This is Part II of the post: What Nurses Expect from Healthcare IT solutions. Part I can be found here.

As mentioned in Part I, if we were to categorize the responsibilities of a nurse into three possible segments (care coordination, digital documentation, and organizational responsibilities), the characteristics of a Health IT solution that nurses will expect to have for performing their duties with ease can be listed accordingly – as shown below:

Care coordination

Everyday activities are mostly concentrated in the medication-administration module of a system. The features of this module should be such that a nurse can:

• view all kinds of orders prescribed by the physician for the inpatient. These can be normal medication orders such as tablets, syrups, sliding-scale regimen, etc.; intravenous fluids, e.g., those pertaining to orders for chemotherapy; any items like syringes or gloves for the patient; and any TPN (total parenteral nutrition) orders. Most CPOE systems have the provision to prescribe normal Rx. However, features to prescribe complicated orders like sliding-scale regimen and IV fluids in fractions are either missing or poorly designed.

• have the ability to view the batch number and expiration date of such orders supplied by the pharmacy. When patients bring their own medicine, there should be a provision to enter information regarding it. For example, consider a patient with a history of hypertension being treated in the emergency room due to symptoms diagnosed as cardiac arrest. The physician treats the patient immediately. Knowing the patient to have a history of hypertension, the nurse asks for details of the medication that the patient usually takes. The patient also hands over a strip of a medication that he or she is supposed to take; the nurse administers the same at the scheduled time and enters the details in the EHR of that patient. Most medication-administration record systems fail to provide the ability to capture information regarding the medications that the patient brings from home. Otherwise, almost every MAR has the provision to capture inventory details like batch number, expiration date, and so on.

• return the medication to the pharmacy via the medication-administration module, with a reason that can be entered and witnessed by the assistant nurse in case of damaged or expired medication. Almost every Health IT systems has the provision to capture details of the returned medication. However, the ensuing workflow, like updating the inventory or capturing the reason or signature of the witness, is mostly missing. Also, even if these features are present, they are poorly implemented, with only the main function of returning medications available with no logical workflow following that.

• have the ability to view allergy details of the patient, enter vital signs, and be able to place an order and amend any prescription upon the consent or authorization (with the help of an authorization screen) from the physician. This holds true for discontinuing and canceling an order as well, and is well-implemented in most systems.

• know that if the patient needs to be administered a scheduled drug, there is a provision for a witness screen that signifies that this controlled or scheduled drug was administered in the presence of witnesses. Most Health IT systems, even if designed efficiently, may not be able to evaluate whether it’s properly and ethically used. For example, despite having to provide a unique username and password to capture the signature of the witness/nurse, there may be scenarios wherein the nurse may ask a colleague to sign on his or her behalf by providing his or her password. The colleague may not even be aware of the reason for signing and may just blindly sign.

• be aware of drug interactions while placing orders via drug-interaction apps integrated into the system. This is mostly well-implemented in many Health IT solutions.

• know that the right dose is being administrated to the right patient at the right time through the right route. This kind of barcode system is mostly implemented but poorly designed.

• have a view of all the patient details (i.e., diagnosis, vital signs, and allergy information). This is mostly well-implemented and well-designed in Health IT systems

• have relevant alerts (in the case of any drug to be administered or ordered from the pharmacy) as an integral part of the module. This includes alerts when the nurse is directly placing an inventory order and the medicine is out of stock. The feature to display alerts is implemented. However, it is poorly designed because the user is not able to understand the purpose of the alert, or essential alerts are not provided.

• Besides orders, alerts in MAR (Medication Administration Record) systems may be used in the following situations:
  • When the nurse forgets/or has yet to administer a given dose of medication within the scheduled time. The system in this case can display an alert or message a few minutes prior indicating that the medication (along with dose and preconditions like before food or after food) is yet to be administered.
  • The Health IT system can throw an alert to indicate that the patient has to be moved or discharged prior to a specified time. For example, assume that the patient has to be moved from a deluxe room to a standard room at 6 p.m. An alert at 5:45 p.m. can be displayed informing the nurse.

Consider the following scenarios:

Scenario (1) : An alert when the nurse administers a medication that the patient is allergic to.

Here, the nurse first administers the medicine and then enters it in the system. The system then displays an alert to the nurse and the clinician in-charge to indicate an emergency situation if the allergy condition is not alleviated.

Scenario (2) : An alert before the nurse administers a medication that the patient is allergic to.

Here, the nurse first enters the details of the medicine in the system before administering that medicine. The Health IT system in this case displays an alert to the nurse, indicating that the patient is allergic to the medicine, and the nurse is informed that this drug is not to be administered. Thus, she refrains from giving the medication to the patient.

In scenario (1), the nurse first administers and then enters it in the system. In scenario (2), the nurse logs in the details of the medicine and, upon cross-verifying, might safely administer the medicine anyway. Here, the appropriate manual intervention at the appropriate time will decide whether the alert is useful or not.

The point to be argued is that one may not have control over what the nurse decides to do. Would the nurse first administer the medicine or enter the details in the system? Is an alert in any way beneficial to the nurse here? Or will proper training help the nurse make informed decisions that make the alert beneficial?

These features are implemented in most of the systems. However, as mentioned in the above paragraph, it is up to the user to use them intelligently in order to derive the benefits. Also, most of the alerts are not implemented well due to their poor design.

The final part of this article can be found here.