Question & Answer: WHAT DID YOU LEARN FROM THE CASE STUDY AND HOW CAN…..

WHAT DID YOU LEARN FROM THE CASE STUDY AND HOW CAN THE PROBLEM BE FIXED?

Questions of Evidence —– Timothy B. Patrick and Norma M. Lang

Chris, an academic health informatics specialist; Parker, a registered nurse; and Alex, a hospital clinical IT manager and strong proponent of electronic health records (EHRs), are discussing the virtues of EHRs for managing nursing care data and information and the obstacles to EHR- based clinical decision support.

Parker: To show you that EHR can’t handle complex situations, listen to this case. Selina Jones is 80 years old. She is admitted to the hospital medical-surgical unit after a fall. It appears she has sustained an injury to her hip that requires a surgical intervention. She is in moderate pain but has difficulty moving, nauseated, calling for her dead husband, very anxious, and not understanding what happened to her—let alone the diagnostic tests, surgical plan, and other treatments she is and will be receiving.

Selina’s daughter has assigned herself the authority to make healthcare decisions for her mom. The daughter agrees to have the surgeon do a surgical pinning of Selina’s hip. The registered nurse is responsible for doing the patient assessment and making the decisions on how best to prepare the patient for surgery, so the nurse has to communicate these decisions to other members of the clinical team. The surgeon and registered nurse are also responsible for creating the postsurgical plans, which include visits to a physical therapist and a social worker.

Surgery is complete. Selina is sent home, but she continues to have unmanaged pain, confusion, anxiety, no ability to participate in her own treatment or therapies, and no understanding of the risk for falls. She can’t sleep well, has acquired pressure ulcers during the hospital stay, and suffers from urinary incontinence.

Chris: Her family situation and medical issues do sound complicated. Alex, you must admit that your EHR is no panacea for this.

Alex: Of course it’s no panacea—nothing is. The golden rule in IT is good data in, good information out. If you enter inadequate data, you can’t expect complete results.

Chris: Good data in, good information out certainly requires more than the usual ICD and CPT codes entered into the EHR. But let’s back up and focus on decision support. At the university hospital, we designed a set of standard nursing practice recommendations for assessment and interventions related to fall risks. So how could we determine whether nursing practice in your hospital conforms to our practice recommendations?

Alex: The data fields in our EHR database have to match the key concepts in your practice recommendations.

Parker: A standard approach matches both sides to a reference vocabulary, like SNOMED CT.

Chris: Suppose your EHR used SNOMED CT or some other standard vocabulary in the first place? Then you really could achieve good data in, good information out.

Alex: Yes, but not everything’s perfect. Besides, operations have its own pressures. As Shakespeare wrote, “There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy.”

Chris: Is matching of data and concepts enough? It seems that your data have to show that the right (according to our recommendations) assessments and interventions were taken. For example, according to our recommendations, incontinence is a risk factor for falls, so if the patient assessment included that, a protocol for fall prevention is implemented.

Alex: Good example—and you could find that in our data.

Chris: Always? And why would you find that?

Parker: Because that protocol for that assessment is common practice?

Alex: Our clinicians are experts and follow good practice in their care plans.

Chris: In our recommendations we cite clinical studies that provide evidence for incontinence as a risk for falls.

Alex: Isn’t it enough if our practice and your recommendations agree on what ought to be done even if the reasons are not strictly the same?

Chris: I don’t think so. Maybe it would be in a simple or an isolated case, but I’m not comfortable with that position in general—and certainly not in a complicated case like that of Selina Jones.