Assessment instruments for e-professionalism in healthcare: a systematic review

Abstract

Introduction

E-professionalism represents the foundation guiding the conduct of health professionals on social media. Assessing e-professionalism allows the identification of gaps in professional education and supports institutions in adapting to technological changes. Gathering evidence on the quality of available instruments may promote reliable and consistent measurement of this construct.

Objective

To evaluate the evidence of validity of instruments used to assess e-professionalism among students and healthcare professionals.

Methods

A systematic review was conducted in accordance with the systematic review manual of the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN). The protocol was registered in PROSPERO (ID CRD42023454825). The search was carried out in July 2025 in the PubMed/MEDLINE, EMBASE, Web of Science, ERIC, and Scopus databases. Articles reporting e-professionalism instruments with evidence of validity were included, targeting professionals and/or students in Nursing, Pharmacy, Medicine, and Dentistry, with no restrictions regarding language or year of publication. The quality of the instruments was analyzed using COSMIN criteria.

Results

The search identified 4,616 studies, with 28 included, published between 2009 and 2025. The instruments were predominantly directed toward students (n = 20; 71.4%). Overall, the instruments contained an average of 26 items and three subscales. Only one of the studies established cutoff points for interpreting the results. Regarding psychometric properties, only four of the nine recommended by COSMIN were investigated, with emphasis on content validity (n = 24; 85.7%), internal consistency (n = 8; 28.5%), and structural validity (n = 7; 25%).

Conclusions

This review highlights the need for greater conceptual consensus regarding e-professionalism and its domains, as well as the development of instruments with robust validity for both students and health professionals. The lack of reliable tools limits comparisons and evaluations, reinforcing the importance of research that integrates theoretical definition with psychometric validation to accurately reflect the construct.

Introduction

Professionalism, from the perspective of the sociology of professions, refers to the way professional groups organize their practice, regulate activities, and construct social recognition, encompassing the norms and values that sustain the relationship between a profession and society, particularly the fiduciary relationship between healthcare professionals and patients [1, 2]. In daily professional practice, it is manifested through actions that reflect responsibility, ethics, social commitment, and adherence to professional values, thereby strengthening public trust and the professional’s prestige within the community [3, 4].

In the health field, the teaching of professionalism is essential, as in professional practice it can contribute to more qualified, responsible, safe, and ethical care, improving relationships among the healthcare team, patients, families, and society [4, 5]. Scientific studies have prioritized understanding how professional dilemmas are addressed in order to guide improvements in care and strengthen the profession [5, 6]. Thus, interest in research on professionalism in healthcare has increased in response to the growing demand for high-quality services [7, 8].

Although historically associated with the formal workplace, the concept has expanded into new spaces with the growth of social media, which have become additional arenas for professional expression [9]. Social media are Web 2.0 digital platforms that integrate interpersonal and mass communication, enabling users to create and share content. They include instant messaging applications, as well as online social networks, such as Instagram, Facebook, and TikTok [10, 11]. In this context, the concept of e-professionalism emerged, proposed by Cain and Romanelli (2009), to describe the extension of the norms and values of traditional professionalism to the digital environment.

The literature highlights that social media exposes professional behaviors to multiple interpretations, reinforcing the need for ethical rigor, attention to privacy, and integrity in online interactions [9, 12]. Their widespread use, reaching five billion users globally, has intensified their integration with professional practice, expanding their influence on how professional communication is produced and perceived, and creating challenges in balancing accessible communication with the ethical responsibilities of healthcare work, while also prompting institutional guidelines to address inappropriate conduct and protect public trust in health professions [13–15].

Studies indicate that social media behaviors influence relationships with patients, professional reputation, and institutional credibility, contributing to the collective image of health professions [16, 17]. Despite its relevance, evidence points to gaps in academic training and institutional policies related to e-professionalism, highlighting the need for educational strategies and guidelines that promote ethical conduct on social media [18, 19]. Therefore, the assessment of e-professionalism helps define professional boundaries in digital interactions and supports institutional adaptations to these emerging challenges [19, 20].

In this context, validated instruments for assessing e-professionalism among health professionals play a central role by enabling the identification of weaknesses, guiding educational practices, and supporting the development of institutional policies [18, 19]. Over the past decade, these instruments have increased in number but vary considerably, limiting comparability across studies. Robust measurement tools are therefore needed to generate reliable and comparable data for evidence-based educational and policy decisions [21, 22]. Accordingly, synthesizing the available evidence is necessary to support the selection of appropriate and theoretically consistent tools, as well as their integration into educational institutions and healthcare systems.

Objective

To evaluate the evidence of validity of instruments used to assess e-professionalism among students and healthcare professionals.

