Abdominal Imaging

, Volume 36, Issue 5, pp 524–531 | Cite as

Physician knowledge and appropriate utilization of computed tomographic colonography in colorectal cancer screening

  • Matthew S. Chang
  • Jessica P. Shah
  • Sunil Amin
  • Susana Gonzalez
  • Joan C. Prowda
  • Joan M. Cheng
  • Elizabeth C. Verna
  • Don C. Rockey
  • Harold Frucht
Original Paper

Abstract

Goals

To assess physician understanding of computed tomographic colonography (CTC) in colorectal cancer (CRC) screening guidelines in a pilot study.

Background

CTC is a sensitive and specific method of detecting colorectal polyps and cancer. However, several factors have limited its clinical availability, and CRC screening guidelines have issued conflicting recommendations.

Study

A web-based survey was administered to physicians at two institutions with and without routine CTC availability.

Results

398 of 1655 (24%) participants completed the survey, 59% was from the institution with routine CTC availability, 52% self-identified as trainees, and 15% as gastroenterologists. 78% had no personal experience with CTC. Only 12% was aware of any current CRC screening guidelines that included CTC. In a multiple regression model, gastroenterologists had greater odds of being aware of guidelines (OR 3.49, CI 1.67–7.26), as did physicians with prior CTC experience (OR 4.81, CI 2.39–9.68), controlling for institution, level of training, sex, and practice type. Based on guidelines that recommend CTC, when given a clinical scenario, 96% of physicians was unable to select the appropriate follow-up after a CTC, which was unaffected by institution.

Conclusions

Most physicians have limited experience with CTC and are unaware of recent recommendations concerning CTC in CRC screening.

Keywords

Guidelines Medical education Survey Colorectal cancer Screening CT Colonography 

Computed tomographic colonography (CTC) is a noninvasive method of detecting colorectal polyps and cancer with high sensitivity and specificity [1, 2, 3, 4, 5, 6, 7, 8]. There are many different modalities available for the detection of adenomatous colorectal polyps, but colorectal cancer (CRC) screening participation remains low. Approximately 50% of eligible patients actually undergo screening [9]. There are many factors that account for this, including socioeconomic and educational [10, 11]. CTC is a non-invasive option for CRC screening that has the potential to improve screening rates and reduce mortality from CRC [4].

There are several reasons why CTC has not become a widely adopted CRC screening modality. While constantly improving, the sensitivity and specificity of CTC varies widely (Table 1) depending on the size of the polyp (diminutive, small, or large), the patient population being studied (average vs. high risk), and the CTC technique employed (2D vs. 3D, stool tagging or not) [12]. A steep learning curve for this highly interpreter-dependent technology has been demonstrated [13, 14, 15]. CTC is also primarily a diagnostic tool without therapeutic capabilities, such as polyp removal. Furthermore, clinical expertise in CTC is currently available in only a limited number of institutions.
Table 1

Per patient sensitivities and specificities according to the literature

Reference

Diminutive polyp

Small polyp

Large polyp

Pickhardt et al. [4] (n = 1233)

≥6 mm

≥10 mm

 Sensitivity (95% CI)

88.7 (82.9–93.1)

93.8 (82.8–92.7)

 Specificity (95% CI)

79.6 (77.0–82.0)

96.0 (94.8–97.1)

Halligan et al. [1] (n = 2610, meta analysis)

≥5 mm

 Sensitivity (95% CI)

45–97

86 (75–93%)

93 (73–98)

 Specificity (95% CI)

26–97

86 (76–93%)

97 (95–99)

Rockey et al. [6] (n = 614)

6–9 mm

≥10 mm

 Sensitivity (95% CI)

51 (41–60)

59 (46–71)

 Specificity (95% CI)

89 (86–92)

96 (94–98)

Mulhall et al. [3] (n = 6393, meta analysis)

<6 mm

6–9 mm

>9 mm

 Sensitivity (95% CI)

48 (25–70)

70 (55–84)

85 (79–91)

 Specificity (95% CI)

92 (89–96)

93 (91–95)

97 (96–97)

Johnson et al. [2] (n = 2531)

≥5 mm

≥10 mm

 Sensitivity (95% CI)

65 (58–73)

90 (84–96)

 Specificity (95% CI)

89 (85.1–92.3)

86 (81.3–90.0)

Graser et al. [8] (n = 307)

>9 mm

 Sensitivity (95% CI)

92.0 (74.0–99.0)

 Specificity (95% CI)

97.9 (95.4–99.2)

Regge et al. [7] (n = 937)

6–9 mm

≥10 mm

 Sensitivity (95% CI)

56.7 (44.0–68.8)

80.3 (72.2–87.0)

 Specificity (95% CI)

