Pharmacists’ readiness to provide naloxone in community pharmacies in West Virginia

Abstract

Objectives

The objective of this study is to assess West Virginia (WV) pharmacists’ stocking and dispensing practices of opioid-related medications and to identify the educational needs relating to providing naloxone in community pharmacies.

Design

A cross-sectional, anonymous, 49-item survey was created and validated to assess the educational needs of WV community pharmacists.

Setting and participants

The data collection instrument was administered to 266 pharmacists currently licensed in WV at six continuing education events throughout the state from March 1^st, 2016 to June 15^th, 2016.

Outcome measures

Pharmacists’ educational needs were determined using the Extended Parallel Process Model, which has four main constructs: perceived severity, perceived susceptibility, response efficacy, and self-efficacy. Pharmacists’ stocking and dispensing of opioids and related medications were also assessed.

Results

Pharmacists completed 157 surveys. They were mostly male (56.1%), full-time employees (67.5%), worked mostly in community pharmacies (69.4%), and had a mean age of 50.19 years (SD=13.62). The newly-adapted opioid perceived efficacy and perceived severity of opioid adverse events scales were tested for reliability and validity. Only 20.4% of the community pharmacists surveyed felt comfortable selling naloxone over-the-counter. As for the other opioid-related medications, only 53.3% stocked buprenorphine and 74.8% stocked buprenorphine/naloxone.

Conclusion

As the most accessible health care providers, community pharmacists are acutely aware of how the opioid epidemic affects their communities. Some pharmacists in WV are hesitant to stock and dispense opioids and opioid-dependence medications. Although this may decrease the flow of potentially abused drugs into the community, it may also restrict access to necessary therapy from patients with opioid use disorder. Furthermore, pharmacists in WV are not yet comfortable stocking and dispensing naloxone. Tailored educational materials can help in controlling the pharmacists’ fear and reinforce the benefits of over-the-counter naloxone use.

Introduction

Opioid misuse and abuse is a health crisis in the United States with 1.9 million Americans abusing prescription opioids in 2014.^1-4 The picture becomes more grim for those living in Appalachia – a cultural region stretching from southern New York to the top of Mississippi.⁵ Appalachia includes all of West Virginia (WV) and parts of twelve other states. In 2015, WV had the most deaths due to drug overdoses with 35.5 deaths per 100,000 inhabitants – twice the national average.⁴ Due to its reliance on community pharmacies to provide access to healthcare professionals and its current high levels of opioid abuse and associated high death rate, WV was chosen as the state to evaluate the capacity of community pharmacies to provide the potentially life-saving drug, naloxone.^4,6

Community pharmacists are the most widely available healthcare professionals⁷ and play a key role in community-based harm reduction strategies to prevent unintentional injuries and reduce the number of deaths related to opioid overdose.^8-11 Community pharmacies throughout the country have taken on naloxone distribution as part of the public health mission to prevent opioid overdose related deaths.⁸ To maximize pharmacists’ effort in this regard states have expanded pharmacists’ legal abilities to furnish naloxone.^8,12-15

Community pharmacists are the gatekeepers to prescription opioids as well as medications used to treat opioid use disorder. As providing naloxone within the community pharmacy is a voluntary act pharmacists’ stocking and dispensing practices must be evaluated to determine the real-world availability of these medications.^16,17 Previous literature suggested that pharmacists’ stocking and dispensing behaviors can be influenced by their perceptions of opioid abuse in their communities.¹⁷ Wright et.al. measured the level of concern about prescription drug abuse in the community among healthcare providers in Indiana. Out of the licensed pharmacists who responded, more than 80% of them dispensed fewer controlled substances, usually out of concern for community opioid abuse.¹⁷

Community pharmacists have been shown to worry about community opioid abuse, so to help mitigate these concerns, provider education must be made available to help pharmacists do their jobs more effectively.¹⁸ Education has been identified as a key component in curbing opioid abuse in the United States.^3,18,19 Educational materials need to be formulated with a validated framework to affect change in certain outcomes. One such framework, the Extended Parallel Process Model (EPPM) has been used extensively to create public communication campaigns by appealing to the individual’s desire to control either danger or fear.^20,21 EPPM has four key constructs which need to be measured in order to categorize individuals into four groups with unique educational goals: (1) perceived severity, (2) perceived susceptibility, (3) response efficacy, and (4) self-efficacy.^20-22 Perceived severity and perceived susceptibility are considered threat variables, and response efficacy and self-efficacy are efficacy variables. By measuring these constructs, one of four unique educational strategies can be identified that best match with the pharmacists (see Table 1). Evaluating the perceptions and actions of the community pharmacists being tasked with providing naloxone in community pharmacies has been identified as an important next step by clinicians and promulgated by the American Pharmacists Association.¹³

Table 1.