Methods

Study design

A systematic literature review was conducted in accordance with the manual for systematic reviews of the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) [23], and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Outcome Measurement Instruments (PRISMA-COSMIN) guidelines [24] (Supplementary file 1). The developed protocol was registered in PROSPERO (ID CRD42023454825) [25] and previously published [11].

Database search

A search strategy was developed to address the following question: “What evidence of validity exists for the instruments used to assess e-professionalism among students and healthcare professionals” The search was conducted in July 2025 in the PubMed/MEDLINE, EMBASE, Web of Science, ERIC, and Scopus databases, using descriptors related to “professionalism,” “e-professionalism,” and “social media,” as well as their combinations and synonyms (Supplementary file 2).

Study selection

The retrieved articles were managed using the Rayyan software [26]. The study selection followed the steps of duplicate removal, title and abstract screening, and full-text assessment. These stages were conducted by two researchers in a blinded and independent manner, and in cases of disagreement, a third researcher was responsible for reaching consensus.

Inclusion criteria

The review included studies that met the following criteria: (a) articles reporting e-professionalism instruments with evidence of validity, addressing at least one of the measurement properties described by COSMIN; and (b) articles whose target population consisted of professionals and/or students in Nursing, Pharmacy, Medicine, and Dentistry – as these professions constitute the core of regulated health professions in most countries [27, 28]. In studies conducted from a multiprofessional perspective, results had to be presented separately by profession. No restrictions were applied regarding language or year of publication.

Studies were considered eligible if they described instruments that explicitly referred to e-professionalism or to professionalism in social media. This criterion was adopted to ensure conceptual alignment with the construct of professionalism, rather than the inclusion of instruments focused solely on general online behaviors. This distinction is important because professionalism encompasses not only patterns of professional behavior but also normative dimensions related to the social role of professions, particularly in sustaining trust between health professions and society.

Literature reviews were excluded to avoid data duplication, as well as studies not published in scientific journals (e.g., theses and dissertations), abstracts from conference proceedings, letters to the editor, and conference papers, due to the limited information they provide, which may compromise the reproducibility and quality of the synthesis.

Data extraction

From the included studies, the following data were extracted: study title, country, language, author and year of publication, construct of interest, origin of the construct, variations in the terminology of e-professionalism, target population (profession), subscales, number of items, response options, score range, and scoring algorithm. To facilitate the identification of the analyzed instruments, those that did not have a defined name in the original articles were assigned a code composed of the first three letters of the first author’s surname and the year of publication.

Quality assessment of the instruments

The quality of the instruments was assessed using the Consensus-based Standards for the Selection of Health Status Measurement Instruments (COSMIN) checklist, which encompasses the following psychometric properties: content validity, structural validity, internal consistency, cross-cultural validity and measurement invariance, reliability, measurement error, criterion validity, hypothesis testing for construct validity, and sensitivity to change and responsiveness [29].

During the application of the COSMIN checklist, each study was assessed independently by two researchers who were previously trained in the use of the instrument. Standardized, preformatted data extraction tables, developed in accordance with the COSMIN manual, were used to ensure consistency in the evaluation process. In cases of discrepancies, a third researcher was responsible for reaching consensus. Each study was evaluated regarding its methodological quality, which was classified as very good (V), adequate (A), doubtful (D), or inadequate (I). In parallel, the results of the measurement properties reported in the studies were classified as sufficient (+), insufficient (–), inconsistent (±), or indeterminate (?).

Subsequently, these findings are synthesized using the GRADE approach adapted by COSMIN, through which the quality of the evidence is classified as high, moderate, low, or very low. This approach allows for a systematic assessment of the quality of the instruments by integrating methodological evidence with the reported results. This classification guides the selection of the most appropriate instrument, which ideally should present favorable results supported by high-quality evidence [29].

Results

The database search identified 4,616 articles. After the removal of 2,346 duplicates, 2,270 studies proceeded to the initial screening stage, during which 2,165 were excluded. A total of 105 articles were assessed for eligibility, of which 77 were excluded. At the end of the process, 28 articles were included (Fig. 1). Additionally, it was observed that no instrument was used in more than one study without modifications or adaptations.

Fig. 1.

Flow of the search and study selection in the databases

Characteristics of the included studies

The articles included in this review were published between 2009 and 2025. Asia accounted for 13 (46.4%) of the included studies. Within this continent, Saudi Arabia stood out with four publications (14.3%) [30–32], followed by Iran with three (10.7%) [33, 34] and China with two (7.1%) [35, 36]. In addition, individual studies were identified from the United Arab Emirates [37], South Korea [38], Singapore [39] and the Philippines [40] (3.5% each).

