Anal cancer and anal cancer screening knowledge, attitudes, and perceived risk among women living with HIV

Abstract

Objectives:

(1) Describe anal cancer knowledge, perceived risk, screening barriers, and acceptability of sample self-collection among WLWH at an integrated safety-net system. (2) Describe differences in demographic and psychosocial variables among a subsample of WLWH with a history of abnormal cervical cytology results versus those with normal results.

Methods:

We conducted telephone surveys with English- and Spanish-speaking WLWH (N=99), and used EHR data to extract insurance type, CD4⁺ cell count, RNA viral load, and cervical cytology results. We calculated descriptive statistics for participant demographics, HIV lab results, and psychosocial variables. Among the subsample of women who completed a recent cervical Pap, we used Fisher’s exact test to assess differences in demographic variables, CD4⁺ counts, RNA viral loads, knowledge, awareness, acceptability, and perceived risk by cervical cytology results.

Results:

Most participants (70%) reported knowing nothing about anal cancer; 28% correctly responded that HIV increases one’s chance of getting anal cancer. Most (68%) never heard of an anal Pap test. Forty percent would get an anal Pap if they could self-collect the sample, while 59% were neutral or disagreed. The two most commonly cited barriers to obtaining an anal Pap were “I don’t know enough about it” (n=15) and “It might hurt” (n=9).

Conclusion:

This study highlights a gap in knowledge and awareness among WLWH regarding their heightened risk for anal cancer. It indicates the need for health education and suggests an opportunity for a self-collection intervention.

PRECIS

Women living with HIV lack knowledge about anal Pap testing, but many are willing to self-collect, which may be a promising tool to increase screening.

INTRODUCTION

Non-AIDS defining cancers are becoming increasingly burdensome among individuals living with HIV as life expectancy increases due to antiretroviral therapies and as the burden of AIDS-defining cancers (e.g., Kaposi’s sarcoma, non-Hodgkin’s lymphoma, and cervical cancer) decreases.¹ Individuals living with HIV are at elevated risk for specific types of non-AIDS defining cancers such as anal cancer, and screening is important as incidence increases.² Women living with HIV (WLWH) often have persistent human papillomavirus (HPV) infections and are consequently, at elevated risk for cervical dysplasia, anal dysplasia, and anal cancer. ^2–6 The incidence of anal cancer among women living with HIV (WLWH) increased from 3.5 to 28.7 per 100,000 person-years from 1996-2000 and 2009-2012.²

Evidence is lacking on whether routine screening and treatment of anal high grade squamous intraepithelial lesions (HSIL) are effective population-level strategies to reduce anal cancer incidence.⁷ The Anal Cancer HSIL Outcomes Research (ANCHOR) Study, a national study enrolling over 5,000 individuals living with HIV aims to fill this research gap.⁸ In the meantime, anal cancer screening recommendations and guidelines vary across health agencies and professional medical associations in terms of screening modalities – digital rectal exam, anal Pap test, or both – and screening frequency.^7,9,10 Some, including the HIV Medicine Association of the Infectious Diseases Society of America (HIVMA), recommend screening for specific high-risk subpopulations, such as women with a history of abnormal cervical Pap test results, rather than recommending screening all WLWH.⁹

Few studies have assessed knowledge and attitudes towards anal cancer and screening among WLWH.^11,12 Findings suggest WLWH have limited knowledge of risk factors and low perceived risk of developing anal cancer, in general. We know of no studies that assess these constructs among WLWH at elevated risk of anal cancer, defined as those with a history of abnormal cervical cytology results, despite some anal cancer screening guidelines focusing on this population. Therefore, our study’s two objectives were to: (1) describe knowledge of anal cancer, perceived risk of anal cancer, awareness of anal Pap testing, screening barriers, and the acceptability of self-collection among WLWH; and (2) describe differences in demographics and psychosocial variables among a subsample of WLWH at elevated risk of anal cancer versus those not at elevated risk (i.e., those with normal cervical cytology results).

