Community Pharmacists

We adopted a similar comprehensive strategy applied in Bangladesh (Haque et al., 2020a (link)), and subsequently used among other Asian countries (Godman et al., 2020a (link); Haque et al., 2020b (link)), building on previous knowledge of activities across Africa just before and after the first cases of COVID-19 were identified in Africa (Ogunleye et al., 2020 (link)). This included a questionnaire survey among community pharmacies in Africa including Ghana, Kenya, Namibia, Malawi, Nigeria, and Zambia as well as among three comparative Asian countries including Bangladesh, Pakistan and Vietnam. Box 1 summarises the open ended questions building on published studies (Haque et al., 2020a (link); Haque et al., 2020b (link)). Impressions were requested if this was the only information available due to issues of confidentiality and culture as no payment was made to pharmacists for the information provided (Godman et al., 2020a (link); Haque et al., 2020a (link)). The objective was to assess the current situation regarding usage patterns, prices, and availability of selected medicines and PPE used in prevention and management of COVID-19 soon after the start of the pandemic. The baseline was early 2020, i.e., just before active preventative measures in a number of the countries. More specific data on actual changes in utilisation and prices was asked if this was available; however, this did not include asking the pharmacists to break down changes in utilisation patterns and prices per month as this was deemed too problematic for this initial study. In addition, it was envisaged there would be limited impact of seasonality in view of year-round influenza activity in some African countries and no real pattern in others (WHO, 2012 ; Dawa et al., 2020 (link); WHO Regional Office for Africa, 2020 ) although research is ongoing in this area to address current knowledge gaps (Sambala et al., 2018 ). In addition, generally limited use of hydroxychloroquine in immunological diseases such as rheumatoid arthritis with concerns over its effectiveness compared with other disease modifying therapies (Rempenault et al., 2020 (link)). A more detailed description of the questionnaire can be found in Haque et al. (2020a) (link).
We did not factor inflation into the impressions as the study period only covered a short time, and per annum inflation in the chosen African countries typically ranged between 6 and 16% per annum (Focus Economics, 2020 ; Trading Economics, 2020a ; Trading Economics, 2020b ; Trading Economics, 2020c ; Trading Economics, 2020d ; Trading Economics, 2020e ).
The Asian countries chosen for comparative purposes had instigated similar activities to prevent the spread of COVID-19 as those seen in Africa and where there can be high patient co-payments for medicines (Godman et al., 2020a (link)).
Convenience sampling was again used to select pharmacists through emails, telephone contact, personal contacts and other mechanisms similar to the studies across Asia (Godman et al., 2020a (link); Haque et al., 2020a (link)). As before, there was no sample size calculation as there was no previous data to base calculations upon at the start of the study and this was a pilot study with a minimum of six pharmacies contacted in all countries apart from Namibia to determine the need for additional follow-up studies.
The findings were subsequently consolidated into categories in a tabular format to aid comparisons between regions and countries, with more specific data available in country publications (Haque et al., 2020a (link); Haque et al., 2020b (link)). We believed that there would be price rises and shortages in other countries apart from Bangladesh. However, the nature and extent would depend on ongoing programmes within the country.
We also explored the situation regarding the preparedness of community pharmacists in Namibia to the pandemic to help enhance future guidance to Governments and other key stakeholder groups. This included questions on i) Key measures/interventions the pharmacy has put in place during the pandemic to curb the spread of COVID-19 in the community (maximum of three from a pre-arranged list of seven known activities); ii) Suggestions on the role (current and new) of pharmacists/pharmaceutical technicians/pharmacist assistants during current and future pandemics (up to three from a pre-arranged list of five potential activities); iii) The main challenges experienced by pharmacy personnel during the pandemic (maximum of three from a pre-arranged list of seven known activities); iv) Changes in utilization, prices and shortages of pertinent medicines and PPE used in the prevention and management of COVID-19 from the beginning of March to end June 2020 (Supplementary Appendix A1). The study was extended to the end of June to provide greater insight.
Potential future guidance for governments, pharmacists and patients will build on the experiences of the pharmacists and others involved in the study in Namibia and across all the studied countries, the co-authors and previous suggestions documented in Ogunleye et al. (2020) (link).
Ethical approval for this study was not required according to our national legislation and institutional guidelines. However, as before in Bangladesh, all pharmacists freely provided the requested information having been given the opportunity to refuse to participate. This is in line with previous studies undertaken by the co-authors in this and related areas including analysis of policies to enhance the use of biosimilars and the rationale use of medicines, pricing policies as well as issues surrounding shortages and generics, which typically involved direct contact with health authority personnel and other key stakeholders (Godman et al., 2014 (link); Godman et al., 2015 (link); Moorkens et al., 2017 (link); Godman et al., 2019 (link); Gad et al., 2020 (link); Godman et al., 2020a (link); Godman et al., 2020b (link); Godman et al., 2020c (link); Godman et al., 2020d (link); Haque et al., 2020a (link); Miljković et al., 2020a (link)).