In addition, recent consensus CRC screening guidelines have issued conflicting recommendations for and against CTC. In March 2008, joint guidelines were issued from the American Cancer Society, U.S. Multi-Society Task Force on CRC, and the American College of Radiology that, for the first time, specifically recommended CTC as an acceptable option for CRC screening [16]. However, in October 2008, the U.S. Preventive Services Task Force published guidelines that determined there was insufficient evidence to assess the balance of benefits and harm of CTC [17]. Largely, as a result of this position piece, in 2009, the Centers for Medicare and Medicaid Services (CMS) decided not to cover screening CTC (https://www.cms.hhs.gov/mcd/viewdraftdecisionmemo.asp?id=220). These opposing statements have caused confusion and controversy regarding the clinical role of CTC in CRC screening [18].

Several studies have evaluated the degree of training required by “readers”, those interpreting CTC, which may include radiologists, gastroenterologists, or other physicians, to accurately interpret CTC [19, 20, 21], but no studies have addressed whether the clinicians ordering a CTC are doing so appropriately, based on an understanding of the utility of CTC in CRC screening. According to the NIH State-of-the-Science Conference: Enhancing Use and Quality of CRC Screening (http://consensus.nih.gov/2010/colorectalstatement.htm), more research is needed to understand physician screening recommendations to patients and adherence to guidelines. We hypothesized that physicians are unfamiliar with recent guidelines and the capabilities of CTC, particularly the high sensitivity for large polyps and high specificity for polyps of most sizes. Therefore, we aimed to assess clinicians’ understanding of the current state of technology of CTC and its role in CRC screening in a pilot study at two institutions, one with and one without routine CTC availability.

Methods

Participant recruitment

Physicians from two academic institutions were asked to complete an anonymous, 10 questions, web-based survey about CTC. One institution had routine, low-to-moderate CTC availability (University of Texas Southwestern Medical Center (UTSW)) with approximately 50–100 studies per year, while the other institution did not have routine availability and performed CTC only on a case-by-case basis (Columbia University Medical Center, (CUMC)) with fewer than 10 studies a year. At both institutions CTC was performed with CO2 insufflation and reviewed in 2D/3D with stool tagging. Survey participants were practicing physicians who perform CRC screening on individuals or refer individuals for screening procedures, such as internists (general, gastroenterologists, oncologists), general surgeons, family practitioners, or radiologists. Trainees were included in the study as CRC screening, including CTC, is extensively covered as part of the training curriculum at both academic institutions. At CUMC, program directors, chief residents, and departmental administrators were asked to distribute an email to clinical trainees and attendings that provided a brief description of the study and a link to the survey, which was hosted by a commercially available survey website (www.surveymonkey.com). At UTSW, individual email addresses were able to be obtained, and physicians were emailed directly with a study description and a link to the same survey. Two reminder emails were sent as follow up to non-responders. Recruitment of participants took place from November 2008 to March 2009. The study was approved by the CUMC institutional review board as qualifying for exemption status.

Web-based survey

The initial survey was written and reviewed by the co-investigators and consisted of 10 multiple choice questions with some questions allowing for free-text responses. A pilot survey was then administered to a test group of 10 physicians, and the responses were reviewed with the participants immediately after survey completion to see if the answers reflected what they intended. Based on this trial, we developed a final version of the survey, which collected baseline information (including gender, level of training, field of training, practice type, and prior CTC experience) and evaluated physician knowledge of CTC (including awareness of the role of CTC in current screening guidelines, application of the guidelines in different clinical scenarios, perceived cost, sensitivity, and specificity) (Table 2). Given the wide range of published sensitivities and specificities of CTC, participants were specifically asked in the survey to provide what they believed was the optimal per patient sensitivity and specificity for CTC at different polyp sizes. Responses were compared to the highest reported sensitivities and specificities in the literature available at the time (Table 1). The purpose of the study was to assess general understanding of CTC and not specifics of CTC techniques. Therefore, for the survey, we considered that CTC had an optimal sensitivity of >90% for large polyps (≥10 mm in size), 50%–90% for small polyps (5–9 mm in size), and <50% for diminutive polyps (<5 mm in size), with a specificity of >90% for large, small, and diminutive polyps. When answering clinical scenarios, participants’ responses were compared with the March 2008 joint American Cancer Society, U.S. Multi-Society Task Force on CRC, and the American College of Radiology guidelines as they were the only guidelines that had specific recommendations regarding the use of screening CTC at the time the survey was administered. Answers to questions were considered to be correct when the physician selected for: (1) no polyp, a repeat CTC or colonoscopy in 5 years, (2) diminutive polyp, a repeat CTC or colonoscopy in 5 years, (3) small polyp, repeat CTC in 5 years OR immediate referral for colonoscopy, and (4) large polyp, immediate referral for colonoscopy. Although the recommended interval for small polyps continues to be debated, we opted to remain consistent with this set of guidelines as there are specific interval recommendations.
Table 2

