Table 3.
Summary of evidence for preventive interventions for post Covid-19 condition, by timing of implementation and follow-up duration.
| Author, country; study design | Intervention/exposure and comparator with sample sizes | Acute phase treatment; Covid-19 illness severity; ≥90% Laboratory confirmed (Y/N) |
Outcomes and results | Conclusions (certainty*) Apply to each outcome seperately |
|---|---|---|---|---|
| Acute-phase interventions with follow-up 12–16 weeks post-Covid 19 | ||||
|
Anastasio 2021, Italy Retrospective controlled cohort |
Steroid use(n = 42) vs. no steroid use (n = 180) |
Hospitalized 100% pneumonia (in analysis) Y |
Dyspnea (MMRC scale): positively correlated p = .05 |
Very uncertain about effects (Very lowa,c) |
|
Qin 2021, China Retrospective controlled cohort |
Corticosteroid use(n = 17) vs. no corticosteroid use (n = 64) |
Hospitalized Moderate-to-critical Y |
Dyspnea: DLCO <80% predicted: IG: 10 (59%) vs. CG: 34 (53%); OR: 1.3 (0.4–3.7), p = .68 | |
|
Xiong 2021, China Retrospective controlled cohort |
Corticosteroid use(n = 138) vs. no corticosteroid use (n = 400) |
Hospitalized Moderate-to-critical Y |
Fatigue(questionnaire item): IG: 39 (28%) vs. CG: 113 (28%) Dyspnea(post-activity polypnoea item): IG: 34 (25%) vs. CG: 81 (20%); RR 1.29 (0.81–2.03) |
|
|
Zhao 2020, China Retrospective controlled cohort |
Low-dose corticosteroid use(n = 7) vs. no corticosteroid use (n = 48) |
Hospitalized Moderate-to-severe Y |
Dyspnea: DLCO <80% predicted: IG: 1 (14.3%) vs. CG: 8 (16.7%); RR 0.83 (0.09–7.90) | |
|
Gherlone 2021, Italy Retrospective controlled cohort |
Antibiotic use(n = 102) vs. no antibiotic use (n = 20) |
Hospitalized Moderate-to-critical (25% ICU) Y |
Adverse events: association between antibiotic use and salivary gland ectasia: aOR 8.34 (95% CI, 1.47–158.19); p = .049 | Very uncertain about effects (Very lowa,b,c,D) |
|
Kataria 2021, India Nonrandomized experimental study (nonconcurrent groups) |
Ayurvedic formulationTinospora cordifolia(Guduchi) and Piper longum (Pippali) twice daily (n = 30) vs. standard care only (n = 30) |
Hospitalized Asymptomatic (32%) to moderate (3%) Y |
Quality of Life(general health much better now): IG 18 (64.3%) vs. 15 (51.7%): RR, 1.24 [0.79, 1.94] Dyspnea(breathlessness during physical activities): IG 6 (21.4%) vs. CG 3 (10.3%); RR, 2.00 [0.55, 7.22] Functional incapacity(work efficiency limited a lot since discharge): IG 0 vs. CG 0 Fatigue: 1 (5%) vs. CG 6 (26.1%); RR, 0.17 [0.02, 1.30] Sleep disturbances: IG 3 (10.7%) vs. CG 1 (3.4%); RR, 3.00 [0.33, 27.12] |
Very uncertain about effects (Very lowa,b,c,d) |
|
Pain Frequent headaches: IG 3 (10.7%) vs. CG 3 (10.3%); RR, 1.00 [0.22, 4.54] Chest pain: IG 1 (5%) vs. CG 4 (17.4%); RR, 0.25 [0.03, 2.10] Dyspnea(oxygen support at home): IG 0 vs. CG 0 |
Very uncertain about effects (Very lowa,c,d) |
|||
|
Feng 2021, China Nonrandomized experimental study |
Human umbilical cord mesenchymal stem cells; 1 intravenous delivery (n = 12) vs. standard medical care only (n = 29) |
Hospitalized Severe Y |
Quality of Life: IG 15.3 ± 3.7 vs. CG 31.9 ± 8.8; MD, −16.60 [−21.23, −11.97] Dyspnea Wheezing: IG 3 (37.5%) vs. 15 (75%); RR, 0.50 [0.20, 1.27] Lung function: FEV1/ FVC ratio <70% IG 1 (12.5%) vs. CG 13 (65%); RR, 0.19 [0.03, 1.24] Adverse events: 0 Serious adverse events: 0 |
