Reported heavy workloads and a lack of dedicated time to integrate naloxone screening into the pharmacy workflow was cited by 6 of 8 respondents as making it difficult to prioritize furnishing naloxone. One respondent noted, “It’s really time. We don’t really have time here to initiate for those implementing naloxone […] unless patients request it.” . With respect to stigma, 5 respondents stated that it was difficult to suggest supplying naloxone to patients due to its association with drug abuse. They indicated that patients perceived offers of naloxone as accusations of opioid abuse. One stated, “[T]here’s always like that lash back from a patient, like oh, I don’t need it because I’m not abusing it. That’s the common phrase.” . Lastly, one respondent reported that the absence of a shared language was a barrier due to lack of understanding and miscommunication, noting that, “[W]e have to get a translator to […] communicate with the patients. Maybe the patient’s not understanding correctly even [as] it’s being translated.” . Examples of responses regarding barriers to furnishing are shown in Table 3. Although California Assembly Bill No. 2760, which passed in 2018, required medical prescribers to offer a naloxone or equivalent prescription to populations at higher risk of opioid overdose, out-of-pocket costs to the uninsured in the United rose 500 percent from 2014 to 2018 for certain brands of naloxone.Generic naloxone has an average wholesale price of $64.80-$75.00.Participants believed that reducing out-of-pocket costs could increase naloxone purchase and use. A chain community pharmacist stated that they went “out of their way to try to find GoodRx discount cards to help bring down the price for patients.” and that doing so reduced patient reluctance to purchase naloxone. Similarly, an independent community pharmacist suggested that “lower [ing] restrictions and mak[ing] it OTC [over the counter]” would increase the likelihood of visitors purchasing naloxone. Participants were also asked to identify facilitators to furnishing.
Responses included collaborating with other health professionals, closer proximity to pain clinics,cannabis grow supplies expanded scope of pharmacy practice in California, supportive corporatepolicies, education and training on naloxone furnishing, and higher demand for naloxone. With respect to collaboration, 2 respondents stated that closer proximity to pain clinics increased the likelihood of pharmacies furnishing naloxone. One pharmacist stated that, “some pharmacies are located in regions [with] higher potential [of] abuse …that can also drive up having … more Naloxone in that location.” . One pharmacist indicated that demand was higher in their region, stating that, “people started asking for it. We dispensed it” . Additional examples of responses regarding facilitators to furnishing are provided in Table 4.Since 2013, California has sought to expand access to care by authorizing pharmacists to furnish medications. Implementing naloxone furnishing by pharmacists in particular provides a potential opportunity to reduce opioid overdoses. These services are especially critical in rural areas like California’s Central Valley that have been disproportionately impacted by the opioid epidemic.We were unable to identify any prior studies that assessed the extent of pharmacist furnishing in rural, HPSAs such as the Central Valley, and our findings suggest that contrary to initial expectations, almost half of contacted pharmacies, including some mail-order pharmacies, furnished naloxone in the Central Valley. In contrast, a study of primarily urban pharmacies in California conducted in 2020 found that 42.4% furnished naloxone.Interviews with pharmacists who furnished naloxone suggested that pharmacies continued to face barriers to successful implementation, many of which have been identified in previous research. These included time restrictions, high out of-pocket costs for purchasers, stigma associated with opioid use, and in 1 case, language barriers. All respondents indicated that out-of-pocket costs were the most critical barrier and that prices varied depending on insurance coverage; this finding is consistent with prior research.
The findings regarding stigma as a barrier to offering and accepting naloxone are also consistent with previous research. This includes a study involving pharmacy students in Tennessee and their perceptions of naloxone use and opioid use disorder patients, which found that although pharmacy students are capable of and predisposed to furnish naloxone, successful furnishing is complicated by limited patient awareness and stigma, specifically the perception that naloxone is for “addicts” only.Another study examining undergraduates’ reactions to fictional vignettes about people with opioid use disorder found addiction was attributed to the opioid user’s character and varied by an user’s socioeconomic status.Studies examining perceptions of take-home naloxone conducted with both healthcare providers and opioid users have found that stigma influences both parties when providing education and seeking out information about naloxone and overdose prevention, respectively.These studies suggest that further interventions in pharmacy education to combat stigma against naloxone use and opioid use disorder might help facilitate increasing naloxone furnishing rates.Limitations to this study include generalizability, variable effects of coronavirus disease 2019, and self-reporting bias. The analysis only considered the 11 counties in the Central Valley, which may limit extrapolation outside of this region. The sample also did not include interview data from pharmacies that furnished but chose not to participate, which may have resulted in a biased sample. Another limitation is that this study was conducted 2 years after the most recent comparison study of naloxone furnishing in California. As a result, the higher furnishing rates observed in this study may have re- flected a time trend or effect of the coronavirus disease 2019 pandemic, such as difficulty securing appointments with physicians encouraging use of pharmacy services, rather than a difference in prevalence.One interview was done through e-mail, rather than a phone call, which limited the ability to probe for clarification and additional detail.
