Risk Behaviors Correlate with Higher Prevalence of Papanicolaou,Human Papillomavirus,and Human Immunodeficiency Virus Screening Among Women in the United States

Abstract

Introduction:

This study assesses whether women with human immunodeficiency virus (HIV) risk behavior have higher Papanicolaou (Pap), human papillomavirus (HPV), and HIV testing, and whether the level of selected variables associated with HIV risk behavior correlate with greater testing. Association between HIV risk situations and HPV vaccination is also evaluated.

Methods:

A cross-sectional assessment was performed in women at age 18 years and older completing the 2018 Behavioral Risk Factor Surveillance System (BRFSS) survey. Independent variables considered and adjusted for, included age, race/ethnicity, marital status, education, annual household income, smoking status, and health care status.

Results:

Prevalence of a Pap test in the past 3 years was 66.2%, of HPV test in the past 5 years was 40.2%, and of HIV test ever was 41.9%. HIV risk situations applied to 4.9% women (15.2% in ages 18–24, 7.2% in 25–44, 1.9% in 45–64, and 0.6% in 65 years and older). Adjusted odds (95% confidence interval) of a Pap, HPV, or HIV test according to HIV risk behavior status were 1.5 (1.3–1.8), 1.6 (1.4–1.8), and 2.6 (2.3–2.9), respectively. The positive association between HIV risk behavior and Pap testing depends on marital status. HIV risk behavior significantly correlates with several variables, which, in turn, correlate with testing. There was no association between HIV risk behavior and HPV vaccination.

Conclusions:

Women with HIV risk behavior are more likely to pursue Pap, HPV, and HIV testing. The significant positive associations are largest for HIV testing and smallest for Pap testing, after adjustment for the selected variables. HIV risk behavior is not associated with HPV vaccination.

Introduction

Human papillomavirus (HPV) is the primary etiological determinant of cervical cancer.^1,2 Approximately 90.6% of cervical cancer in the United States is attributed to a strain of HPV.³ About 80% of sexually active women will contract the HPV infection by age 50.⁴ Although most women will develop an immune response and clear the infection, some will not but rather experience an increased risk of dysplasia and carcinoma.⁵

The highest incidence rates of HPV and cervical cancer in the world are in sub-Saharan Africa.⁶ Diseases associated with HPV are much greater in areas with higher levels of human immunodeficiency virus (HIV).⁷ Like HPV, women infected with HIV are at an increased risk of cervical cancer.⁸ One study estimated that HIV patients are three times more likely to be diagnosed with cervical cancer.⁹

Infection with HIV compromises the immune system such that it is more difficult for the body to fight viral infections that can lead to cancer.^10,11 For example, HIV patients are twice as likely to have HPV infection and half as likely to clear the HPV infection.¹² There is also evidence that HPV infection may increase the risk of HIV.^13,14

In an effort to detect precancerous and cancerous lesions in the cervix, the Papanicolaou (Pap) test is recommended, with specific guidelines provided by the American Cancer Society.¹⁵ The US Preventive Services Task Force recommends Pap testing every 3 years for women 21–65 years of age. In addition, HPV testing is also recommended, but every 5 years for women 30–65 years of age.¹⁵ Because of the relationship among HIV, HPV, and cervical cancer, HIV testing may also help identify precancerous lesions.¹⁶ Prevention efforts to avoid HPV and HIV together include receiving the HPV vaccine and preexposure prophylaxis (PrEP). HPV vaccines are currently the only way to prevent infection of certain types of HPV, if received before being exposed to the virus.^17,18 HIV infection may be further reduced by PrEP among sexually active or injection drug users.

Given HPV and HIV are both infections that can be transmitted sexually and are associated with increased risk for cervical cancer,^19
–21 efforts to prevent or identify and treat these infections is important. Monitoring the prevalence of Pap, HPV, and HIV testing, and HPV vaccination can help inform strategies for increasing their utilization and preventing cervical cancer. A recent study used the Behavioral Risk Factor Surveillance System (BRFSS) data to evaluate trends in Pap testing and HPV vaccination in the United States from 2007 to 2016.²² Decrease in Pap testing and increase in HPV vaccination rates were observed over the 10-year period, overall and within age and race subgroups, indicating a movement from secondary to primary prevention. As HPV vaccination reduces the risk of cervical cancer, vaccinated women may start screening at later ages and less frequently.²³

This study describes the prevalence of Pap, HPV, and HIV testing according to selected demographic and lifestyle variables. Of interest is whether women with HIV risk behavior have higher testing, and whether the level of the selected variables associated with HIV risk behavior correlate with greater testing. We also explore the association between HIV risk behavior and HPV vaccination.

Methods

Study population

The study population is adult women in the United States, at age 18 years and older. This population represents a diverse range of culture, race, religion, age, education, and socioeconomic status. As these factors associate with sexual risk behaviors, they, in turn, may influence screening practices for cervical cancer and HIV.

Data source

The BRFSS is a project wherein American states, the District of Columbia, and certain territories collaborate with the Center for Disease Control and Prevention (CDC). The BRFSS involves ongoing health-related telephone surveys that are designed to collect information about health risk and screening behaviors, chronic health conditions, and access to preventive services. Survey information was intended to be used in public health research and for developing and monitoring programs and influencing health policies. The BRFSS is an annually conducted survey administered to noninstitutionalized adults in the United States, at age 18 years and older.²⁴

This study is based on data from the 2018 BRFSS survey. In this survey, health departments in all 50 states, the District of Columbia, Guam and Puerto Rico participated in asking questions related to cervical cancer screening. Institutional review board approval was not required as the study used a publicly available dataset.

