Recruitment strategies for community-based pharmacy research

Abstract

Background

Rural communities are disproportionately burdened by opioid use disorder (OUD) and other opioid-related harms. Community pharmacies are an accessible health care touchpoint in rural areas and are well-positioned to provide prevention and treatment services. However, there is limited guidance on how to effectively engage rural pharmacies and patients in OUD research. Our objective is to describe research recruitment strategies for pharmacists and patients.

Methods

We engaged a rural practice-based research network to recruit pharmacists and patients for surveys and interviews aimed at improving understanding of opioid overdose prevention and treatment services. To maximize survey completion and ensure diverse participation in interviews, we used a multi-modal strategy designed for pharmacists, including email, personalized survey QR codes with mailed gift baskets, and phone calls. Patient recruitment was pharmacist mediated.

Results

With 78 of 150 pharmacists (52%) completing the survey, we met our response goal of 50%. Completed surveys varied by recruitment method, including 51 (65%) from email outreach and 19 (24%) from personalized QR codes with mailed gift baskets. Fifty-four of the 78 pharmacists who completed the survey (69%) expressed interest in the interview and purposive sampling resulted in 16 completed interviews. A total of 73 patients were interested in interviews. Of those, 46 were fully screened, 45 met eligibility, and 20 completed interviews.

Conclusion

Successful recruitment for rural community pharmacists and patients for opioid-related research requires intentional, relationship-centered strategies. Multi-modal, low-burden contact improved rural pharmacist recruitment and pharmacists were key intermediaries for patient recruitment.

1. Introduction

Opioid use disorder (OUD) and opioid misuse remain persistent and prevalent public health challenges in the United States (US). In 2024, an estimated 4.8 million people aged 12 and older met criteria for OUD and 7.8 million misused opioids during the prior year.¹ The burden of opioid-related harm is especially challenging in the southeastern US. Centers for Disease Control and Prevention (CDC) provisional drug overdose mortality data suggest that US Region 4 (Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, and Tennessee) has, since 2020, consistently experienced the highest number of various drug overdose deaths among all US regions.² Many of the counties in this region are rural, where communities carry a disproportionate burden of opioid-related harms due to limited health care access, including emergency services and primary care.³ Despite these access gaps, 68% of rural residents live within 10 miles of a pharmacy,⁴ positioning pharmacies as one of the few consistently accessible health care touchpoints in these areas.

Community pharmacies are well-positioned to deliver naloxone, buprenorphine, and opioid risk reduction services.⁵ Additional research with pharmacies and their patients could improve the availability and delivery of these services, especially in rural communities. Yet, conducting research in rural areas can be challenging, especially for topics such as OUD, which is highly stigmatized in rural communities.6, 7 Existing research demonstrates that community pharmacy research engagement requires trusted relationships, constituent involvement, clear value to pharmacies and patients, workflow-sensitive procedures, and substantial resources for recruitment and retention.8, 9, 10, 11, 12, 13 However, less is known about how these principles can be operationalized in rural community pharmacies to facilitate recruitment of pharmacists and patients into research studies.14, 15, 16 There remains a need for guidance on effective recruitment strategies for engaging rural populations in stigmatized research topics. This work aims to address this gap by highlighting a partnership with the Rural Research Alliance of Community Pharmacies (RURAL-CP), a seven state practice-based research network in the southeastern US.¹⁷ The purpose of this paper is to identify which recruitment strategies yielded the greatest engagement of rural community pharmacists and patients in opioid research using this practice-based network.

2. Methods

2.1. Study design and setting

This research was part of a larger, multi-method study that included an online pharmacist survey and semi-structured interviews with pharmacists and patients aiming to 1) assess community pharmacists' knowledge, attitudes, and practices regarding opioid overdose prevention and treatment services and 2) identify barriers to pharmacy-based opioid overdose prevention and treatment services. The study was deemed exempt by the Institutional Review Board (IRB) at the University of North Carolina at Chapel Hill (IRB #: 23–2576).

