Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 https://doi.org/10.1186/s12875-020-1081-x RESEARCH ARTICLE Open Access Efficiency as a determinant of loyalty among users of a Community of Clinical Practice: a comparative study between the implementation and consolidation phases David Lacasta Tintorer1,2,3, Josep Maria Manresa Domínguez2,3, Ana Jiménez-Zarco4, Teresa Rodríguez-Blanco3,5, Souhel Flayeh Beneyto1, Pere Torán-Monserrat2,6, Xavier Mundet Tuduri3,7 and Francesc Saigí-Rubió8* Abstract Background: A community of clinical practice called the Online Communication Tool between Primary and Hospital Care (ECOPIH) was created to enable primary care and specialist care professionals to communicate with each other in order to resolve real clinical cases, thereby improving communication and coordination between care levels. The present work seeks to analyse whether ECOPIH makes it possible to reduce the number of referrals. To that end, the objectives are: (1) To find out the degree of loyalty among ECOPIH users, by comparing the medical professionals’ profiles in the tool’s implementation phase to those in its consolidation phase. (2) To evaluate the degree of fulfilment of users’ expectations, by establishing the determining factors that had an influence on the physicians’ intention to use ECOPIH in the implementation phase and observing whether its use had an effective, direct impact on the number of patient referrals that primary care physicians made to specialist care professionals. Methods: Two studies were conducted. Based on a survey of all the physicians in a Primary Care area, Study 1 was a descriptive study in ECOPIH’s implementation phase. Study 2 was a randomised intervention study of ECOPIH users in the tool’s consolidation phase. The results from both studies were compared. Various bivariate and multivariate statistical techniques (exploratory factor analysis, cluster analysis, logistic regression analysis and ANOVA) were used in both studies, which were conducted on a sample of 111 and 178 physicians, respectively. Results: We confirmed the existence of an ECOPIH user profile stable across both phases: under-50-year-old women. Regarding the second objective, there were two particular findings. First, the discriminant factors that had an influence on greater ECOPIH use were habitual Social media website and app use and Perceived usefulness for reducing costs. Second, PC professionals who were ECOPIH members made fewer referrals to SC professionals in Cardiology, Endocrinology and Gastroenterology than older PC professionals who were not ECOPIH members. Conclusions: The use of a community of clinical practice by primary care and specialist care professionals helps to reduce the number of referrals among medical professionals. Keywords: Remote consultation, Primary health care, Problem solving, Telemedicine, Referral and consultation, Continuing medical education * Correspondence: fsaigi@uoc.edu 8 Faculty of Health Sciences, Universitat Oberta de Catalunya, Barcelona. Av. Tibidabo, 39-43, 08035 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 2 of 12 Background based on a Web 2.0 platform. It facilitates communica- In the current context of healthcare spending contain- tion between PC and SC professionals respectively work- ment, the role of primary care (PC) is fundamental be- ing at a number of PC centres and hospitals in the cities cause, when managed effectively, it can prevent of Badalona and Sant Adrià de Besòs in greater Barce- unnecessary referrals and reduce waiting lists [1–3]. lona, Spain [44]. It enables PC and SC professionals to However, people with multiple and complex health share up-to-date information that is relevant to their in- problems are cared for in PC clinics [4]. This means that terests, and PC professionals to raise clinical cases for physicians have to deal with several clinical aspects of consultation with specialists to improve patient manage- patients at once; physicians may have doubts about how ment and to reduce the number of referrals to the next to manage complex patient needs in day-to-day clinical care level. After a two-year follow-up period (2011– practice [5–8]. PC professionals therefore need an effect- 2012), 1000 interventions had been made across six spe- ive system that allows them to perform searches and find cialities through ECOPIH. Contributions had been read the necessary information to enhance their knowledge 12,200 times (each contribution approximately 10 times) and find suitable solutions [9]. and 209 clinical cases had been raised for consultation. Face-to-face or telephone discussions with specialist Presented in this article are the results from two ECO- care (SC) professionals enable PC professionals to ad- PIH follow-up studies conducted between 2011 and dress particular clinical concerns that crop up during pa- 2012, coinciding with the respective implementation tient care. However, given that the health system is at (first year of ECOPIH use) and consolidation phases of saturation point, communication between PC and SC that CoCP (end of the second year of use). Study 1 eval- may be difficult, slow, and ineffective [10–13], and it uated the discriminant factors that had an influence on leads to many referrals to SC (hospitalisation or special- the intention to use ECOPIH, and Study 2 performed a ist outpatient clinics). In turn, this leads to excessive de- characterisation of ECOPIH users and analysed the im- lays for appointments [14, 15] and to a significant pact of ECOPIH use on referrals. By comparing the re- increase in financial, time and psychological costs to sults obtained from the two studies conducted, the physicians and patients. As Horner et al., have pointed present work seeks to analyse whether ECOPIH makes it out, 65% of referrals are inappropriate and up to 30% of possible to reduce the number of referrals to SC. To that them could be avoided [16]. end, the objectives are (Fig. 1): Among the factors associated with a higher referral (1) To find out the degree of loyalty among ECOPIH rate are the little coordination between care levels and users. Specifically, by identifying and then comparing the the lack of training [10, 17–19]. Improving coordination medical professionals’ profiles in the tool’s initial imple- between care levels would not only enhance healthcare, mentation phase (Study 1) to those in its consolidation but also be of considerable educational value and lead to phase (Study 2) to see if they remained stable. a more cost-effective use of health services [2, 20]. Tele- (2) To evaluate the degree to which users’ expectations medicine can improve communication between PC and are fulfilled by establishing the determining factors that SC, and thereby improve efficiency, cost-effectiveness, influence physicians’ intention to use ECOPIH in the and medical care quality [21–27], with a high degree of implementation phase (Study 1) and analysing whether patient satisfaction [28–30].. In addition, telemedicine the expectations identified in that phase were fulfilled by can reduce the number of supplementary tests and refer- observing whether its use had an effective, direct impact rals to SC (by between 