METHODS

Data were collected as part of a larger project focused on cervical cancer screening among WLWH accessing care at a large, urban, integrated safety-net healthcare system.¹³ The study was approved by the University of Texas Southwestern Medical Center’s Institutional Review Board (STU 102016-074).

Setting

The county integrated safety-net system is the only public health care provider for approximately one million under- and uninsured individuals located in a major metropolitan area in Texas. The system provides care through federal, state, and local payer programs including Medicaid, Ryan White, the National Breast and Cervical Cancer Early Detection Program, family planning grants, and a county tax-supported assistance program. The system operates a network of 12 community-based primary care clinics and four HIV specialty clinics serving approximately 1,800 WLWH, 80% of whom are Hispanic or African American. The safety-net system began offering anal Pap testing for women in 2016.

Sample

We used the Population-Based Optimizing Screening through Personalized Regimens (PROSPR) I cohort to identify potential participants.^13,14 The PROSPR cohort includes females 18-64 years with a cervix, no history of cervical cancer, and at least one in-person primary care or women’s health clinic visit with the safety-net system between January 2010 to May 2016.

To oversample WLWH at elevated risk of anal cancer, we applied the follow eligibility criteria: 1) PROSPR cohort member with an HIV diagnosis based on ICD 9/10 codes, electronic health record (EHR) evidence of HIV labs, and record of ≥1 HIV specialty clinic visit; and 2) either under-screened (defined as due for a cervical Pap test between December 2014 and May 2016 and did not complete the test), or with abnormal cervical cytology results (defined as atypical squamous cells of undetermined significance [ASC-US] or worse) between December 2014 and May 2016. We excluded women who did not prefer English or Spanish language. Thus, the sample for our first objective included English- and Spanish-speaking women under-screened for cervical cancer and those at elevated risk for anal cancer between December 2014 and May 2016.

To assess differences in demographics and psychosocial variables by risk status (elevated risk versus normal cervical cytology results) for our second objective, we included only the subsample of participants who completed a cervical Pap test between June 2016 to survey completion (i.e., after the original sampling frame and within18-months prior to study participation).

Data collection and variables

Survey.

We used the six survey items about anal cancer and anal cancer screening from the University of North Carolina Men’s Health Survey.¹⁵ Items assessed: 1) knowledge of anal cancer; 2) knowledge of HIV infection as a risk factor for anal cancer; 3) awareness of anal Pap tests; 4) acceptability of anal self-sampling; 5) perceived risk of anal cancer; and 6) barriers to obtaining an anal Pap test. Items from the Men’s Health Survey did not include pronouns and were not changed for female participants. We also ascertained participant age.

Study materials were developed in English, translated into Spanish, and validated by a team of bilingual staff in a systematic, evidence-based process to verify ≤8^th grade literacy level, accuracy, and cultural appropriateness.¹⁶ Bilingual research staff invited eligible women to participate by telephone, obtained verbal informed consent, and administered the ~25-minute survey in English or Spanish, per participant preference. Participants received a $15 gift card by mail upon completion. Surveys were conducted between February and August 2018.

Electronic health record data.

We collected the following from electron health records: insurance/public payer type, CD4⁺ cell count, RNA viral load, and cervical cytology results for the second objective. Cervical cytology results were coded as normal (i.e., negative for any evidence of dysplasia), or abnormal (i.e., atypical squamous cells of undetermined significance [ASC-US] or worse).

Data analysis

We calculated descriptive statistics for participant demographics, HIV lab results, and psychosocial variables. Among the subsample of women who had completed a recent cervical Pap test (defined as 18 months prior to survey date), we used Fisher’s exact test to assess differences in demographic variables, CD4⁺ cell counts, RNA viral loads, knowledge, awareness, acceptability, and perceived risk by cervical cytology results. We thematically categorized free text responses to the “other” barriers to screening option and built consensus on categories through group discussion.

Role of Funding Source

Funding sources had no role in study design, data collection, data analysis, or interpretation.

