This section reports on the cross-case analysis, with the underlying, within-case descriptions available in Additional File 3. Table 2 summarizes how the six identified collaborative affordances of the EHR facilitated and/or constrained collaboration within and between disciplines and medical specialties.
Insert Table 2 here
Affordance 1: Portability
A widely shared view among the interviewed representatives of the clinics was that information from each specialty was integrated in the EHR. The former (legacy) system had already provided this functionality but the EHR ensured that notes by Medical Specialists were now also included. Several functional groups voiced the importance of collecting data from all specialties since this resulted in a comprehensive overview of the available information. Based on this shared overview, the medical specialists were better able to develop mutual awareness. This was most strongly expressed in the most intensively collaborating outpatient clinics (A, C, and D).
In three clinics (A, C, and E), the Medical Specialists argued that photographs imported into the EHR were sometimes still only accessible by certain clinical specialties. As a result, medical specialists were not able to discuss these images during meetings. This was experienced as a negative influence on collaboration. Moreover, it was argued that, in all cases, the health records of hospitalized patients were still tied to specific medical domains. As a consequence, medical specialists could only access these health records if they logged onto specific domains. However, no one commented that this negatively affected collaboration.
Interviewees from three clinics (B, D, and E) commented that data could not be shared with neighboring-hospitals through the EHR. Since these clinics receive many patients from other hospitals, this interfered with communication with medical specialists from the neighboring hospitals which hindered mutual awareness. Further, outpatient clinic B works intensively with external parties such as research institutions and its Business Manager indicated the difficulties in sharing relevant EHR data with these parties: “I find it a big disadvantage that we cannot easily get reports from the EHR. We can’t do that ourselves: we are constantly dependent on others! [...]. But, we as [specific specialty] have to share lots of data with external agencies and we have struggled with that for a long time.” - [B-BM1]
Affordance 2: Co-located access
In all the clinics, respondents mentioned that the EHR enabled simultaneous access to health records, but also complained that the EHR prevented users simultaneously modifying health records. Several medical specialists from all the outpatient clinics mentioned that they could not place orders when a colleague was working on the same health record.
It was also striking that the different functional groups who collaborated in the same office or clinic were frequently hindered. For example, three medical specialists and medical residents of outpatient clinic E mentioned that they, temporarily, could not complete their work during joint consultations with nursing specialists. In such situations, only one professional could have access to the health record. As a consequence, another professional was denied access and was therefore unable to process orders or relevant data in the EHR. Whether this should be an issue was questioned by one medical specialist since he was convinced that EHR users would often be working on different parts of the EHR database and could therefore not imagine that co-located access needed to be obstructed. Moreover, he argued that it would only make sense to impede co-located access when professionals were trying to work on the same part of the EHR database.
A very large number of notes were created in the EHR, in part because these could only be changed by their owners. Concerns were expressed in outpatient clinics A, C, D, and E about the quality of the shared overview since this was complicated by dozens of notes by various specialists. On a different but related issue, medical specialists were hindered in collaborating during multidisciplinary meetings when they had access to only one desktop because they then had to switch between medical results and the notes of the meeting which made it difficult to remember which patient was being discussed. Some had already seen that this could resolved by using a second desktop.
Affordance 3: Shared overview
As mentioned above, the large number of notes negatively affected the quality of the shared overview. As a result, medical specialists of outpatient clinics A, C, D, and E commented that they were hindered in gaining a mutual awareness of other specialists’ notes. Since data were ordered on priority and not on the chronology of events, all the interviewed medical specialists felt impeded in easily understanding what had occurred in the medical timeline of their patients. Interviewees from outpatient clinics B, D, and E commented that handwritten notes were something from the past, because notes were now entered in the EHR. Therefore, they argued, medical specialists should “finally” be able to understand the notes of their colleagues.