Individualized educational strategies based on constructs measured as part of the Extended Parallel Process Model.²²

	High Efficacy: High belief in the effectiveness of the intervention and their ability to provide the intervention (Response efficacy and Self-efficacy)	Low Efficacy: Low belief in the effectiveness of the intervention and their ability to provide the intervention
High Risk: High belief that the threat is harmful (Perceived severity and Perceived susceptibility)	Pharmacists believe that community naloxone use is highly effective and the risk in their community is high. Education needs to provide a call to action and an infrastructure to deliver the intervention.	Pharmacists believe that their communities are at risk, but do not believe that they are capable providing naloxone or in the effectiveness of its use. Education needs to focus on their ability to act and the role they can play.
Low Risk: Low belief that the threat is harmful	Pharmacists believe in the effectiveness of the intervention, but are not convinced the risk of opioid death in their community warrants their involvement. Education needs to focus on the risk of opioid misuse and abuse in their communities.	Pharmacists believe that their communities are not at risk and they are unable to provide the intervention. Education will need to focus on both their ability to act and the risks in their community.

With this in mind, the current study was guided by the leading research question surrounding pharmacist behavior and practices: what is the community pharmacists’ readiness regarding stocking and dispensing of opioid-related medications in WV? Based on the EPPM, it was expected that the methods for conferring education to WV pharmacists will need to be individualized to certain groups falling into the aforementioned categories in Table 1. With the recent passage of a law allowing pharmacists to potentially act as first responders and naloxone providers, effective educational materials are needed quickly.²³

Objectives

The objective of this study is to identify the educational needs relating to over-the-counter naloxone and to assess WV pharmacists’ stocking and dispensing practices of opioid-related medications.

Methods

A cross-sectional, anonymous, 49 item survey was created to assess the educational needs of community pharmacists in WV (see full survey in the Appendix). Content validity and comprehension of the survey was assessed by four pharmacists and revised based on their feedback. The survey items were separated into four sections. The first section collected pharmacists’ demographics and practice characteristics. Demographics included age and gender. Practice characteristics included licensure status, licensure state, primary employment county, hours worked per week, position, year of first licensure, and type of pharmacy of primary employment. The categories for licensure decade, types of pharmacy employment, and position were incorporated from the 2014 National Pharmacist Workforce Survey.²⁴

The second section contained 26 items including an approximation of opioid-related medication processing, and a group of items which used a five-point Likert scales ranging from 1 representing “strongly disagree” to 5 representing “strongly agree. Those items included opioid perceived efficacy scale (six items), buprenorphine and buprenorphine/naloxone efficacy (three items), opioid adverse event scale (six items), naloxone self-efficacy (one item, 5-point Likert scale), response efficacy (one item), and other opioid-related medication perceptions (6 items). The two opioid scales were adapted from the Clinicians’ Attitudes and beliefs about Opioids Survey (CAOS) perceived effectiveness and impediments/concerns subscales²⁵ for use with pharmacists. Despite the perceived effectiveness of opioids and the impediment/concerns items being valid content areas for pharmacists, the CAOS instrument was originally developed for physicians.²⁵ To assure the appropriateness of these items for pharmacists, validity and reliability testing were performed in the pharmacist sample. The mean scale scores, scale statistics, and the individual items are listed in Table 3.

Section three contained 13 items and was only to be completed by community pharmacists. This section focused on stocking practices (two items), new prescriptions processed (one item), dispensing practices (eight items), and estimated community opioid-related medication misuse/abuse (2 items). The final section contained one open-ended question allowing all pharmacists to expand on their thoughts about opioid use and prescribing in their county.

