Nurse practitioners as antibiotic stewards: Examining prescribing patterns and perceptions

Abstract

Background:

Advanced practice providers in the outpatient setting play a key role in antibiotic stewardship, yet little is known about how to engage these providers in stewardship activities and what factors influence their antibiotic prescribing practices.

Methods:

We used mixed methods to obtain data on practices and perceptions related to antibiotic prescribing by nurse practitioners (NP) and Veteran patients. We interviewed NPs working in the outpatient setting at one Veterans Affairs facility and conducted focus groups with Veterans. Emerging themes were mapped to the Systems Engineering Initiative for Patient Safety framework. We examined NP antibiotic prescribing data from 2017 to 2019.

Results:

We interviewed NPs and conducted Veteran focus groups. Nurse practitioners reported satisfaction with resources, including ready access to pharmacists and infectious disease specialists. Building patient trust was reported as essential to prescribing confidence level. Veterans indicated the need to better understand differences between viral and bacterial infections. NP prescribing patterns revealed a decline in antibiotics prescribed for upper respiratory illnesses over a 3-year period.

Conclusion:

Outpatient NPs focus on educating the patient while balancing organizational access challenges. Further research is needed to determine how to include both NPs and patients when implementing outpatient antibiotic stewardship strategies. Further research is also needed to understand factors associated with the decline in nurse practitioner antibiotic prescribing observed in this study.

Published by Elsevier Inc. on behalf of Association for Professionals in Infection Control and Epidemiology, Inc.

Appropriate prescribing of antibiotics is necessary for addressing antibiotic resistance. Antibiotic prescribing that is not concordant with guidelines can result in overprescribing and can impact patient safety by increasing the risk for Clostridioides difficile infection (CDI) and bacterial resistance.^1–4 Although efforts are increasing to address appropriate prescribing of antibiotics, the primary focus has been interventions targeting physicians^5–10 and little is known about advanced practice providers such as nurse practitioners (NPs) prescribing and participation in antibiotic stewardship efforts in the outpatient setting. A recent Society for Healthcare Epidemiology of America white paper outlined high-value targets for antibiotic stewardship efforts. One such target was identifying stewardship strategies for primary care, urgent care and emergency departments.¹¹

In the United States, NPs can prescribe medications, including antibiotics, with varying degrees of independence based on individual states’ NP practice authority. In early 2019, the Veterans Health Administration (VHA) granted independent prescribing authority of advanced practice nurses (which includes NPs), even in states that require physician oversight. Concurrently, there has been a significant increase in the number of NPs hired by the VA health system to address expanded capacity of primary medical care. Current data indicate an estimated 7,000 NPs working for the VHA alone, with an estimated 2,500 working in primary care. This makes NPs well-positioned as stewards of antibiotics.

Studies focused on NP prescribing have been conducted primarily in pediatric or long-term care settings.^12–14 There have been studies comparing other prescribers to physician prescribers. These studies indicate that antibiotics are more frequently prescribed during visits with NPs and physician assistants compared to physician-only visits.^15–17 Suda et al also examined trends in antibiotic prescriptions by provider type from 2005 to 2010 and showed that, during this time period, overall prescriptions for broad-spectrum antibiotic agents decreased for physicians and increased for NPs and physician assistants. These data suggest increasing trends in the United States for prescribing broad-spectrum antibiotics may be, in part, attributable to prescribers other than physicians.¹⁸

Uncovering factors influencing NP decision-making is essential to optimize guideline-concordant antibiotic prescribing. However, patients are also key stakeholders in antibiotic stewardship efforts. Literature indicate that patients recognize antibiotic resistance as a serious public health problem but may not perceive themselves as personally affected by it.¹⁹ There is also evidence that patients do not understand when an antibiotic is needed or not needed and can pressure prescribers for an antibiotic prescription.^20,21 Further investigation is needed to elucidate specifics on how patients could engage as good stewards to improve prescribing practices.

The goal of our study was to identify barriers and facilitators to guideline-concordant prescribing among NP prescribers in a VHA outpatient setting. We also wanted to explore perspectives about perceived roles in antibiotic stewardship efforts—both NPs and patient roles.

METHODS

We used a convergent parallel mixed methods design to uncover barriers and facilitators that can guide strategies to improve guideline-concordant prescribing of antibiotics. Data analyses (qualitative and quantitative) occurred separately, and neither analysis depended on the other during the final analysis. Findings were not compared or consolidated until analyses were completed.^22–24 As a recent study of methods used to examine NP prescribing indicated the need for a theoretical framework,²⁵ we used a systems approach—the Systems Engineering Initiative for Patient Safety (SEIPS) model (Fig 1)—to examine workflow and organizational factors that hinder or help NPs in making decisions about prescribing antibiotics. In this study, we examined process outcomes related to each aspect of the work system (people—those involved in the process of prescribing; tasks—diagnosing, guideline retrieval, and prescribing; tools—the electronic medical record, guidelines, and testing equipment; organization—the policies related to antibiotic prescribing and leadership aspects of implementing stewardship activities; environment—the physical environment of the VA ambulatory clinics). The SEIPS model emphasizes that any one work system component can affect changes in all other components.^26,27

Fig 1.

