Health Care Access and Utilization among Native Hawaiians and Pacific Islanders Persons in the United States, 2014

August 18, 2017

Questions for Carla Zelaya, Survey Statistician and Lead Author on “Health Care Access and Utilization among Native Hawaiians and Pacific Islanders Persons in the United States, 2014

Q: What did your report find on health care access for the Native Hawaiian and Pacific Islander (NHPI) population?

CZ: Some of the main highlights of the report were NHPI adults were more likely to be insured, have public health coverage, have a usual place of care and a flu vaccination in the past 12 months compared with all U.S. adults. However, NHPI adults were also less likely to have private insurance, and those aged 65 and over were less likely to have ever received a pneumococcal vaccination compared with all U.S. adults.

Q: Is this the first published health data on NHPI population?

CZ: The Native Hawaiian and Pacific Islander National Health Interview Survey (NHPI NHIS), is the first federal survey designed exclusively to measure the health of the civilian non-institutionalized NHPI population of the United States. It was conducted by NCHS, and this is the third report published from the survey.

The first two reports are:

Citation #1: Galinsky AM, Zelaya CE, Simile C, Barnes PM. Health conditions and behaviors of Native Hawaiian and Pacific Islander persons in the United States, 2014. National Center for Health Statistics. Vital Health Stat 3(40). 2017.

Citation #2: Galinsky AM, Zelaya CE, Barnes PM, Simile C. Selected health conditions among Native Hawaiian and Pacific Islander adults: United States, 2014. NCHS data brief, no 277. Hyattsville, MD: National Center for Health Statistics. 2017.

Q: How do you collect data on the NHPI population?

CZ: To conduct the 2014 NHPI NHIS, NHIS field interviewers visited houses previously identified in the American Community Survey (ACS) to have at least one resident of any age with an NHPI racial identity (either alone or in combination with one or more other racial identities).

Q: Is there anything in the data that surprised you?

CZ: Within the NHPI population it was especially interesting that Native Hawaiians and Pacific Islanders adults differed in prevalence of uninsurance and private coverage (but not public coverage).

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

CZ: The prevalence of many indicators of access and utilization of health care among NHPI people differ from the total U.S. population and more specifically from Asians in the U.S. There was also diversity in access and utilization of health care within the NHPI population.

Drug Overdose Deaths Among Adolescents Aged 15-19 in the United States: 1999-2015

August 16, 2017

Questions for Sally Curtin, Statistician and author of “Drug Overdose Deaths Among Adolescents Aged 15-19 in the United States: 1999-2015

Q:  Do trends in overdose deaths among teens reflect the trends of older adults in the U.S.?

SC: There are some similarities, but also differences.  Both teens and older adults experienced the sharp increases from 1999 through the mid-2000s.  But unlike older adults, whose rates continued to increase, teenagers actually had a decline in drug overdose death rates through 2014, before an upturn in 2015.  All of this decline was for males as the rates for females stabilized from 2004-2013 before increasing again.


Q: Do we know why trends for teens dropped during the first several years of the millennium?  And why they increased sharply in 2015?

SC: There are many public health initiatives to combat the rising drug overdose death rates.  While we do not know the exact reason for the decline, we know the specific drugs that were involved—opioids, cocaine, and benzodiazepines.  For the opioids, it was the frequently prescribed drugs—methadone and natural and semisynthetic (oxycodone, morphine) that had declines for teens since the mid 2000s.  Other opioids such as heroin and synthetic opioids (including fentanyl) fluctuated but generally increased over the 1999-2915 study period.  The continued rise in drug overdose deaths involving heroin and synthetic opioids from 2014 to 2015 contributed to the uptick between those years.


Q: What are the differences in overdose deaths by gender and race?

SC: We did not examine race in this report because the numbers were too small for some groups.  By gender, the drug overdose death rate for males was higher for females for every year of the 1999-2015 period and was 70% higher in 2015.  While males had a greater increase in drug overdose death rates than females between 1999 and the mid-2000s, they also declined by about a third between 2007 and 2014 before increasing again.  The rate in 2015 was still lower than the 2007 peak.  Females had an increase, albeit smaller than for males, and then their rate stabilized between 2004-2013 before increasing again.


Q:  What type of drugs are killing these teens?

SC: As for the population at large, the majority of drug overdose deaths involve opioids.  When we examined the specific type of opioid involved, heroin is the leading drug involved and rose fairly steadily throughout the study period.  Synthetic opioids (including fentanyl) were lower than other opioid drugs through the early years of the period, but then doubled between 2014 and 2015.  This large increase for synthetic opioids has also been observed for the population at large.

We did not look at combinations of drugs.  Often, there is more than one drug involved so the categories we show are not mutually exclusive.

