Attitudes and current practices of primary care clinicians in performing anal cancer screening

Current guidelines regarding anal cancer screening are limited. We sought to determine the anal cancer screening practices among physicians in primary care specialties. In 2017, a survey was conducted among 578 physicians from the primary care specialties of internal medicine (IM), family medicine (FM) and obstetrics and gynecology (ob/gyn) regarding their anal cancer screening practices. The survey questions focused on patient demographics, practice setting, history and physical examination practices, and physician’s personal confidence in anal cancer screening. There were 103 (17.8%) respondents and 89% completed the entire survey. Physician specialties were most commonly FM (39%) or IM (39%), compared to ob/gyn (22%). Ob/gyn physicians more frequently took patient sexual histories more than once per year (ob/gyn: 39% vs FM: 3%, IM: 5%; p<0.005) and asked questions regarding receptive anal intercourse practices, anal condylomata, cervical dysplasia or sexually transmitted diseases (p<0.005). IM and ob/gyn physicians also more commonly performed anorectal exams once per year (ob/gyn: 64%, IM: 69% vs FM: 43%, p=0.06). Overall, 40% physicians reported that they referred patients due to lack of training and 15% referred due to lack of equipment. A large proportion reported lower personal confidence (1 or 2 out of 5 on Likert scale) with their anal cancer screening practices (ob/gyn: 41%, FM: 50%, IM: 53%) There is significant variability among primary care physicians regarding their practices of anal cancer screening. Lack of training and limited guidelines are likely factors contributing towards inadequate anal cancer screening. *Correspondence to: Sharon Stein, University Hospitals Cleveland Medical Center, Department of Surger 11100 Euclid Ave Cleveland, OH 44106, USA, Tel: (216) 844-2873; E-mail: sharon.stein@uhhospitals.org


Introduction
Anal squamous cell carcinoma is a relatively rare malignancy, affecting approximately two per 100,000 people in the United States, but the incidence of anal cancer is rising [1,2]. Anal cancer is analogous to cervical cancer; the human papilloma virus (HPV) has a close association with dysplasia and neoplasia in both cancers. Immunosuppression, including infection with human immunodeficiency virus (HIV), increases the risk of progression from dysplasia to cancer [2][3][4][5]. In high-risk populations, including HIV-positive individuals and men who have sex with men, the rate of anal cancer is higher than the rate of cervical cancer in women in the general population [6][7][8].
Unlike cervical cancer, recommendations and implementation of standard screening guidelines are not well established, despite evidence that 5 year survival rates are improved with early detection [9,10]. Most of the data for anal cancer screening relates to surveillance for patients who have developed anal dysplasia and recommendations for monitoring patients for progression [11][12][13]. Studies demonstrating screening practices for patients who may be at risk are limited. Data highlighting attitudes of physicians treating patients with HIV found that although many were aware of the risks of anal cancer, they did not routinely screen patients [14,15]. To our knowledge, there are no published data on screening in general practice patients or highlighting referral patterns for screening.
Currently, there is a knowledge gap regarding the clinical practices for anal cancer screening by primary practitioners. Further understanding of current practices, and obstacles to screening will enable us to better create screening guidelines and focus educational efforts about anal cancer. We sought to determine the existing practices and patterns of anal cancer screening among these providers at a large health care system, where primary care providers frequently serve as the first point of contact for the patients who may be susceptible for anal cancer.

