Pragmatic and open science-based solution to a current problem in the reporting of living systematic reviews

Optimizing process and methods for a living systematic review: 30 search updates and three review updates later

OBJECTIVE

To describe the living systematic review (LSR) process and to share experience of planning, searches, screening, extraction, publishing and dissemination to inform and assist authors planning their own LSR. Many LSR do not publish more than one update, we hope this paper helps to increase this.

STUDY DESIGN AND SETTING

A Cochrane LSR with an international author team that has been 'living' for two years, with monthly search updates and three full updates published in this time. LSRs are regularly updated systematic reviews that allow new evidence to be incorporated as it becomes available. LSR are ideally suited to policy-relevant topics where there is uncertainty and new evidence will likely impact the interpretation and/or certainty of outcomes.

RESULTS

The key features of the process that require consideration are: specifying the frequency of searches and triggers for full updates in the protocol; stakeholder input; publishing and disseminating monthly search findings. A strong team, incorporating methodological and topic expertise, with core members that meet regularly is essential. Regular search updates make it important to have a clear cyclical schedule of activity. To achieve timely updates this process should be streamlined, for example, using automated monthly searches, and systematic reviewing software for screening. LSR provide a unique opportunity to incorporate stakeholder feedback.

CONCLUSIONS

We recommend that LSRs should be: justified; carefully planned including the timing of search updates, triggers for publication and termination; published in a timely manner; have a clear dissemination plan; and a strong core team of authors.

Nirmatrelvir combined with ritonavir for preventing and treating COVID-19

BACKGROUND

Oral nirmatrelvir/ritonavir (Paxlovid) aims to avoid severe COVID-19 in asymptomatic people or those with mild symptoms, thereby decreasing hospitalization and death. It remains to be evaluated for which indications and patient populations the drug is suitable.

OBJECTIVES

To assess the efficacy and safety of nirmatrelvir/ritonavir plus standard of care (SoC) compared to SoC with or without placebo, or any other intervention for treating COVID-19 or preventing SARS-CoV-2 infection. To explore equity aspects in subgroup analyses. To keep up to date with the evolving evidence base using a living systematic review (LSR) approach and make new relevant studies available to readers in-between publication of review updates.

SEARCH METHODS

We searched the Cochrane COVID-19 Study Register, Scopus, and World Health Organization COVID-19 Research Database, identifying completed and ongoing studies without language restrictions and incorporating studies up to 15 May 2023. This is a LSR. We conduct update searches every two months and make them publicly available on the open science framework (OSF) platform.

SELECTION CRITERIA

We included randomized controlled trials (RCTs) comparing nirmatrelvir/ritonavir plus SoC to SoC with or without placebo, or any other intervention for treatment of people with confirmed COVID-19 diagnosis, irrespective of disease severity or treatment setting, and for prevention of SARS-CoV-2 infection. We screened all studies for research integrity. Studies were ineligible if they had been retracted, or if they were not prospectively registered including appropriate ethics approval.

DATA COLLECTION AND ANALYSIS

We followed standard Cochrane methodology and used the Cochrane RoB 2 tool. We rated the certainty of evidence using the GRADE approach for the following outcomes: 1. to treat outpatients with mild COVID-19; 2. to treat inpatients with moderate to severe COVID-19: mortality, clinical worsening or improvement, quality of life, (serious) adverse events, and viral clearance; 3. to prevent SARS-CoV-2 infection in postexposure prophylaxis (PEP); and 4. pre-exposure prophylaxis (PrEP) scenarios: SARS-CoV-2 infection, development of COVID-19 symptoms, mortality, admission to hospital, quality of life, and (serious) adverse events. We explored inequity by subgroup analysis for elderly people, socially-disadvantaged people with comorbidities, populations from low-income countries and low- to middle-income countries, and people from different ethnic and racial backgrounds.