All pharmacies included in this study were members of RURAL-CP which operates in the southeastern US (Alabama, Arkansas, Georgia, Mississippi, North Carolina, South Carolina, and Tennessee).¹⁷ For RURAL-CP, rurality is defined using US Department of Agriculture Rural–Urban Commuting Area (RUCA) codes of 4 or higher, where higher RUCA codes indicate greater rurality.17, 18. However, given the limitations of using RUCA codes to identify rural areas,¹⁹ RURAL-CP also evaluates pharmacies for inclusion on a case-by-case basis using alternative rural criteria, including a combination of the following: the town has a population of less than 10,000, the county has a rural population greater than 50%, the town has fewer than four pharmacies, the pharmacy is located 30 min or more from the nearest city, or the area has a rural or underserved designation. Only four of the 150 active pharmacies in RURAL-CP had RUCA codes less than four at the time of this study.

When pharmacies join the RURAL-CP network, pharmacists are asked to share the pharmacy's email address so study notifications and reminders can be sent to those emails. Based on our experience, emails sometimes bounce back and pharmacists do not routinely check email while providing patient care. Therefore, the research network started collecting phone numbers for each pharmacy; however, these phone numbers were a mix of direct lines to the pharmacy or personal mobile numbers of a team member. The communication database for the network did not specify the type of phone number collected.

2.2. Participant inclusion criteria

Pharmacists were included in the study if they worked full-time at a RURAL-CP pharmacy. Patients were eligible if they were at least 18 years of age, spoke English, used a pharmacy in a rural community, and were currently taking chronic opioid medications or were caretakers of someone who was currently taking or previously had taken chronic opioids.

2.3. Recruitment procedures

2.3.1. Pharmacist survey

We used a multi-modal, escalating contact strategy intentionally designed for rural pharmacists who often experience a high workload burden and inconsistent access to email communication. The web-based survey was distributed in December 2024 via email using Qualtrics (Provo, UT) to the 150 RURAL-CP pharmacies, with the goal of having one pharmacist from 75 pharmacies complete the survey. The survey invitation described the purpose of the study, provided a survey link, and offered a $50 gift card upon completion. It also informed pharmacists about an optional follow-up interview with an additional $50 incentive.

A modified Dillman approach was used to optimize response rates.²⁰ The first email reminder was sent in late January 2025 (approximately one month after the initial invite). A second email reminder was sent in mid-February. When the response rate remained below goal, the survey deadline was extended with a new incentive for survey completion. An incentive basket that included North Carolina snack products, valued at $47.99 (tax exempt with free shipping), was mailed to all RURAL-CP pharmacies in late February 2025. The basket included a cover letter and a personalized QR code for immediate survey access. A third reminder email was sent during the same week, alerting pharmacists to expect the mailing. Members of the research team called survey non-responders during the first week of March to confirm receipt of the mailed basket, troubleshoot survey access, and encourage participation before survey closure. Survey data collection concluded in March 2025.

2.3.2. Pharmacist interviews

At the end of the survey, pharmacists were asked whether they wanted to participate in a follow-up qualitative interview. Pharmacists who agreed completed an interview contact form. Their demographic information from the form including pharmacy state, years in practice, frequency of naloxone and buprenorphine dispensing as well as information compiled by the study team (e.g. pharmacy rurality based on RUCA code and the percentage of Black/African American residents in the pharmacy's county based on 2020 Census data²¹), were used to guide purposive sampling. The goal was to obtain at least two pharmacists from each of the seven RURAL-CP states, representing a range of rurality levels, practice experience, and county demographic characteristics. Effort was made to include pharmacists practicing in counties with higher proportions of Black/African American residents, as these communities are often underrepresented in pharmacy research. Outreach was prioritized first among states, followed by years of practice, then RUCA codes, and finally percentage of Black/African American residents. These priorities were established before recruitment began. The sampling approach remained flexible, allowing the study team to adapt recruitment decisions as needed to maximize variation and ensure representation across the interview sample.

Phase 1 of recruitment for pharmacist interviews began in January 2025 before survey data collection was complete. Prioritized outreach using demographic information occurred via phone calls to the number provided by the RURAL-CP network communication database, followed by an email containing a link to a 60-min Zoom interview with two members of the study team if the pharmacist was interested in participation. Follow-up emails and phone calls were made to reschedule as needed. Phase 2 of recruitment began in April 2025. As in Phase 1, pharmacists were first contacted via phone calls. If the pharmacist agreed to participate in the interview, they were asked to confirm or provide their mobile number and consent to receive a text reminder prior to their scheduled interview. Recruitment continued until our target of two pharmacists per state was met. A thank you email was sent in June 2025 to pharmacists who had already completed the interview to encourage additional distribution of flyers informing patients of the opportunity to participate in study interviews. Interviews concluded in August 2025, and all participants received a $50 gift card.

