Sage Journals: Discover world-class research

Abstract

Background

Artificial intelligence (AI) is rapidly reshaping healthcare, including pharmacy practice through drug interaction screening or treatment monitoring applications. In internet-restricted settings such as Syria, access to most AI-based platforms and digital tools is limited, which may hinder their adoption in community pharmacy practice. However, adoption of AI in such contexts remains largely unexplored.

Objective

To evaluate the level of AI adoption among community pharmacists in Syria and compare their familiarity and openness toward AI technologies with that of pharmacy students.

Methods

A total of 400 participants based in Syria were included: 200 community pharmacists and 200 pharmacy students. Data were collected through a combination of paper-based surveys (pharmacists) and online questionnaires (students). To enhance understanding, a five-slide presentation explaining basic AI concepts was made available to participants unfamiliar with the topic.

Results

Nearly half of respondents (55%; n = 219) reported using AI tools in daily life, with ChatGPT being the most recognized (53%, n = 211). Pharmacists demonstrated higher AI knowledge than students (5.7% vs. 2.5%; χ² (4, N = 400) = 19.39, p = .001). However, students showed significantly greater willingness toward integrating AI into future practice (χ² (2, N = 396) = 33.35, p < .001). Pharmacists with either less than two years or more than six years of experience were less likely to use AI, whereas those with limited prior Generative AI knowledge appeared more willing to adopt such tools compared with those with higher prior knowledge (OR = 9.0, 95% CI [1.36–59.78], p = .023).

Conclusion

Our findings revealed a clear coexistence of interest in and concern about AI among both community pharmacists and pharmacy students. The study underscores the importance of developing clear regulatory guidance, structured training, and context-appropriate oversight to ensure the safe integration of AI into pharmacy practice, particularly in internet-restricted settings.

Keywords

Artificial intelligence pharmacy students pharmacists Syria developing country ChatGPT

Introduction

Artificial intelligence (AI) refers to computer systems designed to perform tasks that typically require human intelligence, such as learning, pattern recognition, and decision-making, by analyzing large volumes of data and adapting their behavior accordingly.¹ In healthcare, AI adoption has grown rapidly, with the potential to revolutionize several disciplines, including precision diagnostics, drug discovery, surgical assistance, and clinical decision support, enhancing workflow efficiency and improving patient outcomes.^2–5

In the pharmaceutical sector, AI plays an important role due to its broad spectrum of potential applications, such as clinical decision support, adverse drug reaction detection, drug development, and personalized medication regimens.^6,7 Community pharmacists frequently analyze extensive data sets to pinpoint overlooked drug interactions, track clinical effectiveness and adverse events of medications, and manage drug substitutions. The integration of AI technologies provides them with tools that help make accurate, evidence-based clinical decisions by rapidly analyzing large amounts of patient-related information, such as medical records, laboratory results, and medication adherence patterns. This enables them to enhance their decision-making processes and deliver personalized care to patients.^6,7

One notable example of AI's impact is the integration of robotic dispensing systems, such as the one at UCSF Medical Center, which processed 3.5 million medication doses without errors. These systems automate routine tasks and empower pharmacists to shift their focus toward person-centered care, such as personalized consultations and the management of complex chronic conditions.⁷ Another example of AI's potential in pharmacy is the development of a deep learning model to address look-alike/sound-alike medication errors. The model achieved over 90% accuracy in identifying blister-packaged drugs. This AI solution can be integrated into hospital systems to automate drug verification, minimize dispensing errors, and optimize pharmacist productivity.⁸ Hybrid AI-driven clinical decision support system, combining machine learning (ML) and rule-based expert systems, was developed to identify high-risk prescriptions in hospitals. The system flagged 74% of prescription orders requiring pharmacist intervention and achieved 74% precision. This performance outperformed traditional tools and reduced alert fatigue, demonstrating AI's potential to enhance medication safety.⁹ While these AI applications improve workflow and safety, their integration into routine pharmacy practice remains limited.¹⁰ Therefore, AI tools require proper governance, transparency, and human oversight to mitigate risks such as misinformation, bias, and unsafe clinical recommendations.¹¹

A new generation of AI applications called generative AI is emerging in the form of conversational tools and chatbots. Generative AI refers to systems that produce human-like text or responses such as large language models including ChatGPT. These tools offer accessible and flexible solutions for answering patient and healthcare professional queries, automating routine tasks and supporting pharmacists in decision-making.^10,12

Pharmacy chatbots are broadly categorized into three types. The first type, rule-based chatbots, is designed to answer medication-related questions, such as drug interactions and side effects. They offer cost-effective and secure solutions but struggle with complex or ambiguous questions. The second type, Natural Language Processing (NLP)-based chatbots, is designed to mimic human conversations by interpreting the intent and context of user inputs. Unlike rule-based systems, they analyze word patterns to generate more flexible responses. The third type, ML-based chatbots, combines NLP with ML algorithms to learn from historical interactions, improving accuracy and effectiveness over time.^10,13

During the COVID-19 pandemic, hospital pharmacy chatbots alleviated staff burden by answering medication-related questions, while community pharmacy chatbots reduced patient calls and visits by providing mask updates. This automation allows pharmacy teams to shift focus toward complex tasks like medication reviews and clinical services, enhancing patient outcomes and satisfaction.^7,10

Medical chatbots intended for diagnosis, treatment, or disease mitigation may qualify as medical devices under current regulatory frameworks. Both the U.S. Food and Drug Administration (FDA) and the U.K. Medicines and Healthcare products Regulatory Agency (MHRA) apply a risk-based regulatory approach to AI-enabled clinical software, which may require registration, premarket review, and postmarket surveillance.^14,15 In contrast, most general conversational AI chatbots that do not make explicit diagnostic or therapeutic claims are not currently regulated as medical devices.¹⁶

However, recent regulatory discussions emphasize that large language model–based chatbots may require medical device approval when used for clinical decision-making or patient management. In the United Kingdom, the MHRA continues to advance its regulatory framework for AI in healthcare, recognizing that tools with a defined medical purpose should comply with established medical device requirements.¹⁵ Platforms such as PharmBot AI are being developed for use within regulated pharmacy environments (e.g. NHS workflows), however, to the best of our knowledge, they do not currently hold formal FDA or MHRA marketing authorization as licensed medical devices.¹⁷

Background: Pharmacy practice and digital health in Syria

Pharmaceutical services in Syria are delivered through both hospital-based and community pharmacies. Hospital pharmacies are mainly located within public healthcare facilities, whereas community pharmacies are predominantly privately owned and often represent the first point of contact for patients. In this role, community pharmacists provide advice on minor ailments, medication guidance, and basic health counseling. Compared with high-income countries, where pharmacists typically operate within well-structured clinical and regulatory frameworks, Syrian community pharmacists frequently assume broader frontline responsibilities due to limited healthcare infrastructure and resource constraints.^18–20 Prolonged conflict, economic sanctions, and financial instability have substantially weakened the Syrian healthcare system, resulting in significant structural and infrastructural challenges. These conditions have adversely affected both the availability and quality of pharmaceutical services across the country.^18,21

Health-related information technology in Syria remains underdeveloped. Most healthcare facilities continue to rely on paper-based records, with limited implementation of electronic health records, electronic prescribing systems, or integrated pharmacy information systems. The absence of unified digital platforms restricts data sharing and limits the adoption of advanced clinical decision support tools. Furthermore, internet connectivity is unstable, particularly in rural and conflict-affected regions. Access to digital health technologies and some widely known AI platforms is additionally constrained by infrastructure damage, local network restrictions, among other factors.^22–25

These infrastructural and technological limitations directly hinder the adoption of AI in pharmacy practice. In addition, many AI-based applications and training resources are primarily available in English, creating a language barrier for Arabic-speaking pharmacists and limiting exposure to AI technologies, including conversational AI tools. Although experimental Arabic-supporting models, such as MedicalBot for Levantine dialects, have been developed, most advanced AI tools still require English-language inputs due to dataset limitations.^23,26 To the best of our knowledge, Syria does not yet have specific national legislation regulating the use of AI in healthcare or pharmacy practice.