Of the included studies, nine (32.1%) were conducted in North America, exclusively represented by the United States [13, 41–47]. Additionally, six studies (21.4%) were conducted in Europe, including two (7.1%) from the United Kingdom [48, 49], three (10.7%) from Croatia [50, 51] and one (3.5%) from Greece [52]. Regarding language, 15 instruments (53.5%) were developed in English [13, 36, 38, 40–47, 49, 53, 54]. Nine instruments (32.1%) did not report the language of development—although they were published in English, they were developed in non-English-speaking countries [30–34, 37, 39, 52]. Additionally, two instruments (7.1%) were developed in Croatian [50, 55] and one (3.5%) in Chinese [35].

Among the 28 articles, 20 (71.4%) targeted students as the study population, demonstrating a clear predominance of investigations focused on individuals in training [13, 30, 33, 36, 38, 40, 41, 43, 44, 46–48, 50, 52, 53, 55, 56]. The distribution by profession showed a greater concentration in Medicine, represented in nine studies (32.1%) [13, 31–34, 39, 47, 56], followed by Pharmacy with eight studies (28.5%) [30, 38, 41–44, 46], Dentistry with four (14.3%) [32, 48, 49, 52], nd Nursing with three (10.7%) [35, 36, 40]. Additionally, four articles (14.3%) included more than one profession [37, 50, 51].

Regarding the construct of interest, 17 studies (60.7%) described e-professionalism as their primary focus, including terminological variations such as “online professionalism,” “social media professionalism,” and “professionalism” [30–33, 35, 37, 38, 42, 44, 47, 49, 53, 57]. Additionally, five instruments (17.8%) focused on patterns of social media use [32, 35, 42, 54], while three (10.7%) identified responsibility or responsible use of social media as the construct of interest [30, 40, 41]. Professionalism in virtual education was addressed in three studies (10.7%) [33, 34].

Regarding structure, the instruments had an average of 25 items, ranging from [53] to 63 [48]. The instruments included an average of three subscales, ranging from instruments without subdivisions [13, 32, 36, 54, 56] to those incorporating up to seven subscales [48]. The most recurrent subscale was e-professionalism, including terminological variations, reported in 16 studies (57.1%) [30, 31, 33, 35, 37, 38, 41, 42, 44, 47, 49, 53, 55]. This was followed by social media use, reported in nine articles (32.1%) [30–32, 37, 39, 42, 47, 49], and responsibility, identified in seven articles (25%) [30, 31, 37, 40, 44, 45, 53]. Subscales related to privacy settings were mentioned in five articles (17.8%) [31, 37, 39, 44, 51], while perceptions or opinions about online profiles appeared in four studies (14.3%) [37, 44, 48, 49] ethical aspects, honesty, and integrity were addressed in three articles (10.7%) [33, 37, 45].

Among the response formats, dichotomous responses were the most frequent, present in 12 instruments (42.8%) [13, 38, 39, 43, 44, 47, 48, 50, 52, 56, 57], followed by the five-point Likert scale, identified in nine instruments (32.1%) [13, 30, 36, 40, 45, 47, 49, 51] and multiple-choice responses, used in eight instruments (28.5%) [13, 32, 35, 39, 42, 49, 52, 56]. It was also observed that six instruments (21.4%) did not report the type of response format used [31, 33, 34, 37, 54]. Among the 28 articles, four (14.3%) mentioned the existence of a score [40, 45, 47, 50], however, only three of them (10.7%) described how this score was calculated [40, 45, 50]. It is noteworthy that only one instrument established a cutoff point to characterize levels of e-professionalism [40], which limits the interpretation and comparison of results across studies. Furthermore, the origin of the construct was described in only three articles (10.7%) [31, 42, 49]. The general characteristics of the included articles are presented in Table 1.

Table 1.

