Optimizing dental prescription through AI: clinical challenges and realities

The management of drug prescription in dental medicine, particularly crucial for antibiotic therapy and the safety of polymorbid patients, constitutes a major lever for reducing clinical errors. While Artificial Intelligence (AI) is perceived as a technological lever capable of securing these therapeutic decisions, its integration into the practice depends directly on the preparation of practitioners. To date, few studies have explored the level of readiness and perceptions of dentists regarding these tools, and no data were available for Pakistani institutions.

Study objectives and hypotheses

This descriptive cross-sectional study, conducted from February 1st to May 30th, 2025, at CMH Lahore Medical College & IOD, aimed to evaluate the awareness, perception, and acceptance of AI applied to prescription by 150 dental healthcare professionals (interns, residents, and consultants). The authors sought to precisely identify the demographic factors influencing this adoption as well as the perceived barriers within this university hospital setting.

The study is based on the hypothesis that, despite a theoretical interest in innovation, structural obstacles — such as the lack of specific academic training, implementation costs, and doubts regarding the reliability of algorithms — hinder the concrete use of AI for medication safety at the chairside.

Methodology of the cross-sectional study on AI in dental prescription

This cross-sectional descriptive study was conducted from February 1st to May 30th, 2025, at the CMH Lahore Medical College & Institute of Dentistry (IOD), a tertiary military teaching institution located in Lahore, Pakistan.

The study population consisted of a sample of 150 dental medicine professionals. Recruitment targeted three distinct practitioner profiles:

• House officers (interns);
• Post-graduate trainees;
• The consultants (experienced practitioners).

The data collection protocol was based on the administration of a validated and self-administered questionnaire, sent to all participants. The study recorded an exceptional response rate of 100% (n=150). The objective was to evaluate the perception, level of awareness, and acceptance of artificial intelligence (AI) applied to drug prescription and therapeutic safety.

Data analysis was performed using IBM SPSS v.25 software. The statistical methods used include:

• Descriptive statistics for demographic variables;
• The chi-square test for group comparisons;
• A binary logistic regression to identify the factors influencing the willingness to adopt.

The statistical significance threshold was set at p < 0.05.

An adoption slowed by a lack of specific knowledge

The study, conducted among 150 dental healthcare professionals (100% response rate), reveals a major paradox between the global awareness of artificial intelligence (AI) and its mastery within the specific therapeutic framework of drug prescription.

Demographic factors and determinants of acceptance

Binary logistic regression analysis shows that the typical profile of a practitioner favorable to AI is young and currently undergoing specialization. The most predictive factor for acceptance remains prior exposure to these technologies (p = 0.001), highlighting that usage precedes trust.

Despite this openness, the medical community expresses significant structural and technical reservations:

• Lack of training: 85.3% of respondents deplore the total absence of a dedicated AI curriculum in dental university programs.
• Clinical reliability: 66.0% are concerned about the accuracy of algorithms and the safety of drug recommendations.
• Economic barrier: 60.7% identify implementation costs as a major barrier to adoption in dental practices or hospital settings.

A near-unanimous consensus (90.7%) emerges on a critical point: AI must not substitute for the practitioner. It is perceived as a clinical decision support tool (CDSS) that must imperatively be used in conjunction with professional judgment to guarantee patient safety.

Analysis of results and clinical scope

The data from this cross-sectional study reveal a critical gap between the overall awareness of AI (74.7%) and the specific knowledge of assisted prescription tools (36%). For the practitioner, this means that while the technology is accepted in principle, its concrete applications for medication safety — such as polypharmacy management or antimicrobial stewardship — remain largely unknown. The high acceptance rate (68.7%), driven by young practitioners (77.8%) and residents (77.6%), indicates a window of opportunity for the integration of clinical decision support systems (CDSS).

Limits and perspective

As the first study of its kind in a dental institution in Pakistan, its scope is rooted in a specific military hospital context (CMH Lahore), which may limit the generalizability of the results to private practice as a whole. The authors highlight major structural barriers: 85.3% of the 150 professionals surveyed point to the lack of training in initial curricula and 66% are concerned about the reliability of the tools. Contrary to popular belief, the obstacle is not technophobic but academic and financial (60.7% cite the cost of implementation).

Implications for daily practice

The near-unanimous agreement (90.7%) that AI should complement, rather than replace, clinical judgment is a pivotal result. Clinically, the use of AI is perceived as a lever to reduce prescription errors in patients with comorbidities. To move from perception to practice, the study suggests that integrating AI literacy into postgraduate education and establishing guidelines by national regulatory bodies are now indispensable.

In concrete terms, for the practitioner:

• Secure your prescriptions: Use AI as a lever for antibiotic stewardship and polypharmacy management, while maintaining your decision-making sovereignty (90.7% of peers agree on the primacy of clinical judgment).
• Anticipate the digital transition: As the lack of training is the main obstacle identified, prioritise self-training on clinical decision support systems (CDSS) to resolve doubts regarding technical reliability (cited by 66% of respondents).
• Optimize workflow: If you manage a team, be aware that young practitioners are the most likely (77.8%) to adopt these tools, thus facilitating their implementation in the dental practice.

Technical lexicon of the study

AI-CDSS (Artificial Intelligence-based Clinical Decision Support System): Clinical decision support systems using artificial intelligence to assist the practitioner in therapeutic choices and the securing of prescriptions.

Antibiotic stewardship: Rational management of antibiotics, identified as a major lever for improvement thanks to AI to optimize prescriptions and combat resistance.

Binary logistic regression: Statistical analysis method used in the study to identify demographic factors and variables influencing practitioners' willingness to adopt AI.

Cross-sectional study: A type of descriptive cross-sectional study used to evaluate, at a specific point in time (from February 1st to May 30th, 2025), the perceptions and acceptance of AI tools by dental surgeons.

Medication safety: Medication safety focused on reducing the risk of prescription errors, particularly crucial when treating patients with underlying systemic pathologies.

AI literacy: Level of technological knowledge and competence in artificial intelligence, the study of which highlights the current absence in post-graduate dental training programmes.

Information intended for healthcare professionals. This content may contain errors or truncated summaries. We recommend always verifying with the original source article. Delynov disclaims all responsibility regarding the use of this information. This document is not intended for patients or the general public.