ABSTRACT
The 4C Mortality Score (4C Score) was designed to risk stratify hospitalised patients with COVID-19. We assessed inclusion of 4C Score in COVID-19 management guidance and its documentation in patients’ case notes in January 2021 in UK hospitals. 4C Score was included within guidance by 50% of sites, though score documentation in case notes was highly variable. Higher documentation of 4C Score was associated with score integration within admissions proformas, inclusion of 4C Score variables or link to online calculator, and management decisions. Integration of 4C Score within clinical pathways may encourage more widespread use.
INTRODUCTION
Early identification of patients with COVID-19 at risk of poor outcome could support early decision-making regarding treatment and escalation of care (1). The 4C Mortality score (4C Score) was developed to better inform clinical decision-making in patients hospitalised with COVID-19 (2), and has been validated in COVID-19 patients hospitalised between August 2020 and February 2021 (3). This study aimed to assess if, and how, the 4C Score was incorporated into the assessment of COVID-19 patients in UK hospitals in January 2021 during the peak of the second wave (4).
METHODS
Between 3 June and 30 September 2021, investigators from all UK hospitals were invited to take part in a cross-sectional study on COVID-19-specific management guidance at their hospital site, and the role of the 4C Score within these documents. Additionally, they were invited to undertake a retrospective case note review of adult patients presenting with community-acquired COVID-19 during a 2-week period (11-24 January 2021) at the peak of the ‘second wave’ to ascertain the proportion of patients that had a 4C Score documented in their case notes. The protocol was developed in collaboration with the UK National Infection Trainee Collaborative for Audit and Research (NITCAR) following STROBE guidance (5). Site investigators were recruited through NITCAR and local trainee networks. Data were collected on a REDCap database (Research Electronic Data Capture, Vanderbilt University, US) (6). This study was classified as service evaluation by the West of Scotland Research Ethics Committee. Caldicott Guardian approval was obtained at each participating site. Details of the study protocol are included as an online Appendix.
RESULTS
Forty-one hospitals in England and Scotland participated in the cross-sectional study of COVID-19 specific guidance, 32 of which took part in the retrospective cohort study (Supplementary Table S1). Characteristics of participating hospitals are summarised in Table 1.
Characteristics and guidance document use in participating hospitals
All but one hospital (97.6%) reported the use of COVID-19-specific guidance. The most common forms of guidance were guidance documents (n=38; 95.0%) and admissions protocols (n=18; 45.0%) (Table 1; see Appendix for definitions). Twelve sites (30.0%) utilised three or more forms of COVID-19 specific guidance.
Thirty (75.0%) hospital sites referenced one or more prognostic scores in their COVID-19-specific guidance; 4C Score was the most commonly included (n=20; 50.0%), followed by Clinical Frailty Scale (CFS; n=18) and NEWS2 (n=17) (Table 1). 4C Score was utilised alongside one or more other prognostic scores at 14 sites, most commonly with CFS (n=12). Thirteen (65.0%) sites that included 4C Score within their COVID-19 guidance used it to inform management decisions, including treatment escalation (n=8), discharge (n=4), admission destination (n=3), treatment limitation (n=3) as well as remdesivir prescription (n=3).
After assessment against eligibility criteria (see Appendix), 4,123 case records from 32 hospital sites were analysed (Figure 1). 4C Score was recorded in >10% patient records in eight (25.0%) sites, ≤10% records in 10 (31.3%), and 0% in 14 (43.8%) sites (Table 2).
Flow chart of eligibility criteria for retrospective case note review
Overall, it was documented in the records of 418 (10.1%) patients, with wide variation in frequency of 4C Score documentation between hospital sites (median 8.2%, range 1.0-74.1% among 18 sites with 4C Score documented in ≥1 case record). The score was mostly documented by junior doctors from foundation to specialty trainees (77.5%, n=324), and predominantly in medical assessment units or medical wards (93.8%), compared to the Emergency Department (3.8%).
