Declines in cervical cancer incidence among young women in the United States

Abstract

We calculated age-standardized cervical cancer incidence rates and trends by histologic type among women aged 15–29 years, using cancer registry data covering approximately 99% of the US population. We examined differences in incidence rates from the earliest to the most recent study period and quantified changes over time using joinpoint regression. Incidence rates for cervical squamous cell carcinoma (SCC) and adenocarcinoma (AC) decreased significantly during 1999–2022 in all age groups (15–20, 21–24, 25–29 years). Among women aged 15–20 years, cervical SCC rates were stable from 1999 to 2010 but declined substantially by 24.2% (95% confidence interval = −44.4 to −18.0) each year from 2010 to 2022, with only a few cases reported during 2017–2022. These findings show that cervical cancer incidence has declined among adolescents and young adult women in the United States, particularly in more recent years, highlighting the population-level impact of cervical cancer prevention measures.

Approximately 12 500 cases of cervical cancer are reported each year in the United States.¹ Most of these cases could be prevented by human papillomavirus (HPV) vaccination, routine cervical cancer screening, and timely follow-up treatment. As a result, declines in the incidence of cervical precancer have been reported in the United States.^2,3 However, available research documenting declines in population-level cervical cancer incidence in the United States is limited in number,^4,5 often dated,⁶ and rarely examines trends by histology.⁴

In this study, we analyzed the latest national data to calculate the population-level incidence rate of cervical cancer among women aged 15–29 years in the United States and examined trends over time by histology. Understanding the incidence rate and trends among younger women—who are eligible for the HPV vaccination during adolescence and cervical cancer screening starting at 21 years—could help inform prevention efforts. This is especially relevant given recent diverging trends, in which use of cytology screening alone decreased from 34.2% in 2013 to 26.4% in 2019 among women aged 21–29 years,⁷ whereas HPV vaccination coverage increased from 62.8% in 2015 to 76.8% in 2020 among female adolescents aged 13–17 years.⁸

We analyzed population-based incidence data from the US Cancer Statistics (USCS) database, which includes the most recent high-quality data from central cancer registries that meet the USCS publication criteria.⁹ During 1999–2022, the database covered 98.8% of the US population. Histologically confirmed cases of invasive cervical cancer were defined according to the International Classification of Diseases for Oncology, Third Edition (ICD-O-3)¹⁰ primary site codes (C53.0-C53.9). Cases were further classified using ICD-O-3 morphology codes as squamous cell carcinoma (SCC; 8050–8086 and 8120–8131), adenocarcinoma, including adenosquamous carcinoma (AC; 8140–8575), and other histology (8000–8049, 8087–8119, 8132–8139, 8576–9992, excluding 9050–9055, 9140, 9590–9992).

Age-standardized incidence rates were calculated for women aged 15–29 years. This age group was selected to reflect eligibility for HPV vaccination and screening while minimizing statistical issues related to the low incidence at younger than age 15 years and adjustment for hysterectomy prevalence older than age 30 years. Rates were standardized to the 2000 US population using the direct method¹¹ and expressed per 100 000 women. Rate ratio (RR) and 95% confidence interval (CI) were estimated to assess the relative difference in incidence rates. Analyses were stratified into 3 age groups (15–20, 21–24, and 25–29 years). Because of sparse data among women aged 15–20 years, particularly toward the end of the study period, diagnosis years were aggregated into 2-year intervals from 1999 to 2016 and two 3-year intervals from 2017 to 2022.

Trends in incidence rates were analyzed using Joinpoint regression, and changes were quantified by annual percent change (APC) and average annual percent change (AAPC). Due to the small case counts in some groups, the weighted Bayesian information criterion method was used for model selection to determine the number and location of joinpoints where trends change. If APC or AAPC differed from zero at an alpha of 0.05, rates were considered to increase or decrease; otherwise, rates were reported as stable. Incidence rates were calculated using the SEER Stat software (version 8.4.5), and trends were analyzed using the Joinpoint Regression Program (Version 5.3.0.0).

During 1999–2022, 15 100 new cases of invasive cervical cancer were reported among US women aged 15–29 years, of which 10 601 (70.2%) were SCC and 3565 (23.6%) were AC. Both case counts and incidence rates decreased across all age groups and for cervical SCC and AC from the earliest to the most recent study period (Table 1).

