesvolunteeradministrationhospitalsmanual.
6.
5
4.2. Data Assessment
Quality of data can play a crucial role as it can be a determinant on the amount of work a firm
may have to do to make sure the data is up to standard. One of the root causes identified in this
case was the lack of good quality data that is causing NextUp Care’s pilot process to slow down.1
To mitigate against this potential problem, the team proposes an initial Data Assessment process
that is added on before taking on any new pilot projects. The Data Assessment process examines
the health care systems current data within its multiple system and identifies factors that can
cause data to be below required quality threshold.5
This process will add additional work to the
NextUp Care pilot procedure, but will pay dividends afterwards with the time saved in the data
cleaning and quality control process. The Data Assessment process can identify whether a
hospital has the required data for scheduling optimization, and whether it is accurate and meets
the NextUp Care requirements. One major drawback to this solution is the extra time that will be
invested in the initial stages of every potential pilot project. For example, if NextUp Care gathers
interest from two hospitals, it will have to do two separate Data Assessments. Moreover, if the
two hospitals fail the assessment, the process will have to start again at a different hospital.
However, without this assessment, NextUp Care would potentially accept two hospitals in its
pilot project only to be faced with daunting data quality issues that may add more time
requirements to the pilot than the time cost incurred for an Data Assessment. 8
The following Data Assessment is recommended to ascertain a better picture of current hospital
data quality. The Data Assessment includes checks on : 9
● Relevance degree at which data meets current and future needs
● Accuracy measure between actual data and data definition (business rules)
● Coherence data understandability
● Consistency how often is good data produced
● Accessibility how easily can the data be accessed
4.3. Endorse Electronic Health Records (EHR) via partnership with Telus
Health
Telus Health has recently acquired Ontario’s largest EMR (Electronic Medical Records)
company, Med Access as of March 3, 2016 making it the dominant provider of EMR systems.
Ontario is already one of the 6 provinces in Canada with an EMR adoption rate of over 50%
making it eligible for a subsidy. Telus Health offers an Electronic Health Records (EHR) 10
solution to hospitals which shares patient data with various hospitals, clinics, community care
8
Collins, M. (2014). How to Build a Data Assessment Business Case ... Retrieved March 21, 2016, from
http://www.enterpriseappstoday.com/datamanagement/howtobuildadataassessmentbusinesscase.html.
9
Bergdahl, M. (2010). Handbook on Data Quality Assessment Methods and Tools. Retrieved from
http://unstats.un.org/unsd/dnss/docsnqaf/EurostatHANDBOOK%20ON%20DATA%20QUALITY%20ASSESSM
ENT%20METHODS%20AND%20TOOLS%20%20I.pdf.
10
Shaw, Gillian (2013). Telus Health acquires Ontario's largest electronic medical ... Retrieved March 18, 2016,
from
http://blogs.vancouversun.com/2013/02/26/telushealthacquiresontarioslargestelectronicmedicalrecordscompa
ny/.
7.
6
centers, regional and provincial authorities. EHRs not to be confused with EMRs can be 11
thought up as the natural evolution of EMRs. While EMRs are essentially digitized versions of
traditional paper medical records in function limited in access to hospitals, EHRs are easily
accessed by organizations outside the boundaries of the hospital which can include NextUp care
if a partnership deal is made with Telus Health. If a partnership deal is struck, relevant patient
data such as check in and out times can be accessed externally by NextUp Care directly without
the need of volunteers among the hospital staff which eliminates the wait time for volunteer
feedback. 12
Over a technology's life cycle, standardization has a significant effect on efficiency. The current
state of health care suffers from excessive customization. Though some degree of 13
customization is expected due to the various things patients are afflicted with, aspects of the
health care system such as MRI and CT scans which rely on machines are things which can be
standardized with no consequences.
Endorsing Telus Health’s EHR system to potential pilot study hospitals before pilot studies begin
is a step towards the standardization of data collection for pilot studies. The aiding in the
standardization of EHR systems across hospitals can eliminate inefficiencies by eliminating
variation in terminology, methodology and outdated manual methods which add to overall
system complexity. However, it will require a bigger buyin from the hospitals, as it may cost
them more upfront versus the standalone NextUp Care application.
5. Solution Evaluation and Recommendation
5.1. Evaluation of Solutions
Given the solutions, including Volunteer Aid, Data Assessment, and Endorse EHR, were
compared using a cost benefit analysis using the objectives that were critical during a pilot study
as advised by the client. A table of comparison is shown as below in Table 1.0:
Pilot Study Time Data Accuracy Cost of
Implementation
Implementation
difficulty
Volunteer Aid Decreased No change High Medium,
designated
training and
supervision
needed
11
N/A, (2014). Health Solutions TELUS Health. Retrieved March 17, 2016, from
https://www.telushealth.co/healthsolutions/.
12
Garett, Peter. (2014). EMR vs EHR – What is the Difference? Health IT Buzz. Retrieved March 18, 2016, from
https://www.healthit.gov/buzzblog/electronichealthandmedicalrecords/emrvsehrdifference/.