Participants were recruited to complete an anonymous paper version of the survey while attending one of six live continuing pharmacy education (CPE) events throughout WV from March 1^st, 2016 through June 15^th, 2016. Eligible participants were pharmacists currently licensed in WV. Participants left completed surveys in a collection box in the back of the room to maintain anonymity. The project was approved by the West Virginia University Institutional Review Board.

Data Analyses

Data were manually entered into an Excel spreadsheet by two of the researchers. To reduce the possibility of an incorrectly entered value, limits were placed on each cell. Data were validated to identify gaps or inaccuracies in the information by looking for outliers. Principal Component Analyses and Cronbach’s alpha statistics were assessed for the two newly adapted scales (opioid perceived efficacy scale and perceived severity of opioid adverse event scale). Principal component analysis (PCA) is a data reduction technique that selects a subset of variables based on correlation or covariance (validity). Cronbach’s alpha is a measure of internal consistency of the scale (reliability). The threshold values of Principal Component Analysis and Cronbach’s alpha were set at 0.4 and 0.7, respectively. The minimum sample size of 50 respondents for scale development was determined using the ratio of 10 respondents per item.^26,27

Descriptive (e.g., frequencies, means, and standard deviations) and inferential statistics were used to describe the data from this cross-sectional research survey. Chi-square analysis was used to examine the differences in demographics between all respondents and the community pharmacist subgroup. This comparison was used to validate the two new scales for all pharmacists. All statistical analyses assumed a significance level of alpha = 0.05. The study data were analyzed using Statistical Package for the Social Sciences (version 24; SPSS, Chicago).

Results

At the six live CPE events, 266 individual pharmacists were in attendance which resulted in 160 returned surveys (60.2% response rate). Three surveys were incomplete which resulted in 59.0% (n=157) usable response rate. The respondents were mostly male (56.1%), full-time employees (67.5%), worked mostly in community pharmacies (69.4%), and had a mean age of 50.19 years (SD=13.62). West Virginia was the primary state of licensure for 80.3% of the respondents. The community pharmacist subgroup used for this study’s analysis and is compared to all respondents in Table 2. There were no significant differences in demographics between all respondents and the community pharmacist subgroup. Pharmacists in other practice settings (hospital, long-term care/home health, retired, or other) were not used in the analysis since the objective was to provide educational materials for community pharmacists.

Table 2.

Demographics for all pharmacists (n=157) and community pharmacist subgroup (n=109).

		All pharmacists (n=157)	Community pharmacists only (n = 109)
Age, Mean (SD)		50.19 (13.62)	48.8 (12.98)
	Missing	2 (1.3%)	2 (1.8%)
Employment status, n (%)
	Full-time	106 (67.5%)	79 (72.5%)
	Part-time	45 (28.7%)	27 (24.8%)
	Missing	6 (3.8%)	3 (2.8%)
Employment position, n (%)
	Owner or part-owner	19 (12.1%)	17 (15.6%)
	Management (includes pharmacist-in-charge)	40 (25.5%)	30 (27.5%)
	Staff	93 (59.2%)	60 (55.0%)
	Missing	5 (3.2%)	2 (1.8%)
Licensed decade, n (%)
	1970 and before	12 (7.6%)	8 (7.3%)
	1971 to 1980	32 (20.4%)	16 (14.7%)
	1981 to 1990	35 (22.3%)	27 (24.8%)
	1991 to 2000	38 (24.2%)	29 (26.6%)
	2001 to 2010	24 (15.3%)	19 (17.4%)
	2011 or later	16 (10.2%)	10 (9.2%)
Gender, n (%)
	Female	69 (43.9%)	48 (44.0%)
	Male	88 (56.1%)	61 (56.0%)

The opioid perceived efficacy and perceived severity of opioid adverse events scale items with mean, standard deviation, component score, and overall Cronbach’s alpha (for each scale) are presented in Table 3. Dimensionality and reliability of the scales were acceptable since both scales loaded onto one component and had Cronbach’s alpha scores over 0.7. Overall, the opioid adverse event scale which represents the perceived severity construct had a mean of 22.3 (SD=3.72).

Table 3.