Selection criteria

Analyses were based on all available data on our topic for women at age 18 years and older who completed the BRFSS survey in 2018. There were 238,911 women who completed the BRFSS during the year. Each state, the District of Columbia, Guam, and Puerto Rico included questions about Pap, HPV, and HIV testing, as well as HIV risk behavior. Only eight states (Alabama, Connecticut, Hawaii, Mississippi, Missouri, New Jersey, Tennessee, and Texas) included questions about HPV vaccination. Other variables considered for the women in this study were age, race/ethnicity, marital status, education, annual household income, smoking status, and body mass index (BMI) weight classifications.

Variables

Primary outcome variables involved the use of the Pap test, HPV test, HPV vaccine, HIV test, and HIV risk behaviors. The BRFSS survey asked a series of questions related to these variables, based on guidelines put forth by the US Preventive Services Task Force on Pap and HPV testing, and the CDC on HIV testing and HPV vaccination.^15

–18 These variables were determined as follows:

Questions asked about the Pap test were “Have you ever had a Pap test?” and “How long has it been since you had your last Pap test?” With these two questions, we created a variable indicating whether they had a Pap test within the past 3 years.

The question about the HPV test began with “An HPV test is sometimes given with the Pap test for cervical cancer screening.” Then, the women were asked “Have you ever had an HPV test?” For those who had an HPV test, the question followed “How long has it been since you had your last HPV test?” With these two questions, we created a variable indicating whether they had an HPV test within the past 5 years.

Participants were told that a vaccine to prevent the human papillomavirus or HPV infection is available and is called the cervical cancer or genital warts vaccine, HPV shot. They were then asked, “Have you ever had the HPV vaccination?”

High-risk situations for HIV were determined by asking whether any of the following applied: “You have injected any drug other than those prescribed for you in the past year.” “You have been treated for a sexually transmitted disease or STD in the past year.” “You have given or received money or drugs in exchange for sex in the past year.”²⁵

Explanatory variables included in this study were age, race/ethnicity, marital status, education, income, smoking status, BMI weight classification, and health care coverage. Age was classified as 18–24, 25–44, 45–64, and ≥65 years; race/ethnicity was classified as white, non-Hispanic, black, non-Hispanic, Hispanic, and other race, non-Hispanic; marital status was classified as married, divorced, widowed, separated, never married, and member of an unmarried couple; education was classified as <high school, high school, some college, and college; annual household income ($) was classified as <25,000, 25,000–49,999, and ≥50,000; smoking status was classified as current smoker, every day, current smoker, some days, former smoker, and never smoker; BMI (kg/m²) weight classification was underweight (<18.5), normal weight (18.5 to <25), overweight (25 to <30), and obese (≥30); and health care coverage was classified as “Yes” versus “No.” These variables were included in the study because of their possible relationship with the outcome variables, as indicated by previous studies.^{26

–39}

Statistical techniques

Data were described using frequencies and percentages. Multiple logistic regression models were used to examine the association between HIV risk behavior and Pap, HPV, and HIV testing, as well as between HIV risk behavior and HPV vaccination, based on estimated odds ratios. Adjusted odds ratios accounted for age, race, marital status, education, annual household income, smoking, BMI, and health care coverage. All statistical analyses corrected for sampling weights in order to generate population estimates. Statistical analyses were performed using SAS 9.4 (2012; SAS Institute, Cary, NC). SAS procedures used in this study were SURVEYFREQ and SURVEYLOGISTIC. Graphs were created in Microsoft Excel, 2016.

Results

Mean age of the participants was 48.4 (standard deviation = 19.5) years. Median age was 48 years. Participants were most likely non-Hispanic whites (75.7%), married (48.5%), college graduates (37.4%), had an income of at least $50,000, never smokers (61.7%), normal weight (35.3%), and had health care coverage (Table 1).

Table 1.

Characteristics of Adult Women in the United States

	n	%	Weighted %
Age (years)
18–24	11,915	5.09	11.88
25–44	50,851	21.74	32.85
45–64	82,882	35.43	32.53
≥65	88,300	37.74	22.75
Missing	4963
Race/ethnicity
White, NH	177,864	75.68	62.68
Black, NH	21,678	9.22	12.49
Hispanic	19,762	8.41	16.75
Other race only, NH	11,260	4.79	6.77
Multiracial, NH	4453	1.89	1.3
Missing	3894
Marital status
Married	115,056	48.5	49.1
Divorced	34,752	14.65	11.48
Widowed	39,521	16.66	10.15
Separated	5470	2.31	2.73
Never married	34,762	14.65	22.02
Member of an unmarried couple	7649	3.22	4.53
Missing	1701
Education
<High school	16,993	7.13	12.49
High school	63,555	26.68	26.35
Some college	68,601	28.8	32.36
College graduate	89,027	37.38	28.79
Missing	735
Annual household income ($)
<25,000	56,117	29.22	30.02
25,000–49,999	47,910	24.94	23.22
≥50,000	88,052	45.84	46.77
Missing	46,832
Smoking status
Current smoker, every day	22,040	9.58	9.61
Current smoker, some days	8675	3.77	3.91
Former smoker	57,298	24.92	20.73
Never smoker	141,953	61.73	65.75
Missing	8945
BMI weight classification
Underweight	4662	2.19	2.42
Normal weight	74,999	35.26	36.81
Overweight	65,320	30.71	29.50
Obese	67,745	31.85	31.28
Missing	26,185
Health care coverage
Yes	221,766	93.14	89.34
No	16,330	6.86	10.66
Missing	815

BMI, body mass index; NH, non-Hispanic.

Prevalence estimates of Pap <3 years, HPV <5 years, and HIV ever are given in Table 2. The table also provides prevalence of HPV vaccination and HIV risk behaviors. Prevalence of the Pap test was ∼1.6 times (60%) greater than that of HPV or HIV tests.

Table 2.