8.9 and 51%) [30–32]. on the number of patient referrals that PC physicians The formation of communities of practice (CoPs) is a made to SC professionals (Study 2). recent approach [33]. Applied to the field of healthcare, communities of clinical practice (CoCPs) are online plat- forms that draw on the advantages of Web 2.0 to con- Methods struct knowledge among healthcare professionals The research presented in this article is the result of a working at different levels of care [34]. Although there is collaboration between the Badalona-Sant Adrià de Besòs limited evidence of their usefulness [35, 36], CoCPs have Primary Care Service (PCS) in Catalonia, Spain, and the been shown to have considerable capacity to enable the Open University of Catalonia (UOC, as abbreviated in transfer of knowledge gained in day-to-day practice [37– the Catalan language), Spain. Some of the results ob- 39], as well as a lot of potential in terms of professionals’ tained from this study relating to physicians’ and nurses’ education, regardless of their care level [36, 40–43]. adoption of the tool have been published elsewhere [45]. Set up in 2009, Eina de Comunicació Online entre Pri- The Badalona-Sant Adrià de Besòs PCS includes nine mària i Hospitalària (ECOPIH as abbreviated in Cata- PC centres and three SC centres: Germans Trias i Pujol lan, or Online Communication Tool between Primary University Hospital, Badalona Municipal Hospital and and Hospital Care as translated in English) is a CoCP the Barcelonès Nord International Health Unit, with a Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 3 of 12 Fig. 1 Objectives and results obtained from the two studies conducted in the present work total of 624 healthcare professionals. These centres serve total of 111 physicians who filled in the questionnaire 227,151 inhabitants. (31.9% response rate). The study was conducted over a Two studies were conducted. Study 1 was a descriptive two-month period, from 1 December 2011 to 31 January study in ECOPIH’s implementation phase. Study 2 was a 2012. As reflected by the above-mentioned percentage, randomised intervention study of ECOPIH users con- the high response rate meant that it was a representative ducted two years later, when use of the tool had become sample of the ECOPIH community population. Also, consolidated. given the population’s finite size, it was possible to work with low margins of error (+ 7.7, 95% confidence level). Study 1 (implementation phase): characterisation of This meant that the results could be extrapolated. Table 1 ECOPIH users during the implementation phase and presents the study specifications, and Table 2 the variables description of discriminant factors that have an influence used in the study. on the intention to use ECOPIH Settings, sampling and type of study Study 1 was a descriptive study on a sample of health- Variables of study 1 care professionals from the Badalona-Sant Adrià de The use of ECOPIH is a dependent variable and the rest Besòs PCS, who had the potential to become regular are independent (explanatory) ones (Table 2). It should be users of ECOPIH. An anonymous, optional survey con- pointed out that principal component analysis was used to sisting of open and closed questions was used to gather obtain the two variables measuring perceived usefulness – data [45]. Perceived usefulness for improving clinical practice quality, The questionnaires were divided into three sections: 1) and Perceived usefulness for reducing costs. It was the sociodemographic and professional background; b) access multidimensional nature of these variables that suggested to and use of Information and Communication Technolo- exploratory factor analysis (EFA) should be performed. gies (ICTs) in professional and personal settings; and c) EFA is a technique to reduce data dimensionality. By ana- perceptions and use of ECOPIH. Information about the lysing a set of original variables, it seeks to determine the scientific objectives and data confidentiality was made fewest dimensions capable of explaining the maximum available to potential respondents. A total of 357 health- amount of information within the data [46]. care professionals were invited, by e-mail, to fill in the In total, nine variables were considered for the pur- questionnaire. All of those professionals were (a) involved pose of extracting the factor dimensions. Each variable in caring for patients in the Badalona-Sant Adrià de Besòs was associated with the healthcare professionals’ per- PCS and (b) could potentially become routine ECOPIH ceived benefits of using ECOPIH. Regarding the particu- users. All professionals were invited to take part by e-mail, lar benefits that ECOPIH could offer its users, some of though it was made clear that they could decline if they these variables referred to quality improvement, while wished. The questionnaires were provided in add- others referred to cost reduction (see Table 3 in Lacasta itional files 1 and 2. The study sample was formed by a et al. [45]). Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 4 of 12 Table 1 Study specifications Study 1 Study 2 Universe 357 healthcare professionals 357 physicians Sample 111 physicians 178 physicians Margin of error 7.7% (p = q) 95% confidence level 5.2% (p = q) 95% confidence level Data collection method Questionnaire Official or institutional electronic records, from the clinical records management program, 2010 to 2012 Sampling method Randoma Randomb Background work December 2011 December 2012 a Based on an anonymous survey. b Randomised intervention study of ECOPIH users. As the 2015 study showed [45], performing a set of the scales. According to Nunnally [47], this indicator statistical tests enabled us to establish the suitability of must have values higher than 0.7 in general and higher the analysis and the reliability of the scale. All of the cor- than 0.6 in the case of new scales. Thus, it is possible to relation matrix’s variables displayed high correlations, assume that the scales used were reliable. In addition, and the value of their determinant was 0.041. The the discriminant, convergent and nomological validity of Kaiser-Meyer-Olkin index value was 0.924 and Bartlett’s the content and construct scales was addressed. Regard- test of sphericity value was 1983.717, with a significance ing the content, the scales were developed following a of 0.000. This analysis explained 86.846% of the variance, major review of the literature (see Table 3 in Lacasta and Cronbach’s alpha values were higher than 0.81 in all et al. [45]). Table 2 Variables used in Study 1 Model variable The use of ECOPIH The healthcare professional uses ECOPIH. Dichotomous variable, where 0 = no and 1 = yes. Perceived usefulness of ECOPIH Perceived usefulness for improving Metric variable obtained from a principal component clinical practice quality (PU1) analysis (see Annex 2) determining the extent to which the healthcare professionals perceived that ECOPIH use improved clinical practice quality. Perceived usefulness for reducing Metric variable obtained from a principal component costs (PU2) analysis (see Annex 2) determining the extent to which the healthcare professionals perceived that ECOPIH use reduced clinical practice costs (in time and effort invested in getting hold of information). Perceived ease of Variable measured on a 5-point Likert scale indicating use of ECOPIH the healthcare professionals’ perceived ease of use of ECOPIH. Security and confidentiality Variable measured on a 5-point Likert scale indicating the level of patient data security and confidentiality that ECOPIH has. Healthcare professional Dichotomous variable indicating the individual’s profile professional profile. 