Scale development for community pharmacists

Opioid Perceived Efficacy Scale Items (n=106)		Mean*	SD	Component 1	Cronbach’s α+
	Opioids are most effective treatments available for persistent pain.	3.01	0.96	0.71	0.742
	In general, opioids are effective for nociceptive pain.	3.33	0.93	0.70
	In general, opioids are effective for mixed pain.	3.27	0.88	0.80
	In general, opioids are effective for neuropathic pain.	2.35	0.87	0.60
	Opioids are effective in controlling chronic non- cancer pain.	3.34	0.93	0.69
	Scale Mean Score	15.30	3.21	-	-
Perceived Severity of Opioid Adverse Event Scale Items (n=105)		Mean*	SD	Component 1	Cronbach’s α+
	Physical dependence is an impediment to taking opioids for long periods of time.	3.72	0.87	0.74	0.787
	Tolerance is an impediment to taking opioids for long periods of time.	3.75	0.84	0.67
	Addiction is an impediment to taking opioids for long periods of time.	3.89	0.84	0.69
	Cognitive functioning side effects are an impediment to taking opioids for long periods of time.	3.66	0.86	0.77
	Taking opioids for long periods of time will decrease their efficacy.	3.80	0.90	0.59
	Long-term opioids are an impediment to physical functioning.	3.50	1.02	0.72
	Scale Mean Score	22.32	3.72	-	-

^*Notes- : Mean scores were estimated using the five-point Likert scale which ranged from 1 representing “strongly disagree” to 5 representing “strongly agree.”

⁺: Cronbach’s alpha (α) is an estimate of how closely a set of survey items are as a group. This value can also be used to assess the reliability of a psychometric test or scale in a survey.

Next, the other three constructs in the EPPM in addition to items related to naloxone preparedness and perceptions of the opioid problem in the community were assessed. Only 20.4% of the community pharmacists felt comfortable selling naloxone over-the-counter (self-efficacy), but 72.5% disagreed with the statement that naloxone is not effective (response efficacy). For perceived susceptibility, the pharmacists estimated that 22.9% (SD=20.7%) of individuals filling prescriptions for opioid medications were misusing or abusing opioids. Most pharmacists (84.4%) also agreed or strongly agreed that opioids were over-prescribed in their counties, they were helping to curb opioid diversion (71.5%), but 42.6% agreed that they were also harming some patients by denying care. After reverse coding the item, 71.3% agreed that letting patients purchase naloxone over-the-counter will increase opioid overdosing. These items are summarized in Table 4.

Table 4.

Community pharmacists’ naloxone preparedness, perceptions, and estimated community misuse/abuse (n=109).

	Strongly Disagree	Disagree	Neither Agree nor Disagree	Agree	Strongly Agree
Naloxone outcomes
Self-efficacy: I am comfortable selling naloxone over the counter in my pharmacy. (n=108)	17 (15.7%)	37 (34.3%)	32 (29.6%)	20 (18.5%)	2 (1.9%)
Response efficacy: Using naloxone for opioid overdose isn’t effective.	23 (21.1%)	56 (51.4%)	24 (22.0%)	6 (5.5%)	0 (0.0%)
Prepared: I do not feel I am adequately trained in the use of naloxone over the counter.	2 (1.8%)	9 (8.3%)	14 (12.8%)	50 (45.9%)	34 (31.2%)
Worsen problem: Letting patients purchase naloxone over the counter will increase opioid overdosing. (n=108)	7 (6.5%)	24 (22.2%)	36 (33.3%)	30 (27.8%)	11 (10.2%)
Perceptions
Opioids are being overprescribed by practitioners in my county.	1 (0.9%)	4 (3.7%)	12 (11.0%)	48 (44.0%)	44 (40.4%)
Pharmacists are curbing opioid diversion and/or abuse by declining to fill some prescriptions for opioids.	0 (0.0%)	12 (11.0%)	19 (17.4%)	65 (59.6%)	13 (11.9%)
Pharmacists are harming some patients who have legitimate pain issues by declining to fill some prescriptions for opioids. (n=108)	3 (2.8%)	32 (29.6%)	27 (25.0%)	43 (39.8%)	3 (2.8%)
Misuse, Mean (SD)
For every 100 patients in your pharmacy who FILL PRESCRIPTIONS for opioid analgesics excluding buprenorphine/naloxone and buprenorphine, how many patients do you estimate misuse or abuse prescription opioids? (n=103)					22.9 (20.7)
For every 100 patients in your pharmacy who FILL PRESCRIPTIONS for buprenorphine/naloxone and buprenorphine, how many patients do you estimate misuse or abuse them? (n=82, excluding those who did not stock)					23.6 (26.8)

