DNP 805 EHR Database and Data Management

Sample Answer

Better clinical outcomes and patient satisfaction are the most important things that every clinician looks forward to. As a doctor of nursing practice prepared nurse, one can be called upon in assisting with the designing of a clinical database of their organization for a better clinical outcome.

Therefore, the purpose of paper is to, “identify a patient clinical problem in which using a database management approach provides clear benefit potential, Identifying the data needed to manage this patient problem using information from the electronic health record (EHR), identifying whether the data is structured or unstructured and providing a complete description of the structured and unstructured data from the EHR that are needed to organize a hypothetical database” (GCU, 2017).

The Clinical problem

Predicting Sepsis Risk and mortality is a clinical problem that can be managed by data from EHRs to provide clear benefit potential. According to Miller, (2016), “Physicians are forever recording information about their patients.

They take vital signs, order lab tests and imaging, prescribe medications, check boxes to define patients’ diagnoses for billing purposes, and write or dictate narrative descriptions of each patient’s status.” All this data is found in structured and unstructured data.

“The widespread adoption of electronic health records by US health care providers is motivating a rapid growth in the use of predictive models to guide clinical decisions, to identify patients at high risk of future events (e.g., 30-day readmission), and to detect disease early, among other applications” (Dey, et al., 2016).

This data, structured or unstructured is used in predicting sepsis in its early stages, which has been found to be one of the “leading cause of death and hospitalization in the United States” (Dey, et al., 2016).
According to Desautels, et al. (2016), “Sepsis is defined as a systemic inflammatory response syndrome (SIRS) due to infection.

” They go on to explain that, “Sepsis, severe sepsis, and septic shock are umbrella terms for a broad and complex variety of disorders characterized by a dysregulated host response to infectious insult and because of the heterogeneous nature of possible infectious insults and the diversity of host response, these disorders have long been difficult for physicians to recognize and diagnose” (Desautels et al, 2016).

The criteria for SIRS is having, “Temperature >38°C or <36°C, Heart Rate >90 bpm, Respiratory Rate >20 Breaths Per Minute, or Arterial carbon dioxide tension <32 mm Hg (equivalent to 4.3 kPa) and White Blood Cell Count >11 or <4 (×109 cells), or 10% immature (band) forms” (Desautels et al, 2016).

Also measuring three elements, “lactate level, blood pressure and respiratory rate can pinpoint the likelihood that a patient will die from the disease”. There are several bedside scoring systems that can help nurses and doctors to predict sepsis so that an early intervention helps to prevent morbidity and mortality in these patients. Some of these bedside scoring systems are

•  “InSight
  
• qSOFA (quick SOFA)
  
• Sequential Organ Failure Assessment (SOFA) score
  
• Modified Early Warning Score (MEWS)
  
• Simplified Acute Physiology Score (SAPS II)
  
• Systemic Inflammatory Response Syndrome (SIRS) criteria” (Desautels et al, 2016).
  

AutoTriage “AutoTriage is designed to detect imbalances in homeostasis through the analysis of correlations between patient vital signs and clinical measurements over time. AutoTriage is designed to continuously sample and analyze patient measurement correlations automatically, and be able to alert clinicians to a deteriorating patient’s state” (Calvert J., et al., 2016).

Structured and Unstructured Data

According to Dey, et al., (2016), “Copious longitudinal structured and unstructured data are captured by EHRs to characterize the patient’s demographic (e.g., age, sex, address), health and treatment status, diagnoses, lab test results, and medication orders”.

They go on to explain that, “As much as 80% of the EHR data is thought to be in unstructured form and to effectively use EHR data it is important to understand how the data comes to be” (Dey, et al., 2016).

Structured data are objective data or hard data that are populated in the EHR, such as vital signs, laboratory values, patient demographics, dates, names, diagnosis codes, identification numbers, and “specific words and short phrases that are often presented in an easy-to-use user point-and-click interface via drop-down boxes with options to select the item of interest” (GCU, 2014).

So therefore in predicting sepsis these data play a vital role, especially the vital signs and laboratory results. On the other hand, unstructured data, are data points “that are typically “free text” in progress notes and care plans, comments embedded in the flow sheet, and the like”(GCU, 2014).

These comments and notes often contain valuable information that expands upon the structured data and that can provide beneficial input for a database designed to map to patient care, but unfortunately, unstructured data are not easily captured in the somewhat inelastic programming processes of a computer system.

That is where innovative and user-based database design comes into play” (GCU, 2014). “Unstructured data is the information that typically requires a human touch to read, capture and interpret properly. It includes machine-written and handwritten information on unstructured paper forms, audio voice dictations, email messages and attachments, and typed transcriptions–to name a few” (DataMark, 2013)

Description of data relationships that apply to the hypothetical database.
According to Shortliffe & Cimino, (2014), “Data provide the basis for categorizing the problems a patient may be having or for identifying subgroups within a population of patients.

