Mortality From Sexually Transmitted Diseases in Reproductive-Aged Women: United States, 1999–2010

Abstract

Objectives. I estimated the sexually transmitted disease–related mortality among US reproductive-aged women from 1999 to 2010.

Methods. I estimated mortality from National Center for Health Statistics’ Multiple Cause of Death data. I defined reproductive age as 15 to 44 years. For diseases partially caused by sexual transmission, I estimated the proportion attributable to sexual transmission from the literature. To calculate mortality rates, I estimated number of deaths from each disease and Census Bureau population for reproductive-aged women for 1999 to 2010.

Results. From 1999 to 2010, the cumulative sexually transmitted disease–related mortality rate decreased by 49%, from 5.3 to 2.7 deaths per 100 000. The primary contributors were HIV and human papilloma virus infections. Mortality from sexually transmitted HIV infection decreased by 62%, from 3.4 to 1.3 deaths per 100 000. Mortality from human papilloma virus–associated gynecologic cancers decreased by 19%, from 1.6 deaths per 100 000 in 1999 to 1.3 deaths per 100 000 in 2010.

Conclusions. Screening and treatment for sexually transmitted diseases may reduce mortality. Research is needed to determine whether sexually transmitted disease–related morbidity among reproductive-aged women has decreased over the past decade.

Deaths from sexually transmitted diseases (STDs) often occur long after acute infection, making their incidence difficult to estimate.^1–3 Some infections, such as syphilis, may directly result in death. By contrast, human papilloma virus (HPV), HIV, and hepatitis more commonly cause death because of secondary sequelae. Genital herpes, gonorrheal infection, and chlamydial infection may cause death from primary infection (e.g., disseminated herpes simplex virus or pelvic inflammatory disease [PID]) or secondary sequelae (e.g., ectopic pregnancy). Because death certificates usually list only the prevalent conditions, STDs are often not recorded on death certificates. In addition, some causes of death, such as liver cancer, may or may not be STD related. STD-related deaths cannot always be counted directly; therefore, direct and indirect methods of estimation must be used.

The methodology for estimating STD-related mortality in women in the United States has evolved. Early estimates used measures of reproductive health mortality.¹ Subsequently, a more comprehensive approach used direct and indirect measures of deaths from all STD-related diseases.² Most recently, disability-adjusted life-years lost because of sexual behaviors were found to be threefold higher in the United States than in other developed countries. Women carry a disproportionately high proportion of the health burden.³

Between 1955 and 1975, STDs such as syphilis and PID were responsible for 20% to 30% of reproductive mortality among women aged 15 to 44 years.¹ Since then, mortality from syphilis and PID has decreased. Cervical carcinoma, HIV, and viral hepatitis infection were the top 3 contributors to STD-related mortality among all women older than 15 years as of 1992.² I have updated the estimated annual STD-related mortality among reproductive-aged women in the United States for 1999 to 2010.

METHODS

I calculated a crude mortality rate (deaths per 100 000 reproductive-aged women per year) by dividing the total number of deaths from each STD by the population of reproductive-aged women for that year. I identified STD-related mortality with the Multiple Cause of Death database available from the Centers for Disease Control and Prevention (CDC) National Center for Health Statistics.⁴ This database provides access to death certificate data for US residents. At the time of death, a single International Classification of Diseases (ICD)⁵ code is provided to define a single cause of death for each decedent. If multiple causes of death are identified, up to 30 ICD codes are listed under multiple cause of death.

I identified the total number of deaths for reproductive-aged women (aged 15–44 years) for whom a STD-related etiology was identified as an underlying cause of death between 1999 and 2010. I identified STDs and sequelae with the ICD-10 codes⁵ listed in Table 1. For infections or sequelae that are partially sexually transmitted, I derived a percentage attributable to sexual transmission from estimates from the literature, consistent with previous mortality analyses (Table 1).^1,2 I considered the following to be universally transmitted by sexual contact in reproductive-aged women: Neisseria gonorrheae, Chlamydia trachomatis, genital herpes simplex virus, HPV, and syphilis. I identified the following as partially sexually transmitted: HIV, hepatitis B virus, and hepatitis C virus. I identified the following partially STD-related sequelae: PID, ectopic pregnancy, chronic liver disease, hepatocellular carcinoma, and cervical, vulvar, and vaginal carcinoma.