Although AI holds considerable potential to improve workflow efficiency, medication safety, and clinical decision-making, its current implementation in Syrian pharmacy practice remains minimal. Consequently, understanding pharmacists’ knowledge, attitudes, and perceptions toward AI is essential to assess readiness, identify barriers, and inform future strategies for effective integration.^23,27

Related work

Although AI demonstrates considerable potential in pharmacy practice, understanding pharmacists’ perspectives on this technology is essential for successful implementation. Several studies from the Middle East and other regions have explored pharmacists’ knowledge, attitudes, and practices (KAP) regarding AI. A study conducted in Jordan among pharmacists practicing in the Middle East reported that 44.8% of participants had a moderate level of knowledge about AI, while 49.1% expressed positive attitudes toward its potential applications in pharmacy practice.²⁸

Another study in Jordan specifically evaluated community pharmacists’ perceptions and willingness to integrate ChatGPT, a generative AI–based conversational tool, into pharmacy practice.²⁹ In healthcare settings, ChatGPT operates through question-answering mechanisms, responding to prompts ranging from open-ended queries to specialty-specific topics such as disease classification, radiology interpretation, and patient education.³⁰

In this study, nearly half of the pharmacists (48.4%) expressed a willingness to incorporate ChatGPT into their professional activities. However, more than 70% reported concerns regarding ChatGPT's inability to exercise human judgment or address complex ethical decision-making. Furthermore, Pharmacists with prior experience using ChatGPT were significantly more inclined to integrate it into practice than those without prior exposure.²⁹

A multinational cross-sectional study across six Arab countries identified key predictors of KAP regarding the use of AI in pharmacy practice. Higher income, advanced educational backgrounds, and prior exposure to AI or related technological innovations were associated with greater acceptance of AI integration.³¹ Similarly, a recent study from Saudi Arabia found that 68.9% of participants perceived benefits from widespread AI adoption.³² Furthermore, 84% agreed that AI could help reduce medication errors in clinical settings. Nonetheless, 12.8% of respondents expressed concerns that AI implementation might lead to job displacement.³²

Research has also focused on pharmacy students’ perceptions of AI and generative AI tools. A study conducted at Afe Babalola University in Ado-Ekiti, Nigeria reported that students demonstrated a good understanding of chat-based AI tools and held generally positive attitudes toward their use. The authors emphasized the importance of integrating AI education into pharmacy curricula to address knowledge gaps and prepare students for future technological advancements.³³

Aims and objectives

The primary aim of this study is to evaluate the use of AI tools among community pharmacists in Syria, alongside their knowledge of and attitudes toward these technologies. A secondary aim is to compare pharmacists’ usage and openness toward AI with that of pharmacy students, and to identify key barriers to its implementation in an internet-restricted context. In Syria, long-standing sanctions, restricted internet access, and limited exposure to AI-based platforms have created a unique context in which both awareness of and engagement with AI remain, to our knowledge, uneven and underexplored. This investigation helps inform education and policy for future advancements in AI within pharmacy.

Design and methods

Study design and participants

A cross-sectional study was conducted between September and November 2024 in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology checklist for observational studies. The study aimed to assess the adoption of AI tools and chatbots among community pharmacists and pharmacy students in Syria. The study also examined their willingness to integrate AI technologies into their daily professional and academic workflows. A total of 400 participants were included, equally divided between licensed community pharmacists and pharmacy students.

Sample size

The sample size was determined by selecting 15% of the total population of community pharmacies officially registered with the Syrian Pharmacists Association in Damascus, which amounts to 1300 pharmacies. A sample size of 200 participants was estimated, which is considered an acceptable percentage, especially given the field constraints in Syria during data collection period and the high response burden, this sample is proportionally representative and maintains acceptable statistical reliability.^34–36 To facilitate a meaningful group comparisons, an equal number of 200 pharmacy students was included, following a 1:1 ratio. Damascus University is considered the biggest and continuously functioning University in Syria during the time of data acquisition. Equal group sizes enhance the power of comparative tests (e.g. chi-square, t-tests) and improve the robustness of parameter estimates.

Inclusion–exclusion criteria and quality control

Participants were required to meet specific inclusion criteria: community pharmacists had to be based in Syria, graduates and actively working in a licensed pharmacy in Damascus, while pharmacy students had to be based in Syria and enrolled at a Syrian university. All participants provided responses to the survey after giving written informed consent. Individuals who did not meet these criteria were excluded from the study. To ensure data quality and reliability, several measures were implemented: (1) direct visits to pharmacies were conducted to collect data and to ensure that responses were provided by licensed pharmacists rather than unauthorized personnel. (2) A five-slide presentation in Arabic was developed to provide brief, neutral, introductory content to explain AI concepts and tools and was made available for all participants who could choose whether or not to view it. (3) The online survey platform included mandatory response fields to minimize missing data. Additionally, data cleaning procedures were applied to exclude invalid responses within individual questionnaires, particularly affecting specific items contradictions in reported AI use. Invalid responses accounted for fewer than 2% of items in both groups and were treated as missing, while the remaining valid data from the questionnaire were retained. No entire questionnaires were excluded in either group; therefore, no replacement strategy was required. A convenience sampling approach was employed to enhance representativeness and reduce selection bias. Invitations to show interest were addressed to all active pharmacists, and the first 200 respondents were included. Similarly, pharmacy students were recruited through academic social groups representing multiple universities and academic years.

Data collection methods, instruments used, and measurements assessed

Data from pharmacists were collected through paper-based questionnaires in Arabic, administered during face-to-face interviews. Before participation, each pharmacist was provided with a detailed explanation of the research purpose and the investigator's identity. Given the fact that pharmacy students actively use online social platforms to communicate, share resources, and exchange study materials, an electronic questionnaire, developed in Arabic via Google Forms, was distributed through targeted social media groups dedicated to pharmacy students. Certain demographic-related questions were modified from the original questionnaire to better suit this group.

To improve comprehension and response validity, a five-page presentation was developed, outlining AI and ML concepts, AI-powered chatbots, and their potential applications in pharmacy practice. The presentation was made available for all participants, who could choose whether or not to view it. This step was taken because preliminary feedback from pilot testing and validation revealed variable familiarity with AI terminology, which could compromise response accuracy. Additionally, Syria remains at an early stage in the introduction of AI technologies, and a proportion of pharmacists and students are still unfamiliar with these tools. The study aimed to include participants regardless of their prior exposure to AI. Therefore, providing a short, standardized overview helped minimize misinterpretation of technical terms. The presentation was presented in printed form to pharmacists who were given time to view it if they wished, and as a shared drive link embedded in the online questionnaire for students, accessible prior to answering the survey questions. Participation was entirely voluntary, with responses remaining anonymous, and respondents were informed of their right to withdraw from the study at any time.

The questionnaire was developed based on the study objectives and the literature review.^31,32 It underwent content validation by a pharmacy faculty member with expertise in pharmacy practice and an IT specialist with AI expertise. The questionnaire was refined iteratively based on their feedback. The validated questionnaire consisted of 29 closed-ended questions divided into five sections: (1) sociodemographic information; (2) current practice and willingness to adopt AI; (3) attitudes toward AI; (4) legal and regulatory considerations; and (5) challenges and barriers to AI adoption. Responses were recorded using multiple-choice questions to facilitate analysis.