1 = physician and 0 = nurse. ICT user profile Mobile device use Categorical variable indicating the extent to which the ICT user uses different types of mobile device. 1 = low, 2 = medium, 3 = high, 4 = very high. Social media website and app use Categorical variable indicating the extent to which the ICT user uses social media technologies (access to social networks). 1 = low, 2 = medium, 3 = high and 4 = very high. Gender Gender of the healthcare professional. 1 = female and 0 = male. Age Age of the healthcare professionals. The variable has four values: 1 = under 40 years old, 2 = between 40 and 49 years old, 3 = between 50 and 59 years old, and 4 = 60 years old or over. Source: Lacasta et al. [45] Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 5 of 12 In order to establish the physicians’ profiles, univariate (Study 1). After the tool’s consolidation, the professionals’ analyses were performed on the different sociodemo- profiles were analysed again, taking into account the ef- graphic and ICT use variables for the selected sample. fective use of the tool (Study 2). By comparing the profiles An important matter was to identify the physicians’ pro- obtained from the two studies, it was possible to find out files by gender and age. To that end, hierarchical cluster if potential users had become actual users (loyalty). analysis was carried out. Regarding the second objective, in order to confirm To identify the variables determining ECOPIH use, whether the tool actually had an influence on the profes- binary logistic regression analysis (Logit) was performed. sionals’ behaviour, an analysis was performed of the rela- tionship between the profiles of the professionals using Study 2 (consolidation phase): characterisation of ECOPIH ECOPIH (Study 1) and the number of referrals they users during the consolidation phase and impact of made in certain specialities (Study 2). Chi-square ana- ECOPIH use on referrals lysis was used to analyse the relationship of dependence Settings, sampling and type of study between the variables. Study 2 was an open, multi-centre, controlled, rando- mised intervention study over a 24-month follow-up Ethics approval period. It was conducted on three PCSs in Barcelona This project adhered to Spanish legislation (Spanish Law Province (Badalona-Sant Adrià de Besòs PCS, SAP Santa 14/2007 of 3 July on Biomedical Research) and to inter- Coloma de Gramenet PCS and Maresme PCS), with 25 national regulations on ethical issues (Declaration of PC centres and 507 PC physicians, all belonging to the Helsinki and Declaration of Tokyo). The research protocol public health system of Catalonia, Spain. (P11/39) was reviewed and approved by the Ethics and The inclusion criteria were: PC clinicians who had Clinical Research Committee of the Primary Care Re- been working for at least 6 months at the same PC search Institute IDIAP Jordi Gol, Barcelona, Spain. All the centre for whom full patient visit and referral data were participants were informed in writing about their partici- available in the official electronic records of the institu- pation in the study and data confidentiality. In order to tion. Since only adult medicine specialities were ana- avoid bias, no information about the intervention was pro- lysed, PC paediatricians were excluded. vided. Written informed consent was obtained from all The population of 357 physicians at the Badalona-Sant participants. Confidentiality was maintained at all levels, Adrià de Besòs, Santa Coloma de Gramenet and Mar- thereby ensuring that professionals and patients could not esme PCSs was the reference point, from which a sample be identified. The patients’ medical records could not be comprising 178 physicians was taken. It should be noted accessed from ECOPIH. Information was obtained from that this random sample was different from the one used the survey responses and existing data related to visits and in Study 1, although the study universe was the same – referrals, and subject-identifying information was coded and finite – in both studies. and anonymised. The features of the intervention meant that it did not have to meet national regulations for clin- Variables of study 2 ical trials. Confidentiality was assured under the Spanish In order to establish the healthcare professionals’ profiles, Personal Data Protection Law (15/1999 of 13 December). hierarchical cluster analysis was performed, taking into ac- count the Gender, Age and ECOPIH member variables. Results The Background work for Study 2 was carried out in First objective: to find out the degree of loyalty among December 2012. Table 3 shows the variables analysed in ECOPIH users that study. Implementation phase: sample profile study 1 In order to respond to the first objective, the healthcare At the start of ECOPIH implementation, sample distri- professionals’ profiles were defined according to the bution by sex and age was fairly balanced: 56.9% women intention to use the CoCP in the implementation phase and 43.1% men. Regarding age, 32.4% were under 40 Table 3 Variables used in Study 2 Model variable Ecopih member The healthcare professional is a member of ECOPIH. Dichotomous variable. 0 = no and 1 = yes. Gender Gender of the healthcare professional. 1 = female and 0 = male. Age Age of the healthcare professional. The variable has four values: 1 = under 40 years old, 2 = between 40 and 49 years old, 3 = between 50 and 59 years old, and 4 = 60 years old or over. Referrals made Dependent variable. The number of patient referrals that the PC professional made in the last year. Categorical variable. 1 = low referral rate (fewer than 7 referrals), 2 = average referral rate (between 7 and 14 referrals), 3 = high referral rate (between 14 and 23 referrals) and 4 = very high referral rate (more than 23 referrals). Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 6 of 12 Table 4 Descriptive statistics of the sample (Study 1) Physicians (111) Profile 1A (64) Profile 1B (47) Gender Female 63 (56.9%) 42 (65.6%) 22 (46.8%) Male 48 (43.1%) 22 (34.4%) 25 (53.2%) Age < 40 years old 36 (32.4%) 36 (56.2%) 0 40–49 years old 28 (25.2%) 28 (43.8%) 0 50–59 years old 33 (29.7%) 0 33(70.2%) ≥ 60 years old 14 (12.6%) 0 14 (29.8%) Mobile device use Low 24 (22.0%) 11 (17.2%) 13 (27.7%) Medium 39 (34.9%) 21 (32.8%) 17 (36.2%) High 40 (35.8%) 27 (42.2%) 14 (29.4%) Very high 8 (7.3%) 5 (7.8%) 3 (6.4%) Social media website and app use Low 43 (38.3%) 26 (41.3%) 31 (65.2%) Medium 66 (59.8%) 36 (56.6%) 16 (34.8%) High 2 (1.9%) 2 (3.2%) 0 Intention to use ecopih Yes 59 (53.2%) 35 (54.7%) 24 (51.1%) No 52 (46.8%) 29 (45.3%) 23 (48.9%) years old, 25.2% were between 40 and 49 years old, implementation phase) coincided with those obtained in 29.7% were between 50 and 59 years old, and only 12.6% the second year of ECOPIH development (Study 2, con- were 60 years old or over. Finally, it should be noted that solidation phase). Thus, a professional profile was ob- the large majority’s Mobile device use was medium-high served in both the implementation and the consolidation (70.7%), whereas their Social media website and app use phases (profiles 1A and 2A) corresponding to healthcare was medium (59.8%) or low (38.3%) (Table 4). professionals who were young, mostly women, and ha- As shown in Table 5, the results obtained indicated bitual users of technology. In another profile (profiles 1B the existence of two different groups. The first profile and 2B in the respective stages), the healthcare profes- (1A) comprised 64 individuals, of whom 65.6% were sionals were older, mostly men, whose ICT use was under-50-year-old women, and the second profile (1B) lower. This explains why the first segment’s use of ECO- comprised 47 individuals, of whom 53.2% were over-50- PIH (profile 1A and 2A) was high (68.1%) and the sec- year-old men. The differences between the two groups ond group’s use of the tool (profile 1B and 2B) was very were significant for both the Age variable and the Gender low (14.2%). We can therefore confirm the existence of variable, with t-test values of 8.708 and 4.437, respect- an ECOPIH user profile – under-50-year-old women ively, at 99 and 95% confidence levels (Table 5). who habitually use ICTs – that was stable in both the When analysing each group’s relationship with tech- implementation phase and the consolidation phase, thus nology, the distribution with respect to Social media maintaining their loyalty to the tool. website and app use was found to be quite homogenous and similar in both, hence the Chi-square value was not significant in either of them. Table 5 Main cluster results at the start of ECOPIH implementation Consolidation phase: analysis of the ECOPIH users’ profiles. Profile 1A Profile 1B Significance (t-test) Sample profile study 2 Age < 40 years old 62 (56.2%) 0 8.708 As shown in Table 6, the results obtained indicated the 40–49 years old 49 (43.8%) 0 (0.004) existence of two different groups. The first profile (2A) 50–59 years old 0 78 (70.2%) comprised 72 professionals under 50 years old (100%), most of whom were women (76.4%) and ECOPIH mem- ≥ 60 years old 0 33 (29.8%) bers (68.1%), and the second group (Profile 2B) com- Gender Male 38 (34.4%) 59 (53.2%) 4.437 (0.032) prised 106 individuals over 50 years old (100%), most of Female 73 (65.6%) 52 (46.8%) whom were women (62.3%) and not ECOPIH members Final cluster centres (85.8%). Gender 1 2 The healthcare professionals’ profiles obtained from Age 1.44 3.30 the analysis of the sample in the first year (Study 1, Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 7 of 12 Table 6 Main cluster results (Study 2) Profile 2A (72) Profile 2B (106) Significance (t-test) Age < 40 years old 32 (44.4%) 0 43.250 (0.000) 40–49 years old 40 (55.6%) 0 50–59 years old 0 86 (81.1%) ≥ 60 years old 0 20 (18.9%) Gender Male 17 (23.6%) 40 (37.7%) Female 55 (76.4%) 66 (62.3%) ECOPIH member Yes 49 (68.1%) 15 (14.2%) No 23 (31.9%) 91 (85.8%) Final cluster centres Gender 2 2 Age 1.56 3.19 ECOPIH member 1 9 Second objective: to evaluate the degree of fulfilment of Gastroenterology, whereas older professionals who were users’ expectations not members of ECOPIH made a high or very high Implementation phase: determinants of ECOPIH use number of referrals. No significant differences were The model’s goodness of fit was confirmed by the values found with regard to the number of referrals made by and level of significance of the Chi-square statistic each group at the start of the study period, so the differ- (68.228, sig. 0.000) and the Hosmer-Lemeshow test ences found could be related to ECOPIH use. (10.224, p = 0.250). In addition, the value of Nagelkerke’s For the remaining specialities (Nephrology, Respiratory statistic indicated that the model obtained explained Medicine and Neurology), no differences were found in 62.1% of the dependent variable’s variance. the number of referrals. However, it should be noted From the analysis in Table 7, it is possible to observe that that the three specialities in which differences were the variables influencing frequency of use are, on the one found were those that involved higher total numbers of hand, the user’s profile in terms of his or her frequency of referrals and healthcare professionals. Social media website and app use (B = 1.933 p = 0.002) and, on the other, Perceived usefulness for reducing costs (time and financial costs) that ECOPIH use entails (B = 1.706 p = Discussion 0.025). No significant differences were found taking into ac- The analysis of the results of the two studies shown has count the professionals’ gender or age. enabled us to respond to the two stated objectives: (1) To find out the degree of loyalty among ECOPIH Consolidation phase: impact of ECOPIH on referrals users, and; Table 8 shows that the PC professionals’ behaviour was (2) To evaluate the degree of fulfilment of users’ significantly different for three of the specialities ana- expectations. lysed. Thus, it was found that, ECOPIH members in PC Both objectives are closely related since one of the professions made a low or average number of referrals main reasons why potential users become actual users is to SC professionals in Cardiology, Endocrinology, and the fulfilment of their expectations from the tool. Table 7 Equation variables (Study 1) B E.T. Wald DF Sig. Exp(B) Perceived usefulness for reducing costs (PU2) 1.706 0.761 5.026 1 0.025 5.508 Perceived usefulness for improving clinical practice quality (PU1) 0.793 0.636 1.553 1 0.213 2.211 Perceived ease of use of ECOPIH −0.075 0.310 0.058 1 0.810 0.928 Security and confidentiality 0.016 0.311 0.003 1 0.958 1.016 Social media website and app use 1.933 0.619 9.748 1 0.002 6.907 Mobile device use −0.011 0.339 0.001 1 0.973 0.989 Constant −3.327 1.751 3.612 1 0.057 0.036 Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 8 of 12 Table 8 Referrals in different specialities, by professional profile (Study 2) Referral rate by speciality Profile 2A (women, < 50 years old, ECOPIH Profile 2B (men, > 50 years old, non-ECOPIH SIGNIFICANCE (F users) users) value) Cardiology Low 11 (15.3%) 5 (4.7%) 0.023 Average 7 (9.7%) 6 (5.7%) High 18 (25.0%) 21 (19.8%) Very 36 (50.0%) 74 (69.8%) high Endocrinology Low 18 (25.0%) 10 (9.4%) 0.025 Average 12 (16.7%) 29 (27.4%) High 21 (29.2%) 29 (27.4%) Very 21 (29.2%) 38 (36.8%) high Gastroenterology Low 9 (12.5%) 5 (4.7%) 0.019 Average 9 (12.5%) 7 (6.6%) High 17 (23.6%) 16 (15.1%) Very 37 (51.4%) 78 (73.6%) high Nephrology Low 66 (91.7%) 94 (88.7%) 0.635 Average 6 (8.3%) 11 (10.4%) High 0 (0.0%) 1 (0.09%) Very – – high Neurology Low 12 (16.7%) 10 (9.4%) 0.455 Average 9 (12.5%) 13 (12.3%) High 19 (26.4%) 36 (34.0%) Very 32 (44.4%) 47 (44.3%) high Respiratory Low 11 (15.3%) 11 (10.4%) 0.443 Medicine Average 15 (20.8%) 17 (16.0%) High 17 (23.6%) 35 (33.0%) Very 29 (40.3%) 43 (40.6%) high Regarding the first objective, after comparing the pro- to a lesser