Methods
Physicians from the University Hospitals Health System in Cleveland, Ohio, USA, were asked to complete an Institutional Review Board-approved internet-based survey via Surveymonkey (SurveyMonkey Inc. San Mateo, California, USA). The University Hospitals Health System is a large health care entity consisting of eighteen hospitals and fifteen ambulatory sites in Northeastern Ohio. Providers work at a variety of health care settings. The survey was sent to family medicine (FM), internal medicine (IM), and obstetrics and gynecology (ob/gyn) providers via email. Completion of the questionnaire was deemed consent for inclusion in the study.
The survey consisted of demographic information including type of practice, years of practice, and practice setting. Providers were asked to characterize their practice regarding patient age ranges, gender, and whether 10% or more of the provider's patient population were "at risk" population. For the purposes of the study, "at risk" was defined as HIVpositive, transplant recipients, on chronic immunosuppression, or men who have sex with men. Physicians were asked about sexual historytaking practices, anorectal exam practices, anal cancer screening practices, and personal confidence in their ability to screening for anal cancer using a Likert Scale. If physicians did not screen patients for anal cancer, they were asked for the reason they did not screen and whether they referred any patients for screening.
For statistical analysis, all categorical and ordinal variables are presented as frequencies with percentages and were compared using the chi square test or Fisher's exact test, where appropriate. All analysis was performed using Stata SE Version 14.2 (StataCorp, College Station, Texas).

Results
The survey was sent to 578 physicians and 103 (17.8%) responded: 40 respondents identified as family medicine physicians, 40 as internal medicine, and 23 as obstetrics-gynecology (Table 1). Ninety-two respondents (89%) completed the survey in its entirety.

Practice Characteristics
The majority of physicians had been in practice for greater than 10 years. Family medicine (97.5%) and internal medicine physicians (90.0%) were more likely to work in an ambulatory setting when compared to ob/gyn physicians (69.6%) (both, p<0.005). Most practitioners were located in a suburban setting (71.8%); 17.5% practiced in an urban setting and 10.7% practiced in a rural setting.

Patient Characteristics
Family medicine physicians were more likely to treat both adult and pediatric populations (FM: 95%, IM: 7.5%, ob/gyn: 30.4%; p<0.001). Only two physicians, both IM, identified having 10% or greater patients in an "at risk" category, limiting further analysis of this subset.
Eighty-percent or more of all physicians performed anorectal exams as part of a routine exam. A screening exam consisted most frequently of a digital rectal exam (ob/gyn: 85.7%, FM: 64.3%, IM: 79.3%). An external exam was performed by most ob/gyn physicians, but just over half of the IM and FM providers (ob/gyn: 92.9%, FM: 53.6%, IM: 55.2%). Ob/gyn and FM physicians were more likely to perform cervical pap smears than IM physicians (Ob: 100%, FM: 92.5%, IM: 57.5%). Additional instrumentation such as anoscopy, flexible sigmoidoscopy or high resolution anoscopy was rarely used  by any provider. Frequency of anorectal exam was varied. IM and ob/gyn physicians performed anorectal exams annually compared to a one-time exam for FM physicians, although the difference was not statistically different (p=0.06) ( Table 3).
Half of FM physicians, 52.8% of IM physicians, and 40.9% of ob/ gyn physicians reported a poor confidence level (score of 1 or 2 on Likert scale) to perform anal cancer screening. A higher percentage of ob/gyn physicians reported their confidence level as a four or five out of five (36.0%) compared to the FM (11.8%) and IM physicians (22.3%) (p<0.05), (Table 4). When asked why they rated themselves as having low confidence, most FM and IM providers stated that they had inadequate training (FM: 45.5%, IM: 30.6%) or the patient population did not warrant screening (FM: 30.3%, IM: 27.8%), while ob/gyn predominantly stated that they had inadequate training (45.5%).
Despite poor confidence for screening, most providers did not refer all patients for screening (Table 5). Sixty-eight percent of FM physicians performed screening on patients they felt were at high risk for anal cancer compared with 44.8% of IM physicians and 35.7% of ob/gyn physicians (p<0.05). Less than a third of providers referred patients who were thought to be at high risk for anal cancer (FM: 29.4%, IM: 16.7%, and ob/gyn: 27.3%). Numbers increased considerably if the patient had an anorectal finding (FM: 76.5%, IM: 63.9%, ob/gyn: 72.7%). When asked which findings would warrant a screening referral, almost all providers would refer for an anal mass, or anal bleeding (FM: 94.1%, IM: 91.7%, Ob/gyn: 95.5%), but a fewer physicians would refer for anal condyloma (FM: 70.6, IM: 75, ob/gyn: 68.2%). If referrals were to be made, the physicians most commonly referred to colorectal surgeons (66%), followed by gastroenterologists (54%), and general surgeons (29%).