Web-based survey on physicians’ understanding of CTC

1. How would you describe yourself? Please select one from drop down menu.

 Field of practice:

  (a) Internal medicine

  (b) Gastroenterology

  (c) Oncology

  (d) General surgery

  (e) Radiology

  (f) Family medicine

 Level of training:

  (a) Intern

  (b) Resident

  (c) Fellow

  (c) Attending (Junior)

  (d) Attending (Senior)

 Sex:

  (a) Male

  (b) Female

 Practice type:

  (a) Academic

  (b) Private

  (c) Combination of both

2. Where have you learned about CT colonography? (select all that apply)

  (a) Medical journals/texts

  (b) Conferences

  (c) Lay media

  (d) Other (please specify) ________

3. What experience do you have with CT colonography in your clinical practice (select all that apply)?

  (a) have ordered before (please indicate how many) ________

  (b) followed a patient who had before

  (c) have read one (please indicate how many) ________

  (d) no prior experience

4. When optimally performed and read, what is the SENSITIVITY of CT colonography for:

 Diminutive polyps (<5 mm)?

  (a) <50%

  (b) 50–60%

  (c) 60–70%

  (d) 70–80%

  (e) 80–90%

  (f) >90%

 Small polyps (5–9 mm)?

  (a) <50%

  (b) 50%–60%

  (c) 60%–70%

  (d) 70%–80%

  (e) 80%–90%

  (f) >90%

 Large polyps (≥10 mm)?

  (a) <50%

  (b) 50%–60%

  (c) 60%–70%

  (d) 70%–80%

  (e) 80%–90%

  (f) >90%

5. When optimally performed and read, what is the SPECIFICITY of CT colonography for:

 Diminutive polyps (<5 mm)?

  (a) <50%

  (b) 50%–60%

  (c) 60%–70%

  (d) 70%–80%

  (e) 80%–90%

  (f)>90%

 Small polyps (5–9 mm)?

  (a) <50%

  (b) 50%–60%

  (c) 60%–70%

  (d) 70%–80%

  (e) 80%–90%

  (f) >90%

 Large polyps (≥ 10 mm)?

  (a) <50%

  (b) 50%–60%

  (c) 60%–70%

  (d) 70%–80%

  (e) 80%–90%

  (f) >90%

6. How do you think CT colonography compares to regular (optical) colonoscopy with regards to:

 Sensitivity?

  (a) worse than optical colonoscopy

  (b) same as optical colonoscopy

  (c) better than optical colonoscopy

 Specificity?

  (a) worse than optical colonoscopy

  (b) same as optical colonoscopy

  (c) better than optical colonoscopy

7. Are you aware of any current colon cancer screening guidelines that take into account CT colonography?

  (a) yes

  (b) no

8. What is the cost of CT colonography from a PATIENT PERSPECTIVE?

  (a) nothing, covered by insurance

  (b) co-pay, <$500

  (c) co-pay, >$500

  (d) full cost to the patient

9. What is the cost of CT colonography compared to regular (optical) colonoscopy from a SOCIETAL PERSPECTIVE?

  (a) less than optical colonoscopy

  (b) same as optical colonoscopy

  (c) more than optical colonoscopy

10. An average risk patient undergoes CT colonography. The findings are as below. The next step in management for each scenario would be:

 For NO polyp:

  (a) repeat CT now

  (b) repeat CT in 1 year

  (c) repeat CT in 5 years

  (d) repeat CT in 10 years

  (e) refer to colonoscopy now

  (f) refer to colonoscopy in 1 year

  (g) refer to colonoscopy in 5 years

  (h) refer to colonoscopy in 10 years

 For a DIMINUTIVEe polyp (<5 mm):

  (a) repeat CT now

  (b) repeat CT in 1 year

  (c) repeat CT in 5 years

  (d) repeat CT in 10 years

  (e) refer to colonoscopy now

  (f) refer to colonoscopy in 1 year

  (g) refer to colonoscopy in 5 years

  (h) refer to colonoscopy in 10 years

 For a SMALL polyp (5–9 mm):

  (a) repeat CT now

  (b) repeat CT in 1 year

  (c) repeat CT in 5 years

  (d) repeat CT in 10 years

  (e) refer to colonoscopy now

  (f) refer to colonoscopy in 1 year

  (g) refer to colonoscopy in 5 years

  (h) refer to colonoscopy in 10 years

 For a LARGE polyp (≥10 mm):

  (a) repeat CT now

  (b) repeat CT in 1 year

  (c) repeat CT in 5 years

  (d) repeat CT in 10 years

  (e) refer to colonoscopy now

  (f) refer to colonoscopy in 1 year

  (g) refer to colonoscopy in 5 years

  (h) refer to colonoscopy in 10 years

Statistical analysis

Data were summarized descriptively as frequencies and percentages. Logistic regression models were fitted to assess predictors of two outcome variables: (1) the awareness of screening guidelines for CTC (defined as a ‘yes’ answer to survey question 7) and (2) 75% proficiency of knowledge-based CTC scenarios (correctly answering at least three of four scenarios). All analyses were performed with SAS 9.2. P values < 0.05 were considered statistically significant.

