Screening Adenoma Detection Rate Research Articles

S252 Screening With FIT-DNA: Impact on Colonoscopy Withdrawal Time, Adenoma Detection and Endoscopist’s Recommendation for Follow-Up

Introduction: Limited data exist on colonoscopy after abnormal fecal immunochemical test (FIT)-DNA (a.k.a. multitarget stool DNA test). We hypothesized that endoscopists perform a more careful exam (e.g., longer withdrawal time (WT)) and recommend early re-screening after FIT-DNA+/negative colonoscopy due to expectations about test performance. We aimed to A) determine adenoma detection rate (ADR) and other lesion detection rates and B) assess endoscopist behavior regarding WT and recommendations after negative colonoscopy for 3 indications: 1) FIT-DNA+, 2) average-risk screening (SCR) and 3) abnormal fecal occult blood test (FOBT+, guaiac or FIT). Methods: Using GIQuIC data (2019-2022) from 727 endoscopy units, we identified patients aged 50-75 years undergoing colonoscopy for FIT-DNA+, SCR or FOBT+. We excluded colonoscopy with any other indications (e.g., family history), inadequate bowel preparation or incomplete exam. If pathology was obtained but results were not available, the record was excluded from pathology-related outcomes. Generalized estimating equations clustered by endoscopist were used to assess the association between indication and outcomes while adjusting for patient characteristics and endoscopist's screening ADR. Results: >1.8 million colonoscopies were included; demographics varied by indication (Table). FIT-DNA+ was associated with higher ADR (59.6%) than SCR (39.3%, p< 0.0001) and FOBT+ (53.8%, p< .0001) and greater detection of advanced neoplasia and sessile serrated lesions (Table). Among those with no pathology obtained on colonoscopy, WT was longer for FIT-DNA+ than for SCR and FOBT+. Among patients aged 50-65 years with no pathology and ASA < IV, a 10-year colonoscopy was recommended in only 80.1% of FIT-DNA+ patients vs. 87.2% of SCR and 86.5% of FOBT+. In multivariable models, compared to FIT-DNA+, SCR and FOBT+ are associated with shorter WT (p< 0.0001 for both), and lower odds of adenoma detection (OR 0.48 (95% CI 0.47-0.50) and 0.71 (0.68-0.74), respectively) and recommendations for re-screening in < 10 years (OR 0.64 (0.57-0.72) and 0.84 (0.74-0.96), respectively). Conclusion: FIT-DNA+ is associated with greater neoplasia detection than FOBT+ and SCR, but also with longer WT and more recommendations for early re-screening after a negative colonoscopy. Despite lower specificity of FIT-DNA vs. FOBT, endoscopists seem to have greater concern for missed pathology in FIT-DNA+/negative colonoscopies, leading to downstream impacts on healthcare utilization. Table 1. - Colonoscopy for Abnormal FIT-DNA vs. Average Risk Screening vs. Abnormal FOBT:Patient Characteristics, Findings and Follow-Up Recommendations After a Negative Exam FIT-DNA+(n=23,046) Average Risk Screening(n=1,760,840) FOBT+(n=26,455) Age, mean (sd) 64.2 (7.1) 58.7 (7.2) 63.8 (7.2) Male (%)Female (%) 9,939 (43.1)13,107 (56.9) 816,630 (46.4)944,210 (53.6) 12,698 (48.0)13,757 (52.0) White (%)Black of African American (%)Asian (%)Other (%)Unknown/Declined (%) 15,777 (68.5)1085 (4.7)211 (0.9)514 (2.2)5459 (23.7) 1,022,859 (58.1)171,348 (9.7)56,644 (3.2)68,966 (3.9)441,033 (25.1) 16,387 (61.9)2112 (8.0)959 (3.6)1224 (4.6)5773 (21.8) Non-Hispanic (%)Hispanic/Latino (%)Unknown/Decline 14,234 (61.8)560 (2.4)8252 (35.8) 1,018,724 (57.9)107,147 (6.1)634,969 (36.1) 16,691 (63.1)2407 (9.1)7357 (27.8) Total endoscopists contributing colonoscopies 2650 4845 3223 Endoscopist mean ADR‡(sd) with minimum 50 colonoscopies 39.1% (10.2)(Endoscopist n=2467) 39.2% (10.9)(Endoscopist n=3878) 39.5% (10.6)(Endoscopist n=3006) Colonoscopy Findings Adenoma Detection Rate (ADR) 59.6% *,§ 39.3% § 53.8% ADR for Males 67.6% *,§ 46.4% § 61.0% ADR for Females 53.6% *,§ 33.2% § 47.2% Advanced Neoplasia Detection± 22.2% *,§ 7.6% § 17.4% Sessile Serrated Lesion Detection 16.2% *,§ 8.2% † 8.8% Adenocarcinoma Detection 1.5% * 0.3% § 1.4% Analysis of Subgroup with No Polyps on Colonoscopy (i.e., No Pathology Obtained) Median Withdrawal Time, Minutes (interquartile range) (n=4211)8.5 (6.9-11.0) *,§ (n=673,519)7.8 (6.4-9.9) (n=7455)7.8 (6.3-10.1) Next Recommended Colonoscopy (limited to age 50-65, ASA< IV, exclude other or none)< 10 Years (%)10 Years (%) (n=2125)422 (19.9)1703 (80.1) (n=567,568)72,403 (12.8)495,165 (87.2) (n=4301)579 (13.5)3722 (86.5) Next Recommended Colonoscopy (limited to age 50-65, ASA< IV, exclude other or none)≤3 Years (%)4 or 5 Years (%)6-9 Years (%)10 Years (%)Excluded intervals n:OtherNone (n=2125)*,§ 76 (3.6)333 (15.7)13 (0.6)1703 (80.1) 1518 (n=567,568)§ 3840 (0.7)65,289 (11.5)3274 (0.6)495,165 (87.2) 2335974 (n=4301)65 (1.5)478 (11.1)36 (0.8)3722 (86.5) 2919 ‡ADR includes average risk screening only, ages 50-75, photodocumentation of the cecum, adequate bowel preparation.*p<0.0001 compared with average risk screening.§p<0.0001 compared with FOBT+.†p=0.0002 compared with FOBT+. ± Includes either 1) adenoma≥10 mm, with high grade dysplasia, or with villous component; OR 2) sessile serrated polyp (SSP)≥10 mm, SSP with dysplasia, or traditional serrated adenoma; OR 3) adenocarcinoma.

