Top Banner
Table of Contents Executive Dashboard Development Guide April 2009 © 2009 Healthcare Information and Management Systems Society (HIMSS)
32

Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

Jun 07, 2018

Download

Documents

truongxuyen
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

Table of Contents

Executive Dashboard

Development Guide

April 2009

© 2009 Healthcare Information and Management Systems Society (HIMSS)

Page 2: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

2

Table of Contents

Overview.......................................................................................................................................3 Defining the Need .........................................................................................................................4 Using Information within the Dashboard .......................................................................................6 Defining the Key Performance Indicators and Metrics................................................................11 Defining the Infrastructure...........................................................................................................16 Transforming Data ......................................................................................................................19 Presenting the Data in an Executive Dashboard ........................................................................22 Environmental/Cultural Factors in Your Organization.................................................................24 ROI/Organization Benefits ..........................................................................................................26 Lessons Learned ........................................................................................................................28 References..................................................................................................................................30 Contributing Authors ...................................................................................................................32

Page 3: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

3

Overview Executive dashboards come in all shapes and sizes. They serve many different purposes and audiences, and use various infrastructures as their foundations. Due to variability, it is not uncommon for an organization to overlook key aspects when beginning to establish a dashboard. In simple terms, an executive dashboard is an electronic display comprised of key performance indicators (KPIs). Yet, who or what determines the KPIs to be used and what needs to be considered? How is the electronic display developed and who determines the amount of information, the level of detail or the display components? Where should this information be stored and what are the options available? This document will try to answer many of these questions. An executive dashboard goes through a lifecycle as it is a project all its own. The following outlines the key steps and as in any lifecycle, the project moves in a continuous loop.

1. Project Initiation Establish governance and decision-making boards Establish change control processes

2. Design Phase Identify the target audience (address executives, managers, clinicians) Determine potential KPIs Evaluate data sources for the KPIs (Do you have the data for your KPIs?) Evaluate current and potentially required software/hardware for the dashboard Address workflow considerations and associated training approaches

3. Build Phase Develop mechanisms for getting the data from the sources (systems, paper, etc.) Set up selected ETL (Extract, Transform, Load) tools and other 3rd party specialty tools

(SAP/Crystal/BO, etc.) Build training materials

4. Test Phase Ensure all the tools work to get accurate results Ensure the updated workflow processes work with the new dashboard

5. Rollout Phase: Conduct training and begin production use 6. Monitor Phase: Review and monitor the dashboard – are the results actionable? Does

the workflow continue to work with the dashboard? Measure and trend data from the dashboard – evaluate impact on the organization. Determine if there are other KPIs that need to be measured.

Page 4: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

4

Defining the Need

Before making a decision, something inside of us convinces our brains of a desire or a need to take an action. Need is defined as “A condition or situation in which something is required or wanted.” That something can be the hiring of a new nurse FTE due to patient volume increases or the expansion of a clinical service to serve an aging population. In both instances need may be generated and influenced by a multitude of internal and external influences, some not in our own ontrol. c

Once we identify the need, we are often faced with many decision points. The process through which we generate conclusions can be described as follows: we collect information, process the data we have available, take a course of action, and then hope for the best. If we are lucky enough, we can quickly evaluate our decisions, collect more information and modify our course. In this section of this resource tool kit, we are considering those factors that create the need for hospital organizations to consider the purchase of Business Intelligence (BI) tools. Without first considering the need for these tools and their importance in guiding us to make decisions, the rest of the content in this resource tool kit loses its relevance. As the healthcare industry begins to build the case for the purchase of BI tools, one only needs to look at internal organizational and external pressures that exist within our space. Listed below are some

xternal factors that may drive your organization’s need to invest in BI tools. e Economic Factors – Changes that negatively impact the revenue generated by the healthcare institution/provider for patient care services:

• Changes in payer mix with a movement away from “higher” paying payers. • Growing number of patients seeking care who do not have insurance. • Closing of hospitals in poorer areas. • Domination by one or a few payers that reduce the reimbursement rate • Process bottlenecks and other workflow impediments impacting cost and timelines of billing.

Governmental Factors – Restrictions and/or demands placed on healthcare institution/provider by government agencies:

• Increased CMS reporting demands on quality. • Changes in reimbursement tied directly into quality indicators. • Elimination of payments for care delivered as a result of medical errors.

Page 5: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

5

Environmental Factors – Other external changes that place pressures on local healthcare systems:

• Increases in malpractice costs causing many physicians to leave the area. • More demands by local employers for hospitals and physicians to reduce medical errors. • Smarter patients who use the Internet before seeking care. • Increase in immigrant population to a geographic area who have had little historical preventive

care. Each one of the above external factors places pressures on internal operations within healthcare entities to adjust. These adjustments may come from:

• Elimination or creation of service offerings. • Re-structuring of payer contracts. • Re-distribution or elimination of staff. • Investment in technology or other capital.

However, in order to approach the right adjustment strategy, tools capable of capturing, analyzing and reporting clinical, operational, financial and satisfaction data are required to guide in the decision-making process. It is clear that many external factors are now placing pressure on healthcare professionals to make internal organizational decisions. To make well-informed decisions, healthcare organizations need the tools to help them with these choices. Once the need for BI has been established and reaffirmed, the next steps include executive sponsorship and an evangelist capable of convincing others of this need and its benefits through out the enterprise.

Page 6: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

6

Using Information within the Dashboard

As noted within the lifecycle, the first step you take in defining an executive dashboard is to identify the target audience and what the information from the dashboard is going to do for these groups. It is important to think about tying back quantifiable business benefits to requirements you would put into a dashboard. An organization may have multiple audiences it is trying to serve, which would result in multiple dashboards. However, before you can begin, it is critical to develop a clearly defined scope document and ask yourself the following:

Who are the intended users? What are you trying to show, learn or understand through the dashboard? What is the soft and hard return on investment (ROI) for the hospital for having a dashboard

with KPI’s? Which departments and which stakeholders stand to gain the most with an initial dashboard

deployment? What is the frequency of the updates? What is the intended lifecycle of the dashboard?