2.3.3. Patient interviews

Patient recruitment was intentionally pharmacist-mediated to leverage the trust pharmacists hold in rural communities²² and to accommodate the sensitive nature of research involving opioid use. A patient-friendly flyer was created that provided a brief introduction to the study, the inclusion criteria, and compensation details. This process had been used in previous studies which also engaged RURAL-CP.¹⁴ Pharmacy personnel were instructed to place flyers inside the prescription bags for patients picking up an opioid or buprenorphine medication and/or display the flyers prominently in their pharmacy. The flyer included a Google Voice phone number and an email address to allow patients to contact the study team directly via call, text, or email. Fifty flyers were initially mailed to each pharmacy in the network.

Interested patients contacted the study team member listed on the flyer. A structured screening protocol was used to confirm eligibility, including purposive screening to ensure the inclusion of patients from a variety of states, demographic and experiential backgrounds. Screening questions assessed: how the patient learned about the study (to confirm receipt of the flyer or seeing it in their pharmacy); whether the patient or someone close to them currently or previously used opioids; whether the patient used a pharmacy in a rural community and if so, the name and zip code of their pharmacy (to confirm rurality); whether anyone had already reached out to them about participating in the study; and basic demographic characteristics, including age, gender, and race/ethnicity. Responses were recorded in a secure database along with patient contact information and relevant notes. Sampling priorities were established prior to recruitment and focused on ensuring representation across states, demographic backgrounds, and lived experiences. We sought to include Black/African American patients and patients residing in communities with more rural RUCA codes, as these populations have historically been underrepresented in research and may face unique barriers to accessing opioid-related services. The sampling approach remained flexible throughout patient engagement, with outreach decisions adapted as needed to maximize diversity across the interview sample.

2.3.4. Expanded screening for diversity of patient experience

After the first two months of patient interviews, additional screening questions were added to improve representation and ensure inclusion of diverse patient experiences. The new questions assessed prior attempts to access naloxone, buprenorphine, or syringes in a rural pharmacy; perceived barriers to accessing these services (e.g., stigma, cost, availability); and experiences with bias, stigma, or discrimination in their community. These additional questions helped identify participants with varied encounters with opioid-related services. Patients meeting inclusion criteria were scheduled for a 40-min, semi-structured interview conducted via Zoom or telephone depending on their internet access capabilities. Patients received a reminder text or phone call the day before or the morning of the interview. Interviews were conducted through the Spring of 2025 and concluded in May once thematic saturation was reached.

2.4. Data analysis

The goal of this study was to achieve at least a 50% survey response rate, consistent with previous studies involving RURAL-CP pharmacists.23, 24, 25 Demographic information from the survey data guided the purposive sampling strategy for pharmacist interviews. The target number of patient qualitative interviews was informed by the principle of thematic saturation, which research suggests can typically be achieved with 9–17 interviews.²⁶ Patient data were analyzed in real-time using a constant comparative method. Recruitment continued until saturation was reached and sufficient variation across patient characteristics was obtained.

3. Results

3.1. Pharmacist survey

Of the 150 pharmacists, 78 completed the survey (response rate 52%).²⁷ Results from survey items examining rural community pharmacists' knowledge and practices have previously been published.²⁷ Nineteen surveys were completed following the initial email invitation, 21 were completed after the 1-month reminder email, and 11 additional were completed after the 7-week reminder email (Fig. 1). Therefore, 65% of all survey responses were received through the modified Dillman method. The remaining 27 were completed after the third reminder. This reminder included the mailed incentive basket with the personalized QR code and a follow-up phone call to nonrespondents. (Fig. 1). Overall, 19 pharmacists completed the survey via the personalized QR code and 59 completed the survey via the emailed link.

Fig. 1.

Cumulative Pharmacist Survey Responses by Recruitment Event.