We adopted a similar multiple strategy approach to the paper of Haque et al. discussing activities in Bangladesh (4 (link)). This included updated information from a pragmatic review of the literature combined with a questionnaire survey among community pharmacies and drug store owners in India, Malaysia, Pakistan, and Vietnam, building on the findings in Bangladesh, to assess the current situation regarding usage patterns, prices, and availability of carefully selected medicines that could potentially be used in the management of COVID-19, as well as PPE in most countries, soon after the start of the pandemic (4 (link)). We also included herbal medicines in Pakistan and Vietnam as we are aware of their use in patients with COVID-19 among a number of LMICs despite concerns (209 (link), 210 (link)). Medicines included pertinent antimalarials such as hydroxychloroquine, antibiotics such as azithromycin, analgesics (general including paracetamol), vitamins and immune boosters such as vitamin C as well as PPE including face masks (Box 1). We just asked for impressions in the first instance for changes in utilization, prices, and shortages, from early March to end of May 2020 if this was the only information available due to issues of confidentiality (4 (link)). The baseline was early 2020, i.e., just before active preventative measures in a number of the countries (Table 1). More specific data on actual changes in utilization and prices was asked if this was available; however, this did not include asking the pharmacists to break down any changes in utilization patterns and prices per month as this was deemed too problematic for this initial study.
Convenience sampling was used to select pharmacists through emails, telephone contact, personal contacts and other mechanisms. Similar to the initial study in Bangladesh (4 (link)), there was no sample size calculation as there was no previous data to base calculations upon at the start of the study. In addition, the studies undertaken in Malaysia, Pakistan, and Vietnam were pilot studies to help determine the need for additional studies. All questions were again open ended with data captured on Excel spreadsheets (Box 1). A more detailed description of the questionnaire can be found in Haque et al. (4 (link)). The replies from the community pharmacists were collated where possible into logical bands for comparisons between countries including the initial analysis from Bangladesh (4 (link)). These bands were not pre-defined as this was an exploratory study, with changes in prices based on local prices and not converted into a single currency such as US dollars using current exchange rates since ascertaining actual prices was not an objective of this study. The changes in utilization and prices were absolute changes during the time period of the study. Suggested strategies going forward for all key stakeholder groups also builds on the combined experiences of the co-authors. We have successfully used this approach before to provide future direction in LMICs (4 (link), 125 (link), 211 (link)–215 (link)).
We believed that there would be price rises and shortages in other countries apart from Bangladesh. However, the nature and extent would depend on ongoing programmes within the country (Table 2) including greater price controls in India and Pakistan (79 , 105 (link)).
Ethical approval for this study was not required according to national legislation and institutional guidelines. However, as before, all pharmacists freely provided the requested information having been given the opportunity to refuse to participate if wished. This is in line with previous studies undertaken by the co-authors in this and related areas including analysis of policies to enhance the rationale use of medicines and biosimilars, pricing policies and issues surrounding generics, which involved direct contact with health authority personnel and other key stakeholders (4 (link), 125 (link), 212 (link), 216 (link)–219 (link)).