Pharmacies that did not furnish naloxone were not included in interviews on the grounds that they would be unable to provide information on facilitators to naloxone furnishing; future studies could investigate if these pharmacies furnish other medications. Additional research could also address potential differences in furnishing practices between independent and chain pharmacies, as well as furnishing rates for other medications in this region. Irrespective of these limitations, the findings provide new information regarding pharmacist furnishing in HPSAs, barriers that prevent the widespread provision of naloxone, and potential strategies that may help overcome those barriers.The American College of Obstetrics and Gynecologists and the American Academy of Pediatrics recommend avoiding cannabidiol and CBD-containing products during breastfeeding because of potential neurodevelopmental risks to the infant. Cannabidiol use is widespread and increasing among adults, especially for medical purposes. However, information on CBD risk to the breastfed infant is largely unknown owing to limited and variable existing evidence. To support the strength of recommendations, more knowledge is required on the dose–exposure–response relationship of CBD in breastfed infants. Such information would lead to a better understanding of whether observed CBD concentrations in milk consumed by infants lead to relevant systemic concentrations in breastfed infants associated with neurodevelopmental delays . Beginning in 2014, Mommy’s Milk Human Milk Biorepository investigators sought to improve the understanding of maternal exposure to various agents, including marijuana and its metabolites, during breastfeeding and the potential for infant exposure to specific agents and subsequent adverse infant outcomes. The HMB is a USAand Canada-wide study that collects human milk samples from mothers who were or were not taking medications and recreational drugs,grow lights for cannabis including marijuana, CBD, and CBDcontaining products. The HMB investigators continue to study breastfeeding exposures and potential infant outcomes through administration of neurodevelopmental questionnaires and face-to-face testing. In the present secondary data analysis, we seek to fill a gap by further defining CBD exposures to breastfed infants. In this work, we leverage real-world CBD concentrations in breastmilk from the HMB, knowledge of breast milk intake as a function of infant age, dose and type of administration, and physiologically based pharmacokinetic models to translate CBD dose through breastfeeding into neonatalexposures. Physiologically based pharmacokinetic modeling is a mathematical tool used to predict drug exposures based on the physicochemical properties of a compound, and the anatomy and physiology of organisms. We sought to answer, among breastfeeding mothers taking CBD based on real-world use, what is the predicted exposure and its associated variability in breastfed infants? The open-source PBPK modeling platform, PK-Sim version 11 was used to perform PBPK modeling. Plot Digitizer version 2.6.8 was used to digitize published pharmacokinetic profiles to obtain concentration–time data. R was used to curate the HMB dataset, analyze subgroups, and simulate infant daily doses.
The HMB was established in 2014 at the University of California San Diego for research purposes. The HMB collects voluntary human milk samples from lactating women who are or are not exposed to any medication, recreational drug, or environmental chemical primarily in the 2 weeks prior to sample collection. Detailed information on recruitment, data collection, and sample preparation and analysis methods have been presented previously. Participants complete an interview to provide their demographics, maternal and child health history, breastfeeding habits, and all exposures focused in the previous 2 weeks prior to sample collection. Exposure information from women who reported marijuana use at any time since giving birth included type of administration, frequency of use, dose, and time since last use before milk sample collection. Milk samples were previously measured for metabolites, including CBD concentrations and the date and time of the milk collection were ascertained. Cannabidiol concentrations in human milk were determined by liquid chromatography with tandem mass spectrometry. The analytical range of the assay was 0.1–200 ng/mL. The method was validated in human milk by establishing the accuracy and precision of three sets of calibration curves and quality-control samples over 3 days. Acceptability criteria for accuracy was ±15% of nominal concentrations except ±20% at the lower limit of quantification . Acceptability criteria for precision was ±15% coefficient of variation, except ±20% coefficient of variation at LLOQ. This present study received ethics clearance from the parent study through the UC San Diego Human Research Protections Program, and for a secondary data analysis through the University of Waterloo Research Ethics Board . Information on maternal demographics, exposures, and measured CBD concentrations in milk collected and assayed by the parent study between 2015 and 2021 were extracted from the existing HMB dataset. The dataset was organized to describe: all concentrations in milk ; and concentrations by self-reported maternal frequency, dose, and type of administration . From the existing data for the sample on quantification of CBD, three methods were assessed to account for below the limit of quantification values: BLQ = LLOQ/2, BLQ are drawn from uniform distributions of 0 to LLOQ, and BLQ = LLOQ. For concentrations from samples with a maternal reported type of administration , only concentrations with one type of maternal administration were retained. Missing end time of exposure was replaced with the time of concentration sample collection and vice versa. The effects of administration type, time after last dose , and dose frequency on concentration were assessed. Administration type was a categorical variable defined as: edible, joint, oil, pipe, or other , and N/A categories. As a continuous variable, TAD was described as time in hours elapsed from the end of maternal administration to milk sample collection for concentration measurement. Time after last dose was calculated by subtracting the date and time of sample collection by date and time of the last reported date of maternal administration. To account for the varying ways in which dose and frequency of CBD and CBD-containing products were consumed , dose frequency was categorized as low, medium, and high based on the data of each week-normalized dose type. To compare these subgroups, the exposure-concentration subset was considered with and without BLQ values. A linear regression model to predict log-concentrations was obtained after testing the significance of subgroups on CBD in milk concentrations including TAD, administration type, and interactions between TAD and administration type, and administration type and dose frequency. Model goodness of fit was evaluated through a standard residual analysis. Post-hoc pairwise comparisons of estimated marginal means of the significant subgroups were performed using various p value adjustment methods owing to the lack of a gold standard method. The pediatric PBPK model was developed according to the workflow of Maharaj et al.. An adult oral CBD PBPK model established from our previous work was scaled to simulate CBD exposure in virtual breastfeeding infants.