General characteristics of the included studies

Assessment instrument				Scope of the instrument			Structure of the instrument
Name	Country	Language	Author (year)	Construct of interest	Origin of the construct	Target population (N)	(Sub)scales (number of items) [Total number of items]	Response options	Score range (per (sub)scale)	Scoring algorithm
WAN2019	China	Chinese	Wang et al. (2019) [35]	Social media usage patterns and e-professionalism	NR	Nurses (658)	Social media usage patterns (14), social media professionalism (16) [27]	Multiple choice	NR	NR
RUSMAS	Philippines	English	Oducado et al. 2019 [40]	Responsible use of social media	NR	Nursing students (298)	Perceptions and attitudes regarding the responsible use of social media [12]	5-point Likert scale	Mean: negative attitude (0.00–2.33); ambivalent (2.34–3.66); positive attitude (3.67–5.00)	Mean
SUVSM	China	English	Zhu et al., (2020)[36, 58]					5-point Likert scale	NR	NR
CAI2009	United States of America	English	Cain et al. (2009) [41]	Responsibility and e-professionalism	NR	Pharmacy students (244)	E-professionalism (13); intention to change behavior (5) [15]	Dichotomous	NR	NR
CAI2013	United States of America	English	Cain et al., (2013) [42]	Social media usage patterns and e-professionalism	Traditional professionalism paradigms	Pharmacists (159)	Social media usage patterns (6); Facebook friending behavior (5); educational and professional use of social media (6); digital professionalism (4); social media and professional students (8) [26]	Multiple choice	NR	NR
GET2013	United States of America	English	Gettig; Lee; Fjortoft, 2013 [43]	E-professionalism	NR	Pharmacists and pharmacy students (296)	Observed scenarios (20); perception of professionalism (10) [27]	Dichotomous and 3-point Likert scale	NR	NR
NES2013	United States of America	English	Ness et al. (2013) [44]	E-professionalism	NR	Pharmacy students (212)	Responsibility (2), hiring decisions (2), profile editing (2), professionalism (2), and privacy settings (2); opinions about one’s online profile (13) [36]	Dichotomous and 3-point Likert scale Dicotômica	NR	NR
YAN2015	South Korea	NR	Yang et al., (2015) [38]	E-professionalism	NR	Pharmacy students (210)	Social media usage patterns (6), e-professionalism (11) [16]	Dichotomous	NR	NR
PSMPI	United States of America	English	Chisholm Burns et al. (2017) [45]	Profissionalismo em mídias sociais	NR	Pharmacy students (187)	Honesty and integrity (4), respect for others (2), responsibility (3), duty (3), and excellence (3) [13]	5-point Likert scale	Each subscale has a possible score range of 3 to 15	The total score is the sum of all 15 items, ranging from 15 to 75 points
AUG2015	United States of America	English	Augustine et al. (2015) [46]	Level of comfort, perceptions, and concerns regarding social media behaviors	NR	Pharmacy students (292)	Knowledge of institutional policies and disciplinary actions related to unprofessional online content (7). Comfort scale (9); concern scale (8) [24]	6-point Likert scale	Comfort scale (−2 to 2); concern scale (−5 to 3)	NR
ALM2021	Saudi Arabia	NR	Almetwazi et al. (2021) [30]	Professionalism and responsibility in social media use	NR	Pharmacy students (239)	Social media use (9); responsibility (6); professionalism (8) [20]	5-point Likert scale	NR	NR
BAR2012	United States of America	English	Barker et al. (2012) [54]	Social media usage patterns	NR	Physicians (66)	Does not describe subscales [11]	NR	NR	NR
JAI2014	United States of America	English	Jain et al. (2014) [47]	Online professionalism	NR	Physicians (1184) and medical students (237)	Demographic information (4), Facebook usage patterns (7), attitudes toward e-professionalism (29) [35]	Dichotomous and 5-point Likert scale	NR	NR
MOS2019	Iran	NR	Mosalanejad; tafvisi; sanaiey (2019) [34]	Professionalism in virtual education	NR	Physicians (NR)	Computer and information literacy (10), adherence to professionalism in interpersonal and group communications in the virtual environment (10), compliance with virtual rules and regulations (6), professional commitment in virtual business (4), attention to personal professionalism (7), professionalism in the healthcare field (4) [36]	NR	NR	NR
LOW2021	Singapore	NR	Low; Tan; Joseph, 2021 [39]	E-professionalism and professionalism	NR	Physicians (119)	Social media use (5); account privacy settings (5); use of instant messaging applications (3); professional cases/practices (7); physician–patient relationship (3); personal use of social media (2) [22]	Dichotomous and multiple choice	NR	NR
MOS2021	Iran	NR	Mosalanejad; Abdollahifard (2021) [33]	Professionalism in virtual education	NR	Medical students (250)	Compliance with laws and regulations governing cyberspace (9), individual professionalism in the use of cyberspace (9), knowledge management and information literacy (5), professionalism in interpersonal and group rules (5), compliance with ethics in the use of cyberspace (5) [30]	NR	NR	NR
MOS2022	Iran	English	Mosalanejad; Abdollahifard (2022) [53]	Professionalism in virtual education	NR	Medical students (157)	Hiring decisions (4), responsibility (2), and professionalism (4) [9]	NR	NR	NR
FIR2023	Saudi Arabia	NR	Firdos et al., (2023) [31]	E-professionalism	Traditional professionalism paradigms	Medical students (560)	Use of social media in education (7), social media use and privacy (7), responsibility and e-professionalism (9) [20]	NR	NR	NR
FAR2025	Saudi Arabia	English	Farsi et al., (2025) [56]	Knowledge of and adherence to the ethical standards of e-professionalism	NR	Medical students (400)	Does not describe subscales [17]	Multiple choice and dichotomous	NR	NR
KIT2016	United States of America	English	Kitsis et al. (2016) [13]	E-professionalism	NR	Physicians (614) and medical students (496)	Does not describe subscales [29]	Dichotomous, multiple choice, and 3-, 4-, and 5-point Likert scales	NR	NR
KEN2016	United Kingdom	English	Kenny; Johnson (2016) [48]	Online professionalism	NR	Dental students (155)	Frequency of social media use (3), social media behaviors (8), perceptions of online behaviors (16), witnessed behaviors (16), self-reported behaviors (16), intentions to change social media profiles (4) [59]	Dichotomous and 5-point Likert scale	NR	NR
DOB2019	United Kingdom	English	Dobson; Patel; Neville. (2019) [49]	E-professionalism	Traditional professionalism paradigms	Dental students (88)	Social media usage patterns (6); knowledge of e-professionalism guidelines (2); perceptions of e-professionalism (21); and accessibility/security on social media (8) [33]	Dichotomous, multiple choice, and 5-point Likert scale	NR	NR
KAR2020	Greece	NR	Karveleas et al., (2020)[52]	E-professionalism	NR	Dental students (512)	Facebook profile settings and main reasons for its use (8); Facebook content shared by dental students (5); student–patient relationship via Facebook (6) [17]	Dichotomous and multiple choice	NR	NR
BAH2021	Saudi Arabia	NR	Bahabri; Zaidan, (2021) [32]	Social media usage patterns and professionalism	NR	Dentists (238)	Does not describe subscales [20]	Multiple choice	NR	NR
ePaci	Croatia	Croatian	Viskić et al., (2022) [50]	E-professionalism	NR	Physicians (507) and dentists (246)	Somewhat unprofessional behavior (6), professional behavior (4), seriously unprofessional behavior (4), illegal and condoning behavior (3) [16]	4-point Likert scale and dichotomous	-1 to 1	The ePaci was obtained by summing the 15 items (−1/0/+2) and standardized to a range of −1 to +1 using:2 × [(i − iₘᵢₙ)/(iₘₐₓ − iₘᵢₙ)] − 1.
HUS2021	United Arab Emirates	NR	Hussain et al. (2020)[37]	E-professionalism	NR	Pharmacy students (325) and medical students (250)	Use of social media in education (3); perceptions of social media in education (7); responsibility and professionalism (8); online privacy (8) [23]	NR	NR	NR
Smeprof-S1scale	Croatia	Croatian	Marelić et al., (2021) [55]	E-professionalism	NR	Medical students (411) and dental students (287)	Ethical aspects (5), risks of social media (5), exclusion of physicians (2), freedom of choice (3), importance of professionalism (3), physicians in the digital era (3), and negative consequences (3) [21]	5-point Likert scale	NR	NR
Smeprof-S2scale	Croatia	Croatian	Marelić et al., 2024 [51]	E-professionalism	NR	Physicians (507) and dentists (246)	Confidentiality (5), privacy (5), contact with patients (4), and fair distribution of resources (1)	3-point frequency rating scale	NR	NR

