Below, interview findings pertaining to social and personal influence are reported, organized by source of influence or referent. For each referent or group of referents information is provided on the direction of influence, the relation of the referent to the physician, the arguments used by the referent to influence physicians, the channels through which influential communication was conveyed, and the means of influence and corresponding psychological processes involved. The findings are summarized in .
Social and personal normative influences on physicians to use electronic health records (EHR).*
3.1. Hospital and clinic administration
The most commonly mentioned source of social influence was a physician’s (local) hospital administration. Physicians with outpatient (i.e., primary care) practices also often mentioned their outpatient clinic administration. Of the 13 physicians explicitly asked about the entity that most influenced their EHR use behavior, six (46.2%) said it was their hospital or clinic. Hospital and clinic influences were exclusively positive—i.e., encouraging EHR use.
Although all physicians were formally affiliated with their respective hospitals, only one physician used personal language (“my employers”) to make explicit her relationship to the hospital. Two physicians spontaneously named specific clinic or hospital leaders (e.g., “I mean … by hospital administration, the person who communicates to us are, is our liaison … they’re the ones who send us the e-mails that say what’s going on”); another did so when prompted to name the face associated with hospital pressure to use EHR. The rest used impersonal language such as “the hospital,”
“the administrative bodies,”
“the leadership,” and “their rules,” suggesting social distance between physicians and the hospital. Four physicians commented that the hospital argued for EHR use by appealing to EHR’s benefits for quality and safety of patient care, for example: “[Hospital Leader], is a big patient safety advocate and will frequently, you know, refer to the fact that, you know, [EHR] is going to make things so much better and safer.” About as many physicians mentioned business reasons (productivity, efficiency, better billing) as the arguments for EHR use communicated by the hospital, and one physician described a third argument: “that’s part of their message, that this is kind of the way medicine is going, and we need to do this…”
The hospitals communicated their encouragement of EHR explicitly using channels such as campaigns, training sessions, and e-mail. Additionally, the hospitals’ massive expenditures on the EHR system and technical support were seen as an implicit message that physicians should use EHR, as illustrated in this exchange:
Interviewer: How do you hear about their approval of it?
Respondent: Well, the fact that they’re buying it #laughter# you know?
The hospital or clinic influenced EHR use predominantly through administrative controls—“they’re kind of making us use the technologies”
—and by eliminating alternative options:
Respondent: I mean, approving is not the right word … they’re coercing me to use it.
Interviewer: So how do you, how do they coerce you?
Respondent: Well, by simply making that the only place where the information is available.
3.2. Fellow physicians
Physicians often described the opinions of their physician colleagues about EHR use and indicated that these physicians may also be a source of social influence. Stated an ophthalmologist, “other than, you know, listening to what I, myself, tell myself, I think I probably listen to my partners the most.” In fact, of the 13 physicians who were asked, six named fellow physicians as the entity most strongly influencing them. Fellow physicians mainly encouraged but sometimes discouraged EHR use, or as one physician put it, “it’s very mixed on the physician side, but I think most physicians you speak to who have been using electronic health record[s] don’t want to go back to the way it was before because of the advantages and I think the potential that they see. So I think in general the physicians are supportive.”
Not surprisingly, respondents used language that revealed social affiliation with fellow physicians (“my colleagues,” “my partners,” “other physicians in my department”). One physician described that she was most strongly influenced to use EHR by a “techie” colleague “because he’s a father figure.”
Describing the integrated and collaborative nature of patient care, respondents explained why fellow physicians might encourage one another to use EHR, for example, “having a clear, consistent line of communication for the next person that is coming on is very important”
or as another physician stated,
“Because we want to be able to share information with each other and I guess if like one third of our colleagues were not using the system, then that brings down the quality of the data, so it makes sense for all of us to be involved in the system. So there is some, I think, there’s some kind of peer pressure to make sure that we’re using it.”
Respondents reported that fellow physicians who opposed EHR used counterarguments including the difficulty of doing data entry and the extra burden that this imposed; that EHR brought about undesirable changes in workflow; that EHR intruded on patient-physician interaction; that the time costs of using EHR reduced productivity and income; and that the EHR software was not yet perfected and “they’re then put in that position of having to help develop the product.”
