Year: 2025
Type: article
Abstract: Our brief report aimed to compare metro/non-metro incidence in Human Papillomavirus (HPV)-associated cancers and identify states with diverging metro/non-metro trends. Nationally, the growing metro/non-metro disparities appear driven by accelerated incidence of HPV-associated oropharyngeal cancer in males and stagnate rates of cervical cancer in females.1 Further complicating the trends, recent data from 2021 revealed that the incidence of HPV-associated cancer was significantly higher in small rural than in non-metro urban communities (two groups often conflated as “rural”).2 Generally, rural areas are often characterized by low population density and distance from major population centers, but here “non-metropolitan urban” refers to smaller cities or towns outside major metropolitan regions. This distinction helps capture varying degrees of remoteness and access to services, which is important for understanding disparities in HPV-associated cancer. Subsequent research has yet to confirm whether this finding reflects a true decoupling between rural and non-metro urban rates, or noisy data in a small population. Most importantly, these regional analyses have largely ignored how HPV-associated cancer trends varied by metro/non-metro status across states. With authority to set health policy and cancer plan agendas, state policymakers may be more motivated to rapidly implement effective cancer prevention and control strategies in states with accelerating metro/non-metro disparities in HPV-associated cancer. Identifying states where non-metro incidence of this preventable cancer is declining could fuel efforts to disseminate effective strategies across states. We analyzed the newly released (April 2025) cancer incidence data from the Surveillance, Epidemiologic, and End Results (SEER) program.3 This SEER-21 (November 2024 submission) data include all cancer diagnoses from 21 cancer registries. We restricted incidence rates to cancers associated with HPV (oropharyngeal squamous cell carcinoma, anal and rectal squamous cell carcinoma, vulvar squamous cell carcinoma, vaginal squamous cell carcinoma, cervical carcinoma, penile squamous cell carcinoma).4 See the supplemental file (Appendix S1) for full histology codes and restriction criteria. Given our focus on rurality, we also excluded incidence data from the Alaska Native cancer registry, which does not reveal rural status. The final data included incidence rates from 13 states (CA, CT, GA, ID, IL, IA, KY, LA, NJ, NM, NY, TX, and UT) and the Seattle Puget Sound cancer registry. The data spanned years 2000-2022. See supplemental file (Appendices S2–S4) for the final data. Age-adjusted incidence rates were reported as cases per 100 000 population. Because of the differences in HPV etiology for specific types of cancer as related to sex, as well as the evidence of certain cancers contributing to sex-specific metro/non-metro disparities, incidence rates were stratified by sex.1,4 Our subgroup variable was SEER’s Rural–Urban Continuum Code (RUCC) 2013.5,6 Using the RUCC definitions, we stratified each sex-specific incidence rate into three mutually exclusive groups: “metro” counties (RUCC 1-3), non-metro “Urban” counties (RUCC 4-7), and non-metro “Rural” (RUCC 8-9). First, we extracted annual incidence rates for each year. Then, within the SEER*Stat program, we calculated incidence rate ratios (IRR) comparing each year’s metro incidence with urban and rural incidences. Utilizing the Tiwari correction method to estimate valid confidence intervals (CI) of incidence rates, we separately tested if the metro rate was statistically significantly different than the urban and rural rates.7-9 To assess how HPV-associated incidence rates have changed over time, within SEER*Stat, we calculated annual percentage change (APC) from 2000 to 2022 using the weighted-least-squares approach.10,11 Here, we conducted two tests of statistical significance. First, we used the result of the APC calculation, with Tiwari Correction, to test if each APC was statistically different than zero. Second, we compared the 95% CI of the APC to test if the metro APC was statistically different than the urban APC and rural APC. Finally, we compared sex-specific incidence rate growth (APCs) across these rural categories and states. The analytic sample included 401 536 HPV-associated cancer cases. Between 2000 and 2022, the overall incidence of HPV-associated cancer increased 0.5% annually. Figure 1A visualizes the incidence rate trends, by sex and rural status. HPV-associated cancer incidence trends. Figure 1 (top row) visualizes the age-adjusted HPV-associated cancer incidence rates for males and females, by rural status. Figure 1 (bottom row) visualizes the statistically significant annual percentage changes (APC) in HPV-associated cancer incidence by sex, rural status, and state. This row excludes APCs, which were suppressed due to low case counts or if APCs were not statistically different from 0 based on weighted-least-squares approach method and Tiwari correction. Rural status based on rural–urban continuum codes (RUCC). Metro = RUCC 1-3, urban = RUCC 4-7, rural = RUCC 8-9. In 2000, the male incidence rate (cases per 100 000 population) of HPV-associated cancer did not differ by rural status (metro = 7.3, CI = 7.1, 7.6; urban = 7.9, CI = 7.2, 8.6; rural = 8.0, CI = 6.0, 10.4) (Table 1). Between, 2000-2022, the incidence rate for HPV-associated cancer in males increased significantly across all groups. However, the metro APC of 1.5% (CI = 1.3, 1.8) was significantly smaller than the urban APC of 2.7% (CI = 2.3, 3.0) and the rural APC of 3.6% (CI = 2.6, 4.7). This growth resulted in statistically significantly lower HPV-associated cancer incidence in metro counties (10.4 CI = 10.2, 10.7) than in urban counties (12.6 CI = 11.8, 13.4) in 2022. The 2022 incidence rate ratio for the urban, compared to metro, rate in males was statistically significant (IRR = 1.21; CI = 1.1, 1.3). There was no difference between the 2022 metro and rural (11.5 CI = 9.3, 14.1) rate. HPV-associated cancer incidence rates and trends (2000-2022), by sex and rural status. Incidence rates are reported as annual percentage change (APC) uses weighted least-squares method (2000-2022). 