Systems Development Life Cycle (SDLC)

Systems Development Analysis

Life Cycle:
Analysis/Planning Maintenance/
Evaluation
Design

Implementation

Systems Development Life Cycle (SDLC) SDLC Analysis/Planning Phase

ANALYSIS Analysis (and planning) – define scope and goals of project • Identify all Project Stakeholders * Organizational Governance will
• Organizational Governance* / Leadership be discussed in further detail
• Project Manager during our week eight lecture
DESIGN
Design (and develop) specifications, determine system • Steering Committees
requirements • Key stakeholders
• Users of the system (i.e., Physicians, nurses, other clinical disciplines)
• Define the System Goals
IMPLEMENTATION Test, train, install, go-live, support, evaluate
• System goal/purpose statement
• Should align with the strategic goals of the organization
MAINTENANCE/ Install regular updates, evaluate system’s ability to continue • Define Project Objectives
EVALUATION to meet the original goals • Project Charter

Sample Project Charter – Reducing preventable allergic reactions

Systems Development
Life Cycle: Design Phase

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]
Systems Development Life Cycle (SDLC) Systems Development Life Cycle (SDLC)
ANALYSIS Analysis (and planning) – define scope and goals of project
Analysis

DESIGN
Design (and develop) specifications, determine system
requirements
Maintenance/
Design
Evaluation IMPLEMENTATION Test, train, install, go-live, support, evaluate

Implementation MAINTENANCE/ Install regular updates, evaluate system’s ability to continue

EVALUATION to meet the original goals

SDLC Design Phase SDLC Design Phase

Workflow Process Mapping Elements
• Determine and prioritize the system requirements
• Develop current state and future state workflow analyses
• If acquiring a new system:
• Screen the marketplace for potential vendors
• Develop and distribute the Request for Information (RFI) or Request
for Proposal (RFP)
• Evaluate & rank vendors
• Contract negotiations

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

SDLC Design Phase SDLC Design Phase

CDS Rule Considerations CDS Rule Considerations
Setting where notification is delivered, rule output recipient, and
• How will the rule be triggered? delivery channel
• How will data be provided to trigger the rule? • Where: ED, hospital care unit, ICU, OR, physician office, wherever recipient is
(i.e., physical location)
• How will you make sure the rule triggers for the right • Who: patient, nurse, pharmacist, physician, other member of care team; internal
to system (i.e., who is the recipient of the CDS when the rule triggers action by
patients? the system)
• How (delivery channel): EHR (specific module, i.e., results review, CPOE, etc.),
• How will the rule results affect the care delivery process? PHR, mobile device
• When (specific point in workflow): i.e., associated with a specific information
system-based task
• What: User interaction with rule output, including rule actions and notifications
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press. Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link] [Link]
SDLC Design Phase SDLC Design Phase
CDS Rule Considerations • Develop / Configure CDS Interventions for greatest usability
• Rule results disposition (e.g., processing user response) and positive impact
• Facilitate specific user action (e.g., ordering an intervention,
documenting)
• Enabling user documentation that patient is an exception to the
rule
• How will rule behavior and effects be monitored?

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press. Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link] [Link]

CDS Intervention Types

• Applying the right CDS intervention format (from the CDS
Five Rights)
CDS Intervention Types • Categorized based on their typical presentation and user
interaction
• Further grouped by typical workflows in which they appear

A. CDS during data-entry tasks

1. Smart documentation forms CDS Intervention Types
2. Order sets, care plans and protocols
3. Parameter guidance
4. Critiques and warnings “immediate alerts” A. CDS during data-entry tasks
CDS B. CDS during data-review tasks 1. Smart documentation forms
5. Relevant data summaries (single-patient) 2. Order sets, care plans and protocols
Intervention 6. Multi-patient monitors
3. Parameter guidance
Types 7. Predictive and retrospective analytics
4. Critiques and warnings “immediate alerts”
C. CDS during assessment and understanding tasks
8. Filtered reference information and knowledge
resources
9. Expert workup and management advisors
D. CDS not triggered by a user task
10. Event-driven alerts (data-triggered) and reminders
(time-triggered)
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical
Decision Support (2nd Edition). CRC Press. [Link]
 Example: Structured Nursing Documentation
Standard Fall Precautions, Morse Fall Scale & Tailored Interventions
CDS Intervention Types
A. CDS during data-entry tasks Benefits:
1. Smart documentation • Provide complete documentation /
forms reduce omission errors
Examples: • Ensure critical data are captured
• Structured documentation forms • Provide coded data for other data-
• Pre-procedure checklists or driven CDS interventions
questionnaires • Provide calculated risk scores
• Medication Administration Record • Provide prompts to acquire specific
(MAR) forms that automatically information in the desired format
track missed meds (i.e., ”kg” for weight)
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

