Trends of colorectal cancer screening methods: an analysis of Behavioral Risk Factor Surveillance System data from 2018–2020

Data source and study population

We utilized a cross-sectional study design utilizing data from the 2018–2020 cycles of the BRFSS, thereby not necessitating Institutional Review Board (IRB) approval. The BRFSS is a random digit-dialed telephone survey of the US civilian population. It is performed yearly among adults 18 years and older and asks participants about health conditions and behaviors. This survey is administered by the CDC and state health departments in the 50 US states, the District of Columbia, and other select US territories. The BRFSS utilizes ranking weights to produce estimates of populations that adjusts for survey noncoverage, nonresponse, and probability of being sampled given geographic location, race, age, and sex [12].

Study variables

First, we utilized the year of data collection for the annual assessment variable. Because the USPSTF guidelines provide Grade A evidence for CRC screening among individuals between 50 and 75 years of age, we included only individuals within this age range for our analysis. To assess whether individuals report having been screened for CRC, we utilized the following BRFSS survey questions:

1. “Have you ever had a colonoscopy?” and reports it having been conducted within the past 10 years.

2. “Have you ever had a sigmoidoscopy?” and reports it having been conducted within the past 5 years.

3. “A blood stool test is a test that may utilize a special kit at home to determine whether the stool contains blood. Have you ever had this test utilizing a home kit?” and reports it having been conducted within the past 3 years.

4. “Have you ever had a virtual colonoscopy?” and reports it having been conducted within the past 5 years. This question was only collected in the 2020 BRFSS.

Questions A–D were extracted as a ‘yes’ or ‘no.’ For overall trends, persons were considered to have met the CRC screening guidelines if they met criteria from any of the above (A–D) questions. Responses of “don’t know/unsure” or “refused” were excluded from our analysis. We also extracted demographic variables of self-reported sex, age, and self-reported race/ethnicity and organized these characteristics according to BRFSS-determined categories.

Statistical analysis

Utilizing Stata 16.1 (StataCorp LLC, College Station, TX), we calculated the prevalence of individual meeting guidelines for CRC screening between 2018, 2019, and 2020 according to the type of CRC screening completed and also by subpopulations of race/ethnicity and sex. We then utilized design-based, chi-square (X ²) tests of independence to discover possible associations between these subpopulations per year. An alpha level of 0.05 for all statistical tests was utilized in this analysis.

Demographics

Total response for colonoscopies across 2018, 2019, and 2020 BRFSS cycles were 200,844, 9,724, and 177,922, respectively; for sigmoidoscopy were 94,966, 2,836, and 171,486; stool samples were 94,966, 2,836, and 171,486. Across types of CRC screening and data cycles, samples of males and females responding to CRC screening questions were comparable.

For all respondents, those who claimed to receive a colonoscopy within the last 10 years trended from 64.1 to 70.1%–63.4%, respectively for the years 2018–2020 (Table 1). For those receiving a sigmoidoscopy within 5 years, a trend of 2.5, 4.0, and 3.8% was noted. For a blood stool sample within 3 years, we noted a trend of 18.7, 17.6, and 20.3% of the respondents performing this test (Figure 1). Deviations in trends for colonoscopy (X ² _{(DF=1.84, 7.1e+05)}=26.31, p<0.001) and sigmoidoscopy (X ² _{(DF=1.64, 4.4e+05)}=39.41, p<0.001) were statistically significant; however, they were not statistically significant for stool sample screenings (X ² _{(DF=1.78, 8.3e+05)}=1.75, p=0.18).

Figure 1:

Trends in colorectal screening type from 2018 to 2020.