NR: Not reported

Assessment of measurement properties

Among the nine measurement properties included in the COSMIN framework, only four were investigated in the articles included in this review. Content validity emerged as the most frequently evaluated property, reported in 24 studies (85.7%), followed by internal consistency, addressed in eight studies (28.5%), and structural validity, examined in seven studies (25%). Reliability, in turn, was the least explored dimension, being reported in only three studies (10.7%) (Table 2).

Table 2.

Classification of the quality of evidence and results of the instruments according to COSMIN criteria

Profession	Instrument	Content validity	Structural validity	Internal consistency	Reliability
Nursing	WAN2019	Very low (?)
	RUSMAS	Very low (+)		High (+)
	SUVSM		Moderate (-)	High (+)	Low (+)
Pharmacy	CAI2009	Very low (±)
	CAI2013	Very low (-)
	GET2013	Very low (±)
	NES2013	Very low (±)
	YAN2015	Very low (±)
	PSMPI	Very low (-)	Moderate (-)	High (+)
	AUG2015	Low (±)	High (?)
	ALM2021	Very low (+)
Medicine	BAR2012	Low (-)
	JAI2014	Very low (?)
	MOS2019	Very low (?)
	LOW2021	Very low (+)
	MOS2021	Moderate (-)	Moderate (-)	High (+)	Low (+)
	MOS2022	Very low (?)	Moderate (+)	Low (+)	Low (+)
	FIR2023			Low (+)
	FAR2025	Very low (±)
	KIT2016	Very low (+)
Dentistry	KEN2016	Very low (±)
	DOB2019	Very low (+)
	KAR2020	Very low (±)
	BAH2021	Very low (?)
Multiprofessional	ePACI		Low (?)
	HUS2021	Very low (±)
	SMePROf-S¹*	Low (+)	Moderate (?)	High (+)
	SMePROF-S²*			Low (+)

Classification of results: Sufficient (+); Insufficient (–); Inconsistent (±); Indeterminate (?)