Results

Demographics

398 of 1655 (24%) participants completed the survey, 235 (59%) from the institution with routine CTC availability (UTSW), and 163 (41%) from the institution without significant CTC use (CUMC). Of those who responded, 57% were men, 52% self-identified as trainees (23% interns, 63% residents, and 14% fellows), 46% as attending physicians, and 15% as gastroenterologists. The vast majority (91%) worked in an academic medical practice setting (Table 3).
Table 3

Respondent characteristics

Characteristic

Number (%)

Institution

 CUMC

163 (41)

 UTSW

235 (59)

Field of practice

 Internal medicine

169 (42)

 Gastroenterology

59 (15)

 Radiology

51 (13)

 Other

109 (27)

 Unspecified

10 (3)

Level of training

 Trainee

207 (52)

 Attending

182 (46)

 Unspecified

9 (2)

Sex

 Male

228 (57)

 Female

159 (40)

 Unspecified

11 (3)

Practice type

 Academic

363 (91)

 Private practice

3 (1)

 Combination

22 (6)

 Unspecified

10 (3)

CTC information sourcea

 Medical journal/texts

297 (45)

 Conferences

183 (50)

 Lay media

154 (39)

 Other

14 (4)

 Unspecified

7 (2)

Prior CTC experience

 Yes

85 (21)

 No

311 (78)

 Unspecified

2 (1)

Type of CTC experiencea

 Ordered CTC

32 (8)

 Followed a patient who had CTC

36 (9)

 Read CTC

5 (1)

 Combination of ordered/followed/read

12 (3)

CUMC Columbia University Medical Center, UTSW University of Texas Southwestern, CTC computed tomographic colonography

a Respondents were permitted to select more than one choice

Experience with CTC

A large proportion of respondents reported learning about CTC from medical journals/texts and conferences (45% and 50%, respectively) while approximately a third of physicians (39%) derived knowledge of CTC from lay media. A substantial majority of physicians (78%) had no personal experience with CTC. Of those who had prior clinical experience with CTC and specified the number of CTC either ordered or read, only 5/52 (10%) had read or ordered more than 10 studies, the most being 25 studies (median 3, interquartile range 3). Only 12% was aware that current screening guidelines included CTC (Table 4).
Table 4

Physician responses to the survey

Awareness of guidelines that incorporate CTC

Number (%)

Question

  Yes

49 (12)

  No

347 (87)

  Unspecified

2 (1)

Knowledge-based clinical scenarios (application of guidelines):

 No polyp

 

  Correct

211 (53)

  Incorrect

187 (47)

 Diminutive polyp (<5 mm)

  Correct

70 (18)

  Incorrect

328 (82)

 Small polyp (5–9 mm)

  Correct

297 (75)

  Incorrect

101 (25)

 Large polyp (≥10 mm)

  Correct

361 (91)

  Incorrect

37 (9)

Optimal sensitivity of CTC for

 Diminutive polyp (<5 mm)

 

  Correct

154 (39)

  Incorrect

244 (61)

 Small polyp (5–9 mm)

  Correct

121 (30)

  Incorrect

277 (70)

 Large polyp (≥10 mm)

  Correct

241 (61)

  Incorrect

157 (39)

Optimal specificity of CTC for

 Diminutive polyp (<5 mm)

 

  Correct

9 (2)

  Incorrect

389 (98)

 Small polyp (5–9 mm)

  Correct

31 (8)

  Incorrect

367 (92)

Large polyp (≥10 mm)

  Correct

165 (41)

  Incorrect

233 (59)

Sensitivity of CTC compared to optical colonoscopy

  Worse

195 (49)

  Same

154 (39)

  Better

49 (12)

Specificity of CTC compared to optical colonoscopy

  Worse

249 (63)

  Same

125 (31)

  Better

24 (6)

Cost of CTC to patient

  Covered by insurance

247 (62)

  Not covered by insurance

142 (36)

  Unspecified

9 (2)

Cost of CTC to societya

  Less than optical colonoscopy

188 (47)

  Same as optical colonoscopy

89 (22)

  More than optical colonoscopy

116 (29)