Table 1.

Cervical cancer incidence rates among women aged 15–29 years, by age and histology—United States, 1999–2022.

	15–20 years						21–24 years						25–29 years
	Total		SCC		AC		Total		SCC		AC		Total		SCC		AC
Yeara	No.	Rate	No	Rate	No.	Rate	No.	Rate	No.	Rate	No.	Rate	No.	Rate	No.	Rate	No.	Rate
1999–2022	403	0.14	197	0.07	144	0.05	2327	1.18	1650	0.84	467	0.24	12370	5.01	8754	3.55	2954	1.20
1999	69	0.30	43	0.19	14	0.06	144	2.03	99	1.40	31	0.44	678	7.08	498	5.20	139	1.45
2000							131	1.81	91	1.26	23	0.32	678	7.20	478	5.08	155	1.65
2001	53	0.22	25	0.11	15	0.06	118	1.57	87	1.15	17	0.23	621	6.76	438	4.77	148	1.61
2002							144	1.85	114	1.46	22	0.28	573	6.27	400	4.38	128	1.40
2003	43	0.18	21	0.09	14	0.06	129	1.62	89	1.12	27	0.34	548	5.96	397	4.32	119	1.29
2004							125	1.55	88	1.09	26	0.32	503	5.37	357	3.81	124	1.32
2005	60	0.24	32	0.13	19	0.08	133	1.65	89	1.11	30	0.37	503	5.25	339	3.54	131	1.37
2006							116	1.44	89	1.10	21	0.26	539	5.45	359	3.63	147	1.49
2007	46	0.18	22	0.09	17	0.07	112	1.38	74	0.91	26	0.32	519	5.14	368	3.64	122	1.21
2008							106	1.30	76	0.93	21	0.26	548	5.34	376	3.66	138	1.34
2009	41	0.16	24	0.09	12	0.05	128	1.56	81	0.99	25	0.30	571	5.52	388	3.75	155	1.50
2010							105	1.27	82	0.99	18	0.22	517	4.99	354	3.42	137	1.32
2011	32	0.13	15	0.06	13	0.05	114	1.34	81	0.95	22	0.26	494	4.75	326	3.14	143	1.38
2012							119	1.37	87	1.00	22	0.25	455	4.37	324	3.11	111	1.07
2013	18	0.07	7	0.03	9	0.04	91	1.03	66	0.75	21	0.24	538	5.14	352	3.36	154	1.47
2014							96	1.08	70	0.79	21	0.24	560	5.27	410	3.86	124	1.17
2015	10	0.04	—b	—b	6	0.02	74	0.84	54	0.61	13	0.15	555	5.12	387	3.57	135	1.25
2016							79	0.91	55	0.64	15	0.17	587	5.31	430	3.89	137	1.24
2017	17	0.05	—b	—b	13	0.03	63	0.74	39	0.46	19	0.22	507	4.51	368	3.28	107	0.95
2018							56	0.66	40	0.47	15	0.18	494	4.38	362	3.21	115	1.02
2019							41	0.49	29	0.35	9	0.11	461	4.09	340	3.02	102	0.91
2020	14	0.04	—b	—b	12	0.03	39	0.47	23	0.28	8	0.10	348	3.14	255	2.30	80	0.72
2021							36	0.43	26	0.31	10	0.12	331	3.04	261	2.39	59	0.54
2022							28	0.33	21	0.25	—b	—b	242	2.24	187	1.73	44	0.41
Rate ratio: 2020–2022 vs. 1999–2000 (95% CI)	—	0.13 (0.07, 0.22)c	—	—b	—	0.53 (0.22, 1.23)	—	—	—	—	—	—	—	—	—	—	—	—
Rate ratio: 2022 vs. 1999 (95% CI)	—	—	—	—	—	—	—	0.16 (0.11, 0.25)c	—	0.18 (0.11, 0.29)c	—	—b	—	0.32 (0.27, 0.37)c	—	0.33 (0.28, 0.40)c	—	0.28 (0.20, 0.40)c

Abbreviations: AC = adenocarcinoma; CI = confidence interval; SCC = squamous cell carcinoma.

^aBecause of sparse data, diagnosis years for women aged 15–20 were grouped into 2-year intervals during 1999–2016 and two 3-year intervals during 2017–2019 and 2020–2022.

