Dementia Mortality in the United States, 2000–2017

March 14, 2019

A new NCHS report presents data on mortality attributable to dementia. Data for dementia as an underlying cause of death from 2000 through 2017 are shown by selected characteristics such as age, sex, race and Hispanic origin, and state of residence.

Trends in dementia deaths overall and by specific cause are presented. The reporting of dementia as a contributing cause of death is also described.

Key Findings:

  • In 2017, a total of 261,914 deaths attributable to dementia as an underlying cause of death were reported in the United States. Forty-six percent of these deaths were due to Alzheimer disease.
  • In 2017, the age-adjusted death rate for dementia as an underlying cause of death was 66.7 deaths per 100,000 U.S. standard population. Age-adjusted death rates were higher for females (72.7) than for males (56.4).
  • Death rates increased with age from 56.9 deaths per 100,000 among people aged 65–74 to 2,707.3 deaths per 100,000 among people aged 85 and over.
  • Age-adjusted death rates were higher among the non-Hispanic white population (70.8) compared with the non-Hispanic black population (65.0) and the Hispanic population (46.0).
  • Overall, age-adjusted death rates for dementia increased from 2000 to 2017.
  • Rates were steady from 2013 through 2016, and increased from 2016 to 2017. Patterns of reporting the individual dementia causes varied across states and across time.

Updated Provisional Drug Overdose Death Data: 12-Month Ending from August 2017-August 2018

March 13, 2019


QuickStats: Death Rates Attributed to Excessive Cold or Hypothermia Among Persons Aged 15 Years or Older, by Urbanization Level and Age Group

February 22, 2019

During 2015–2017, death rates attributed to excessive cold or hypothermia increased steadily with age among those aged 15 years or older in both metropolitan and nonmetropolitan counties.

The rate for persons aged 85 years or older reached 3.8 deaths per 100,000 in metropolitan counties and 7.3 in nonmetropolitan counties.

The lowest rates were among those aged 15–24 years (0.2 in metropolitan counties and 0.5 in nonmetropolitan counties). In each age category, death rates were lower in metropolitan counties and higher in nonmetropolitan counties.

Source: National Center for Health Statistics, National Vital Statistics System, Mortality Data 2015–2017.

QuickStats: Death Rates for Motor Vehicle Traffic Injury, by Age Group — National Vital Statistics System, United States, 2015 and 2017

February 15, 2019

From 2015 to 2017, death rates for motor vehicle traffic injury increased for persons aged 15 years or older.

For infants and children younger than 15 years there was no statistically significant change from 2015 to 2017, and this group had the lowest death rate (2.0 deaths per 100,000) in 2017.

The highest death rate in 2017 was for persons aged 75 years or older (19.1), followed by a 15.3 death rate for persons aged 15–34 years, and 12.8 for persons aged 35–54 and 55–74 years.

Source: National Vital Statistics System. Underlying cause of death data, 1999–2017.

Drugs Most Frequently Involved in Drug Overdose Deaths: United States, 2011–2016

December 12, 2018

Questions for Lead Author Holly Hedegaard, M.D., M.S.P.H., Health Statistician, and author of “Drugs Most Frequently Involved in Drug Overdose Deaths: United States, 2011–2016.”

Q: Is there a specific finding in this report that surprised you?

HH: During the six years of the study, the relative ranking of the drugs most frequently involved in drug overdose deaths changed. In 2011, the drug most frequently involved in drug overdose deaths was oxycodone, in 2012-2015 was heroin and in 2016 was fentanyl. In 2016, fentanyl was involved in nearly 30% of the drug overdose deaths in the United States.

The drugs most frequently involved in drug overdose deaths also varied by the intent of the death. In 2016, the drugs most frequently involved in unintentional (accidental) drug overdose deaths were fentanyl, heroin and cocaine, while the drugs most frequently mentioned in suicides by drug overdose were oxycodone, diphenhydramine, hydrocodone, and alprazolam.

Q: How is the data in this report different from the recently released drug overdose data brief and provisional drug overdose numbers produced by NCHS?

HH: The drug overdose data brief and the provisional drug overdose numbers produced by NCHS involve analysis of death certificate data coded using the International Classification of Diseases, Tenth Revision (ICD-10). One limitation of this classification system is that, with a few exceptions, ICD–10 codes reflect broad categories of drugs rather than unique specific drugs.

In the National Vital Statistics Report, NCHS uses data from the literal text on death certificates to identify the specific drugs involved in the death. Using this method, we can look at the number of deaths involving specific drugs, such as oxycodone, hydrocodone, or fentanyl, for example, rather than be limited to the broader categories found with ICD-10 coded data, such as natural and semi-synthetic opioids or synthetic opioids other than methadone.

Q: What did your report find on the percentage of drug overdose deaths mentioning at least one specific drug or substance?

HH: Using the literal text to identify the specific drugs involved is dependent on whether or not the specific drugs are reported on the death certificate. The specificity of reporting has improved in recent years. In 2011, the specific drugs or drug classes involved were reported for 78% of drug overdose deaths; in 2016, the reporting increased to nearly 88% of drug overdose deaths.

Q: Do you have data that goes further back than 2011?

HH:  A previous report looked at the drugs most frequently involved in drug overdose deaths in 2010-2014. That report is available at

Q: Do you have data on drugs most frequently involved in drug overdose deaths that goes up to 2017?  If not, when do you expect that will be available?

NCHS does not currently have information on the drugs most frequently involved in drug overdose deaths in 2017. NCHS is currently preparing the data files for analysis. The results for 2017 will be available in 2019.

