DefinePK

Abstract

Background: Artificial Intelligence (AI) is revolutionizing healthcare worldwide by improving diagnostics, treatment planning, and patient management. However, its effective integration into healthcare systems in low- and middle-income countries (LMICs) like Pakistan remains limited. Successful implementation largely depends on the knowledge, attitudes, and practices of healthcare professionals. This study aims to assess these factors among healthcare workers at a tertiary care hospital in Mardan, Pakistan, offering valuable insights from an LMIC context.
Objective: To evaluate the knowledge, attitudes, and practices of healthcare professionals regarding the use of Artificial Intelligence (AI) in healthcare settings.
Methods: A cross-sectional study was conducted from July to December 2024. Healthcare workers were consecutively surveyed using a validated KAP questionnaire with strong reliability (Cronbach’s alpha: 0.89 for KAP, 0.79 for awareness/behavior). Scores ranged from 0–12 for knowledge, 0–18 for attitude, and 0–16 for practice. Data were analyzed using descriptive statistics and group comparisons (t-tests, Chi-square, Mann–Whitney U) based on age, gender, occupation, and qualification; significance was set at p < 0.05.
Results: Among 250 participants (mean age 32.1 ± 6.1 years; 67% male), 58.3% were doctors, 30.9% paramedics, and 10.9% nurses. Mean scores were: knowledge 9.2 ± 2.9, attitude 12.8 ± 3.6, and practice 11.7 ± 3.6. Good knowledge, positive attitudes, and adequate practice were observed in 60%, 54.8%, and 64.8% of participants, respectively. Female participants and paramedics had comparatively higher KAP scores than their counterparts. Attitude differed significantly by occupation and education (p=0.01), while practice varied among diploma holders and paramedics.
Conclusion: Healthcare professionals in this LMIC setting demonstrate moderate knowledge, positive attitudes, and fair practices towards AI. Targeted education and policy initiatives are essential to improve AI readiness and integration in healthcare systems like Mardan.