RESULTS

Ninety-nine women with HIV who accessed care at the safety-net system (12% response rate) completed surveys. Most participants were 30-49 years (56%), non-Hispanic Black (78%), and spoke English (89%) (Table 1). Half of all participants (51%) had their healthcare covered by federal, state, or county payer programs. Ten percent of participants had CD4+ cell counts between 0-199 μ/l indicating acquired immunodeficiency syndrome (AIDS), and 18% had RNA viral loads over 200 copies/ml.

Table 1.

Demographic and psychosocial characteristics of women living with HIV, overall and by recent cervical cytology results

	n(%)
	Overall	Normal cervical cytology	Abnormal cervical cytology
Variable	N=99	n=26	n=11
Demographics
Age (years)
18-29	6 (6.06)	1 (3.85)	1 (9.09)
30-49	55 (55.56)	13 (50)	7 (63.64)
50-64	38 (38.38)	12 (46.15)	3 (27.27)
Highest level of education
Grade school or less	12 (12.12)	4 (15.38)	1 (9.09)
Some high school	16 (16.16)	6 (23.08)	3 (27.27)
High school diploma/GED	33 (33.33)	7 (26.92)	3 (27.27)
Some college or technical/vocational degree	32 (32.33)	8 (30.77)	4 (36.36)
College graduate	6 (6.06)	1 (3.85)	0
Race/ethnicity
Non-Hispanic black	77 (77.78)	20(76.92)	9 (81.82)
Non-Hispanic white	4 (4.04)	2 (7.69)	0
Hispanic	17 (17.17)	4 (15.38)	2 (18.18)
Other	1 (1.01)	0	0
Preferred language
English	89 (89.90)	23 (88.46)	10 (90.91)
Spanish	10 (10.10)	3 (11.54)	1 (9.09)
Insurance typea
Medicare	8 (8.08)	2 (7.69)	0
Medicaid	18 (18.18)	5 (19.23)	3 (27.27)
Federal, state, or government	51 (50.52)	10 (38.46)	3 (27.27)
Commercial	4 (4.04)	1 (3.85)	1 (9.09)
Charity/other	18 (18.18)	8 (30.77)	4 (36.36)
HIV lab resultsb,c
CD4+ cell count (cells/μl)
0-199d	10 (10.10)	0	2 (18.18)
200+	89 (89.90)	26 (100)	9 (81.82)
RNA viral load (copies/ml)
0-199	81 (81.82)	22 (84.62)	7 (63.64)
200+	18 (18.18)	4 (15.38)	4 (36.36)
Psychosocial variablese
Knowledgec
How much would you say you know about anal cancer?
Nothing at all	69 (70.41)	16 (61.54)	6 (60.00)
Some knowledge	29 (29.59)	10 (38.46)	4 (40.00)
How do you think having HIV or AIDS affects the chances of getting anal cancer?
Increases chances	28 (28.28)	12 (46.15)	1 (9.09)
Other	71 (71.72)	14 (53.85)	10 (90.91)
Awareness of anal Papc
Have you ever heard of an anal Pap test or anal Pap smear?
Yes	32 (32.33)	8 (30.77)	4 (36.36)
No	67 (67.68)	18 (69.23)	7 (63.64)
Perceived riskc
Do you feel you will get anal cancer in the future if you don’t do the anal Pap test regularly?
Yes	31 (32.29)	8 (32.00)	3 (30.00)
No	65 (67.71)	17 (68.00)	7 (70.00)
Acceptability of self-samplingc
It is more likely that I would get the anal Pap test if I could collect the sample myself.
Agree	40 (40.82)	11 (44.00)	4 (36.36)
Neutral/Disagree	58 (59.18)	14 (56.00)	7 (63.64)
Perceived barriers to screeningc
If a local doctor or clinic provided anal Pap tests, what would be your reasons for not getting one?
I don’t know enough about it	15 (15.15)	4 (15.38)	2 (18.18)
Asking for test or getting it might be embarrassing	3 (3.03)	2 (7.69)	0
It might hurt	9 (9.09)	3 (11.54)	0
Test might not be accurate	0	0	0
Costs too much	1 (1.01)	0	0
Don’t have transportation to get to doctors office	1 (1.01)	0	0
I don’t want this information in my medical chart	0	0	0
None; I am completely willing to get anal Pap test	46 (46.46)	9 (34.62)	6 (54.55)
Other	29 (29.29)	9 (24.32)	3 (8.10)

