Scale and pace
From March 21st to March 28th, 18 DGHs in Wales received several hundred hardcopy guideline posters, subsequently distributed in areas where relevant HCPs could easily access them, such as COVID-19 wards, medical assessment units and emergency departments. Figure 3 highlights the publication of the guideline coincided when total confirmed COVID-19 inpatients and COVID-19 deaths were low. Registration rates increased substantially around 28th March in response to a range of alignment and facilitation activity, including email campaigns, formal guideline on-boarding, and discussions promoting adoption with executive teams. New registration rate slowed commensurate with a reduction in the rate of patients admitted to hospital and dying from COVID-19 (Fig. 3). Total registrants reached 4521 during the first wave (Fig. 4).
Penetration of the target
The primary target audience, consultants, accounted for the greatest proportion of professionals registered with the guideline platform (23%). We observed uptake across allied health professionals (including physiotherapists, pharmacists, dieticians and occupational therapists) accounted for 21.4%, and nurses 20.6%. We next evaluated uptake for consultants predicted to manage patients admitted with COVID-19, estimated as the sum of all ED, Respiratory, intensive care, and palliative care consultants across Wales (Supplemental Table A). We next compared this to all consultants. From a possible 2505 consultants employed in Wales (7), 1131 (45%) registered with the guideline. As the total number employed within each HB was known, we used this to normalise uptake between HB and derive a penetration ratio (Fig. 5). HB6 showed the greatest penetration, with 73.7% (325 of a possible total 440) of all consultants registered, followed by 51.9% (111 of 214) within HB5. Uptake was lowest in HB1 at 30.5% (131 of 429 consultants) but still superseding the original target (x3.6). This variation in consultant uptake between HBs was highly significant (Chi-squared testing, p < 0.0001).
Given the potential impact of sickness and staff transfers on this estimate of guideline uptake, we conducted additional sensitivity analysis to derive penetration ratios, using publicly available figures for total HB catchment population, total number of clinical staff, number of acute beds, and COVID-19 admissions (Supplementary Table B). This confirmed the observed trend in guideline registration between HBs. Remarkably; we observed a ratio of four HCPs registering within HB6 for every COVID-19 admission. Within HB1 (with the lowest penetration), this fell to approximately one HCP for every two COVID-19 admissions.
To better understand the potential influencers for guideline registrations we assessed facilitator activity. We calculated the number of unique interactions (number of logged in page visits) with the GFD. There were 972 interactions in total. Of these, HB6 had the highest-most interactions (642, 67%) (Fig. 6). The HB with least interactions was HB1, with nine interactions (0.9%). This is a similar figure to HB5 (1.4%) suggesting penetration accounted to more than facilitator activity alone.
To explore further reasons for the low penetration in HB1 we investigated the burden of COVID-19 within each HB at the time of guideline publication. This estimated the degree of organisational readiness across each HB. At the time of guideline launch, HB1 had the greatest number of COVID-19 inpatients when compared to the other HBs. This equates to a greater percentage of the peak number of inpatients (37.4%) when compared to other HBs (4.9%, 10.4%, 6.3%, 5.7%, and 5.9%, respectively) (Table 1). This suggests HB1 had low organisational readiness at the time of guideline release.
Table 1
– Table of figures representing inpatient numbers at time of guideline launch against its peak for each HB.
Health Board
|
COVID-19 inpatients around the time of the guideline launch (data from 22nd March)
|
Maximum COVID-19 inpatient count during the first wave
|
COVID-19 inpatients around the time of the guideline launch as percentage of peak
|
HB1
|
107
|
286
|
37.4%
|
HB2
|
12
|
243
|
4.9%
|
HB3
|
26
|
250
|
10.4%
|
HB4
|
13
|
208
|
6.3%
|
HB5
|
5
|
88
|
5.7%
|
HB6
|
12
|
202
|
5.9%
|
Survey Responses
In total, 178 healthcare professionals responded to the survey representing 3.9% of the total number of registrants at the time. Of all responses 33.9% of these were consultants, with 23.1% nurses and 26% reported as ‘other’. The average rating of the guideline platform was 4.01 out of a maximum of five stars. Of all respondents from the survey, 68% had encouraged others to use the guideline platform, 28% had not. The majority of respondents reported using the guideline weekly (26.6%), 23.2% using it 2–3 times per week and a further 22.6% used it daily. The majority of respondents accessed the guideline most often whilst on duty at work (53.2%), whilst 26.3% accessed it most often from home, whilst not on duty, and a further 19.3% during work, whilst off duty. The majority of respondents accessed the guideline using a hospital computer (57.3%), 24.0% accessed using mobile phones, 20.5% using their personal computer, and 7.0% using a tablet device. A commonly reported complaint to the central guideline management team however, was accessibility issues using hospital computers, where local firewalls blocked access to the guideline website and/or video play function. The mean sliding scale score for the extent to which the guideline informed their practice was 63 out of 100 indicating the majority of reported the guideline informed their clinical practice. Eighty one per cent of respondents indicated that they would like the update emails to continue.
Volume of content available
At the time of writing, 7 national pathways and approximately 260 information pages are freely available to HCPs online (www.covid-19hospitalguideline.wales.nhs.uk). More than 180 pre-recorded video tutorials featuring 45 clinical specialists are freely available to registrants, which have resulted in 31,000 video plays. During the first wave alone, there were nearly 170,000 page views from those signed in. Google analytics show approximately 40,000 sessions, consisting on average of 4.2 page views per session and average session duration of over 5 minutes. The guideline continues to provide weekly updates throughout the second wave. There were 101 email campaigns (mail-out clinical updates and video synopses, to registrants). In total, 207 registrants unsubscribed to the emails (4.6% of total registrants). This equates to 2.5% unsubscribes per email campaign, or 0.04% of all users per email campaign. Of the un-subscribers who entered their job title (n = 189, 91% of un-subscribers), the highest proportion came from Medical Students – 21% (n = 40), followed by ‘Other Healthcare Professional’ – 17% (n = 33) and Registrars – 16% (n = 31).