AI applications knowledge assessment instrument, answer key, and scoring

Knowledge of AI applications in pharmacy practice was assessed using the questionnaire item: “How can AI be utilized in pharmacy practice?,” participants were presented with multiple response options and could select all that applied.

Responses were classified based on an a priori answer key. The following options were considered correct:

Researching drug interactions,

Understanding the mechanism of action of a specific drug,

Finding alternative drugs with similar effects, and

Understanding indications and contraindications for medications;

while the following options were considered incorrect:

Identifying specific clinical conditions,

Searching for a specific drug for a minor condition, and

Calculating drug dosages.

Selections under the “Other” option were reviewed by the authors on a case-by-case basis and classified as correct or incorrect.

For scoring and statistical coding, participants’ responses were grouped into four ordinal categories reflecting increasing levels of knowledge. Responses were assigned a score of 1 when all selected answers were incorrect, 2 when a mix of correct and incorrect answers was selected, 3 when some correct answers were selected but not all, and 4 when all selected answers were correct. Statistical coding followed the same scoring scheme.

The complete questionnaires for pharmacists and students, along with the five-slide presentation, are provided as Supplemental Material to this article.

Ethical considerations and informed consent

This study adhered to the ethical principles outlined in the World Medical Association's Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board (IRB) of the Faculty of Pharmacy and the Scientific Research Ethics Committee at Damascus University (Approval No. WD 4591, dated 25 August 2024). All participants provided written informed consent before data collection was initiated. Participation was entirely voluntary, and respondents were informed of their right to withdraw from the study at any stage. No remuneration or reward was offered to participants for taking part in this study.

Statistical analysis

Data analysis was conducted using the Statistical Package for the Social Sciences (SPSS), version 25 (IBM SPSS^® Statistics for Windows; IBM Corp., United States). Descriptive and inferential statistics were applied. Frequencies and percentages were used to summarize the collected data, providing insights into adoption rates, attitudes, and willingness to integrate AI technologies among the study population. Inferential analysis included Chi-square tests to compare categorical responses between groups and binary logistic regression to examine factors influencing AI usage. A p-value < .05 was considered statistically significant.

Results

The study included a total of 400 participants, comprising 200 community pharmacists and 200 pharmacy students. Among the pharmacists, 128 (64%) were already familiar with AI tools and chatbots and, therefore, did not view the introductory presentation. Similarly, 108 (54%) of the pharmacy students reported prior familiarity with AI concepts and tools.

Demographic characteristics

The demographic characteristics of the participating pharmacists are detailed in Table 1. The majority were between 26 and 30 years old (38%; n = 76), predominantly female (66.3%; n = 133), and had 0–2 years of experience in pharmacy practice (31%; n = 62). Additionally, most pharmacists reported a moderate level of pharmacy profits (59.6%; n = 115) and had an upper-intermediate English proficiency (34.7%; n = 69).

Table 1.

Demographic profile of participating pharmacists.

Variable		Total pharmacists, n = 200 N (%)*
Age group	20–25	72 (36)
	26–30	76 (38)
	31–35	21 (10.5)
	36–40	11 (5.5)
	41–50	9 (4.5)
	> 50	11 (5.5)
Gender	Male	66 (33.2)
Gender	Female	133 (66.8)
Years of practice	0–2	62 (31)
	2–4	53 (26.5)
	4–6	27 (13.5)
	> 6	58 (29)
Pharmacy ownership	Yes	46 (23)
Pharmacy ownership	No	154 (77)
Pharmacy profit level	Minimal	8 (4.1)
	Low	39 (20.2)
	Moderate	115 (59.6)
	High	21 (10.9)
	Excellent	10 (5.2)
Do you dispense prescriptions from hospitals, health centers, or clinics near the pharmacy?	Yes	94 (47)
	No	11 (5.5)
	Sometime	95 (47.5)
Is there a stable internet connection in the pharmacy at all times?	Yes	159 (79.9)
	No	40 (20.1)
Do you use the internet to stay updated on new pharmaceutical laws and practices?	Yes	143 (71.5)
	No	5 (2.5)
	Sometimes	52 (26)
English proficiency level	Weak	3 (1.5)
	Intermediate	53 (26.6)
	Upper intermediate	69 (34.7)
	Very good	59 (29.6)
	Fluent	15 (7.5)

Note: * percentages are calculated based on valid responses per variable, and the item-level missing data accounts for the observed discrepancies.

The distribution of pharmacy students is presented in Table 2. Of the 200 surveyed students, 92% (n = 185) were female, and 85.5% (n = 171) were enrolled at Damascus University, a public institution. The majority were in their fifth academic year (73%; n = 146), had achieved exemplary academic performance rated as Good or higher (94.5%; n = 189), and had completed practical training in a pharmacy (70.5%; n = 141).

Table 2.

Demographic profile of participating pharmacy students.

Variable		Total students, n = 200 N (%)*
Gender	Male	15 (7.5)
Gender	Female	185 (92.5)
University	Damascus	171 (85.5)
	AIU	5 (2.5)
	Yrmouk	1 (0.5)
	Sham	8 (4)
	IPU	4 (2)
	Other	11 (5.5)
Academic year	First	2 (1)
	Second	11 (5.5)
	Third	23 (11.5)
	Fourth	18 (9)
	Fifth	146 (73)
Academic performance rating (percentage)	Fair (60–64.99)	11 (5.5)
	Good (65–74.99)	97 (48.5)
	Very good (75–84.99)	83 (41.5)
	Excellent (85–94.99)	7 (3.5)
	Honor (95–100)	2 (1)
Have you completed practical training in a pharmacy?	Yes	141 (70.5)
Have you completed practical training in a pharmacy?	No	59 (29.5)
Was there a stable internet connection in the pharmacy where you trained or are currently undergoing training?	Yes	89 (51.1)
	No	85 (48.9)
English proficiency level	Weak	1 (0.5)
	Intermediate	61 (30.5)
	Upper intermediate	71 (35.5)
	Very good	61 (30.5)
	Fluent	(3)

Note: * percentages are calculated based on valid responses per variable, and the item-level missing data accounts for the observed discrepancies.

Practice and knowledge of AI tools

The usage and practice forms of AI tools among participants are summarized in Figure 1. A total of 54.8% (n = 219) participants reported using AI tools for various purposes in their daily lives (27.5%; n = 55 for pharmacists vs. 82%; n = 164 for students). Similarly, 49.8% (n = 199) participants had utilized AI in pharmacy settings (23%; n = 46 of pharmacists) or to obtain medical and pharmaceutical information during their studies (77%; n = 153 of students). A statistically significant difference was observed between pharmacists and students in this regard (p-value < .001).

Figure 1.

Patterns of AI chatbot use among participants (p-value < .001).

Among the AI-powered chatbots, ChatGPT was the most frequently used, with 52.9% (n = 211) users in the sample, followed by Copilot and Gemini (formerly known as Google Bard), which both had 12.6% (n = 50) of users. Again, a significant difference was noted between pharmacists and students regarding the usage of all chatbot platforms assessed in the survey. Notably, a higher proportion of students verified the information obtained from AI tools compared to pharmacists (χ²(2, N = 399) = 96.97, p < .001).

Regarding knowledge of AI applications in pharmacy practice, responses were categorized into four groups: (1) all selected answers were incorrect, (2) a mix of correct and incorrect answers was selected, (3) some correct answers were selected, but not all, and (4) all selected answers were correct. Different response patterns were observed between pharmacists and students (χ²(4, N = 400) = 19.39, p = .001). Both groups demonstrated generally limited knowledge of AI applications, with only 5.7% (n = 11) of pharmacists and 2.5% (n = 5) of students selecting exclusively correct answers. Among pharmacists, (30.4%; n = 59) selected mostly correct options while leaving some correct responses unselected, compared with (19%; n = 38) of students in the same category. These distributions highlight differences in how each group conceptualizes AI applications in pharmacy practice. A detailed breakdown is presented in Figure 2.

