COVID 19 UPDATE
Newsletter. May 19, 2020
Deborah Bickel – In residence in San Miguel
It was a shock for all of us in San Miguel de Allende to wake up to find our longest standing private hospital closed and emptied of 80 staff members and major equipment over night without notice. This was a day or two after the Chief of the hospital presented the expatriate news site for COVID related information with an operational plan for how they were going to work with the other local hospitals and the community to handle acute COVID cases.
There is a lesson to learn in this surprising event in the middle of a pandemic: We expatriates have very little idea what is really going on in our adopted country. Transparency in both business and government has seldom been highly valued in Mexico. I am not saying it is much different in the United States especially over the last four years but at least when on our home ground we are more accustomed to reading between the lines and a robust press has often but not always served us well when government leaders were withholding information, lying or simply acting against the average citizens’ interests.
The citizens of the world, facing a pandemic unlike anything we have seen in modern times, have a critical need for transparency and reliable information more than ever before. Millions of lives and world -wide economies are at stake. Transparent, nonpartisan, well informed guidance has never been more important. In this week’s New England Journal of Medicine, the press was credited with being “instrumental in creating one of the best databases on testing in the United States to fill in the gap left by the CDC’’. For more information on just how the CDC’s failure among other things led to untold preventable deaths in the early weeks of the epidemic in the US see May 15th NEJM Failing the Test: The Tragic Data Gap Undermining the U.S. Pandemic Response Eric C. Schneider, M.D.
Most of the local information expatriates read in SMA about COVID-19 comes from volunteer web and face book sites publishing the same data found on official government web sites. You will also find a long discussion of the modeling methodology Mexico is using which relies on “sentinel” data. It isn’t a very complex model. It is used when there is limited testing capacity and was selected before much was known about how changeable and infectious COVID-19 has proved itself to be.
Sentinel data is essentially limited testing done all over the country and then an agreed upon value is attached to each positive test. This value should be representational of the real incidence of the disease. It is multiplied by a factor that is agreed upon in advance based on size, homogeneity of the population and what is known about the character of the disease.
As someone who has worked for more than 25 years with epidemics such as HIV, tuberculosis and cholera for institutions such as UNICEF, Save The Children, USAID and Ministries of Health in at least 7 countries, I am very familiar with predictive modeling and its usefulness (or not) in helping us effectively respond to the impact diseases may have in the near and distant future. Modeling is meant to help us know the best way to stop the spread of an infectious disease in the face of whatever limitations exist.
In order for any model to be successful, but especially sentinel modeling, the data must be very reliable and the population density and representative nature of the population being tested must be transparently agreed upon. There must be a consensus on how many patients each positive test represents. Finally, the test itself must reliably identify the positive from those members of the population who remain negative.
I do not see that any of these conditions exist currently in Mexico. The numbers of affected and dying in the earliest and hottest spots in the capital with over 21 million people was not released until enough public outrage and press coverage forced the release of information. It finally became clear through interviews with dying or sick health workers working with inadequate protective equipment, the examination of proxy data supplied by morgues, funeral homes and cemeteries, and finally by main stream reporting and investigation from well – respected local and international press, that the epidemic was so out of control that the collection of reliable data was impossible.
As a practically trained epidemiologist, a clinician and as a public health professional, I helped shape country and regional responses to epidemics. I am very aware of the need for sustainable models that help us predict the course of a particular disease based on health data, the characteristics of a disease, the impact of culture, political climate and finally the economic limitations that nearly always exist in the economic south AKA developing countries.
I only mention my background to lend some credence to my growing anxiety about the inadequate information we are receiving from official government sources that persists in using a model that is clearly under reporting the number of COVID deaths and infections among the most densely populated parts of the country. Not many of you may know that the daily counts we look at are up to two weeks old by the time they reach a report. In a pandemic moving as rapidly as this one, that is a long delay. Even the Deputy Minister of Health, who designed and continues to promote this system, admits the model fails to identify COVID deaths because often patients die before testing is done. He also says that each positive test represents 8 to 10 positive cases when other experts place the numbers closer to 15 or 20 positive cases per one identified.
This is no small problem. A common way to assess the quality of data being reported is for health investigators to go to an area where there is an excess of deaths from diseases similar to the one under investigation. The Mexican MOH does not have the resources nor inclination to do these investigations. The Wall Street Journal reported on May 17ththat in their own review of 105 death certificates in civil registries in Mexico City, 4 were listed as COVID deaths but 64 deaths were listed as “atypical pneumonia” and despite the notation on many certificates that they were probable COVID related deaths these numbers will not be included in official COVID data.
Even if one does not take into account the limitations of Mexico’s data, the incidence of COVID appears to be on the steep side of an upward turn or at a minimum a slight dip in an otherwise steep curve. Click on this link to see the curve of officially reported COVID incidences and deaths. https://ourworldindata.org/coronavirus-testing
A clear consensus among epidemiologists and other health analysists from countries that beat back at least the first wave of the epidemic, is that mass testing and contact tracing is the only way to tell with any certainty what phase of the epidemic a country is in. Resources to do this are of course costly but the cost in human lives of not demanding a higher quality and quantity of data is staggering.
Mexico’s rank as having the lowest rate of testing among any of thecountries in the OECD – Organization for Economic Cooperation and Development (less than 1 as compared to an average of 22) is a result ofits early commitment to a sentinel survey model. Mexico is not only severely limited in its capacity to test, the data it is reporting is probably deeply flawed if the case of Mexico City is any example. Despite the limited capacity for testing and the poor quality of the data, the government claims it is ready to begin opening key industries. The closure of all but essential services has had a deep impact on the economy and the Mexican community is anxious to return to work. People are going to begin to be hungry soon and pressure will be put on city and municipal government to open up, so, why not open up now?
The Vice Minister of Health says his survey data is robust enough to detect a second wave of infection and adjust economic actions accordingly. There is no other agency from the WHO, CDC and the OECD that supports anything other than massive testing and contact tracing as a precursor to opening up the economy as well as a critical tool to check for secondary waves of infection.
Mexico cannot possibly build the capacity for wide spread testing in a short period of time and especially without the financial support of trading partners. Opening up the economy on however a limited scale without the one tool we know may work to identify new waves of disease, will drive further waves of infection that will kill the most vulnerable and put workers at risk of mass exposure to infection. The impact of these deaths and illnesses will add an untenable burden on an already poorly prepared public health infrastructure and limited pools of well-trained or adequately protected health personnel.
So, you may ask why does San Miguel de Allende appear so unaffected by the surrounding storm? All of us need to take into account that one reason we see so little evidence of death from COVID is that the mayor and the city acted well in advance of other small towns to prevent the spread of the disease at its earliest onset. Mandatory sheltering in place and city-wide enforcement of hygiene measures had to have limited the number of cases of COVID. If we open up now without the capacity to greatly increase testing and contact tracing, the five or more buses coming in daily from Mexico City and the reunification of multiple generational families alone will no doubt create evidence of the pandemic surrounding us and lead to many unnecessary deaths in our community.
DEBORAH BICKEL REPORTING FROM SAN MIGUEL DE ALLENDE