After a string of major epidemics—novel H1N1 pandemic influenza in 2009, MERS in 2012, Ebola in West Africa in 2014, Zika in 2015—pointing out that infectious diseases are increasingly posing serious threats to the health of populations across the globe seems almost unnecessary. But while headlines—and peer-reviewed literature—attest to the fact that the incidence of emerging and reemerging deadly infectious diseases are on the rise, what do we know about endemic infectious disease deaths that occur regularly and locally? Under the International Health Regulations, countries are required to develop the ability to, at the “community and primary public health level...detect illnesses and deaths above expected levels.” Across the globe, however, countries struggle to count the total number of cases of infectious diseases and the deaths caused by them. This is true, even in countries like the United States where public health surveillance capacities are relatively advanced. While national programs may exist to track the total number of deaths, there are often limited data about how these deaths vary over time at the local level.
In a recently published JAMA article, researchers from the Institute of Health Metrics at the University of Washington report broader trends in deaths due to infectious diseases in the United States, covering several decades. Previous studies have documented declines in overall deaths due to infectious diseases at the national and state levels, due in part to improved health care and vaccination nationwide. This study identified a similar decline nationally but also identified some rather troubling trends at the local level. Below, we review the results of this study and discuss why improved understanding of the burden of routine infectious diseases is critical for outbreak detection and response.
The researchers analyzed death records from the National Center for Health Statistics to identify temporal and geographic trends in infectious disease mortality in the US. The study examined the percentage of deaths for which infectious diseases were the underlying cause at the county level for the years 1980-2014. The study also includes disease-specific analyses for 6 categories of infectious diseases that each account for at least 1% of the infectious disease deaths in the US from 1980-2014: lower respiratory infections, diarrheal diseases, HIV/AIDS, meningitis, hepatitis, and tuberculosis. Though the US CDC publishes annual summaries of nationwide notifiable infectious disease data, this study is the first to look at county-level deaths due to the above infectious diseases. As the authors point out, examining county-level data “can help identify geographic differences and their root causes.”
What did they find?
More than 4 million deaths due to infectious diseases were recorded in the US from 1980 through 2014. While the total number of deaths attributed to infectious diseases increased (from 72,220 in 1980 to 113,650 in 2014), the researchers found that percentage of all deaths due to infectious diseases actually decreased once they adjusted for population growth and overall deaths. The adjusted infectious disease mortality during the study period decreased by more than 18%, falling from 41.95 deaths per 100,000 persons in 1980 to 34.10 in 2014.
The researchers also identified several different significant disparities in infectious disease mortality. First, the percentage of deaths attributable to all infectious disease varied significantly between US counties—ranging from 12.2 per 100,000 people to 135.11. Additionally, the percentage of infectious disease deaths were significantly higher among men than women. Although interestingly, the researchers noted that the decline in infectious disease mortality from 1980 to 2014 was greater among men than women.
Among all of the recorded infectious diseases during the study period, more than 95% were attributed to the 6 disease types listed above. Among these, lower respiratory infections were the most frequent cause of death—accounting for nearly 79% of all infectious disease mortality. Compared to other infectious disease causes, deaths attributable to lower respiratory infections had the “highest absolute inequality among counties.” Nationally, deaths due to lower respiratory infections decreased by more than 25% during the study period, but there were statistically significant increases in deaths due to lower respiratory infections in more than 12% of US counties, predominantly clustered in the southeastern portion of the US.
The second most common cause of infectious disease mortality was diarrheal diseases. Though they only accounted for only 7.07% of all infectious disease deaths in 2014, diarrheal disease mortality increased in more than 99% of US counties over the course of the study period. Similar trends were observed in deaths attributable to HIV/AIDS. In 2014, the percentage of infectious disease deaths due to HIV/AIDS was 7.04%, representing a significant increase in more than 85% of US counties since 1985. This increase, however, was likely due to the fact that very few counties saw HIV/AIDS deaths in the early phase of the epidemic. Notably, there was sharp decline in deaths after the epidemic peaked in the US in 1994, likely due to the availability of HIV treatments. Additionally, HIV/AIDS deaths in 2014 were higher in men than women, a finding which suggests a need to improve targeting of men in HIV control efforts. HIV/AIDS also exhibited a shift away from wealthy West Coast counties in the 1980s and toward less affluent counties in the southeastern US in 2014.
Infectious disease mortality due to the next three categories—meningitis (1.21% of 2014 infectious disease deaths), hepatitis (0.97%), and TB (0.85%)—were considerably lower than the top three causes discussed above. Nationally and in all US counties, deaths from meningitis and TB decreased from 1980 through 2014; however, there remained geographic disparities for both. The highest mortality both diseases was found in the southeast and Alaska. Across the US, the percentage of infectious disease deaths due to hepatitis—excluding chronic hepatitis B and C, which are classified differently—also decreased during the study period, but there were statistically significant increases in some counties in the south and West Coast of the US.
While the researchers did not attempt to determine the true drivers of increasing or decreasing incidence of infectious diseases in this study, they did offer some hypotheses for the trends they observed. Because early diagnosis and proper treatment of infectious diseases have such a substantial impact on patients’ survival, one potential driver of the observed disparities in mortality is differences in patients’ ability to access health care and the quality of available health care between counties. Additionally, differences in the prevalence of risk-associated behaviors (eg, illicit drug use) and socioeconomic variation between US counties may also factor into some of the observed inequalities in infectious disease deaths.
Why is this study important for outbreaks?
The US CDC defines an outbreak or epidemic as “the occurrence of more cases of disease than expected in a given area or among a specific group of people over a particular period of time.” Without a good understanding of the baseline disease incidence, it is difficult to identify an outbreak among the normal background noise. This study is the first to show county-level, disease-specific trends in infectious disease deaths over the span of several decades. The results help improve our understanding of expected levels of infectious disease mortality over time and at a level of geographic granularity that could support the detection of localized outbreaks. By better understanding the expected number of disease- and county-specific deaths that typically occur within a given time period, public health officials can improve their chances of noticing the onset and making sense of off-normal events. For example, a cluster of deaths in a county may not noticeably change national or even statewide surveillance numbers, but it could indicate the emergence of an outbreak. Additionally, if deaths start occurring in different patient populations than expected (eg, HIV/AIDS mortality in women begins to rise faster than in men), it could indicate a changing pattern of disease transmission that needs to be investigated so as to understand whether the disease itself is changing or if new control strategies are needed.
This study is a good reminder that we are better able to detect and respond to outbreaks when we have a solid and nuanced understanding of expected infectious disease outcomes. This speaks to the importance of building core public health capacities to collect, analyze, and share data generated by surveillance activities in order to understand spatial, temporal, and other infectious disease trends to support the detection of and response to infectious disease events of all kinds.
Outbreak Observatory aims to collect information on challenges and solutions associated with outbreak response and share it broadly in near-real time to allow others to learn from these experiences in order to improve global outbreak response capabilities.
Photo courtesy of Pixabay.