^aInsurance type is reflective of participant’s insurance status at cohort entry for parent HIV study;

^bMost recent CD4+ count and RNA viral load within 18 months prior to or on cohort entry to parent HIV study,¹³ may be measured on different days;

^cNo significant difference among WLWH by cervical cytology results;

^dIndicates acquired immunodeficiency syndrome (AIDS);

^eKnowledge of anal cancer response options “a little,” “a moderate amount,” and “a lot” collapsed to “some knowledge.” Knowledge of HIV infection as a risk factor response options “decreases chances,” “has no effect,” and “don’t know” collapsed to “other.” Acceptability of self-sampling response options “strongly agree” and “agree” collapsed to “agree” and response options “neutral,” “disagree,” and “strongly disagree” collapsed to “neutral/disagree;”

^fMore than one response option allowed;

percentages will not equal 100

Of the 99 participants, 37 (37%) completed a cervical Pap test within 18 months prior to completing the survey. Of the 37, eleven participants’ most recent cytology was abnormal and the remaining 26 had normal results.

Knowledge, awareness, and perceived risk

The majority of participants lacked knowledge about anal cancer and about HIV as a risk factor for anal cancer. About 70% of participants reported knowing nothing at all about anal cancer, and only 28% correctly responded that HIV increases an individual’s chance of getting anal cancer. Among women with a recent cervical Pap test, only 1 of 11 women (9%) with an abnormal result believed that HIV increases chances of getting anal cancer, compared to 12 of 26 women (46%) with a normal result.

Most participants (68%) had never heard of an anal Pap test. About one third of participants (32%) responded that they could get anal cancer in the future if they did not get an anal Pap test regularly.

Acceptability of self-sampling and perceived barriers to screening

Forty percent of participants agreed they would get an anal Pap test if they could self-collect the sample while 59% were neutral or disagreed.

Of the listed barriers to obtaining an anal Pap, the two most commonly cited items were “I don’t know enough about it” (n=15) and “It might hurt” (n=9). None of the participants endorsed the barrier response options of “The test might not be accurate” and “I don’t want this information in my medical chart.” For the open-ended “other barrier” question, the two most common thematically categorized responses were “discomfort” (n=6) and “low perceived risk” (n=10). Specific text statements categorized as “discomfort” included “uncomfortable and embarrassing” and “wouldn’t want anything in [there].” Participant remarks categorized as “low perceived risk” included “I don’t have anal sex” and “cancer doesn’t run in my family.” Almost half of all participants (46%) responded having no barriers to screening and were “completely willing” to get an anal Pap test if a doctor or clinic offered it. Among women who completed cervical Pap tests, most women with abnormal results (n=9/11) reported they were willing to get an anal Pap test compared to approximately one third (n=9/26) of those with normal cervical results.

Differences by risk status

We found no significant differences by cervical cytology results for demographic variables, HIV lab results, or psychosocial constructs among the subsample of women who completed a cervical Pap test within 18 months prior to survey completion.

DISCUSSION

WLWH at increased risk of anal cancer or under-screened for cervical cancer, who accessed care within an urban, integrated safety-net system, had limited knowledge about anal cancer and about HIV as a risk factor for anal cancer, limited awareness of anal Pap tests, and low perceived risk of developing anal cancer. Compared to previous studies among WLWH, these findings suggest WLWH who are at elevated risk of anal cancer and under-screened for cervical cancer have less knowledge and lower perceived risk of anal cancer than WLWH in general.^11,12 Despite this, nearly half of participants were “completely willing” to obtain an anal Pap test if it was offered, and 40% would self-sample. Results on the acceptability of self-sampling are consistent with previous studies reporting a high rate of acceptability toward anal self-sampling among WLWH.¹⁷ Further, self-sampling studies have also shown high concordance with physician-collected anal samples.^17,18 Collectively, this suggest self-sampling may be acceptable and feasible option for anal cancer screening among WLWH.