^bCounts, rates, and rate ratios not reported when based on fewer than 6 cases.

^cRate ratios differed significantly.

Among women aged 15–20 years, the total number of cases fell from 69 during 1999–2000 to 14 during 2020–2022, with fewer than 6 cases of cervical SCC reported in the latter period. In both age groups of 21–24 and 25–29 years, incidence rates decreased from 1999 to 2022. Among women aged 21–24 years, incidence rates declined from 2.03 to 0.33 per 100 000 (RR = 0.16; 95% CI = 0.11 to 0.25), and among women aged 25–29 years, rates declined from 7.08 to 2.24 per 100 000 (RR = 0.32; 95% CI = 0.27 to 0.37).

The overall incidence rate declined by 5.5% annually from 1999 to 2004, remained stable between 2004 and 2017, and decreased sharply by 12.9% per year during 2017–2022 (Figure S1). Incidence rates for cervical SCC and AC decreased from 1999 to 2022 in all age groups (Table 2). Among women aged 15–20 years, cervical SCC rates were stable from 1999 to 2010 but declined substantially by 24.2% each period from 2010 to 2022, whereas cervical AC rates decreased by 6.0% each period from 2006 to 2022. Among women aged 21–24 years, SCC rates decreased by 2.8% annually from 1999 to 2012 and sharply by 11.5% annually during 2012–2022. Among women aged 25–29 years, incidence rates of both SCC and AC declined during 2016–2022. Incidence rates and trends by histological type other than SCC and AC are presented in Table S1 and Figure S2.

Table 2.

Cervical cancer incidence rate trends among women aged 15–29 years, by age and histology—United States, 1999–2022.^a

		Trend 1		Trend 2		Trend 3		Trend 4		Overall Trendb
Age (years)	Histology	Years	APC (95% CI)	Years	APC (95% CI)	Years	APC (95% CI)	Years	APC (95% CI)	AAPC (95% CI)
15–20c	Overall	1999–2010	−5.2 (−8.8 to 8.6)	2010–2022	−14.4 (−34.0 to −10.7)d					−10.2 (−15.0 to −7.8)d
	SCC	1999–2010	−5.5 (−10.0 to 2.6)	2010–2022	−24.2 (−44.4 to −18.0)d					−15.8 (−23.1 to −13.1)d
	AC	1999–2006	2.8 (−3.3 to 17.7)	2006–2022	−6.0 (−14.4 to −4.0)d					−3.6 (−6.0 to −1.6)d
	Other	1999–2016	−9.0 (−19.2 to 2.5)	2016–2022	−94.6 (−97.7 to −78.8)d					−53.6 (−61.8 to −42.7)d
21–24	Overall	1999–2012	−2.8 (−4.1 to −0.8)d	2012–2022	−11.4 (−16.5 to −8.8)d					−6.6 (−8.2 to −5.7)d
	SCC	1999–2012	−2.8 (−4.1 to −1.1)d	2012–2022	−11.5 (−16.0 to −9.1)d					−6.7 (−8.0 to −5.8)d
	AC	1999–2017	−3.3 (−5.0 to 13.1)	2017–2022	−17.3 (−39.0 to −4.9)d					−6.6 (−10.2 to −3.1)d
	Other	1999–2022	−9.6 (−23.7 to 3.5)							−9.6 (−23.7 to 3.5)
25–29	Overall	1999–2005	−5.2 (−12.8 to −2.6)d	2005–2018	−0.9 (−1.8 to 3.7)	2018–2022	−15.7 (−22.6 to −10.5)d			−4.7 (−5.7 to −4.1)d
	SCC	1999–2005	−5.7 (−12.0 to 1.5)	2005–2012	−1.9 (−9.2 to 7.4)	2012–2016	4.6 (−14.7 to 9.8)	2016–2022	−10.6 (−15.2 to −5.5)d	−4.1 (−5.0 to −3.5)d
	AC	1999–2016	−1.1 (−2.3 to 0.6)	2016–2022	−14.4 (−25.4 to −8.4)d					−4.8 (−7.0 to −3.5)d
	Other	1999–2017	−3.6 (−5.2 to −7.1)	2017–2022	−15.9 (−33.7 to −5.1)d					−6.5 (−9.1 to −3.9)d

Abbreviations: AAPC = average annual percent change; AC = adenocarcinoma; APC = annual percent change; CI = confidence interval; SCC = squamous cell carcinoma.