Discussion
With the increasing prevalence of anal cancer risk factors and increasing incidence of anal cancer, the health care system needs to establish guidelines for screening for anal cancer. To date, there is minimal evidence regarding methods of screening for anal cancer. This study helps to elucidate the practice patterns of anal cancer screening among first line providers.
Most of our providers are not comfortable screening for anal cancer. Almost half of the providers surveyed reported confidence levels of 1 or 2 on a 5-point Likert scale. This is not surprising given the lack of information on screening available in the literature. Unlike cervical cancer, for which a robust educational program has significantly increased screening patterns, parallel recommendations for anal cancer have been lacking. After the introduction of the Pap smear in 1928 and colposcopy in the 1970's, the rates of cervical cancer were decreased in half due to aggressive screening and treatment programs [16,17].
There is a wide variety of screening practices elucidated in the study, including both whom should be screened and how they should be screened. More providers did a digital rectal exam than external visual exam. This is surprising as a visual perianal exam, in addition to digital rectal exam, is recommended for anyone with a perianal complaint [18]. Providers rarely used a scope to better visualize changes in the anal mucosa. It was also interesting to note that providers were more likely to refer patients for rectal bleeding rather than for condyloma. The majority of anal condyloma cases are caused by low-risk serotypes of HPV [19,20]. However, these lesions are risk factors for anal cancer and visual changes that accompany anal condyloma infection may represent pre-cancerous lesions [20,21]. It has been suggested that anal condyloma above the dentate line, in MSM, or in HIV-positive individuals warrants consideration of biopsy [22].
The current study is unique because unlike most previous reports that focused on anal cancer screening patterns among specialists, the physician cohort of our study represents the first-line encounter for patients who could possibly require screening for anal cancer. It is difficult to set expectations for primary care providers when we lack standards and algorithms for screening for anal cancer. In lieu of official guidelines, many articles provide recommendations given the lack of resounding evidence for the use of anal cancer screening practices. Nearly all of these recommendations are for high-risk patients, rather than the general patient population. Some advocate for yearly digital anorectal exams in high-risk groups, while others for anal cytology followed by high resolution anoscopy (HRA) if findings are noted, and others recommend HRA alone be used to screen for high-grade squamous intraepithelial lesions and anal cancer [23][24][25].    Findings from the present study should be interpreted in the context of limitations. As with many surveys, there is potential bias in who chose to participate in the study. Our study was sent through official University Hospitals email, but many providers may not open or read the email, and we did not have the ability to determine if the email was opened or read. In addition, there may be inherent bias in practitioners who complete the email secondary to increased interest in the topic. The low response rate makes generalization from the study difficult, but it does provide us with a limited view of the scope of practice, education, and gaps in our literature. Our survey was created for this study alone and did not undergo extensive validation or testing procedures.

Family medicine, n (%) Internal Medicine, n (%) Obstetrics & Gynecology, n (%) p-value
This study demonstrates a need for better screening recommendations, education, and guidelines. The rate of anal cancer is increasing, perhaps in part to more people living with immunosuppression long enough to develop a cancer. But unlike other cancers which are amenable to screening, screening parameters have not yet been developed for anal cancer. It is conceivable that with active screening, the incidence of anal cancer could be dramatically reduced, or even eradicated.

Conclusion
There is great variability among physicians practicing in primary care specialties regarding their practices in the screening for anal cancer. Variability exists in many aspects of screening including historytaking, physical exam procedures, and referral practices. Primary care providers may benefit from evidence-based guidelines, widespread education, and increased training in order to identify pre-cancerous and early-stage cancer lesions in order to improve patient outcomes.