This research has employed a multi-phase sampling technique. First, a cluster sampling method was used where five regions in Saudi Arabia i.e., Eastern Province, Riyadh, Asir, Madinah, and Northern Borders were selected for data collection as each belong to a different part of the country. Potential participants who visited the community pharmacy were invited to take part in the study and fill out the self-administered questionnaire through five community pharmacists working in the different above-mentioned regions. Second, in attempts to increase the response rate a convenience sampling method was used which involved advertising the questionnaire link through social media platforms, including WhatsApp and Telegram groups in Saudi Arabia.
The minimum recommended sample size was estimated to be 385 participants based on a population size of 36,000,000 persons (10 (link)), with a 5% margin of error, a 95% confidence level and a 50% response distribution.

The questionnaire was adapted from previous research (1 (link), 5 (link), 9 (link)) and it involved a total of four sections. Section One collected sociodemographic information, i.e., age group, gender, nationality, marital status and educational level. Section Two involved questions related to the use of the community pharmacy and it assessed participants' satisfaction with the community pharmacy services in Saudi Arabia. Section Three evaluated participants' willingness to use different pre-determined community pharmacy services. Section Four gathered data on barriers that would prevent patients from approaching a community pharmacist for help or advice.
The original version of the questionnaire was prepared in the English language, and then translated into Arabic. In order to ensure the validity of the translation, a back translation technique was undertaken by the study investigator and two translators who have excellent proficiency in both languages.
In order to ensure the clarity of the questions, the questionnaire was pilot-tested with seven participants who met the inclusion criteria of the study. Modifications were made on the questionnaire based on the pilot test feedback. The data from the pilot study were excluded from the final results. The final questionnaire was distributed in the Arabic language, which is the official language of Saudi Arabia.
The questionnaire was created using the Google forms platform. An introductory statement inviting eligible subjects to participate in the study, and highlighting the aim of the study and the inclusion criteria, was added to the survey link. The cover page of the questionnaire displayed a participant information sheet with detailed information about the conduct of the study.

An online survey of community pharmacists (e.g., big-box, chain, independent, or grocery pharmacies) in the State of Connecticut was conducted between July 29–September 29, 2020.
Four-hundred respondents were sought for participation as a convenience sample. Respondents included pharmacists licensed in Connecticut who attested to (1) having practiced on average at least 20 hours a week after March 1, 2020, in a brick-and-mortar community pharmacy physically located in Connecticut; and (2) having regular and physical face-to-face interactions with the public in their provision of pharmacy services. No other requirements, such as time licensed or degree earned were required. Pharmacists working remotely, in medical marijuana dispensaries, or in other non-community pharmacy settings were excluded from the study. This study was reviewed and approved by the University of Connecticut Institutional Review Board. Participants' responses were collected anonymously.
Survey items that measured perceptions of confidence were developed from domains in the Emergency Risk Communication (ERC) framework.⁷ (link) Developed in 2017 by Savoia, Lin, and Gamhewage, the ERC was chosen as this study's conceptual framework as it serves to guide the development of studies that assess communication outcomes related to public health emergencies. Domains of the ERC are broadly categorized into three groups, including (1) information environment, (2) populations, and (3) public health systems. Respective to each domain, three correlating items questioned respondents' confidence in (1) proving information to the public, (2) managing their own and others' mental health, and (3) performing point of care testing (POCT) mitigation. A fourth item sought to explore if stress (i.e., “risk perception, emotions and trust” in the ERC) related to pharmacists' avoidance of risk communication (i.e., “knowledge, attitudes and practices” in the ERC).
Respondents were also asked questions from the Perceived Stress Scale (PSS-10), developed by Cohen et al. in 1983.⁸ (link) In the PSS-10, respondents were asked how often they feel a certain way on a five-point scale from ‘never’ (1) to ‘very often’ (5). To calculate a total PSS-10 score, responses to the positively stated items were reversed and then summed across all items. Higher scores indicate higher levels of perceived stress. Other items included a mix of multiple-choice and Likert scales; free-text boxes were included without forced response for the participant to provide comments or clarification.
Two community pharmacists matching the inclusion/exclusion criteria beta-tested the survey for clarity and content before distribution; their data were not included in the analysis.