MAIN RESULTS

As of 15 May 2023, we included two RCTs with 2510 participants with mild and mild to moderate symptomatic COVID-19 in outpatient and inpatient settings comparing nirmatrelvir/ritonavir plus SoC to SoC with or without placebo. All trial participants were without previous confirmed SARS-CoV-2 infection and at high risk for progression to severe disease. Randomization coincided with the Delta wave for outpatients and Omicron wave for inpatients. Outpatient trial participants and 73% of inpatients were unvaccinated. Symptom onset in outpatients was no more than five days before randomisation and prior or concomitant therapies including medications highly dependent on CYP3A4 were not allowed. We excluded two studies due to concerns with research integrity. We identified 13 ongoing studies. Three studies are currently awaiting classification. Nirmatrelvir/ritonavir for treating people with asymptomatic or mild COVID-19 in outpatient settings Nirmatrelvir/ritonavir plus SoC compared to SoC plus placebo may reduce all-cause mortality at 28 days (risk ratio (RR) 0.04, 95% confidence interval (CI) 0.00 to 0.68; 1 study, 2224 participants; low-certainty evidence) and admission to hospital or death within 28 days (RR 0.13, 95% CI 0.07 to 0.27; 1 study, 2224 participants; low-certainty evidence). Nirmatrelvir/ritonavir plus SoC may reduce serious adverse events during the study period compared to SoC plus placebo (RR 0.24, 95% CI 0.15 to 0.41; 1 study, 2224 participants; low-certainty evidence). Nirmatrelvir/ritonavir plus SoC probably has little or no effect on treatment-emergent adverse events (RR 0.95, 95% CI 0.82 to 1.10; 1 study, 2224 participants; moderate-certainty evidence), and probably increases treatment-related adverse events such as dysgeusia and diarrhoea during the study period compared to SoC plus placebo (RR 2.06, 95% CI 1.44 to 2.95; 1 study, 2224 participants; moderate-certainty evidence). Nirmatrelvir/ritonavir plus SoC probably decreases discontinuation of study drug due to adverse events compared to SoC plus placebo (RR 0.49, 95% CI 0.30 to 0.80; 1 study, 2224 participants; moderate-certainty evidence). No studies reported improvement of clinical status, quality of life, or viral clearance. Nirmatrelvir/ritonavir for treating people with moderate to severe COVID-19 in inpatient settings We are uncertain whether nirmatrelvir/ritonavir plus SoC compared to SoC reduces all-cause mortality at 28 days (RR 0.63, 95% CI 0.21 to 1.86; 1 study, 264 participants; very low-certainty evidence), or increases viral clearance at seven days (RR 1.06, 95% CI 0.71 to 1.58; 1 study, 264 participants; very low-certainty evidence) and 14 days (RR 1.05, 95% CI 0.92 to 1.20; 1 study, 264 participants; very low-certainty evidence). No studies reported improvement or worsening of clinical status and quality of life. We did not include data for safety outcomes due to insufficient and inconsistent information. Subgroup analyses for equity For outpatients, the outcome 'admission to hospital or death' was investigated for equity regarding age (less than 65 years versus 65 years or greater) and ethnicity. There were no subgroup differences for age or ethnicity. For inpatients, the outcome 'all-cause mortality' was investigated for equity regarding age (65 years or less versus greater than 65 years). There was no difference between subgroups of age. No further equity-related subgroups were reported, and no subgroups were reported for other outcomes. Nirmatrelvir/ritonavir for preventing SARS-CoV-2 infection (PrEP and PEP) No studies available.

AUTHORS' CONCLUSIONS

Low-certainty evidence suggests nirmatrelvir/ritonavir reduces the risk of all-cause mortality and hospital admission or death in high-risk, unvaccinated COVID-19 outpatients infected with the Delta variant of SARS-CoV-2. There is low- to moderate-certainty evidence of the safety of nirmatrelvir/ritonavir. Very low-certainty evidence exists regarding the effects of nirmatrelvir/ritonavir on all-cause mortality and viral clearance in mildly to moderately affected, mostly unvaccinated COVID-19 inpatients infected with the Omicron variant of SARS-CoV-2. Insufficient and inconsistent information prevents the assessment of safety outcomes. No reliable differences in effect size and direction were found regarding equity aspects. There is no available evidence supporting the use of nirmatrelvir/ritonavir for preventing SARS-CoV-2 infection. We are continually updating our search and making search results available on the OSF platform.

Paper 4: Search methods and approaches for living guidelines

OBJECTIVES

To describe the key features of a continual evidence surveillance process that can be implemented for living guidelines and to outline the considerations and trade-offs in adopting different approaches.

STUDY DESIGN AND SETTING

Members of the Australian Living Evidence Consortium (ALEC), National Institute of Health and Care Excellence (NICE), and the US GRADE Network (USGN) shared their practical experiences of and approaches to establishing surveillance systems for living guidelines. We identified several common components of evidence surveillance and listed the key features and considerations for each component drawn from case studies, highlighting differences with standard guidelines.

RESULTS

We developed guidance that covers the initial information needed to support decisions around suitability for living mode and the practical considerations in setting up continual search surveillance systems (search frequency, sources to search, use of automation, reporting the search, ongoing resources, and evaluation). The case studies draw on our experiences with developing guidelines for COVID-19, as well as for other conditions such as stroke and diabetes, and cover a range of practical approaches, including the use of automation.

CONCLUSION

This paper highlights different approaches to continual evidence surveillance that can be implemented in living guidelines.