Page 289 - SAHCS HIVMed Journal Vol 20 No 1 2019

Page 5 of 9 Original Research

TABLE 1: Usefulness of the group as a learning tool (n = 91).
Variable Strongly agreed Agreed Neutral Disagreed Strongly disagreed
n % n % n % n % n %
New clinical insights 51 56 32 35 7 8 1 1 - 0
Practical application of pre-existing knowledge 49 54 34 37 8 9 - 0 - 0
Guidance according to national or international guidelines 57 63 31 34 4 4 - 0 - 0
Recommend to colleagues 58 64 23 25 8 9 - 0 2 2

clinical insights on HIV/TB, that the information discussed TABLE 2: Increased clinician confidence in managing patients and levels of group
engagement.
in the group chat was according to national guidelines Variable Greater clinical OR 95% Confidence p
and international best practice principles and they would confidence (%) interval
recommend a similar case discussion platform to other Followed group 93 8.44 2.33–35.23 < 0.05
colleagues (Table 1) Posted questions 93 3.8 1.02–18.48 0.02
Posted responses 92 3.36 0.96–13.55 0.03
Clinicians’ knowledge and use of informed OR, odds ratios
consent in the group
(OR 48.13, 95% CI 4.99–464.49); those who referred to old
The last objective was understanding the clinicians’ knowledge chat cases were 21.42 times more confident (OR 21.42, 95%
of informed consent when sharing patient information on CI 4.39–104.84); there was also an increase in confidence in
social media. From the responses, 89% of the participants participants who reported that they had gained new clinical
reported that they were aware that according to HPCSA insights while participating in the group (OR 23.75, 95%
regulations, they needed to obtain documented patient CI 3.95–142.88).
consent when posting a patient-related image on social
media. However of those that reported posting questions, When looking at recommending the group to colleagues as
only half obtained consent (52%) versus 48% not obtaining an outcome, participants who report gaining new clinical
consent, when posting patients’ photographs or other insights were 17.33 times more likely to recommend the group
medical images on the group (even if patient identity was not (95% CI 3.13–96.01). Those who reported that the group
revealed). When asked if they obtained documented patient helped them to practically apply pre-existing knowledge and
consent when posting patients’ laboratory results on the felt that the guidance given was according to national or
group (even if patient identity was not revealed), around international guidelines were also, respectively, 12.82 (95%
two-thirds (68%) of participants who had posted said they CI 2.55–64.56) and 20 (95% CI 1.63–245.63) times more likely
had in fact not obtained consent, less so than when posting to recommend the group to other colleagues as a case
other medical images. discussion platform (Table 4).
In terms of group engagement and recommending the group
Bivariate analysis
to others, those who followed the group regularly were 4.79
Using a bivariate analysis, with cross-tabulation in Epi times more likely to recommend it (95% CI 1.19–21.10). There
Info, any statistically significant associations were looked was no difference in those who posted questions and responses
for in those clinicians who reported feeling more confident (Table 5).
in managing their patients after group participation and
whether they would recommend the group as a learning Lastly, we looked at whether Internet access impacted
platform to other colleagues. clinicians’ reported ability to follow the group, but there was
no clinically significant association found.
Table 2 looks at any association between group engagement
as an exposure variable, and increased clinician confidence as Discussion
an outcome. In doctors who followed the group regularly, Group engagement and its usefulness
there was a clinically significant increase in OR (8.44, 95% as a learning tool
CI 2.33–35.23), participants being 8.44 times more likely to
have increased confidence in managing their patients. Those The responses from participants in this study were
who posted questions also had an increase in OR, 3.8 times favourable in the reported use of the WhatsApp group and
more likely to have an increase in their clinical confidence its application as a learning tool. The majority of the
(95% CI 1.02–18.48). participants firstly cited regular Internet connectivity, which
facilitated the uninterrupted use of the application and
Other associations were found in increased clinician confidence communication in real time. They also reported using the
19
in managing patients as an outcome, cross-tabulated with group discussions as a guide to further manage other
participant perceptions of the usefulness of the group as a patients, referred back to old chat discussions and were
learning tool (Table 3). Of statistical significance, participants satisfied at the timeous response to cases (including
who used the chat group guidance to manage their patients the peer responses themselves). Group engagement, or
were 48.13 times more likely to be confident afterwards participation, measured by following of the group and

http://www.sajhivmed.org.za 282 Open Access