Adenoma Detection Rate (ADR) Irrespective of Indication Is Comparable to Screening ADR: Implications for Quality Monitoring

Adenoma detection rate (ADR) is a key measure of colonoscopy quality. However, efficient measurement of ADR can be challenging because many colonoscopies are performed for non-screening purposes. Measuring ADR without being restricted to screening indication may likely facilitate more widespread implementation of quality monitoring. We hypothesized that the ADR for all colonoscopies, irrespective of the indication, would be equivalent to the ADR for screening colonoscopies. We reviewed consecutive colonoscopies at two Veterans Affairs centers performed by 21 endoscopists over 6 months in 2015. We calculated the ADR for screening exams, non-screening (surveillance and diagnostic) exams, and all exams (irrespective of indication), correcting for within-endoscopist correlation. We then performed simulation modeling to calculate the ADRs under 16 hypothetical scenarios of various indication distributions. We simulated 100,000 trials with 3,000 participants, randomly assigned indication (screening, surveillance, diagnostic, and FIT+) from a multinomial distribution, randomly drew adenoma using the observed ADRs per indication, and calculated 95% confidence intervals of the mean differences in ADR of screening and non-screening indications. Among 2628 colonoscopies performed by 21 gastroenterologists, the indication was screening in 28.9%, surveillance in 48.2% and diagnostic in 22.9%. There was no significant difference in the ADR, 50% (95%CI: 45-56%) for all colonoscopies vs 49% (95%CI: 43-56%) for screening exams (p=.55). ADRs were 56% for surveillance and 38% for diagnostic exams. In our simulation modeling, only one out of 16 scenarios (screening 10%, surveillance 70%, diagnostic 10% and FIT+ 10%) resulted in a significant difference between the calculated ADRs for screening and non-screening indications. In our study, the overall ADR computed from all colonoscopies was not significantly different than the conventional ADR based on screening colonoscopies. Assessing ADR for colonoscopy irrespective of indication may be adequate for quality monitoring, and could facilitate the implementation of quality measurement and reporting. Future prospective studies should evaluate the validity of using overall ADR for quality reporting in other jurisdictions before adopting this method in clinical practice.