Example: VTE Prophylaxis Order Set with Risk Assessment

CDS Intervention Types

A. CDS during data-entry tasks Benefits:
2. Order sets, care plans • Promote adherence to care standards
and protocols • Facilitate compliance with
Examples: organization’s policies, best practices
and clinical pathways
• Order sets and care plans for standard • Often rate high in physician
ICU inpatient admissions acceptance, increasing the likelihood
• Conditional order sets that adjust for of being used and benefits realized
patient conditions, allergies, • Multi-step protocols can include
comorbidities or other variations collected order sets or care plans that
• Protocol helpers that lead a clinician are designed to be used in sequence.
to the various ‘order sets’ that are
part of a multi-step pathway

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

Example: Medication Dosage Alert with Parameter Guidance for Med

Dose Range and Frequency
CDS Intervention Types
A. CDS during data-entry tasks Benefits:
3. Parameter guidance • Proactive interventions help
Examples: ensure that drug doses,
• Guided dose algorithms based on
frequencies and other
weight, body surface area (BSA) and ordering parameters are
other parameters complete and correct
• Forms that facilitate entering complex • Parameter guidance may be
orders such as insulin sliding scales provided as simple lists (such
• Template to ensure complete as for recommended drug
documentation of reasons and dosages) or via more complex
alternative options when ordering
physical or chemical restraints algorithms

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]
 Example: Pregnancy Warning -Drug Conflict Alert with References

CDS Intervention Types

A. CDS during data-entry tasks Benefits:
4. Critiques and warnings • Reactive interventions prompt the
“immediate alerts” user to immediately address
Examples: possible errors related to new
data or orders just entered in the
• Drug allergy warnings system
• Drug interaction warnings (i.e., • Help enforce care standards and
interactions with other drugs, prevent adverse events
pregnancy, lab results, or food
• May provide critiques to help
• Inappropriate therapeutic ensure that additional
duplication warning documentation or orders
• Underdose/overdose alert necessary to make an order safe
and complete are entered
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

CDS Intervention Types CDS Intervention Types

B. CDS during data-review tasks B. CDS during data-review tasks Benefits:
5. Relevant data summaries • Concise presentations of important
5. Relevant data summaries (single-patient) (single-patient) data
6. Multi-patient monitors Examples: • Optimize decision making for a
• Health maintenance flowsheet showing specific patient’s circumstances by
7. Predictive and retrospective analytics fulfilled and needed screening tests ensuring all pertinent data are
• Chart that displays a patient’s noted and considered.
immunization status and highlights • Organize complex data sets to help
vaccines that are due care team members better
• ICU electronic “status board” that understand the overall clinical
provides a filtered view of a patient’s picture and to highlight needed
critical physiologic parameters, ventilator
settings, IV med deliver rates and other actions
info to facilitate ”at-a-glance” decision
making.
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

Example: Single Patient Summary with Clinical Reminders

CDS Intervention Types

B. CDS during data-review tasks Benefits:
6. Multi-patient monitors
• Provide a filtered, organized “at-a-
Examples: glance” look at key parameters and
• Emergency department tracking facilitate using the right information
system highlighting new patients, for decision making from a large data
critical lab results, admission and
disposition process set covering several patients
• Operating Room status monitor • Proactive intervention prevents many
showing the procedure progress in omission errors and permits faster
each room
response to important events
• Inpatient unit status monitor that lets
the nurse see which patients have • Does not necessarily require an
pending orders, due medications, or immediate clinician response
transfer/discharge events
Nelson R., PhD, RN-BD, Anef, Staggers F. A. N., PhD, RN, & FAAN Health Informatics: An Interprofessional Approach. [VitalSource Bookshelf]. Retrieved from Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link] [Link]
Example: ED Status Board

CDS Intervention Types

B. CDS during data-review tasks Benefits:
7. Predictive and retrospective • Display data over time, compare them
analytics
to benchmarks, enable users to drill
Examples: down into data details to understand
• Retrospective displays of root causes, and facilitate quality
performance, control charts, improvement projects
benchmarks for strategic guidance
and problem-spotting • Can be used to help find the causes
• Syndromic surveillance and bio for results that CDS interventions are
surveillance tools producing and to monitor
• Predictive tool to stratify pressure improvement trends
ulcer risk • Predictive analytics – risk stratify
individual patients
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link] [Link]

Example: Hospital Management Dashboard

CDS Intervention Types

C. CDS during assessment and understanding tasks
8. Filtered reference information and knowledge resources
9. Expert workup and management advisors

[Link]