Figure 2.

Participants’ knowledge of AI applications in pharmacy (p-value < .001).

Viewing the AI presentation was not associated with participants’ knowledge of AI applications in pharmacy practice (χ²(3, N = 394) = 1.03, p = .795).

Attitudes toward AI in pharmacy practice

Attitudes toward the integration of AI in pharmacy practice are summarized in Table 3. Pharmacy students demonstrated a greater willingness to incorporate AI into their future professional practice compared to pharmacists (χ²(2, N = 396) = 33.35, p < .001). However, some pharmacists (13.1%; n = 52) outright rejected the idea of using AI, while (7.8%; n = 31) remained uncertain about its adoption.

Table 3.

Attitudes toward integrating AI into community pharmacy practice.

Variable		Total N (%)	Pharmacists N (%)	Students N (%)	p-value
Would referencing sources increase your trust in AI-generated information?	Yes	374 (95.2)	178 (92.2)	196 (98.0)	.008
	No	19 (4.8)	15 (7.8)	4 (2.0)	.008
Are you willing to incorporate AI into your future practice?	Not sure	31 (7.8)	31 (15.5)	0	<.001*
	Yes	313 (79.0)	147 (73.5)	166 (84.7)
	No	52 (13.1)	22 (11.0)	30 (15.3)
Do you think AI will replace a pharmacist?	Yes	37 (9.3)	26 (13.0)	11 (5.5)	<.001
	No	185 (46.3)	106 (53.0)	79 (39.5)
	In routine tasks only	178 (44.5)	68 (34.0)	110 (55.0)
Can you rely on AI-generated information?	Yes	36 (9.0)	24 (12.0)	12 (6.0)	.03
	No	80 (20.0)	45 (22.5)	35 (17.5)
	Only if verified	284 (71.0)	131 (65.5)	153 (76.5)
To what extent do you trust AI-generated information?	Not reliable	8 (2.0)	7 (3.5)	1 (0.5)	<.001
	Reliable after verification	146 (36.5)	94 (47.0)	52 (26.0)
	Fully reliable	5 (1.3)	1 (0.5)	4 (2.0)
	Reliable for straightforward tasks only	241 (60.3)	98 (49.0)	143 (71.5)
Do you think AI is more accurate than human experts?	Yes	95 (24.0)	44 (22.4)	51 (25.5)	.477
Do you think AI is more accurate than human experts?	No	301 (76.0)	152 (77.6)	149 (74.5)	.477
Would you like to have an AI-powered application designed for Syrian pharmacists?	Yes	338 (84.5)	155 (77.5)	183 (91.5)	<.001
	No	62 (15.5)	45 (22.5)	17 (8.5)	<.001
Do you think AI can assist with clinical decision-making?	Yes	289 (72.6)	131 (66.2)	158 (79.0)	.004
Do you think AI can assist with clinical decision-making?	No	109 (27.4)	67 (33.8)	42 (21.0)	.004
Would you be more motivated to use AI if recommended by drug regulatory authorities?	Yes	333 (84.3)	154 (77.4)	179 (91.3)	<.001
	No	62 (15.7)	45 (22.6)	17 (8.7)	<.001

Notes: AI: artificial intelligence.

*p-value refers to the comparison between pharmacists and students.

Viewing the AI presentation was not associated with willingness to use AI in the future (χ²(2, N = 396) = 0.10, p = .949).

A majority of pharmacists (53%; n = 106) expressed the belief that AI could not replace the pharmacist's role in pharmacy practice and this distribution differed significantly between pharmacists and students (χ²(2, N = 400) = 19.93, p < .001). However, a majority of students (55%; n = 110) of students believed AI could assist pharmacists in routine tasks but could not fully take over their responsibilities. In both groups, the most frequent response was that AI is reliable for straightforward tasks only; however, the distribution of responses differed significantly between pharmacists and students (χ²(3, N = 400) = 26.79, p < .001).

Additionally, compared to pharmacists (77.5%; n = 155), most students (91.5%; n = 183) expressed a strong desire for an AI-powered application specifically designed for pharmacy practice (χ²(1, N = 400) = 14.97, p < .001). A significant difference was observed between the two groups regarding most attitude-related questions. For example, the perception that AI could assist with clinical decision-making (χ²(1, N = 398) = 8.25, p = .004). However, when asked about AI accuracy, both pharmacists and students unanimously agreed that AI-generated responses were not more accurate than those provided by human experts, with no statistically significant difference observed between the two groups (χ²(1, N = 396) = 0.51, p = .477).

Legal aspects of AI implementation

Regarding the legal implications of AI usage in pharmacy, both pharmacists and students shared similar perspectives. They supported the establishment of official regulations by authoritative organizations to oversee AI integration and ensure accountability. Participants also agreed that healthcare institutions, such as the Ministry of Health should be made aware of AI adoption (69.3%; n = 138 of pharmacist vs. 71.5%; n = 143 of students) and that individuals using AI should be held legally responsible for any misinformation obtained from AI tools (80.8%; n = 323 of pharmacist vs. 84.5%; n = 169 of students). No significant differences were observed between the two groups concerning AI-related legal considerations; for example, regarding the need for regulations governing the use of AI in pharmacy practice (χ²(1, N = 400) = 1.30, p = .255). A detailed breakdown of responses is presented in Figure 3.

Figure 3.

Participants’ perspectives on the legal implications of AI in pharmacy.

Obstacles to AI adoption and proposed solutions

Table 4 shows obstacles facing AI adoption in pharmacy practice along with proposed solutions to address them. The most significant barrier to AI adoption in pharmacy practice was the lack of a stable internet connection within pharmacies (64.1%; n = 255), followed by limited technological skills and knowledge (56.3%; n = 224), and the restricted access to AI tools and chatbots due to international sanctions imposed on Syria (55.3%; n = 220). Both pharmacists (65.3%; n = 126) and students (84%; n = 168) identified the need for AI tools that support the Arabic language as a key solution to enhancing AI adoption in pharmacy practice.

Table 4.

Reported barriers to AI adoption in pharmacy and suggested solutions.

Variable		Total n = 400	Pharmacists n = 200	Students n = 200
Obstacles	Lack of stable internet access	255 (64.1%)	97 (49.0%)	158 (79.0%)
	Restrictions due to international sanctions	220 (55.3%)	105 (53.0%)	115 (57.5%)
	Limited proficiency in English	198 (49.7%)	91 (46.0%)	107 (53.5%)
	Insufficient technical skills and AI knowledge	224 (56.3%)	113 (57.1%)	111 (55.5%)
	Lack of trust in AI-generated information	206 (51.8%)	93 (47.0%)	113 (56.5%)
	Other challenges (e.g. paid access to certain tools, absence of scientifically validated references, etc.)	11 (2.8%)	8 (4.0%)	3 (1.5%)
Solutions	Introducing more courses on AI and machine learning in pharmacy education	266 (67.7%)	131 (67.9%)	135 (67.5%)
	Enhancing AI tools to support Arabic pharmaceutical and medical content	294 (74.8%)	126 (65.3%)	168 (84.0%)
	Other solutions (e.g. user-friendly AI interfaces for all age groups, training AI to prioritize medically verified information with reliable references)	24 (6.1%)	12 (6.2%)	12 (6.0%)

AI: artificial intelligence.

Factors influencing AI usage in pharmacy

The factors influencing AI usage among pharmacists are detailed in Table 5. Pharmacists with 0–2 years had significantly lower odds of utilizing AI tools in pharmacy practice than pharmacists with more than six years of experience (reference category) (OR = 0.075, 95% CI [0.013–0.427], p = .003). However, pharmacists with lower knowledge of AI applications in pharmacy settings were significantly more likely to report using AI tools in their practice compared to those with those demonstrating higher AI applications knowledge (reference category) (OR = 9.0, 95% CI [1.355–59.783], p = .023).