TABLE 1—

ICD-10 Code Designation and Estimates for Proportion Attributable to Sexual Transmission for Selected Diseases: United States, 1999–2010

Disease	ICD-10 Code	Sexually Transmitted, %	Range, %
Viral hepatitis B	B16, B18, B18.1	26	22–336–10
Viral hepatitis C	B17.1, B18.2	9	6–157,11–14
Hepatocellular carcinoma	C22	10	8–1215,16
Chronic liver disease	K74	5	4–82,17
Chlamydia trachomatis	A56	100	. . .
Neisseria gonorrheae	A54	100	. . .
Pelvic inflammatory disease	N70–N76	65	40–9018–20
Ectopic pregnancy	O00	4321	43–7221–24
HIV	B20–B24	80	70–9025–27
Genital herpes	A60, B00	100	. . .
Cervical cancer	C53	10028–30	. . .
Vulvar and vaginal cancer	C51, C52	5031	. . .
Syphilis	A50–A53	100	. . .
Other (lymphogranuloma venereum, chancroid, granuloma inguinale, trichomoniasis, anogenital warts, other predominantly sexually transmitted disease not elsewhere classified)	A55, A57–A59, A63, A64	100	. . .

Note. HSV = herpes simplex virus; ICD-10 = International Classification of Diseases, 10th revision.⁵

I estimated the number of reproductive-aged women each year from 1999 to 2010 from the US Census Bureau.⁴ If the number of deaths was too small to calculate a reliable mortality rate, I reported the total number of deaths. An independent reviewer repeated the data extraction and analysis to confirm accuracy.

RESULTS

Between 1999 and 2010, an estimated 31 351 STD-related deaths occurred among reproductive-aged women (Table 2). The primary contributors to STD-related mortality were HIV (60%) and HPV (34%). During this interval, a 49% decrease occurred in the cumulative STD-related mortality rate, from 5.3 to 2.7 deaths per 100 000 women (Figure 1).

TABLE 2—

Sexually Transmitted Disease–Related Deaths in Reproductive-Aged Women: United States, 1999–2010

Year	HIV, No.	HPV, No.	Hepatitis B and C, No.	Gonorrhea/Chlamydia, No.	Pelvic Inflammatory Disease, No.	Ectopic Pregnancy, No.	Genital Herpes, No.	Syphilis, No.	Other, No.	Total, No.
1999	2071	993	139	3	27	9	24	3	1	3270
2000	2043	995	138	3	24	12	19	3	1	3238
2001	1942	977	135	2	23	10	30	4	0	3122
2002	1898	941	145	3	25	8	21	3	1	3045
2003	1782	892	130	1	23	12	17	2	0	2858
2004	1702	900	117	1	17	7	21	4	0	2768
2005	1548	845	118	1	14	8	20	5	0	2558
2006	1463	864	101	0	17	7	16	5	0	2472
2007	1330	890	96	0	16	7	15	4	0	2358
2008	1149	820	90	1	14	9	8	1	0	2092
2009	965	787	81	0	20	10	19	3	0	1885
2010	787	779	82	1	13	10	13	0	0	1684
Total	18 679	10 680	1371	16	233	110	223	37	3	31 351

Note. HPV = human papilloma virus.

Source. Centers for Disease Control and Prevention.⁴

FIGURE 1—

Annual sexually transmitted disease–related mortality in women aged 15–44 years: United States, 1999–2010.

Note. HPV = human papilloma virus.