*Although they share the same name and authorship, they do not correspond to the same instrument. In the SMePROf-S2 version, the items were modified to better suit the target population. Therefore, their results cannot be synthesized

Content validity

Content validity, comprising instrument development and content validation studies, showed limited methodological quality. Instrument development studies were predominantly inconsistent, whereas content validation studies were rated as doubtful (Table 3). In addition, most content validity studies presented inconsistent results (n = 9; 37.5%) [35, 42, 49–51, 54, 59–61] and the majority showed a very low level of evidence (n = 20; 83.3%) [15, 33, 35, 36, 38–40, 42–47, 49, 50, 53–55, 59, 61]. Content validity studies were more frequent in instruments developed for Pharmacy [33, 43, 46, 47, 49–51, 61] and Medicine (n = 8; 33.3% each) [15, 35, 37–39, 44, 52, 53], followed by Dentistry (n = 4; 16.6%) [36, 55, 59, 62], Nursing (n = 2; 8.3%) [40, 45] and multiprofessional instruments (n = 2; 8.3%) [42, 57].

Table 3.

Risk of bias assessment according to the COSMIN criteria

Instrument	Content validity		Structural validity		Internal consistency		Reliability
	Quality of the development study	Quality of the validation study	Result	Study quality	Result	Study quality	Result	Study quality
WAN2019	I	D
RUSMAS	I	D			Cronbach’s α = 0,76	V
SUVSM			EFA: 59.4% of the total variance	A	Cronbach’s α = 0,80	V	ICC = 0,85	D
CAI2009	I	D
CAI2013	I	D
GET2013	I
NES2013	I
YAN2015	I
PSMPI	I	D	EFA: 61.0% of the total variance	A
AUG2015		D	1.4 > MNSQ < 0.6	V
ALM2021	I
BAR2012		D
JAI2014	I
MOS2019		D
LOW2021
MOS2021		D	EFA: 51.0% of the total variance	A	Cronbach’s α = 0,78	V	ICC = 0,81	D
MOS2022		D	EFA: 71.8% of the total variance	A	Cronbach’s α = 0,94	D	ICC = 0,82	D
FIR2023					Cronbach’s α = 0,92	D
FAR2025	D
KIT2016	I
KEN2016	I
DOB2019	I
KAR2020	I
BAH2021	I	D
ePACI			EFA: 54.4% of the total variance	D
HUS2021	I
SMePROf-S¹		D	EFA: 65.5% of the total variance	A	Cronbach’s α = 0,72	V
SMePROF-S²					Cronbach’s α = 0,70	D

Study quality classification: Very good (V); Adequate (A); Doubtful (D); Inadequate (I)

EFA: Exploratory Factor Analysis

Six instruments presented sufficient results. Of these, two (8.3%) corresponded to instruments developed for Medicine [15, 44], while Nursing [45], Pharmacy [33], Dentistry [55] and one multiprofessional instrument [57] each had one instrument (4.1%) with sufficient results. Although these instruments demonstrated positive results, they were classified as having a very low level of evidence. The only instrument that showed comparatively better performance was the SMePROf-S¹ [57], developed for multiprofessional contexts, which obtained a sufficient result, although it was still classified as having low-quality evidence (Table 2).

Internal consistency

Studies evaluating internal consistency involved instruments for Medicine (n = 3; 37.5%) [34, 37, 38], Nursing (n = 2; 25%) [41, 45], multiprofessional instruments (n = 2; 25%) [57, 58] and Pharmacy (n = 1; 12.5%) [50], with no records for Dentistry. The majority of studies reported sufficient results with high-quality evidence. The MOS2022 and FIR2023 instruments demonstrated sufficient results; however, they were classified as having low-quality evidence [34, 38]. Additionally, among the evaluated instruments, SMePROf-S² showed the least favorable performance, with insufficient results and low-quality evidence [58] (Table 2).

Structural validity

Structural validity was most frequently investigated in Pharmacy (n = 2; 25%) [50, 51] and Medicine (n = 2; 28.5%) [37, 38], while the other professions were represented by only one study each. The majority of studies (n = 5; 71.4%) demonstrated adequate methodological quality according to the COSMIN risk of bias checklist [37, 38, 41, 50, 58] (Tabela 3), resulting in a moderate level of evidence (Table 2). Nevertheless, most results were classified as insufficient [37, 41, 50] or indeterminate [56, 57]. Among the evaluated instruments, MOS2022 showed the best performance, classified as sufficient with a moderate level of evidence [38].