They also help a physician to decide what additional information is needed and what actions should be taken to gain a greater understanding of a patient’s problem or most effectively to treat the problem that has been diagnosed” There is a need for accurate prediction of mortality risk and patient deterioration in the acute care units.

Advanced warning of patient deterioration is crucial for timely medical intervention and patient management, and accurate risk assessment aids in allocating the limited resources in these acute care units. Clinical Decision Support Systems (CDSS) have been used in these acute care units for predicting patient outcome and to score the severity of patient condition.

The vast majority of prediction models currently in use are based on aggregate baseline patient characteristics. These systems usually rely on a weighted linear combination of features, “such as age, type of admission, and vital sign measurements.

However, the most commonly used CDSS such as the Modified Early Warning Score (MEWS), the Sequential Organ Failure Assessment (SOFA), and the Simplified Acute Physiology Score (SAPS II), have suboptimal specificity and sensitivity when applied to patient mortality prediction” (Calvert J., et al., 2016) . These CDSS assessments assume that risk factors are independent from one another, and, therefore, they are not sensitive to the underlying complex homeostatic physiologies of patients.

Additionally, they do not account for variations in individual patient physiologies and trends in patient information. The increasing prevalence of EHR has evidence based practice providing a great opportunity to extract clinically relevant patient vital signs and laboratory results for increased predictive value in patient outcome (Calvert J., et al., 2016).

Conclusion

As a doctor of nursing practice prepared nurse, called upon in assisting with the designing of a clinical database of an organization for a better clinical outcome, is very important. In this paper it has been shown that there are bedside scoring systems that can predict sepsis at an early stage, leading to a beneficial outcome for both the patient and the clinician.

The data can be obtained from both structured and unstructured data. According to McCann, (2014), “Septicemia is currently responsible for the deaths of 36,000 people each year, according to data from the Centers for Disease Control and Prevention, making it the No. 11 leading cause of death in the U.S.” He goes on to explain that, “Officials estimate average sepsis mortality rates to be more than 16 percent nationwide.

In addition to the human death toll, the disease also costs the industry a pretty penny financially. It persists as the No. 1 most expensive hospital condition, costing more than $20 billion annually” (McCann, 2014). So having a designed database in the EHRs that can help clinicians predict this deadly clinical problem is very important.

Resourses

• Calvert J., Mao Q., Hoffman J.L., Jay M., Desautels T., Mohamadlou H., Chettipally U., Das R.
  (2016) . Using electronic health record collected clinical variables to predict medical intensive care unit mortality Annals of Medicine and Surgery, 11 , pp. 52-57.
  
• DataMark, (2013)Unstructured Data in Electronic Health Record Systems Challenges and
  Solutions http //insights.datamark.net/white-papers/unstructured-data-in-electronic-health-record-systems-challenges-and-solutions
  Desautels, T., Calvert, J., Hoffman, J., Jay, M., Kerem, Y., Shieh, L., … Das, R. (2016).
  Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data A Machine Learning Approach. JMIR Medical Informatics, 4(3), e28. http //doi.org/10.2196/medinform.5909
  
• Dey, S., Wang, Y., Byrd, R. J., Ng, K., Steinhubl, S. R., deFilippi, C., & Stewart, W. F. (2016).
  Characterizing Physicians Practice Phenotype from Unstructured Electronic Health Records. AMIA Annual Symposium Proceedings, 2016, 514–523.
  
• Grand Canyon University(GCU). (2014). Lecture notes. Retrieved on 3/25/17 from
  https //lcgrad3.gcu.edu/learningPlatform/user/users.html?operation=loggedIn#/learningPlatform/loudBooks/loudbooks.html?viewPage=past&operation=innerPage&topicMaterialId=607e9ca1-cd23-42f1-968b-1788cf9283c0&contentId=ebff44f5-7775-4830-b2f2-3eb94bd12979&
  
• Grand Canyon University(GCU). (2017), EHR Database and Data Management
  https //lc grad3.gcu.edu/learningPlatform/user/users.html?operation=loggedIn#/learningPlatform/user/users.html?operation=studentHome&classId=b8ffe6a2-0942-40fc-b88c-88fc1ee76d19&
  
• McCann, E (2014). Analytics project slashes sepsis deaths. Retrieved from
  http //www.healthcareitnews.com/news/data-analytics-strategy-slashes-sepsis-death-rates
  Miller, K. (2016) Learning from Patients’ Health Records
  http //biomedicalcomputationreview.org/content/learning-patients’-health-records Shortliffe, E., H. & Cimino, J., J. (2014). Biomedical informatics Computer applications in
  health care and biomedicine (health informatics) (4th ed.). New York NY Springer Science + Business Media. ISBN-13 9781447144748