All pharmacists participating in the study stocked opioids in their pharmacies; however only 53.3% stocked buprenorphine and 74.8% stocked buprenorphine/naloxone. The vast majority of pharmacists reported stocking the same or higher quantities of all three medication types this year compared to the previous year (89.8% for opioids, 83.1% for buprenorphine, and 88.9% for buprenorphine/naloxone). The comparative stocking was only considered if the pharmacist marked that their pharmacy stocked the medication of interest. Nearly all (89.6%) of the community pharmacists declined to fill an opioid prescription at least 1-2 times per week, but that was only the case 58.1% and 60.8% of the time for buprenorphine and buprenorphine/naloxone, respectively. The two geographic factors assessed (out-of-state prescribing and out of local area prescribing) were significantly associated with whether (or not) the pharmacist would fill the prescriptions for opioids. The results for each item are provided in Table 5.

Table 5.

Stocking and dispensing outcomes by community pharmacists (n=109).

Do you order and stock the following products for your primary employment site?
	Yes	No
Opioid analgesics (n=107)	107 (100%)	0 (0.0%)
Buprenorphine (n=107)	57 (53.3%)	50 (46.7%)
Buprenorphine/naloxone (n=107)	80 (74.8%)	27 (25.2%)
Have you ever DECLINED to fill a prescription for a patient for the following products?
	Rarely, if ever	≥ 1-2 times per week
Opioid analgesics (n=106)	11 (10.4%)	95 (89.6%)
Buprenorphine (n=93)	39 (41.9%)	54 (58.1%)
Buprenorphine/naloxone (n=97)	38 (39.2%)	59 (60.8%)
How likely are you to fill a prescription for the following products written by an out-of-state practitioner?
	Not at all likely	Somewhat likely or higher
Opioid analgesics (n=108)	40 (37.0%)	68 (63.0%)
Buprenorphine (n=102)	75 (73.5%)	27 (26.5%
Buprenorphine/naloxone (n=103)	66 (64.1%)	37 (35.9%)
How likely are you to fill a prescription for the following products for a patient who does not live within your pharmacy’s local area (e.g. patient lives outside a 20-mile radius of your pharmacy)?
	Not at all likely	Somewhat likely or higher
Opioid analgesics (n=107)	60 (56.1%)	47 (43.9%)
Buprenorphine (n=102)	84 (82.4%)	18 (17.6%)
Buprenorphine/naloxone (n=103)	77 (74.8%)	26 (25.2%)

In the past year, has your pharmacy stocked more, less, or about the same quantity of the following products compared to the previous year?
	More	Less	About the same
Opioid analgesics (n=108)	38 (35.2%)	11 (10.2%)	59 (54.6%)
Buprenorphine (n=59, excluding those who did not stock)	24 (40.7%)	10 (16.9%)	25 (42.4%)
Buprenorphine/naloxone (n=81, excluding those who did not stock)	46 (56.8%)	9 (11.1%)	26 (32.1%)

Discussion

Only one fifth of the pharmacists surveyed are comfortable selling naloxone over-the-counter in their pharmacy and three quarters of the respondents do not feel they are adequately trained in the use of naloxone. There is clearly an educational need regarding naloxone use in WV pharmacists. By allowing well-trained pharmacists to dispense naloxone over-the-counter, many of the patients’ barriers to accessing naloxone in the community⁸ can be removed. However, the access to naloxone will only increase if the pharmacists choose to be trained¹⁵ and to stock the naloxone products.

