tetano
Editor, Senior Moderator
BMJ Open
. 2026 Jan 13;16(1):e107660.
doi: 10.1136/bmjopen-2025-107660. Data-driven strategies for model-informed decision-making during the COVID-19 pandemic: a systematic review
Mehdi Lotfi[SUP] 1 [/SUP], Lars Kaderali[SUP] 2 [/SUP]
Affiliations
Objectives: To systematically review data-driven modelling studies that evaluated the effectiveness of interventions implemented during the COVID-19 pandemic and to identify which measures were most frequently reported as effective in controlling disease spread.
Design: Systematic review of modelling studies focused on data-driven, model-informed decision-making for COVID-19 interventions.
Data sources: A comprehensive literature search was conducted in PubMed, Web of Science and Embase, covering publications from 1 January 2020 to 16 October 2024.
Eligibility criteria: Studies were included if they: (1) used real-world data; (2) had sufficient sample sizes and (3) assessed at least one intervention with measurable outcomes.Meta-analyses and purely theoretical modelling studies were excluded. Papers were further filtered using a structured screening process to ensure empirical and intervention-based modelling.
Data extraction and synthesis: Data were extracted from eligible studies and categorised according to modelling approaches, data sources, intervention types and reported effectiveness. Descriptive synthesis was performed to summarise modelling trends and intervention performance. Studies were classified into major intervention categories, including tracing, testing and isolation (TTI); physical and social distancing (PSD); vaccination; lockdowns; mask-wearing; home office or stay-at-home (HOSH) and health infrastructure enhancement (HIE).
Results: Out of 2297 studies identified, 126 met inclusion criteria. Compartmental models were the most frequently used approach, primarily relying on case and death counts to assess intervention impact. The most commonly reported effective interventions were TTI, PSD, vaccination, lockdowns, mask-wearing and HOSH. When considering effectiveness relative to study frequency, the top six interventions were TTI, HOSH, mask-wearing, HIE, PSD and lockdowns. The relatively lower representation of vaccination reflects that most included studies were conducted during the early stages of the pandemic, before widespread vaccine rollout and availability of empirical vaccination data.
Conclusions: This review highlights the critical role of data-driven models in guiding COVID-19 response strategies. Evidence supports the combined effectiveness of non-pharmaceutical interventions, robust testing and tracing systems and health infrastructure strengthening. Real-world impact, however, remains dependent on local healthcare capacity, socioeconomic conditions and cultural contexts. Continued research is essential to refine adaptive modelling approaches and strengthen preparedness for future public health emergencies.
Keywords: COVID-19; Decision Making; Public health.
. 2026 Jan 13;16(1):e107660.
doi: 10.1136/bmjopen-2025-107660. Data-driven strategies for model-informed decision-making during the COVID-19 pandemic: a systematic review
Mehdi Lotfi[SUP] 1 [/SUP], Lars Kaderali[SUP] 2 [/SUP]
Affiliations
- PMID: 41529879
- DOI: 10.1136/bmjopen-2025-107660
Objectives: To systematically review data-driven modelling studies that evaluated the effectiveness of interventions implemented during the COVID-19 pandemic and to identify which measures were most frequently reported as effective in controlling disease spread.
Design: Systematic review of modelling studies focused on data-driven, model-informed decision-making for COVID-19 interventions.
Data sources: A comprehensive literature search was conducted in PubMed, Web of Science and Embase, covering publications from 1 January 2020 to 16 October 2024.
Eligibility criteria: Studies were included if they: (1) used real-world data; (2) had sufficient sample sizes and (3) assessed at least one intervention with measurable outcomes.Meta-analyses and purely theoretical modelling studies were excluded. Papers were further filtered using a structured screening process to ensure empirical and intervention-based modelling.
Data extraction and synthesis: Data were extracted from eligible studies and categorised according to modelling approaches, data sources, intervention types and reported effectiveness. Descriptive synthesis was performed to summarise modelling trends and intervention performance. Studies were classified into major intervention categories, including tracing, testing and isolation (TTI); physical and social distancing (PSD); vaccination; lockdowns; mask-wearing; home office or stay-at-home (HOSH) and health infrastructure enhancement (HIE).
Results: Out of 2297 studies identified, 126 met inclusion criteria. Compartmental models were the most frequently used approach, primarily relying on case and death counts to assess intervention impact. The most commonly reported effective interventions were TTI, PSD, vaccination, lockdowns, mask-wearing and HOSH. When considering effectiveness relative to study frequency, the top six interventions were TTI, HOSH, mask-wearing, HIE, PSD and lockdowns. The relatively lower representation of vaccination reflects that most included studies were conducted during the early stages of the pandemic, before widespread vaccine rollout and availability of empirical vaccination data.
Conclusions: This review highlights the critical role of data-driven models in guiding COVID-19 response strategies. Evidence supports the combined effectiveness of non-pharmaceutical interventions, robust testing and tracing systems and health infrastructure strengthening. Real-world impact, however, remains dependent on local healthcare capacity, socioeconomic conditions and cultural contexts. Continued research is essential to refine adaptive modelling approaches and strengthen preparedness for future public health emergencies.
Keywords: COVID-19; Decision Making; Public health.