Table 5.

Factors influencing AI usage among pharmacist respondents.

Factor		p-value	OR*	95% CI for OR
Factor		p-value	OR*	Lower	Upper
Age group	20–25	.795	1.480	0.077	28.575
	26–30	.986	1.026	0.060	17.462
	31–35	.542	0.426	0.027	6.613
	36–40	.844	0.730	0.032	16.811
	41–50	.210	0.144	0.007	2.991
	> 50 (REF)	.495
Gender	Male	.867	1.081	0.432	2.706
Gender	Female (REF)
Years of practice	0–2	.003	0.075	0.013	0.427
	2–4	.053	0.198	0.038	1.021
	4–6	.184	0.321	0.060	1.715
	> 6 (REF)	.027
Pharmacy ownership	Yes	.500	0.695	0.241	2.005
Pharmacy ownership	No (REF)
Pharmacy profit level	Minimal	.999	2343731389
	Low	.141	3.929	0.636	24.276
	Moderate	.153	3.264	0.644	16.538
	High	.563	1.743	0.265	11.461
	Excellent (REF)	.512
Do you dispense prescriptions from hospitals, health centers, or clinics near the pharmacy?	Yes (regularly)	.192	1.711	0.764	3.832
	No	.447	0.536	0.107	2.676
	Sometimes (REF)	.236
Is there a stable internet connection in the pharmacy at all times?	Yes	.829	0.890	0.310	2.559
	No (REF)
Do you use the internet to stay updated on new pharmaceutical laws and practices?	Yes (regularly)	.014	0.269	0.095	0.768
	No	.999	244705500.7	.000
	Sometimes (REF)	.049
English proficiency level	Weak	.812	0.677	0.027	16.936
	Intermediate	.613	1.482	0.322	6.817
	Upper intermediate	.378	1.909	0.454	8.032
	Very good	.923	1.072	0.261	4.408
	Fluent (REF)	.759
What are the applications of AI chatbots in pharmacy?	All answers are incorrect	.023	9.0	1.355	59.783
	Correct and incorrect answers selected	.023	4.383	1.227	15.657
	Not all correct answers selected	.063	3.520	0.937	13.224
	All answers are correct (REF)	.086

*OR: odds ratio; AI: artificial intelligence..

Factors influencing AI implementation willingness in pharmacy

As shown in Table 6, two variables were found to be statistically significant factors influencing pharmacists’ willingness to implement AI technology. Years of practice (p = .002) showed a notable association, with a significant proportion of participants expressing a willingness to integrate AI tools into their future work. Additionally, the prescription dispensing variable (p = .016) also demonstrated a significant impact on pharmacists’ openness to adopting AI in their practice.

Table 6.

Factors associated with pharmacists’ willingness to implement AI technology.

Variable		Are you willing to incorporate AI technology in pharmacy settings in the future?			p-value
Variable		Yes n (%)	No n (%)	Not sure n (%)	p-value
Age group	20–25	51 (70.8)	7 (9.7)	14 (19.4)	.551
	26–30	57 (75)	7 (9.2)	12 (15.8)
	31–35	17 (81)	2 (9.5)	2 (9.5)
	36–40	10 (90.9)	1 (9.1)	0 (0)
	41–50	5 (55.6)	2 (22.2)	2 (22.2)
	> 50	7 (63.6)	3 (27.3)	1 (9.1)
Gender	Male	44 (66.7)	9 (13.6)	13 (19.7)	.321
Gender	Female	102 (76.7)	13 (9.8)	18 (13.5)	.321
Years of practice	0–2	39 (62.9)	6 (9.7)	17 (27.4)	.002
	2–4	44 (83.0)	1 (1.9)	8 (15.1)
	4–6	18 (66.7)	6 (22.2)	3 (11.1)
	> 6	46 (79.3)	9 (15.5)	3 (5.2)
Pharmacy ownership	Yes	32 (69.6)	9 (19.6)	5 (10.9)	.085
Pharmacy ownership	No	115 (74.7)	13 (8.4)	26 (16.9)	.085
Pharmacy profit level	Minimal	6 (75)	2 (25)	0 (0)	.376
	Low	27 (69.2)	4 (10.3)	8 (20.5)
	Moderate	89 (77.4)	12 (10.4)	14 (12.2)
	High	13 (61.9)	2 (9.5)	6 (28.6)
	Excellent	6 (60)	1 (10)	3 (30)
Do you dispense prescriptions from hospitals, health centers, or clinics near the pharmacy?	Yes (regularly)	76 (80.9)	4 (4.3)	14 (14.9)	.016
	No	6 (54.5)	4 (36.4)	1 (9.1)
	Sometimes	65 (68.4)	14 (14.7)	16 (16.8)
Is there a stable internet connection in the pharmacy at all times?	Yes	123 (77.4)	13 (8.2)	23 (14.5)
	No	24 (60)	9 (22.5)	7 (17.5)
Do you use the internet to stay updated on new pharmaceutical laws and practices?	Yes (regularly)	103 (72)	15 (10.5)	25 (17.5)	.693
	No	4 (80)	1 (20)	0 (0)
	Sometimes	40 (76.9)	6 (11.5)	6 (11.5)
English proficiency level	Weak	1 (33.3)	1 (33.3)	1 (33.3)	.434
	Intermediate	44 (83)	5 (9.4)	4 (7.5)
	Upper	50 (72.5)	8 (11.6)	11 (15.9)
	Very good	43 (72.9)	5 (8.5)	11 (18.6)
	Fluent	9 (60)	2 (13.3)	4 (26.7)

AI: artificial intelligence.

Discussion

This study provides the first comprehensive assessment of AI use, knowledge, and attitudes among community pharmacists and pharmacy students in a resource-constrained setting such as Syria. Overall, AI use was more common among pharmacy students than practicing pharmacists, pharmacists demonstrated slightly higher knowledge of validated AI applications, and both groups expressed generally positive attitudes toward future AI integration despite substantial concerns regarding reliability, regulation, and infrastructure. These findings should be interpreted in light of the study setting. The study was conducted in Damascus, Syria's principal urban, administrative, and educational center, which hosts a socioeconomically heterogeneous population due to internal displacement and blurred boundaries between urban and peri-urban healthcare practice.³⁷

This complex context, further shaped by prolonged sanctions, infrastructural instability, and language barriers, is likely to influence exposure to and engagement with AI technologies in pharmacy practice.

Use of AI tools

The study revealed a statistically significant difference in AI usage between pharmacists and students. While 64% of pharmacists reported prior familiarity with AI tools, only 27.5% actively used them in daily practice. In contrast, 82% of students reported using AI tools for academic and various daily tasks. This difference likely reflects generational disparities in technological familiarity. As members of Generation Z, students have grown up in a digitally driven environment, which influences learning styles, communication patterns, and openness to emerging technologies, thereby facilitating greater AI use to enhance academic and operational efficiency.³⁸

On the other hand, pharmacists may tend to be more cautious about integrating digital technologies due to their established commitments and concerns about increasing working hours or workload.³⁹ A previous Syrian study assessing AI KAP among medical students and physicians reported that 89.3% had never used AI in any activity, while only 10.6% had ever used AI in practice. Interestingly, graduates used AI more frequently than undergraduates, a trend that contrasts with our findings.⁴⁰ Additionally, a nationwide U.S. study among pharmacy students and faculty found no significant difference in AI usage between faculty and students.⁴¹

Notably, a higher proportion of students in our study reported verifying AI-generated information compared to pharmacists. In contrast, a study in India revealed that only 26.2% of students cross-checked the accuracy of data collected using AI-powered tools and search engines, while the majority (73.8%) did not verify information obtained from AI-driven sources.⁴²

Knowledge of AI chatbots applications

Participants’ knowledge of AI applications in pharmacy practice was assessed by distinguishing between validated and nonvalidated uses of AI tools in community pharmacy settings (Figure 2). Applications considered appropriate included identifying drug–drug interactions (DDIs), understanding the mechanism of action of a particular drug, finding alternative drugs with the same effects, and identifying indications and contraindications. These uses align with AI's current role as a clinical decision-support tool rather than an autonomous decision-maker.