13
Tassey, G. (1999). Standardization in TechnologyBased Markets National ... Retrieved from
http://www.nist.gov/director/planning/upload/researchpolicypaper.pdf.
8.
7
Data Assessment Decreased per
individual pilot
Increased Medium Low
Endorsement of
EHR
Decreased No change High Medium, many
hospitals have
already equipped
with EHR
Table 1 CostBenefit Analysis of Potential Solutions Based off Client Criteria
5.2 Recommendation
The team recommends adding the Data Assessment process to help analyze a hospital’s data
quality and its infrastructure. Using the costbenefit analysis in table 1, all three solutions do well
against three objectives, which are Pilot Study Time, Data Accuracy, and Implementation
Difficulty. However, the cost of implementation metric is the deciding factor in choosing the
recommended design. The Data Assessment requires small upfront work to evaluate the data
system and data quality. The assessment will take only a fraction of the time (and cost) relative
to the time taken if the assessment was not done, and a commitment was made to a hospital with
poor quality of data. The goal of the Data Assessment is to help make sure any future
commitments for pilot studies do not become liabilities for NextUp Care, but rather assets to help
them roll out its product faster.
6. Conclusion
Our client, NextUp Care is interested in shortening their pilot study time at hospitals. Our team
identified the root causes as the lack of dedicated time and poor data quality. Three solutions are
selected to improve the situation: Volunteer Aid, Data Assessment, and the endorsement of EHR.
This report recommends Data Assessment for its low cost of implementation as it requires a
small amount of time upfront to analyze a hospital’s data quality and data infrastructure. This
assessment can help make more informed decisions as to which hospitals will be better suited for
the pilot study and will have less cost (i.e time cost) on NextUp Care, thus rolling out pilot
projects faster.
7. Future work
Firstly, to measure the time Data Assessment solution can save in a pilot project, the baseline of
4 months should be used, which is a conservative estimate of the duration of a pilot project with
no data assessment. Analyze the time it takes to do the data assessment on a hospital that passes
the data quality evaluation, and the time it takes to do the actual pilot project. If the total time
from initial data assessment to completion of the pilot project takes less than 4 months, then the
solution will have been successful.
Data Assessment allows for the identification of hospitals with the good data infrastructure to
yield a short pilot study length. However, if the data assessment fails to approve any hospital’s
data quality, NextUp Care should lower the standards of data quality for the assessment, and start
over again.
11.
10
Attribution Table
Presentation/Project: MIE 561 Case 3 Supervisor: Prof. Mike Carter
Document Name: Case 3 Report Date: March 22, 2016
This form must be filled out and signed for each teamwritten document. The completed form must be
attached to the document, though not included as part of the document (not included in the table of
contents, page numbers, or word count limits). It should accurately reflect each team members’
contribution to the document and in what way they contributed.
If there are irreconcilable differences that are preventing all team members from signing the attribution
table then each team member must write a letter (max 800 words) explaining their position on the
difference and suggest a solution.
Making fraudulent claims in an attribution table displays intent to deceive and is a serious academic
offence.
Section
Student Names
Ather Mohammad Michael Zhang Zack Pan
Executive Summary ET RD MR
Introduction ET RD MR, ET
Problem Description ET RD, RS ET
Problem Analysis RD ET ET
Solutions RD, RS, OR1, OR2 RD, RS, OR1, OR2 RD, RS, OR1, OR2
Solution Evaluation
and Recommendation
RD ET ET
Conclusion RD ET ET
Future Work RD ET
All FP FP FP
Fill in abbreviations for roles for each of the required content elements. You do not have to fill in every
cell. The “all” row refers to the complete report and should indicate who was responsible for the final
compilation and final read through of the completed document.
RS – research
RD – wrote first draft
MR – major revision
ET – edited for grammar and spelling
FP – final read through of complete document
for flow and consistency
CM – responsible for compiling the elements
into the complete document
OR – other
If you put OR (other) in a cell please put it in as OR1, OR2, etc. Explain briefly below the role referred to:
12.
11
OR1: PrePlanning
OR2: Solution Generation
By signing below, you verify that you have:
● Read the attribution table and agree that it accurately reflects your contribution to the associated
document.
● Written the sections of the document attributed to you and that they are entirely original.
● Accurately cited and referenced any ideas or expressions of ideas taken from other sources according
to an accepted standard.
● Read the University of Toronto Code of Behaviour on Academic Matters and understand the
definition of academic offense includes (but is not limited to) all forms of plagiarism. Additionally
you understand that if you provide another student with any part of your own or your team’s work,
for whatever reason, and the student having received the work uses it for the purposes of committing
an academic offence, then you are considered an equal party in the offence and will be subject to
academic sanctions.
Print Name:
Ather Mohammad Signature:
Print Name:
Michael Zhang Signature:
Print Name: Zack Pan Signature:
