Acceptability of Naloxone Dispensing Among Pharmacists

Abstract

Background

Addressing the opioid epidemic requires participation from the entire healthcare system, including pharmacists. The San Francisco Department of Public Health initiated naloxone prescribing at six safety net clinics. We evaluated this intervention, demonstrating that naloxone prescribing from primary care clinics is feasible and acceptable.

Objective

To evaluate acceptability of naloxone dispensing to patients prescribed opioids among pharmacists serving clinics participating in a naloxone intervention.

Methods

We surveyed 58 pharmacists at pharmacies that serviced San Francisco safety net clinics. In-person surveys collected information on demographics, experiences in dispensing naloxone, and interest in prescriptive authority. Surveys were completed November 2013 through January 2015. We conducted descriptive analyses and assessed bivariate relationships.

Results

Most respondents were staff (56.9%) or supervising pharmacists (34.5%). Most (92.9%) were aware their pharmacy stocked naloxone and 86.8% felt it should be prescribed to some or all patients on long-term opioids. Most (82.1%) dispensed naloxone at least once in the past 12 months. More than half were comfortable providing naloxone education. Nearly half (43.4%) indicated they would want authority to furnish without a prescription. Over half (55.2%) reported no problems dispensing. The common problem was lack of sufficient naloxone knowledge. Only 12% reported more than one problem in dispensing naloxone to patients, which was associated with being uncomfortable with educating patients (p=0.03).

Conclusion

Naloxone dispensing was acceptable among pharmacists. Their most cited problem was lack of sufficient naloxone education. This may be resolved with improved instructional materials, incentives for extensive patient education, or mandatory training.

Introduction

Over 42,000 deaths in the United States were attributed to opioid overdose in 2016.¹ Deaths related to opioids jumped from 1.5 to 5.9 deaths per 100,000 persons between 2000–2014.² Distribution of the opioid antagonist naloxone has become common practice at low-threshold drug user services, such as syringe access programs, and has been associated with a reduction in opioid overdose mortality.³ More recently, there has been an effort to reach those at risk of opioid analgesic overdose by integrating take-home naloxone prescribing into primary care practice to be used on the patient in the event of an opioid overdose. The Centers for Disease Control and Prevention’s Guidelines for Prescribing Opioids for Chronic Pain now recommend prescribing naloxone to patients who may be at risk of overdose, including patients who take ≥50 morphine milligrams equivalents per day, use benzodiazepines concomitantly, or have a history of substance use disorder.⁴

Changing practice, however, requires active participation from multiple elements of the healthcare system, including pharmacists. Given that patients who receive opioids for chronic pain are at increased risk for overdose, having pharmacists dispense naloxone to this demographic is a potential strategy to decrease mortality rates. A recent study from Australia indicated that pharmacy-based naloxone dispensing was acceptable among pharmacists; however, it also suggested that pharmacists may need more training in dispensing and educating patients around naloxone for take-home use.⁵ Another study in West Virginia posited that naloxone education for community pharmacists could better prepare them to decrease opioid-related deaths.⁶

In response to data demonstrating that most opioid overdose decedents in San Francisco were prescribed opioids through primary care⁷, the San Francisco Department of Public Health in 2013 initiated naloxone prescribing at six safety net clinics. These clinics are locations that furnish health services to uninsured or publicly insured individuals. We conducted a multi-faceted evaluation of this intervention, demonstrating that naloxone prescribing from primary care clinics is feasible and acceptable among pharmacists. In addition, we surveyed pharmacists from community pharmacies that had dispensed naloxone.

Objective

The aim of this study was to evaluate pharmacists’ acceptability of naloxone dispensing to patients prescribed opioids for chronic pain at clinics participating in a naloxone co-prescribing intervention and to assess the feasibility of naloxone prescribing through primary care clinics. In this study, acceptability referred to the belief and attitude that naloxone should be prescribed.