Reliability

This was the least explored property among the instruments included in this review, being assessed in only three studies. Of these, two were conducted for Medicine (n = 2; 66.7%) [37, 38] and one for Nursing (n = 1; 33.3%) [41]. In all cases, the results were classified as sufficient, although the quality of evidence was considered low. Overall, regardless of the target population, the instruments exhibited similar methodological gaps.

Most studies employed instruments applied in a single cross-sectional assessment and without clear interpretation criteria, which limits their reproducibility. This approach reduces the ability to generalize findings and hinders consistent comparisons across different contexts and populations, consequently weakening the available evidence base for the assessment of e-professionalism.

Discussion

In the context of e-professionalism, understood as the extension of traditional professionalism principles to digital environments, the results of this review allow an analysis of how the construct has been operationalized among health professionals and students [10]. The findings indicate that the earliest studies on this topic emerged in a context characterized by the expansion of social media use and the need for guidelines to guide professional conduct in digital settings.

In this scenario, aspects such as ethical behavior, appropriate communication, and responsibility in disseminating health-related information become particularly important. Furthermore, although professionalism in healthcare is, from an ontological perspective, linked to the fiduciary relationship between professional and patient, differences in digital inclusion, healthcare systems, and institutional policies may influence the understanding and operationalization of e-professionalism across different contexts [10, 63, 64].

A significant increase in scientific production on e-professionalism has been observed since 2019. This growth appears to be associated with the intensified use of social media for professional communication and the expanded circulation of information in digital environments [63]. The COVID-19 pandemic further accelerated this trend, consolidating digital platforms as tools for interaction and knowledge sharing [65]. This context also reinforced academic interest in the topic, reflecting society’s concern with ensuring ethical, competent, and responsible conduct by health professionals in virtual settings [65, 66].

This review indicates that dentistry has a lower volume of research on e-professionalism. The literature highlights a greater emphasis on the commercial dimensions of dental practice, such as e-branding and the activity of influencer dentists, revealing tensions between clinical practice and professional marketing. Furthermore, e-professionalism in dentistry is relatively more recent compared to other health professions and is still in a developmental stage, with a gradual shift from normative approaches toward strategies based on self-regulation and the formation of digital professional identity [67, 68].

The geographic distribution of the studies indicates interest in e-professionalism across different contexts, particularly in the United States and in Asian countries, which represent the largest global user base of social media [69]. In some Asian countries, social, cultural, and political factors may influence forms of online participation, whereas in Western democracies, greater freedom of expression tends to favor digital interactions [70, 71]. These differences affect not only social media use but also research priorities and the development of instruments to assess e-professionalism, which may reflect both the need to monitor practices in more restrictive contexts and to guide conduct in more open digital environments.

The predominance of studies involving students can be partly explained by the ease of recruitment and data collection in academic settings [72, 73]. In addition, students are particularly relevant for research on professionalism, as they are in the process of developing and consolidating their professional identity [7, 74]. Nevertheless, the relative scarcity of studies involving practicing professionals highlights an important gap in the literature, given that e-professionalism also manifests significantly in clinical practice and everyday professional interactions.

The instruments identified in this review converge on a set of traits associated with e-professionalism, such as responsibility, ethics, integrity, and privacy settings. These findings are consistent with the systematic review by Guraya, Guraya, and Yusoff (2021). However, the authors note that there is no consensus on ideal behaviors, although they describe unprofessional behaviors on social media among health professionals. These behaviors include violations of patient confidentiality, lack of awareness of privacy settings, inappropriate descriptions of professional roles, use of obscene language, and derogatory or prejudiced comments [20].

Despite this thematic convergence, e-professionalism has been operationalized in heterogeneous ways across the available instruments. In some cases, the construct is treated as the central focus of assessment, while in others it appears as a subscale integrated into broader instruments, whose dimensions often include traits inherent to e-professionalism. This pattern reinforces previous findings, which highlight the lack of conceptual clarity as a persistent challenge in the field [23]. It is important to note that the absence of conceptual clarity impacts the content validity of instruments [75]. This is reflected in the results of this review, which indicate a low level of evidence in development studies, particularly due to the limited coverage of key aspects of e-professionalism.

Although e-professionalism is conceptually derived from traditional professionalism, only a small portion of studies describe the theoretical frameworks that guided the development of the measures. This theoretical gap generates limited understanding of the construct, which compromises various stages of the measurement process, including content validity and the definition of instrument structure. The literature describes reflective instruments, in which items reflect a single underlying construct, and formative instruments, in which distinct items collectively define the construct. Furthermore, complex constructs may adopt hybrid models [75, 76]. Considering the multidimensional nature of e-professionalism, this construct is best represented by a formative structure composed of multiple traits, each of which can be assessed through reflective subscales.