In healthcare, most dashboards display three kinds of indicators. These indicators are financial, clinical, and operational indicators. The financial indicators address the financial health of the organization while the clinical indicators speak to both quality care and the operational aspects of a clinical department. For consideration, we have provided background information on how an organization may use a dashboard for both financial and clinical information. Financial Information An organization can use a dashboard (or multiple dashboards) to monitor its financial health on an ongoing basis. The dashboard should convey fundamentals about the business, pinpoint areas of growth as well as areas of concern, and most importantly, aid decision-making. The KPI’s defined for the organization serve as the foundation for the dashboard. The KPI’s can be shared between different organizational levels or they can be unique to a given level or a given department. The examples below illustrate the use of KPI’s at three different levels. Executive Level:

• Average daily census • Average length of stay • Case mix index • Payer mix • Inpatient cost per discharge • Inventory cost per adjusted discharge • Cash position of the organization • Days in accounts receivable • Admissions/procedures by various services • Average daily revenue • Actual revenue in relation to budgeted revenue • Operating expenses in relation to budget

Middle Management Level:

• Unit census

Page 7: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

7

• Admissions, discharges, and transfers • Inpatient cost per discharge • Case mix index • Procedures by day, week, month-to-date • Visits per day, week, month-to-date • Labor costs per visit • Supply costs per visit • Actual revenue in relation to budgeted revenue for department • Operating expenses in relation to budget for department • Cash position of the organization • Overtime hours

Staff Level:

• Unit census • Admissions, discharges, and transfers • Procedures by day, week, month-to-date • Visits per day, week, month-to-date • Cash position of the organization

For illustration, let’s look at the KPI “cash position of the organization.” This particular indicator is viewed by three different levels of an organization. In a typical hospital, estimated cash collections are budgeted monthly. In order to meet the estimated target, personnel in the patient accounting department must bill accounts in a timely manner; follow up with insurance carriers on any problematic accounts; and collect residual balances and self-pay claims from patients. The goal of the department is to meet or exceed the estimated cash collection target monthly. In order to stay on target, the patient accounting director monitors this KPI daily on his/her dashboard. To keep the staff apprised of where they stand in relation to the target, the director posts a graph produced from the dashboard on the departmental bulletin board daily. The staff monitors the graph on a daily basis in an effort to prioritize activity and ensure departmental success. If the daily results are positive, the director take steps to reward the productivity of the staff. If the daily results indicate that the month-end target may not be met, the director can work with staff to help prioritize work, eliminate obstacles or take additional action to reverse the negative trend. Just as the information is reported to the director and staff, the information is also reported to the chief financial officer and the chief executive officer on their dashboards so that they are aware of how the department is doing over the course of a month as it relates to cash collections. In addition, at month-end, the information is used to report to the board on the cash position of the hospital. In another example, “procedures by day, week, month-to-date” is an important indicator at the department level. This indicator is important to a clinical manager because it drives the manager’s resource needs including personnel and supplies. It also predicts what the manager may anticipate in the way of revenue for his/her operations. Likewise, this is an important indicator for the departmental staff as it defines workload and the type of skill needed for the demands of the department. When using this indicator on a departmental dashboard, along with known scheduled appointments, the manager can adjust staffing to meet patient care needs. The manager can also increase or decrease the supply inventory of the department based upon the actual growth or contraction of activity within the department. As demonstrated by the above examples, monitoring this type of information on a continuous basis would allow the management and staff of an organization to more rapidly recognize changes in financial position so that corrections could be instituted, when needed.

Page 8: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

8

Clinical Information In any operational setting, the presentation of prioritized clinical quality KPI’s can drive continuous improvement in the areas of quality of care and safety. A by-product of improvements to patient care and safety processes and systems can also bring about financial heath and stability to the entire organization. Up-to-date KPI information displayed on a dashboard empowers staff to initiate continual process improvement. A practical, in situ explanation of how a transformation culture would form and grow is reviewed below. Here, we will follow the Clinical Quality KPI “Fall Risk Assessment” through an Emergency Department (ED) to Executive Management to see how it is used to improve the quality of patient care. Before beginning the explanation, it is appropriate to consider the form that a unit-centric dashboard might take. In an ED, the traditional dashboard elements such as graphs, gauges, or statistics are likely replaced by a spreadsheet-style view (roughly similar to the venerable grease board). These elements are distributed to the workstations throughout the department with each workstation having its view customized to display only data that is pertinent for that area of the department. In this view, each line represents a particular patient, and the columns contain patient demographics, assigned resources, orders, KPIs, and other information relative to the patient on that line. This view is in a constant state of change as people move about, orders are completed, and finally, a patient disposition is made. Now, assume that a new patient who presented in the ED with a severe headache has been triaged then placed into examination room one. Triggered by the triage process, raw data begins to be processed into information for presentation via each localized dashboard. This information now includes triage information, patient history, current medications, and so forth. As raw data is processed, it is recognized that the patient is over 60 years old, has osteoporosis, and is taking medication to reduce anxiety. All three data points indicate a significant risk that this patient will fall and suffer a bone fracture if allowed to freely ambulate. Appropriately, data displayed on the dashboard in the hall outside of exam room one is highlighted, made to blink, or in some other way indicate that a high-risk situation has emerged. This is noted by the attending nurse who immediately initiates the standard protocol used when a high-risk patient is present. The raw data feedback through the Clinical Information System (CIS) contributes to updated information, which in turn signals the dashboard that the risk has been mitigated, allowing it to relax the graphic display of the fall risk KPI for that patient. All of this information is compiled and stored for use in more traditional dashboards and reporting by other stakeholders in the organization. For example:

1. Later that same day, a unit supervisor views information showing trends and response times for all fall risks seen in the ED during that day. The supervisor concludes that the standard protocol could be improved if the workstation local to the patient displays the risk in a different form and requests the change from the IT department.

2. At the beginning of the next week, the Director of Nursing examines trends and response times for this risk across all departments in that facility for the previous week. The Director of Nursing notes that improvements in certain departments have earned a reward for the supervisors of these departments because they have met the target for this measure. The Director of Nursing notifies the other supervisors that they will meet during the following week

Page 9: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

9

to review current fall-risk protocols in each department and more closely align them to the protocols of the departments that successfully met the target measure.

3. At the beginning of each month, the Chief Nursing Officer (CNO) compares the total number of falls and other quality KPIs for the previous month to the same month last year. The CNO notes the improvements in quality KPIs, then awards the Director of Nursing the predetermined reward.

4. At the beginning of the quarter, the Chief Financial Officer (CFO) records all costs associated with falls and other clinical quality KPIs, correlates them with the most recent patient satisfaction ratings, and determines that the benchmark for the predetermined executive level awards has been met.