Methods

Naloxone Prescribing Program

Starting in 2013, primary care providers at the six participating clinics were trained and encouraged to prescribe naloxone to patients on long-term opioids (≥3 months) or otherwise at risk of experiencing or witnessing an opioid overdose. The CDC guidelines define pain as chronic if it lasts more than 3 months, so this study used the same timeframe to characterize long-term opioid use.⁸ Providers were encouraged to prescribe the off-label intranasal naloxone to be used with a nasal atomizer, as the naloxone auto-injector and intranasal spray had not yet been approved by the Food and Drug Administration. Clinic staff gave patients a kit containing a naloxone instructional brochure and a nasal atomizer. Additionally, clinic staff taught patients how to assemble the device with a demonstration kit. A naloxone prescription was sent electronically to the patient’s preferred pharmacy and the patient was instructed to bring the other items to the pharmacy. The parent study identified chronic pain patients – patients prescribed opioids for equal to or more than 3 consecutive months for chronic non-cancer pain – to determine whether a patient would receive naloxone and naloxone education by clinic staff.

Before clinics began prescribing naloxone, the study pharmacist received a list of pharmacies most frequently utilized by clinic patients. The study pharmacist contacted each pharmacy at least one month prior to program initiation to alert them to expect naloxone prescriptions. The study pharmacist offered information by on-site presentation, telephone, or fax (see Appendix 1), with most pharmacists electing fax. Education included information on naloxone pharmacokinetics and pharmacodynamics, ordering information including national drug codes, how to assemble the medication, formulary status, and how to educate patients on opioid overdose risk factors and identifying and managing an opioid overdose. Because this information was often disseminated via fax, we do not know how many pharmacists at each pharmacy received the information. The study pharmacist was available on an as-needed basis for additional technical support throughout the project, usually initiated by clinicians who had difficulty getting a naloxone prescription filled at a given pharmacy. Study activities were approved by the University of California San Francisco Committee on Human Research (CHR# 13–11168).

Survey Instrument

The paper survey was developed by study staff in consultation with the study pharmacist. The study pharmacist acted as the content expert and assisted in developing the survey instrument. The survey was expected to take approximately 5 minutes to complete and consisted of 15 questions related to demographics, experience prescribing naloxone, barriers to filling naloxone prescriptions, personal opinions around naloxone prescribing, and interest in prescriptive authority (See Appendix 2). Pharmacists could report up to ten pre-specified problems with naloxone dispensing, including the need for more training to educate patients on naloxone, insurance coverage issues, and device complications. Questions regarding attitudes, beliefs, and thoughts on naloxone dispensing assessed pharmacist acceptability, while those about experience with dispensing evaluated the feasibility of naloxone prescribing via primary care clinics. Since the maximum number of problems reported was 3 and since the data for this variable were skewed, responses were summarized as a binary variable in which pharmacists reported ≤1 problem or >1 problem with naloxone dispensing. Number of years licensed to practice pharmacy was coded as less than 5 years, 5 to 20 years, and more than 20 years. Pharmacist opinion regarding if patients should receive take-home naloxone was collapsed to yes (to all or some patients) and no (no patients or unsure) due to small strata sizes. Preference for prescriptive authority and comfort in educating participants were binary, “yes-no”, variables.

Procedure

One-hundred and thirty-four pharmacies were identified as locations utilized by patients at the six participating safety net clinics. Study staff successfully contacted 103 of the 134 pharmacies used by the six participating clinics, 33 of which reported that at least one pharmacist had dispensed naloxone at least once. Study staff then distributed surveys in-person to pharmacists at these 33 pharmacies. Full-time and at-least half-time pharmacists were asked to participate in the survey regardless of if they had personally dispensed naloxone. Fifty-eight eligible pharmacists completed the survey. Pharmacists were given an informed consent form that explained the purpose of the study. The survey was self-administered and participation was voluntary. Participants were offered a $5 gift card for completion of the survey. The surveys were conducted between November 2013 and January 2015.

Analysis

Survey results were entered into Qualtrics Survey Software (Qualtrics, Provo, UT) by study staff immediately after survey completion. The results were exported into Stata version 14 (College Station, TX) for analysis. We ran descriptive statistics on demographic and pharmacist acceptability of naloxone dispensing. We used Fisher’s exact test to assess the crude associations between problems experienced by pharmacists in dispensing naloxone or number of years licensed to practice pharmacy and the following variables: preference for authority to furnish naloxone without prescription, beliefs that naloxone should be available to patients, and comfort in educating patients about naloxone.

Results

Pharmacist characteristics

There was a total of 73 full or at least half-time pharmacists eligible to complete the survey, and 58 participated. Over three-quarters (79.4%) of the pharmacists at the 33 eligible pharmacies completed the survey.