Distribution of Adult Women in the United States According to Pap, HPV, and HIV Testing, HPV Vaccination, and HIV Risk Behavior

	n	%	Weighted %
Pap test
<3 Years	133,197	61.12	66.16
≥3 Years	71,237	32.69	23.32
Never	13,502	6.2	10.52
Missing	20,975
HPV test
<5 Years	52,771	32.96	40.25
≥5 Years	13,349	8.34	7.33
Never	93,964	58.70	52.41
Missing	78,827
HPV vaccine
Yes	1556	19.71	21.60
No	6337	80.29	78.40
Missing	231,018
HIV test
Ever	72,774	34.05	41.90
Never	140,951	65.95	58.10
Missing	25,186
HIV risk situations apply
Yes	6825	3.08	4.86
No	214,871	96.92	95.14
Missing	17,215

HPV, human papillomavirus; HIV, human immunodeficiency virus; Pap, Papanicolaou.

Source: Behavioral Risk Factor Surveillance System (BRFSS, 2018).

Prevalence estimates of Pap, HPV, and HIV testing, HPV vaccination, and HIV risk behavior are given according to selected variables in Table 3.

Table 3.

Association Between Pap, HPV, and HIV Testing, HPV Vaccination, with HIV Risk Behavior and Selected Variables

	Pap <3 years, %	aOR	95% CI	HPV <5 years, %	aOR	95% CI	HPV vaccine, %	aOR	95% CI	HIV ever, %	aOR	95% CI	HIV risk behavior, %	aOR	95% CI
Age (years)
18–24	44.54	—		33.37	—		44.63	—		35.29	—		15.24	—
25–44	83.13	4.41	4.00–4.89	60.84	2.46	2.20–2.76	17.97	0.34	0.23–0.49	61.88	2.81	2.53–3.12	7.2	0.49	0.42–0.58
45–64	73.05	2.20	1.98–2.44	39.85	1.05	0.93–1.18	4.54	0.07	0.03–0.15	43.25	1.37	1.22–1.52	1.93	0.13	0.11–0.16
≥65	41.97	0.65	0.58–0.72	12.72	0.26	0.23–0.30				15.12	0.35	0.31–0.39	0.6	0.05	0.03–0.08
p	<0.0001			<0.0001			<0.0001			<0.0001			<0.0001
Race/ethnicity
White, NH	63.59	—		37.2	—		24.14	—		35.29	—		4.26	—
Black, NH	74.62	2.01	1.84–2.20	49.44	1.67	1.53–1.82	20.32	0.84	0.57–1.24	65.56	3.67	3.39–3.98	6.62	1.05	0.89–1.25
Hispanic	72.41	1.67	1.52–1.83	45.74	1.41	1.28–1.55	16.68	0.73	0.48–1.11	50.88	1.83	1.68–1.99	5.82	1.02	0.87–1.21
Other, NH	60.4	0.71	0.62–0.82	35.34	0.73	0.62–0.86	23.45	0.89	0.53–1.49	34.99	0.84	0.73–0.98	4.68	0.87	0.61–1.22
Multiracial, NH	66.07	1.13	0.96–1.32	49.84	1.45	1.23–1.70	24	0.70	0.41–1.17	57.92	2.16	1.86–2.50	9.19	1.35	1.03–1.75
p	<0.0001			<0.0001			0.0035			<0.0001			<0.0001
Marital status
Married	72.7	—		41.55	—		14.23	—		38.35	—		2.13	—
Divorced	65.02	0.84	0.80–0.92	41.71	1.24	1.14–1.34	11.19	0.70	0.45–1.07	51.65	1.81	1.67–1.97	4.77	2.49	2.09–2.97
Widowed	42.2	0.66	0.61–0.71	15.01	0.73	0.65–0.92	13.79	1.31	0.50–3.39	19.23	0.90	0.81–0.99	1.25	1.31	0.89–1.92
Separated	73.61	1.02	0.87–0.18	48.5	1.34	1.14–1.58	19.15	2.36	1.09–5.14	58.65	1.65	1.41–1.93	7.36	2.59	2.01–3.33
Never married	60.42	0.68	0.63–0.74	43.68	0.99	0.91–1.08	32.89	2.06	1.45–2.93	49.32	1.24	1.15–1.34	11.36	2.40	2.06–2.80
Unmarried couple	73.18	1.01	0.89–1.16	53.17	1.31	1.13–1.51	30.45	2.58	1.45–4.59	58.88	1.66	1.47–1.86	10.67	2.42	1.97–2.98
p	<0.0001			<0.0001			<0.0001			<0.0001			<0.0001
Education
<High school	60.3	—		32.04	—		11.82	—		40.15	—		4.61	—
High school	58.18	1.01	0.91–1.13	33.78	1.14	1.01–1.29	22.4	1.25	0.67–2.32	36.27	1.02	0.92–1.14	5.21	0.91	0.71–1.17
Some college	64.83	1.20	1.07–1.33	41.24	1.40	1.24–1.58	25.31	1.58	0.86–2.90	43.57	1.40	1.26–1.57	5.57	1.09	0.84–1.40
College graduate	77.18	1.74	1.56–1.95	49.03	1.71	1.51–1.94	21.25	1.64	0.88–3.07	45.93	1.63	1.45–1.82	3.87	1.12	0.85–1.47
p	<0.0001			<0.0001			0.0002			<0.0001			<0.0001
Annual household income ($)
<25,000	60.96	—		38.4	—		19.93	—		47.05	—		6.7	—
25,000–49,999	63.62	1.03	0.96–1.10	40.27	1.04	0.96–1.13	23.59	1.28	0.88–1.87	41.8	0.83	0.77–0.89	5.59	0.95	0.82–1.11
≥50,000	74.61	1.21	1.13–1.31	46.82	1.15	1.06–1.24	23.11	1.44	1.00–2.09	43.14	0.85	0.79–0.92	3.75	0.83	0.72–0.96
p	<0.0001			<0.0001			0.3105			<0.0001			<0.0001
Smoking status
Daily smoker	62.56	—		45.41	—		21.99	—		56.09	—		9.35	—
Occasional smoker	69.63	1.40	1.22–1.61	49.38	1.13	0.97–1.32	21.2	0.86	0.32–2.30	58.74	1.06	0.93–1.20	9.31	0.97	0.79–1.19
Former smoker	64.33	1.27	1.17–1.39	37.54	0.95	0.86–1.05	19.92	0.89	0.54–1.48	42.18	0.84	0.77–0.92	3.84	0.71	0.60–0.85
Never smoker	67.04	1.22	1.15–1.34	39.79	0.81	0.75–0.88	21.88	0.83	0.54–1.27	38.77	0.49	0.45–0.53	4.28	0.43	0.37–0.50
p	<0.0001			<0.0001			0.9274			<0.0001			<0.0001
BMI weight classification
Underweight	51.99	0.74	1.15–1.34	34.32	0.90	0.75–1.09	42.5	1.91	0.78–4.67	37.33	0.88	0.75–1.03	7.37	1.14	0.84–1.53
Normal weight	65.19	—		40.56	—		25.61	—		39.87	—		5.3	—
Overweight	67.23	1.09	1.03–1.16	39.47	0.97	0.91–1.04	21.93	1.08	0.78–1.51	41.27	1.01	0.95–1.08	4.18	0.97	0.84–1.11
Obese	66.84	0.95	0.90–1.02	41.49	0.94	0.88–1.01	17.44	0.89	0.65–1.23	45.99	1.02	0.96–1.09	5.17	1.13	0.99–1.29
p	<0.0001			<0.0001			<0.0001			<0.0001			<0.0001
Health care coverage
Yes	66.99	—		40.48	—		23.53	—		41.25	—		4.54	—
No	59.82	0.50	0.45–0.55	38.94	0.68	0.62–0.76	15.27	0.88	0.56–1.38	47.97	0.87	0.79–0.96	7.64	1.17	0.97–1.39
p	<0.0001			0.2529			0.0013			<0.0001			<0.0001