^aYear 2020 was excluded from the model due to interruptions from COVID-19.

^bOverall trend includes years 1999–2022.

^cBecause of sparse data, diagnosis years for women aged 15–20 were grouped into 2-year intervals during 1999–2016 and two 3-year intervals during 2017–2019 and 2020–2022.

^dDifferent from zero at an alpha of 0.05.

In this investigation of population-based cervical cancer incidence over 24 years, both the number of cases and incidence rate declined among US adolescents and young adult women, with the steepest reductions in more recent years. Declines were evident across histology types, with a substantial reduction in SCC among women aged 15–20 years.

In the United States, HPV vaccination has been routinely recommended for girls aged 11–12 years (with catch-up through age 26 years) since 2006.¹² Notable reductions in cervical cancer incidence following the introduction of the HPV vaccine, especially among vaccine-eligible cohorts, may reflect population-level effects of vaccination. These patterns are encouraging and mirror the decreasing trends in HPV prevalence,¹³ cervical precancers,² invasive cancers,⁴ and mortality.¹⁴

Cervical cancer screening has not been recommended for women younger than 21 years since 2009 by the American College of Obstetricians and Gynecologists (ACOG)¹⁵ and since 2012 by the US Preventive Services Task Force.¹⁶ Delaying the recommended screening initiation age to 21 years may offer an alternative explanation for the declining cervical cancer incidence among those aged 15–20 years, as fewer cancers would be diagnosed in this younger age group. In such a case, we might expect to observe increasing incidence rates among women aged 21–24 and 25–29 years. However, we did not observe an increase in incidence rates; instead, rates either declined or remained stable. For example, overall cervical cancer and cervical SCC rates among women aged 21–24 years declined throughout the study period, decreasing by more than 11% per year since 2012. Furthermore, cervical AC incidence rates among women aged 15–20 years began declining around the time the HPV vaccine was introduced. Since cervical ACs are not as easily detected by cytology because they are generally located higher in the endocervix,^17,18 this decline is unlikely due to screening.

Among women aged 21–29 years, cytology-based screening was recommended during the study period; however, the recommended frequency and screening uptake in the age group declined.⁷ The uptake of primary HPV testing among women aged 25–29 years, which has been recommended by the American Society for Colposcopy and Cervical Pathology since 2015¹⁹ and ACOG since 2016,²⁰ remained below 1% in the United States.^7,21,22 This suggests that a shift from cytology to HPV testing is unlikely to meaningfully affect cervical cancer screening rates among women younger than 30 years. Reduced screening use may lower the detection of invasive cancer in the short term.

The findings in this study are subject to at least 5 limitations. First, low case counts observed among women aged 15–20 years could lead to increased variability and unstable trends. However, trend results remained consistent with alternate models and tests. Second, our study focused on population-level cervical cancer incidence rates and trends, primarily due to data availability and constraints. Although it would have been ideal to examine incidence rates by vaccination or screening status or both, national-level linkage of cancer incidence data with immunization registries, screening records, or other risk factor data is not routine. Therefore, cervical cancer incidence rates could not be examined by vaccination history, screening status, or related factors, limiting our ability to characterize these relations and to account for potential confounding or effect modification. Third, the impact of cervical cancer screening on incidence is complex, but changes in screening initiation age, modalities, and uptake over the study period may have influenced incidence rates. Fourth, cervical cancer incidence rates were not adjusted for hysterectomy; however, the impact is likely minimal given the low prevalence of hysterectomy among younger women (0.6% among those aged 18–29 years).²³ Fifth, incidence rates for 2020 were excluded from trend analyses because of reduced diagnoses during the COVID-19 pandemic.²⁴

In this study of population-based cancer registry data, national cervical cancer incidence rates declined in recent years, particularly among women aged 15–20 years, possibly reflecting the effect of HPV vaccination. Improvement of HPV vaccination and cervical cancer screening coverage among eligible groups and timely follow-up and treatment could further reduce incidence rates, moving the nation closer to meeting the global target for cervical cancer elimination.

Supplementary Material

Supplementary material is available at JNCI: Journal of the National Cancer Institute online.

Data availability

We analyzed de-identified cancer incidence data from the United States Cancer Statistics (USCS). Investigators can access the USCS public use databases after completing data-use agreements.