Existing evidence supports cautious use of AI for medication-related information. For instance, a comparative evaluation of conversational AI tools (ChatGPT-3.5, ChatGPT-4, Bing AI, and Bard) demonstrated that Bing AI showed superior accuracy (78.8–89.0%) and specificity for DDIs. However, the authors emphasized that further refinement is required before such tools can be considered reliable for routine clinical practice.⁴³ Conversely, other studies reported substantially lower accuracy rates for ChatGPT in detecting DDIs (30%) and adverse drug reactions (65%), highlighting inconsistencies in AI performance and reinforcing the necessity of pharmacist oversight.⁴⁴

AI tools also demonstrated notable limitations in tasks requiring numerical precision. Previous research reported poor performance of ChatGPT in drug dosage calculations, with correct response rates as low as 35%, likely due to the complexity of dosage determination, which depends on patient-specific variables such as age, weight, comorbidities, and medical history.^44,45 These findings justify the classification of dosage calculation as a nonvalidated AI application in our assessment.

Regarding medication selection, AI tools demonstrated relatively better performance in suggesting therapeutic alternatives, with reported success rates of up to 85%.⁴³ Nevertheless, because large language models are trained primarily on general text data, their recommendations remain insufficiently reliable for some specific medical questions or suggesting medications for minor conditions.⁴⁶

Analyzing responses to this question identifies gaps in participants’ understanding of AI's validated roles versus its limitations. Among pharmacists, 55% selected a mix of correct and incorrect answers, while 30% selected mostly correct answers. In contrast, pharmacy students demonstrated a different response pattern. A substantially higher proportion (72%) selected a mix of correct and incorrect answers, while 19% selected mostly correct answers. Overall, both groups showed gaps in identifying validated AI applications in pharmacy practice; however, pharmacists demonstrated a slightly higher proportion of fully correct responses compared with students. This is consistent with prior research showing that Syrian graduate doctors possessed greater knowledge compared to students.⁴⁰

The observed differences in knowledge between pharmacists and students may be attributed to variations in professional exposure, learning context, and patterns of AI use. While students are more frequently exposed to AI tools through academic and informal use, this exposure may not translate into accurate recognition of validated pharmacy-specific applications. Pharmacists, on the other hand, may benefit from practice-based reasoning grounded in routine medication-related tasks, which could explain their slightly higher proportion of fully correct responses.

However, other studies have reported contrasting findings. For example, a study involving students and faculty from six Middle Eastern countries found that students scored significantly higher in AI knowledge than faculty members.³¹ This difference may reflect variations in educational systems, as some countries may have begun integrating AI into pharmacy curricula, whereas in Syria, AI knowledge appears to be acquired primarily through professional experience rather than formal education.

Attitudes toward AI in pharmacy practice

Both pharmacists and students demonstrated generally positive attitudes toward AI, with students expressing a greater willingness to integrate AI into their future professional practice. These findings are consistent with prior research conducted in Syria, Jordan, and Ethiopia.^40,47,48 Despite this positive attitude, concerns regarding the reliability of AI-generated information persist, as most respondents (71%) reported relying on AI only after verification.

Participants largely viewed AI as an assistive tool rather than a replacement for pharmacists. Over half of pharmacists (53%) rejected the notion of AI replacing their roles, while 55% of students believed AI could assist with routine tasks. This perception aligns with global healthcare trends emphasizing the complementary role of AI alongside human clinical judgment.⁴⁹ Similarly, studies from the United States indicate that pharmacy students anticipate AI as part of future practice and education. They acknowledge that while AI cannot replace human creativity or judgment, it can support evidence-based medicine and enhance professional efficiency.⁵⁰

Two variables were found to be statistically significant factors influencing pharmacists’ willingness to implement AI technology. Pharmacists in the two to four years category demonstrated the highest willingness (83.0%), which could be attributed to their openness toward innovation at a career level marked by reinforcement of skills and responsiveness to time-saving technology. Conversely, the four to six years cohort was distinguished by the highest resistance (22.2%) and lowest interest (66.7%), potentially due to established workflows or skepticism toward disruptive technologies. Notably, Pharmacists with more than six years of experience also exhibited high willingness, perhaps because they were confident in balancing AI benefits against clinical expertise or recognized the potential to ease workload burdens.

The uncertainty observed among early-career pharmacists highlights the need for targeted education to bridge knowledge gaps and build confidence in AI applications. Additionally, the prescription dispensing practices also seem to influence pharmacists’ openness to AI adoption in their practice.

While AI adoption is moderate, skepticism regarding the accuracy of AI-generated information underscores the importance of verification protocols, especially in an environment such as Syria, where access to scientifically validated references is restricted.⁵¹ Importantly, The strong demand for Arabic-language AI applications reflects the need for locally relevant tools that can support pharmacy practice in the region.

Legal aspects of AI implementation

Regarding the legal implications of AI usage in pharmacy, both pharmacists and students shared similar perspectives. Participants strongly supported legal responsibility for AI errors (80.8%) and regulatory governance (92.5%). Currently, Syria lacks specific regulations for AI in pharmacy practice. Adopting elements from the EU AI Act, FDA guidance, and WHO recommendations, such as risk-based classification, human oversight, and monitoring, could enhance patient safety and promote ethical use.^52,53

These concerns mirror findings from a cross-sectional study among 501 pharmacy professionals in the Middle East and North Africa countries. The respondents were concerned about issues related to patient privacy, cybersecurity, job displacement, and the absence of legal regulation.⁵⁴ Without adequate governance, unregulated AI use in fragile healthcare systems may exacerbate risks related to bias, misinformation, and barriers stemming from language limitations and underdeveloped digital infrastructure.

Obstacles to AI adoption

The study identified three primary obstacles to AI adoption: the lack of a stable internet connection within pharmacies (64.1%), followed by limited technological skills and knowledge (56.3%), and restricted access to AI tools and chatbots due to international sanctions imposed on Syria (55.3%). The lack of technical skills may stem from educational gaps, including the absence of AI-focused courses in pharmacy curricula, limited professional training opportunities, and challenges related to Arabic-language localization of AI tools.

In addition, policy-related shortcomings, such as the absence of national strategies for digital health integration, may further hinder the development of necessary competencies. These vary from those in other nations due to the unique situation of Syria. For example, in Jordan, a nearby nation that does not suffer from issues of war or economic sanctions, the most common obstacles included lack of AI-specific hardware and software, the need for human supervision, and high operational costs.⁴⁷

Factors influencing AI implementation willingness in pharmacy

Our findings reveal different patterns in AI usage among pharmacists based on experience and knowledge. Using pharmacists with more than six years of experience as the reference category, those with zero to two years of practice demonstrated a lower odds ratio of using AI tools in their future pharmacy practice. This finding is generally consistent with the Jordanian study, in which those with 6–10 years of practice were significantly more likely to use AI than those with <1 year of practice.⁴⁷ This could be due to the fact that fresh graduate may be more cautious about integrating potentially noncredible sources into pharmacy practice.

Interestingly, pharmacists with lower levels of AI knowledge were more inclined to employ AI tools. This likely reflects random or uninformed use, as limited understanding prevents recognition of inaccuracies or inappropriate applications. These findings underscore the need for structured education and guidance to ensure AI adoption is both informed and safe.