  Unspecified

5 (1)

CTC computed tomographic colonography

Responses rounded to nearest percent

Physician beliefs about CTC

In general, the literature reports that CTC, when optimally performed and interpreted, is nearly as sensitive and roughly as specific as optical colonoscopy, which improves as the polyp size increases. In our survey, 49% and 63% of physicians believed that CTC was less sensitive and less specific than optical colonoscopy, respectively. When asked to approximate the optimal sensitivity of CTC, physicians generally believed CTC to be worse even at larger lesion sizes with only 39% correctly identifying the sensitivity of CTC for large lesions (≥10 mm), when compared to estimates found in the literature. Physicians were much worse at correctly identifying the specificity of CTC for diminutive and small polyps, with only 2% and 8% correct, respectively, although there is considerable variability and controversy in the literature regarding polyps of these sizes. For large polyps, 59% believed CTC to be less specific than the literature indicates. Even when expanding the definition of a “correct” answer, from an optimal range of ≥90% down to ≥80%, only 14% and 38% of physicians were able to correctly identify the specificity of CTC for diminutive and small polyps.

While Medicare and most medical insurances do not cover screening CTC, 62% of respondents believed that it was covered. 70% believed the societal cost of CTC would be the same or less than that of colonoscopy, which is consistent with some computer model predictions, although this is debated [22, 23, 24, 25, 26, 27].

Awareness and application of guidelines

The final multivariable model assessing physician awareness of CTC screening guidelines is presented in Table 5. By the Hosmer–Lemeshow goodness-of-fit test, the model predicts the outcome accurately in this data set (P = 0.59). After controlling for institution, level of training, sex, and prior CTC experience, we found that gastroenterologists had a 3.5-fold greater odds of being aware of screening guidelines than non-gastroenterologists (95% CI 1.68–7.27). Furthermore, physicians who reported previous experience with CTC had a 4.8 times greater odds of being aware of the guidelines than those who did not.
Table 5

Multiple logistic regression to predict awareness of screening guidelines for CT colonography (CTC)

 

Estimate (SE)

Odds ratio

95% CI

P

Intercept

−3.45 (0.86)

   

Institution

 CUMC

−0.26 (0.35)

0.78

0.39–1.55

0.472

 Reference group: UTSW

    

Field

 GI

1.25 (0.37)

3.49

1.68–7.27

<0.001

 Reference group: Non-GI

    

Level of training

 Attending

0.08 (0.35)

1.09

0.54–2.17

0.814

 Reference group: trainee

    

Sex

 Female

0.15 (0.35)

1.17

0.59–2.29

0.659

 Reference group: male

    

Practice type

 Academic

0.7 (0.82)

2.03

0.40–10.23

0.390

 Reference group: combination

    

Prior CTC experience

 Yes

1.57 (0.36)

4.81

2.39–9.68

<0.001

 Reference group: no

    

CUMC Columbia University Medical Center, UTSW University of Texas Southwestern

When presented with four clinical management scenarios, as described in Table 2, only 4% of physicians were able to select the appropriate follow-up interval for all four scenarios (without a polyp or with a diminutive, small, or large polyp), the most difficult scenario being for a diminutive polyp, where participants opted to repeat CTC or refer for optical colonoscopy earlier than recommended (Table 4). Specifically, 28% would repeat CTC within 1 year (recommended screening interval is every 5 years), and 46% referred for optical colonoscopy within 1 year (colonoscopy should be offered to patients whose largest polyp is 6 mm or greater) (Table 6). In a multivariable regression model to predict 75% proficiency in the knowledge-based scenarios, no covariates were significantly predictive, including institution, sex, field, awareness of guidelines, level of training, practice type, and prior experience with CTC (data not shown).
Table 6

Physician responses to colorectal cancer screening scenarios according to size of polyp detected on CT colonography

Scenario response

No polyp, n (%)

Diminutive polyp, n (%)

Small polyp, n (%)

Large polyp, n (%)

CTC now

3 (0.8)

3 (0.8)

4 (1.0)

7 (1.8)

CTC 1 year

13 (3.3)

109 (27.4)

39 (9.8)

17 (4.3)

CTC 5 years

211 (53.0)

70 (17.6)

15 (3.8)

4 (1.0)

CTC 10 years

127 (31.9)

12 (3.0)

2 (0.5)

2 (0.5)

OC now

1 (0.3)

120 (30.2)

297 (74.6)

361 (90.7)

OC 1 year

2 (0.5)

62 (15.6)

31 (7.8)

6 (1.5)

OC 5 years

21 (5.3)

17 (4.3)

9 (2.3)

1 (0.3)

OC 10 years

20 (5.0)

5 (1.3)

1 (0.3)

0 (0.0)