Q: What is the take home message for this report?

HH: The patterns in the specific drugs most frequently involved in drug overdose deaths can change from year to year. Complete and accurate reporting in the literal text on death certificates of the specific drugs involved provides critical information needed for understanding and preventing drug overdose deaths.

Mortality in the United States, 2017

November 29, 2018

Questions and Answers from the authors of the recently released 2017 mortality data.  The data can be found in the following reports, “Mortality in the United States, 2017, ” “Drug Overdose Deaths in the United States, 1999–2017, ” and “Suicide Mortality in the United States, 1999–2017.”

Q: Why did life expectancy decline in 2017?

A: Mortality rates increased for 7 out of the 10 leading causes of death in the U.S., including a 5.9% increase in the flu/pneumonia death rate, a 4.2% increase in the accidental/unintentional injury death rate, and 3.7% in the suicide rate. Many of the accidental/unintentional deaths were from drug overdoses, which continued to increase in 2017.

Q: Isn’t this the third straight year that life expectancy declined?

A: Estimated life expectancy at birth in 2017 was 0.3 years lower than in 2014 and 0.1 years lower than in 2016. The 2016 life expectancy estimate was revised to 78.7 years, up from an estimated 78.6 years, which was reported a year ago. This means that the 2016 life expectancy estimate is the same as the 2015 estimate, which also was revised to 78.7 years, down from an estimated 78.8 years, originally reported two years ago. As a routine matter, for the highest degree of accuracy, NCHS blends Medicare data for people ages 66 and over with our vital statistics data to estimate life expectancy. However, the two data sets are released on different schedules. When Medicare data for a year aren’t available at the time we release our final mortality statistics, we use the most recent Medicare data available at the time. We later revise life expectancy estimates when updated Medicare data become available.

Q: How many deaths in 2017 were attributed to opioids?

A: In 2017, 47,600 drug overdose deaths mentioned involvement of any type of opioid, including heroin and illicit opioids, representing over two-thirds of all overdose deaths (68%).

Q: Why is the 70,237 number of overdose deaths smaller than what CDC has previously reported for 2017?

A: The 70,237 number is a final, official number of overdose deaths among U.S. residents for 2017 whereas the previously reported (and slightly higher) numbers were provisional estimates. In August of 2017, CDC began calculating monthly provisional data on counts of drug overdose deaths as a rapid response to this public health crisis, in order to provide a more accurate, closer to “real-time” look at what is happening both nationally and at the state level. These monthly totals are provisional counts, and they include all deaths occurring in the U.S. – which include deaths among non-residents (i.e., visitors here on business or leisure, students from abroad, etc). These counts also do not include deaths that are still under investigation. As a result, the monthly numbers are provisional or very preliminary, and the final 2017 number of 70,237 deaths is an official number that only include deaths among U.S. residents and account for any previously unresolved deaths that were under investigation.

Q: Does this mean that the 70,237 total does not include deaths to undocumented immigrants here in the U.S.?

A: We don’t get immigration status off the death certificates, so we wouldn’t know how many of the deaths were to undocumented immigrants.

Q: In comparing the 2017 numbers with 2016 and past years, is the crisis of drug overdose deaths growing or about the same?

A: From 2016 to 2017, the number of drug overdose deaths increased from 63,632 deaths to 70,237, a 10% increase, which is a smaller increase compared to the 21% increase from 2015 to 2016, when the number of drug overdose deaths increased from 52,404 deaths to 63,632 deaths. Over a longer period of time, from 1999 through 2017, the age-adjusted rate of drug overdose deaths increased on average by 10% per year from 1999 to 2006, by 3% per year from 2006 to 2014, and by 16% per year from 2014 to 2017. So the trend is continuing, although the increase in 2017 was not as large as in previous years.

Q: Are there any other trends of significance when looking at the types of drugs attributed to overdose deaths?

A: The rate of drug overdose deaths involving synthetic opioids other than methadone, which include drugs such as fentanyl, fentanyl analogs, and tramadol, increased 45% in one year, from 6.2 per 100,000 in 2016 to 9.0 per 100,000 in 2017. In 2017, 40%(?) of all drug overdose deaths mentioned involvement of a synthetic opioid other than methadone.

Q: Has fentanyl overtaken heroin as the major cause of overdose death?

A: The data brief on drug overdose deaths does not specifically address fentanyl. However the rate of drug overdose deaths involving synthetic opioids other than methadone, which includes fentanyl, increased 45% 2016 and 2017 whereas the overdose death rate from heroin did not change (4.9 deaths per 100,000).

Q: There is a lot of stark news in these three reports. Are there any positives to report?

A: The cancer mortality rate declined between 2016 and 2017, and although estimated life expectancy declined in 2017, life expectancy for people at age 65 actually increased. Also, regarding drug overdose deaths, the rate of increase in drug overdose deaths slowed between 2016 and 2017, although the increases that occurred were still very significant.

QuickStats: Age-Adjusted Death Rates from Lung Cancer by Race/Ethnicity — National Vital Statistics System, United States, 2001–2016

August 6, 2018

During 2001–2016, the lung cancer death rates for the total population declined from 55.3 to 38.3 as well as for each racial/ethnic group shown.

During 2001–2016, the death rate for the non-Hispanic black population decreased from 63.3 to 41.2, for the non-Hispanic white population from 57.7 to 41.5, and for the Hispanic population from 23.9 to 16.6.

Throughout this period, the Hispanic population had the lowest death rate.

SOURCE: CDC/National Center for Health Statistics, National Vital Statistics System, 2001–2016, Mortality. CDC Wonder online database.