Most respondents were staff (56.9%) or supervising (34.5%) pharmacists and the majority worked at chain (69.0%) or hospital (20.7%) pharmacies. Duration of time in practice varied, with 28.3% of pharmacists licensed for less than 5 years, 41.5% for 5–20 years, and 30.2% for over 20 years (Table 1).

Table 1:

Pharmacist Characteristics

	N	(%)*
Total	58
Type of Pharmacy
Chain	40	(69.0)
Hospital	12	(20.7)
Independent	5	(8.6)
Other	1	(1.7)
Position
Owner	2	(3.4)
Managing/supervising pharmacists	20	(34.5)
Assistant pharmacy manager	1	(1.7)
Staff pharmacist	33	(56.9)
Other	2	(3.4)
Years Licensed To Practice Pharmacy (N=53)
≤5	15	(28.3)
5–20	22	(41.5)
>20	16	(30.2)

Acceptability of Naloxone Prescribing

Most respondents (92.9%) were aware that their pharmacy stocked naloxone and 86.8% felt that naloxone should be prescribed to some or all patients on long-term opioids. Eighty-two percent of respondents had dispensed naloxone at least once in the past 12 months, and 35.7% dispensed naloxone to five or more patients in the past 12 months. Seventy-five percent of respondents dispensed at least one intranasal naloxone device and 8.9% dispensed at least one intramuscular naloxone device. Most respondents (46%) had dispensed naloxone to one to four patients (Table 2).

Table 2:

Pharmacist Acceptability of Naloxone Dispensing and Comparison of Select Acceptability Characteristics by Problems Dispensing Naloxone and Years Licensed to Practice Pharmacy

			Problems With Naloxone Dispensing				Years Licensed To Practice Pharmacy
	Overall		1 problem or less		More than 1 problem		<5 years		5–20 years		>20 years
	N	(%)	N	(%)†	N	(%)†	N	(%)†	N	(%)†	N	(%)†
Total (unless specified)	58		51		7		15		22		16
Aware pharmacy stocked naloxone (N=56)
Yes	52	(92.9)
No	4	(7.1)
Has dispensed naloxone in past 12 months (N=56)	46	(82.1)
Intranasal	42	(75.0)
Intramuscular	5	(8.9)
Number of patients dispensed naloxone to (N=56)
0	10	(17.9)
1–4	26	(46.4)
≥5	20	(35.7)
Received education around dispensing naloxone (N=54)
Yes	25	(46.3)
No	29	(53.7)
Comfortable educating patients about naloxone (N=54)
Yes	33	(61.1)	32	(62.7)	1	(14.3)
No	21	(38.9)	16	(31.4)	5	(71.4)
Provided naloxone education to at least one patient (N=50)
Yes	26	(52.0)
No	34	(68.0)
Should naloxone prescription be offered by prescription to patients on longterm opioids (N=53)
Yes to all/some	46	(86.8)					13	(86.7)	15	(68.2)	14	(87.5)
No/unsure	7	(13.2)					1	(6.7)	5	(22.7)	1	(6.3)
If legal, would want to exercise pharmacist prescriptive authority to prescribe naloxone‡ (N=53)
Yes	23	(43.4)	21	(41.2)	2	(28.6)	8	(53.3)	7	(31.8)	8	(50.0)
No	30	(56.6)	26	(51.0)	4	(57.1)	5	(33.3)	13	(59.1)	8	(50.0)
Problems with Naloxone Dispensing Listed on Survey
No problems	32	(55.2)
I need more training to educate patients on naloxone	9	(15.5)
Not all insurance covers naloxone	7	(12.1)
The device was too complicated	3	(5.2)
Incorrect formulation sent from provider	3	(5.2)
I’m not comfortable discussing overdose with pain patients	2	(3.4)
Educating patients about naloxone should be done by someone else	2	(3.4)
Patients reacted poorly	1	(1.7)
Filling the naloxone prescription takes too much time	1	(1.7)
I don’t think patients should be prescribed naloxone	1	(1.7)
Educating patients about naloxone is too time consuming	0	0.0

^†Percentages are column-calculated

^‡These problems were pre-defined and specifically asked about as part of the survey.

