The Latin term “Pro re nata,” or “PRN” as it is better known, translates to “as the situation demands” or “according to circumstance” or “as needed”[1]. The practice of writing PRN orders is common for medication prescribing in hospitals, other skilled medical facilities, and even home use. Balancing the convenience of PRN medication prescribing with patient safety is complex: defining the difference between a PRN medication and an anticipatory medication is not clear to most providers or nurses. Standing orders have demonstrated benefit in improving vaccination rates and also improving preventative screening processes [2]. Other uses for PRN medications include the anticipation of likely medical events or patient needs occurring in the absence of a physically present physician. A PRN order for a stool softener, for example, can help with patient comfort and avoid delay in care or management. Literature supports the use of anticipatory orders in the settings for palliative care and end of life comfort. In general, medical practice reliance on PRN medications has extended well beyond anticipatory

medications such as stool softeners and use of opiates in the terminally ill, to treatment of dynamic medical conditions [3]. The role of anticipatory medication in treating known, stable conditions is of clear benefit to the patient and treating practitioner alike. Less clear is the reliance on PRN medications in unknown and variable clinical scenarios, which may result in duplication of PRN medications further complicated with limited prescriber direction provided. One of the advantages PRN orders provide is flexibility in dosing based on the individual patient’s needs. A potential drawback to this type of prescribing practice is opportunity for interpretation of the requested order when guidelines for administration are limited [4].There is limited literature providing input on how to safely use this routine prescribing tactic. In a study examining the use of as needed psychiatric medications, 73% of the time PRN psychiatric medications were initiated by nurses [5]. Nurses have tremendous impact on the use of PRN medications, from administration to educating the patient of treatment options.

Following an undesired outcome involving the use of PRN medications as described in the case below, a simple medication use evaluation of PRN medication use by one service was evaluated.

Case

​
A 77-year-old white female admitted to the non-teaching service from an outside nursing home following a two day history of progressive and worsening agitation, altered mental status, and suspected urinary tract infection. Past medical history included hypertension, mood disorder, Alzheimer’s Dementia, neuropathy, myelodysplastic syndrome, gastric reflux disease, depression, seizures, and chronic kidney disease stage II. During hospitalization, the patient was prescribed 13 scheduled medications and 13 PRN medications. Refer to Table 1 for a listing of the medications.

The PRN medications were prescribed within the first five hours of admission. On the first hospital day the patient’s morning blood pressure was 170/95 mmHg. At this time, the patient received scheduled amlodipine, as well as the available PRN medications: clonidine, hydralazine, and haloperidol. Both PRN blood pressure medications included parameters for use with systolic blood pressure (SBP) greater than

Table 1: (BID: twice daily: 1000 and 1800; daily: 1000; QHS: 2200; TID (0800, 1300, and 1800); Q hours: every # hours permitted dosing)*APAP: acetaminophen

160 mmHg, but lacked guidance for preferred administration order or instructions for repeat blood Pressure monitoring before giving additional medications. Two hours later, noon vital signs revealed a BP of 74/38 mmHg. A subsequent BP assessed by the physician read 50/30 mmHg and the patient was noted to be unresponsive which resulted in an intensive care unit transfer. The patient was hemodynamically stabilized and transferred back to the general medicine floor two days later.

Additional safety concerns are the potential for drug interactions due to the number of concomitant medications available. For example, the number of different formulations of acetaminophen products may allow for the 24 hour administered dose to exceed a safe quantity. The multiple agents with overlapping side effect profiles creates concern for additive potential or enhanced effects leading to an altered central nervous system, sedation, and/or respiratory depression. Cardiovascular effects are also a concern due to combinations with risk of an increased potential for QT interval prolongation, exacerbation of orthostatic regulation disturbances and increased arrhythmogenic potential. This is not an all descriptive list of potential adverse effects but further insinuates the concerns with this particular case presentation.