3.2. Pharmacist interviews

More than two-thirds (n = 54, 69%) of the 78 pharmacists who completed the survey expressed interest in participating in a follow-up interview. Interested participants represented all seven RURAL-CP states, including 4 from AL, 13 from AR, 13 from GA, 6 from MS, 9 from NC, 3 from SC, and 6 from TN. During initial outreach (Phase 1), three pharmacists per state (21 total) were contacted. Of these, 15 interviews were scheduled, three pharmacists declined participation, and three pharmacists did not respond to outreach attempts. All pharmacists who declined to participate cited time constraints as their reason for not participating. Among the 15 scheduled interviews, five pharmacists did not attend and were classified as lost to follow-up.

After completing 10 interviews, all states were represented, and outreach continued to achieve the recruitment goal of at least two pharmacists per state. During Phase 2 of outreach, seven additional pharmacists were scheduled for interviews. One pharmacist did not attend and was considered lost to follow-up. Participation rates improved with the incorporation of text message reminders, with no-shows reduced from 33% to 14%. Over the whole recruitment period, 16 interviews were completed. Fig. 2 displays recruitment progress across contact, scheduling, and completed interviews.

Fig. 2.

Pharmacist Interview Recruitment.

From the 16 complete interviews, there were three pharmacists interviewed from both AL and NC and two pharmacists from each of the remaining five states. All pharmacists' practice sites were in areas with a RUCA code higher than four. Of the 16 interviewed pharmacists, five had 26 or more years in practice, six had 11 to 25 years, and the remaining five pharmacists had 10 or fewer years. Twelve pharmacists practiced in counties with a Black/African American population of 0-25%, three practiced in areas with a Black/African American population of 26-50%, and one practiced in an area with a 51-65% Black/African American population.

3.3. Patient interviews

A total of 73 patients contacted the study team to express interest in participating in the interview. Purposive sampling yielded 46 patients who were fully screened, of which 45 met eligibility criteria. One participant was ineligible because they had no history of opioid use. Three patients were from pharmacies that did not have a pharmacist participate in a survey or an interview, but saw the flyer. The eligible patients were recruited from 20 rural community pharmacies.

Twenty-three patient interviews were scheduled as determined by the purposive sampling. Two patients did not show and could not be rescheduled. One interview ended early due to limited engagement with the questions, yielding 20 completed patient interviews. These 20 participants represented 14 rural community pharmacies in six states (3 AL, 3 AR, 3 GA, 1 MS, 6 NC, 4 TN), with pharmacy ZIP codes corresponding to a RUCA code of 4 or higher. There remained 27 patients who expressed interest but were not needed for the interview. Our team communicated with each patient to let them know that we reached our recruitment goal, thank them for their interest and encourage their participation in future research. Fig. 3 summarizes patient recruitment and participation.

Fig. 3.

Patient Recruitment and Participation Summary.

4. Discussion

This study demonstrated that recruitment of rural pharmacists and their patients into opioid-related research is feasible using intentional relationship-centered strategies tailored to community pharmacies. Findings reinforced three takeaways: practice-based research networks (PBRNs) are facilitators for research engagement; multi-modal, low-burden contact improves engagement; and pharmacists are key facilitators for patient recruitment.

4.1. Practice-based research networks as key facilitators for engagement

The primary facilitator for this study's success was leveraging the RURAL-CP practice-based research network. RURAL-CP provided an established mechanism for engaging pharmacies in research and adapting interventions to rural contexts. Because the Director and Program Manager of RURAL-CP were study team members with a longstanding relationship with member pharmacists, rural pharmacists likely viewed this study as credible, relevant, and low-risk. Community pharmacies are increasingly recognized as important partners in practice-based and implementation research. Existing literature suggests that successful engagement requires constituent-informed research priorities, value to pharmacies, attention to workflow and resource constraints, active communication and support, and infrastructure such as PBRNs.8, 9, 28, 29 Our findings build on emerging evidence that engaging community pharmacies in research is itself a resource-intensive implementation process. Stolte et al. found that recruitment through pharmacy PBRNs required repeated communication, substantial time, and dedicated resources, with pharmacists' perceptions of the value of the intervention for both the pharmacy and its patients influencing recruitment and retention.¹¹ These findings are consistent with our experience that successful pharmacy engagement requires more than identifying eligible sites. Our findings extend this literature by examining these considerations in the context of rural community pharmacies and patient recruitment, highlighting the potential role of pharmacies as trusted, community-based partners in reaching rural populations.