Very uncertain about effects (Very lowa,c,D) |
| Fatigue: IG 4 (50%) vs. CG 14 (70%); RR, 0.71 [0.34, 1.51] | Very uncertain about effects (Very lowa,b,c,D) | |||
|
Jain 2021, U.S. Prospective uncontrolled cohort |
Regionalin-patient rehabilitation following acute hospitalization |
Hospitalized Severe (% ICU NR; 94% required oxygen) NR |
Hospital re-admissions(within last 2–3 weeks of follow-up): 0 | Very uncertain about effects (Very lowa,c,D) |
| Acute-phase interventions with follow-up >16 weeks post-Covid 19 | ||||
|
Blomberg 2021, Norway Retrospective controlled cohort |
Antibiotic use(n = 31) vs. no antibiotic use (n = 262) |
Hospitalized or home isolated In hospitalized: 9% on respirator Y |
Fatigue(Chalder fatigue scale (score ≥4)): IG 17 (55%) vs. CG 91 (35%); aOR 0.42 (0.10–1.75), p > .05 |
Very uncertain about effects (Very lowa,c) |
|
Frontera, 2021, US Retrospective controlled cohort |
Azithromycin use(n = 246) vs. no azithromycin use (n = 136) |
Hospitalized 8.3% ICU Y |
All p > .10, unadjusted Functional incapacity: (mRS 3–6): OR 0.398 Functional incapacity(Barthel Index <100): OR 0.078 Cognitive impairment(Telephone Montreal Cognitive Assessment [score < 18]): OR 0.188 Anxiety(subscale of Neuro-QoL [T-score > 50]): OR 0.606 Depression(subscale of Neuro-QoL [T-score > 50]): OR 0.104 Fatigue (subscale of Neuro-QoL [T-score > 50]): OR 0.633 Sleep disturbances (subscale of Neuro-QoL [T-score > 50]): OR 0.819 Return to work: OR 0.458 |
|
|
Frontera, 2021, US Retrospective controlled cohort |
Corticosteroid use(n = 101) vs. no corticosteroids (n = 281) |
Hospitalized 32% ICU Y |
Unless stated, all p > .10 unadjusted; all non-significant with multivariate (data not shown) Functional incapacity: (mRS 3–6): OR 0.13 Functional incapacity(Barthel Index <100): OR 0.480 (0.29–0.80), p = .005 Cognitive impairment(Telephone Montreal Cognitive Assessment [score < 18]): OR 0.98 Anxiety(subscale of Neuro-QoL [T-score > 50]): OR 0.82 Depression(subscale of Neuro-QoL [T-score > 50]): OR 0.37 Fatigue (subscale of Neuro-QoL [T-score > 50]): OR 2.2 (1.3–3.9), p = .004 Sleep disturbances (subscale of Neuro-QoL [T-score > 50]): OR 1.7 (1.0–2.8), p = .064 Return to work (self-report, Yes/No): OR 0.4 (0.2–0.8), p = .008 |
Very uncertain about effects (Very lowa,c) |
|
Huang 2021, China Retrospective controlled cohort |
Corticosteroid use(n = 398) vs. no corticosteroid use (n = 1335) |
Hospitalized 4% ICU Y |
Dyspnea: (Lung function): aOR 1.18 (0.60–2.34), p = .63 | |
|
Anxiety or depression: aOR 1.23 (0.88–1.72), p = .22 Fatigue or muscle weakness: aOR 1.04 (0.77–1.42), p = .78 | ||||
|
Frontera 2021, US Retrospective controlled cohort |
Hydroxychloroquine use(n = 272) vs. no hydroxychloroquine use (n = 110) Therapeutic anticoagulation use(n = 134) vs. no therapeutic anticoagulation use (n = 248) Zinc use(n = NR) vs. no zinc use (n = NR) |
Hospitalized 32.2% ICU, 51.3% neurological COVID-19 disorder Y |
