Lau 2020d.

Study characteristics
Patient Sampling	Purpose: To diagnose Covid‐19 acute‐phase infection and convalescent‐phase infection Design: Multi‐group study to estimate sensitivity and specificity [1] Test group ‐ Confirmed COVID patients, residual leftover sera (n = 415) [2] Control group (n = 715): [2a] Non‐Covid control; current healthy healthcare workers (HCWs) (n = 597); [2b] Cross reactivity group‐ 315 HCWs from group [2a] who received their annual influenza vaccination 4 weeks prior to testing and 118 non‐Covid patients who had antibody positive samples [dengue, hepatitis C (HCV), hepatitis B (HBV), syphilis, antinuclear antibody (ANA), double‐stranded DNA antibody (anti ds‐DNA), rheumatoid factor (RF)] from ambulatory patients (n = 433) Recruitment: [1] Test samples: Residual serum samples from cases with suspected or confirmed infection from April to June 2020. Recruited from 3 institutions in Singapore: Changi General Hospital, Khoo Teck Puat Hospital, and Sengkang General Hospital, 415 excess serum samples (from 280 individual patients) [2a] Control group: 597 samples from consenting healthy (no self‐reported respiratory symptoms) healthcare workers (HCWs) were collected (laboratory staff, nurses, and housekeeping staff) [2b] Cross‐reactivity group: non‐Covid samples. Except for dengue, all other samples for cross‐reactivity analysis were from excess serum samples from before November 2019. Plus 315 from group [2a] who had their annual influenza jab 4 weeks prior to the antibody test. Prospective or retrospective: Both [1] test samples ‐ retrospective [2a] Healthy HCWs group ‐ prospective [2b] Cross‐reactivity ‐ retrospective, prospective for HCWs with influenza vaccination, unclear for dengue fever patients Sample size: 1130 (415) of which 785 (70) were eligible for our review (279 + 66 confirmed COVID cases without eligible time split excluded) Further detail: Inclusion: [1] Subjects who had positive RT‐PCR from April to June 2020, from 3 institutions in Singapore: Changi General Hospital, Khoo Teck Puat Hospital, and Sengkang General Hospital. [2a] Healthcare workers (HCWs) consenting healthy (no self‐reported respiratory symptoms) (laboratory staff, nurses, and housekeeping staff) [2b] HCWs with recent influenza vaccination, samples that tested positive for dengue fever or other antibodies [Anti‐HCV, Hepatitis B, anti‐nuclear antibody (ANA), double‐stranded DNA antibody (ds‐DNA), rheumatoid factor (RF), syphilis]. Except for dengue, all other samples for cross‐reactivity analysis were from excess serum samples from before November 2019. Exclusion: [1] Test group ‐ PCR‐negative samples [2] Not stated [3] Not stated
Patient characteristics and setting	Setting: Hospital (Not stated if inpatients only or also outpatients) Location: Changi General Hospital, Khoo Teck Puat Hospital, and Sengkang General Hospital Country: Singapore Dates: [1] Test samples ‐ from April to June 2020 Symptoms and severity: not mentioned Demographics: Data for 349 samples included in analyses: Age 49.8 (95% CI 47.7 to 51.8), range 23‐97 years; 282 males, 67 females Exposure history: Not mentioned Non‐Covid group 1: [2b] cross‐reactivity group Source: excess serum samples from before November 2019 (except for dengue) and samples from healthy HCWs who recently received influenza vaccination Characteristics: dengue N = 74, HCV N = 3, HBV N = 13, syphilis N = 1, ANA N = 16, anti‐ds‐DNA N = 4, and RF N = 7 315 healthy HCWs with recent influenza jab Non‐Covid group 2: [2a] current, healthy HCWs Source: Not mentioned, possibly HCWs (laboratory staff, nurses, and housekeeping staff) from the same 3 hospitals in Singapore Characteristics: 597 consenting healthy (no self‐reported respiratory symptoms) healthcare workers (HCWs) Characteristics only for [2a] and [2b] combined (n = 715): Age 40.4 (95% CI 38.9 to 41.9), range 19‐81 years; 126 males, 589 females
Index tests	Test name: Roche Elecsys anti‐SARS‐CoV‐2 assay Manufacturer: Roche Antibody: Total antibodies Antigen target: Undisclosed epitope Evaluation setting: Laboratory Test method: Sandwich immunoassay (where biotinylated SARS‐CoV‐2 specific recombinant antigens and SARS‐CoV‐2 specific recombinant antigens labelled with ruthenium form a sandwich complex with anti‐SARS‐CoV‐2). Electrochemiluminescent‐immunoassay Timing of samples: 0‐6 days: 189 7‐13 days: 90 >= 14 days: 70 Samples used: Serum [1] Leftover serum [2] Serum [3] Stored serum before 2019 or serum Test operator: Lab personnel from hospital laboratories (For serology, Changi and Sengkang hospitals employed the Roche Cobas e801 while Khoo Teck Puat used the Cobas e602 immunoassay analyser) Definition of test positivity: cut‐off index (COI) of 1.0 for a positive sample Blinding reported: Not stated Threshold predefined: Yes
Target condition and reference standard(s)	Reference standard: real‐time polymerase chain reaction (PCR) test systems that targeted at least 2 viral epitopes of SARS‐CoV‐2 Samples used: Not stated Timing of reference standard: Not stated Blinded to index test: Yes performed prior to index test Incorporated index test: No Definition of non‐COVID cases: [2a] Untested, no reported respiratory symptoms [2b] Pre‐pandemic, unclear for 74 dengue patients Samples used: [2a] Untested [2b] Pre‐pandemic or untested Timing of reference standard: [2a] Untested [2b] Pre‐pandemic or unclear Blinded to index test: yes Incorporated index test: No
Flow and timing	Time interval between index and reference tests: [1] 0–6 days ‐189, 7–13 days ‐ 90, 14+ days ‐ 70, 21+ days ‐36 [2] Untested [3] Pre‐pandemic or unclear All patients received same reference standard: No Missing data: A total of 415 excess serum samples (from 280 individual patients) that tested positive for SARSCoV‐2 by PCR were identified for sensitivity analysis. Of these, 66 were residual samples from inpatients not initially suspected of having COVID‐19 but who subsequently tested positive for SARS‐CoV‐2 PCR and were excluded from the sensitivity analysis. 279 Covid samples excluded from review as no eligible time split. Uninterpretable results: not stated Indeterminate results: No cases with indeterminate or missing results were used in our study. Unit of analysis: [1] Samples (349 samples from 205 individual patients) [2a] Unclear [2b] Unclear
Comparative
Notes	Funding: Research funding ‐ none declared; Temasek Holdings Pte Ltd sponsored the laboratory testing kits used in this study. Publication status: Published paper Source: JALM‐ Journal of Applied Laboratory Medicine Author COI: None declared, Honoraria: T.C. Aw, Abbott Diagnostics, Roche Diagnostics, Beckman‐Coulter
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	No
Did the study avoid inappropriate exclusions?	Unclear
Did the study avoid inappropriate inclusions?	No
Could the selection of patients have introduced bias?		High risk
Are there concerns that the included patients and setting do not match the review question?			High
DOMAIN 2: Index Test (All tests)
DOMAIN 2: Index Test (Antibody tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Unclear
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Unclear risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Low concern
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	No
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
The reference standard does not incorporate the index test	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		High risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			High
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Unclear
Did all patients receive the same reference standard?	No
Were all patients included in the analysis?	No
Did all participants receive a reference standard?	No
Were results presented per patient?	No
Could the patient flow have introduced bias?		High risk