Bolded aORs are significant at the 0.05 level. The p-values are based on the Rao–Scott chi-square. NH other race consists of 41.21% Asian, 36.38% American Indian/Alaskan Native, and 22.42% other race.

aOR, adjusted odds ratio; CI, confidence interval.

Source: Behavioral Risk Factor Surveillance System (BRFSS, 2018).

Age

The prevalence of Pap, HPV, and HIV testing was greatest in the age range 25–44 years. Although testing dropped with older age, 42.0% of women received a Pap <3 years who were at least 65 years of age. HPV vaccination and HIV risk were greatest in the age range 18–24 years.

Race/ethnicity

Prevalence of Pap testing was greatest in non-Hispanic blacks and Hispanics. Prevalence of HPV and HIV testing was greatest in non-Hispanic blacks, Hispanics, and multi-race. Non-Hispanic other race has the lowest levels of testing. HIV risk was greatest among non-Hispanic blacks and multi-race, but the variables adjusted for in the multiple regression model accounted for higher HIV risk behavior in blacks, but not in multi-race. HPV vaccination was greatest among non-Hispanic whites and multi-race and lowest among non-Hispanic blacks and Hispanics. In the age group 18–24 years, the prevalence of HPV vaccination was 53.9% for non-Hispanic whites, 32.3% for non-Hispanic blacks, 35.2% for Hispanics, 44.9% for non-Hispanic other race only, and 35.7% for non-Hispanic multiracial. Racial/ethnic differences in HPV vaccination were explained by other variables in the multiple regression model.

Marital status

Prevalence of the Pap test was greatest in women who are married, separated, or part of an unmarried couple; prevalence of the HPV test was greatest for women who are divorced, separated, or part of an unmarried couple; and prevalence of the HIV test was greatest in women who are divorced, separated, never married, or part of an unmarried couple. Prevalence of HPV vaccination was greatest in women who are separated, never married, and members of an unmarried couple. HIV risk was greatest in women who are not married.

Education

Prevalence estimates of Pap, HPV, and HIV were positively associated with education. HPV vaccination increases with education. College graduates have the lowest HIV risk behavior. The lower HIV risk behavior in college graduates is explained by the other variables in the multiple regression model.

Annual household income

Prevalence estimates of Pap and HPV testing were greatest in those in the highest income group. On the contrary, HIV testing and HIV risk behavior were greatest in the lowest income group. HPV vaccination was not significantly associated with income.

Smoking status

Prevalence of Pap testing was significantly lower among current daily smokers. Prevalence of HPV and HIV testing were higher among smokers. Smoking status was not significantly associated with HPV vaccination. Women with HIV risk behaviors were also significantly more likely to smoke.

BMI weight classification

Underweight women have the lowest prevalence of Pap, HPV, and HIV testing, but the highest level of HPV vaccination and HIV risk. Weight was only significant in the model involving Pap testing, after adjustment for other variables. Specifically, compared with women of normal weight, women who are underweight are less likely to have a Pap test and women who are overweight are more likely to have a Pap test.

Health care coverage

Health care coverage was associated with greater Pap and HPV testing, but lower HIV testing. Health care coverage was also associated with lower HIV risk behavior. This latter result is explained by other variables in Table 3. Women who had health care coverage were 1.5 times (50%) more likely to get an HPV vaccination.

Multiple logistic regression models showed the significant and relative contribution of the variables (Table 4). Each of the variables were significant in the models involving Pap, HPV, and HIV testing, with the exception of weight classification for HIV testing. Only age, marital status, and income were significant variables in the model involving HPV vaccination and only age, smoking status, marital status, and income were significant in the model involving HIV risk behavior. Age has the largest influence in each of the models.

Table 4.

