Understanding the costs of Healthcare programs

27 January 2014 | Andy Pitman, Microsoft Government Solutions

Improving State-level Healthcare Analytics

Top of mind for health and human service leaders and the public at large is the Affordable Care Act.  Regardless of where one’s opinion lands regarding this topic, the reality is that state and federal government involvement in healthcare is growing and there’s a critical need to understand the costs and effectiveness of these programs.

There are multiple constituencies involved in these programs:  patients/taxpayers, providers, payers, and government administration.  Massive amounts of data is generated regarding for each of these constituencies, but that data typically resides in separate repositories.  Bringing that data together can yield important insights regarding program effectiveness, cost, and even fraud.  Not surprisingly, governments, particularly at the state level, are embarking on accessing that data through All Payer Claims Database (APCD) initiatives.

These APCD solutions are complex, time consuming to implement, and can be expensive.  Additionally, without inclusion of clinical data they may not yield the most effective performance measures.  Fortunately, there are vendors that provide comprehensive data solutions that are proven and effective at reasonable cost.  An example is Microsoft partner, PluralSoft.  PluralSoft builds upon SQL Server and SQL Server Reporting Services to create a user-friendly mash up of various data sources, enabling more informed decision-making by government officials responsible for administering and managing health-related initiatives such as, but not limited to, Medicaid.  PluralSoft’s following explanation explains the value of comprehensive data in APCD solutions:

The Elusive 360-Degree View of Your Data

There has been a continuing practice for over three decades now of using payer databases as the source for quality analytics. It is not a surprising phenomenon because that claims data is relatively structured, fairly standardized, and easily accessible. Unfortunately, this approach is much like making a peanut butter and jelly sandwich, but forgetting to add the jelly.  Yes, it is still a sandwich; but it is lacking an important element!

Let’s undertake a simple analysis of claims data as a sole source for healthcare analytics, as follows:

What can you measure with an all payer claims data?

Performance measures often accessible from claims data include, but not limited to:

  • Mortality rates
  • Complications and co-morbidities
  • Access to appropriate health services
  • Length of stay (duration of treatment)
  • Cost of care at various levels
  • Compliance to certain practice guidelines (can be determined with CPT codes)
  • Adherence to certain medication guidelines
  • Different types and levels of utilization
  • Cost and frequency of episodes
  • Network census

But what can’t you measure with an all payer claims data?

Performance measures NOT accessible from claims include, but are not limited to:

  • Biomedical measures based on clinical observation/results/findings (e.g., results of laboratory tests and vital signs)
  • Disease specific symptoms (severity, duration, frequency, physical exam and psychological/cognitive/emotional) based on patient self-report and clinician observation/report
  • Quality of life/morbidity (e.g., pain/discomfort, treatment side-effects, family and social interactions, sexual function, functional status/independence) based on patient self-report and clinician observation/report
  • Safety (avoidable adverse events due to errors and omissions)
  • Patient-reported satisfaction with care
  • Patient knowledge through education of one’s condition
  • Patient Self-management
  • Changes in lifestyle (e.g., tobacco use, food choices, physical activity, use of alcohol and illicit drugs)
  • Reasons for variance from practice guidelines
  • Comorbidity influences

So, what happens when all payer claims data is merged with clinical data?

  • Both types of data are needed from a single view to assess effectiveness, access, utilization and cost of care.
    • Is the diagnosis correct (i.e., fits the biomedical measures and symptoms)
    • How much the patient’s biomedical measures improved (or worsened)
    • How much the patient’s symptoms improved (or worsened)
    • How much the patient’s quality of life improved (or worsened)
    • Process of care errors and omissions (gaps in care)
    • The usefulness of particular guidelines for particular types of patients
    • Whether the patient was satisfied with care
    • If the patient was educated adequately in self-care (well-care)
    • If the patient is complying with the plan of care and if the plan of care is appropriate for the patient
    • If a provider has a good reason for varying from the practice guideline and if such variance improves (or worsens) outcomes
    • Impact of coexisting conditions on outcomes
  • Detection of fraudulent activity
  • Data quality assessment

As we can all see, combining clinical and claims data provides the insight that costs cannot be contained without an understanding of quality factors. Similarly, quality cannot be enhanced without an understanding of cost factors. In short, your peanut butter and jelly sandwich has to contain both the peanut butter and the jelly! Bon appetit!

Andy Pitman
Microsoft Government Solutions