Simplifying Measurement of Adenoma Detection Rates for Colonoscopy.

Adenoma detection rate (ADR) is the colonoscopy quality metric with the strongest association to interval or "missed" cancer. Accurate measurement of ADR can be laborious and costly. Our aim was to determine if administrative procedure codes for colonoscopy and text searches of pathology results for adenoma mentions could estimate ADR. We identified US Veterans with a colonoscopy using Current Procedure Terminology (CPT) codes between January 2013 and December 2016 at ten Veterans Affairs sites. We applied simple text searches using Microsoft SQL Server full-text searches to query all pathology notes for "adenoma(s)" or "adenomatous" text mentions to calculate ADRs. To validate our identification of colonoscopy procedures, endoscopists of record, and adenoma detection from the electronic health record, we manually reviewed a random sample of 2000 procedure and pathology notes from the 10 sites. Structured data fields were accurate in identification of colonoscopies being performed (PPV = 0.99; 95% CI 0.99-1.00) and identifying the endoscopist of record (PPV of 0.95; 95% CI 0.94-0.96) for ADR measurement. Simple text searches of pathology notes for adenoma mentions had excellent performance statistics as follows: sensitivity 0.99 (95% CI 0.98-1.00), specificity 0.93 (95% CI 0.92-0.95), NPV 0.99 (95% CI 0.98-1.00), and PPV 0.93 (0.91-0.94) for measurement of ADR. There was no clinically significant difference in the estimates of overall ADR vs. screening ADR (p > 0.05). Measuring ADR using administrative codes and text searches from pathology results is an efficient method to broadly survey colonoscopy quality.

Variations in Screening Adenoma Detection Rate by Specialty of Physicians in a Predominately African American Population

Background: Screening colonoscopy aims to interrupt the adenoma-carcinoma sequence by removing all precancerous adenomatous polyps. Adenomatous polyp detection rate (ADR) can vary between endoscopists as well as between race, age, and risk of colorectal cancer (CRC). The purpose of this study was to compare ADR among academic gastroenterologists (A-GI), non-A-GI, and surgeons for endoscopies performed in the same endoscopic suite of a large medical center with a predominately African American (AA) population.Methods: All screening colonoscopies performed in 2014 for patients aged 62-76 years were identified using the electronic medical records data. Patients with average risk and high risk of CRC defined as having a 'personal history of polyps' or 'family history of CRC', and history of ulcerative colitis and Fecal Occult Blood Test/Fecal Immunochemical Test (FOBT/FIT) positivity were included. Patients with incomplete colonoscopy (defined as failing to achieve cecal intubation or poor preparation) and unrecovered tissue biopsy were excluded. ADR was calculated for three groups of endoscopists: A-GIs, non-A-GIs, and surgeons.Results: A total of 573 screening colonoscopies was analyzed. The endoscopists comprised five A-GIs, eight non-A-GIs, and six surgeons. The majority of patients were of AA decent (71%), female (54%) with an average age of 66 years. Patients classified as average risk comprised 79% of the population. Most of the colonoscopies were performed by A-GI (n=339), followed by non-A-GI (n=144), and surgeons (n=90). The ADR for A-GI was 50% as compared to 32% for non-A-GI (p<0.001) and 25% for surgeons (p<0.001). Also, A-GI were more likely to identify ≥3 adenomas during screening colonoscopies. Significant differences were observed (p<0.001) in the mean time of colonoscopy for A-GI (30 mins) non-A-G (14 mins), and surgeons (18 mins).Conclusion: Significant variation in the ADR between endoscopists belonging to different specialties were observed. Although all appear to achieve acceptable ADR (ie at least 25 for men and 15 for women), academic gastroenterologists had better performance than non-academic GI and surgeons. This may be explained by a significantly longer average duration of procedures for the highest ADR group.

Effect of time of day and specialty on polyp detection rates in Australia.