Functional incapacity: (mRS 3–6): Hydroxychloroquine: OR 0.98 Therapeutic anticoagulation: OR 0.18 Zinc: OR 0.67 (0.44–1.03), p = .066 Functional capacity(Barthel Index <100): Hydroxychloroquine: OR 0.97 Therapeutic anticoagulation: OR 0.078 Zinc: OR 0.41 Cognitive impairment(Telephone Montreal Cognitive Assessment [score < 18]): Hydroxychloroquine: OR 0.13 Therapeutic anticoagulation: OR 0.99 Zinc: OR 0.52 Anxiety(subscale of Neuro-QoL [T-score > 50]): Hydroxychloroquine: OR 0.95 Therapeutic anticoagulation: OR 0.91 Zinc: OR 0.91 Depression(subscale of Neuro-QoL [T-score > 50]): Hydroxychloroquine: OR 0.50 Therapeutic anticoagulation: OR 0.98 Zinc: OR 0.62 Fatigue (subscale of Neuro-QoL [T-score > 50]): Hydroxychloroquine: OR 0.22 Therapeutic anticoagulation: OR 1.8 (1.1–3.1), p = .022 Zinc: OR 0.27 Sleep disturbances (subscale of Neuro-QoL [T-score > 50]): Hydroxychloroquine: OR 0.52 Therapeutic anticoagulation: OR 1.7 (1.0–2.8), p = .051 Zinc: OR 0.17 Return to work (self-report, Yes/No): Hydroxychloroquine: OR 0.19 Therapeutic anticoagulation: OR 0.31 (0.16–0.60), p = .001 Zinc: OR 2.3 (1.2–4.4), p = .016 |
Very uncertain about effects (Very lowa,c) |
|
Huang 2021, China Retrospective controlled cohort |
Intravenous immunoglobulin use(n = 345) vs. no intravenous immunoglobulin use (n = 1388) |
Hospitalized 4% ICU Y |
Dyspnea(Lung function): aOR 0.94 (0.49–1.79), p = .85 |
Very uncertain about effects (Very lowa,c,d) |
|
Anxiety or depression: aOR 0.77 (0.54–1.10), p = .15 Fatigue or muscle weakness: aOR 0.96 (0.70–1.31), p = .78 |
Very uncertain about effects (Very lowa,b) |
|||
|
Vetrici 2021, US Pilot RCT |
Photobiomodulation adjunctive anti-inflammatory treatment; daily × 4 days (n = 5) vs. standard (supportive) care only (n = 5) |
Hospitalized Moderate/severe (0% IV at baseline) Y |
Dyspnea(any respiratory symptoms): 0/5 (0%) vs. 4/5 (80%); RR, 0.11 [0.01, 1.64] Treatment-related adverse events: 0/5 vs. NA |
Very uncertain about effects (Very lowa,b,c,D) |
|
Wu 2021, China Prospective controlled cohort |
Oseltamivir use(n = 53) vs. no oseltamivir use (n = 30) Ganciclovir use(n = 42) vs. no ganciclovir use (n = 41) |
Hospitalized Severe illness (0% IV) Y |
Dyspnea: DLCO <80% predicted association with Oseltamivir: unadjusted OR 0.75 (95% CI 0.29–1.92) DLCO <80% predicted association with Ganciclovir: unadjusted OR 1.34 (95% CI 0.53–3.38) |
Very uncertain about effects (Very lowa,c,d) |
| Early post-acute interventions with follow-up 12–16 weeks post-Covid 19 | ||||
|
Amini 2021, Iran Prospective uncontrolled cohort |
Cognitive-motor-therapyfor 4 weeks; performed twice per week; measured at baseline (n = 42) and follow-up (n = 42) |
Hospitalized Non-severe (COVID-19 symptoms stage 1) NR |
Quality of Life: IG (baseline) 47.8 ± 1.6 vs. IG (3 mo) 46.9 ± 1.2; MD 0.86 (p = .006) (<22 pathological symptoms) Cognitive impairment: IG (baseline) 17.9 ± 3.3 vs. IG (3 mo) 19.7 ± 2.2; MD −1.72 (p = .001) (<23 threshold for impairment) *Several subscales of the MMSE were found significant at 3-month follow-up (attention and calculation, recall & action performance, MD range −0.79 to −0.33), all other domains were found to be non-significant (orientation, information encoding, & lingual skills, MD range −0.15 to −0.03). Depressive symptoms: IG (baseline) 8.6 ± 1.7 vs. IG (3 mo) 8.2 ± 1.17(<6 pathological symptoms) Anxiety symptoms: IG (baseline) 13.7 ± 2.448 vs. IG (3 mo) 12.622 ± 2.292; MD 1.155 (p = .001) (<6 = pathological symptoms) Functional incapacity(each <6 pathological symptoms) Physical symptoms: IG (baseline)12.66 ± 1.34 vs. IG (3 mo) 11.63 ± 2.31; MD 1.028 (p = .001) Social performance: IG (baseline) 15.05 ± 2.46 vs. IG (3 mo) 14.68 ± 2.41; MD 0.37 (p = .001) |
Very uncertain about effects (Very lowa,c,d) |