Example: Infobutton User Interface

CDS Intervention Types

C. CDS during assessment and Benefits:
understanding tasks • Filtering is often accomplished
8. Filtered reference information through info buttons or evidence
and knowledge resources links
Examples:
• Ideally will filter then deliver
• Info button linking from problem list entry to
evidence-based treatment overviews for that information from reference sources
problem (i.e., textbooks, journal articles,
• Link from an immunization flowsheet to table guidelines and other resources)
listing standard immunization intervals into clinician and patient workflow.
• Health risk assessment results that generate a
set of information resources and educational
programs to be used by the patient
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press. Nelson R., PhD, RN-BD, Anef, Staggers F. A. N., PhD, RN, & FAAN Health Informatics: An Interprofessional Approach. [VitalSource Bookshelf]. Retrieved from
[Link] [Link]
 Example: Visual Dermatology guide

CDS Intervention Types

C. CDS during assessment and Benefits:
understanding tasks
9. Expert workup and management • Typically use patient data from an
advisors
EHR or from direct user entry to
Examples:
suggest diagnoses that may not
• Diagnostic decision support program that
accepts symptoms, signs and general test have been considered
findings and suggests a ranked list of
possible diagnoses • Suggest tests and workup steps
• Workup advisor that suggests best next that can further clarify the
tests for patients, given their clinical diagnosis
findings
• Visual dermatology guide that helps • Same model can be used to advise
clinicians identify rashes based on
characteristics such as color, size, and on recommended therapies
texture specific to the patient’s situation
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link] [Link]

CDS Intervention Types CDS Intervention Types

D. CDS not triggered by a user task D. CDS not triggered by a user task Benefits:
10. Event-driven alerts (data-
10. Event-driven alerts (data-triggered) and reminders (time- triggered) and reminders (time- • Raise awareness about events that are
triggered) occurring outside routine, patient-
triggered) Examples: specific workflows
• EHR, text message or pager alert about an
abnormal and critical lab result • Can also detect important non-events
• Reminder that a patient is due for an (when important events have not
immunization occurred) in a prescribed time period
• Alert that a critical follow-up lab test has or as indicated (i.e., screening, disease
not been performed monitoring tests, or health
• Program that generates standard letters to maintenance exams)
patients about their lab results
• Disease management alert • Prevent omission errors and promote
faster response to critical situations
Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F., & Sittig, D. F. (2012). Improving Outcomes with Clinical Decision Support (2nd Edition). CRC Press.
[Link]

Example: Text Message Alert for a Critical Lab Value

CDS Design Features

[Link]
Knowledge based vs. Non-knowledge based CDSS Knowledge-based CDSS

Knowledge based CDSS Non-knowledge based CDSS

• Contain rules, guidelines, or other compiled knowledge
derived from the medical literature
• If-Then Statements

Sutton, R. T., Pincock, D., Baumgart, D. C., Sadowski, D. C., Fedorak, R. N., & Kroeker, K. I. (2020). An overview of clinical decision support systems: benefits, risks, and
strategies for success. NPJ digital medicine, 3, 17. [Link]

Knowledge-based CDSS Non-Knowledge Based CDSS

Example: • Artificial Intelligence (AI)
• Big Data / Data Analytics
IF current order’s medication is in Nephrotoxic Drug group • Natural Language Processing
↓ • Machine Learning
AND Creatinine clearance > Age-specific threshold • Deep Learning
↓
THEN Display warning “Use caution when prescribing
nephrotoxic drugs in this patient”

Non-Knowledge Based CDSS Non-Knowledge Based CDSS

Artificial Intelligence (AI) Big Data/Data Analytics
• When computers and other machines mimic human cognition, and
are capable of learning, thinking, and making decisions or taking Four categories:
actions • Descriptive Analytics: create simple reports and visualizations that show what
occurred at a particular point in time. Least advanced analytics tools.
• Diagnostic Analytics: allow analysts to dive deep into the data and determine
AI in Healthcare - the use of machines to analyze and act on medical root causes for a given situation.
data, usually with the goal of predicting a particular outcome. • Predictive Analytics: use highly advanced algorithms to forecast what might
• Clinical use cases: happen next. Often these tools make use of artificial intelligence and machine
learning technology.
• Machine learning for medical diagnosis purposes
• Use of predictive analytics to develop improved preventative care • Prescriptive Analytics: tell organizations what they should do in order to achieve
a desired result. These tools require very advanced machine learning capabilities,
recommendations for patients and few solutions on the market today offer true prescriptive capabilities.

[Link]
Non-Knowledge Based CDSS
Natural Language Processing
• Allows machines to read, understand, and interpret human
language.
• Examples: Siri, Alexa, Google Assistant, Google Translate
• Healthcare-related examples: Speech-text dictation to aid with
clinical documentation, computer-assisted coding, clinical trial
matching

[Link]

[Link]