The low number of participants (n=37) completing a cervical Pap test within 18-months prior to survey completion highlights the difficulty of identifying WLWH at elevated risk of anal cancer based on cervical cytology results. This underlines the importance of self-sampling as a potential screening strategy and the need for studies focused on assessing self-sampling among WLWH. Studies have examined attitudes towards HPV cervico-vaginal self-sampling among women, attitudes towards anal self-exams (i.e., feeling for palpable anal lumps) among women and among men who have sex with men (MSM), and barriers to anal self-sampling among MSM living with HIV.^19–22 However, it is unclear if findings would be generalizable to WLWH or if findings related to self-exams would be generalizable to self-sampling behaviors, which include collecting a specimen for testing. Future studies are needed to explore the utility of self-sampling among WLWH, potential barriers to self-sampling, and effective educational tools to teach patients how to conduct the procedure, areas with limited research.

We found no significant differences in anal cancer knowledge, anal Pap test awareness, and perceived risk of anal cancer by cervical cytology results. While others have not assessed these constructs related to anal cancer, one study focused on cervical cancer found those with a history of abnormal cervical cytology, regardless of HIV status, were significantly more knowledgeable about HPV compared to those without an abnormal history.¹² The authors hypothesized this difference was due to patient education about HPV for women with abnormal cytology. Because HIVMA guidelines recommend WLWH with abnormal cervical cytology results receive anal cancer screening, we anticipated similar differences in knowledge, awareness, and perceived risk by cytology results due to patient education and/or exposure to anal Pap tests. However, only one of the 11 women with abnormal results believed that HIV increases the risk of anal cancer, compared to nearly half of women with a normal result. While we did not measure patient education, it is possible that women with abnormal results received less education about anal cancer risk due to a clinical focus on the cervical abnormality. Future studies are needed that develop and assess multilevel interventions that focus on provider communication about anal cancer and on patient education.

While a strength of this study is its unique patient population, limitations include the small sample size from a single safety-net healthcare system and limited generalizability. Our study is similar to those previously assessing anal cancer knowledge and attitudes among WLWH which also include small sample sizes (N≤200).^11,12 Studies assessing these constructs among women, with and without HIV, are often limited to single clinical or geographic settings,^11,12,23,24 and national studies, such as those that have assessed anal cancer knowledge, attitudes, and beliefs among men,¹⁵ are needed for women. Findings from our subgroup analyses by cervical cytology results should be considered preliminary and interpreted with caution due to the small sample size (n=37).

Our study is limited in generalizability in two ways. First, we sampled only WLWH accessing care within a safety-net healthcare system. Anal cancer knowledge, attitudes, and beliefs may differ among WLWH accessing care across different healthcare settings (e.g., practices serving mostly private/commercially insured patients versus patients covered by federal and state insurance programs). It is important to note, however, that most (~73%) people living with HIV in the U.S. access care through Ryan White-funded clinics like our safety-net system.²⁵ Second, our findings are not generalizable to all WLWH since we restricted our sample to only women at increased risk of anal cancer (defined as overdue for a cervical Pap test or with a history of abnormal cervical Pap test results). Current guidelines recommend anal cancer screening for WLWH at increased risk only.⁷ Knowledge, attitudes, and perceived risk may differ among women who were adherent to cervical screening guidelines, and they may differ based on prior experience with anal Pap tests, a factor we did not measure. Further studies are needed to assess these potential differences.

CONCLUSION

This study highlights a critical gap in knowledge and awareness regarding heightened risk for anal cancer among WLWH. This points to an important need for health education on this topic, especially given the willingness of women to be screened for anal cancer and their acceptance of anal self-sampling.

ABBREVIATIONS

AIDS

Acquired immunodeficiency syndrome

HIV

human immunodeficiency virus

WLWH

women living with HIV

HSIL

high grade squamous intraepithelial lesions

ANCHOR

Anal Cancer HSIL Outcomes Research Study

HIVMA

HIV Medicine Association of the Infectious Diseases Society of America

PROSPR

Population-based Optimizing Screening through Personalized Regimens Study

EHR

electronic health record

ASC-US

atypical squamous cells of undetermined significance