Systems Engineering Initiative for Patient Safety (SEIPS) model adapted to this study.

Study population

The primary study population was VA-employed NPs working in an outpatient capacity at one VA medical facility (William S. Middleton VA Hospital, Madison, WI) that includes 6 satellite ambulatory sites known as Community-based Outpatient Clinics (CBOCs). Veterans receiving care at any VA facility made up the study population of patient focus groups. Participation was voluntary for both NPs and Veterans. The institutional review board (IRB) determined this study met criteria for exempt human subjects research.

Qualitative data—NP interviews

We conducted face-to-face semistructured interviews to explore NP decision-making for patients presenting with nonspecific upper respiratory tract infections. We recruited NPs for study participation by presenting our study during a staff meeting. Study fact sheets were distributed during this meeting. The Associate Director of Patient Care Services attended this meeting and endorsed using NP protected time for the interview. We obtained names for all facility-level NPs who work in outpatient settings. An email request with the fact sheet was sent by the primary investigator to all NPs. Once the NP agreed to participate, the lead investigator called or emailed to coordinate a date and time to meet at the outpatient setting or, if an in-person interview was not feasible, a phone call interview was scheduled.

Interviews were conducted at 6 CBOCs located in rural areas of 2 states. At the time of this study (2019), there were 31 NPs working in the 6 CBOCs. We were able to reach 14 NPs (45%) for an in-person interview, with 2 interviews conducted by phone due to NP preference. All NPs were interviewed prior to February 2020—prior to the COVID-19 pandemic shutdown. We used both inductive and deductive methods to analyze interview data. Statements were extracted from transcripts and organized into themes and mapped to SEIPS elements. Interviews were recorded, transcribed by a VA-approved transcriptionist, coded for recurring themes, and mapped to the SEIPS framework (Fig 1). We used a survey conducted with NPs in 2012 as a starting point for developing the interview questions⁵ and consulted with a local VA infectious disease physicians. Questions were vetted by 2 investigators (KS and KB) and the primary investigator pilot tested the questions with 3 NPs at a rural VA facility not participating in the project. Modifications were made to questions based on the pilot test. The interviews covered topics such as appropriate prescribing, perceptions and ideas about the NP role in antibiotic stewardship, participating in developing an intervention targeting NPs, the VA policy mandating antibiotic stewardship, and barriers and facilitators to concordant prescribing in the outpatient VA setting. We also included questions on level of prescribing confidence, personal prescribing versus other NP prescribing, concern about antibiotic resistance, broad spectrum antibiotics versus using narrow spectrum agents and antibiotic stewardship program elements. We covered the “should you treat” and “how you treat” aspects of prescribing.

Qualitative data—Veteran focus groups

Focus groups were conducted with Veterans to generate multiple perspectives in an interactive group setting. Three focus groups were held between November 9, 2019 and January 9, 2020 (prior to COVID-19 restrictions). Recruitment consisted primarily of word of mouth. Inclusion criteria consisted of any Veteran, with no age or gender restrictions. A skilled focus group leader led all three focus groups (KB), while the primary investigator (MJK) took notes by hand. Focus group questions were generated from a combination of general literature related to patient engagement in antibiotic stewardship and the SEIPS framework. We piloted focus group questions with an existing health care-associated infection patient engagement group, which included Veterans. Order of questions and language were modified based on feedback from this group. At the conclusion of each focus group, the facilitator circulated examples of patient-centered antibiotic stewardship materials available through the Centers for Disease Control and Prevention (CDC) website. Focus group participants provided feedback on these materials. Duration of each focus group was approximately 1.5 hours. Veterans were paid $150 for participating to cover time and driving expenses. All 3 focus groups were conducted at the Madison VA facility.

Quantitative data

To assess guideline concordance with antibiotic prescribing, we extracted data over 3 calendar years (2017–2019) on NP prescribing for diagnoses categories (acute unspecified upper respiratory infections) in which, according to guidelines, antibiotics are not indicated. For those Veterans prescribed an antibiotic, we examined 2 diagnoses—acute bronchitis unspecified and acute respiratory infection unspecified for the type of antibiotic prescribed. Current CDC adult treatment recommendations indicate no antibiotics are needed for these 2 diagnoses. We also extracted patient demographic data including distance the Veteran lives from the CBOC (using zip codes). We also examined day of week and time of day of the clinic visit related to Veterans prescribed an antibiotic and those who were not prescribed an antibiotic, separating Monday-Wednesday, from Thursday and Friday visits. Finally, we investigated whether Veterans prescribed an antibiotic for upper respiratory illness had a positive CDI test within 90 days of the antibiotic prescribed.