This pattern contrasts with findings from Jordan, where greater AI knowledge was associated with positive attitude and adoption.⁴⁷ This discrepancy may reflect differences in how AI is perceived and used in resource-constrained settings such as Syria, where informal or superficial use of readily accessible tools may occur without comprehensive understanding. In contrast, in more stable settings, greater knowledge may translate into more confident and structured adoption.

Other factors like age, gender, and English level did not affect AI usage. However, a study conducted in the United States revealed notable differences in AI usage and familiarity based on gender and age, with male participants reporting higher usage rates and greater familiarity with AI technologies compared to females. Additionally, younger respondents showed higher levels of AI use and familiarity than their older counterparts.⁴¹

Study's strengths and limitations

To our knowledge, this study is the first to directly compare AI adoption between community pharmacists and students in Syria, providing novel insights pertinent to this country and contributing original knowledge to the global literature in this field.

However, it has several limitations. First, the overrepresentation of Damascus University students (85.5%) and urban pharmacists might limit the generalizability of our findings to rural areas or other Syrian governorates that differ in cultural context or internet infrastructure. Although Damascus hosts individuals displaced from various Syrian regions, the study population may still reflect comparatively better access to resources and connectivity. Consequently, the results should not be assumed to represent pharmacists and students in more remote or resource-limited settings. Second, all study variables were self-reported and are therefore subject to inherent biases, which include social desirability bias, inaccurate recall, and the potential overstatement of willingness to adopt AI technologies. As a result, reported levels of adoption and openness may be inflated and should be interpreted with caution. Third, the use of mixed data collection modes (paper-based surveys for pharmacists and online questionnaires for students) may have introduced mode-related response bias. Differences in format and perceived anonymity can influence how participants respond and may affect comparability between groups. Finally, the introductory AI presentation provided to participants may have introduced a degree of information bias by influencing their subsequent responses. Future research should therefore examine differences in knowledge and attitudes between pharmacists who are already familiar with AI and those with no prior exposure, without the use of pre-survey educational materials, to better capture baseline perceptions.

Conclusion

This study assessed AI adoption among community pharmacists in Syria and compared their knowledge and attitudes with those of pharmacy students. Using a cross-sectional survey of 400 participants, this study revealed a clear coexistence of interest in and concern about AI among both community pharmacists and pharmacy students. Within the pharmacist group, AI use varied by years of experience, with early-career pharmacists being significantly less likely to use AI compared to those with more than six years of practice. While enthusiasm for future use is evident, gaps in knowledge and confidence highlight the need for supportive conditions that enable safe and meaningful adoption.

Our findings reveal uneven readiness for digital transformation within pharmacy practice in low-resource, internet-restricted settings. Effective integration of AI will depend on the availability of Arabic-language tools and on reliable mechanisms for validating AI-generated information to safeguard against erroneous advice. To ensure safe and equitable adoption, regulations should be developed rapidly, addressing specific challenges such as inadequate infrastructure, limited access to technology due to sanctions, and existing regulatory gaps, particularly within the healthcare sector.

Supplemental Material

sj-docx-1-dhj-10.1177_20552076261432730 - Supplemental material for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country

Supplemental material, sj-docx-1-dhj-10.1177_20552076261432730 for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country by Areej Kahwaji, Reem Ismael, Tala Arnouk, Alaa Alhomsy and Tamim Alsuliman in DIGITAL HEALTH

Supplemental Material

sj-docx-2-dhj-10.1177_20552076261432730 - Supplemental material for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country

Supplemental material, sj-docx-2-dhj-10.1177_20552076261432730 for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country by Areej Kahwaji, Reem Ismael, Tala Arnouk, Alaa Alhomsy and Tamim Alsuliman in DIGITAL HEALTH

Supplemental Material

sj-pdf-3-dhj-10.1177_20552076261432730 - Supplemental material for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country

Supplemental material, sj-pdf-3-dhj-10.1177_20552076261432730 for Adoption rates and knowledge of generative artificial intelligence in pharmacy practice: A comparative study in an internet-restricted country by Areej Kahwaji, Reem Ismael, Tala Arnouk, Alaa Alhomsy and Tamim Alsuliman in DIGITAL HEALTH

Footnotes

Acknowledgments

Not applicable.

ORCID iD

Areej Kahwaji

Ethical considerations

This study adhered to the ethical principles outlined in the World Medical Association's Declaration of Helsinki. Ethical approval was obtained from the IRB of the Faculty of Pharmacy and the Scientific Research Ethics Committee at Damascus University (Approval No. WD 4591, dated 25 August 2024).

Consent to participate

All participants provided a written informed consent at the beginning of the questionnaire prior to collecting any data and participated on a voluntary basis. No remuneration or reward was offered to participants for taking part in this study.

Author contributions statement

Areej Kahwaji: conceptualization, data curation, formal analysis, methodology, project administration, writing—original draft, and writing—review and editing; Reem Ismael: investigation, resources, writing—original draft, and writing—review and editing; Tala Arnouk: conceptualization, data curation, and methodology; Alaa Alhomsy: data curation and investigation; Tamim Alsuliman: conceptualization, methodology, supervision, validation, and writing—review and editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The anonymized dataset is available at a reasonable request from the corresponding author.

Supplemental material

Supplemental material for this article is available online.

References

Bajwa

Munir

Nori

, et al. Artificial intelligence in healthcare: transforming the practice of medicine. Future Healthc J 2021; 8: e188–e194.

Kermany

Goldbaum

Cai

, et al. Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell 2018; 172: 1122–1131.e9.

Topol

. High-performance medicine: the convergence of human and artificial intelligence. Nat Med 2019; 25: 44–56.

Esteva

Kuprel

Novoa

, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2017; 542: 115–118.

Varghese

Harrison

O’Grady

, et al. Artificial intelligence in surgery. Nat Med 2024; 30: 1257–1268.

Chalasani

Syed

Ramesh

, et al. Artificial intelligence in the field of pharmacy practice: a literature review. Explor Res Clin Soc Pharm 2023; 12: 100346.

Raza

Aziz

Noreen

, et al. Artificial intelligence (AI) in pharmacy: an overview of innovations. Innov Pharm 2022; 13: 2.

Ting

Chung

Chen

, et al. A drug identification model developed using deep learning technologies: experience of a medical center in Taiwan. BMC Health Serv Res 2020; 20: 12.

Corny

Rajkumar

Martin

, et al. A machine learning–based clinical decision support system to identify prescriptions with a high risk of medication error. J Am Med Inform Assoc 2020; 27: 1688–1694.

10.

Ramadhani

. Chatbots in pharmacy: a boon or a bane for patient care and pharmacy practice? Sci Pharm 2023; 2: 3.

11.

Ethics and governance of artificial intelligence for health: Guidance on large multi-modal models. Accessed January 26, 2026. https://www.who.int/publications/i/item/9789240084759.

12.

Raza

Venkatesh

Kvedar

. Generative AI and large language models in health care: pathways to implementation. Npj Digit Med 2024; 7: 62.

13.

Adamopoulou

Moussiades

. An overview of chatbot technology. In: Maglogiannis I, Iliadis L, Pimenidis E, eds. Artificial intelligence applications and innovations. Springer International Publishing; 2020:373-383. doi:10.1007/978-3-030-49186-4_31.

14.

Health C for D and R. Artificial intelligence-enabled device software functions: lifecycle management and marketing submission recommendations. January 6, 2025. Accessed January 26, 2026. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/artificial-intelligence-enabled-device-software-functions-lifecycle-management-and-marketing.

15.

Regulation of AI in Healthcare. GOV.UK. February 2, 2026. Accessed January 26, 2026. https://www.gov.uk/government/calls-for-evidence/regulation-of-ai-in-healthcare.

