A detailed description of the implementation of inpatient insulin orders with a commercial electronic health record system.

From Clinfowiki
Revision as of 00:50, 17 November 2015 by Tom4891 (Talk | contribs)

Jump to: navigation, search

The following is the review from the article “A detailed description of the implementation of inpatient insulin orders with a commercial electronic health record system” from Neinstein et al. (2014). [1].

Introduction and Background

Due to the support for implementation of meaningful use and guidelines, the clinical setting is rapidly changing the manner in which it delivers its health service to the patient, as well as the chain of events within it. An example of such change is present in the treatment of Diabetes Mellitus (DM) patients at the inpatient setting. Glycemic management is a crucial part of the treatment of chronic patients in this scenario since an uncontrolled follow-up can bring consequences to the health of the patient. Although, the clinical setting has at its disposal protocols and programs for this, most of them are paper-based and still lack several aspects within the management of the intervention program itself. For this reason the integration of a glycemic management program within the CPOE of an EHR can bring optimal benefits for the management of the disease in the inpatient setting.


Implementation of the current glycemic management program and shifting it and integrating it into the CPOE of the different EHRs of the clinical setting.


  • Step 1: Preparation and requirements gathering.
  • Step 2: Design and build.
  • Step 3: Implementation and dissemination.
  • Step 4: The complete optimization of the program.


  • Step 1: Conducted with the help of a clinical pharmacist and clinical informaticist. Both of them trained in the development and implementation of order sets of EHRs. This step suffered major problems with vendors providing sample insulin order sets (did not support needs).
  • Step 2: Encountered major problems from the transition of a paper-based entry into the system. Mainly, due to the fact that the paper entry required additional information which couldn’t be present simultaneously by the CPOE entry screen. During this phase, as a way to avoid different entry formats, they tried to emulate the formats of other institutions. However, they realized that this was not possible since both formats paper and virtual entry were different from several other institutions compared to them. Due to this, several changes were required to be done to satisfy the needs required for their specific clinical setting.
  • Step 3: Consisted basically on educating the prescribers according to their health care areas on how to order insulin using the order sets. The main areas of focus were pharmacist and nurses.
  • Step 4: Consisted on evaluating the system once it began according to the evolution of the glycemic management program. Several issues were noticed in the program such as inability to eliminate duplicates, unable to combine insulin doses into single ones through CPOE commands (Free-text), and the range of discrete dose set by the CPOE was sensitive so it created many alerts which in most cases were false-positives.


Transition didn’t occur as expected due to several conditions and factors not met or taken into account. Appearance of the order set in nurse vs. pharmacist. Stakeholders were not always present. Also, no data was present to compare from before the intervention vs. the point in time after it was set. Impact of the reliability of the study majorly is consequence of these last condition.


  1. Neinstein, A., MacMaster, H. W., Sullivan, M. M., & Rushakoff, R. (2014). A Detailed Description of the Implementation of Inpatient Insulin Orders With a Commercial Electronic Health Record System. Journal of diabetes science and technology, 8(4), 641-651. http://www.ncbi.nlm.nih.gov/pubmed/24876450