RESULTS

Fourteen NPs were interviewed; demographic information is included in Table 1. A breakdown of recurring themes categorized by SEIPS elements including barriers and facilitators is depicted in Table 2. NPs appear to be strong advocates for better triage to reduce unnecessary office visits and to decrease unnecessary travel for Veterans and thus, providing access to Veterans needing a higher level of care. The factors most influencing NP comfort level or confidence with prescribing were patient-provider relationships, aversion to risk (potentially avoiding risk) and access to clinical decision support tools, including pharmacists. NPs identify as patient educators and building patient relationships was identified as essential to building trust between patients and providers and this promoted guideline-concordant prescribing by NPs. Time constraints had the least influence on prescribing confidence level, since NPs overwhelming stated that VA provides adequate patient visit time.

Table 1.

Nurse practitioner characteristics (n = 14)

Category	n (%)
Age group (years)
18–39	5 (36)
40–59	6 (43)
>60	3 (21)
Years since graduate school
<5	5 (36)
5–10	1 (7)
>10	8 (57)
Years in practice as NP
<5	5 (36)
5–10	1 (7)
>10	8 (57)
Years in practice at VA(NP or not NP)
<5	7 (50)
5–10	5 (36)
>10	2 (14)

Table 2.

NP interview recurring themes mapped to SEIPS elements

Recurring theme	SEIPS element	Barrier or facilitator
VA resources available	Organization	Facilitator
Ready access to pharmacists and infectious disease specialists	Organization	Facilitator
Lab results delayed if patient visit is late in day	Organization	Barrier
Veteran driving distance to outpatient clinic	Organization	Barrier
NPs see themselves as educators	People	Facilitator
Patient-provider relationships (communication) are critical to prescribing	People	Facilitator
Time with patient is adequate	Organization	Facilitator
Clinical decision support tools are available	Organization	Facilitator

A total of 15 Veterans (10 males, 5 females) participated in focus groups. Table 3 outlines recurring themes heard from Veterans with corresponding SEIPS element. These data were integrated with NP interview data to present a comprehensive picture of antibiotic use from the perspectives of both stakeholders—patient and prescriber.

Table 3.

Veteran focus group themes mapped to SEIPS elements

Recurring theme	SEIPS element/work system or patient factors
Knowing the difference between a viral and bacterial infection is important	People Patient knowledge
Antibiotics seem to be overprescribed and don’t always work; fear of “superbugs”	Tools and technology Patient knowledge
VA experience different than other health care facilities	Organization/work system
Pharmacists play a key role in communication with patients about antibiotics	Organization/work system
Proper triage can eliminate a trip to the clinic; access issues	Organization; tools and technology/work system

Examination of prescribing data for 3 years (2017–2019) indicated a sharp downward trend in the proportion of antibiotics prescribed by VA outpatient NPs (Table 4). During these 3 years, acute bronchitis unspecified was diagnosed 94 times with a prescription provided to the patient. In that same time frame, acute upper respiratory infection unspecified was diagnosed 72 times with a prescription provided to the patient. We examined one broad spectrum antibiotic—azithromycin—for these 2 diagnoses. Azithromycin is the most commonly prescribed antibiotic agent.²⁸ For acute bronchitis unspecified, azithromycin was prescribed in 46% (43/94) of visits over a 3-year time period, with a dramatic drop in both 2018 and 2019. However, for acute upper respiratory infection unspecified, azithromycin was prescribed in 28% (20/72) of patient visits but did not drop during the three years (Table 5).

Table 4.

Veterans diagnosed with nonspecific URI by NP (outpatient) versus prescribed an antibiotic (facility-level only)

Year	No. diagnosed with nonspecific URI	No. prescribed an antibiotic (%)
2017	595	237 (40%)
2018	666	68 (10%)
2019	609	28 (5%)

Table 5.

Percentage of visits in which azithromycin was prescribed for 2 different diagnoses

Year	Acute bronchitis unspecified	Acute upper respiratory infection unspecified
2017	53% (40/75)	27% (12/45)
2018	14% (2/14)	30% (5/17)
2019	20% (1/5)	30% (3/10)

There were no appreciable differences in the average ages of Veterans prescribed an antibiotic and those who were not prescribed an antibiotic (ages ranged from 63 to 67). There were also no appreciable differences in prescribing related to average distance traveled to the outpatient clinic. Distance ranged from 16.61 to 23.89 miles. Examining prescribing related to time of week, we found that NPs did not prescribe differently at the end of the week (Thursday-Friday) versus the beginning of the week (Monday-Wednesday) In fact, for all 3 years, examining prescribing for Thursdays and Fridays, 80% (296/369) of visits ended without an antibiotic being prescribed. Our data pull also indicated no incidences recorded in VA records of a CDI diagnosis during the 3 years for those Veterans prescribed an antibiotic.