Another practical example that an acute care facility would want to monitor closely is that of pressure ulcer given that it is one of the eight “reasonably preventable" hospital-acquired conditions that the Centers for Medicare and Medicaid Services (CMS) will no longer pay for if not present on admission. Pressure ulcers can cause serious injury or death and result in increased cost to treat. A pressure ulcer that is present on admission to an acute care facility must be identified and documented by the clinical staff and by the attending physician. Obtaining hospital reimbursement payment is based upon the physician documentation. Under the new CMS payment plan that took effect on 10/1/2008, pressure ulcers that are present on admission will qualify for a higher reimbursement only if the presence of Stage III or IV ulcer is noted in the medical record within two days of an inpatient admission. If the documentation of an initial skin assessment is automated as part of the CIS, then the raw data feed from that system would enable an acute care facility to proactively monitor and inform the clinical team and leadership of any required follow-up actions. For example:

1. Whenever a patient is admitted to an inpatient unit, the CIS can be configured to trigger an alert reminding the nurses and physicians assigned to the patient’s care that an initial skin assessment has not been completed within a defined period of time (e.g., within 8-12 hours of a patient’s admission, depending on an institution’s policy).

2. The same information can be viewed by the Nurse Manager or Supervisor to see if there are any skin assessments that have yet to be documented in the CIS, so that they can follow-up to ensure their timely completion in the system.

3. The data feed regarding skin assessments, Braden scores, pressure ulcer staging, and the stage of the worst pressure ulcer will also be reviewed by the Wound Care Clinical Nurse Specialist with the nurses and the interdisciplinary team. A key focus will be on clinically managing patients with pressure ulcers that are present on admission and on proactively identifying patients who may be at risk for developing pressure ulcers.

4. The Director of Nursing reviews the dashboard data with his/her respective Nurse Managers on a routine basis to determine the number of patients with pressure ulcers, the number of patients with pressure ulcers that are present on admission, and the number of patients with pressure ulcers that are hospital-acquired.

5. The CNO reviews and compares the dashboard data with the Director of Nursing on a quarterly basis, noting among other things, the incidence and distribution of pressure ulcers by inpatient units and clinical services, the timeliness and efficiency of managing the pressure

Page 10: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

10

ulcers, and their overall impact on staffing and resources.

6. The CNO and CFO review the costs associated with pressure ulcers and related clinical quality KPIs with benchmark data from peer organizations to compare quality data and care outcomes.

As we can see, the process of compiling raw data into information, then presenting it in the appropriate form, at the right time, and to the right entity creates a quality improvement tool. This tool provides a healthcare organization the means and methods to engage existing clinical processes, then steadily refine and improve them, thus producing safer and more effective systems. The ongoing cycle of compiling, presenting, and analyzing KPIs on regular intervals initiates, then continues to fuel, ongoing transformational change.

Page 11: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

11

Defining the Key Performance Indicators and Metrics Once an organization has defined a need for an executive dashboard, they must go through the process of defining what they will include in their dashboard. While the terms ‘KPIs’ and ‘metrics’ are often used interchangeably, there is a difference:

KPIs are the indicators that are included in the dashboard. They are the “what” that is being reviewed.

Metrics are “how” the KPIs will be developed. Metrics have a numerator and a denominator and have a result.

The process for identifying the KPIs and developing the metrics can be quite in-depth. The following outlines the key activities. Identifying the Key Performance Indicators

• Purpose of the dashboard o What is the purpose of the dashboard and how will it be used? o Is it to provide an historical view of the data or real time data? o Will the dashboard be solely for reporting financial or operational performance

information or will items such as marketing, clinical quality, service and performance indicators be included?

o Will the dashboard be used to highlight results from a specific project or implementation?

o How many KPIs should be included? It is important to resist the urge to include every data into a dashboard as doing so can result in information overload.

o What are the trends? Are there lagging indicators?

• Define the Users o Who are the stakeholders? o Based on the end users, the KPIs will vary. Who is the intended audience of the

executive dashboard? It is solely the executive staff, or does it include department managers?

o If you are a multi-facility organization, do the corporate executives desire a different view than the hospital level executives?

• Accessibility and Frequency

o What is the availability and accessibility of the data? o Based on the defined need, what is the frequency of reporting (e.g., daily real time,

month, quarterly)? • Process and Accountability

o What will be the defined process for completing the dashboard? o Will all metrics be automated? o Will there be an audit process? o Will a quality assurance (QA) process occur at key intervals? o Who is accountable for ensuring the metrics are accurate and properly reported?

Page 12: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

12

Developing the Metrics

• Defining Once the KPIs have been determined, each of the proposed metrics must be defined.

o What are the defined numerator and denominator? o What are the inclusions and exclusions? o What are other critical parameters that must be established? o Is historical data required? If so, how much? o How is the metric calculated (develop the actual calculation)? o Consideration of how this definition may vary from other existing calculations? o Validation of the definition with the stakeholders and Data Quality/Decision Support

departments • Availability

o What is the source of the data? o Do the data elements (numerator and denominator) currently exist? o What processes need to change to collect the data?

Evaluating Data Sources From the metrics and KPIs identified in the previous section, several categories are noted: (1) financial; (2) operational; (3) human capital; (4) satisfaction measures; (5) clinical quality; (6) marketing; and (7) service/performance. The data often comes from different sources. Some may originate from clinical systems whereas others are located in administrative systems. Data warehouses could be an additional data source, while individual departments may be manually maintaining databases. It is critical to understand where the data will be coming from and how it is getting into the dashboard as data accessibility will have an impact on how the dashboard is designed. The number of metrics and KPIs to consider, including the compilation of data from a multitude of sources, can be overwhelming. Depending on the core focus, a dashboard can be further grouped into: (1) Strategic, which would be more of an executive level that charts progress against strategic objectives and goals; (2) Tactical, which would be on a departmental level that monitors performance and charts progress against budget and other goals; and (3) Operational, which monitors performance of operational activities and tasks. Once the scope is determined, an organization can identify the key metrics to include. However, many organizations will define their KPIs and associated metrics simultaneously to define their scope as it helps to answer many of the questions. The chart below provides an organization with sample indicators based on the category and audience level. This might be a good format to consider with your organization.

