Medical specialists have quite varied attitudes towards eHealth; some are strong advocates while others remain pessimistic. Underlying these attitudes are their perceptions of the benefits and risks associated with adoption. These attitudes are strong determinants of adoption rates in each specialty segment. But the segments are not homogeneous: in each there will be specialists who are quite resistant to eHealth applications, and those that verge on being eHealth evangelists. Identifying potential resistors and catalysts for change will be critical for advancing adoption, as will uncovering ways to deploy eHealth’s strongest advocates in a way that helps influence their more hesitant peers.
The most commonly agreed eHealth benefits include collaboration, continuity of care, and efficiency (Exhibit 12). However, these perceptions are strongly held by less than half of all specialists. Moreover, most of the remaining benefits had just 15–30 percent of specialists in strong agreement. Given this variation in attitude, we undertook additional analysis to understand the underlying drivers and primary reasons for differences between specialists.
When evaluating attitudinal readiness across purely demographic dimensions (e.g. age, geographic region, gender), the differences between groups were not overly strong (Exhibit 13). Viewing results by specialty segment offered greater variation, but we still observed strong advocates and resistors within most segments. As discussed in the methodology section, we decided to identify clusters, or distinct groupings of similar specialists, based on specialists’ attitudes towards computers and perceptions of eHealth benefits and barriers. This approach provided rich and robust insights on attitudes and needs, but in the absence of a demographic component, the clusters were difficult to identify and reach. Therefore, we added the percentage of income derived from the private sector to create a hybrid clustering approach, which combined all of the attitudinal dimensions from the original analysis with a demographic attribute. The resulting clustering grouped medical specialists to form reachable, interpretable and distinct clusters so that specific engagement strategies could be targeted to relevant practitioners. (Detail on strategy development is provided in Section 7.)
Cluster analysis of attitudes and driversApplying the approach described above, we identified five overarching eHealth attitudinal clusters of medical specialists (Exhibit 14). Comparing specialists across clusters, they are distinct in the benefits they see in eHealth applications, the barriers they perceive, and their primary influences.
Clusters exhibit some differences in demographics (Exhibit 15). eHealth entrepreneurs and network adopters skew younger (20 percent and 23 percent over age 55, respectively) while the uninterested cluster skews older (64 percent are over age 55). The uninterested cluster is predominantly male. Network adopters are most commonly found in the public hospitals; the remaining clusters skew more heavily towards private practice, especially the apprehensive followers (69 percent of income from the private sector) and the uninterested (89 percent). Geographic dispersion is relatively consistent, though the eHealth entrepreneurs are the least likely to be in a major city location (21 percent based outside of a major city, versus 13 percent for the uninterested). Internet use for personal reasons is consistent between most clusters except the eHealth entrepreneurs, who are more likely to spend over 10 hours a week online (21 percent versus 12 percent of all specialists). The uninterested are most likely to be native English speakers (89 percent); the network adopters are the least (75 percent).
Although specialist segments often have one or two dominant clusters, most segments still contain representation across all five, indicating that the clinical workflow needs do not negate the need for eHealth solutions. However, the clinical needs of each sector and some demographic factors explain the cluster distributions depicted in Exhibit 16. While there will be strong eHealth supporters in each specialist segment, how they are leveraged will differ in each segment’s engagement and adoption strategy, particularly in those segments where the more resistant clusters (3–5) are in the strong majority.
We describe the clusters below, in order of their likelihood to adopt, and influence others to adopt, eHealth solutions. The relevant research findings are then summarised in Table 5 and Exhibits 17–21.
- Cluster 1: eHealth entrepreneurs (24 percent of all medical specialists). These specialists have the strongest interest in and use of eHealth solutions, and are willing eHealth participants. Anticipating strong benefits in efficiency, continuity of care and practitioner collaboration, their main concerns are IT compatibility and reliability issues. They are almost twice as likely as other clusters to expect that eHealth will improve patient engagement, relationships and satisfaction. Though internal medicine practitioners are the most prominent eHealth entrepreneurs, the cluster is well-represented in nearly all specialty groups.