Colleagues related their opinions of EHR mainly through interpersonal communication. However, few of the communications were described as explicit attempts to influence peers. Colleagues are nevertheless likely to influence one another’s opinions and behavior through less direct processes, including informational social influence, conformity, and social learning (Fulk 1993
) (see ). Informational social influence refers to “an influence to accept information obtained from another as evidence
about reality” (Deutsch and Gerard 1955
, p.629). In this case, it would mean that physicians who heard about fellow physicians’ experiences and opinions of EHR may have interpreted their colleagues’ accounts as evidence about the EHR system, thus altering their own opinions and subsequently their behavior. This is especially likely to occur when one trusts the source of information (e.g., a close colleague) and when one has little personal experience with the target behavior (e.g., has not used a certain EHR feature). Take for instance the following comment by an obstetrician whose hospital’s EHR had no data entry functionality yet, remarking on his fellow physicians’ influence: “So they’re not trying to influence me as much as they, you know, group-think is like, ‘Oh, this … [data entry] is going to make it slower for me.’”
It is possible that an accumulation of such comments from fellow physicians will lead one to believe that using EHR with data entry functionality will truly slow down work, predisposing a negative attitude that might affect how and perhaps even whether one will use data entry when it becomes available. Similarly, others’ behavior may be interpreted as accurate information about an EHR. One physician commented, “I’m telling you, I know that there are docs, I know docs who have retired from practice earlier because this was sort of the last straw.”
Physicians might infer from colleagues’ premature retirement that using EHR is undesirable; they might alter their behavior to match, perhaps by resisting implementation, by not using the full extent of EHR features, or by working around the system.
Theories of conformity further suggest that by observing the behavior of others around them, physicians may learn about and attempt to conform to what appear to be the social norms for behavior (Cialdini and Trost 1998
). One motivation to conform is to gain or preserve social status. Thus, being around fellow physicians who extensively use EHR may influence one to do the same in order to fit in. One physician described such a situation:
“The fact that we’re in an environment where people may not want to use the systems, but do use the systems, you know, they actually rely on them now… And there’s sort of a cultural issue here, so I guess if there wasn’t a culture where everyone was cool with accessing these systems, I would be much less excited about using this system.”
Many physicians described how physicians around them worked around EHR (e.g., writing orders on paper and having a nurse enter them electronically), used it selectively, or used it poorly (e.g., not updating information in the EHR, expecting others to do it). If such behaviors were common enough among important social agents, they may have been interpreted as social norms worth conforming to or at least not blatantly violating.
A final way that physicians in the study may have been influenced by their physician colleagues without being directly persuaded to alter their EHR use behavior is through social learning. This refers to the acquisition of behavioral patterns by observing others (Bandura 1977
, Fulk 1993
). By observing (or being shown) a fellow physician using EHR in a seemingly advantageous way, one may emulate that behavior not only to improve one’s social standing (as in the case of conformity) but also because it may improve one’s work. For example, physicians described how “we sort of teach each other”
and that “the things you don’t know … you stumble upon by working with someone else or working with a colleague that says, ‘Oh, you can get this information and then [they] show you how to do it.’”
Research on social learning shows that even when observed behaviors are not meant to be taught they may still be learned. Thus, when and how surrounding physicians use EHR may influence fellow physicians’ own use. That influence can be desirable or undesirable. For example, several physicians noted that too many colleagues copy and paste whole clinical notes rather than importing only the most important contents from prior notes and adding only new, relevant information to subsequent notes. It is possible, therefore, that through observing others’ note writing, some physicians will find the practice of indiscriminate copy-and-pasting bothersome and will not do it themselves whereas others may see it as a useful time-saving behavior worth replicating.