Type 3 Analysis of Effects F Values

Pap <3 years			HPV <5 years			HPV vaccine			HIV ever			HIV risk behavior
	F	Pr > F		F	Pr > F		F	Pr > F		F	Pr > F		F	Pr > F
Age	971.27	<0.0001	Age	699.86	<0.0001	Age	35.17	<0.0001	Age	766.85	<0.0001	Age	151.06	<0.0001
Health care	175.55	<0.0001	Health care	50	<0.0001	Marital	6.9	<0.0001	Race/ethnicity	289.36	<0.0001	Smoking	50.23	<0.0001
Education	88.48	<0.0001	Race/ethnicity	49.96	<0.0001	Income	3.28	0.0376	Smoking	186.79	<0.0001	Marital	37.23	<0.0001
Race/ethnicity	87.14	<0.0001	Education	44.23	<0.0001	Education	1.11	0.3435	Education	66.5	<0.0001	Income	3.31	0.0367
Marital	35.17	<0.0001	Marital	20.94	<0.0001	Weight	1.08	0.3571	Marital	61.31	<0.0001	Health care	2.8	0.0943
Income	16.69	<0.0001	Smoking	16.03	<0.0001	Race/ethnicity	0.7	0.5915	Income	14.07	<0.0001	Education	2.57	0.0526
Smoking	14.13	<0.0001	Income	5.99	<0.0001	Smoking	0.65	0.5815	Health care	8.48	0.0036	Weight	1.9	0.1272
Weight	11.48	<0.0001	Weight	1.35	0.0001	Health care	0.42	0.5192	Weight	1.23	0.2974	Race/ethnicity	1.5	0.1992

Pr > F refers to the p-value associated with the F statistic.

Source: Behavioral Risk Factor Surveillance System (BRFSS, 2018).

Adjusted odds (95% confidence interval [CI]) of a Pap, HPV, and HIV test among women with HIV risk behavior were 1.5 (1.3–1.8), 1.6 (1.4–1.8), and 2.6 (2.3–2.9), respectively. Interaction terms involving HIV risk were added to each model, but only an interaction involving marital status was significant in the Pap model. The association between HIV risk behavior and Pap testing within the last 3 years is given by marital status in Figure 1. HIV risk behavior was not associated with Pap testing in women who are married, separated (marginally insignificant), or part of an unmarried couple. However, women with HIV risk behavior had significantly greater odds of Pap testing if they were divorced, widowed, or never married. The higher odds was most pronounced in never married women.

FIG. 1.

Adjusted odds of Pap test within 3 years according to HIV risk behavior by marital status. Pap, Papanicolaou.

HIV risk behavior was not significantly associated with HPV vaccination (adjusted odds ratio = 1.9, 95% CI = 0.97–3.8).

Discussion

This study explores whether women with HIV risk behavior have higher prevalence of Pap, HPV, and HIV testing. We found that testing is more common in women where HIV risk situations apply, more so for HIV, followed by HPV and then Pap. Interaction terms involving HIV risk behavior in each of the models were insignificant, with the exception of an interaction involving marital status in the Pap model. The positive association between HIV risk behavior and Pap testing is dependent on marital status, only significant for women who are divorced, widowed, or never married. It is marginally insignificant for women who are separated. Significant positive associations exist for divorced, widowed, and never married, but not for married, separated, and unmarried couples. Perhaps HIV risk situations in divorced, widowed, and never married women have a longer duration of occurrence, wherein the perceived need for testing is greater.

Age has the largest influence on testing, with observed prevalence of Pap, HPV, and HIV in line with recommended guidelines.^15

–18 Drop in testing among women in their 40s and older is consistent with observations in other research,²⁶ and correlates with lower HIV risk behavior in these older women. Although women older than 65 years do not generally need Pap testing, some situations warrant continued screening (e.g., abnormal prior Pap and/or HPV test, history of cervical cancer, sexually active with multiple partners, and if exposed to diethylstilbestrol before birth). Yet, it is unclear whether these reasons fully explain why almost 42% of women 65 years and older reported having had a Pap test in the prior 3 years, especially because only 0.6% of these women reported HIV risk behavior and 0.14% indicated a history of cervical cancer.

Non-Hispanic blacks, Hispanics, and non-Hispanic multi-race have the highest prevalence of HIV risk behavior. They also have the highest prevalence of Pap, HPV, and HIV testing, more so for HPV and HIV. Higher prevalence of Pap, HPV, and HIV testing in non-Hispanic blacks, Hispanics, and non-Hispanic multi-race is consistent with other research.^26

–29 It may be that higher risk of infection and cervical cancer causes women to perceive a greater need for testing and/or be more likely to be recommended for testing. It may also be that higher testing among these racial/ethnic groups in the United States reflect successful efforts from the National Breast and Cervical Cancer Early Detection Program.⁴⁰ The program specifically provides funding to all states to help minority and underserved women gain access to cancer screening.

Along with higher HIV risk behavior, non-Hispanic blacks in the United States have higher cervical cancer rates and HIV prevalence,^41
–43 which may also contribute to higher Pap, HPV, and HIV testing. Pap testing is least common in non-Hispanic other race. Asians have the largest representation in this racial/ethnic group. Lower Pap testing here is consistent with previous research,^27,44 and likely reflects these women having lower perceived risk for cervical cancer. Lower HIV testing among non-Hispanic Asian women is also consistent with their having comparatively low levels of HIV.⁴²

Prevalence of the Pap test is greatest for women who are married, separated, or part of an unmarried couple. Several studies have shown comparatively high prevalence of Pap testing among married women.^30

–33 This may be explained by a husband encouraging his wife to be screened, which is more likely if the man is knowledgeable about the importance of screening.³¹ Higher Pap testing among unmarried couples may similarly involve a knowledgeable partner providing encouragement for screening. Why is there a higher level of Pap testing in women who are separated, is unclear.