|
Benzakour 2021; Switzerland Prospective uncontrolled cohort (pre-post) |
Screening and treatment for psychiatric symptoms (n = 109; 64 follow-up) |
Hospitalized 16.5% ICU (in n = 64 with CoviCare follow-up data) NR |
PTSD: T0: 15 (14.6%) vs. T1: 7 (10.6%) Depressive symptoms: T0: 20 (18.5%) vs. T1; 6 (10.0%) Anxiety symptoms: T0: 17 (15.7%) vs. T1 6 (10.0%) |
Very uncertain about effects (Very lowa,c,d) |
|
Li 2021, China Retrospective controlled cohort |
Chinese medicinefor 28 days after discharge; 2 different oral formula twice daily, depending on type of syndrome (pathogen residue or qi and yin deficiency) (n = 64) vs. Western medicine or no medicines (n = 32) |
Hospitalized All severities (22% severe or critical) Y |
Fatigue Dyspnea Insomnia Pain: Chest tightness No significant difference in the improvement rates of symptoms, including fatigue, between the two groups (p > .05) |
Very uncertain about effects (Very lowa,b,c,d) |
|
Martin 2021, Belgium Prospective controlled cohort |
Telerehabilitation programmevideoconferencing; supervised by a physiotherapist; home–based individual and group 50-min endurance and strength exercises 2/week for 6 week with 2–3/week unsupervised (n = 14) vs. patients refusing programme (n = 13) |
Hospitalized Severe or critical illness (22% ICU) |
Dyspnea: Sit-to-stand (< 50th %ile): IG pre 14/14 (100%) vs. post 13/14 (93%) vs. CG pre 13/13 (100%) vs. post 13/13 (100%); RR, 0.93 [0.77, 1.13] Change since baseline in dyspnea after sit-to-stand: IG 2.5 (−2–7) vs. CG 2.0 (0–6); p = .56 Adverse events: none reported |
Very uncertain about effects (Very lowa,c,D) |
| Early post-acute interventions with follow-up >16 weeks post-Covid 19 | ||||
|
Meije 2021, Spain Prospective uncontrolled cohort |
Outpatient assessment with referrals(standard follow-up protocol checklist of symptoms and adverse events, medical history, physical examination, laboratory testing including chest x-ray) medical follow-up as required (50.3% patients; 27.1% to pulmonologist) (n = 294) |
Hospitalized Variable severity (1% IV; ICU 8.9%) N |
At post discharge assessment vs. 7 mo follow-up: Proportion with persistent symptoms: 228 (77.6%) vs. 147 (50%) Pain: Migraine 19 (6.5%) vs. 12 (4.1%) Chest pain 30 (10.2%) vs. 8 (2.7%) Psychopathology: Need for psychological medication 35 (11.9%) vs. 30 (10.2%) Fear of relapse 68 (23.1%) vs. 86 (29.3%) |
Very uncertain about effects (Very lowa,c) |
|
Dyspnea: 88 (29.9%) vs. 28 (9.5%) Insomnia: 62 (21.1%) vs. 54 (18.4%) Functional disability: 60 (20.4%) vs. 57 (19.4%) |
Very uncertain about effects (Very lowa,b,c) |
|||
Abbreviations: CG: control group; DLCO: diffusing capacity of the lungs for carbon monoxide; ICU: intensive care unit; IG: intervention group; IV: invasive ventilation; MD: mean difference; MMRC: modified Medical Research Council dyspnea scale; MMSE: mini mental state examination; mRC: modified Rankin Scale; NR: not reported; OR: odds ratio; QoL: quality of life; RR: relative risk; vs.: versus.
*Assessed using GRADE. Certainty was rated down by 0, 1, or 2 levels for risk of bias (a), indirectness in outcome (b), inconsistency/lack of consistency (c), and imprecision/ wide confidence intervals (d); capital letters indicates very serious concern. The certainty rating applies similarly to each outcome within each row.