A newly adapted psychometric instrument based on the Technology Acceptance Model provides a statistically reliable and structurally valid method for evaluating how hospital physicians perceive and adopt remote electronic consultation platforms. Developed and evaluated within the public healthcare infrastructure of the Egyptian Ministry of Health and Population, the modified questionnaire demonstrated robust internal consistency and clear structural dimensionality when administered to medical practitioners receiving virtual specialist input. The validation analysis confirmed that the measurement tool accurately captures two distinct primary domains: perceived usefulness and perceived ease of use. By establishing strong internal reliability coefficients exceeding 0.92 across both domains and verifying the structural fit through advanced latent variable modeling, the investigation offers healthcare administrators a verified mechanism for diagnosing adoption barriers, assessing clinician satisfaction, and guiding future implementations of digital consultation workflows in resource-constrained hospital environments.
The expansion of digital health infrastructure has become an essential strategy for international healthcare systems striving to optimize specialized clinical delivery and advance health equity in geographically isolated and under-resourced regions. The World Health Organization has identified electronic consultation as a cost-effective, secure modality that facilitates clinical communication, professional medical education, and collaborative patient care across institutional boundaries. In Egypt, early limited-scale deployments mirrored similar initiatives across developing digital ecosystems, including programs established in Saudi Arabia, China, and India. The operational landscape shifted dramatically following the global outbreak of the COVID-19 pandemic in 2020, during which public quarantine facilities relied on virtual consultations to coordinate multidisciplinary patient management while mitigating contagion risks. Following that acute crisis, institutional utilization data prompted the Egyptian Ministry of Health and Population to double its operational footprint from 150 to 300 specialized e-consultation units across 26 national governorates, completing more than 27,000 formal clinical teleconsultations by late 2022, primarily concentrated in dermatology, internal medicine, and pediatrics.
Despite this rapid institutional scaling, healthcare planners possessed limited empirically grounded data describing how frontline physicians in public facilities experienced these digital platforms. In response to this evaluative gap, researchers adapted the foundational Technology Acceptance Model originally formulated by Davis, which models technological adoption primarily through the dual psychological constructs of perceived usefulness and perceived ease of use. While extensive literature documented diverse applications of this conceptual framework across telehealth settings in Arab countries, existing measurement instruments frequently addressed broader patient-facing telemedicine or generic population samples rather than physician-to-physician hospital interactions. Furthermore, among the four historical variations of the original model reviewed by Lewis, investigators selected version 4, which employs a seven-point Likert scale formatted with magnitude of agreement progressing sequentially from left to right. This specific ordering aligns with regional cultural preferences and captures authentic experiential usability rather than abstract likelihood of use. In adapting the framework, investigators strategically omitted the third theoretical construct of behavioral intention to preserve instrument brevity and lessen survey burden for overloaded clinicians.
The development of the specialized assessment tool was executed through a rigorous four-phase methodological progression designed to guarantee semantic clarity, conceptual equivalence, and cultural suitability. During the initial design phase, the selected items from the fourth version of the Technology Acceptance Model were translated into Arabic and adapted specifically to reflect the workplace realities of physicians requesting specialist guidance. To enrich the instrument beyond quantitative metrics, researchers introduced supplementary evaluative sections, culminating in a comprehensive 23-item questionnaire. This structural format comprised eight sociodemographic identifiers, twelve core psychometric scale items measuring perceived usefulness and perceived ease of use, a binary close-ended inquiry regarding overall service satisfaction, and two open-ended qualitative prompts asking respondents to outline operational challenges and suggest concrete system enhancements. In the third methodological phase, five iterative rounds of formal panel review involving multidisciplinary specialists in public health, behavioral science, and medical informatics established robust face and content validity, while the fourth phase secured parallel English translation through two independent bilingual professionals.
Prior to conducting nationwide psychometric validation, the investigative team executed a preliminary pilot evaluation to test administrative feasibility, comprehensibility, and formatting. This preliminary phase examined a targeted convenience sample of 17 practicing physicians based at a governmental hospital receiving routine e-consultation services in Alexandria. The pilot cohort exhibited a median age of 43 years, ranging between 38 and 55 years, alongside a median of 13 years of professional medical practice. Females constituted 70.6 percent of the pilot participants, and 94.1 percent reported urban residential status. Administration of the 23-item paper instrument required an average completion duration of approximately ten minutes. Detailed critical feedback gathered during these sessions guided essential modifications to phrasing, syntactic structure, and layout, eliminating ambiguity before final deployment in the larger multi-center study.