Considering that formative and reflective models differ in terms of the direction of causality, the pattern of correlations among indicators, and the way scores are composed, selecting the appropriate model for an instrument is crucial to ensure measurement validity and accurate interpretation of results [75, 77]. However, none of the identified studies clearly specify the model used, which complicates the interpretation of both the measures and the results. Consequently, the wide variation in subscales and response formats across instruments hampers comparability between studies, as each instrument ultimately measures the construct in different ways.

Regarding structural validity, the results of this review highlight weaknesses in the evaluated instruments. Overall, the findings were unsatisfactory, with only moderate methodological quality. Under these conditions, there is no assurance that the items adequately reflect the dimensionality of the construct [75, 78]. Furthermore, most studies reporting structural validity do not provide scores or cutoff points, which prevents meaningful interpretation of the scores. As a consequence, the instruments have limited utility, as they do not allow for score interpretation or for evidence-based decision-making in research or professional practice [79, 80].

In this review, internal consistency was the psychometric property that showed relatively more favorable results and higher quality of evidence among the evaluated instruments. However, in several studies, this property was reported in isolation, without adequate evidence of structural validity or in the presence of unsatisfactory results for that property. Under these circumstances, the interpretation of scores becomes limited, compromising confidence in the measures [32]. Regarding reliability, although the studies that assessed it reported satisfactory results, the low quality of evidence indicates significant limitations. Thus, there is no guarantee that the instruments demonstrate consistent stability and precision across different conditions or time points [75, 81].

Although e-professionalism instruments present significant limitations, these weaknesses are not unique to this field. Studies indicate that measurement instruments in various areas of health face similar challenges, often with limited evidence and, in most cases, insufficient methodological quality [81–83]. In such cases, it is recommended that instruments be used with caution and on a provisional basis. Furthermore, emphasis should be placed on strengthening the theoretical foundation and content validity before undertaking new efforts for psychometric validation [75, 83].

According to COSMIN criteria, none of the identified instruments can be formally recommended [32]. However, some instruments showed relatively better psychometric results than others. Still, they are considered only for provisional use because they present insufficient or low-quality evidence, which does not allow for a strong recommendation either in favor or against their use [83]. As these instruments are more recent than others, this was also considered in the classification and may partly explain the limited evidence on validity [27].

These instruments are: RUSMAS (Nursing), ALM2021 (Pharmacy), LOW2021 (Medicine), DOB2019 (Dentistry), and SMePROf-S¹ (multiprofessional contexts). Among these, only LOW2021 was developed for practicing professionals, highlighting the need to explore e-professionalism in professional practice. Furthermore, Given the emergent nature of the construct and the absence of instruments supported by robust evidence, their use should be cautious, with interpretations restricted to their specific scope [82].

It is important to note that the present study has some constraints. The search strategy and inclusion criteria employed may have excluded potentially relevant studies. In addition, the lack of conceptual and theoretical consensus on e-professionalism remains a challenge for evaluating instruments in this area, as the specific nuances of the construct may not be fully captured by standardized criteria. Regarding strengths, to the best of our knowledge, this is the first systematic review to compile and analyze instruments for assessing e-professionalism in health. Moreover, the use of COSMIN provided rigor to the analysis and allowed for the identification of both the strengths and limitations of the evaluated instruments.

Conclusion

Although no instrument was recommended according to the COSMIN criteria, this review identified promising instruments that may be used but still require the strengthening of their validity evidence through additional studies. The findings also highlight the need for conceptual consensus regarding e-professionalism, given the existing terminological diversity, and reinforce the importance of a solid theoretical foundation to guide the development of reliable and replicable instruments.

Future research should prioritize the conceptual definition and theoretical validation of e-professionalism, ensuring that instruments accurately represent the construct and its multiple dimensions. Such advances will provide support for educational institutions and professional organizations to improve their guidelines and training strategies for professional conduct in digital environments, promoting ethical, safe, and practice aligned with contemporary demands.

Authors’ contributions

All authors contributed to the study. LGS, FCAN, TMAT and DPLJ made substantial contributions to the conception and design of this study. LGS, DMS, TMAT and LTCNB collected, analyzed, and interpreted the data. LGS, LJSCV and FCAN wrote the manuscript. LJSCV adjusted the manuscript in accordance with the terms of the journal. FCAN, ARM, ADDOF and DPLJ revised the manuscript and made important contributions to the structure and content of it. All authors accepted the final version of the manuscript.

Funding

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001.

Data availability

As this is a review study, the analyzed data were obtained from previously published studies available in the scientific literature. The synthesized data are presented in this article. The data extraction spreadsheets, as well as the assessments of methodological quality and instrument properties conducted by the authors, are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.