In all the outpatient clinics, medical specialists argued that the medical history and problem lists of patients were not useful in gaining a mutual awareness of the issues with other specialties. Two reasons emerged from the interviews. First, within specialties, there are different views on which information was important for providing high quality care. Therefore, within specialties, data were entered in different ways, resulting in specialty-specific information that was less useful for other specialties. Second, it was even argued that, within some specialties, no use was made of the medical history and problem lists, leading to friction between specialties when patients were referred with an empty health record when it came to certain specialties. Interestingly, some specialties involved in outpatient clinics A and C had developed a uniform policy for the use of the medical history and problem lists. As a result, all the medical specialists of these specialties entered the required data.
The importance of integrated information resources in providing high quality care was expressed by many medical specialists. For example, for some outpatient clinics (D and E), the medical history and problem lists were seen as highly important since these clinics often treat patients with an extensive medical history. However, as already mentioned, there were concerns about the quality of the data. Indeed, some interviewees of outpatient clinics D and E said that they did not make use of the medical history in their medical consultations and surgeries because they simply did not trust the data stored in the system. Specifically, some medical specialists and medical residents commented that important information was occasionally missing from the medical history and problem lists. Consequently, these interviewees explicitly read letters (as contained in the EHR) to develop a mutual awareness with other medical specialists.
Data included in the medical history and problem lists were tied to certain codes of the EHR’s vocabulary. As a result, symptoms that were not in its vocabulary, or symptoms that were misspelt could not be added. In this respect, the medical specialists of outpatient clinic E highlighted that the EHR did not include an adequate search functionality, impeding them in connecting the correct diagnosis with an appropriate code. This was considered to decrease the quality of the shared overview.
In all the clinics, it was argued that the provision of a shared overview is a requirement for collaboration between specialties as it increases their mutual awareness. However, this would only be effective if all the hospital’s specialties used the EHR consistently, which was not the case.
Affordance 4: Mutual awareness
In each of the clinics, interviewees argued that medical specialists were impeded in developing a mutual awareness between specialties because information was not clearly represented in the EHR. Two causes were offered: (1) the shared overview was not clear because each specialty entered the data differently, which negatively influenced the mutual awareness between medical specialists of different specialties, and (2) the data in the EHR were sorted on priority what impeded specialists in seeing what had happened in the medical timelines of their patients. As discussed earlier, the mutual awareness of some medical specialists from outpatient clinics A, C, and E was decreased because images were still tied to certain departments.
On the other hand, the EHR could support medical specialists in improving their mutual awareness of patients’ medical timelines since this process was now more transparent. Moreover, the mutual awareness between medical specialists was increased due to portable notes. With all the specialties of the hospital integrated in the EHR, interviewees in four outpatient clinics (B, C, D, and E) commented that patients could be referred more easily between different specialties by means of the orchestrating affordance. The use of the messaging affordance was also seen as an important component in supporting mutual awareness between medical specialists.
No single functionality of the EHR could be directly linked to the mutual awareness of healthcare professionals. However, all the other collaborative affordances had an influence on actors’ mutual awareness. Therefore, the mutual awareness between different healthcare professionals was seen by many interviewees as a highly important factor in collaboration. However, the mutual awareness between medical specialists depended on uniform use of the EHR.
Affordance 5: Messaging
The advantages of the messaging affordance were experienced differently in each outpatient clinic. However, interviewees in all the clinics appreciated the benefit of having the possibility to attach health records to messages. Previously, patient-related matters were discussed through Outlook. This frequently led to misunderstandings between medical specialists because health records could not be attached to an email. In three outpatient clinics (A, C, and D), it was mentioned that some specialties had developed a policy that required the use the messaging option. Through this, various functional groups within these specialties could be assured that the “receiver” had actually read their message. Accordingly, the healthcare professionals in these specialties were better equipped to gain a mutual awareness.
Conversely, in clinics B and E, there was no observed shift to adopt messaging. Specialties in these outpatient clinics did not adopt a uniform policy for the use of the messaging affordance. In these cases, the collaborative advantages depended on the medical specialists’ individual decisions to use the afforded messaging. Certainly, some of the medical specialists in outpatient clinic E did not use messaging. Interviewees from both clinics B and E commented that some medical administrators did use messaging, but that some medical specialists did not. As a result, the medical administrators’ messages were not answered.