CTC computed tomographic colonography, OC optical colonoscopy

Discussion

We evaluated physicians’ understanding of the use of CTC for CRC screening according to current guidelines in a pilot study at two different academic institutions. The findings demonstrate that physicians have a limited understanding of CTC screening guidelines and lack first-hand clinical experience with CTC. While gastroenterologists were much more likely to have prior clinical experience with CTC and were more aware of guidelines than non-gastroenterologists, they were equally unable to select an appropriate follow-up after CTC when given a clinical scenario. A large proportion of the respondents were trainees at major academic institutions who may have less clinical experience overall, but are often the most familiar with the latest advances in CRC screening, including CT colonography, as it is an integral part of the curriculum in all training institutions and a core topic covered in board certification exams. Physicians were found to have many misperceptions regarding the accuracy of CTC, often citing lower sensitivities and specificities when compared to colonoscopy. This may be in part due to debate over the accepted sensitivity and specificity in the literature, which varies depending on the technique used and size cut offs [1, 2, 3, 4, 5, 12]. In addition, this confusion over CRC screening guidelines is probably not limited to CTC alone and likely encompasses other screening modalities, as has been demonstrated previously for fecal occult blood testing [28].

Surprisingly, institutional availability of CTC did not affect physician understanding. This may partly be explained by the finding that most physicians learned about CTC from a combination of the medical literature and conferences, as opposed to firsthand clinical experience. These findings may differ in an institution with remarkably high CTC volume; however, at present there are fewer than five such institutions. In addition, it was a commonly held belief that CTC was covered by most medical insurers, which is incorrect. While most insurance companies reimburse for CTC performed after an incomplete colonoscopy, fewer insurers cover CTC for screening indications. As a result, CTC currently has such a limited role that even within an institution where CTC is readily available it seems that only a few physicians are familiar with the utility of the procedure. It may be that physicians underutilize CTC because they perceive it to be unfamiliar and unreliable, holding it to a higher standard than other CRC screening examinations, such as colonoscopy or stool DNA [18].

From our findings, even if CTC were unequivocally found to be as sensitive, specific, and cost-effective as optical colonoscopy, there is a significant risk of inappropriate utilization in clinical practice. Although our study was limited to only two academic institutions, the vast majority of physicians surveyed were unable to identify key features of CTC and did not follow current guidelines when presented with a clinical scenario. As a result, physicians referred patients for repeat CTC earlier than recommended, and for optical colonoscopy more often than suggested by cost-effectiveness studies; this would significantly increase costs and potentially overload endoscopy suites. Our results, while far from conclusive, suggest that physicians would tend to refer patients with diminutive polyps (<6 mm), which is contrary to guidelines (should consider referral for ≥6 mm), and also refer at intervals earlier than the recommended 5 years. Previous studies have demonstrated similar tendencies among physicians to over-refer for repeat optical colonoscopy [29].

One significant limitation to our study was that it was a pilot study limited to only two academic institutions with overall low-to-moderate CTC use. Also, we had a low response rate, which might reflect a general lack of knowledge or interest regarding CTC among physicians, and may have possibly biased the results, as physicians who either have an interest in or are knowledgeable about CTC are more likely to respond. If we had a higher response rate, it is possible that our study would have identified an even more significant knowledge deficit among physicians, although this is speculative. This was potentially confounded by the fact that individual physician emails were only accessible to the investigators at one of the sites (UTSW) and not the other (CUMC), which hindered the ability to send targeted follow-up reminders and likely resulted in a reduced response rate. Our response rate may have been improved if the survey had been more limited in distribution to only selected individuals within a few departments. Some of the questions, particularly regarding cost, could have been better clarified, specifically stating that we were interested in screening CTC, not CTC for other indications, and better defining terminology, such as “societal perspective” in question 9. Another possible confounder, although it is unlikely to have affected the final survey results, was defining a “small polyp” as 5–9 mm in size, based on older literature, which has since become more standardized using a 6–9 mm cutoff. Also, it would have improved the internal validity of the survey to compare responses about CTC with responses for already accepted CRC screening modalities, such as colonoscopy, and to compare physician knowledge of established guidelines. Some of the questions included in the survey were intended to provide a general snapshot of the current beliefs and practices of physicians and were not meant to be the basis for extensive conclusions. Overall, however, most physicians had no experience with CTC and had only a superficial understanding of its capabilities. Of the physicians that did respond, there was a broad representation by institution, field of practice, level of training, and gender which allowed for some generalizability.

There is also significant variation in the sensitivities and specificities reported in the literature, which in the past often depended on the techniques and polyp size cutoff used. In an attempt to limit response variation, we asked physicians to report what they believed to be the optimal sensitivities and specificities. Although these values may not reflect actual practice at all centers, it demonstrates the potential that CTC holds if adopted more broadly. However, even with choosing lower values for sensitivity and specificity, we still found that physicians were unaware of CTC’s high specificity at smaller polyp sizes.