Category and Indicator Multi-Facility Corporate Executive

Hospital Executives

Department Managers

Potential Sources

Financial Inpatient Cost per Discharge Finance Financial Summary to Budget Finance Labor Cost per Visit Finance, Nursing, Ancillary Supply Cost per Visit Finance, Supply Chain Medication Cost per Case Finance, Pharmacy Net Income (and by Facility) Finance Inventory Cost per Adjusted Finance

Page 13: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

13

Category and Indicator Multi-Facility Corporate Executive

Hospital Executives

Department Managers

Potential Sources

Discharge Case Mix Index (CMI) Finance Payer Mix Finance Total Labor Premium Costs Finance, Payroll Overtime Hours (and/or rate) Finance, Payroll Operational Average Length of Stay Clinical Information System Number of Patient Days or Discharges

Decision Support

Patient Wait Times (e.g., Scheduling/ ED)

Clinical Information System, ED System, Scheduling System

Employee Turnover Rate Nursing/Nurse Recruitment, Human Resource

Lost Film Rate Radiology Patient Turnover Rate Operating Room System,

Clinical Information System Nursing Hours per Patient Day Nursing Average Inpatient Daily Census Clinical Information System,

Decision Support Percent of Patients Discharged by Target Time

Clinical Information System, Decision Support

Bed Turnover Rate Clinical Information System, Decision Support

ED Length of Stay – Admitted Patients

Clinical Information System, ED System, Decision Support

Average Acuity – by Unit Nursing, Decision Support Sick Time Utilization Nursing, Payroll, Finance Percent of Contract Hours Nursing Human Capital Time to Fill Positions (e.g., RN) Nursing/Nurse Recruitment,

Human Resource Staff Vacancy Rates and Fill Rates

Nursing/Nurse Recruitment, Human Resource

Skill Mix Nursing Turnover Rates Nursing/Nurse Recruitment,

Human Resource Length of Service Nursing, Human Resource Training and Development Expenditure per FTE

Organizational Development, Nursing Education

Percentage of High Performers with Formal Development Plans

Organizational Development, Nursing

Total Worker’s Compensation Claims per 100 FTEs

Employee Health, Human Resource

Staff Injuries per 100 FTEs Employee Health, Human Resource

Satisfaction Measures Overall Patient Satisfaction Patient Satisfaction Survey Satisfaction with Nursing Care Patient Satisfaction Survey Satisfaction with Physicians Patient Satisfaction Survey

Page 14: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

14

Category and Indicator Multi-Facility Corporate Executive

Hospital Executives

Department Managers

Potential Sources

Physicians Satisfaction Rate Patient Satisfaction Survey Likelihood to Recommend Hospital

Patient Satisfaction Survey

Clinical Quality Adverse Drug Event Nursing, Pharmacy Mortality Rate Clinical Information System Nursing Medication Administration Time

Clinical Information System

Order Change Rate Clinical Information System Never Events Clinical Information System Medication Errors Clinical Information System Medication Reconciliation Clinical Information System Utilization of Two Patient Identifiers

Clinical Information System, Patient Chart Review

Responsiveness to Critical Test Results

Clinical Information System, Critical Results Log

Overall Nosocomial Infection Rate

Infection Control, Nursing

Universal Protocol Compliance Infection Control, Nursing Hand Hygiene Compliance Infection Control, Nursing Unplanned Deaths Due to Healthcare Associated Infection

Infection Control, Nursing

Central Line-Associated Primary Bloodstream (CLAB) Infection Rate

Infection Control, Nursing

Nosocomial MRSA/VRE Infection Rate

Infection Control, Nursing

Nosocomial Urinary Tract Infection (UTI) Rate

Infection Control, Nursing

Ventilator-Associated Pneumonia (VAP) Rate

Infection Control, Nursing

Surgical Site Infection (SSI) Rate Infection Control, Nursing Rapid Medical Response Team Utilization

Nursing, Resuscitation Committee

Failure to Rescue Nursing, Resuscitation Committee

Falls Incidence Rate Nursing Falls with Injury Rate Nursing Falls Risk Assessment Nursing Pressure Ulcer Incidence Nursing Hospital-acquired Pressure Ulcer Prevalence

Nursing

Skin Assessment, Including Braden/Norton Scale

Nursing

Risk Assessment Completed on Admission

Nursing

Admission Assessment Complete within 24 hours

Nursing

Pain Management Nursing Restraint Prevalence Nursing Percent compliance with all AMI Indicators

Clinical Effectiveness

Page 15: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

15

Category and Indicator Multi-Facility Corporate Executive

Hospital Executives

Department Managers

Potential Sources

Percent of Patients Receiving Perfect AMI Care

Clinical Effectiveness

Adult Smoking Cessation Counseling (AMI)

Clinical Effectiveness

Percent Compliance with all CHF Indicators

Clinical Effectiveness

Percent of Patients Receiving Perfect CHF Care

Clinical Effectiveness

Discharge Instructions (CHF) Clinical Effectiveness Adult Smoking Cessation Counseling (CHF)

Clinical Effectiveness

Percent Compliance with All Pneumonia Indicators

Clinical Effectiveness

Percent of Patients Receiving Perfect Pneumonia Care

Clinical Effectiveness

Oxygenation Assessment Respiratory Therapy Pneumococcal Screening and/or Vaccination

Clinical Information System

Adult Smoking Cessation Counseling (Pneumonia)

Clinical Effectiveness

Marketing (Front End Required Data) Referral Information – Site/facility, Payer

ADT/Registration, Physician Referral Service

Disease Specific Outcomes Measures

Clinical Effectiveness

Referral Information by Referring Person

ADT/Registration, Physician Referral Service

Referral Information by Provider/Physician

ADT/Registration, Physician Referral Service

Referral/Patient Source Information - Reason, Referral Source

ADT/Registration

Service/Performance Physician Order Entry Rate Clinical Information System Pre-Registration Rate ADT/Registration Patient, Physician, Staff Satisfaction Rate

Patient Satisfaction Survey, Employee Satisfaction Survey, NDNQI RN Satisfaction Survey

Discharge Processing Time Clinical Information System Process Efficiency Rate (Specific Productivity Reporting)

Clinical Information System, Clinical Effectiveness

** Regulatory measures may include: CMS, Joint Commission, core measures, state agencies. Note: Multi-entity executives would be interested in regulatory compliance results.