Prevalence of HPV testing is higher among women who are not married or widowed. Greater HPV testing in these women may be because of higher perceived HPV risk. In general, unmarried adults are at greater risk for HPV infection.^45,46 Other research has identified comparatively high sex-risk behaviors in unmarried women.¹⁹ Not being married may also contribute to higher HIV risk behavior in non-Hispanic blacks, Hispanics, and non-Hispanic multi-race. Note that in our data prevalence of marriage for non-Hispanic whites, non-Hispanic blacks, Hispanics, non-Hispanic other race only, and non-Hispanic multi-race is 54.84%, 27.61%, 43.28%, 53.05%, and 36.81%, respectively.

Women with a college education have the lowest HIV risk behavior. They also have the highest level of Pap, HPV, and HIV testing. Higher educated women are more likely to know that HIV risk behavior can lead to HPV infection, cervical cancer, and HIV, through discussions with their health care provider and other resources.^27,34,35 Likewise, women with an annual household income of at least $50,000 per year have the lowest HIV risk behavior. They also have higher prevalence of Pap and HPV testing. However, both HIV risk behavior and HIV testing are more common among women in the lowest income group. This is consistent with research showing that the burden of HIV in the United States is greatest among those below the poverty level.³⁶

Current smokers have significantly higher prevalence of HIV risk behavior. Current daily smokers have significantly lower prevalence of Pap testing and current smokers have higher prevalence of HPV and HIV testing. Other research has also observed lower Pap testing among current smokers, despite their having higher risk of developing cervical cancer.⁴⁷ Higher HPV and HIV testing among current smokers may be because of the widely known link between HPV and smoking,⁴⁸ and because individuals with HIV are more likely to develop the harmful consequences of certain diseases associated with smoking (e.g., heart disease, cancer, serious lung diseases and infections such as pneumonia and other illness).^49,50 Furthermore, the significant association between current smoking and HIV risk behavior is consistent with women who smoke having a constellation of inferior health practices (e.g., poorer diet, less physical activity, and high-risk sexual behaviors).^29,51
–53

Underweight women have the greatest prevalence of HIV risk behavior. This higher prevalence is explained by other variables in the multiple regression model. Weight classification was only significantly associated with Pap testing, after adjustment. Compared with women of normal weight, underweight women are less likely and overweight women are more likely to have a Pap test. Lower screening in underweight women is consistent with other research, and may reflect generally poorer health practices in these women.^37,38 In contrast to a previous study that found that overweight and obese women were more likely to delay Pap testing,³⁷ we did not observe significantly lower prevalence of Pap testing in these women.

Health care coverage correlates with lower prevalence of HIV risk behavior. This correlation is explained by other variables in the multiple regression model. Health care coverage is associated with significantly higher prevalence of Pap, HPV, and HIV testing in the adjusted models.

HPV vaccination did not correlate with HIV risk behavior after accounting for age, marital status, and annual household income. In the multiple regression model, only age, marital status, and income significantly contributed to HPV vaccination. HPV vaccination is much higher in the youngest age group, 18–24 years. Although HPV vaccine can be given in the age range 9–45 years, most individuals only benefit from it if received before age 27.⁵⁴ HPV vaccination is more common in never married or unmarried couples. These women may perceive themselves at greater risk for HPV and, thus, get the HPV vaccination. Finally, the increase in HPV vaccination prevalence with higher income is consistent with ability to pay for the vaccination.

A previous study assessed how certain explanatory variables may influence cervical cancer screening among women with HIV risk behaviors.⁵⁵ Like our study, they found that women with HIV risk behaviors have a relatively high level of cervical cancer screening. They also found that this greater level of cervical cancer screening was most pronounced in the age group 23–34 years, decreasing thereafter, college graduates, regular physical exercisers, never smokers, insured, and recently tested for HIV. Our study did not consider physical exercise, but for the other variables among all women, not just those with HIV risk behaviors, similar findings were observed. An assessment of statistical interactions indicated that these variables correlate with cervical cancer screening similarly between women with and without HIV risk behaviors. An exception was that the significant positive association between HIV risk behaviors and Pap testing depended on marital status (Fig. 1), as already discussed. Some limitations exist in this study. First, response rates average 49.8%. Second, this is a cross-sectional survey where recall and honest reporting may be an issue. Third, interpretation is limited to discussing associations rather than cause–effect relationships.

Conclusion

Women with HIV risk behavior have higher prevalence of Pap, HPV, and HIV testing than women without HIV risk behavior, with the influence of HIV risk behavior greatest for HIV testing and smallest for Pap testing. The lower positive association involving Pap may be because this test is more routinely received and not as motivated by just risk behaviors. The positive associations between HIV risk behavior and testing do not vary across the levels of age, race/ethnicity, marital status, education, annual household income, smoking status, weight classification, or health care coverage, with the exception of marital status in the Pap model. Here a positive association exists for divorced, widowed, and never married, but not for married, separated, or unmarried couples. It may be that divorced, widowed, and never married have greater duration of HIV risk behavior, thereby promoting a greater perceived need for testing.

Levels of selected demographic and lifestyle variables that correspond with higher HIV risk behavior are ages 18–24 years; non-Hispanic blacks, Hispanics, and non-Hispanic multi-race; not married; less than a college education; <$50,000 annual household income; current smokers; underweight; and without health care coverage.