Currently, there are even more options for CRC screening than ever before, including fecal occult blood testing, flexible sigmoidoscopy, double contrast barium enema, colonoscopy, stool DNA testing, and fecal immunochemical testing, with more options on the horizon. As a CRC screening modality, CTC allows for a total colon evaluation using state of the art technology with few downsides. First, it is non-invasive and does not require sedation, which allows patients to return to work the same day and eliminates the requirement of a post-procedure escort. Second, it is well tolerated by patients with a low risk of perforation, which is further improved with the use of CO2 insufflation [30]. Third, CTC can be interpreted remotely and has the potential to improve access to CRC screening. Finally, CTC technology continues to improve and, with the advent of stool and fluid tagging, has the potential to achieve a long, sought-after goal of CRC screening—the minimal colonic preparation [31].

Although CTC has the potential to improve CRC screening rates and ultimately reduce mortality, it is clear that physicians have a limited understanding of CTC’s current capabilities and limitations. Future studies with a broader sampling of physicians and institutions (including private practice) across the country are needed to further assess physician knowledge and beliefs. We conclude that greater availability and substantial physician education would improve understanding of the appropriate role of CTC in CRC screening and might enhance CRC screening compliance.

Notes

Conflicts of interest statement

None to report.

References

  1. 1.
    Halligan S, Altman D, Taylor S, et al. (2005) CT colonography in the detection of colorectal polyps and cancer: systematic review, meta-analysis, and proposed minimum data set for study level reporting. Radiology 237:893–904PubMedCrossRefGoogle Scholar
  2. 2.
    Johnson C, Chen M, Toledano A, et al. (2008) Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med 359:1207–1217PubMedCrossRefGoogle Scholar
  3. 3.
    Mulhall B, Veerappan G, Jackson J (2005) Meta-analysis: computed tomographic colonography. Ann Intern Med 142:635–650PubMedGoogle Scholar
  4. 4.
    Pickhardt P, Choi J, Hwang I, et al. (2003) Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 349:2191–2200PubMedCrossRefGoogle Scholar
  5. 5.
    Pickhardt P, Nugent P, Mysliwiec P, et al. (2004) Location of adenomas missed by optical colonoscopy. Ann Intern Med 141:352–359PubMedGoogle Scholar
  6. 6.
    Rockey DC, Paulson E, Niedzwiecki D, et al. (2005) Analysis of air contrast barium enema, computed tomographic colonography, and colonoscopy: prospective comparison. Lancet 365:305–311PubMedGoogle Scholar
  7. 7.
    Regge D, Laudi C, Galatola G, et al. (2009) Diagnostic accuracy of computed tomographic colonography for the detection of advanced neoplasia in individuals at increased risk of colorectal cancer. JAMA 301:2453–2461PubMedCrossRefGoogle Scholar
  8. 8.
    Graser A, Stieber P, Nagel D, et al. (2009) Comparison of CT colonography, colonoscopy, sigmoidoscopy and faecal occult blood tests for the detection of advanced adenoma in an average risk population. Gut 58:241–248PubMedCrossRefGoogle Scholar
  9. 9.
    Meissner H, Breen N, Klabunde C, et al. (2006) Patterns of colorectal cancer screening uptake among men and women in the United States. Cancer Epidemiol Biomarkers Prev 15:389–394PubMedCrossRefGoogle Scholar
  10. 10.
    McAlearney AS, Reeves KW, Dickinson SL, et al. (2008) Racial differences in colorectal cancer screening practices and knowledge within a low-income population. Cancer 112:391–398PubMedCrossRefGoogle Scholar
  11. 11.
    Denberg TD, Melhado TV, Coombes JM, et al. (2005) Predictors of nonadherence to screening colonoscopy. J Gen Intern Med 20:989–995PubMedCrossRefGoogle Scholar
  12. 12.
    Rockey DC (2009) Computed tomographic colonography: current perspectives and future directions. Gastroenterology 137:7–14PubMedCrossRefGoogle Scholar
  13. 13.
    Burling D, Halligan S, Altman DG, et al. (2006) CT colonography interpretation times: effect of reader experience, fatigue, and scan findings in a multi-centre setting. Eur Radiol 16:1745–1749PubMedCrossRefGoogle Scholar
  14. 14.