All community pharmacists indicated their pharmacies stocked opioid medications; however, the availability of medications for opioid use disorder (buprenorphine and buprenorphine/naloxone) is less than opioids themselves. Pharmacists are trusted healthcare providers, trained in the safe and effective uses of medications. This trust and relative autonomy in stocking and dispensing decisions emphasizes the importance of community pharmacists in combating community public health crises like opioid abuse. This research provides insight into the practices and attitudes which drive those practices for WV’s pharmacists. It is not readily apparent why pharmacists would not be comfortable providing naloxone over-the-counter. However, the huge range of estimated opioid misuse or abuse and the high percentage of pharmacists who believe letting patients purchase naloxone over-the-counter will increase overdosing do not help to alleviate this stress. These results indicate that research into their tailored educational needs is required to optimize naloxone’s availability and use in the state.

The community pharmacists surveyed had an above average perceived severity and perceived efficacy scores as measured by the newly adapted and validated scales. Both of these constructs make up the threat variables of the EPPM, and represent the perceived threat relating to opioid abuse in their community. Self-efficacy and response efficacy make up the efficacy variables in the EPPM. Both of these items had low scores indicating that the community pharmacists participating believe that their ability to practice effective naloxone distribution in the community is low.

Community pharmacists also had high threat and low efficacy levels according to the EPPM. Based on the description in Table 1, pharmacists believe that their communities are at risk (high threat constructs), but do not believe that they are currently capable of providing naloxone or in the effectiveness of its use in the community (low efficacy constructs). These needs are in stark contrast to the currently mandated²³ educational materials provided by the WV Office of Emergency Medical Services. The training materials were created for first responders, are not pharmacist-specific, and emphasize emergency use of naloxone rather than the provision of it in the community for future use.²⁸ Based on this model, educational needs should focus on controlling pharmacists’ fear in order to increase pharmacists’ ability to act and emphasize the role they can play to help patients in WV.

With the exponential growth in opioid prescribing and high death rate due to opioid overdoses in WV, community pharmacists are in a unique position to bring about change as the epidemic continues to be battled. One important aspect of care that community pharmacists can influence is access to medications used to treat opioid use disorder, specifically buprenorphine, buprenorphine/naloxone, and naloxone. All three of these medications have proven efficacy in the treatment of opioid use disorder.^29-31 If access to these medications is not provided, there are disparities in the availability of effective treatments for opioid use disorder. The risk of opioid-related injury and death is shared by many stakeholders including patients, prescribers, and pharmacists.⁸

Limitations

There were potential limitations to this study. First, survey results of pharmacists from one state may not be generalizable to pharmacists in other states. However, these results are being used to develop educational materials tailored to the needs of pharmacists in WV based on the EPPM. Second, the validity of the results may be affected by the usual limitations of self-report questionnaires and thus may not fully reflect the respondents’ beliefs, attitudes, or actual practices. Another potential limitation is that a convenience sample of pharmacists was used for this survey. The demographics of the non-responders at the CPE events could not be compared to those who did complete the survey. Future research that utilizes a sampling frame could assess this non-response bias.

Conclusion

The risk for opioid-related injury and death has steadily and markedly increased over the past decade as a result of a confluence of factors. As the most accessible health care providers, community pharmacists, are acutely aware of how this epidemic affects their communities. Some pharmacists in WV are hesitating to stock and dispense opioids and opioid-dependence medications. Although this may decrease the flow of potentially abused drugs into the community, it may also restrict access of necessary therapy from patients with opioid use disorder.

Furthermore, pharmacists in WV are not yet comfortable stocking and dispensing naloxone. Tailored educational materials can be helpful to control the fear and reinforce the benefits of over-the-counter naloxone use. Ongoing research will allow for the results to be demographically weighted to improve generalizability so that statewide education can be implemented quickly and effectively.

Key points.

Background

• Opioid prescribing, misuse, and abuse has increased significantly over the past two decades to become a health crisis in the United States.

• West Virginia leads the nation with 35.5 deaths due to prescription drug overdose per 100,000 inhabitants – twice the national average.

• Community pharmacists can have a great individual responsibility and control over what medications are stocked and dispensed in their pharmacies.

Findings

• Few community pharmacists felt comfortable selling naloxone without a physician’s prescription.

• Based on the Extended Parallel Process Model, tailored educational materials need to focus on controlling the pharmacists’ fear and reinforcing the benefits of increased naloxone availability through community pharmacies.