DISCUSSION

We uncovered a downward trend in antibiotics being prescribed by NPs during the years 2017–2019. This appears to be contradictory to advanced practice providers’ prescribing trends seen in the private sector.²¹ This downward trend occurred during the same time period when NPs were given increased prescribing autonomy. An investigation into our local stewardship activities uncovered numerous pharmacy-led stewardship activities during 2018 and 2019. Two examples of these included nurse practitioner education on antibiotic stewardship and antibiotic stewardship reminder in the weekly pharmacy newsletter. Details of these initiatives are provided as Supplementary Appendix A. Although, no definitive correlation can be made, we believe these stewardship activities contributed to local downward trends in antibiotic prescribing and may be replicated at other facilities. This connection between increased outpatient stewardship activities and decreased prescribing of antibiotics is supported by the fact that we found no appreciable difference in antibiotic prescribing related to patient visit day, time of day, age of patient or distance traveled (access) to the clinic.

Including nurses as key stakeholders in antibiotic stewardship has been studied in the past. However, most studies have focused on stewardship opportunities for bedside nurses.^29–33 Examining prescribing patterns among outpatient NPs was recently recommended by the nursing profession.^34,35 Our study was unique in that we not only examined prescribing patterns, but we also examined NP and Veteran perceptions about antibiotic use and their respective roles in antibiotic stewardship.

First, we found that trust between the NP and patient was viewed as facilitating guideline-concordant prescribing—whether for upper respiratory conditions or other health problems. This important relationship factor between nurses and patients has been previously identified in the context of antibiotic stewardship.³⁶ Second, we learned that organizational factors such as delayed lab results and challenges with triage (patients coming in when not necessary) were viewed as barriers to concordant prescribing. Organizational factors were also outlined in a study on inpatient nurses.³¹ Nurse practitioners see themselves as patient educators, not just prescribers. Third, there is a need for patient health literacy related to use of antibiotics.³⁷ This harmonizes with current literature related to NPs working in both outpatient and inpatient settings.^38–41 Lastly, NPs expressed satisfaction with resources such as clinical decision support tools and access to pharmacists and infectious disease specialists. We found that both NPs and Veterans view pharmacists as an integral part of decision-making related to prescribing of antibiotics. This also harmonizes with current literature related to the increased role of pharmacists in antibiotic stewardship.^42–44

Although education efforts by the CDC about antibiotic use and antibiotic resistance has been in place for over 2 decades,⁴⁵ the need exists to communicate with patients about differences between viral and bacterial infections and why or why not an antibiotic may be needed.³⁷ Veteran patients appeared to understand the consequences of antibiotic overuse—talking about “superbugs”—but did not appear to understand the reason for the overuse. We also learned Veteran patients appreciate time spent with providers and indicated pharmacists as important communicators about the drugs they have been prescribed or are currently using. Veterans also see triage as important to saving driving time—which in rural areas can be significant. Our findings are similar to a recent study on provider perspectives on stewardship in the outpatient setting including the need for further patient education.⁴⁶

Our findings can be used to address human factors and work system barriers that exist for providers prescribing antibiotics in the outpatient setting—issues which ultimately have clinical implications.⁴⁷ The marked decline in antibiotic use over a 3-year time period indicates the need for ongoing and consistent stewardship activities targeting advanced practice providers in the outpatient setting.

We did not extract co-morbidities of Veterans presenting with unspecified upper respiratory infections. This could be considered a limitation; however, co-morbidities are not considered rationale for antibiotic prescription for viral respiratory infections. Another limitation is the small number of sites and participants in our study and all NP participants were employed by the same regional facility. However, these preliminary findings related to prescribing practices in conjunction with NP and Veteran perceptions can be used to both scale up and test interventions to increase guideline-concordant prescribing—interventions that are practical and easily integrated into routine care of Veterans seen in the outpatient setting. Stewardship committee activities, as described in Supplement A, can be replicated providing a strategy for those working on outpatient antibiotic stewardship issues.

CONCLUSION

NPs working in the outpatient setting rely not only rely on clinical judgment and reciprocal provider-patient trust, but also on work system tools and technology, such as clinical decision support tools and pharmacists to make prescribing decisions. NPs practice patient-centered care focused on communicating with and educating patients while balancing organizational challenges. This, coupled with increasing numbers of NPs in the outpatient setting, signals the need to include NPs as pivotal players within antibiotic stewardship programs. Further investigation is needed to determine how to increase both NP and patient engagement in outpatient antibiotic stewardship activities.