In the Field with our Experts

Agile Analytics: Put Your SaaS in the Sand

By Rob Paller, Consultant

Recently at a client, the data warehouse administrator was asked to define a sandbox environment in the production data warehouse for  analysts and developers working on a small project. The idea behind this sandbox was to allow the team a working area for collaboration and intermediate storage of results while working with the data in a purely ad hoc capacity. Instantly it was recognized this could be the start of something bigger within the organization—something that could not currently be provided by the incumbent business intelligence tools. The response had to be formulated quickly in order to avoid stifling the creativity of the analysts—or worse, the progress of the project—but care had to be taken as well; if managed incorrectly it could get out of hand and become a waste of system resources and a drain on human resources that had already been spread thin.  The business unit in question is looking to move from the confines the current business intelligence environment and push the edges.

This was a group of analysts that wanted to get their hands dirty and weren’t afraid to fail. They wanted to mash data together that previously could not be done by the business intelligence tools in their controlled ad hoc environments. This was data mining for the next set of KPIs that would shape the way business moves forward.

The concept of agile analytics is not new, eBay presented on and blogged about this concept in 2008. The idea at this client was simple. By leveraging the existing enterprise data warehouse system to house their sandbox environment the duplication of data is all but eliminated. Groups interested in sharing data between their sandbox environments are strongly discouraged until the data has been properly integrated into the production environment. The sandbox environments would also be given a short life expectancy at their inception to prevent the prototypes from becoming production and data ending up in a wasteland. This all sounded great on paper.

In the midst of a development architecture overview, a brief conversation among a few enterprise architects uncovered the potential Screw-Me Scenario that could bring the concept of agile analytics to an untimely demise. “The users of the data warehouse are not permitted to write ad hoc queries outside of a controlled business intelligence tool. They might write a bad query.” Thanks for the warning, we’ll be sure to refine our pitch to the enterprise architects to diffuse this scenario before it turns ugly.

In Oliver Ratzesberger’s presentation for eBay’s Analytics as a Service, he acknowledges that the metrics we already know are cheap and the unknown metrics are expensive. But the known metrics are not pushing the edges. Known metrics are found in the middle of the box. Agile analytics is about pushing the edges about how your enterprise data warehouse is used to improve response to the needs of the business. It is about the evolution of the user community from one who plays in controlled ad hoc environments to encouraging them to experiment with new ideas and not to fear failing along the way. Agile analytics is about encouraging your users reach out for the edges and P U S H. Only once the edges are stretched can the middle of the box redefined.

photo by edenpictures via Flickr (Creative Commons License)

---

Rob Paller is an expert at business analytics and database administration. Since joining Baseline, Rob has been responsible for developing a case analysis system to streamline the oversight of food assistance benefits, implementing a common citizen data model, and assisting in the rollout of a new public assistance data model integrating data from over 10 years of legacy with a new benefit eligibility determination system.

The Role of Logical Data Models

By Caryn Maresic, Senior Consultant

Logical Data Models (LDMs) were the standard means of recording business rules and data definitions back in the ‘80s and ‘90s.  Business and IT partnered to learn the art of data modeling and 3rd normal form in hopes of finding a common ground to record requirements.  Over time, it became evident that in today’s fast-paced development environment the business doesn’t have the time to digest all the nuances in a data model.  It is IT’s job to gather information via interviews.  Many times, we skip the LDM task because we don’t have time and LDMs are too academic and we find ourselves asking…

Are Logical Data Models really necessary?  Yes!

LDMs drive BI design by defining business rules independent of physical implementation.  There are many tools (CA ERwin Data Modeling, Embarcadero Technologies), and techniques (UML, IDEF, Bachman) in use today. To illustrate the difference between the benefits of a PDM and an LDM, let’s look at an example:  Customer/Address relationships.

The PDM above (created using) supports many Customer/Address business rules, but it does not tell me the Customer/Address relationships as defined by the business.

In contrast, the following LDM is more rigorous and definitive:

What does the LDM above tell me that the PDM did not? 

• There are two types of Addresses – Physical and Mailing.
• The Customer Bill To Address can be either a Physical or Mailing Address
• Customer Ship To Addresses must be Physical Addresses.
• Every customer must have a Bill To Address.
• A Customer has zero, one or many Ship To Addresses.

All of these relationships can be implemented using the PDM, but they aren’t dictated by the PDM.  The LDM explicitly defines each relationship and prompts us to ask better questions of the business, such as:

• Are there any Customers who do not have a Bill To Address?
• Are there any Customers who do not have a Ship To Address?
• When there are multiple Ship To Addresses, how do you know which one to use?
• Are there any Customers who have more than one Bill To Address?
• Do you know of any upcoming business initiatives that might change these rules?
• Was there ever a time when the business rules were different?