^*p<0.05 in bivariable analyses

Over half of the respondents (53.7%) had not received education around dispensing naloxone; nonetheless, a majority of the sample (61.1%) reported feeling comfortable providing naloxone education to patients. Among the respondents not comfortable providing education to patients, 57% said this was due to lack of training, 10% said hospital policy indicated that providers conduct the naloxone trainings, the remaining respondents did not provide explanation. Fifty-two percent of respondents reported providing naloxone education to at least one patient. Nearly half of respondents (43.4%) indicated that they would want authority to furnish naloxone without a prescription.

Over half of respondents (55.2%) reported experiencing no problems with naloxone dispensing. The most widely noted problems included not having enough training to educate patients on naloxone (15.5%) and problems with insurance coverage (12.1%). No other problems were endorsed by more than 10% of pharmacists (Table 2). Twelve percent reported more than one problem dispensing naloxone, which was associated with not being comfortable educating patients about naloxone (p=0.03). Preference for authority to furnish naloxone without a prescription was not significantly associated with either number of years licensed to practice pharmacy (p=0.32) or problems dispensing naloxone (p=0.69). In addition, there was no statistically significant association between the number of years licensed and the belief that naloxone should be made available to patients (p=0.30) (Table 2).

Discussion

With more than three quarters of pharmacists believing naloxone should be prescribed to patients on long-term opioids, results suggest high levels of pharmacist acceptability. Over half of pharmacists reporting feeling comfortable providing naloxone education to patients, which points to the feasibility of naloxone prescribing through primary clinics. The study also shows an interest in furnishing the product without a prescription among pharmacists. These results suggest a substantial interest in the topic, particularly because this study was conducted prior to the implementation of California Assembly Bill 1535 in April 2015 which authorized pharmacists to furnish naloxone without a prescription.

The most frequently reported problems with dispensing naloxone were not having received enough training to educate patients on naloxone and issues with insurance coverage. The prior study in Australia found similar concerns about insufficient knowledge, an issue that should be partially resolved as FDA approved formulations, with appropriate patient counseling information, have emerged. These new formulations should require less extensive education by providers or pharmacists, as the medication comes with standard educational materials. Nonetheless, training in opioid overdose prevention and management may help pharmacists to more effectively dispense, or furnish without a prescription, naloxone. With regard to insurance coverage, California Medicaid covers naloxone as a fee-for-service medication, which means managed care Medicaid plans do not cover the drug. Prior authorization requests sent to managed care health plans could have potentially led to delays in service.

This study has limitations. First, this was a convenience sample and survey respondents may have been more favorable regarding naloxone dispensing than non-respondents. Second, due to the nature of the parent study’s implementation design, some pharmacists that were surveyed may not have been contacted by the study pharmacist, and – because the study pharmacist generally communicated with only one person – not all pharmacists had received education on the topic. Third, surveys were self-administered which resulted in incomplete surveys and missing data. Future studies that include an education component may want to consider a more systematized means of educating pharmacists. Fourth, the small sample size used in this study is a limitation, as it restricted the statistical analyses we were able to perform. Furthermore, the survey was brief and certain questions were limited. This may have prevented respondents from fully expanding upon their experiences and interests in naloxone dispensing. For future data collection efforts, we suggest using a template that is more extensive and that includes more open-ended responses. Finally, pharmacists serving a population of safety-net patients where naloxone is actively dispensed may interact with patients potentially at risk of overdose more frequently and may be more favorable to a naloxone intervention, limiting the generalizability of results. Moreover, the sample of pharmacists was gathered within the scope of the parent naloxone co-prescription intervention study where a study pharmacist contacted community pharmacists to alert and educate them about naloxone prescriptions. This may have further influenced pharmacist acceptability of naloxone prescribing. Overall, these observations highlight the need for studies of pharmacist acceptability of naloxone dispensing in other pharmacy and geographical settings.

In summary, this survey of pharmacists at sites actively dispensing naloxone found high levels of acceptability for providing naloxone and substantial interest in furnishing the product without a prescription. The most significant barrier was lack of knowledge to educate patients, a problem at least partially resolved with the approval of naloxone products intended for lay usage and containing appropriate patient education materials. Future efforts should focus on increasing education for pharmacists around naloxone dispensing and patient education and should assess different education techniques. Furnishing naloxone is an important element of pharmacists’ efforts to respond to the opioid crisis.