Decline in in � uenza cases in Mexico after the implementation of public health measures for COVID-19

Arellanos-Soto Daniel Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0002-2707-9754 Padilla-Rivas Gerardo Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0003-3243-8843 Ramos-Jimenez Javier Department of Internal Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0002-4701-7972 Galan-Huerta Kame Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0002-3495-0528 Lozano-Sepulveda Sonia Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0002-6690-7428 Martinez-Acuña Natalia Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0001-7147-9635 Treviño-Garza Consuelo Department of Pediatrics, Hospital Universitario “Dr. Jose E. Gonzalez”, Autonomous University of Nuevo Leon, Secretariat of Health of Nuevo Leon state https://orcid.org/0000-0002-6544-3780 De la O-Cavazos Manuel Department of Pediatrics, Hospital Universitario “Dr. Jose E. Gonzalez”, Autonomous University of Nuevo Leon, Secretariat of Health of Nuevo Leon state https://orcid.org/0000-0002-8123-1294 Montes de Oca-Luna Roberto Department of Histology, Hospital Universitario “Dr. Jose E. Gonzalez”, Autonomous University of Nuevo Leon, Secretariat of Health of Nuevo Leon state https://orcid.org/0000-0001-8253-680X Rivas-Estilla Ana Maria (  amrivas1@yahoo.ca ) Department of Biochemistry and Molecular Medicine. School of Medicine, Autonomous University of Nuevo Leon https://orcid.org/0000-0003-2311-2534


Introduction
Public health measures, including public education and physical distancing, were implemented in Mexico to reduce transmission of COVID-19 after the rst few cases were reported in epidemiological week 3 (January 13, 2020) and public awareness was increased. 1 We examined the effect of these COVID-19 measures on in uenza incidence as a proxy for determining the overall potential reduction in respiratory virus transmission.
Mexico, like other Latin American countries, established population con nement and a reduction in its economic activity to face the coronavirus pandemic and the respiratory disease it generates, COVID-19. The Mexican Ministry of Health established a package of prevention measures called "National Protocol for Healthy Distance", effective as of Monday, March 23, 2020, to prevent the spread of COVID-19 in the community. 2 These control strategies focused on social distancing and basic prevention measures and were aligned with those used for other common respiratory viral infections such as in uenza: use of face masks, avoiding contact when greeting, hand washing, cough etiquette, temperature monitoring, home con nement, and avoiding close contact with sick individuals. In addition to this, cancellation of large-scale events, the lockdown of activities at all levels of education and workplaces (e.g., segregated teams and home-o ce working wherever possible) and non-essential activities of the public, social, and private sectors were temporarily suspended, and the reprogramming of massive events was raised. Also, intensive public health education on public and personal hygiene and social responsibility was established. 3 We compared in uenza activity indicators in 2020 before and after public health measures were taken to reduce COVID-19 with the corresponding indicators from the three preceding years and the immediately previous one, as well as the involvement in the decrease of seasonal in uenza cases.

Methods
We obtained routine sentinel surveillance data on in uenza-like illnesses (ILI) from the national network of the National Health Secretariat and the National Centre of epidemiological surveillance for all weeks since 2016. 4 ILI were de ned as fever (>38°C) and cough. ILI samples were tested and registered per week, and the percentage of in uenza positivity was determined.
We evaluated whether the in uenza trends in the 2019-2020 season were different before and after the COVID-19 National Protocol for Healthy Distance was implemented compared to previous seasons by using a quasi-experimental difference-in-difference design. We extracted weekly reports of seasonal in uenza data from the 2016-17 season to the 2019-20 season from epidemiological (EPI) week 40 to EPI week 30, using the open-access databases of the public health authority. 4 We compared the number of in uenza cases in season 2019-2020 and the average of the corresponding periods in the previous three in uenza seasons (2016-2019). T-tests were performed for each EPI week using IBM SPSS Statistics for Windows, Version 23.0. 5

Results
In Mexico, in uenza incidence follows a yearly bimodal pattern: the in uenza season and the inter-seasonal period. The in uenza season spans from October to March, including autumn and winter (two years, weeks 40 to 20).
We compared indicators of in uenza transmissibility (con rmed cases and deaths) in 2020 against the average from corresponding periods in the three preceding in uenza seasons (2016-2020). In Mexico, the number of seasonal in uenza cases in the 2019-20 season was lower after COVID-19 transmission compared to previous years ( Fig. 1 and 2A). We observed the peak of in uenza cases in week 4. Particularly, the average of in uenza cases at week 4 was signi cantly different (P <0.05) from the previous two in uenza seasons (587 vs 357.33 ± 99.6; respectively).
The number of in uenza cases decreased below the average of previous years in EPI week 12 (2017-2019). In addition, we observed that after EPI week 12, in Mexico, 9 weeks had passed since the detection of the rst SARS-CoV-2 case and the number of COVID-19 cases was increasing ( Fig. 2A). The healthy distance protocol was o cially established in EPI week 13. It was observed that simultaneously the cases of in uenza were decreasing and were unusually low after week 20 compared to previous years. Even when measures to contain the transmission of the new respiratory disease were established (EPI week 13), the number of COVID-19 cases increased while prevalence of in uenza declined ( Fig. 2A). In fact, the number of in uenza cases in 2020 decreased signi cantly (P <0.05) from EPI week 22 to 30 compared to the seasons 2016-2019 (Fig. 2B). It is important to highlight that this phenomenon indicates a signi cant difference in the transmission mechanisms and the pathogenesis of these two diseases.
We performed a weekly paired difference t-test using IBM SPSS Statistics for Windows, (Version 23.0). 5 We found that the percentage of in uenza cases decreased by 64% (p=0.001) and the estimated daily number of in uenza cases decreased by 76% (p=0.002) in EPI week 20 of season 2019-2020 compared with the preceding years (Fig. 2B). Additionally, before EPI week 20, the number of deaths in 2020 from in uenza per EPI week was statistically similar (P> 0.05) to the average of the seasons 2016-2019 ( Fig. 1 and 2C). In contrast, we saw no signi cant changes in deaths, only in the percentage of in uenza positivity (31%; p = 0.008), in EPI weeks 21 to 30 of 2019-2020 compared with preceding seasons (Fig. 1 and 2C).
The COVID-19 epidemic has altered social and health behaviors, resulting in an unexpected reduction of seasonal in uenza cases. In uenza disease has not been detected by our national laboratories since May 8th. This dramatic decrease is unlikely to be related to climatic variations as the number of laboratorycon rmed in uenza A were similar during the same weeks in 2019 (relative risk, 1.3; 95% CI, 0.4-3.8, P=0.7), which was consistent with the typical transmission pattern in Mexico.
In addition, both the number of in uenza strains isolated from clinical specimens in commissioned laboratories and the positivity rate dropped drastically in 2020; the trends were different from 2019 (p<0.05 for both). In the 2019-2020 in uenza season (weeks 40 to 30), circulation of in uenza A (H1N1) pdm09 (47%), B (30%), A (H3N2) (19%) and in uenza A non-subtypeable (4%) was observed. In week 20, the cumulative positivity rate was 8.9%, which decreased by week 30 with only two cases of non-subtypeable In uenza A identi ed (Figure 3).

