Municipality- level predictors of COVID-19 mortality in Mexico: a cautionary tale

Background

Inequalities and burden of comorbidities of the Coronavirus disease 2019 (COVID-19) vary importantly inside the countries. We aimed to analyze the Municipality-level factors associated with a high COVID-19 mortality rate of in Mexico.

Methods

We retrieved information from 142,643 cumulative confirmed symptomatic cases and 18,886 deaths of COVID-19 as of June 20^th, 2020 from the publicly available database of the Ministry of Health of Mexico. Public official data of the most recent census and surveys of the country were used to adjust a negative binomial regression model with the quintiles (Q) of the distribution of sociodemographic and health outcomes among 2,457 Municipality-level. Expected Mortality Rates (EMR), Incidence Rate Ratios (IRR) and 95% Confidence Intervals are reported.

Results

Factors associated with high MR of COVID-19, relative to Quintile 1 (Q1), were; diabetes prevalence (Q4, IRR=2.60), obesity prevalence (Q5, IRR=1.93), diabetes mortality rate (Q5, IRR=1.58), proportion of indigenous population (Q2, IRR=1.68), proportion of economically active population (Q5, IRR=1.50), density of economic units that operate essential activities (Q4, IRR=1.54) and population density (Q5, IRR=2.12). We identified 1,351 Municipality-level without confirmed COVID-19 deaths, of which, 202 had nevertheless high (Q4, Mean EMR= 8.0 deaths per 100,000) and 82 very high expected COVID-19 mortality (Q5, Mean EMR= 13.8 deaths per 100,000).

Conclusion

This study identified 1,351 Municipality-level of Mexico that, in spite of not having confirmed COVID-19 deaths yet, share characteristics that could eventually lead to a high mortality scenario later in the epidemic and warn against premature easing of mobility restrictions. Local information should be used to reinforce strategies of prevention and control of outbreaks in communities vulnerable to COVID-19.

Key messages

• Predictors of COVID-19 mortality varied importantly between Municipality-level.

• Municipality-level factors associated with high mortality of COVID-19 were the prevalence of obesity and diabetes, mortality rate of diabetes, the proportion of indigenous and economically active population and population density.

• Municipality-level with high case-fatality rates of COVID-19 are likely undergoing insufficient testing and should improve its availability.

• Identified predictors ought to be considered by local governments to reinforce tailored strategies to prevent casualties in populations vulnerable to COVID-19, as mortality is expected to be eventually high in some Municipality-level that may not have reached the apex of the epidemic yet.