Adenoma detection rate (ADR) is an important quality metric in colonoscopy. However, there is conflicting evidence around factors that influence ADR. This study aims to investigate the effect of time of day and endoscopist background on ADR and sessile serrated adenoma/polyp detection rate (SSA/P-DR) for screening colonoscopies. Consecutive patients undergoing colonoscopy in 2016 were retrospectively evaluated. Primary outcome was the effect of time of day and endoscopist specialty on screening ADR. Secondary outcomes included evaluation of the same factors on SSA/P-DR and other metrics and collinearity of ADR and SSA/P-DR. Linear regression models were used for association between ADR, time of day, and endoscopist background. Bowel preparation, endoscopist, session, patient age, and gender were adjusted for. Linear regression model was also used for comparing ADR and SSA/P-DR. Chi-square was used for difference of proportions. Two thousand six hundred fifty-seven colonoscopies, of which 558 were screening colonoscopies, were performed. The adjusted mean ADR (screening) was 36.8% in the morning compared with 30.5% in the afternoon (P<0.0001) and was 36.8% for gastroenterologists compared with 30.4% for surgeons (P<0.0001). For every 1-h delay in commencing the procedure, there was a reduction in mean ADR by 3.4%. Using a linear regression model, a statistically significant positive association was found between ADR and SSA/P-DR (P<0.0001). Morning and afternoon sessions and gastroenterologists and surgeons achieved the minimum standards recommended for ADR. Afternoon lists and surgeons were associated with a lower ADR compared with morning and gastroenterologists, respectively. Additionally, SSA/P-DR showed collinearity with ADR.

Sa1049 VALIDITY FOR CALCULATING ADR FOR ALL INDICATIONS CORRELATES WELL WITH THE STANDARD OF THE SCREENING ADR IN THE AUSTRALIAN POPULATION

Determining adenoma detection rate (ADR) is important to ensure quality colonoscopy. The minimum standard of 25% in screening patients aged >50 years is recommended by professional organisations. Despite reasonable uptake of the National Bowel Cancer Screening Program, most patients in Australian endoscopy units are symptomatic (diagnostic) or have a personal history of polyps (surveillance). While patients with a 10 family history of colorectal cancer <55 years fit screening criteria, due to overwhelming referrals of symptomatic patients, pure family history screening numbers are very low. Thus, screening colonoscopies are less representative of the actual work carried out by Australian endoscopists. Rex (2017) demonstrated that combination of screening, surveillance, and diagnostic ADR correlated well with pure screening ADR. However, as this novel review was based at one site, it needs replication. Aim Assess if the ADR in the screening population correlates significantly with the overall ADR for all indications. Method Colonoscopy records from July 2014 to September 2017 were retrospectively examined at two regional hospital endoscopy units in Australia. All procedures were allocated to Screening (positive FOBT or familial history of cancer), Surveillance (previous history of colonic polyps or cancer), or Diagnostic (symptomatic). If patients had more than one indication, diagnostic indications took precedence over surveillance indications, which took precedence over pure screening indications. Additionally, patients with IBD, polyposis syndromes, and preoperative cancer, with inadequate bowel prep, an anastomosis, or obstructive mass, or the caecum was unable to be intubated or aged <50 were excluded. Endoscopists were included in the analysis if they had completed at least 200 colonoscopies in the study period. Results 6315 colonoscopies by 14 endoscopists were analysed. Table 1 shows the ADR for each indication and overall ADR. 51% of procedures completed were for diagnostic indications, then screening (25%) and surveillance (24%).Screening ADR was higher than diagnostic ADR for all endoscopists. 38% of endoscopists had a higher screening ADR than surveillance ADR. Screening ADR was positively correlated with all ADR (r=0.8178, p=0.0003).Using the minimum recommendation of 25% for mixed gender population > 50 years, all endoscopists exceeded this target for all indication groupings individually and using the combined ADR. Discussion Despite the surveillance ADR being slightly below the screening ADR in this population, overall findings correlate with the Rex (2017) study. This supports expansion of ADR calculations to include all procedures. Additionally, it makes calculation of ADR simpler, efficient and more representative for endoscopists who work in units with proportionally high numbers of diagnostic colonoscopies.

Adenoma detection rate metrics in colorectal cancer surveillance colonoscopy.