    Taylor SA, Halligan S, Burling D, et al. (2004) CT colonography: effect of experience and training on reader performance. Eur Radiol 14:1025–1033PubMedCrossRefGoogle Scholar
  15. 15.
    Rockey DC, Barish M, Brill JV, et al. (2007) Standards for gastroenterologists for performing and interpreting diagnostic computed tomographic colonography. Gastroenterology 133:1005–1024PubMedCrossRefGoogle Scholar
  16. 16.
    Levin B, Lieberman D, McFarland B, et al. (2008) Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin 58:130–160PubMedCrossRefGoogle Scholar
  17. 17.
    U.S. Preventive Services Task Force (2008) Screening for Colorectal Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med 149:627–637Google Scholar
  18. 18.
    Garg S, Ahnen DJ (2010) Is computed tomographic colonography being held to a higher standard? Ann Intern Med 152:178–181PubMedGoogle Scholar
  19. 19.
    Halligan S, Altman DG, Mallett S, et al. (2006) Computed tomographic colonography: assessment of radiologist performance with and without computer-aided detection. Gastroenterology 131:1690–1699PubMedCrossRefGoogle Scholar
  20. 20.
    Bodily KD, Fletcher JG, Engelby T, et al. (2005) Nonradiologists as second readers for intraluminal findings at CT colonography. Acad Radiol 12:67–73PubMedCrossRefGoogle Scholar
  21. 21.
    Dachman AH, Kelly KB, Zintsmaster MP, et al. (2008) Formative evaluation of standardized training for CT colonographic image interpretation by novice readers. Radiology 249:167–177PubMedCrossRefGoogle Scholar
  22. 22.
    Ladabaum U, Song K, Fendrick AM (2004) Colorectal neoplasia screening with virtual colonoscopy: when, at what cost, and with what national impact? Clin Gastroenterol Hepatol 2:554–563PubMedCrossRefGoogle Scholar
  23. 23.
    Vijan S, Hwang I, Inadomi J, et al. (2007) The cost-effectiveness of CT colonography in screening for colorectal neoplasia. Am J Gastroenterol 102:380–390PubMedCrossRefGoogle Scholar
  24. 24.
    Hur C, Chung DC, Schoen RE, et al. (2007) The management of small polyps found by virtual colonoscopy: results of a decision analysis. Clin Gastroenterol Hepatol 5:237–244PubMedCrossRefGoogle Scholar
  25. 25.
    Pickhardt PJ, Hassan C, Laghi A, et al. (2008) Small and diminutive polyps detected at screening CT colonography: a decision analysis for referral to colonoscopy. AJR Am J Roentgenol 190:136–144PubMedCrossRefGoogle Scholar
  26. 26.
    Pickhardt PJ, Hassan C, Laghi A, et al. (2007) Cost-effectiveness of colorectal cancer screening with computed tomography colonography: the impact of not reporting diminutive lesions. Cancer 109:2213–2221PubMedCrossRefGoogle Scholar
  27. 27.
    Sonnenberg A, Delco F, Bauerfeind P (1999) Is virtual colonoscopy a cost-effective option to screen for colon cancer? Am J Gastroenterol 94:2268–2274PubMedCrossRefGoogle Scholar
  28. 28.
    Nadel M, Shapiro J, Klabunde C, et al. (2005) A national survey of primary care physicians’ methods for screening for fecal occult blood. Ann Intern Med 142:86–94PubMedGoogle Scholar
  29. 29.
    Mysliwiec PA, Brown ML, Klabunde CN, et al. (2004) Are physicians doing too much colonoscopy? A national survey of colorectal surveillance after polypectomy. Ann Intern Med 141:264–271PubMedGoogle Scholar
  30. 30.
    Shinners T, Pickhardt P, Taylor A, et al. (2006) Patient-controlled room air insufflation versus automated carbon dioxide delivery for CT colonography. Am J Roentgenol 186:1491–1496CrossRefGoogle Scholar
  31. 31.
    Iannaccone R, Laghi A, Catalano C, et al. (2004) Computed tomographic colonography without cathartic preparation for the detection of colorectal polyps. Gastroenterology 127:1300–1311PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Matthew S. Chang
    • 1
  • Jessica P. Shah
    • 2
  • Sunil Amin
    • 1
  • Susana Gonzalez
    • 1
  • Joan C. Prowda
    • 3
  • Joan M. Cheng
    • 3
  • Elizabeth C. Verna
    • 1
  • Don C. Rockey
    • 2
  • Harold Frucht
    • 1
  1. 1.Muzzi Mirza Pancreatic Cancer Prevention and Genetics Program, and Division of Digestive and Liver Diseases, Department of MedicineColumbia University College of Physicians and SurgeonsNew YorkUSA
  2. 2.Division of Digestive and Liver Diseases, Department of MedicineUniversity of Texas Southwestern Medical CenterDallasUSA
  3. 3.Department of RadiologyColumbia University College of Physicians and SurgeonsNew YorkUSA

Personalised recommendations