The search on pandemics and LHSs yielded 353 results, while the search on climate change and LHSs yielded four results. After duplicate removal, the title and abstract of 182 articles were screened against the inclusion and exclusion criteria. Of these, 117 articles did not meet eligibility criteria, including the four papers identified from the climate change and LHS search. Sixty-five full texts were assessed for eligibility. Thirteen additional papers were identified via backwards citation searching of included full texts and underwent the same screening process. Thirty-two papers were included after the final screening (Fig. 1).

PRISMA flow diagram for study selection process. LHS Learning health system

Of the 32 included publications that described an LHS response to a pandemic, 12 (38%) were case studies, 9 were narrative descriptive articles (28%), 9 were empirical studies (28%), 1 was a protocol (3%), and 1 was a policy-focused publication (3%). Two papers discussed “long COVID” [31, 36] and one paper discussed the COVID-19 pandemic and another pandemic disease (tuberculosis [37]). The other 29 papers (90%) discussed experiences during the COVID-19 pandemic.

The 32 included studies covered nine countries: the USA (n = 18, 56%), Canada (n = 5, 16%), UK (n = 3, 9%), and one paper from each of France, Guinea, Kenya, Nigeria, South Africa, and Spain (3% for each paper). Settings included hospitals and medical centers (n = 7, 22%), primary care (n = 5, 16%), large health networks (n = 11, 34%), such as the US’s Veterans Health Administration, and several community health services (e.g., prevention, vaccination) (n = 7, 22%). Some studies covered multiple settings. Details of all included LHSs are reported in Table 1 and Additional file 1: Table S3.

The SANRA was used to appraise 15 papers and the MMAT was appropriate for 17 papers. The JBI tool was not appropriate for any papers. Most of the papers were appraised as high (n = 15, 47%) or moderate quality (n = 16, 50%), only one paper was of low quality (3%) [39]. This paper was still included because it provided specific data about real-world LHS responses to the COVID-19 pandemic (Additional file 1: Table S4).

Twenty papers included a definition of an LHS [2, 31, 37, 38, 40, 41, 44, 46, 48, 52, 54, 57,58,59,60,61,62,63,64,65] (Table 2). Twenty-seven unique references were used to cite these LHS definitions. The most cited references were those originating from the IoM (cited in N = 9 papers, 28%) [31, 40, 44, 57,58,59,60, 62, 65] or with authorship from Friedman (cited in N = 7 papers, 22%) [41, 43, 52, 54, 58, 63, 65].

Of 17 studies (53%) that reported use of an LHS framework, 11 used existing frameworks that were not explicitly modified from their original structure [37, 38, 40, 41, 43, 46, 48, 51, 56, 57, 62], 4 adapted an existing framework in response to the COVID-19 pandemic [44, 49, 59, 64], and 2 developed new frameworks [52, 54] (Table 2).

Science and Informatics (n = 31, 97%), Continuous Learning Culture (n = 26, 81%), and Structure and Governance (n = 23, 72%) were the most frequently discussed LHS dimensions. Incentives (n = 21, 66%) and Patient-Clinician Partnerships (n = 18, 56%) received less attention (Table 2). Twenty-nine papers (91%) discussed benefits or opportunities arising from the societal and health system conditions created by the COVID-19 pandemic compared to 22 papers (69%) which discussed challenges presented by a pandemic to the development or advancement of an LHS.

Thirty-one articles (97%) described an LHSs response to current or future pandemics that involved the Science and Informatics dimension of the framework (Table 2). Within this dimension, four sub-themes were discerned (Table 3). The most frequently discussed sub-theme was the use of healthcare information systems (n = 26, 84%), including eHRs, machine learning/AI, and clinical prediction/decision making tools [31, 36,37,38, 40,41,42,43,44, 46, 47, 49,50,51,52, 54,55,56,57,58, 60, 61, 63,64,65]. Data-driven research and knowledge translation was reported in 21 articles (68%) [2, 37,38,39,40,41,42, 44,45,46,47,