The answers to these questions may require changes to the model, but without the LDM, we may not have asked the right questions!  The best part about the LDM is how valuable it is to the rest of the development process.  It will be used in production of training materials, test plans and test cases, data profiling and migration efforts and, finally, physical database design.

---

Caryn has over 20 years experience in providing high-quality data solutions to clients in the areas of Business Intelligence, Data Warehousing and System Integration.  Caryn has expertise in across industries with an emphasis in Pharmaceutical, Manufacturing, and Insurance.  Prior to joining to Baseline, she ran her own consulting company.

To Infinity and Beyond! — Healthcare in the Cloud (Part 3)

By Carol Newcomb, Senior Consultant

This is the final installment in a 3-part blog series, discussing the opportunities that cloud computing offers in healthcare. I present futuristic scenarios from each healthcare contingent’s vantage point:  patients, providers and payers.   A myriad of technologies exist today.  It will be up to healthcare organizations of all types to get their data ready to meet the demands for data integration, security, portability, transparency and accountability in this brave new world.  Mature data governance systems and enterprise-wide data integration will be critical in this endeavor.

III: Payers

I’ve worked in the insurance industry for 30 years now, and I’ve never seen anything like this!  For years, we had consultants running around here, designing portals, databases, dashboards and training us on new software to use in running analyses.  But half the time, we just went back to using the antiquated systems that most of our data servers were designed to use.  In the last couple of years, we have completely stopped talking about those different hardware platforms.  Now, the Risk Management division, Actuarial, Medical, Quality, Marketing and Claims Processing all use one software package; all the data management (since we get millions of transactions and claims each day) is handled offsite.  It’s never been this simple!  We still have our Data Governance Committee that works with each department to flush out any issues, but things have really improved.Since all the data is now integrated on what they call “The Cloud”, we get daily dashboard updates on our desktops, showing use trends, costs, disease prevalence around our region, geographic mappings of where the large employers have families, results of marketing activities, and new health and wellness information that is also shared with our subscribers.  We had a period of some pretty large layoffs, since our entire IT department is now outsourced, and all those consultants have disappeared.

Our Fraud and Risk Management analysts used to have hundreds of spreadsheets on their desktop computers, which they would constantly be updating and piecing together to look for patterns or gaps.  Now they can use more sophisticated statistical detection algorithms because the data is updated and fed to them each night.  The simplest algorithms are actually run offsite, and those results are delivered to them daily.  From those, they can then drill into services or providers that look suspect.  They can now turn around reports in about 2 days, where it used to take them 2 months!

Our business model has shifted from using claims history to cut anticipated high-cost cases, to using the data from those same claims to design health and wellness programs stressing prevention and care management.  If we see pockets of infectious diseases in one region, for example, we do further analysis on the saturation and specialty mix of our network providers, the employer group mix, and the educational factors that may be contributing to higher medical costs.  We then work with the schools, employers and practitioners to combat the spread of that particular disease.  We have been able to avert hundreds of hospital admissions through lower cost prevention measures, which we track through our “Population Health” Program.  Across our company, there is a competition to target and drive down claims rates, and each group gets an annual part of their bonus based on their results.  It’s pretty interesting.

Even though we don’t get as much data through claims-processing as the local hospitals get in their clinical systems, we participate in national research studies of medical effectiveness.  Some procedures that we used to consider experimental, we now collect data on and enlist our Actuarial Department to help with statistical analysis and cost-accounting.   Our Medical Quality Department works with other government-sponsored research groups to compare the results of our findings, and often they lead to some pretty surprising conclusions.  Where we thought we were saving money in the past, we were actually driving up utilization in higher cost medical facilities, but now we’re helping encourage subscribers to get medical attention sooner, which prevents some pretty catastrophic claims.  We couldn’t do this before because our Utilization Review desk would deny those services that we now know help reduce lifetime costs.

Probably the best benefit we’ve gained from using Cloud computing services is that our claims processing is now practically effortless.  Twenty-five percent of our cost of business was dedicated to claims handling, denials, reviews, exceptions, and remediation.  Not only have we saved money, we have reduced premiums to our subscribers, and the healthcare providers have reduced the cost of their back-office operations, which used to handle all the churn of rejected and resubmitted claims.  They now get paid faster and we save money.  How’s that for amazing! We’ve also noticed another side benefit: our subscriber turnover rate has dropped by 10%!  Customer satisfaction rates have never been higher!

---

Integration of claims data with population-based data and actuarial model results has significant business impact.  Clearly, departmental governance representation was required to mesh all the different data types.  Several large data warehousing systems can be housed and integrated using Cloud technology, but the rules in how to match and align different data collected for entirely different purposes is the key to analytic power in this example.  As companies offload daily transaction processing, which can be automated and scaled, business dollars can be better deployed to more strategic purposes, and those resources then achieve more with the data at hand.

photo by Simone Ravella via Flickr (Creative Commons license)

---

Carol Newcomb is a Senior Consultant with Baseline Consulting. She specializes in developing BI and data governance programs to drive competitive advantage and fact-based decision making. Carol has consulted for a variety of health care organizations, including Rush Health Associates, Kaiser Permanente, OSF Healthcare, the Blue Cross Blue Shield Association and more. While working at the Joint Commission and Northwestern Memorial Hospital, she designed and conducted scientific research projects and contributed to statistical analyses.