extent, as identified in Study 2. This segment fessional profiles in the implementation phase to those bears considerable similarity to Profile 1B observed in in the consolidation phase, two professional profiles Study 1 (except for gender), in which the professionals were found to remain similar over time. In one profile, rated information security higher than cost reduction. the healthcare professionals were young, mostly women, In order to respond to the second objective, the deter- and habitual users of technology. In another, the health- mining factors that had an influence on the use of ECO- care professionals were older, mostly men, whose ICT PIH in the implementation year were analysed. The use was lower. impact of the tool’s use on the number of referrals in an The findings suggest that the degree of loyalty (from uncontrolled real-life setting was then evaluated, based on initial use to consolidated use) was high among the the medical professionals’ voluntary use thereof. This en- group of younger female professionals. This seems to be abled us to establish whether the expectations created at supported by the fact that those professionals in the seg- the start of ECOPIH use (specifically, that PC profes- ment comprising mostly under-50-year-old women who sionals felt that ECOPIH use would allow costs associated stated their intention to use ECOPIH in the implemen- with clinical practice to be reduced), had been fulfilled. tation phase were actual members of it two years later. Concerning the factors determining the adoption of The results obtained show that the professionals in the ECOPIH, our study revealed that two factors explained over-50-year-old profile were those who used ECOPIH physicians’ use of this tool. Firstly, professionals’ ICT Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 9 of 12 user profiles influenced intention to use ECOPIH (B = and as seen in other studies [30], the degree of tool use 1.933 p = 0.002). All professionals, regardless of their age has an influence on the tool’s potential usefulness; this is and whether or not they were ECOPIH users, habitually linked to Perceived usefulness for reducing costs (time used mobile devices and extensively used social media and financial costs) that ECOPIH use entails. The seg- websites and apps. It is logical to think that those who ment of physicians who predominantly used ECOPIH habitually used social media/online platforms would be regularly had lower referral rates in those specialities for more likely to use a CoP in a clinical setting because which it had been used the most. This particularly rein- they are already more comfortable with online platforms. forces the idea that, for this group of professionals, the Furthermore, the majority of under-50-year-old profes- tool fulfilled their expectations in terms of its ability to sionals would have used ICTs intensively at various reduce costs associated with clinical practice. These stages throughout their higher education and profes- findings are consistent with the results from other stud- sional development. In contrast, many professionals ies on the use of telemedicine applied to consultations among the over-50-year-old generation could be classi- among professionals [32, 52]. These results indicate that fied as late adopters of ICTs, mainly because such adop- the tool has great potential because we are on the point tion occurred in the workplace. Consequently, some of a generational changeover. Given today’s user profile, professionals were reluctant to use ICTs because they use of ECOPIH, and therefore its usefulness, are ex- saw it as an obligation and considered them hard to use pected to increase in the near future. and not particularly useful. It should be noted that these results were only ob- In the explanation of the physicians’ ECOPIH use, sec- tained in three of the six specialities evaluated (Cardi- ond in order of importance was Perceived usefulness for re- ology, Endocrinology and Gastroenterology, and not ducing costs (time and financial costs). Physicians decided Respiratory Medicine, Nephrology or Neurology). This is to use the CoCP because they considered that it could be- probably due to two reasons. First, the latter three speci- come an effective tool for reducing various costs (B = alities were incorporated into ECOPIH later, and that 1.706 p = 0.025). Considering time strain, it is reasonable might have hindered its use. Second, they are specialities to think that healthcare professionals would opt for the in which fewer referrals are made, probably because the development of more efficient professional activities. most common disorders within them are more protoco- Thus, it is understood that the intention to use ECOPIH lised and less individualised, thereby facilitating an inde- is conditional upon it being perceived as a tool that en- pendent handling of them by PC professionals. ables a correct diagnosis to be made while minimising the Finally, the results obtained show that the profes- amount of time, effort and financial cost involved for both sionals in the over-50-year-old profile are those who the physicians and the healthcare institutions [9]. How- used ECOPIH to a lesser extent and, in turn, made the ever, this tendency appears to be more apparent among most referrals. This segment, as identified in Study 2, younger, especially female professionals than among older bears considerable similarity to Profile 1B observed in male professionals. The setting’s cultural history might ex- Study 1 (except for gender), who used ICTs to a lesser plain the gender differences between the two groups iden- extent and rated information security higher than cost tified. Thus, the group of older professionals is mostly reduction. Hence, it can be seen that the number of re- male because, at the start of the second half of the twenti- ferrals made remained unchanged. eth century in Spain, women’s access to certain types of higher education – such as medicine or engineering – was quite limited. The medical profession underwent a gradual Limitations feminisation, meaning that those generations of healthcare This study has a number of limitations. First, the diffi- professionals trained at the end of the last century in- culty of recording the impact of such tools should not cluded a high percentage of women. In the early twenty- be overlooked. The impact of CoCP tools should also first century, 70% of new medical students were women; take into account the quality of referrals, the physician’s this has since risen to 85% [48, 49]. Moreover, the younger trust, and interprofessional communication. Further- generation of professionals is very aware of patients’ ser- more, use and effectiveness of ECOPIH may have been vice experience because of the more active role that pa- influenced by other factors not considered in our study. tients play in healthcare provision models (empowerment Examples of such factors include: (a) availability of and decision-making) [50] [51]. other PC-SC consultation systems; (b) each centre’s An important aspect that affects the professionals’ de- healthcare workload, and; (c) being or not being a cision to use the tool continuously over time is its ability teaching centre for resident physicians. The results ob- to fulfil the expectations created in the implementation tained from this study need to be complemented by a phase. Regarding the impact of the tool’s use on the qualitative evaluation in order to assess the tool com- number of referrals in an uncontrolled real-life setting, prehensively [37, 53–55]. Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 10 of 12 Second, we are aware that both the number of ECO- tools offer advantages for clinical efficiency. The poten- PIH users (65 members) and the referral rate (number tial of the tool increases as more and more young pro- of referrals per professional) limit the study’s statistical fessionals use it. We also believe that its use should be power. In order to solve both of these problems, we strengthened because of the advantages it offers in terms could have conducted a randomised controlled study on of efficiency, learning and spreading knowledge. one specific group of professionals, that is to say, those who were enthusiastic about and committed to using Supplementary information ECOPIH. If we had done so, results of greater magni- Supplementary information accompanies this paper at https://doi.org/10. 1186/s12875-020-1081-x. tude might have been obtained, though it would have af- fected the external validity of our study, which was Additional file 1. Questionnaire to Primary Care professionals. conducted in a real-life setting of clinical practice and Additional file 2. Questionnaire to Specialist Care professionals. gave the professionals the freedom to use the tool as they wished, which we believe is one of the greatest Abbreviations strengths of our study. CoCP: Community of clinical practice; CoP: Community of practice; Finally, during the study design phase, the inclusion of ECOPIH: Eina de Comunicació entre Primària i Hospitalària (Online clinical variables as a measure of the effect of ECOPIH Communication Tool between Primary and Hospital Care); ICT: Information and communication technologies; PC: Primary care; PCS: Primary Care use was considered. That option was ultimately rejected Service; SC: Specialist care; UOC: Open University of Catalonia owing to the difficulty of isolating the effect of the tool’s use from other influencing factors (e.g., courses taken by Acknowledgements This project received a grant from the 7th 2011 Award for Primary Health the professionals) and of finding a single clinical variable Research of the Barcelona Health Region, Government of Catalonia. to encapsulate the improved clinical control of patients, This study was also supported through funding from the UAB–Novartis Chair since ECOPIH is a platform on which any type of clin- at the Autonomous University of Barcelona. It did not play any role in the study design. ical case can be raised for consultation. The English version of this work was language-edited by Steven Norris thanks to the support of the eLearn Center at the Open University of Future implications Catalonia. We believe that further research should be done on the Authors’ contributions impact of CoCPs for professionals working in different All the authors have made substantive intellectual contributions to a areas of healthcare to communicate with one another, published study. DLT (ECOPIH Community Manager) contributed to research from the perspective of both the financial implications (a question formulation, literature search, study design, analysis and interpretation of results, and manuscript drafting. JMM participated in study reduction in referrals and visits, and cost analysis) and conception and design, acquisition of data, and analysis and interpretation the clinical outcomes. We propose that longer-term of data, and helped with manuscript drafting. TRB performed statistical follow-ups should be done and that the use of the tool analysis and interpreted the findings. AJZ contributed to research question formulation and study design, performed statistical analysis, interpreted the should be more actively promoted and encouraged, findings and participated in manuscript drafting. SFB (ECOPIH Community while ensuring that its use is never made compulsory Manager) participated in data collection, interpretation and dissemination of [56]. In order to do that, it will be necessary to ensure results, and helped with manuscript drafting. PTM was involved in critically reviewing the manuscript for important intellectual content, giving final that the firm managing the tool guarantees its continu- approval of the version to be published, and the collection and analysis of ity, that users are given time to actually use it, and that results. XMT participated in protocol design, interpretation of results and solutions to any technological aspects representing bar- article drafting. FSR contributed to research question formulation, study design coordination, literature search, study design, interpretation of results riers to its use are found. As a future strategy, and in and article drafting. He is the guarantor of the article. All the authors have keeping with the recommendations of some authors, it read, revised and approved the final manuscript. might be appropriate to push ahead with the tool’s dis- Funding semination, presenting it in a way that facilitates its use. Not applicable. This would strengthen the available evidence and the relative advantages of using ECOPIH, which would sig- Availability of data and materials All data generated or analysed during this study are included in this nificantly help to increase its use [57]. published article. Finally, it would be interesting to expand the research by looking into the impact of a CoP in a PC clinical setting Ethics approval and consent to participate as a novel tool for training based on real clinical cases. The research protocol (P11/39) was reviewed and approved by the Ethics and Clinical Research Committee of the Primary Care Research Institute IDIAP Jordi Gol, Barcelona, Spain. Conclusions ECOPIH and other CoCPs can be used to raise clinical Consent for publication Not applicable. cases for consultation and share information among PC and SC professionals; such tools may reduce the number Competing interests of referrals to SC. Furthermore, ECOPIH and similar The authors declare that they have no competing interests. Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 11 of 12 Author details 15. Montero Ruiz E, López-Álvarez J. La interconsulta médica: problemas y 1 Centre d’Atenció Primària Gran Sol, Gerència d’Àmbit d’Atenció Primària soluciones. Med Clin (Barc). 2011;136(11):488–90 Available from: http:// Metropolitana Nord, Institut Català de la Salut. Avinguda del Doctor Bassols, linkinghub.elsevier.com/retrieve/pii/S0025775309011099. 112 - 130, 08914 Badalona, Spain. 2Unitat de Suport a la Recerca 16. Horner K, Wagner E, Tufano J. Electronic consultations between primary and Metropolitana Nord, IDIAP Jordi Gol, CAP El Maresme. Camí del Mig, 36 specialty care clinicians: early insights. Issue Brief (Commonw Fund). 2011; planta 4ª, 08303 Mataró, Spain. 