Sex

Regarding CRC screening trends between sexes, colonoscopy usage in males trended from 62.3 to 68.6%–62.5% for the years 2018–2020, respectively (Table 1). For females, the trend for colonoscopies was 65.8–71.4% to 64.3%. Sigmoidoscopy screening in males trended from 2.6 to 4.4%–4.6%. Female sigmoidoscopy trends were 2.4–3.5% to 3.1%. For the blood stool test, males trended from 18.3 to 15.2%–17.4%. Females trended from 19.0 to 19.6%–19.1% for blood stool tests (Figure 2). Differences in proportions of males and females completing colonoscopy (X ² [DF=1, 198,831]=48.55, p<0.001, X ² [DF=1, 9,664]=4.09, p<0.04, X ² [DF=1, 176,343]=9.14, p<0.003) were significant among all cycles of BRFSS. Only in 2019 and 2020 were these differences between genders significant for blood stool screenings (X ² [DF=1, 9,504]=14.67, p<0.001, X ² [DF=1, 248,852]=16.07, p<0.001). Differences in proportions of males and females completing sigmoidoscopies was significant only in 2020 (X ² [DF=1, 169,910]=39.66, p<0.001).

Figure 2:

Trends of colorectal screening by type and sex.

Race

All race/ethnicity groups most often completed colonoscopies, with all but individuals identifying as Hispanic (48.6%) having higher than 56% completion rates (Table 2). However, individuals reporting as Hispanic received more blood stool tests than other races at 23.4%. Assuming that individuals meeting the USPSTF guidelines in one category did not overlap in another screening method, average CRC screening among all methods showed that 89.7% of individuals who reported as being White completed screening, 91.3% of those self-reporting as Black, and 81.9% with race not listed. Individuals identifying as Asian (74.4%), American Indian/Alaska Native (AI/AN [79.2%]) and Hispanic (78.1%) had lower rates of screening overall (Figure 3).

Figure 3:

Trends in colorectal screening methods by race/ethnicity.

Clinical implications and recommendations for osteopathic practice

The importance of screening for CRC necessitates that all populations should receive screenings regardless of race, gender, or year. Based on the gaps discovered in our study, our team recommends that efforts be made by osteopathic physicians to improve CRC screening, especially in minority populations who remain at increased risk. A study done by Sajaadi et al. investigated internet search trends regarding colon cancer after the death of Chadwick Boseman, a prominent actor in the United States and in the African American community [22]. Their findings showed an increase in CRC website traffic as well as an increase in revenue to prominent cancer organizations following Chadwick Boseman’s death [22]. Their results suggest that public awareness may improve CRC screening among priority populations. We recommend that more efforts be made to increase public awareness of CRC and the availability of screening tools to prevent this disease.

Furthermore, a systematic review by Naylor et al. [23] found that educating patients on the process of CRC screening, such as appointment setting, bowel preparations, and appointment reminders, all contributed to improvements in CRC screening rates among the minority populations tested. Another study found that health experience variables are most important in increasing CRC screening when compared to demographics or attitudes toward CRC screenings [24]. This study proposed that improving the healthcare experience should be addressed first over improving patient attitudes and beliefs in order to improve the rates of CRC screening for minority groups. The variables measured in this experience included whether patients had a primary care doctor, whether they had ever received a doctor recommendation for a CRC screening test, and their satisfaction with their doctor or previous provider [24]. Greiner et al. [25] investigated CRC screening trends in patients with multiple comorbidities. Their results are interesting, revealing that patients with five or more comorbidities were significantly less likely to be screened for colon cancer when compared to patients with two to four conditions [25]. This could suggest that preventative measures like CRC screening could fall low on the priority list as physicians tackle more pressing, acute problems in patients with extensive health conditions. Nevertheless, based on the results of the previously mentioned studies, our team recommends addressing these population gaps in CRC screening by improving patient education in the CRC screening process and healthcare experiences to increase CRC screening in minority populations.

Limitations

Our study’s strengths include the use of a large, nationally representative sample of the US population. In addition, our study is not without its limitations. First, due to the nature of data collection being in survey form, the respondents’ answers are subject to response bias. BRFSS is also a survey dependent on respondents’ understanding of their own medical care, which is unverified by survey providers. Additionally, the pandemic of COVID-19 likely contributed to data trend changes, leading to potential confounding for interpreting the data trends for the year 2020. Future studies should examine the rates of CRC screening usage utilizing a longitudinal study design. Additionally, future studies should assess specific reasons for why certain populations are more likely to choose inferior CRC screening methods. Another potential area of future interest is determining the cause of lower BRFSS reporting for all CRC screening in the year 2019.