The comprehensive validation survey was implemented as a cross-sectional investigation conducted across 42 governmental hospitals under the administrative oversight of the Ministry of Health and Population, spanning 16 separate Egyptian governorates. The survey was hosted on an online platform utilizing Secure Sockets Layer encryption to protect data integrity during transmission and preserve strict participant anonymity. Recruitment took place over a two-week sampling window extending from January 30 to February 12, 2023, leveraging dedicated professional hospital communication groups on WhatsApp. Snowball sampling was employed to broaden participation, wherein initially identified physicians forwarded the secure survey link to eligible peers who were not integrated into primary administrative chat channels. Participation was strictly bounded by formal eligibility criteria, restricting enrollment to licensed physicians possessing at least one year of continuous clinical service in governmental hospitals who actively utilized the national telemedicine platform to obtain external specialist guidance, while explicitly excluding academic consulting physicians who provided rather than received teleconsultations.
Understanding the operational environment of the national e-consultation architecture provides critical contextual insight into the instrument’s ecological validity. The Ministry of Health and Population operates a centralized, dedicated telemedicine platform engineered for synchronous video interactions between clinical sites. Individual consultations follow structured 20- to 30-minute appointments dedicated to formal case presentation, interactive diagnostic deliberation, and therapeutic planning. Attending hospital physicians initiate consultations primarily for complex clinical scenarios or when essential subspecialty expertise is locally unavailable, such as specialized pulmonary facilities requiring urgent diagnostic input from cardiology, oncology, or gynecology. Prior to each scheduled session, the requesting physician prepares a comprehensive clinical record abstract, while facility information technology staff members upload diagnostic records, laboratory panels, and imaging files directly to the portal. Following the live session, the consulting specialist logs formalized written recommendations, triggering algorithmic workflow updates that classify the case as resolved, requiring supplementary diagnostics, or necessitating patient transfer to a tertiary center, with oversight from a dedicated administrative unit.
The nationwide analytical cohort encompassed 120 completed physician responses, satisfying statistical sample size adequacy guidelines that advocate a ratio of approximately ten respondents per scale item for robust factor analytic procedures. The demographic profile revealed a diverse clinical sample with a mean participant age of 37.3 years and a standard deviation of 7.6 years. Females accounted for 51.67 percent of the cohort, while 55.83 percent held senior specialist status, and 85.83 percent resided in urban administrative zones. The median professional medical experience among respondents was 10 years, spanning an overall range from 1 to 32 years of clinical practice. Participating practitioners were distributed across multiple hospital classifications, with 76.67 percent practicing in general hospitals, 11.66 percent in tropical and fevers hospitals, 7.50 percent in chest and pulmonary hospitals, 2.50 percent in pediatric institutions, and 1.67 percent in ophthalmology facilities. Geographically, respondents represented 16 governorates, led by South Sinai with 18 physicians, Alexandria with 16, and Daqahliya with 5, whereas the Red Sea governorate registered 1 participating physician. Internal medicine and pediatrics emerged as the primary medical specializations, representing 27 and 18 respondents respectively, alongside smaller representations across dermatology, intensive care, pulmonology, and neurology.
Rigorous statistical screening confirmed that the collected dataset fully satisfied all mathematical requirements for latent variable analysis. Assessment of univariate distributional assumptions demonstrated that all 12 core items produced mean ratings exceeding 5 on the seven-point continuum, reflecting generally favorable assessments, with the item assessing ease of understanding the system yielding the highest individual mean score of 5.92. Tests of sampling adequacy affirmed suitability for exploratory factor analysis, yielding a Kaiser-Meyer-Olkin metric of 0.925, well above the recommended 0.60 threshold. Simultaneously, Bartlett’s test of sphericity achieved high statistical significance, rejecting the identity matrix hypothesis at a probability level below 0.001. Exploratory factor analysis employing maximum likelihood extraction and principal component analysis paired with Varimax orthogonal rotation and Kaiser normalization extracted two major factors with eigenvalues exceeding unity, matching the scree plot inflection profile. These two extracted factors accounted for 81.17 percent of total data variance, with rotated factor loadings spanning from 0.661 to 0.912 and all item communalities comfortably exceeding the 0.50 retention threshold, ranging between 0.736 and 0.882.