Several physicians spontaneously mentioned patients as a source of social influence; those who did not were asked explicitly about patients’ opinions about physicians’ use of EHR. Physicians believed that when patients had opinions about EHR use, those opinions were predominantly positive (“I think patients in general are very happy about it”) and that patients wanted their physicians to use EHR (“They think that that’s the way that it should go”). Reasons for patients’ approval of EHR use, as related by physician respondents, included access to records that go “pretty far back;” access to their own medical records via a patient portal; possibly improved safety and quality of care; the perceived clarity, thoroughness, and portability of the record; the potential reduction in duplicate tests; and because patients “like what they think is new and efficient.” One physician commented, “I think the patients feel like they’re getting better care by having [EHR]. I don’t think they necessarily understand what that truly means.”
Three physicians explained why patients may instead discourage EHR use. Said a psychiatrist, “my paranoid patients hate me putting anything on the Internet about them because they’re just sure that their enemies are going to get it.”
The remaining physicians explained patient’s disapproval in terms of worsened patient-physician interaction:
“I think some patients don’t like it. It changes the interaction with patients, so that you may be looking at a screen and typing while they’re talking … It certainly made it a little more challenging, and I know some patients have expressed that they don’t feel as much of the personal connection that used to be there.”
Although physicians may not seek acceptance into the social group from patients as they do from fellow physician peers, they may nevertheless have a high motivation to comply with patients’ wishes. One physician, when asked which social entity influenced him the most, replied, “well, it ought to be patients,” and it is possible that other physicians share this sentiment.
However important the patient opinion might be, eight physicians said that patients, especially hospital inpatients, were not even aware of their physicians’ EHR use. “The patient doesn’t see me do anything with the [EHR],”
said one anesthesiologist. “You know, it isn’t to the point where I have a computer in front of me while I’m talking to a patient.”
Similar comments were made by a rheumatologist, cardiologist, obstetrician, ophthalmologist, orthopedist, and two family medicine physicians, suggesting that lack of awareness was not strictly confined to one specialty or patient population. As one respondent summed up:
“They don’t know. I mean, I think the patients aren’t aware, at least my patients aren’t aware, and I even have savvy patients. I don’t think it affects how theyperceive the hospital and the care. They’re more concerned about nurses seeing them on time and getting their bedpans and their needs met.”
Physicians believed that inpatient’s unawareness could partially be attributed to physicians using EHR in the hall or at the nurses’ station, rather than in the patient’s room. This could be contrasted to the outpatient clinic, where “we’re seen with a computer terminal right there as we’re talking to them, very different.”
Additionally, some physicians were not aware of their patients’ opinions:
“You know, patients would be a very important constituency, and I have no idea, actually, whether they’re positive, negative, or neutral. And I guess I can say that I really never asked any of my patients, like kind of how do you feel about me having access to all this information about you? So, I really don’t know, and I imagine there’d be a wide spectrum of responses. So I guess I should know more, but I don’t know about that constituency.”
3.4. Intra-organizational clinical groups
Three physicians identified the internal quality improvement group as a source of social pressure. Other mentioned groups were the hospital compliance staff, a group responsible for integrating inpatient and outpatient information, the physician’s clinical department, and the physician practice group. As with the hospitals and clinics, those groups only encouraged and did not discourage EHR use.
In contrast to the impersonal language physicians used to refer to their hospitals and clinics, physicians spoke of “our department”
and “our practice groups,”
suggesting belonging to those groups. One physician spoke of being an insider of a quality improvement group whereas others referred to such groups as outsiders (e.g., “the quality people,” “the compliance folks”
). According to physicians, the groups had different motives and thus used different arguments to achieve compliance. Quality improvement groups, stated one physician, are
“people who are trying to get us to do really good chronic disease management, people who are looking at outcomes and trying to figure out how can you measure outcomes… now we have the capability because … the information is captured in a way that most of the time is searchable and sort of dissectible … So those people are ecstatic about this.”
