SAP for Healthcare Blogs
Discover insights and practical tips to optimize your healthcare operations and improve patient outcomes with SAP. Share your own experiences in a blog post.
cancel
Showing results for 
Search instead for 
Did you mean: 

By Martin Kopp, Werner Eberhardt, and Dominik Bertram

In many countries, healthcare costs are out of control. Yet as procedures get more sophisticated, the associated costs continue rising. And the problem is only growing as populations age. The solution? Focus on prevention and more effective treatments. By improving patient care in this way, healthcare organizations could see fewer readmissions and lower their costs.

To make the right – and smarter – decisions, physicians and others in the healthcare market need information at the point of care. And that information should be aggregated so that hospitals can better treat patients and better decide on the right course of care or predict potential outcomes of a procedure or prescription.

However, healthcare organizations are up against a uniquely vexing problem when it comes to making use of all the data in their midst. For one, their data silos can number into the hundreds when it comes to the different clinical systems and subsystems that form the IT backbone in hospitals and other healthcare-related organizations. Moreover, the unstructured data in these environments includes physician letters and notes containing critical clinical information. All this data about patient cases is a rich treasure trove, but in its unstructured, siloed state, it’s useless to other healthcare providers and caregivers. Even the structured data is confounding at scale because of the various classification systems and taxonomies applied to different data sets.

  

In a perfect world, all healthcare-related data would be digitized and presented in a standardized structure within a common database that physicians and others can
access and query in near real time. By applying advanced, predictive analytics to this data, healthcare organizations can make better evidence-based decisions, opening up a world of possibilities. Consider these scenarios.

  

1. Reducing readmission rates. Healthcare systems want to avoid costs by lowering re-admissions. They can do so by analyzing parameters around readmission rates and cross-referencing this with patient information to predict the likelihood of the patient returning. They can then take steps to decrease this likelihood based on what has worked with other patients fitting that profile. For instance, perhaps a hospital knows that patients undergoing surgery for a certain condition return if they are released without first undergoing a particular test. With this insight, the hospital’s physicians could order the test before those patients are released to proactively curtail
the likelihood of readmission.

2. Managing population health. By analyzing health and sickness patterns across a certain population of patients, healthcare organizations can make recommendations designed to keep that group of people healthy. For instance, a hospital may see that elderly patients with a certain diet are experiencing numerous bowel problems. Rather than wait for these patients to suffer from these issues and seek treatment, the hospital could launch a campaign designed to educate them about ways to prevent those problems from occurring.

3. Improving drug usage and efficacy. In today’s world, healthcare organizations and pharmaceuticals can tap into Big Data around genomic sequencing. By comparing information about modified DNA to healthy DNA, physicians can determine on a molecular level which drug and dosage will prove most effective for a patient. This will
improve treatments for all types of diseases. And when the efficacy of drugs is improved, patients require fewer treatments - which in turn lowers healthcare
costs.

For example, though less than 50% of chemotherapies are effective, most healthcare organizations spend a fortune on cancer drugs and treatments. Instead of prescribing a standard course of treatment for each cancer patient, it would be more effective to compare the profile of a cancer patient that has already been treated effectively with that of a similar patient needing treatment. In this way, the physician could recommend the perfect drug, dosage, delivery route, and frequency for that patient and increase the likelihood of an effective treatment.

  

Pharmaceuticals could also better warn physicians and patients about dangers associated with certain drugs. Consider a drug that is safe for most patients but can lead to
death in an older patient with kidney problems. Healthcare organizations could reduce these deaths with a platform in place that allowed physicians to check all dependent criteria and fully understand the risks.

4. Streamlining medical research. Today most drugs are researched in clinical trials designed to validate a hypothesis. However, it’s challenging and time-consuming to find suitable patients, randomize and run these trials, and collect and analyze the data. The process would be much more efficient if pharmaceuticals could access a large and well-organized clinical data warehouse. In some cases, pharmaceutical companies might be able to produce statistically valid treatment evidence without recruiting new patients for trials.

With today’s technologies, these scenarios are real possibilities. To take advantage of these opportunities for predicting and improving outcomes, technical experts
and healthcare providers must co-innovate to devise the most fitting data and predictive models and outputs.

What is your experience when it comes to using healthcare data to make better treatment decisions? How else can physicians and pharmaceuticals tap into data in context to improve patient care? Please share your thoughts in the comments.