A target goal for screening adenoma detection rate (S-ADR) of ≥ 25% has been set to define high-quality colonoscopy performance. However, there is no current accepted target goal for ADR in colorectal cancer (CRC) surveillance. This makes quality assessment challenging when physicians perform cancer surveillance colonoscopy but minimal screening procedures. In this cohort study, consecutive colonoscopies performed at either Rush University Medical Center or Rush Oak Park Hospital by a gastroenterologist or colorectal surgeon in average risk screening population and CRC surveillance population were reviewed retrospectively from 2006 to 2012 and prospectively from 2013 to 2016. ADR in first surveillance colonoscopy following surgical resection of CRC (CRC-ADR) was reported in high-quality detectors (HQD) or low-quality detectors (LQD) based on achievement of 25% ADR in consecutive screening colonoscopy in average risk patients. Pearson's correlation was used to describe the association between individual S-ADR and CRC-ADR for colonoscopists. There was a very strong positive correlation (r = 0.88, p = 0.002) between ADR in average risk screening and first time CRC surveillance. For HQD as defined by S-ADR ≥ 25% (n = 10 colonoscopists), the CRC-ADR was 37.7% (78/207, SD 8%) which was very similar to their respective S-ADR of 33.4% (816/2440, p = 0.22). For LQD (n = 5 colonoscopists), the CRC-ADR was 20.2% (40/198) which was similar to their respective S-ADR of 20.1% (119/591, p = 0.99). The CRC-ADR was significantly higher for HQD than for LQD (37.7 vs. 20.2%, p < 0.0001). The major finding of this study is a defined CRC-ADR for HQD based on the ability to achieve S-ADR ≥ 25%. S-ADR strongly correlates with CRC-ADR. CRC-ADR is quite similar to the colonoscopists' respective S-ADR for both HQD and LQD. For colonoscopists who perform limited screening colonoscopies but do perform CRC surveillance colonoscopies, ADR metrics similar to S-ADR to assess quality in colonoscopy could be considered.

Factors Associated with Surveillance Adenoma and Sessile Serrated Polyp Detection Rates.

Adenoma detection rate (ADR) and sessile serrated polyp detection rate (SSPDR) data in surveillance colonoscopy are limited. Our aim was to determine surveillance ADR and SSPDR and identify associated predictors. A retrospective review of subjects who underwent surveillance colonoscopy for adenoma and/or SSP at an academic center was performed. The following exclusion criteria were applied: prior colonoscopy ≤3years, incomplete examination, or another indication for colonoscopy. Patient, endoscopist, and procedure characteristics were collected. Predictors were identified using multivariable logistic regression. Of 3807 colonoscopies, 2416 met inclusion criteria. Surveillance ADR was 49% and, SSPDR was 8%. Higher ADR was associated with: age per year (OR 1.03; 95% CI 1.02-1.04), male gender (OR 1.55; 95% CI 1.29-1.88), BMI per kg/m2 (OR 1.02; 95% CI 1.01-1.04), withdrawal time per minute (OR 1.09; 95% CI 1.07-1.10), and endoscopists' screening ADR (OR 1.01; 95% CI 1.00-1.03). Years since training (OR 0.99; 95% CI 0.98-0.99) was associated with lower ADR. Family history of CRC (OR 1.58; 95% CI 1.02-2.27) and endoscopists' screening ADR (OR 1.40; 95% CI 1.15-1.74) were associated with higher SSPDR. African-American race (OR 0.36; 95% CI 0.10-0.75) and diabetes (OR 0.41; 95% CI 0.21-0.76) were associated with lower SSPDR. For surveillance colonoscopy, nearly half of patients had an adenoma and one in twelve had an SSP. In addition to established factors, BMI, endoscopists' screening ADR, and years since training were associated with ADR, whereas African-American race and diabetes were inversely associated with SSPDR. Further studies are needed prior to integrating surveillance ADR and SSPDR into quality metrics.

Calculating the adenoma detection rate in screening colonoscopies only: Is it necessary? Can it be gamed?