An interviewee who was a member of the quality improvement group also identified easy analysis of a large set of outcomes data as the group’s reason for encouraging EHR use, and especially emphasized that the group encouraged uniformity in EHR use (“if you have everyone autonomously using the electronic health record in their own way you can’t coalesce data”
). Not surprisingly, this physician strongly personally believed that using EHR had quality improvement benefits, suggesting that through his affiliation with the group he shared or internalized its rationale for EHR use. For the compliance group, the rationale was to integrate documentation and billing so that patients were charged for the actual treatment they received. The group integrating inpatient and outpatient information argued for using EHR to gain operational efficiency and cost savings. For the hospital physician practice group, the arguments were multiple: “the promise for greater efficiency, better outcomes, especially in an environment where public reporting of those things is becoming more prominent.”
Unlike the hospital and clinics, intra-organizational groups were not spoken of as controlling or forcing physicians to use EHR. It is possible, therefore, that these groups sought compliance rather than obedience.
3.5. Extra-organizational professional groups
Five physicians mentioned professional organizations such as the national and state Academy of Family Physicians, the American College of Physicians, and the American Medical Association. Physicians referred to the professional societies as “interested in quality and outcomes”
and “more around the outcomes and safety piece.”
For the most part professional groups were described as advocates of EHR as a useful tool if not necessarily persuading physicians’ use behavior. A physician explicitly identifying herself as an orthopedist described how her professional society might still indirectly influence her behavior:
“Most orthopods are very technology-oriented, because we need to [use it for] our surgery and everything, and so the Academy [American Association of Orthopaedic Surgeons] also is very technology-oriented. And not that they say, ‘Oh, you know, electronic records are, are good,’ but there’s a lot of information out there about electronic records, and even in our own offices that … yeah, would influence you to say this is a good way to go.”
Another physician’s comments showed identification with a professional group and the internalization of the group’s preference for EHR:
Respondent: This is a group of psychiatrists who are techies. #indicating on computer # There’s my boys. And so--
Interviewer: This is an organization?
Respondent: Yeah, I’m going to join it … Very, very fun. So, they encourage it.
3.6. Extra-organizational advocacy, legal, and regulatory groups
Physicians mentioned local and national organizations of which they were not members that advocated EHR use, including the local patient safety collaborative; the state collaborative for healthcare quality; the state medical society; and the Leapfrog Group, a national patient safety advocacy group. For the most part those groups were believed to pressure hospitals and clinics to adopt EHR systems, although they may also have influenced physicians’ beliefs:
Respondent: … Leapfrog made recommendations, and I think the hospitals are listening to those recommendations, because they’re a powerful group of people …
Interviewer: Do you feel like they have any impact on your own opinion, or …
Respondent: I mean, I think that they have some very interesting suggestions. I mean, it’s been a while since I’ve read the information they put out, but, I think it makes me think that some of the stuff that’s being implemented is worthwhile.
Physicians perceived social pressure from malpractice carriers, whose ability to defend physicians increased with more legible documentation, and from payers (insurance companies, Centers for Medicare & Medicaid Services [CMS]), who also benefited from clinical documentation that was easier to access, read, and track. Another physician mentioned that CMS and the Joint Commission, an accrediting organization, see “potentially some of the safety and quality potential of [EHR], and so I think there’s strong support.”
Finally, nine physicians believed that “the government is obviously encouraging this.” The influence of the government, particularly the federal government, was especially salient for physicians at Hospital 2, where interviews took place in January and February 2009, during which time EHR was a key topic for the Obama administration and in debates surrounding the American Recovery and Reinvestment Act, with over $20 billion eventually allocated to stimulate health IT adoption. As one physician put it, “Well, I know the Obama administration is very pro-technology, and I know, I believe I read that there was gonna be some push for subsidizing conversion to [EHR]. And so that creates the expectation that we all should go to [EHR].” Other physicians believed that government support would soon be supplemented by requirements, e.g., “the Feds are coming after us. #sighs# You know, there are going to be governmental requirements—Feds, state, who knows—for electronic medical records.”
3.7. Personal normative influence
The source of personal normative influence is one’s self, although it is shaped by the social groups to which one belongs or has belonged. Personal norms might thus reflect, for example, the internalized opinions of a professional organization or other sources described above. At the same time, because personal norms may be jointly shaped over long periods of time by many social agents and not exclusively professional ones, they may have a unique influence on professionals’ use of technology. Personal norms include (a) moral beliefs, or ethical positions that motivate a particular behavior and (b) role beliefs, or one’s perceptions of the behavior appropriate to someone in their role.