Page 16: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

16

Defining the Infrastructure

When determining infrastructure, the organization will need to consider this on two levels. The first level is the technology itself. The organization will need to identify what technology will be used to support the development and use of an executive dashboard. The second level to consider is the data itself. Some data that will be measured may not be automated. This is especially true of clinical data which is in varying stages of automated development within healthcare institutions. If that is the case, then provision will need to be made for manual collection and conversion of this manual data into an automated means. This second level of infrastructure is often overlooked or the effort to support the measurement is underestimated. Technology Infrastructure From a technology perspective, questions to be asked include:

• Should my organization build its own internal data warehouse or should I look to an outside vendor? This question should be coupled with an analysis of the complexity of the dashboard needed, the cost to build, the technical staff available for building purposes, and the organization’s ability to support an internally-developed system.

• If a decision is made to look to an outside vendor, does my current healthcare information system vendor offer a decision support system that provides for an executive dashboard or must I look to other vendors for this technology?

• Will the system being considered or in use at the facility allow for reporting both inpatient and outpatient data?

• Are there other modules available for the system that should be considered to enhance reporting capabilities?

• Will the system being considered or in use at the facility interface with both vendor and non-vendor systems?

• Does the system provide capabilities for the use of comparative data reporting with industry benchmarks?

• Does the system use industry-standard technologies to ease the interfacing of data between systems?

• Does the system provide for flexible scheduling of data interfaces based on internal need? • Is the system flexible enough to provide different views and reports based on the intended

audience? • Does the system provide for both text and graphical reporting? • How intuitive is the user interface? What is the learning curve to become comfortable with

using the system? • Has the organization defined the business retention and legal requirements relating to data

classification and lifecycle?

Page 17: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

17

Data Infrastructure The infrastructure needed to support data collection should begin during the data definition phase of an executive dashboard project. It is at that point that information about the data should be defined. This will lend credibility to the measures and ensure that everyone is interpreting the data in the same manner. As shown in the table below, a matrix/table may want to be considered for documenting this information. This will facilitate both data definition and data infrastructure development. No. Data Data Definition Data

Owner Data

Maintained Method

Collection Sampling

1. Days in Accounts Receivable

An analysis of the aging of accounts receivable using the discharge date of the patient as the base point.

Patient Accounting

Patient Accounting System

Automated 100%

2. Adult Smoking Counseling

Counseling provided to a patient by a nurse on smoking cessation during an inpatient stay.

Nursing Nursing Clinical Documentation System

Automated 100%

3. Reason for No Lipid Lowering Therapy

Reason stated in medical record by medical specialist / primary care doctor for not prescribing lipid lowering therapy.

Department Medicine

Hand-written in Medical Record / Maintained in HIM Department

User-defined field within the medical record abstract will be used to collect indicator. HIM abstractors will collect data during discharge abstracting.

100%

Once information on the data has been defined and a method to collect all data has been identified, then development of the corresponding data infrastructure can take place. Questions to ask when developing this infrastructure include:

• What systems will feed the Decision Support system for executive reporting purposes? • What systems with user-defined fields can be tapped within the organization to collect manual

data? Examples include the Medical Record Abstracting system, the Quality Management system, and the Risk Management System. Are there other systems within the organization that can be used?

• Who will be responsible for collecting this data and at what point in a hospital stay or patient visit should the data be collected? Some data can be collected concurrently. With other data, it may be necessary to wait until a patient is discharged or released.

• What controls will be put into place to ensure the timeliness of data reporting? Should closing dates for a given period be established for all data and not just for financial data?

• What controls will be put in place to ensure the integrity of reported data? Should there be a periodic review of collected data, especially manual data, to ensure its validity?

• How often should information in the Decision Support system be updated in order to make it relevant and useful for decision-making?

• How should information be summarized, aggregated, and reported? Should this information be the same or different based on management position and level within the organization?

• How should data be reported to the medical staff? • How should data be reported to the organization’s governing body? • What data should be shared with all employees?

Page 18: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

18

• Should this data be online or in written reports? Systems are also available today that can read hand-written documentation and collect data without human intervention. This can be very advantageous for hand-written clinical documentation. These systems use probabilistic analysis algorithms shaped by medical domain knowledge which is gleaned from medical experts and external guidelines. This technology has been shown to produce results with a high degree of accuracy and is available today from a limited number of vendors. This type of analysis will become more prevalent in the future.

Page 19: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

19

Transforming Data Dashboard development rests on the creation of thoughtful KPIs that frame how raw data is turned into information for decision-making. This information must then be thoughtfully presented to decision-makers in a way that is both understandable and useful.

Understanding the Raw Data If decision-makers are to trust a dashboard to aid them in their decision-making, the underlying data driving the dashboard must also be trustworthy. If the data populating the dashboard are of low quality, so will the information gleaned from the dashboard. The answers to the questions below must be addressed by those creating the dashboard, but should also be transparent to the decision-makers who rely on the dashboard data to make organizational decisions.

• How often do raw data need to be collected, sampled or merged? • How is data integrity ensured?

o Within each data source o Across data sources

• Are the data complete and valid? o Statistical significance

Turning Raw Data into Information As KPIs for a dashboard are developed, it is essential to remember that KPIs are meant to accurately reflect organizational priorities or values. By asking the following questions, your organization can start to ensure that your organization is “measuring what it intends to measure”:

• What data are needed to accurately represent each KPI? • Is the KPI a single measure or a composite of several measures? • Are there industry standards (or industry disagreements) about how a KPI is measured? • Have we documented what data contribute to, and the measurement approach for, each KPI? • Is the construction of each KPI transparent to decision-makers?

Presenting Information through Dashboard Design While KPIs reflect a piece of an organization’s performance, when grouped together and presented to decision-makers they tell a story about the performance of the organization as a whole. The way this story is presented to decision-makers is likely to affect both how well decision-makers understand the information presented, as well as their resulting choices. The following suggestions and questions will help your organization develop a dashboard that decision-makers find understandable and useful.

• Is the design of the dashboard based on known principles?

o Edward Tufte’s Web site (http://www.edwardtufte.com/) and books give advice on how to display information effectively.

o There are Web sites to help you pick a graph style to meet your needs such as: http://extremepresentation.typepad.com/blog/files/choosing_a_good_chart.pdf

Page 20: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

20

• Are you using a consistent format to present each KPI? • Have you kept the presentation simple and removed unnecessary clutter? • Are decision-makers able to easily navigate between levels of indicators – i.e., aggregated

indicators and sub-indicators? • Is it transparent in the display of information who is responsible for the performance of each

indicator? • Does the display of information allow decision-makers to easily compare current performance

to target levels of performance? • When changes in practice are made to improve an indicator, is it clearly displayed what

changes were made and when they were made so the organization can determine if the changes were successful?