Antidepressant Use in Persons Aged 12 and Over: United States, 2011-2014

August 16, 2017

Questions for Laura Pratt, Psychiatric Epidemiologist and Author of “Antidepressant Use in Persons Aged 12 and Over: United States, 2011-2014

Q:  Are more people taking antidepressants now vs. in the past?

LP:   Yes, in our data brief, figure 4, you can see how antidepressant use has increased over time from 1999-2002 to 2011-2014.  Slightly less than 8% of the U.S. population took antidepressants in 1999-2002 while almost 13% took antidepressants in 2011-2014.  This is an increase of about 65%.  The rates of increase were similar for males and females, but twice as many females took antidepressants as males at all time points.


Q: Is there any particular age group in which antidepressant use is higher?

LP:   Among all persons and among females, antidepressant use was highest in persons 60 years of age and older.


Q:  Does this mean that rates of mental illness are on the rise?

LP:   Our report does not look at rates of mental illness.  But in general, prescription drug use is also related to healthcare access and utilization, and, in mental health particularly, many studies have shown high rates of under-treatment.  The situation with a large percent of people with depression, for example, not receiving treatment has improved over time.  Increases in healthcare utilization and treatment of depression would result in a higher rate of antidepressant use whether or not the rates of mental illness increased.


Q:  What are the risks or dangers of antidepressant use?

LP:   The first antidepressants that were available had many side effects and could cause overdose death.  The vast majority of overdose deaths related to these drugs were intentional (suicides). The newer antidepressants in use today have fewer side effects and have a much lower risk of overdose.  Antidepressants do not produce a “high” and are not drugs of abuse.


Q:  Any other findings you feel are noteworthy?

LP:   It was very noteworthy that non-Hispanic white persons ages 12 and older continue to have rates of antidepressant use that are between 3 and 5 x higher than persons in other race and Hispanic origin groups. I was also surprised to see that 25% of people who take an antidepressant have taken it for more than 10 years.  In our first antidepressant data brief, the percent of people taking an antidepressant for more than 10 years was 13.6%.  Interestingly, the percent of persons taking antidepressants who took them for more than 2 years was 61% in 2005-08 and increased to 68% in 2011-2014.

Condom Use During Sexual Intercourse Among Women and Men Aged 15-44 in the United States: 2011-2015 National Survey of Family Growth

August 10, 2017

Questions for Casey Copen, Ph.D., M.P.H., Statistician and Author of “Condom use during sexual intercourse among women and men aged 15-44 in the United States: 2011-2015 National Survey of Family Growth

Q: Why did you decide to examine condom use in this report?

CC: There are about 20 million new sexually transmitted infections (STI) in the United States each year. These infections can lead to long-term health consequences, such as infertility. Condoms can provide protection against most STIs but are often used incorrectly and inconsistently. Similarly, condoms can reduce the risk of pregnancy but have one of the highest rates of contraceptive failure of any contraceptive method.

The purpose of this report was to examine recent trends in condom use during vaginal (sexual) intercourse and measure the prevalence of condom use, alone or in combination with another contraceptive method. This information can be used to plan health services and educational programs in the U.S.

Q: Has the data in this report been previously published?

CC: Data on condom use has been collected for women since 1973 and for men, since 2002. Starting in 2013, questions on problems with condom use were asked of women aged 15-44 who used a condom during intercourse in the past 4 weeks. Periodically, the National Survey of Family Growth (NSFG) publishes reports on current contraceptive use, which includes condoms, to measure how contraceptive use among women has changed over time.

Last month, NSFG released a report on contraceptive use among female and male teens aged 15-19 that included information on condom use at first and most recent intercourse ( However, the data on condom use problems have not been previously published.

Q: What do you think were the main findings of your report?

CC: In 2011-2015, 23.8% of women and 33.7% of men aged 15–44 used a condom at last sexual intercourse in the past 12 months. Among condom users aged 15-44, the majority of women and men used only a condom and no other method during last intercourse in the past 12 months (59.9% of women and 56.4% of men); another 25.0% of women and 33.2% of men used condoms plus hormonal methods; and 15.1% of women and 10.5% of men used condoms plus non-hormonal methods. Almost 7% of women aged 15–44 who used a condom in the past four weeks said the condom broke or completely fell off during intercourse or withdrawal and 25.8% said the condom was used for only part of the time during intercourse.

Q: How has the percentage of condom use in the U.S. during sexual intercourse changed since 2002?

CC: The percentages of women aged 15-44 who used condoms at last sexual intercourse in the past 12 months were similar from 2002 to 2011-2015, but for men aged 15–44, the percentages increased from 29.5% in 2002 to 33.7% in 2011–2015. Percentages of female and male condom users aged 15–44 who used condoms only at last intercourse in the past 12 months decreased from 67.9% of women and 63.0% of men in 2002 to 59.9% and 56.4% in 2011-2015. Alongside this decline, percentages of female condom users aged 15–44 who used condoms plus non-hormonal methods at last intercourse increased from 11.9% in 2006–2010 to 15.1% in 2011–2015.