Although six physicians reported not having any moral beliefs pertaining to the use of EHR, two others believed that using EHR was a moral obligation and the right thing to do because using EHR improved patient safety and quality of care. In the words of one anesthesiologist:
“Personally, my view of this is that this is the future of medicine. [Not using EHR] would be like saying fifty years ago ‘Wow, I’m not really sure I believe in the germ theory, and this whole thing about antibiotics,’ you know, this is where medicine is going. I think that it’s an incredibly powerful tool that can tremendously enhance the quality of healthcare in this country.”
Another anesthesiologist, however, questioned whether it was always right to computerize certain aspects of care. Other moral reservations about using EHR included the resultant ability to track or monitor physicians’ actions; the possibility that an unauthorized individual could access information when a physician stepped away from the computer after logging in; and the possibility of a security breach. In fact, 10 of the 17 physicians who were asked directly about moral norms expressed “reservations and concerns about … the ability to safeguard the information,”
particularly that “somebody could hack in there”
or that “way too much information collected about way too many people”
could otherwise fall into the wrong hands. At the same time, physicians believed that compared to regular e-mail, EHR enabled a secure portal for sending patients confidential messages and that compared to paper records the EHR better restricted access of patient information to authorized personnel.
With respect to role beliefs, physicians variously identified themselves in personal terms such as a typist; an average system user; a savvy user; an adventuresome computer user; someone who liked computers; and a member of “the new generation” of physicians. Mentioned professional identities included a physician; a medical director; a highly paid, highly trained professional; a professional in a complex job; and a member of a professional group resistant to change. Role theory indicates that to the extent that EHR use is a typical behavior for someone in a particular role, a person identifying themselves with that role will be influenced to use EHR. Thus, the respondent self-identifying as a younger, “new generation” physician might have been more compelled to use EHR than his older, more experienced counterparts. A medical director might similarly have been influenced to use EHR extensively to set an example. Physicians who were comfortable with computers in general may have been influenced to fully explore EHR, as expected of “a computer person.”
That certain EHR use behaviors are expected of certain roles was made evident by study physicians as they named the characteristics associated with users and non-users of EHR. According to interviewed physicians, the following types of people were more likely to use EHR and to be facile with it: younger people; physicians just starting their practice; those who grew up with computers or used computers in their private lives; people comfortable with computer use; faster typists; people open to change; people with more respect for the team; physicians engaged in quality improvement; physicians in certain specialties (primary care physicians; technology-driven specialties such as radiology, anesthesiology, and orthopedics).
The opposite types (e.g., older, less comfortable with technology, members of certain specialties such as surgery) were perceived by interviewed physicians to be less likely to accept EHR or use EHR effectively. Accordingly, several physicians explained that their self-identified role was incompatible with certain aspects of EHR use, as the following three examples illustrate.
- A cardiologist noted that a physician must “appear professional” (a role characteristic) but that the EHR’s template-driven clinical notes give the opposite impression because they are obviously computer-generated. To convince physicians to use EHR would require that the notes “don’t look like kindergarten output.”
- An anesthesiologist explained his reservations about learning to type in orders rather than giving them verbally during surgery by stating that he was not an office worker, but rather had a “clinical-based job … where you can’t always just hang up the phone and go attend to that task.”
- Several physicians invoked their roles as highly-paid medical professionals to make the point that direct data entry into the EHR was not a task befitting them (see Holden in press), for example, “you’re taking your highly paid … physicians who really, I think from a pure business standpoint, their job should be to make medical decisions, not to do data entry” or “someone who’s hired to do data entry: big deal. That’s what they’re hired for. That’s what—you’re not hired for that.” (Many interviewed physicians expressed difficulty taking on a data entry role because of the cognitive demands of data entry including difficulty concurrently entering data and attending to patients [see Holden in press].)