Learn more about Mike Doyle

Author Bio

Mike Doyle

Mike Doyle joined Health Catalyst in May of 2013 as a Vice President. He has been connected with the Health Catalyst senior leadership team since 2006. Prior to Health Catalyst, Mike led the Enterprise Data Warehouse (EDW) program at Allina Health as Director of Healthcare Intelligence. He helped Allina grow its EDW program from a nascent clinical improvement initiative to an enterprise-wide strategic asset, in heavy demand by thousands of users across all of Allina’s 11 hospitals and 100+ clinics. Prior to his work with Allina, Mike was employed on the Northwestern Medicine campus in Chicago, beginning as a Systems Administrator at the Medical School and eventually leading the Analytics and Systems Integration team at Northwestern Medical Faculty Foundation. In addition to his experience building strong technology teams, Mike has experience in technical roles such as database administrator, web programmer, data architect, and business intelligence developer. Mike holds a Master of Music degree from Northwestern University and a Bachelor of Fine Arts degree from Carnegie Mellon University.

Read articles by Mike Doyle


Mike Doyle

How to Avoid the 8 Most Common Pain Points in Becoming a Data-Driven Healthcare Organization

What does it mean to be a data-driven organization? What are the advantages of achieving this status? How can my organization get there? These are all great questions, but they are outnumbered by the eight common pain points every healthcare system encounters along the way:

Conflicting versions of the truth
Lacking a culture that supports data transparency
A lack of trust in the data
Data volume and overload
Struggling to find an effective system
“We already tried that”
A dearth of data-savvy staff
Lack of executive sponsorship

Anticipating and absorbing those pain points is part of the secret to success. This article explores them further, as well as the advantages of letting data guide critical healthcare business decisions.

Mike Doyle

The Value of Claims Versus EHR Data in Care Management and Population Health Analytics Strategies

If EHR data have the breadth and depth of a pond, then claims data are just lily pads on the surface. In other words, the volume of EHR data is far more substantial than that of claims data. Population health strategists should adopt the AND-BOTH approach, rather than the EITHER-OR, when it comes to working with both types of data. This best-of-both-worlds tactic offers data that is standardized, accessible, discrete, AND real-time, detailed, and physician based. Given what some might view as an overwhelming volume of data when working with both, it’s wise to develop a data governance framework supported by some type of flexible data management platform, such as an enterprise data warehouse (EDW).

Mike Doyle

Early- or Late-binding Approaches to Healthcare Data Warehousing: Which Is Better for You?

Most industries use enterprise data warehouses (EDWs) to create meaningful analytics on their operations and processes. Healthcare has long struggled with implementing and maintaining EDWs. One reason for this is that a lot of the data healthcare uses is unstructured, meaning there are few to no restrictions on it. And this unstructured data can exist in several systems within the organization. Additionally, health systems must pull data from many sources, such as EMRs, financial systems, and patient satisfaction data. The early-binding approach to data warehousing makes the binding decisions early in the process and, thus, lacks the agility healthcare needs to respond to ever-changing business rules and requirements. This approach can also take a long time to implement. Late-binding data warehousing has a much faster time-to-value and allows users to create analytics based on what-if scenarios. Plus, it can change to reflect the always-moving world of healthcare analytics needs.

Mike Doyle

3 Phases of Healthcare Data Governance in Analytics

Healthcare data governance is a broad topic and covers more than data stewardship, storage, and technical roles and responsibilities. And it’s not easy to implement. It’s necessary, though, for health systems that are entering the world of analytics because the governance structure will enable the organizations to drive higher-quality, low cost care. In order for healthcare data governance to be most effective however, it needs to be adaptive because real healthcare data governance is much more fluid than any plan laid out on paper. Typically there are three phases that characterize successful analytics implementations: the early stage, the mid-term stage, and the steady state. As health systems begin to determine the effectiveness of their data governance strategy, it’s important to look at key metrics from their analytics implementations that will either trend up, remain solid, or trend down.

Mike Doyle

Clinical Data Warehouse: Why You Really Need One

From time to time, people question whether they really need a clinical data warehouse. The first wave of data warehousing projects didn’t deliver much value and healthcare CIOs had plenty of tasks to accomplish in the meantime. But today’s technology, data, and regulations scream for an analytics solution and a clinical data warehouse can offer this. A hospital should consider reporting requirements and technical requirements. Finally, when an organization has access to its data, leaders are empowered to make informed decisions.

Mike Doyle

Build vs. Buy a Healthcare Enterprise Data Warehouse: Which is Best for You?

One of the most common questions we hear is some flavor of  “should we build or buy a data warehouse?”   This is a very understandable question.  And there are multiple reasons  for either building your own data warehouse or buying it.  We acknowledge that.  And even though we know that we naturally advocate the advantages of buying a data warehouse, we also know that buying a data warehouse is not for everyone, and we understand the circumstances in which health systems will choose to build their own data warehouse.  In this article, we do our best to collect the pros and cons of each scenario, and even suggest a hybrid solution, so that you can better assess your situation, understand your options and make the best decision for you.

Mike Doyle

Data Warehousing in Healthcare: Is It Necessary?

Wondering if you really need a healthcare data warehouse or if the technology you already possess if enough? Data warehousing enables healthcare analytics. It will help fulfill reporting requirements so your analysts can concentrate on analyzing data. It can offer near real-time answers to many questions, whether financial, clinical, or technical. Eventually, you’ll wonder “How did I ge tby so long without it?”

Mike Doyle

5 Myths You Won't Need to Worry About When Adopting a Clinical Data Warehouse

If you’ve been thinking about implementing a clinical enterprise data warehouse (EDW), chances are you have a few questions about the possible problems you’ll encounter. In this Insight, Mike addresses some fears that are actually common myths including: 5) I can’t provide broad access to my EDW, 4) users don’t need or want the ability to write SQL queries, 3) I don’t need an EDW — my BI tool does everything I need, 2) EDWs are too expensive, and 1) EDWs take too long to complete.

Mike Doyle

6 Surprising Benefits of Healthcare Data Warehouses: Getting More Than You Expected

Recently, I invited a group of my colleagues to share some examples of unexpected benefits they had witnessed at healthcare organizations that feature powerful, thriving EDW initiatives. The number of responses I received was overwhelming; more than I could possibly hope to include in one blog post. With a goal of hopefully sharing all of them within a continuing series, here are some excerpts, reprinted with permission and in the words of the “EDW Elders” within our company. These include 1) negotiating with insurance companies, 2) Stage 1 Meaningful Use self-certification, 3) data quality issues, 4) financial data comparisons, 5) EMR user log data, and 6) employee satisfaction data.