Background Current recommendations are to calculate the adenoma detection rate (ADR) in screening colonoscopies only. The need to confine the measure to screening has not been established. Methods We retrospectively assessed our quality database for whether calculating ADR from screening, surveillance, and diagnostic colonoscopies (overall ADR) would alter conclusions about the performance of colonoscopists, compared to using an ADR based only on screening colonoscopies. We also prospectively tested the extent to which one physician could corrupt the screening-only ADR by changing the procedure indication after reviewing the examination findings. Results For 15 physicians, screening ADRs differed from the overall ADR by a mean of 2.6 percentage points (range 0 - 6.9 percentage points). Using the overall ADR rather than screening ADR changed the ADR from just below to just above the recommended screening threshold for one physician. In the prospective assessment, a single expert colonoscopist utilized indication gaming in patients with both screening and diagnostic indications and was able to increase his apparent screening-only ADR from 48.4 % to 55.1 %. Conclusions Use of an overall ADR rather than screening-only ADR could simplify ADR measurement, increase the number of examinations available to measure ADR, seldom affect whether a doctor meets recommended ADR thresholds, and eliminate the potential for gaming the ADR by changing the colonoscopy indication.

Tu1019 Adenoma Detection Rate (ADR) Irrespective of Indication Is Comparable to Screening ADR: Implications for Quality Monitoring

Tu1019 Adenoma Detection Rate (ADR) Irrespective of Indication Is Comparable to Screening ADR: Implications for Quality Monitoring

Sa1411 Performance Characteristics of Ultra-THIN Versus Standard Colonoscope for Routine Colonoscopy- a Systematic Review and Meta-Analysis

Sa1410 Reporting Adenoma Detection Rate Confidence Intervals Provide Better Feedback About Endoscopists’ Performance Rebecca S. Slack*, William A. Ross, Phillip Lum, Selvi Thirumurthi, Brian R. Weston, Ethan Miller, Jeffrey Lee, Mehnaz A. Shafi, Marta L. Davila, Robert S. Bresalier, Patrick M. Lynch, Gladis Shuttlesworth, Lopa Mishra, John R. Stroehlein, Gottumukkala S. Raju Gastroenterology, Hepatology and Nutrition, University of Texas MD Anderson Cancer Center, Houston, TX; Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX Introduction & Aims: Adenoma detection rate (ADR) is widely accepted as a quality metric for colonoscopists. However it is unclear what is an adequate sample size to calculate ADR, how to present data, and what numbers are required as adjustments are made to ADR thresholds set higher in the future. Methods: 95% Clopper-Pearson binomial confidence intervals were created for ADRs at various sample sizes and cutoffs to demonstrate the variability around the estimate. This is in consideration of how many colonoscopies are needed to reliably estimate the ADR, since confidence interval length varies by sample size and by proportion (wider near 0.50, narrower closer to 0 and 1). The probability of having an ADR at or above the required cut-off at given hypothetical detection rates were created for cut-offs ranging from 5%-40% to represent all possible cut-offs for ADR, men, women and potentially higher cut-offs. Results: A. Sample size to determine ADR: Table 1 presents confidence intervals for current ADR cut-off rates at 25% & 15% from men and women respectively as well as cut-offs of 40% and 30% (achieved in some practices) respectively.a. 50 patients is a reasonable number of colonoscopies to be performed in a year for each gender (100 total colonoscopies, assuming 50 men and 50 women) with current cut-off rates of 25% and 15% in men and women.b. 110 patients is the smallest number of screenings that produce confidence intervals within +/10% at a 40% detection rate (widest interval considered).c. 410 patients is the first sample size to be within +/5% at a 40% detection rate. B. Physician Performance Probability of Reaching Target ADR: Table 2 presents the probabilities of having an annual ADR at 15% in women.a. If a physician’s true ADR is 15%, then only about 50% of the time would the annual ADR actually be 15% or higher regardless of sample size.b. If the physician’s true ADR is 10% (with 50 screening colonoscopies in women in a year), there is a 12% chance of seeing an ADR that year of atleast 15%, but no chance if the true ADR is only 5%.c. If the physician’s true ADR is 20%, there is a 19% chance of having a reported ADR that year below the 15% cut-off. Conclusions: ADRs need to always be presented with confidence intervals to show the reliability of the estimate for each selected time period. Any cut-off selected will come with a risk of an incompetent physician achieving it and a competent physician not achieving it due to random chance. Any decision about cut-offs and consequences of those cut-offs need to keep these situations in mind. Table 1. Confidence Limits at Hypothetical Numbers of Screening Colonoscopies & ADR No. of colonoscopies www.giejou Current ADR Threshold rnal.org Higher ADR Threshold Men 25% Women 15% Men 40% Women 30% N (%) Detected 95% CI ADR N (%) Detected 95% CI ADR N (%) Detected 95% CI ADR N (%) Detected 95% CI ADR