3Universitat Autònoma de Barcelona, 23:1–14 Available from: http://mobile.commonwealthfund.org/~/media/ Bellaterra, Cerdanyola del Vallès. Campus de la UAB, Plaça Cívica, s/n, 08193 Files/Publications/Issue Brief/2011/Oct/1554_Horner_econsultations_ Bellaterra, Barcelona, Spain. 4Faculty of Economics and Business, Universitat primary_specialty_care_clinicians_ib.pdf. Oberta de Catalunya, Barcelona, Spain. 5Institut Universitari d’Investigació en 17. Forrest CB, Nutting PA, von Schrader S, Rohde C, Starfield B. Primary care Atenció Primària (IDIAP Jordi Gol), Gran Via Corts Catalanes, 587, àtic, 08007 physician specialty referral decision making: patient, physician, and health Barcelona, Spain. 6Departament de Ciències Mèdiques, Universitat de Girona, care system determinants. Med Decis Mak. 2006;26(1):76–85 Available from: C/ Emili Grahit, 77, 2n, 17003 Girona, Spain. 7Unitat de Suport a la Recerca http://www.ncbi.nlm.nih.gov/pubmed/16495203. Barcelona Ciutat, IDIAP Jordi Gol, Carrer Sardenya 375, 08025 Barcelona, 18. Blank L, Baxter S, Woods HB, Goyder E, Lee A, Payne N, et al. What is the Spain. 8Faculty of Health Sciences, Universitat Oberta de Catalunya, evidence on interventions to manage referral from primary to specialist Barcelona. Av. Tibidabo, 39-43, 08035 Barcelona, Spain. non-emergency care? A systematic review and logic model synthesis. Heal Serv Deliv Res. 2015;3(24):1–430. Received: 27 December 2018 Accepted: 6 January 2020 19. Brez S, Rowan M, Malcolm J, Izzi S, Maranger J, Liddy C, et al. Transition from specialist to primary diabetes care: a qualitative study of perspectives of primary care physicians. BMC Fam Pract. 2009;10:39 Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2704171&tool= References pmcentrez&rendertype=abstract. 1. Murray M, Berwick DM. Advanced access: reducing waiting and delays in 20. Harrison R, Clayton W, Wallace P. Can telemedicine be used to improve primary care. JAMA. 2003;289(8):1035–40 Available from: http://www.ncbi. communication between primary and secondary care? BMJ. 1996;313(7069): nlm.nih.gov/pubmed/12597760. 1377–80-1 Available from: http://www.pubmedcentral.nih.gov/articlerender. 2. Musgrove H, Creese A, Preker A, Baeza C, Anell A, Prentice T. The World fcgi?artid=2352902&tool=pmcentrez&rendertype=abstract. health report 2000 : health systems: improving performance, vol. 78. 21. Carey M, Noble N, Mansfield E, Waller A, Henskens F, Sanson-Fisher R. The Geneva: The World health report 2000; 2000. Available from: http://www. Role of eHealth in Optimizing Preventive Care in the Primary Care Setting. J who.int/whr/2000/en/whr00_en.pdf Med Internet Res. 2015;17(5):e126 Available from: http://www.jmir.org/2 3. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems 015/5/e126/. and health. Milbank Q. 2005;83(3):457–502 Available from: http://www.ncbi. 22. Saigi-Rubió F, Jiménez-Zarco A, Torrent-Sellens J. Determinants of the nlm.nih.gov/pubmed/16202000. intention to use telemedicine: Evidence from Primary Care physicians. Int J 4. Beasley JW, Hankey TH, Erickson R, Stange KC, Mundt M, Elliott M, et al. Technol Assess Health Care. 2016;32(1–2):29–36. How many problems do family physicians manage at each encounter? A WReN study. Ann Fam Med [Internet]. 2004;2(5):405–10 Available from: 23. Díaz-Chao Á, Torrent-Sellens J, Lacasta-Tintorer D, Saigí-Rubió F. Improving http://www.annfammed.org/content/2/5/405.short. Integrated Care: Modelling the performance of an online community of 5. Davidoff F, Miglus J. Delivering clinical evidence where it’s needed: building practice. Int J Integr Care. 2014;14(1) Available from: http://www.ijic.org/ an information system worthy of the profession. JAMA. 2011;305(18):1906–7 article/10.5334/ijic.1200/. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21558524. 24. Bashshur RL, Shannon G, Krupinski EA, Grigsby J. Sustaining and realizing the 6. Smith R. Strategies for coping with information overload. BMJ. 2010;341: promise of telemedicine. Telemed J E Health [Internet]. 2013;19(5):339–45 c7126 Available from: http://www.ncbi.nlm.nih.gov/pubmed/21159764. Available from: http://online.liebertpub.com/doi/abs/10.1089/tmj.2012.0282. 7. Grant RW, Ashburner JM, Hong CS, Hong CC, Chang Y, Barry MJ, et al. 25. Waldura JF, Neff S, Dehlendorf C, Goldschmidt RH. Teleconsultation Defining patient complexity from the primary care physician’s perspective: a improves primary care clinicians’ confidence about caring for HIV. J Gen cohort study. Ann Intern Med. 2011;155(12):797–804 Available from: http:// Intern Med. 2013;28(6):793–800. www.ncbi.nlm.nih.gov/pubmed/22184686. 26. Liddy C, Rowan MS, Afkham A, Maranger J, Keely E. Building access to 8. Del Fiol G, Workman TE, Gorman PN. Clinical questions raised by clinicians at specialist care through e-consultation. Open Med. 2013 Jan;7(1):e1–8. the point of care: a systematic review. JAMA Intern Med. 2014;174(5):710–8 27. Vimalananda VG, Gupte G, Seraj SM, Orlander J, Berlowitz D, Fincke BG, et al. Available from: http://archinte.jamanetwork.com/article.aspx?articleid=1846630. Electronic consultations (e-consults) to improve access to specialty care: a 9. Lacasta Tintorer D, Manresa Domínguez JM, Pujol-Rivera E, Flayeh Beneyto systematic review and narrative synthesis. J Telemed Telecare. 2015;21(6):323–30. S, Mundet Tuduri X, Saigí-Rubió F. Keys to success of a community of 28. Angstman KB, Adamson SC, Furst JW, Houston MS, Rohrer JE. Provider clinical practice in primary care: a qualitative evaluation of the ECOPIH satisfaction with virtual specialist consultations in a family medicine project. BMC Fam Pract. 2018;19(1):56 Available from: https://bmcfampract. department. Health Care Manag (Frederick). 2009;28(1):14–8 Available from: biomedcentral.com/articles/10.1186/s12875-018-0739-0. http://www.ncbi.nlm.nih.gov/pubmed/19225331. 10. Alonso Pérez de Ágreda JP, Febrel Bordejé M, Huelin Domeco de Jarauta J. 29. Stoves J, Connolly J, Cheung CK, Grange A, Rhodes P, O’Donoghue D, et al. Factores Asociados a la Derivación Inadecuada Entre Atención Primaria y Electronic consultation as an alternative to hospital referral for patients with Especializada: Estudio Cualitativo en Médicos de Atención Primaria. Gac Sanit. chronic kidney disease: a novel application for networked electronic health 2000;14(2):122–30 Available from: http://apps.elsevier.es/watermark/ctl_ records to improve the accessibility and efficiency of healthcare. Qual Saf servlet?_f=10&pident_articulo=13008602&pident_usuario=0&pcontactid= Health Care. 2010;19(5):e54 Available from: http://www.ncbi.nlm.nih.gov/ &pident_revista=138&ty=119&accion=L&origen=elsevier&web=www.elsevier. pubmed/20554576. es&lan=es&fichero=138v14n2a13008602pdf001.pdf. 30. Zanaboni P, Scalvini S, Bernocchi P, Borghi G, Tridico C, Masella C. 11. Terraza Núñez R, Vargas Lorenzo I, Vázquez Navarrete ML. Coordination among Teleconsultation service to improve healthcare in rural areas: acceptance, healthcare levels: systematization of tools and measures. Gac Sanit. 2006;20(6): organizational impact and appropriateness. BMC Health Serv Res. 2009;9:238. 485–95 Available from: http://scielo.isciii.es/pdf/gs/v20n6/revision.pdf. 31. Kim-Hwang JE, Chen AH, Bell DS, Guzman D, Yee HF, Kushel MB. Evaluating 12. O’Malley AS, Reschovsky JD. Referral and consultation communication electronic referrals for specialty care at a public hospital. J Gen Intern Med. between primary care and specialist physicians: finding common ground. 