16.

Gilbert

Harvey

Melvin

, et al. Large language model AI chatbots require approval as medical devices. Nat Med 2023; 29: 2396–2398.

17.

About Us. Phambot AI. Accessed January 27, 2026. https://www.pharmbotai.com/about-us/.

18.

Stabilizing Syria. Rehabilitating Syria’s public health system in a fragile transition. Published online July 29, 2025. Accessed January 27, 2026. https://www.csis.org/analysis/stabilizing-syria-rehabilitating-syrias-public-health-system-fragile-transition.

19.

A ES Nahed. Pharmacists in Aleppo help families get the medicines they need. Accessed January 27, 2026. https://www.emro.who.int/syria/news/pharmacists-in-aleppo-help-families-get-the-medicines-they-need.html.

20.

Al alawneh

Nuaimi

Basheti

. Pharmacists in humanitarian crisis settings: assessing the impact of pharmacist-delivered home medication management review service to Syrian refugees in Jordan. Res Soc Adm Pharm 2019; 15: 164–172.

21.

El-Hammadi

. Career preferences of Syrian students and their attitudes toward a number of practice areas: will community pharmacy continue to dominate the profession? Curr Pharm Teach Learn 2013; 5: 373–380.

22.

Mansour

Al-Kayali

. Community pharmacists' role in controlling bacterial antibiotic resistance in Aleppo, Syria. Iran J Pharmaceut Res : IJPR 2015; 16: 1612–1620. e125038.

23.

Alzaydi

Abedalrhman

. Digital innovations in healthcare: harnessing artificial intelligence, IoT, and big data analytics for personalized medicine and improved patient outcomes—insights from the Syrian healthcare sector. Appl Sci Biotechnol J Adv Res 2025; 4: 28–47.

24.

Sam. HeRAMS. Accessed January 27, 2026. https://www.emro.who.int/syr/herams/herams.html.

25.

Alkhalil

Abbara

Grangier

, et al. AI in conflict zones: the potential to revitalise healthcare in Syria and beyond. BMJ Glob Health 2024; 9: e015755.

26.

Abdelhay

Mohammed

Hefny

. Deep learning for Arabic healthcare: MedicalBot. Soc Netw Anal Min 2023; 13: 71.

27.

Jarab

Abu Heshmeh

Al Meslamani

. Artificial intelligence (AI) in pharmacy: an overview of innovations. J Med Econ 2023; 26: 1261–1265.

28.

Jaber

Hasan

Abutaima

, et al. The impact of artificial intelligence on the knowledge, attitude, and practice of pharmacists across diverse settings: a cross-sectional study. Int J Med Inf 2024; 192: 105656.

29.

Abu-Farha

Fino

Al-Ashwal

, et al. Evaluation of community pharmacists’ perceptions and willingness to integrate ChatGPT into their pharmacy practice: a study from Jordan. J Am Pharm Assoc 2023; 63: 1761–1767.e2.

30.

Dada

Puladi

, et al. ChatGPT in healthcare: a taxonomy and systematic review. Comput Methods Programs Biomed 2024; 245: 108013.

31.

Hasan

Jaber

Tabbah

, et al. Knowledge, attitude and practice among pharmacy students and faculty members towards artificial intelligence in pharmacy practice: a multinational cross-sectional study. PLOS ONE 2024; 19: e0296884.

32.

Syed

Al-Rawi

MBA

. Community pharmacists awareness, perceptions, and opinions of artificial intelligence: a cross-sectional study in Riyadh, Saudi Arabia. Technol Health Care 2024; 32: 481–493.

33.

Orok

Okaramee

Egboro

, et al. Pharmacy students’ perception and knowledge of chat-based artificial intelligence tools at a Nigerian university. BMC Med Educ 2024; 24: 1237.

34.

Bartlett

IIJE

Kotrlik

Higgins

. Organizational research: determining appropriate sample size in survey research. Inf Technol Learn Perform J 2001; 19: 43–50.

35.

Lwanga

Lemeshow

. Sample size determination in health studies : a practical manual. Accessed January 27, 2026. https://iris.who.int/items/9c2e5da4-3785-4fec-9dbc-841e4ae0d98c.

36.

Delice

. The sampling issues in quantitative research. Educ Sci Theory Pract 2010; 10: 2001–2018.

37.

2019 EASO COI Report Methodology. Country of origin information report | Syria - socio-economic situation: Damascus City. EUAA; 2021. doi:10.2847/95783.

38.

Chan

CKY

Lee

KKW

. The AI generation gap: are gen Z students more interested in adopting generative AI such as ChatGPT in teaching and learning than their gen X and millennial generation teachers? Smart Learn Environ 2023; 10: 60.

39.

Borges do Nascimento

Abdulazeem

Vasanthan

, et al. Barriers and facilitators to utilizing digital health technologies by healthcare professionals. Npj Digit Med 2023; 6: 1–28.

40.

Swed

Alibrahim

Elkalagi

NKH

, et al. Knowledge, attitude, and practice of artificial intelligence among doctors and medical students in Syria: a cross-sectional online survey. Front Artif Intell 2022; 5: 1011524.

41.

Gustafson

Berman

Gavaza

, et al. Pharmacy faculty and students perceptions of artificial intelligence: a national survey. Curr Pharm Teach Learn 2025; 17: 102344.

42.

Risana

Shirin

Purayil

, et al. Artificial intelligence and pharmacy education: a survey to assess the knowledge, application, and perspective of B. Pharm. Students from India. Discov Educ 2024; 3: 13.

43.

Al-Ashwal

Zawiah

Gharaibeh

, et al. Evaluating the sensitivity, specificity, and accuracy of ChatGPT-3.5, ChatGPT-4, Bing AI, and Bard against conventional drug-drug interactions clinical tools. Drug Healthc Patient Saf 2023; 15: 137–147.

44.

Salama

. The promise and challenges of ChatGPT in community pharmacy: a comparative analysis of response accuracy. Pharmacia 2024; 71: 1–5.

45.

Shin

Ramanathan

. Evaluation of prompt engineering strategies for pharmacokinetic data analysis with the ChatGPT large language model. J Pharmacokinet Pharmacodyn 2024; 51: 101–108.

46.

Vaishya

Misra

Vaish

. ChatGPT: is this version good for healthcare and research? Diabetes Metab Syndr Clin Res Rev 2023; 17: 102744.

47.

Jarab

Al-Qerem

Alzoubi

, et al. Artificial intelligence in pharmacy practice: attitude and willingness of the community pharmacists and the barriers for its implementation. Saudi Pharm J 2023; 31: 101700.

48.

Sendekie

Limenh

Abate

, et al. Artificial intelligence in community pharmacy practice: pharmacists’ perceptions, willingness to utilize, and barriers to implementation. Explor Res Clin Soc Pharm 2024; 16: 100542.

49.

Alowais

Alghamdi

Alsuhebany

, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ 2023; 23: 89.

50.

Knobloch

Cozart

Halford

, et al. Students’ perception of the use of artificial intelligence (AI) in pharmacy school. Curr Pharm Teach Learn 2024; 16: 102181.

51.

Saadi

Abbas

Turk

, et al. Medical research in war-torn Syria: medical students’ perspective. Lancet 2018; 391: 2497–2498.

52.

Linera MÁ

Meuwese

. Regulating AI from Europe: a joint analysis of the AI act and the framework convention on AI. Theory Pract Legis 2025; 13: 1–20.

53.

Ethics and governance of artificial intelligence for health: WHO guidance. 1st ed. World Health Organization; 2021.

54.

Hasan

Jaber

Khabour

, et al. Ethical considerations and concerns in the implementation of AI in pharmacy practice: a cross-sectional study. BMC Med Ethics 2024; 25: 55.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.24 MB