Women in the youngest and oldest age groups have the lowest Pap, HPV, and HIV testing. This is consistent with recommended screening guidelines recommending that Pap screening begins at age 21 years and HPV begins at age 30 years. Women who are non-Hispanic blacks, Hispanics, and non-Hispanic multi-race have the highest Pap, HPV, and HIV testing. Perhaps health care professionals recommend screening more so among these women and/or because of programs specifically aimed to provide funding assist minority and underserved women gain access to screening. Women who are not married have the highest prevalence of Pap testing and the lowest prevalence of HPV and HIV testing. Married women may perceive their risk of HPV and HIV as lower, thereby being less inclined to undergo testing for these viruses. Women with less than a college education have lower Pap, HPV, and HIV testing. This is consistent with more educated women having a better overall health orientation. Women with <$50,000 annual household income have lower Pap and HPV testing, but higher HIV testing. The link between lower income and higher HIV testing is consistent with lower income women historically being at greater risk for HIV, whereby recommendation for HIV testing for these women is greater. Women who are underweight have lower Pap testing. This may be because underweight women have poorer health orientation in general. Finally, women without health care coverage have lower Pap, HPV, and HIV testing. This is consistent with ability to undergo screening in general.

Footnotes

Data Availability

The data used to support the findings of this study are available in the public domain.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Bosch

, Lorincz

, Munoz

, Meijer

, Shah

. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol, 2002; 55:244–255.

Walboomers

, Jacobs

, Manos

, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol, 1999; 189:12–19.

Viens

, Henley

, Watson

, et al. Human papillomavirus-associated cancers—United States, 2008–2012. MMWR, 2016; 65:661–666.

Centers for Disease Control and Prevention. Genital HPV infection fact sheet. Rockville, MD: Division of STD Prevention, Centers for Disease Control and Prevention, 2004.

Stanley

, Sterling

. Host responses to infection with human papillomavirus. Curr Probl Dermatol, 2014; 45:58–74.

De Vuyst

, Alemany

, Lacey

, et al. The burden of human papillomavirus infections and related diseases in sub-Saharan Africa. Vaccine, 2013; 31(Suppl S5):F32–F46.

Pantanowitz

, Michelow

. Review of human immunodeficiency virus (HIV) and squamous lesions of the uterine cervix. Diagn Cytopathol, 2011; 39:65–72.

Grulich

, van Leeuwen

, Falster

, Vajdic

. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: A meta-analysis. Lancet, 2007; 370:59–67.

Hernández-Ramírez

, Shiels

, Dubrow

, Engels

. Cancer risk in HIV-infected people in the USA from 1996 to 2012: A population-based, registry-linkage study. Lancet HIV, 2017; 11:e495–e504.

10.

Dubrow

, Silverberg

, Park

, Crothers

, Justice

. HIV infection, aging, and immune function: Implications for cancer risk and prevention. Curr Ipin Oncol, 2012; 24:506–516.

11.

Shiels

, Cole

, Kirk

, Poole

. A meta-analysis of the incidence of non-AIDS cancers in HIV-infected individuals. J Acquir Immune Defic Syndr, 2009; 52:611–622.

12.

Looker

, Rönn

, Brock

, et al. Evidence of synergistic relationships between HIV and human papillomavirus (HPV): Systematic reviews and meta-analyses of longitudinal studies of HPV acquisition and clearance by HIV status, and of HIV acquisition by HPV status. J Int AIDS Soc, 2018; 21:e25110.

13.

Smith-McCune

, Shiboski

, Chirenje

, et al. Type-specific cervico-vaginal human papillomavirus infection increases risk of HIV acquisition independent of other sexually transmitted infections. PLoS One, 2010; 5:e10094.

14.

Houlihan

, Larke

, Watson-Jones

, et al. Human papillomavirus infection and increased risk of HIV acquisition. A systematic review and meta-analysis. AIDS, 2012; 26:2211–2222.

15.

Smith

, Andrews

, Brooks

, et al. Cancer screening in the United States, 2017: A review of current American Cancer Society guidelines and current issues in cancer screening. CA Cancer J Clin, 2017; 67:100–121.

16.

Branson

, Handsfield

, Lampe

, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep, 2006; 55:1–17.

17.

Wuerthner

, Avila-Wallace

. Cervical cancer: Screening, management, and prevention. Nurse Pract, 2016; 41:18–23.

18.

Meites

, Szilagyi

, Harrell

, et al. Human papillomavirus vaccination for adults: Updated recommendations of the advisory committee on immunization practices. MMWR Morb Mortal Wkly Rep, 2019; 68:698–702.

19.

Giuliano

, Anic

, Nyitray

. Epidemiology and pathology of HPV disease in males. Gynecol Oncol, 2010; 117(2 Suppl):S15–S19.

20.

Auvert

, Buvé

, Ferry

, et al. Ecological and individual level analysis of risk factors for HIV infection in four urban populations in sub-Saharan Africa with different levels of HIV infection. AIDS, 2001; 15(Suppl 4):S15–S30.

21.

Stanley

. Pathology and epidemiology of HPV infection in females. Gynecol Oncol, 2010; 117(2 Suppl):S5–S10.

22.

Zhang

, Sun

. Pap testing in the USA and potential association with HPV vaccination: A cross-sectional analysis of the BRFSS data (2007–2016). J Gen Intern Med, 2019; 34:17–19.

23.

Kim

, Burger

, Sy

, Campos

. Optimal cervical cancer screening in women vaccinated against human papillomavirus. J Natl Cancer Inst, 2017; 109:djw216.

24.

Centers for Disease Control and Prevention. Behavior risk factor surveillance system. Overview: BRFSS 2016. Available at: https://www.cdc.gov/brfss/annual_data/2016/pdf/overview_2016.pdf Accessed December 15, 2019 .

25.

Centers for Disease Control and Prevention. LLCP 2018 codebook report. Available at: https://www.cdc.gov/brfss/annual_data/2018/pdf/codebook18_llcp-v2-508.pdf Accessed December 15, 2019 .

26.

Llanoa

AAM

, Tsui

, Rotter

, Toler

, Stroup

. Factors associated with high-risk human papillomavirus test utilization and infection: A population-based study of uninsured and underinsured women. BMC Womens Health, 2018; 18:162.

27.

