Colorectal Cancer Screening Practices among Obstetrician/Gynecologists and Nurse Practitioners

Introduction

Colorectal cancer (CRC) is the third leading cause of cancer and cancer-related deaths in women. ¹ In 2007, there will be an estimated 74,630 new diagnoses of colorectal cancer and 26,180 associated deaths in women. ¹ The current recommended CRC screening tests for cancer prevention and cancer detection based on the most recent U.S. multisociety task force guidelines (American College of Gastroenterology, American College of Gastrointestinal Endoscopy, American College of Physicians/Society of Internal Medicine, and American Cancer Society) include colonoscopy every 10 years; flexible sigmoidoscopy, CT colonography, or barium enema every 5 years; yearly fecal occult blood test (FOBT) or fecal immunohistochemical test; and stool DNA, interval unknown. ² Despite these recommendations, CRC screening compliance among patients remains low. Data from the National Behavioral Risk Factor Surveillance System (BRFSS) in 2004 demonstrated only a 57% adherence rate for all CRC screening modalities. ³

Women are less likely than men to be up-to-date with their CRC screening. ^{4 –11} Therefore, it is critical to examine the CRC screening practices of physicians and nurse practitioners (NPs) who routinely provide primary care to women. Studies have shown that 38%–93% of gynecologists report providing primary care to their patients. ^{12 –14} Since 1996, the Council on Resident Education in Obstetrics and Gynecology has recognized the importance of their discipline's role in women's primary healthcare and has required 6 months of primary care in their curriculum. In addition, NPs are use regularly in obstetrician/gynecologist (Ob/Gyn) practices and routinely perform health maintenance examinations. Thus, it is critical to ensure that CRC screening has been incorporated into both the Ob/Gyns' and NPs' preventive screening practices.

The objective of this study is to examine current CRC screening practices relative to nationally accepted guidelines and other preventive screening practices, such as mammography and cervical cancer screening, in a nationally representative sample of Ob/Gyns and NPs.

Materials and Methods

Results

Discussion

Improving CRC screening practices is an initiative promoted by several professional societies, including the American College of Obstetricians and Gynecologists (ACOG). This study is the first to describe current CRC screening practices and knowledge among Ob/Gyns and NPs in light of the most recent practice guidelines. ² Our findings suggest that both provider knowledge and screening practices need improvement, particularly with regard to FOBT use.

Our findings are similar to those of studies of other physicians. In two national surveys of primary care physicians, 85.1%–98% of participants recommended CRC screening. ^16,17 In contrast, screening rates among gastroenterologists are reported to be as high as 98%. ¹⁵ However, NPs in our study reported lower screening rates than in previous studies. In 2000, Shaheen et al. ¹⁸ conducted a survey of NPs and physician assistants. They reported 69% of primary care NPs reported recommending flexible sigmoidoscopy and 92% reported performing FOBT.

Current recommendations regarding FOBT testing include two samples from three consecutive stools that are not rehydrated using either a guaiac-based test with dietary restriction (avoiding red meat) or an immunochemical test without dietary restriction. ¹⁹ For both Ob/Gyns and NPs in our sample, FOBT (66%) was the diagnostic modality most often used for screening, but it appears that many are not using it as recommended. Almost one third of participants reported relying on a single stool sample obtained by in-office digital rectal examination as their screening modality. Similarly, Musinski ²⁰ reported in a retrospective study of gynecology patients that 43% of CRC screening consisted of a single digital FOBT. Single-sample digital rectal examination appears to be a common practice of other primary care providers as well. ²¹ The sensitivity of single digital FOBT is markedly lower (4.9%) than that of home FOBT kits (23.9%), which assess two samples from three consecutive stools. ²¹ Further, 66% of practitioners in our study rehydrated stool samples, which is not recommended in the current guidelines because it decreases the performance of the assay. ¹⁹ Finally, 43% of those surveyed failed to include dietary restrictions when using guaiac-based FOBT, and a similar proportion reported withholding NSAIDs, although there are no recommendations to do so. ¹⁹

The effectiveness of FOBT as a CRC screening tool is determined in part by additional testing, specifically, ordering a colonoscopy when a positive test occurs. One fifth of our study participants would repeat the FOBT after a positive result, which is similar to respondents elsewhere. ²¹ Because important lesions may bleed only intermittently, confirmatory FOBT could miss clinically important lesions. ²² Therefore, follow-up of all positive tests with colonoscopy is indicated. ^19,22