Q: What is the take-home message of your report?

CC: This report showed that the majority of women and men aged 15-44 in each NSFG survey period did not use a condom at all during last sexual intercourse. It is important to note that there are many factors associated with condom nonuse, such as being in a monogamous relationship, using hormonal or other methods to prevent pregnancy or trying to get pregnant. Although this report could not address each of these factors directly, these findings indicate there are differences in condom use by age, education, Hispanic origin and race, relationship at last sexual intercourse and number of sexual partners in the past 12 months. About one-quarter of women aged 15-44 who used a condom during intercourse in the past 4 weeks reported that the condom was used for only part of the time during intercourse (25.8%), suggesting that condoms are often used inconsistently.

Quarterly Provisional Estimates for Selected Birth Indicators, 2015—Quarter 1, 2017 Quarterly Provisional Estimates of Infant Mortality, 2014—Quarter 3, 2016 Vital Statistics Rapid Release from the National Vital Statistics System

August 8, 2017

Lauren_RossenQuestions for Lauren Rossen, Ph.D., Health Statistician and Lead Author of “Quarterly Provisional Estimates for Selected Birth Indicators”and “Quarterly Provisional Estimates of Infant Mortality

Q: What findings in your new data analyses on births and infant mortality most surprised you and why?

LR:  These latest quarterly provisional estimates suggest that the steady decline in teen birth rates that we have seen over the past several years is continuing into 2017, which is good news. What is of concern is the recent uptick in preterm birth rates, a trend that emerged in 2015 and that has unfortunately continued into 2016 and early 2017.

Q: What is the difference between the Rapid Release provisional estimates on births released today and the report from your office released last month, “Births: Provisional Data for 2016”? And how are these two provisional data analyses different from your office’s “preliminary data” released in the recent past?

LR:  The most recent Quarterly Provisional Estimates provide an update to some of the data released in the recent report, Births: Provisional Data for 2016. That report is similar to previous “preliminary birth data” reports, but is redesigned and released under our Vital Statistics Rapid Release (VSRR) program. We hope that the VSRR program can be a one-stop-shop for our provisional vital statistics data. The Quarterly Provisional Estimates describe very recent trends in key indicators of maternal and infant health from the birth and mortality data the report, Births: Provisional Data for 2016, provides some critical context for understanding these recent trends.  report also describes some additional demographic and reproductive health indicators that aren’t yet available in the Quarterly Provisional Estimates, such as birth rates by race and Hispanic origin, as well as the timing of prenatal care.

Q: What in your data analyses can be attributed to no change in infant mortality in the last few quarters?

LR: is another surprising and concerning finding, because infant mortality rates have generally been declining over the past decade, at least through 2014. These declines seem to have leveled off more recently, according to our provisional estimates. We can’t speak to why infant mortality rates might no longer be declining, but we are planning future research to help us better understand this troubling trend.    

Q: What differences did you see among various age groups of mothers?

LR: There is a great deal of detail in the recent report, Births: Provisional Data for 2016, discussing how age-specific birth rates have changed recently. Generally, both that report and our recent Quarterly Provisional Estimates show that maternal age is increasing. Birth rates among younger women (under 30) are going down, while those among women 35 and up are increasing.

Q: What are seasonal fluctuations in the number of infant deaths and births, and what do you mean by accounting for seasonality as you described in your report’s preface?

LR: People may not think that there are seasonal patterns to births, but it turns out that there are more babies born in the third quarter of the year, from July-September, than during other parts of the year. There are seasonal patterns in other indicators as well. For example, preterm birth rates dip slightly in the third quarter compared with other quarters, while infant mortality rates tend to be a bit higher in the beginning of the year than toward the end of the year. So to ensure that any differences we find aren’t influenced by seasonal fluctuations, we only compare the most recent quarter with the same quarter from the previous year. We also present 12 month-ending estimates, which include all seasons of the year, and thus aren’t subject to seasonal ups and downs.

Stat of the Day – August 7, 2017

August 7, 2017

QuickStats: Suicide Rates for Teens Aged 15–19 Years, by Sex — United States, 1975–2015

August 7, 2017

The suicide rate for males aged 15–19 years increased from 12.0 to 18.1 per 100,000 population from 1975 to 1990, declined to 10.8 by 2007, and then increased 31% to 14.2 by 2015.

The rate in 2015 for males was still lower than the peak rates in the mid- 1980s to mid-1990s.

Rates for females aged 15–19 were lower than for males aged 15–19 but followed a similar pattern during 1975–2007 (increasing from 2.9 to 3.7 from 1975 to 1990, followed by a decline from 1990 to 2007).

The rates for females then doubled from 2007 to 2015 (from 2.4 to 5.1). The rate in 2015 was the highest for females for the 1975–2015 period.