Discussion
According to the WHO, the lethality of COVID-19 is higher than for seasonal in uenza (usually well below 0.1%). COVID-19 mortality seems to differ according to the region. 6 Until now, it is estimated that the crude mortality ratio (the number of reported deaths divided by the reported cases) is between 3-4%; therefore, the infection mortality rate (the number of reported deaths divided by the number of infections) will be lower.
However, for both diseases, mortality is determined to a large extent by regional access to and quality of healthcare services.
In our country, the functional healthcare and surveillance systems installed, the government's efforts to identify ILI cases during the COVID-19 pandemic, and adequate laboratory capacity guarantee an appropriate in uenza testing and reporting of results. As reported by other countries, public health initiatives to control COVID-19 spread probably helped to decrease in uenza transmission and number of cases in May 2020 because both viruses share the same transmission mechanisms through the respiratory route and contact, although with different e ciency giving a basic reproduction number (R 0 ) for COVID-19 higher than that of seasonal in uenza, then a reduction in transmission could reduce the impact of COVID-19, thereby preventing the increase of mortality. The modeling of the effective reproduction number for COVID-19 in Mexico in May 2020 (National Secretariat of Health) at 0.5-1, is lower (55-77% less) than the mean estimated R 0 for this virus (2.2). 7 The observed 61% reduction in in uenza transmission is consistent with the information above. 8 Due to the measures implemented in our country to contain the COVID-19 cases, there were severe changes in the availability of medical care and assistance, including hospitalization. Avoidance of medical care during this period may be a major confounder in interpreting our results. It is important to note that due to the similarity in symptoms between COVID-19 and in uenza and the low number of COVID-19 patients in Mexico (<200 cases as of March 21, 2020), ILI patients would seek help for a differential diagnosis. The reduction of the availability of medical care also did not explain the lower number of severe in uenza cases seen in 2020 ( Figure 2C). Therefore, we believe that the decline in in uenza activity in Mexico in 2020 is the result of the strict control measures that were put in place in response to COVID-19.  11 We can mention some limitations in our study. First, due to the seasonal bimodal pattern of in uenza incidence in Mexico, it is expected that in uenza cases diminish around March-May. However, this year the number of cases fell to almost zero. Second, it is important to consider an underdiagnosis of ILI´s due to the fear of leaving home in search of medical attention and catching coronavirus, as well as being classi ed as COVID-19, all this contributes to generating an altered pattern 11 Third, if we consider a similar dynamic for COVID-19, we will also have to record a higher number of cases for COVID-19 due to the underdiagnosis and the presentation of asymptomatic cases. 12,13 In conclusion, we found a marked decline in in uenza cases in Mexico after the implementation of public health measures for COVID-19. De nitively, the containment measures to reduce the risk of contagion for the new respiratory disease was additionally e cient for in uenza, but with greater magnitude. This is partly explained by the different transmission mechanisms of these two respiratory diseases, the virulence of each viral agent, the absence of immunity in the population, and the diverse virological factors inherent in each pathogen. Our results suggest that such measures are useful in reducing the spread of viral respiratory diseases and their establishment mitigate the impact of the COVID-19 pandemic, especially now that we are entering into a new in uenza season.

Declarations
Author contributions Daniel Arellanos-Soto and Gerardo Padilla-Rivas conceived and designed the study, acquired data, and made the analysis and interpretation of data, and together with Ana María Rivas-Estilla drafted the paper. Daniel Arellanos-Soto made the gures. Javier Ramos-jimenez, Kame Galán-Huerta, Sonia Lozano-Sepulveda, Natalia Martínez-Acuña, Consuelo Treviño-Garza, Manuel de la O-Cavazos and Roberto Montesde-Oca-Luna revised it critically for important intellectual content. Ana María Rivas-Estilla gave the nal approval of the version to be submitted.

Additional information
Funding: The authors declare that the research was conducted in the absence of any commercial or nancial relationships that could be construed as a potential con ict of interest.
Con icts of interest/Competing interests: The authors declare there are not con icts of interest.
Ethics approval: Not necessary, the data analyzed is publically available on Secretariat of Health of México databases.
Consent to participate: Not necessary. We analyzed data publically available on Secretariat of Health of México databases.