95% CIs are reported in brackets. Incidence rate ratio test-statistic reveals if the urban or rural incidence rate was significantly different from the metro rate. Rural status based on Rural–Urban Continuum Codes (RUCC). Metro = RUCC 1-3, urban = RUCC 4-7, rural = RUCC 8-9. P < .01. P < .001. P < .0001. APC was statistically different than the metro APC based on 95% CI. In every state, the incidence of male HPV-associated cancer increased between 2000 and 2022 (Figure 1B; Table S1). However, these APCs differed by state and rural status, ranging from a 1.1% (CI = 0.8, 1.4) increase in metro California to a 6.1% (CI = 3.5, 8.7) increase in rural Iowa. In 2000, the female incidence rate (cases per 100 000 population) of HPV-associated cancer did significantly differ between metro and non-metro urban, as well as among rural counties (metro = 14.5 CI = 14.2, 14.8; urban = 16.2 CI = 15.3, 17.2). The 2000 incidence rate ratio for the urban, compared to metro, rate in males was statistically significant (IRR = 1.12, 1.1, 1.2). There was no difference between the metro and rural rates (14.2 CI = 11.6, 17.5). Between 2000 and 2022, the incidence of HPV-associated cancer in females decreased in metro counties (APC = −0.3%; CI = −0.4, −0.1). Conversely, between 2000 and 2022, the incidence of HPV-associated cancer in females increased in urban (APC = 0.7%; CI = 0.4, 1.0) and rural (APC = 1.2%; CI = 0.6, 1.9) counties. This divergence resulted in a wider gap between metro (12.5 CI = 12.3, 12.8) incidence and urban incidence (16.6 CI = 15.6, 17.5) in 2022. The incidence rate ratio comparing urban to metro counties increased to 1.32 (CI = 1.2, 1.4) in 2022. Also, the 2022 incidence for rural (17.1 CI = 14.0, 20.7) was statistically significantly larger than the metro rate based on the 2022 incidence rate ratio (IRR = 1.37; CI = 1.12, 1.66). Contrary to the consistent growth observed in male incidence, the change in incidence of HPV-associated cancer in females differed across states and rural status (Figure 1B; Appendix S5). From 2000 to 2022, the incidence of HPV-associated cancer in females declined in metro counties in California, Hawaii, Illinois, New Jersey, and New York, as well as urban Hawaii—which showed the largest decline (APC = −2.5%; CI = −4.7, −0.2). Conversely, incidence increased in urban counties in Connecticut, Iowa, Kentucky, Louisiana, New Mexico, and Texas, as well as rural Iowa, Kentucky, Texas, and Louisiana—which showed the largest increase (APC = 3.7%; CI = 0.8, 6.7). As a descriptive analysis, we did not attempt to explain these diverging metro/non-metro trends. We also did not attempt to understand intrastate variation in HPV-associated cancer rates (ie, racial/ethnic or socioeconomic status subgroups). While we analyzed the latest data, the three-year delay in SEER data availability limits our ability to inform cancer control policies in real-time. Finally, the limited counts of HPV-associated cancer in rural (RUCC 8-9) counties yielded wide CIs and imprecise test statistics, ultimately limiting our ability quantify statistically significant differences between non-metro urban and rural rates (even if such differences exist). Our study reaffirmed the persistent, diverging incidence of HPV-associated cancer in non-metro America. Prioritizing effective risk reduction policies should be considered especially urgent in states with the > 3% annual growth in non-metro HPV-associated cancer incidence (Illinois, Iowa, Kentucky, Louisiana, Texas). We predict that the divergence in rates and rising trends in non-metro regions are due to stagnate declines in smoking rates and lower adherence to recommended preventative measures. For males, reducing HPV-associated cancer will require state and local health systems to continue investing in oropharyngeal prevention and control activities. HPV-associated cancer prevention advocates and providers cannot ignore the role of smoking, especially as smoking prevalence remains high in rural males.12,13 While HPV is a crucial part of the etiology of HPV-associated oropharyngeal cancer, other prevention activities such as smoking and tobacco cessation could further reduce this burden.14,15 Although the future generation of non-metro males could benefit from policies increasing adherence to HPV vaccination recommendations, the current generation could reduce their risk of HPV-associated cancer if they participated in programs to stop smoking cigarettes.16 While non-metro females could also benefit from other risk reduction strategies, given the lower prevalence of smoking and earlier age of onset for cervical cancer, females stand to benefit most from rapid uptake in HPV vaccine efforts and secondary prevention screening services.12,13,17 Successful approaches have been extensively investigated for improving adherence to vaccination and screening recommendations in non-metro females.18-20 From tobacco taxes and cessation to vaccination and cervical cancer screening, states have pursued different approaches preventing and controlling to HPV-associated cancer.21-25 This vast landscape of state policies influencing current and future risk of HPV-associated cancer may be contributing to the diverging metro/non-metro trends. We need rapid diffusion of policies improving primary and secondary prevention practices in non-metro communities. Without such urgency, non-metro males and females will continue shouldering a growing burden of HPV-associated cancer while their metro neighbors potentially realize eliminating these preventable, yet lethal cancers. Jason Semprini (Conceptualization [equal], Data curation [equal], Formal Analysis [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Software [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]) Supplementary material is available at AJE Advances: Research in Epidemiology online. None. None to disclose. Included as supplemental file.
Author: Jason Semprini
Institution: Des Moines University
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $2,625
OpenAlex ID: W4412767850