- Cluster 2: Network adopters (17 percent of all medical specialists). These specialists commonly work in an environment, such as public hospitals, where computer use is expected. In these settings, they anticipate eHealth improvements in collaboration and continuity and quality of care, so that internal IT compatibility and reliability are their main concerns. While enthusiastic about the benefits, they are difficult to influence directly because they have restricted influence over their operating environment and so are less likely to control purchasing and adoption decisions within their practices. Accordingly, use of eHealth applications by others in their network is a much stronger driver for adoption than financial incentives or patient demand. Anaesthesia, emergency medicine, and diagnostics specialists are over represented in this cluster.
- Cluster 3: Capable but unconvinced (13 percent of all medical specialists). These specialists are reasonably tech-savvy, have financial resources and IT support, and are relatively unconcerned by potential barriers such as IT compatibility, cost or privacy. They currently use a range of eHealth applications, but only the ones with clear benefits. Relative to the first two clusters, they are much less interested in adopting the eHealth solutions that they do not currently use. For example, 44 percent of the cluster responded that it didn’t need interactive decision support for ordering tests versus 18 percent of eHealth entrepreneurs and 23 percent of network adopters, even though computerised test ordering rates were similar across all three clusters. They are unimpressed by the surveyed adoption drivers, with only peer practitioner use holding sway with them, and even then not greatly. Practices for which IT failures or delays would be either costly or critical, such as surgeons and emergency specialists, are over represented in this cluster.
- Cluster 4: Apprehensive followers (30 percent of all medical specialists). These specialists see some benefits in eHealth applications, but are less enthusiastic about them than the first two clusters, and so are waiting for others to adopt them first. The perceived benefits are, for these followers, heavily outweighed by perceived barriers. For example, they are more than four times more likely to be concerned about malfunction risk than Cluster 3 specialists. They have the most serious privacy concerns, are financially constrained from new investments, and have limited access to IT support. This cluster contains a fairly even mix across all specialties.
- Cluster 5: Uninterested (16 percent of all medical specialists). These specialists have the lowest IT usage rates and have negative perceptions of all eHealth benefits. They do not operate in an environment in which computer use is expected, and face many barriers to adoption. They strongly disagree that eHealth applications will bring better patient relationships, engagement or outcomes (in fact, believe the reverse). The cluster is skewed towards older practitioners, with retirement looming, and solo practitioners. No adoption lever is likely to influence them to adopt eHealth solutions. Psychiatrists and surgeons are over represented in this cluster; emergency and diagnostics specialists are almost nonexistent. Though active adoption cannot be expected, the cluster cannot be ignored, as some of its members will influence their peers and public opinion.
Table 5: Cluster use, perceptions and drivers(number in brackets reflects percent of segment that use or strongly agree):
|Viewing pathology results (86%)
Viewing/recording notes during consultations (53%)
Electronic health record (59%)
Paperless records (31%)
|Continuity of care (72%)
|External compatibility (30%)
Internal compatibility (26%)
Malfunction and downtime risk (17%)
|Financial incentives (56%)
Professional body endorsement (50%)
|Viewing pathology results (89%)
Viewing/recording notes during consultations (43%)
Electronic health record (53%)
Paperless records (9%)
| Collaboration (77%)
Continuity of care (71%)
Quality of care (65%)
Patient safety (61%)
|Internal compatibility (32%)
Malfunction and downtime risk (23%)
External compatibility (19%)
Privacy breaches (13%)
|Other practitioner use (16%)
Support staff use (9%)
Professional bodies (9%)
|Capable but unconvinced
|Viewing pathology results (71%)
Viewing/recording notes during consultations (39%)
Electronic health record (49%)
Paperless records (15%)
| Efficiency (41%)
|Internal compatibility (15%)
Malfunction and downtime risk (13%)
External compatibility (10%)
|Other practitioner use (12%)
|Viewing pathology results (46%)
Viewing/recording notes during consultations (31%)
Electronic health record (33%)
Paperless records (12%)
| Continuity of care (34%)
|Malfunctions and downtime risk (58%)
Prefer to wait until technology proven (51%)
Privacy breaches (47%)
|Financial incentives (30%)
Professional body endorsement (17%)
|Uninterested and IT challenged
|Viewing pathology results (20%)
Viewing/recording notes during consultations (18%)
Electronic health record (16%)
Paperless records (6%)
|Minimal||Malfunctions and downtime (40%)
Prefer to wait until technology proven (36%)
Privacy breaches (31%)