To Infinity and Beyond! — Healthcare in the Cloud (Part 2)

By Carol Newcomb, Senior Consultant

This is the second installment in a 3-part blog series, discussing the opportunities that cloud computing offers in healthcare. I present futuristic scenarios from each healthcare contingent’s vantage point:  patients, providers and payers.   A myriad of technologies exist today.  It will be up to healthcare organizations of all types to get their data ready to meet the demands for data integration, security, portability, transparency and accountability in this brave new world.  Mature data governance systems and enterprise-wide data integration will be critical in this endeavor.

II:  Providers

It’s 11:30pm on a Sunday night, and I’m working a double shift in the Emergency Room.  It’s been a little slow tonight, so I’ve been online researching some reasons why we’ve been seeing more Somali children here recently.  The hospital keeps the latest demographic figures on the intranet.  Turns out that at first just a few families settled here, but more relatives are joining them and many of the children are orphaned.  When they came through immigration, though, they were given medical checkups and those records are available for each of the kids I have treated recently, since they were each given a universal medical record number.  I can see the results of any tests they had, and whether they were given any medications or treatment.  There’s also a public health database through CDCP that tracks the incidence of infectious diseases, and some primary prevention steps we should prescribe at the time we treat these kids.  It’s very helpful stuff.  I’m going to share what I learn through the local Health Information Exchange (HIE) with the social workers in the neighborhood clinics.

Whoa!  Suddenly things are picking up!  There’s a little old lady with severe dementia who fell and knocked out her tooth, and a teenager who got into a brawl with a black eye.  Where are his parents!?  First I try talking to the lady, but she’s confused.  At least she is wearing an ID tag, so I look up her name (Dorothy Xu) in the HIE portal.  She’s never been to this hospital, but I see that she has spent some time in a local nursing home, and was recently discharged to the care of Dr. Jones.  We start a new hospital record for her, instantly downloading the data from the nursing home and Dr. Jones’ office.  I also see that her daughter lives nearby and that her husband died less than a year ago, so I get her daughter’s telephone number and give her a call.  I’m surprised that the ambulance techs didn’t do this already, but they were making sure she didn’t have any other injuries.  When her daughter arrives, she has brought a list of all of Dorothy’s medications, and one of her favorite nightgowns.  I send an email requesting an urgent appointment with one of the dentists in her insurance network first thing in the morning, and then make sure she has a prescription for some pain medications to get her through the night.  Poor thing.  Also, I talk to her daughter about whether Dorothy has seemed depressed since her husband’s death.  I give her daughter a printout of some good psychiatrists in in the local area, and then she takes her Mom home.

So what about this black eye?  Jason is 15 and it looks like he already has a police record.  He refuses to talk to me, so we get his fingerprint and match it to his medical record.  He doesn’t want the police to know he was in another fight, but I’m going to have to call his parents to pick him up.  He seems concerned, but refuses to say anything more.  I scan through his chart and his Mom’s, and I see that she has also been treated for a fractured wrist and a black eye about 4 weeks ago.  Either this family is really clumsy, or something is going on here!  Apparently, a social worker at school also noted a few things in his record, and the pieces are beginning to suggest some family abuse problems.

When his Mom arrives, we talk privately for a few minutes.  I tell her I am concerned, but don’t really have very much information.   As we talk, she becomes visibly upset.  Eventually, she admits that her husband, who was recently laid off, has been taking his anger out first at her son, and also at her.  I ask if she needs any other assistance or resources, but she refuses.  So I treat her son’s eye and send them on their way.  I also make note in the husband’s record that he may have some history of abusive behavior, and I notice that he was recently in the hospital as a result of a car wreck.  At least this information will be available to the nurses in case they are treated here again.

---

Offloading disparate datasets to integrated Cloud servicers that will also perform basic analytics can greatly empower healthcare providers with ready access to data.  Remotely hosted Health Information Exchanges will empower social workers, schools and hospitals, even law enforcement authorities, to work together in assisting their local communities.   This is the profile of the new healthcare provider as a result of Cloud technology.

photo by dawvon via Flickr (Creative Commons license)

---

Carol Newcomb is a Senior Consultant with Baseline Consulting. She specializes in developing BI and data governance programs to drive competitive advantage and fact-based decision making. Carol has consulted for a variety of health care organizations, including Rush Health Associates, Kaiser Permanente, OSF Healthcare, the Blue Cross Blue Shield Association and more. While working at the Joint Commission and Northwestern Memorial Hospital, she designed and conducted scientific research projects and contributed to statistical analyses.