The exploratory factor analysis distributed the 12 items across two clearly delineated latent domains that diverged slightly from the original Technology Acceptance Model structure. Factor 1, representing perceived usefulness, consolidated eight items reflecting how electronic consultations enhance clinical decision-making, optimize task efficiency, and facilitate consultant communication to improve patient care outcomes. Factor loadings for this primary domain ranged from 0.661 for clarity of interaction to 0.890 for general clinical work support, successfully incorporating two items originally theorized to measure ease of use. Factor 2, representing perceived ease of use, retained four items measuring operational ergonomics and learning demands, with loadings ranging from 0.785 for platform flexibility to 0.912 for ease of learning the interface. Subsequent confirmatory factor analysis utilizing structural equation modeling in R confirmed this two-factor specification without modifications, producing standardized structural loadings between 0.80 and 0.95, a Comparative Fit Index of 0.95, and a Root Mean Square Error of Approximation of 0.084, indicating acceptable model fit to the observed data.
Psychometric evaluation demonstrated exceptional internal consistency for the finalized 12-item questionnaire, yielding an overall scale Cronbach’s alpha of 0.95. When evaluated independently, the eight-item perceived usefulness subscale attained an internal consistency coefficient of 0.963, while the four-item perceived ease of use subscale demonstrated an alpha coefficient of 0.918. Aggregated construct scores exhibited positive skewness toward favorable perceptions, with perceived usefulness generating a mean score of 42.18 with a standard deviation of 12.8 across an empirical range of 8 to 56, and perceived ease of use averaging 23.1 with a standard deviation of 5.18 across an observed span of 4 to 28. Bivariate correlational analyses revealed a statistically significant positive relationship between perceived usefulness and perceived ease of use, with a Pearson correlation coefficient of 0.626. While chronological age and cumulative professional experience displayed an expected strong collinear correlation of 0.896, neither age nor professional experience exhibited any statistically significant association with either usefulness or ease of use ratings, indicating that technology acceptance was uniform across career stages.
While confirming the validity and practical utility of the modified assessment instrument, the investigators identified several inherent methodological boundaries that contextualize the findings. Restricting participation exclusively to governmental hospitals operated by the Egyptian Ministry of Health and Population restricts direct extrapolation to private healthcare providers, academic university medical centers, or non-governmental health networks. Furthermore, relying on convenience and snowball recruitment strategies introduced potential self-selection bias and constrained sample representativeness, while the self-administered survey format remains vulnerable to social desirability influences and subjective reporting discrepancies. Nevertheless, the resulting 12-item modified instrument, bolstered by supplementary qualitative sections and dual Arabic and English linguistic availability, provides health system administrators with a validated, efficient, and psychometrically sound framework to evaluate clinician engagement, address operational bottlenecks, and enhance digital consultation programs across developing healthcare environments.
Subject of Research: Validation of a modified Technology Acceptance Model questionnaire for hospital e-consultation services
Article Title: A reliable tool for assessment of acceptance of e-consultation service in hospitals: the modified e-consultation Technology Acceptance Model (TAM) questionnaire
Article References: Ashmawy, R., Zeina, S., Kamal, E., Shelbaya, K., Gawish, N., Sharaf, S., Redwan, E. M., & Mehanna, A. (2025). A reliable tool for assessment of acceptance of e-consultation service in hospitals: the modified e-consultation Technology Acceptance Model (TAM) questionnaire. Journal of the Egyptian Public Health Association, 100(1), Article 6. https://doi.org/10.1186/s42506-025-00187-x
Image Credits: AI Generated
DOI: 10.1186/s42506-025-00187-x
Keywords: Technology Acceptance Model, e-consultation, telemedicine, hospital physicians, construct validity, exploratory factor analysis, confirmatory factor analysis, perceived usefulness, perceived ease of use, Cronbach alpha, Egyptian Ministry of Health and Population, health equity, remote consultation, structural equation modeling, internal consistency, healthcare delivery
News Source: Ophelia Keating. (November 20, 2026). Modified Technology Acceptance Model Accurately Evaluates Hospital Physician E-Consultation Adoption. Scienmag.