• Does your organization periodically evaluate how decision-makers understand and are using the dashboard through interviews, surveys and/or usability evaluations?

• Does your organization change the design of the dashboard based on these evaluations? Making Decisions through Trend Interpretation Many dashboards use trend lines as a way to represent how KPIs are changing over time. It is challenging for decision-makers to compare trends across KPIs when each is measured using a different scale and positive trends move in different directions. As previously stated, the information in a dashboard tells a story to decision-makers about how the organization is performing. Making these comparisons easily and accurately will aid decision-makers as they form this story of organizational performance.

• How does a decision-maker know which direction a trend should be heading, without them having to think about it? For some trends, such as mortality rates, downward trends are desirable and for others, such as patient satisfaction, upward trends are desirable. Some dashboards display phrases such as “down is good” next to the chart while others show an upward or downward arrow with “down (or up) is good” written inside.

• What value range should the y-axis for a trend line cover? If the y-axis covers a small range of values, changes in the indicator will appear larger. If the y-axis covers a large range of values, changes in the indicator will appear smaller. These choices will likely have an effect on how decision-makers evaluate and respond to changes in trends.

Making Decisions Based on the Story Conveyed through the Dashboard Once the dashboard is created, it is important to test the dashboard with different audiences to ensure that the information is interpreted as intended. Following are some suggestions on how to test your dashboard to ensure that it tells the correct story:

• Document the story you intend for the dashboard to tell. Set up meetings with individuals at various levels (if applicable) in your organization. Provide the dashboard and have them tell you what story they think the information in the dashboard conveys. Compare their story to yours and analyze any differences. Also, document their overall impressions of the dashboard, including what elements work well, what could be improved, what information is missing, etc. The dashboard should stand on its own, with decision-makers able to interpret and use it without background information.

• If the dashboard is in a Web-based or software format, set up usability testing with individuals who will be using the dashboard to understand how well they can navigate through the system and how they interpret the data.

Page 21: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

21

The goal of a dashboard is to have the KPIs presented in such a way that a story is told simply and effectively without much verbiage. The information should be clear, concise and credible, adding value to the user’s decision-making process.

.

Page 22: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

22

Presenting the Data in an Executive Dashboard An effective Business Intelligence (BI) platform should present actionable information in alignment with the organization structure of the hospital. Depending on their role and area of responsibility, each decision-maker should be presented with a personalized view of the KPIs most relevant to his/her role, as well as an enterprise perspective to clearly show how these granular metrics fit into the overall company strategy. Furthermore, presenting data to non-technical business and clinical users requires intuitive data visualization and graphical presentation of information. Most BI applications accomplish this through a combination of the following presentation layers:

• Enterprise scorecards • Role-based dashboards • Dynamic report tables • Proactive alerts and targets

Scorecards, dashboards, and dynamic reports built on a consolidated source of data are the best mechanisms to ensure that everyone in the hospital is focused on the same targets and looking at the latest numbers. These types of capabilities ensure there are no issues with multiple versions and manual maintenance as with spreadsheets. When outliers or areas of interest are identified, it is important to be able to drill down to the underlying detail to perform root-cause analysis. Scorecards, dashboards, dynamic reports, and proactive alerts work together to:

• Proactively identify the existence of a problem (scorecards) • Understand the magnitude and historical context (dashboards) • Determine the root cause of the problem (dynamic reports) • Set alerts, targets, and reminders to be notified of changes

Enterprise Scorecards – Manage performance by measuring actual results toward organizational goals and targets. Scorecards keep different areas of the hospital focused on a common plan by monitoring real world execution and mapping the results back to specific goals and targets. Good scorecards use a simple color or graphical system to provide quick insight for the user. The GPS system in your car, for example, shows progress towards your destination. Keywords used in thinking of scorecards: manage, align, strategic, and balance. Role-Based Dashboards – Monitor performance by displaying specific graphs for KPIs for specific areas of hospital operations. Dashboards fall one level down from scorecards in the business decision-making process and focus more on monitoring operational results. Typically the dashboard is tied to operational goals and is less strategic than a scorecard. The data reflected in dashboards are usually timely (daily, weekly, monthly) in nature and are depicted in a graphical format. Think of your car’s dashboard. The speedometer tells how fast you are going right now but it does not tell you about progress towards your destination. Keywords used in thinking of dashboards: measure,

Page 23: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

23

understand, tactical, and trend.

Dynamic Reports – Display data in flexible tables that users pivot and filter to answer whatever questions they have while determining root causes. Report views can be quickly changed by a non-technical business user without running a new query. Dynamic reports are different from static or canned reports – they allow the user to “interrogate” the data, to change dimensions, and modify the hierarchy of data on-the-fly. If a user sees a trend or potential outliers, he/she can quickly look at the data in multiple different ways to better understand the root causes. The best scorecards and

dashboards drill into dynamic reports. Keywords used in thinking of dynamic reports: analyze, discover, root cause, investigate, and dimensions. Alerts, Targets, and Reminders - Proactive notification is integral to performance management. Your scorecards, dashboards, and dynamic reports should be able to alert you on progress (or lack of progress) to targets whether or not you are looking at the data. Proactive alerts and reminders can be set to automatically push updates via email based on business rules and thresholds set by the decision-makers. For example, when a very specific financial metric such as unbilled Medicare receivables goes beyond a specific target value of $500,000, then the revenue cycle director responsible for this metric would receive an email automatically to notify him/her of this problem. An effective performance management application would have hundreds of these alerts set on key metrics, ensuring that nothing slips through the cracks. The graphic below illustrates how scorecards, dashboards, reports, and alerts can work together to present data and facilitate action. Through a strategic enterprise scorecard, a hospital executive notices that a key organizational metrics is “Red” indicating a performance problem (Step 1). Using the associated dashboard, a department manager reviews underlying operational metrics to understand the severity, trend and impacted areas (Step 2). An analyst then performs detailed root-cause analysis to determine what change or correction in the process is required (Step 3). With this correction in place, positive change in performance is monitored by the senior executive, department manager, and analysts using a proactive alert to notify everyone immediately if the problem reoccurs (Step 4).