The eight sites that recorded 4C Score for >10% of reviewed patients were more likely to include the score within the admissions proforma (62.5% (5/8) vs. 16.7% (4/24), p= 0.02), detail the variables or incorporate an electronic link to the online calculator (87.5% (7/8) vs. 33% (8/24), p=0.008), and utilise the score to inform management recommendations (75.0% (6/8) vs. 20.8% (5/24), p=0.01).
DISCUSSION
No prior studies have explored the use of COVID-19-specific guidance or the application of 4C Score in the assessment of COVID-19 patients. In this nationwide study, most surveyed hospitals (97.6%) had COVID-19-specific guidance documents; around two-thirds employed multiple forms of guidance. The 4C Score was the most frequently referenced prognostic score, often in conjunction with CFS. The latter is the only clinical stratification tool referenced in the NICE COVID-19 management guidance (7). The 4C Score was chiefly used to inform treatment escalation, but was also used to inform decisions on patient destination (i.e. discharge or admission location) as well as treatment limitation.
Although half of participating sites incorporated the 4C Score within guidance, documentation of the score varied substantially between sites. Audits of documentation of established mortality risk scores such as CURB-65 (for patients with community-acquired pneumonia) have also demonstrated variable recording (8,9). Nevertheless, our findings suggest that documentation of the 4C Score may be encouraged by incorporation into clinicians’ workflow through admissions proformas, detailing 4C Score variables or online calculator within guidance, in addition to linking the score to management recommendations.
This project demonstrates the capacity of national trainee networks to facilitate multi-centre studies. Collaboration with NITCAR and other trainee networks enabled the rapid recruitment of 41 UK sites and completion of >4,000 case note reviews within a short timeframe.
This study has several limitations. Although NITCAR is a national organisation, site participation was not representative, with a preponderance towards Scottish sites and no sites included from Wales or Northern Ireland. We assumed that guidance documents reported by sites between June and September 2021 were in use in January 2021 (period selected for retrospective case note review); some of these documents may not have been in use at that time or may have since been updated. We equated documentation with use of the 4C Score; this would have omitted occasions when a score was calculated but undocumented in the case notes, or conversely, decisions that were made without the value being considered.
In summary, this national study highlights that 4C Score and other clinical stratification tools were widely adopted in COVID-19-specific management guidance. Ongoing validation of the 4C Score in the context of vaccinated patients and emerging SARS-CoV-2 variants is needed. However, ensuring that the score is accessible to clinicians through integration within clinical pathways could lead to more widespread deployment.
Data Availability
All data produced in the present study are available upon reasonable request to the authors
Contributors
AH is guarantor and corresponding author for this work and attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted. Contributor role taxonomy (CRediT): conceptualisation: AH, AEB, AM, JSP; data curation: AEB, JSP, AH; formal analysis AH; investigation: JSP, AA, SB, KB-E, SB-M, RB, TC, DC, AC, TC-B, JE, JG-B, BH, HH, AK, SL, CL, KM, CM, BM, CM, EM, JN, DP, LP, MS, DS, SS, JT, FT, SW, EW, RW; methodology: AH, AEB, AM, JSP; project administration: JSP, AEB, NITCAR, AH; writing – original draft: AEB, JSP, AH; writing – review and editing: AA, SB, KB-E, SB-M, RB, TC, DC, AC, TC-B, JE, JG-B, BH, HH, AK, SL, CL, KM, CM, BM, CM, EM, JN, DP, LP, MS, DS, SS JT, FT, SW, EW, RW, AM, NITCAR.
Funding
The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-project sectors.
Competing interests
AH has received grant funding from UKRI for work that included the development and validation of 4C Mortality Score. Presented data included in this manuscript at the St. Andrew’s Day Symposium: Updates on Acute Medicine, Royal College of Physicians Edinburgh.
Patient consent for publication
Not required.
Acknowledgements
The authors would like to thank the following for their contribution to this study. NITCAR committee for support and assistance in facilitating this nationwide study. Vanesa Anton-Vazquez, Ali Alam, Matthew Kennedy, Imran Khan, Darver Malik, Andy Nicholson, Linda Provan, Matthew Stevens, Omolola Wilson for contributing hospital site data regarding COVID-19 specific guidance.
Footnotes
↵* Joint first authors