High homicide rate decreases average life expectancy in Mexico

In some northern states men live three years less on average and the situation may be even worse in other Latin American countries.  

HiramGisela Morales, the newly sworn in Temixco mayor was gunned down and killed in her home on Saturday morning. She is of the latest of many victims of rampant violence in Mexico.

Now a study in the journal Health Affairs shows that homicide rates are affecting life expectancy, especially for Mexican men. We speak with Hiram Beltrán-Sánchez, one of the study’s authors.

ResearchGate: How did you come to study the effect of homicides on life expectancy in Mexico?

Hiram Beltrán-Sánchez: There has been a lot of media attention on the increasing violence in Mexico in the last 10 years, particularly from homicides. Homicides tend to occur almost in any country, particularly among young people, but it is quite unusual to have homicide rates as high as those reported in national surveys in Mexico. For example, the homicide rate in Mexico increased from 9.5 homicides per 100,000 people in 2005 to more than 22.0 per 100,000 people in 2010. This high homicide level prompted us to think that homicides were probably having an impact on overall life expectancy in the country.

RG: Which data did you base your study on?

Beltrán-Sánchez: We base our study on national data from the Mexican National Statistical Office that collects number of deaths and classifies them by cause (i.e. cancer, diabetes, homicides).  These are the databases typically used to estimate mortality levels in a country.

RG: What did you find? What was the average life expectancy a decade ago, and what is it now?

Beltrán-Sánchez: We found that life expectancy among men stagnated over the decade and remained at about the same level in 2000 and 2010 (at about 72 years of life) while that of women increased at a very slow pace (from 77.2 years of life in 2000 to 77.8 years in 2010). More importantly, when we looked at life expectancy in each state of the country we found remarkable results. For example, male life expectancy in about two-thirds of the states ended up with lower life expectancy in 2010 than they had had ten years earlier. People living in the north of the country had the worst experience; male life expectancy fell by up to three years in Chihuahua, Sinaloa and Durango during the latter half of the decade (2005-2010). The importance of this result is that life expectancy in Mexico had been growing by an average of 5.5 years per decade from 1940 up to 2000; it was thus a surprise to see a reversal in life expectancy among men in most of the states in the country between 2000 and 2010.

RG: How can you be sure that homicides are the main reason for the drop?

Beltrán-Sánchez: In our study we included all deaths reported in the Mexican National Statistical Office and created eight groups of causes of death, including homicides as a separate category. Because we included all possible causes of death, this classification allow us to identify the cause of death that had the largest impact on life expectancy. Our results indicate that homicides had the largest impact on life expectancy, and we also show that in northern states homicides reached very high levels that male life expectancy fell by up to three years in some states (e.g., Chihuahua).

RG: You assume that your results may be underestimating the effect. How so, and by how much?

Beltrán-Sánchez: There are at least three problems with the data we use that make it unlike for homicides to be accurately reported in the Mexican National Statistical Office. First, it has been reported that not all homicides are registered in the statistical office and this may lead to an underestimate of the actual number of deaths. Second, it has also been reported that some deaths due to homicides are wrongly coded (for example, some reports indicate that some of these deaths are coded as suicides); because we do not have access to the death certificate it was impossible for us to ascertain the cause of death. Third, there have been many reports indicating a large number of missing people associated with the drug on wars in Mexico and it is unclear how many of them may have died from homicides.

We don’t know by how much the above factors may be underestimating the effect of homicides. To our knowledge there has not been a study that estimates the impact of the above factors on homicide mortality.

RG: How many less murders would it take to have life expectancy rise again in Mexico?

Beltrán-Sánchez: An increase in life expectancy results from mortality declines in many causes of death, it is thus difficult to ascertain a specific reduction in homicides that could lead to an increase in life expectancy. For example, we showed that despite the increase in homicide mortality other causes of death declined during the period 2005-2010 that lead to increases in life expectancy; this increase in life expectancy, however, was largely offset by the detrimental impact of homicides.  It is likely that if homicides decline to the levels shown before the implementation of the war on drugs in 2005 (9.5 homicides per 100,000 people), there would be a major gain in life expectancy in Mexico. For example, a reduction of 56% on the homicide rate in 2010 would have led to a rise of more than half a year (0.6) in male life expectancy.

RG: Do you know if there’s a similar trend in other countries?

Beltrán-Sánchez: We know that the homicide rate in Mexico in the period of study (2000-2010) was lower than in other Latin American countries — for example Honduras, Belize, El Salvador, Colombia and Brazil. Based on our results from Mexico, one would expect homicides to have a greater impact in these countries.

RG: Where will these results take you from here?

Beltrán-Sánchez: There is an urgent need to document the impact of homicides on the Latin American population. We are continuing this line of work in Central American countries to ascertain the impact of homicides on average years of life in these populations.

RG: What should the consequences of your findings be?

Beltrán-Sánchez: Second, that the main message of our study is to raise awareness of the large impact of homicides on average years of life in the Mexican population and to emphasize that the policies implemented in the past decade have not controlled the high levels of violence in Mexico.  We suggest that these policies should be discontinued, and new strategies incorporating a public health perspective should be implemented to minimize the effects of violence on the health status of the Mexican population.