Page 24: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

24

Environmental/Cultural Factors in Your Organization The most difficult part most organizations face in implementing Business Intelligence is not in defining the technological, data/KPIs, or presentation requirements, but rather is pushing back against the tide of political resistance and resiliency of those who fear change. Business Intelligence has the promise and expectation of improved processes and decision making throughout the organization, but in order for it to be successful, people must embrace new tools and change how they do their jobs. So, you thought you were a technologist? This section of the guide discusses how you must also be a promoter, marketer, sales person and political strategist; all in one. In prior sections of this guide, reference has been made regarding the need to:

• Define the users • Determine metric accessibility and frequency • Develop process and accountability

Within each bullet point above, the focus must be on the WHO rather than the WHAT when considering the environmental impact of Business Intelligence in your organization. When dealing directly with the people whose buy-in determines success or failure of a Business Intelligence project, there are some simple guides and personal behaviors you can follow to reduce resistance and political push-back.

• Get business people and clinical leaders involved early. o Focus on the needs of the people before you talk about the technical requirements. o Develop a user group consisting of department heads and clinicians at the beginning of

your Business Intelligence project. Purposely include those who you know are most resistant to change but have

the greatest to gain. The user group should include physician and nursing leadership, even if these

clinicians are not employees of the hospital or organization. o Make everyone feel they have a stake in Business Intelligence success and its

potential failure. o Don’t overlook the opportunity to use dashboards to facilitate working discussions,

collaborative research and planning with business and clinical leaders. Use the tool in a non-threatening and engaging manner with those who might be intimated or reluctant to move away from their legacy spreadsheets.

• Listen to users concerns and apprehensions. o Learn how “things” are done today and never dismiss processes nor label them as

“inefficient” or “archaic.” o Do not assume nor promote Business Intelligence as the answer to everything.

• There is no room for ego.

o Technologists have a reputation for assuming they are smarter than everyone else and are often perceived as arrogant. Watch how you present Business Intelligence and take great care in how you interact with your non-technical peers.

o Learn to take criticism even if it is not warranted. o Never become defensive and learn to admit fault when Business Intelligence proves to

not add value; day one.

Page 25: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

25

• Do not expect change to just happen. o Learn what works today and do not try to change things for the sake of change. o If departmental employees are resistant to change, take the time to show them how

new tools and ways of doing things will not threaten their jobs, but rather will improve the quality of their work-lives.

• Focus on small wins and built off on them.

o Let business and clinical employees share or take credit when progress is made and results are produced.

o Use the enthusiasm of a few happy “customers” to drive acceptance in other areas of the organization.

• Always make “it” about the patient.

o Constantly reinforce the reason why you are bothering with Business Intelligence is because of the desire to improve patient care services.

• Obtain senior management and physician/nurse buy-in and backing.

o When all else fails, employees will do what is being asked by senior managers.

With so many demands being placed on our healthcare system these days, organizations and their employees need to embrace new thoughts and ideas if they are going to survive. Business Intelligence has the promise of assisting that success. However, with all this promise comes the fear of something new. To be successful in your Business Intelligence endeavor, you need to make sure this project is more about people and processes and less about technology and tools. It is critical to make Business Intelligence an integral part of an organization’s strategic planning and decision-making process. IT and business leaders must demonstrate value to all levels of the organization.

Page 26: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

26

ROI/Organization Benefits

Because the benefits of implementing an executive dashboard are varied and not necessarily quantifiable, many organizations struggle with defining the return on investment (ROI) of pursuing this endeavor. Traditionally, ROI is defined as the total benefit derived from a project divided by the total cost of deployment. Although total cost to implement is more readily available and quantifiable, the benefit is not making the ROI an elusive metric for many institutions. Total cost of implementation can be defined to include all time spent by internal and external resources to plan, develop, and implement the tool. Specifically, the following may be considered:

• Data harvest, scrubbing and preparation time • Software purchase or development • System maintenance costs • Hardware • Initial and ongoing training

The benefits of executive dashboards may be broken into both hard and soft measures. Normally, a hard measure would include tangible, quantifiable items such as time reductions and manual process step elimination. Specific items to consider are:

• Reduced time to consolidate disparate data reports • Time saved preparing for regulatory reporting • Process improvements through elimination of duplicate processing of the same data set • Improved productivity as a result of having improved access to data • Cost savings associated with elimination of report printing and distribution

On the other hand, soft measures are qualitative advantages experienced by all users that provide abilities and business insights that were not previously available. Some more benefits to take into account are:

• Ease of monitoring change • Ability to assess more complex interrelated processes • Improved visibility and accountability • Ability to align operations with strategic goals and objectives • Improved decision-making ability through increased flexibility and potential real-time updates

to adjust to what is happening in operations • Benchmark against industry standards, and regulatory requirements • Discover previously unknown or unnoticed relationships between various indicators • Empowered front-line to take action on the data and move forward with changes/improvement

efforts • User satisfaction with the tool and access to information • Elimination of multiple report development when the scorecard is customizable by each user

on their desktop

Finally, when attempting to calculate the ROI of an executive dashboard it is helpful to estimate the cost of not having or using the tool. For example, an organization may experience a competitive

Page 27: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

27

disadvantage because it cannot respond as easily or quickly as necessary to support the organization’s goals and objectives.

Page 28: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

28

Lessons Learned • Source Data: For an executive dashboard to be truly effective, careful thought should go into the

structure and quality of the source data that will feed the data warehouse. In some instances a business intelligence solution could be setup to connect to data stores operationally, helping skip or prolong the need for a formal data warehouse. The time spent in deciding on whatever data model you use needs to be well thought out, as it is like pouring the foundation for a house. The output will only be as good as the input provided.

• Policies: Organizations should establish policy and ensure resources that will allow for the timely

entry of data into the various systems used. Monthly close-out dates should be established for both the clinical and financial areas and adhered to by participating parties.

• Define the Purpose: Much like the data model that should be considered carefully, defining

requirements in a project like this is extremely important. An organization should spend adequate time defining the intentions of the executive dashboard. Many organizations jump right into a dashboard without the adequate preparation time, resulting in a great deal of re-work.

• Start Small and Iterate: A common pitfall for business intelligence implementations is attempting

to deliver all things to all people at once. Start with three to five meaningful metrics and fully integrate them into the process and feedback loop of your hospital. Another way to manage scale is to start with one region, facility, or department and use the knowledge gained in the process when building the enterprise scorecard. With an iterative process, you can set your organization up for long-term success by building positive momentum and better understanding the significance of key performance metrics. This approach borrows from leading edge technology practices general referred to as “agile software development” first introduced in the late 1990s. (For more information see http://en.wikipedia.org/wiki/Agile_software_development.) An “agile” methodology generally promotes rapid implementation and multiple iterations of software, rather than long-term planning and large-scale development prior to roll out and testing. This model is more easily adapted to changing requirements and user feedback. This approach can be very helpful when considering a BI strategy or building an enterprise scorecard. Rather than spending two years designing the perfect metrics in a vacuum, start with just a few metrics in one department and get them into production immediately. This will create an iterative feedback loop that is essential to both determining the best possible metrics for the organization, but also ensuring business user acceptance along the way.