Between 1999 and 2010 the sexually acquired HIV mortality rate among women aged 15 to 44 years decreased by 62%, from 3.4 deaths per 100 000 women in 1999 to 1.3 deaths per 100 000 women in 2010. The mortality rate from HPV-associated gynecologic cancers (cervical, vulvar, and vaginal) decreased by 19%, from 1.6 deaths per 100 000 women in 1999 to 1.3 per 100 000 women in 2010 (Figure 1).

Acute viral hepatitis B and hepatitis C infections are sexually acquired about 26% and 9% of the time, respectively.^9–17 Although the primary infection for sexually acquired hepatitis B virus and hepatitis C virus is commonly acquired during a woman’s reproductive years, morbidity and mortality may occur much later in life. Hepatocellular carcinoma and chronic liver disease are severe complications of chronic hepatitis B and C infections. I estimated that hepatocellular carcinoma was caused by hepatitis B virus in 32% of cases and hepatitis C virus in 15% of cases.^18,19 I assumed that nonalcoholic chronic liver disease was caused by hepatitis B virus in 11% of cases and hepatitis C virus in 26% of cases.^3,20 According to these estimates, the mortality rate from sexually acquired hepatitis and sequelae decreased from 0.2 deaths per 100 000 women in 1999 to 0.1 deaths per 100 000 women in 2010 (Figure 2).

FIGURE 2—

Annual mortality rate from sexually acquired hepatitis in women aged 15–44 years: United States, 1999–2010.

Chlamydial and gonococcal infections are not commonly reported as a cause of death, but these pathogens are associated with PID and ectopic pregnancy. Approximately 65% of PID cases are attributable to 1 or both of these sexually transmitted pathogens.^27–31 I estimated that 43% of ectopic pregnancies are attributable to chlamydial or gonococcal infection.^25–28 Between 1999 and 2010, 65% of chlamydial and gonococcal infection–related deaths occurred in the setting of PID, and 31% were in the setting of ectopic pregnancy. The overall mortality rate from chlamydial and gonococcal infections, PID, and ectopic pregnancy was less than 0.1 deaths per 100 000 women in women aged 15 to 44 years. The median number of deaths was 28 per year (range = 23–40).

Syphilis-associated death among reproductive-aged women in the United States was rare. Between 1999 and 2010 a total of 37 women died from syphilis. The number of deaths remained stable with a median of 3 deaths per year (range = 0–5).

Between 1999 and 2010 a total of 223 reproductive-aged women had genital herpes simplex virus–associated deaths. The overall number of deaths each year remained stable, with a median of 19 deaths per year (range = 8–30).

DISCUSSION

With federal data on multiple causes of death and the number of women of reproductive age, I have demonstrated decreasing STD-related mortality rates among reproductive-aged women over the past decade. In 1999, STD-related mortality accounted for 6% of total mortality among reproductive-aged women. By 2010, only 4% of mortality in this age group was attributable to sexually transmitted infections.⁴

In 2010, approximately 300 000 women in the United States were living with HIV infection. Between the years 1999 and 2010, 70% to 90% of new diagnoses of HIV among women were sexually acquired. This estimate varies by population demographic and year.^24–26 The previous increase in HIV-related mortality documented by Ebrahim et al. was not observed during 1999 to 2010.³ This was likely because of the widespread application of highly active antiretroviral therapy.

More than 40 types of HPV can infect the genital area. Infection with high-risk, oncogenic HPV types, such as 16 and 18, increase a woman’s risk of anogenital cancer.³² The prevalence of high-risk HPV in the United States was 23% per a 2003 to 2005 report.³³ The highest prevalence was among reproductive-aged women. Because non–HPV-associated cervical cancer among reproductive-aged women is rare, cervical cancer in this age group is considered entirely attributable to sexually acquired high-risk HPV infection.^28–30 High-risk HPV causes approximately 50% of vulvar and vaginal cancers.³² In 2010, mortality data from the CDC demonstrated that about 17% of all cervical cancer deaths occurred among women of reproductive age. The majority of cervical cancer deaths in 2010 occurred in women older than 45 years.⁴ Cervical cancer screening programs, with consequent earlier detection and treatment, may have contributed to the decrease in cervical cancer mortality among younger women.³⁴