This case illustrates the complexity of using PRN medication orders and the adverse outcomes associated with their undirected use. It also provides an opportunity for improved design of electronic order sets and careful review of systems of safety using electronic order sets. The nurse executed the physician orders as written and gave all PRN medications indicated by the orders, in this case “blood pressure.” The patient received clonidine and hydralazine in addition to scheduled amlodipine. The patient was also agitated, so the nurse gave the haloperidol at the same time as the three blood pressure medications, following her interpretation of the orders as written.

The decisions for use of PRN orders are multifaceted and include choosing between medications prescribed for the same indication, timing of administration, and route of administration. A literature search to review current practice and guidelines for PRN medication orders was conducted. Search terms used were PRN medication use, safety of PRN medications, PRN parameters, anticipatory orders, and nurse PRN. There is a paucity of published literature regarding PRN medication safety use, but an article closely related to this case evaluated the use of PRN psychiatric medication use as well as nurses’ opinions on appropriate administration of pain medications [4,5]. A study conducted by Usher and colleagues evaluated various factors influencing psychiatric PRN medication use by retrospective chart review [5]. One variable of importance, highlighted in a study done in 2001, was that the documented reason for medication administration was absent 36% of the time [5]. In 2008, the Joint Commission on Accreditation of Healthcare, or JCAHO, required an indication for all PRN medications [6]. A study conducted by Gordon and colleagues evaluating nurses’ opinions on timing of PRN pain medications and the strengths given, showed one-third to one-half of all participating nurses chose to administer a dose that did not adequately cover the patients’ need [4]. The nurses’ reasoning was based on their discomfort with the potential adverse effects associated with overdosing, but they were acting within the range provided by the limited PRN order guidance and were left to interpret what was felt to be the safest for patient care [4]. Twenty-one percent of nurses indicated that they would give less than what was ordered by the physician, in attempt to avert harmful effects to the patients [4]. The researchers of this study felt the results accurately portrayed the need for prescribers to write appropriate range orders to guide the nurses’ practice.

Due to the lack of literature and the potential impact PRN medication use can have on patient safety, we were provided with an opportunity to evaluate the PRN prescribing practices, use of parameters, and guidance within our institution.

This project was designed as an institutional review board (IRB) approved concurrent observational evaluation of the PRN prescribing practices for a non-teaching service within a large academic medical center. The non-teaching service consists of two daytime physicians along with a nurse practitioner and one physician overnight. Data was collected for all patients admitted to the service over a five day course from Monday to Friday. Inclusion criteria were patients prescribed a PRN medication. Other information collected included dosing parameters, instructions or additional direction provided, and indication. Standard demographics were collected to define the type of patients on the service, including age, race, sex, number of co-morbid diseases, and number of scheduled medications. Descriptive statistics were used for interpretation of the

Information collected. The project was approved by the IRB as part of our medication use evaluation program.

During the specified five day window of observation, 45 patients with at least one PRN medication were identified on the non-teaching service. Table 2 illustrates the types of patients on the service. On average, each patient was prescribed seven PRN medications with less than one PRN medication per patient containing parameters or additional administration directions beyond indication. Graph 1 depicts the more commonly prescribed PRN medication indications. For the patients prescribed a PRN medication, if more than one PRN order existed per indication (n=57 occurrences), 28% of those medications prescribed included parameters or additional directions for administration. For those orders that did include parameters, most were part of an alcohol withdrawal or hyperglycemic management protocol. Forty-two patients had a total of 88 PRN orders written for pain medications. Thirty-two of those patients were prescribed more than one PRN pain medication, with only two of them containing parameters for selection guidance. During two instances, the PRN medication was converted to a scheduled medication based on frequency of use data.

Table 2. Demographics with number of medication

PRN, standing orders, by protocol, or anticipatory orders are equivalent in their exemption of prescriber presence to implement their use. Unfortunately, parameters, boundaries for use, and guidelines are often omitted from these orders that may provide personnel with necessary details for implementation. When instruction for use is lacking, there is room for alternative interpretation, self-discretion, and clinical judgment which will vary among individuals. Parameters set by the ordering prescriber are beneficial; however guidance for their use should also be included to minimize the chance for misinterpretation and responsibility felt on behalf of personnel to determine the appropriate course of action to address the patients’ needs. An example is depicted in Table 3.