• Leverage Industry Best Practices: Fortunately, there is no need to “re-invent the wheel” when selecting best-practice metrics. Various government agencies, clinical quality initiatives, benchmarking organizations, industry associations, and your peer hospitals have all published metrics that have set the standards for how hospitals measure performance. 1. Industry associations and publications regularly publish articles and books on best practice

metrics:

Healthcare Financial Management Association - HFMA http://www.hfma.org/NR/rdonlyres/B2ED396F-26B1-4863-8120-018F6FF56E54/0/400625.pdf

Page 29: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

29

American College of Healthcare Executives (ACHE) http://www.ache.org/pubs/redesign/productcatalog.cfm?pc=WWW1-2039

2. Industry thought leaders: Many individuals and companies with deep healthcare domain

expertise have also developed templates for hospital performance metrics:

Cleverly and Associates: http://www.cleverleyassociates.com/Library/MetricsThatWork.pdf Mede Analytics: https://www.medefinance.com/healthcare-analytics-resources/documents/2008-10-Mede-Provider-White-Paper-Scorecard.pdf

3. Government organizations: Various government agencies and affiliates have established

guidelines and requirements for measuring clinical quality and patient safety.

The Joint Commission http://www.jointcommission.org/NR/exeres/5A8BFA1C-B844-4A9A-86B2-F16DBE0E20C7.htm Agency for Healthcare Research and Quality (AHRQ) http://www.ahrq.gov/qual/measurix.htm

• Consider Software as a Service: Many hospitals have started using a hosted, outsourced model to manage their business intelligence infrastructure. Gartner defines software as a service (SaaS) as follows:

SaaS is software owned, delivered and managed remotely by one or more vendors. If the vendor requires user organizations to install software on-premises using their infrastructures, then the application is not SaaS. SaaS delivery requires a vendor to provide remote, outsourced access to the application, as well as maintenance and upgrade services for it. The infrastructure and IT operations supporting the applications must also be outsourced to the vendor or another provider.

• Gartner Hype Cycle for Software As A Service, 2008

Solutions delivered in a SaaS model, which is a proven technology in other industries, are now getting traction in healthcare. A hosted BI platform incorporates a Web-based analytics portal delivered in a vendor-hosted environment that requires no hardware, no software, and little to no IT resources. By leveraging an outsourced, Web analytics application hospitals can have a robust performance dashboard up and running in less than 90 days. SaaS solutions are typically delivered as a subscription fee, without a large up-front capital investment.

Page 30: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

30

References

Articles and Publications Shutt JA. Balancing the health care scorecard. Managed Care. September, 2003. Pieper SK. How to strategically manage with scorecards. Healthcare Executive. May/June, 2005. Johnson M. Steps Toward an automated scorecard system. Healthcare Financial Management. February, 2006. Cleverley WO, Cleverley JO. Scorecards and dashboards: Using financial metrics to improve performance. Healthcare Financial Management. July, 2005. Kaplan RS, Norton DP. The balanced scorecard: Measures that drive performance. Harvard Business Review. July/Aug, 2005. Brown AM. Measuring a healthy hospital: Metrics-based tools for improving operational performance. HCT Project. Volume 2; July, 2004. Kirby S. Start small and build toward business intelligence. Healthcare Financial Management. January, 2009. Hammer D. UPMC's metric-driven revenue cycle. Healthcare Financial Management. September, 2007. Books Sower V. Benchmarking for Hospitals: Achieving Best-in-Class Performance without Having to Reinvent the Wheel. Milwaukee:American Society for Quality; 2007. Berger S. The Power of Clinical and Financial Metrics: Achieving Success in Your Hospital. Chicago: Health Administration Press; 2005 Kaplan RS, Norton DP. The Balanced Scorecard: Translating Strategy into Action. Cambridge:Harvard Business School Press; 1996. Howson C. Successful Business Intelligence: Making BI a Killer App. New York: McGraw Hill Osborne Media; 2007.

Page 31: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

31

Reference Web Sites BeyeNETWORK: http://www.b-eye-network.com/channels/1021 BeyeSearch: http://www.beyesearch.com/directory/Industry_Business_Intelligence_Solutions/Healthcare/ B_-BESTPRACTICES.COM: http://www.bi-bestpractices.com/ BIScorecard: http://www.biscorecard.com/index.asp Six Sigma Healthcare: http://www.healthcare.isixsigma.com/ Health Industry Insights: http://www.healthindustry-insights.com/HII/home.jsp SerchDataManagement.com: http://searchdatamanagement.techtarget.com/

Page 32: Executive Dashboard Development Guide - Amazon S3s3.amazonaws.com/.../Content/files/ExecutiveDashboardDevGuide.pdf · Executive Dashboard Development Guide . ... clinical department.

© 2009 Healthcare Information and Management Systems Society (HIMSS)

32

Contributing Authors

The HIMSS Management Engineering-Performance Improvement (ME-PI) Community formed a “hot topic” group focused on executive dashboards. As its deliverable, the group opted to create a development guide for executive dashboards. Through our experience, we found that many organizations consider only portions of developing and implementing an executive dashboard, but rarely the compilation of the topics you will find in this guide. The Executive Dashboard Development Guide has been developed with the expertise of more than ten individuals. These individuals have a wide array of experience – from provider organization to consulting. We will continue to enhance our guide and share with members of HIMSS, healthcare organizations, and others who may have an interest. The contributing authors include:

Name Credentials Company Cecilia Backman MBA, RHIA, CPHQ Manuel C. Co Jr. MSN, MS, RN, CPHIMS NYU Langone Medical Center Matt Dobski MedeAnalytics Karen Fairchild Fairchild Consulting, Inc. Chuck Henck Cynthia McKinney MBA, FHIMSS IBM Jenna Maquard PhD University of Massachusetts Brenda Mollohan MHA, JD Jesse Purdue Information Builders Jonathan Rothman MBA Emergency Medical Associates of NJ/NY Henry Tenarvitz Versus Technology, Inc. Brooke Wessman MSEM Henry Ford Health System