In 1975, the mortality rate from PID was 0.29 deaths per 100 000 women aged 15 to 44 years.¹ The PID mortality rate has decreased by 90% over the past 35 years to just a few deaths per year. This dramatic improvement is presumably because of continued efforts for chlamydia and gonorrhea screening, early diagnosis, and multiple-drug therapy.³²

This analysis has both strengths and weaknesses. Among the strengths is the use of federal, population-based data for the United States. I included the predominant contributors to STD-related death, offering a comprehensive assessment of STD-related mortality in reproductive-aged women. This study also has several weaknesses. Estimating the contribution of some infections, particularly hepatitis B and C, to STD-related mortality is difficult. The proportion of diseases that are sexually transmitted likely varies temporally and geographically. I relied on the estimates used in previous analyses and the available literature.^7–14 Using the attributable proportion of sexual transmission to estimate mortality from disease is necessarily imprecise.

Population mortality data from death certificates are inherently at risk for misclassification and underreporting.³⁵ The natural history of a disease may affect whether reporting will change over time. A disease that is more lethal will likely be more frequently reported as an underlying cause of death. However, a disease that is less lethal is likely to be underreported.³⁵ In 1999, mortality data were reported on the basis of ICD-10 codes, a change from previous years. It is unknown whether medical examiners were less likely to report an underlying cause of death in 2010 as compared with 1999. Overall, all estimates should be viewed with caution. Although some misclassification is inevitable, this is unlikely to have changed substantially over time. Hence, I believe that these temporal trends are real and not because of artifacts of reporting.

To calculate the percentage of diseases attributable to sexual transmission, I used the same estimates as those in previous analyses. This approach informs changing mortality over time. The methodological issues of this analysis are unlikely to have accounted for the large decreases in deaths observed, particularly among the primary contributors to STD-related mortality from 1999 to 2010: HIV and HPV-related genital cancer.

My results are useful as estimates of STD-related mortality to compare with previous mortality analyses.^1,3 The next step, an analysis of STD-related morbidity for 1999 to 2010 would be achieved by calculating disability-adjusted life-years.² The likelihood of sexual acquisition, morbidity, and mortality from selected diseases may be different among certain subgroups. Future analyses should evaluate temporal trends in the disability-adjusted life-years lost because of STD and identify particularly vulnerable groups of women.

Although I have described an important reduction in STD-related mortality among reproductive-aged women in the United States, I did not address deaths of children from perinatal acquisition of STDs. Diseases that are prevalent in women of reproductive age, such as N. gonorrheae, C. trachomatis, and genital herpes, are associated with STD-related neonatal complications from preterm birth, chorioamnionitis, and disseminated infection.^36–39 On the other hand, mortality from perinatal HIV transmission will presumably have decreased over the past decade because of perinatal screening and treatment.⁴⁰ Less prevalent conditions, such as syphilis, also have severe consequences. For example, the CDC reported 431 congenital syphilis cases, and 25 case-related stillbirths, in 2008.⁴¹ Future analyses should describe STD-related mortality among neonates over the past decade.

All STD-related deaths are preventable. Despite the steady decrease in STD-related deaths among US reproductive-aged women from 1999 to 2010, the fact that more than 1000 women still die of STD each year underscores the need for sustained efforts to maintain and strengthen STD control. Currently, the predominant contributors to STD-related mortality among reproductive-aged women are HIV and HPV-related genital cancers. Behavioral interventions target high-risk groups to prevent HIV transmission. Improving HPV vaccine coverage is the key to HPV-related cancer prevention but has been far below goals.⁴²Efforts to increase STD prevention should be continued.