The effectiveness of RURAL-CP was reflected in patient engagement. Rural residents have been identified as a hard-to-reach population,³⁰ and previous research has illuminated numerous reasons why recruitment and participation of rural residents can be challenging, including unique cultural/social aspects of rural settings, distrust of outside researchers, and lack of anonymity.¹⁴ Therefore, researchers looking to engage rural residents often need to employ unique strategies. Pharmacists in our study were willing to distribute recruitment materials (e.g. flyers) to eligible patients and display study materials prominently, but these behaviors would be unlikely without an existing trusted relationship with researchers. In addition, many of these community pharmacies participated in patient recruitment even if their pharmacists did not participate in the survey or interview themselves. This further demonstrates that a rural PBRN is a facilitator for research since the pharmacists helped with the research recruitment even though they did not feel like they had time to participate.

Despite using a PBRN, challenges still emerged. Instability in community pharmacy settings occurred due to pharmacy openings and closures³¹ and pharmacy staff turnover. This required additional time to verify mailing addresses and, at some pharmacies, the correct pharmacist-in-charge. Prior to this study, RURAL-CP updated their communication database on an annual basis. During this study, the team realized this frequency needs to be at least twice a year or quarterly due to staff turnover. We discovered that we need consistency in obtaining and labeling mobile numbers and direct pharmacy numbers. We also realized that mail delivery options are limited in rural areas and depending on what is being mailed (i.e., letters versus packages), the address may be different. Community-based researchers should confirm the preferred mailing address for the type of delivery that is being sent. We discovered that the gift baskets required a different mailing address (e.g. street address or post office box) compared to a mailed letter. The communication database should include all types of mailing addresses. Researchers should also confirm the best point of contact for the pharmacy to receive information. In some cases, it may be a pharmacy owner, manager, or lead technician.

Future studies should be mindful of the impacts of pharmacy business operations and staffing turnover on recruitment efforts and engagement. Maintaining a database that is frequently updated is essential for community-based research to reduce verification burden. Researchers should also verify contact information prior to the start of any new study.

4.2. Multi-modal, low-burden outreach improves rural pharmacist engagement

Rural community pharmacies face substantial workload pressures, including staffing shortages,³² increased workload,33, 34 and financial threats,³⁵ making research participation difficult. We found that using multiple, low-burden contact methods meaningfully increased engagement.

Mailed gift basket incentives with personalized QR codes generated over a third of the total number of survey completions, with 70% of those responses occurring through QR code access. To our knowledge, no existing studies involving community pharmacists have documented the use of personalized QR codes with gift baskets as an incentive. While gift baskets are uncommon and most survey studies involving community pharmacists offer gift card or gift card raffle incentives,23, 25, 35, 36, 37 our finding aligns with evidence that tangible mailed materials, such as a “survey kit”, can drastically improve response rates (80.9%).³⁸ The novelty of receiving mailed recruitment materials with a gift may have helped this study rise above the “noise” of crowded email inboxes.

Another low-burden outreach method used in this study was text messaging. To our knowledge, there are no published studies describing the use of text messaging to aid in the recruitment and retention of pharmacist participants. There is a limited body of evidence that shows text messaging has been used successfully as a low-cost, convenient strategy to facilitate contact and reduce attrition among patients; however, these studies did not have a pharmacy or a rural component.16, 39, 40, 41 This is the first study that we are aware of that demonstrates text message outreach boosted completion of pharmacist interviews by reaching those who seldom check email. Our no-show rates dropped from 33% to 14% after implementing text message reminders. Reminders were beneficial for busy pharmacists, as workload often distracts pharmacists from being aware of the current time or accidentally missing scheduled events. This study highlights that text messaging can be integrated into a trusted, pharmacy-based rural recruitment strategy.

Rural pharmacists' unpredictable workflow and competing responsibilities complicate scheduling of interviews. Therefore future studies should incorporate text-based communication early, routinely request mobile numbers, and design flexible scheduling processes such as including night and weekend times for interviews or an online calendar to reschedule if needed.

4.3. Pharmacists as key facilitators for patient recruitment in rural OUD research

In terms of patient recruitment for interviews, pharmacists played a critical role in connecting researchers with rural patients. Rural communities often hold mistrust toward research14, 16, 42 and experience stigma related to substance use.14, 43 However, pharmacists are highly trusted²² and previous research has found success recruiting rural patients through their pharmacists.¹⁴ By displaying flyers and placing them in patient prescription bags, pharmacists enabled high-yield, targeted recruitment, producing more than three times the number of interested patients needed for our study.