As such, the delivery of the collaborative advantages offered by the messaging affordance depended on its uniform use in and between clinics. It was widely argued that the messaging opportunity currently led to an information overload, mostly due to it being used for uninformative medical results or letters. A decrease in the use of the messaging option had been noted in outpatient clinics B and E. Side effects of the messaging system were also expressed. For example, verbal communication between collaborating disciplines was lost in most cases, and interviewees saw this as negatively impacting on their collaboration.
Affordance 6: Orchestrating
All the business managers and medical administrators argued that the use of the orchestrating affordance was more efficient than the use of paper notes because orders were processed immediately and sent to the correct actor. On the other hand, most medical specialists argued that the use of the orchestrating affordance led to time consuming digitalized healthcare processes and marginalized verbal communication with colleagues, patients and the medical administration. As reported by previous studies [17-18], the time spent to use the EHR was experienced to decrease the time spent on verbal communication with colleagues and patients. Here, this was seen to negatively affect the collaboration between different functional groups and the quality of healthcare. This was expressed by a Medical specialist as follows: “The disadvantage is that the EHR takes away the interaction between people and I think that is actually a drawback: interactions between people are often more useful than the parametric recording of data. We have to make sure that we don't diminish this interaction too much." - [D-MS3]. Hence, making use of the orchestrating affordance was seen as boosting efficiency from a managerial perspective but was perceived as less desirable from a patient care standpoint.
In four outpatient clinics (B, C, D, and E), it was voiced that different disciplines were not able to work in a natural way because all processes were now based on digitalized orders. Examples were provided of some disciplines not collaborating without an order as this was the hospital’s policy. For example, daily face-to-face collaborative processes between nurses and medical specialists were hindered by this in outpatient clinics B, D, and E.
Most interviewees argued that the orchestrating affordance did not easily support them in arranging multidisciplinary consultations with other specialties. Although the nature of these consultations was not affected, the orchestrating opportunity did not properly guide financial information flows within the hospital. As a result, some outpatient clinics did not receive financial compensation for organizing these meetings. As a consequence, one of outpatient clinic A’s specialties was in debt for organizing these consultations. Moreover, medical specialists from outpatient clinics D and E mentioned that it was too difficult to arrange multidisciplinary consultations through the EHR. Consequently, these interviewees did not make use of this functionality.
Some medical specialists raised the difficulty of inviting other medical specialists to consultations. One medical specialist argued that this could be seen more as a future potential of the EHR to further support the collaboration between different specialties. However, several business managers and medical specialists expressed the view that multidisciplinary meetings were better registered in the EHR and that the results of these meetings could more easily be found than before.
Influence of organizational choices and policies
EHR users interviewed from all the clinics agreed that the strict role authorization and different system representations in the EHR hindered interdisciplinary collaboration. The hospital had only authorized medical specialists to enter patient-related data in the EHR. These strict role authorizations limited interdisciplinary collaboration as a business manager explained: "The Board of our organization decided that only medical specialists would be authorized to enter patient-related data. Therefore, the medical administrators are no longer authorized although it is, of course, a team that is collaborating." [C-BM1]. Before the introduction of the EHR, medical administrators were authorized to enter patient-related data and, therefore, a shift was perceived in the administrative burden from the medical administration to the medical specialists.
Various medical specialties found that they were not able to collaborate well because each specialty worked in a different medical context in the EHR. As a result, EHR users were hindered in understanding what had taken place when something went wrong in terms of orchestrating because the two parties had different system representations.
In two outpatient clinics (A and D), the lack of organization-wide policies made it difficult to actualize the collaborative affordances. However, all the medical specialties involved were required to use the EHR according to the department’s own policy. As such, all the specialties were assured that certain affordances were used. As a result, actors were guaranteed that their messages would be actually read by the right actor. In the other outpatient clinics (B, C, and E), no indications were found that a department-wide policy had been implemented. As a consequence, collaborative affordances were perceived and actualized differently by the various medical specialists.