Despite the fact that FOBT was the most commonly used modality, providers preferred colonoscopy for themselves. This discordance could be explained by a number of factors, including convenience of FOBT, cost, availability or risks of colonoscopy, and patient compliance. Klabunde et al. ²³ demonstrated from the 2000 National Health Interview Survey that 46% of primary care providers cited cost and lack of insurance as one of the major barriers to CRC screening. In a recent survey of primary care providers in Wisconsin, 68.5% cited cost as the major reason for not recommending colonoscopy, and 33.2% and 26.1% reported a lack of availability or risk of perforation, respectively, as barriers to colonoscopy. ²⁴ Providers' expectation of low patient compliance might also explain the relatively low use of colonoscopy. However, providers are not particularly good at determining which features of a diagnostic test are important to patients and affect their compliance. ²⁵

Clearly, opportunities for improving knowledge about CRC screening exist. For instance, only slightly more than half (54.6%) of our respondents were able to identify the correct age to initiate screening in average-risk patients. Performance in our study was only slightly better than in a 1997 study of primary care providers (48.5%) but worse than a 1999 survey of gastroenterologists (71%). ^15,16 Our findings are disheartening, as CRC screening guidelines for average-risk populations have been available since 1997. As demonstrated by our and others' findings, further education is needed to improve screening for patients at increased risk for CRC. ^24,26

We found three provider factors that appeared to be positively associated with CRC screening. After controlling for other influencing factors, practitioners in multispecialty groups were 2.62 times more likely to report routine CRC screening compared with those in single specialty private practices. This may be due to enforcement of practice standards as a result of managed care or group economics. We also found that providers in practice between 5 and 10 years were more likely to report screening than other groups, which is consistent with other studies suggesting recent graduates are more likely to screen for CRC than those in practice >10 years. ^27,28 In previous work, provider gender has also been shown to affect patient CRC screening adherence. Menees et al. ²⁹ demonstrated that having a female physician was associated with increased prevalence of CRC screening at the time of upper endoscopy (EGD) and increased CRC screening completion in the following 6 months. Other studies have suggested that patients of female physicians were more likely to undergo breast and cervical cancer screening than those with male physicians. ^30,31 Although female providers in our study were not more likely to report screening than males, they were more likely to report using a recommended method. This result supports previous findings, where women physicians have been found to be more focused on prevention than their male counterparts and have more favorable views and beliefs on prevention. ^{32 –34}

This study sampled practicing Ob/Gyns and NPs from the American Medical Association's membership list. However, this group may not represent the larger group of practicing Ob/Gyns or NPs in this country. A more representative sample of Ob/Gyns would probably be obtained by sampling members of the ACOG, but their membership lists were not available. When compared with studies of ACOG members, our sample tended to be slightly older but had a similar proportion of female physicians. Further, respondents in our study and in other studies of ACOG members tend to be in practice for ≥10 years. ^35,36

We also acknowledge several other study limitations. First, the low response rate (29%) limits the generalizability of our findings, but it is reassuring that our responses were fairly consistent with studies of other provider types, some of which might be expected to be more knowledgeable about CRC screening than Ob/Gyns. Additionally, as those who respond to surveys are typically thought to have a particular interest or knowledge in the study topic, we would expect that our sample would perform better than nonresponders. Therefore, our results probably do not underestimate screening practices or knowledge about CRC screening. Additionally, reporting bias may be present in these data because we are relying on self-reported data. Again, we would expect that influence to result in an overestimate of CRC screening practices. Other studies have shown that clinicians overestimate their cancer screening practices. ^{37 –39}

Conclusions

Our results demonstrate important CRC screening patterns among Ob/Gyns and NPs. Routine screening for CRC among Ob/Gyns and NPs is less frequent than screening for other cancers, such as breast cancer or cervical cancer. Even though up to 80% of surveyed providers reported performing routine CRC screening, only 59% routinely screen following recommended guidelines, and 53% were able to identify the correct age to initiate screening. The incorrect use of FOBT is prevalent especially among Ob/Gyns, as is inadequate follow-up of positive FOBT results. As many women seek primary care from Ob/Gyns and NPs, further education of both healthcare providers and patients may improve compliance with current CRC screening guidelines.