Furthermore, including the eligibility criteria on the flyer and relying on pharmacists to determine which patients should receive bag stuffers minimized ineligible inquiries. The burden of implementing bag stuffers on pharmacy staff was minimal, as pharmacists were asked to provide the flyer for either an opioid or buprenorphine prescription, therefore minimizing any workflow disruptions to the pharmacist. As a result, nearly all screened patients (45 of 46) were eligible, demonstrating the precision and appropriateness of pharmacist-mediated recruitment. Using a Google Voice number for participants offered anonymity and flexibility so that pharmacies did not provide patient information directly to the study team, which is especially valuable in rural settings where privacy concerns may be heightened.14, 16, 44, 45 By using a new, virtual phone number, Google Voice also provides anonymity for the research team so personal cell numbers do not need to be shared. The flexibility of this technology is in the ability to call or text, which is convenient for both participant and researcher, and texting may be less intimidating for some participants.

Although patient recruitment was successful, challenges remained. Conducting interviews required reliable internet or phone access, which presented a barrier for many rural participants. Indeed, limited access to technology, including reliable internet and cell phone service, has been widely cited as a barrier for rural research participants.14, 16, 46, 47 Additionally, restricting eligibility to English-speaking individuals further limited participation. In terms of operational constraints, a single study coordinator was responsible for managing a high volume of patient inquiries. This resulted in delayed and incomplete screening. Similarly, having only one qualified interviewer limited scheduling flexibility and interview availability. The study team also continued to receive inquiries from interested patients after study conclusion, suggesting both sustained interest and unmet research team capacity.

Future studies should anticipate high patient interest and allocate multiple staff for communication, screening, and interviewing. Incorporating bilingual study team members would also broaden participation and strengthen trust in linguistically diverse rural communities. Finally, flyers for future patient recruitment should include a study end date to reduce the chance of additional inquiries post-study.

4.4. Limitations

Because all participating pharmacies in this study were members of an established PBRN, our recruitment success may not be generalizable to all community pharmacies or pharmacies that are part of non-rural PBRNs. With regard to patient recruitment, our study achieved saturation; however, a limitation is that we did not monitor how consistently participating pharmacies distributed or displayed patient recruitment flyers. Consequently, the variation in pharmacy engagement may have influenced patient recruitment.

5. Conclusion

This study demonstrates the feasibility and value of engaging rural community pharmacists and patients in research on opioid overdose prevention and treatment services using recruitment methods that are relationship-centered, flexible, and tailored to community pharmacies. Leveraging an established PBRN, engaging in multi-modal, low-burden contact, and using pharmacists as trusted intermediaries were critical to achieving a 52% pharmacist survey response rate and achieving saturation for the patient interviews across a diverse geographic and demographic rural sample. These strategies offer a practical roadmap for future research addressing stigmatized issues in rural communities.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this work the authors used ChatGPT to perform light editing. After using this tool, the authors reviewed and edited the content as needed, and they take full responsibility for the content of the published article.

Funding acknowledgement

This publication is supported by the Food and Drug Administration (FDA) of the U.S. Department of Health and Human Services (HHS) as part of a Center of Excellence in Regulatory Science & Innovation award (U01FD007857) from the Center for Drug Evaluation and Research (CDER). The contents are those of the authors and do not necessarily represent the official views of, nor an endorsement by, FDA/HHS or the U.S. government.

CRediT authorship contribution statement

Abigail G. Shackley: Writing – review & editing, Writing – original draft, Project administration, Investigation, Data curation. Stefanie P. Ferreri: Writing – review & editing, Validation, Supervision, Methodology, Funding acquisition, Conceptualization. Jennifer M. Cavender: Writing – review & editing, Writing – original draft, Project administration, Investigation, Data curation. Kaileigh D. Yen: Writing – review & editing, Investigation, Data curation. Lori T. Armistead: Writing – review & editing, Validation, Investigation. Delesha M. Carpenter: Writing – review & editing, Validation, Methodology, Conceptualization. Tamera D. Hughes: Writing – review & editing, Visualization, Validation, Supervision, Methodology, Investigation, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.