2010;25(10):1123–8. Available from: http://www.pubmedcentral.nih.gov/ Arch Intern Med. 2011;171(1):56–65 Available from: http://archinte.ama-assn. articlerender.fcgi?artid=2955477&tool=pmcentrez&rendertype=abstract. org/cgi/reprint/171/1/56. 32. Straus SG, Chen AH, Yee H, Kushel MB, Bell DS. Implementation of an 13. Bodenheimer T. Coordinating care--a perilous journey through the health electronic referral system for outpatient specialty care. AMIA Annu Symp care system. N Engl J Med. 2008;358(10):1064–71 Available from: http:// Proc. 2011;2011:1337–46 Available from: http://www.pubmedcentral.nih. amcp.org/WorkArea/DownloadAsset.aspx?id=12358. gov/articlerender.fcgi?artid=3243286&tool=pmcentrez&rendertype=abstract. 14. Louro González A, Fernández Obanza E, Fernández López E, Vázquez Millán 33. Wenger E. Communities of practice: learning, meaning, and identity: P, Villegas González L, Casariego VE. Análisis de las dudas de los médicos de Cambridge University Press; 1999. 318 p. Available from: https://books. atención primaria. Aten Primaria. 2009;41(11):592–7. google.com/books?id=heBZpgYUKdAC&pgis=1. Lacasta Tintorer et al. BMC Family Practice (2020) 21:15 Page 12 of 12 34. Gabbay J, le May A. Evidence based guidelines or collectively constructed qualitative evaluation. J Am Med Inform Assoc. 2003;10(6):596–604 Available “mindlines?” Ethnographic study of knowledge management in primary from: http://www.researchgate.net/publication/10607429_Understanding_ care. BMJ. 2004;329(7473):1013 Available from: http://www.ncbi.nlm.nih.gov/ the_Normalization_of_Telemedicine_Services_through_Qualitative_ pmc/articles/PMC524553/. Evaluation_Table_1. 35. Li LC, Grimshaw JM, Nielsen C, Judd M, Coyte PC, Graham ID. Use of 54. Greenhalgh T, Russell J. Why do evaluations of eHealth programs fail? An communities of practice in business and health care sectors: a systematic alternative set of guiding principles. PLoS Med. 2010;7(11):e1000360 review. Implement Sci. 2009;4(1):27 Available from: http://www.ncbi.nlm.nih. Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid= gov/pubmed/19445723. 2970573&tool=pmcentrez&rendertype=abstract. 36. McGowan BS, Wasko M, Vartabedian BS, Miller RS, Freiherr DD, 55. Lacasta Tintorer D, ; Manresa Domínguez JM; Pujol-Rivera E; Flayeh Beneyto Abdolrasulnia M. Understanding the factors that influence the adoption and S; Mundet Tuduri X; Saigí-Rubió F. Keys to success of a Community of meaningful use of social media by physicians to share medical information. Clinical Practice in primary care: A qualitative evaluation of the ECOPIH J Med Internet Res. 2012;14(5):e117 Available from: http://www. project. BMC Fam Pract. 2018;19(1):56. pubmedcentral.nih.gov/articlerender.fcgi?artid=3510763&tool= 56. Wenger E, McDermott RA, Snyder W. Cultivating communities of practice: a pmcentrez&rendertype=abstract. guide to managing knowledge: Harvard Business School Press; 2002. 37. Wenger E, Snyder W. Communities of Practice: The Organizational Frontier. Available from: http://dl.acm.org/citation.cfm?id=560383. Harv Bus Rev. 2000:139–45 Available from: https://hbr.org/2000/01/ 57. Masso M, Thompson C. Attributes of innovations and approaches to communities-of-practice-the-organizational-frontier. Accessed 3 June 2018. scalability - lessons from a national program to extend the scope of 38. Norman GR, Schmidt HG. Effectiveness of problem-based learning curricula: practice of health professionals. J Multidiscip Healthc. 2016;9:401–10 theory, practice and paper darts. Med Educ. 2000;34(9):721–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27616889. 39. OSMA Legal Services Group. Social networking and the medical practice. J Okla State Med Assoc. 2010;103(10):517–26 Available from: http://www.ncbi. nlm.nih.gov/pubmed/21189831. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in 40. Maloney EJ. What Web 2.0 Can Teach Us About Learning. The Chronicle of published maps and institutional affiliations. Higher Education, vol. B26; 2007. Available from: http://www.chronicle.com/ article/What-Web-20-Can-Teach-Us/8332 41. Cain J, Policastri A. Using Facebook as an informal learning environment. Am J Pharm Educ. 2011;75(10):207 Available from: http://www. pubmedcentral.nih.gov/articlerender.fcgi?artid=3279026&tool= pmcentrez&rendertype=abstract. 42. Yamout SZ, Glick ZA, Lind DS, Monson RAZ, Glick PL. Using social media to enhance surgeon and patient education and communication. Bull Am Coll Surg. 2011;96(7):7–15 Available from: http://www.ncbi.nlm.nih.gov/ pubmed/22315896. 43. Paton C, Bamidis PD, Eysenbach G, Hansen M, Cabrer M. Experience in the use of social media in medical and health education. Contribution of the IMIA Social Media Working Group. Yearb Med Inform. 2011;6:21–9 Available from: http://www.ncbi.nlm.nih.gov/pubmed/21938320. 44. Lacasta Tintorer D, Flayeh Beneyto S, Alzaga Reig X, Mundet Tuduri X, De la Fuente JA, Manresa JM, et al. Impact of the implementation of an online network support tool among clinicians of primary health care and specialists: ECOPIH Project. BMC Fam Pract. 2013;14(1):146 Available from: http://www.biomedcentral.com/1471-2296/14/146. 45. Lacasta Tintorer D, Flayeh Beneyto S, Manresa JM, Torán-Monserrat P, Jiménez-Zarco A, Torrent-Sellens J, et al. Understanding the discriminant factors that influence the adoption and use of clinical communities of practice: The ECOPIH case. BMC Health Serv Res. 2015;15(1):373. 46. Fabrigar L, Wegener D, MacCallum R. Evaluating the use of exploratory factor analysis in psychological research. Psychological. 1999; Available from: http://psycnet.apa.org/journals/met/4/3/272/. 47. Nunnally JC. Psychometric theory. University of Michigan, US: McGraw-Hill; 1978. p. 701. 48. Arrizabalaga P, Bruguera M. The feminization and the profession of Medicine. Med Clin (Barc). 2009;133(5):184–6 Available from: http://www. ncbi.nlm.nih.gov/pubmed/19426998 PMID: 19426998. 49. Paik JE. The feminization of medicine. JAMA. 2000;283(5):666 Retrieved August 8, 2019. Available from: http://jama.jamanetwork.com/article. aspx?doi=10.1001/jama.283.5.666-JMS0202-2-1). 50. Torrent-Sellens J, Díaz-Chao Á, Soler-Ramos I, Saigí-Rubió F. Modeling and predicting outcomes of eHealth usage by European physicians: multidimensional approach from a survey of 9196 general practitioners. J Med Internet Res. 2018;20(10):e279. https://doi.org/10.2196/jmir.9253. 51. Lee Y, Lin J. Do patient autonomy preferences matter? Linking patient- centered care to patient-physician relationships and health outcomes. Soc Sci Med. 2010;71(10):1811–8. 52. Vimalananda VG, Gupte G, Seraj SM, Orlander J, Berlowitz D, Fincke BG, et al. Electronic consultations (e-consults) to improve access to specialty care: a systematic review and narrative synthesis. J Telemed Telecare [Internet]. 2015;21(6):323–30 Available from: http://www.pubmedcentral.nih.gov/ articlerender.fcgi?artid=4561452&tool=pmcentrez&rendertype=abstract. 53. May C, Harrison R, Finch T, MacFarlane A, Mair F, Wallace P, et al. Understanding the normalization of telemedicine services through