This study was reviewed by the University of Utah Institutional Review Board (IRB) and was determined not to meet the definitions of Human Subjects Research according to Federal regulations (IRB_00142577). Therefore, the study did not require IRB oversight.

Data analysis was performed using Stata/IC (version 16.1; StataCorp). Daily test results were provided in the Trend database. Days when no GI Panel tests were performed in a state were excluded from the analysis. The daily test results were summed to weekly test results, and subsequently, weekly positivity percentages for each state and target were analyzed.

An interrupted time series analysis (ITSA) was performed for each of the 3 pathogens of interest and each state, using a downloadable Stata package (Figures 1-6) [22]. Figures 1-6 show the percent positivity before social distancing mandates went into effect (solid black dots before the vertical dashed line) and after social distancing mandates went into effect (solid black dots after the vertical dashed line). The best fit positivity percentage is represented by the solid horizontal line. We were most interested in the difference between the best fit positivity percentage before social distancing mandates and immediately after social distancing mandates. The time of intervention for the ITSA was the week that social distancing mandates went into effect for each individual state. In most states, multiple mandates went into effect in the same week; thus, the ITSA was performed only once for mandates occurring in the same week. Mandates occurring in separate weeks required a separate ITSA.

A total of 84,223 tests from 9 states were included in the final data set (Table 1). Norovirus GI and GII had a higher overall positivity percentage compared to adenovirus F40 and 41 and E coli O157.

All states included in the analyses mandated social distancing policies in March of 2020 between the weeks beginning March 15, 2020, and March 29, 2020 (Table 2). Social distancing mandates occurred across two separate weeks for most states, with the exception of California and Illinois, where all mandates were announced in a single week. According to the Kaiser Family Foundation, Nebraska was the only state in the analysis that did not mandate nonessential business closures or a stay-at-home order.

With the exception of E coli O157 in Kansas, Michigan, and Nebraska, we observed an immediate decrease in positivity percentage during the week of social distancing mandates. Norovirus GI and GII showed the most notable drop in positivity, whereas E coli O157 appeared to be least impacted by social distancing mandates (Figure 7; Table 3). Although we acknowledge the analysis has a multiple testing problem, the majority of our significant results showed significance even below the .01 level (Table 3).

The results from this study indicate that public health measures meant for COVID-19 can initially decrease GI pathogen positivity. In most instances, we observed an immediate decrease in positivity percentage, suggesting that social distancing measures can very quickly decrease GI pathogen transmission. Similar results have been observed in previous studies, demonstrating that these public health measures can decrease transmission of other pathogens transmitted through the respiratory route, such as influenza, respiratory syncytial virus, and respiratory enterovirus; this study additionally demonstrates that they are also effective against GI pathogens on a national scale [8-13].

To provide context for our results on the effects of the mode of transmission of the pathogens, we chose 3 pathogens with different routes of transmission. Interestingly, our results show norovirus GI and GII was most impacted by the social distancing mandates. This may be due to the contagious nature of the pathogen, suggesting that social distancing mandates may be most effective against highly contagious GI pathogens that are most commonly spread person to person. The reduced person-to-person contact resulting from social distancing mandates could have decreased the transmission of norovirus GI and GII. Conversely, social distancing mandates showed a minimal impact on E coli O157 positivity rates. The results of E coli O157 positivity may be due to the incubation period of the bacterial infection, typically longer than viral infections, as we did not account for lag time in our analysis; however, this could be investigated in future studies. Additionally, as restaurants closed to dine-in service, take-out service typically remained available, which could have contributed to the minimal impact observed in our results. The social distancing mandates exhibited a moderate effect on the positivity percentage of adenovirus F40 and 41, more so than on E coli O157, but not as extreme as on norovirus GI and GII. It is possible that limited person-to-person contact impacted one route of transmission for adenovirus F40 and 41, but the virus was still spreading through other routes of transmission (eg, a fecal-oral route) leading to the observed moderate impact.

Our results show variability in the change of positivity percentage between states. This study did not aim to analyze this variability; however, differences in health behaviors of state residents and the enforcement of social distancing mandates may be contributing causes of this variability. An area of focus for future studies could be investigating differences between states.

The most notable strength of this study is the database itself, Trend. This study is a novel use of this unique diagnostic database that allowed us access to a large sample size of routinely collected, deidentified data. The large sample size consisting of test results across the nation should allow for generalizability to many communities. During the COVID-19 pandemic, social distancing mandates were universally used as a public health measure to control transmission, permitting us to use them as a variable in the analysis and compare states. The ITSA analysis allowed us to investigate the immediate impact of COVID-19–related social distancing measures on the positivity percentage of GI pathogens.

Although this study presents interesting findings using a unique diagnostic database, it is not without limitations. The BioFire Trend database is expansive, and not all laboratories in a state may be participating, introducing the possibility of selection bias. However, the data set represents data from 9 states in different regions across the United States, allowing for some generalizability. There is possibility for diagnostic bias if a clinician chooses not to use the GI Panel, but most provider institutions will have testing algorithms established, that likely include the GI Panel if the patient is showing acute GI illness symptoms. Additionally, key social distancing measures were only measured on a statewide basis, which does not account for differences between counties or cities, nor did we have available data about compliance with social distancing mandates. However, given the time frame of our study, compliance is likely high, since our analysis period covers the time when the only measures available to prevent COVID-19 were social distancing measures. Hand hygiene was not investigated in this study, and it may have impacted the transmission of GI pathogens during the COVID-19 pandemic; future studies should investigate the effect of hand hygiene. BioFire Trend does not collect demographic data, and further studies are needed to investigate the effects of gender, age, race or ethnicity, and other demographic variables on GI pathogen positivity percentages.

We investigated the impact of social distancing mandates for COVID-19 on GI pathogen positivity, and we discovered that social distancing measures did, in fact, decrease GI pathogen positivity. Our results show the possible utility of social distancing measures to reduce the spread of GI pathogens. The use of similar measures may prove useful in GI pathogen outbreaks. In addition to anecdotal evidence of decreased illness during the COVID-19 pandemic, the findings from this study can be used to reinforce that social distancing interventions can be used to reduce GI pathogen transmission.

The use of a unique diagnostic database, Trend, exhibits the potential for its use as a public health surveillance tool. We have only demonstrated one use of this routinely collected data, but we imagine it could be used for algorithms, models, and tools for early detection of diseases and monitoring the impact of different interventions to control outbreaks. Further research should not only investigate additional applications of Trend but also the impact of different public health measures between different communities. Further studies could assess whether GI pathogen positivity also decreased on a global scale during the COVID-19 time frame, further investigate the impact of COVID-19 mandates on pathogen positivity with the addition of contextual information, and observe long-term GI pathogen positivity after social distancing mandates went into effect.