Monnat

. Race/ethnicity and the socioeconomic status gradient in women's cancer screening utilization: A case of diminishing returns?. J Health Care Poor Underserved, 2014; 25:332–356.

28.

Bennett

, Probst

, Bellinger

. Receipt of cancer screening services: Surprising results for some rural minorities. J Rural Health, 2012; 28:63–72.

29.

King

, Chen

, Garza

, Thomas

. Breast and cervical screening by race/ethnicity: Comparative analyses before and during the great recession. Am J Prev Med, 2014; 46:359–367.

30.

Moretto

, Canil

, Weberpals

. Correlation between marital status and adequate Pap test screening among Canadian women: An analysis of sociodemographic factors from the Canadian community health survey, 2007–2008. J Clin Oncol, 2017; 30:e15537.

31.

Ncube

, Bey

, Knight

, Bessler

, Jolly

. Factors associated with the uptake of cervical cancer screening among women in Portland, Jamaica. N Am J Med Sci, 2015; 7:104–113.

32.

Suzu

, Elizabeth

, Adejumo

. Husbands' knowledge, attitude and behavioral disposition to wives screening for cervical cancer in Ibadan. Afr J Psychol Stud Soc Issues, 2014; 17:167–176.

33.

Williams

, Amoateng

. Knowledge and beliefs about cervical cancer screening among men in Kumasi, Ghana. Ghana Med J, 2012; 46:147–151.

34.

Olusola

, Banerjee

, Philley

, Dasgupta

. Human papilloma virus-associated cervical cancer and health disparities. Cells, 2019; 8:e622.

35.

Muyunda

, Musonda

, Mee

, Todd

, Michelo

. Educational attainment as a predictor of HIV testing uptake among women of child-bearing age: Analysis of 2014 demographic and health survey in Zambia. Front Public Health, 2018; 6:192.

36.

Centers for Disease Control and Prevention. HIV infection among heterosexuals at increased risk—United States, 2010. MMWR, 2013; 62:183–188.

37.

Frontaine

, Heo

, Allison

. Body weight and cancer screening among women. J Womens Health Gend Based Med, 2001; 10:463–470.

38.

Charkhchi

, Schabath

, Carlos

. Breast, cervical, and colorectal cancer screening adherence: Effect of low body mass index in women. J Womens Health, 2020; 29:996–1006.

39.

Levy

, Bruen

, Ku

. Health care reform and women's insurance coverage for breast and cervical cancer screening. Prev Chronic Dis, 2012; 9:120069.

40.

Miller

, Plescia

, Ekwueme

. Public health national approach to reducing breast and cervical cancer disparities. Cancer, 2014; 120(Suppl 16):2537–2539.

41.

Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Incidence—SEER 21 Regs Limited-Field Research Data + Hurricane Katrina Impacted Louisiana Cases, Nov 2018 Sub (2000–2016) <Katrina/Rita Population Adjustment>—Linked To County Attributes—Total U.S., 1969–2017 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2019, based on the November 2018 submission.

42.

Centers for Disease Control and Prevention. HIV surveillance report. Available at: https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-report-2018-vol-30.pdf Accessed May 29, 2020 .

43.

Singh

, Azuine

, Siahpush

. Widening socioeconomic, racial, and geographic disparities in HIV/AIDS mortality in the United States, 1987–2011. Adv Prev Med, 2013; 2013:657961.

44.

Kagawa-Singer

, Pourat

, Breen

, et al. Breast and cervical cancer screening rates of subgroups of Asian American women in California. Med Care Res Rev, 2007; 64:706–730.

45.

Sellors

, Mahony

, Kaczorowki

, et al. Prevalence and predictors of human papillomavirus infection in women in Ontario, Canada. Survey of HPV on Ontario women (SHOW) group. CMAJ, 2000; 163:503–508.

46.

Cercato

, Mariani

, Vocaturo

, et al. Predictors of human papilloma virus (HPV) infection in Italian women. J Med Virol, 2010; 82:1921–1927.

47.

Lam

, Stewart

, Ho

. Smoking and high-risk sexual behavior among young adults in Hong Kong. J Behav Med, 2001; 24:503–518.

48.

, Koutsky

, Castle

, et al. Relationship between cigarette smoking and human papilloma virus types 16 and 18 DNA load. Cancer Epidemiol Biomarkers Prev, 2009; 18:3490–3496.

49.

U.S. Department of Health and Human Services. AIDS.gov: Smoking. Available at: https://www.hiv.gov/hiv-basics/staying-in-hiv-care/other-related-health-issues/smoking Accessed May 29, 2020 .

50.

Helleberg

, Afzal

, Kronborg

, et al. Mortality attributable to smoking among HIV-1-infected individuals: A nationwide, population-based cohort study. Clin Infect Dis, 2013; 56:727–734.

51.

MacLean

, Cowan

, Vernarelli

. More to gain: Dietary energy density is related to smoking status in US adults. BMC Public Health, 2018; 18:365.

52.

Heydari

, Hosseini

, Yousefifard

, Asady

, Baikpour

, Barat

. Smoking and physical activity in healthy adults: A cross-sectional study in Tehran. Tanaffos, 2015; 14:238–245.

53.

Hansen

, Kjaer

, Munk

, et al. Early smoking initiation, sexual behavior and reproductive health—A large population-based study of Nordic women. Prev Med, 2010; 51:68–72.

54.

Boersma

, Black

. Human papillomavirus vaccination among adults aged 18–26, 2013–2018. NCHS Data Brief No. 354. Hyattsville, MD: NCHS, January. 2020.

55.

Zhang

, Advani

, Huchko

, Braithwaite

. Impact of healthcare access and HIV testing on utilisation of cervical cancer screening among US women at high risk of HIV infection: Cross sectional analysis of 2016 BRFSS data. BMJ Open, 2016; 10:e031823.