Improving measurement of the adenoma detection rate and adenoma per colonoscopy quality metric: the Indiana University experience

The adenoma detection rate (ADR) is a validated marker of colonoscopy quality. However, the optimal measurement method is unclear. The aims of our study were to (1) define benchmarks for the number of adenomas per screening colonoscopy (APC) quality metric; (2) study the association between ADRs for screening, surveillance, and diagnostic indications; and (3) explore the association of the screening ADR with an overall ADR inclusive of all colonoscopy indications. Retrospective study. University hospital and associated ambulatory surgery center endoscopy units. Patients aged≥50 years who underwent colonoscopy for screening, surveillance, or diagnostic indications by 20 endoscopists between January 1, 1999 and April 30,2012. Colonoscopy. ADR, APC for screening, surveillance, and diagnostic indications. A total of 21,766 colonoscopies were included. The indication was screening in 7434 (34.2%), surveillance in 8338 (38.3%), and diagnostic in 5994 (27.5%). The screening ADRs and APCs were significantly correlated (R= 0.91; P< .0001). For men, an ADR of 25% corresponded to an APC of 0.46 (95% confidence interval [CI], 0.35-0.57); for women, an ADR of 15% corresponded to an APC of 0.20 (95% CI, 0.13-0.27). Overall, the ADR stratified by colonoscopy indication was highest for surveillance, followed by screening, then diagnostic. For men, a screening ADR of 25% corresponded to a surveillance ADR of 31.9% (95% CI, 24.8%-38.9%); for women, an ADR of 15% corresponded to a surveillance ADR of 24.3% (95% CI, 18.3%-30.5%). The corresponding diagnostic ADRs were 17.0% (95% CI, 12.4%-21.6%) and 15.4% (95% CI, 11.5%-19.3%), respectively. There was significant correlation between screening ADR and an overall ADR inclusive of all colonoscopy indications. External generalizability, retrospective design. We propose minimum screening APC detection benchmarks of 0.50 for men and 0.20 for women. ADRs for screening, surveillance, and diagnostic colonoscopy are correlated and can be used to derive a simplified overall ADR inclusive of all colonoscopy indications.

Defining Benchmarks for the Adenoma per Colonoscopy Quality Metric: Presidential Poster

Purpose: The adenoma detection rate (ADR), defined as the proportion of screening colonoscopies where at least one adenoma is found, is currently the prime surrogate measure of colonoscopy performance quality. Minimum ADR targets in average-risk individuals are 25% for men and 15% for women. Unanswered questions remain about the optimal method for measurement of adenoma detection. The mean number of adenomas per colonoscopy (APC) is thought to be less prone to corruption than the ADR and provide a more comprehensive assessment of quality. However, the minimum benchmarks for this metric are unknown. The aim of this study was to define minimum detection benchmarks for the number of adenomas per screening colonoscopy quality metric. Methods: We performed a retrospective review of a prospectively updated database at a University hospital and associated ambulatory surgery center endoscopy units. Reports of screening colonoscopies performed by 20 attending gastroenterologists between 1/1/1999 and 4/30/2012 were reviewed. Colonoscopies performed on patients 50 years or older were included. ADR was defined as the proportion of screening colonoscopies in which at least one adenoma was detected. The mean number of adenomas per screening colonoscopy (APC) was defined as the total number of adenomas detected divided by the number of screening colonoscopies. Pearson correlation coefficient was calculated to evaluate the association of ADR and APC. Linear regression was used to model the relationship between the two rates and to derive mean APC rates which correspond to ADR of 25% in men and 15% in women, respectively. Results: A total of 7,434 screening colonoscopies were included. The screening ADR and APC for each endoscopist are shown in Table 1. The ADRs ranged from 15.7% to 46.2% (2.9-fold difference), while APCs ranged from 0.2 to 1.08 (5.4-fold difference). The screening ADR and APC were significantly correlated (R=0.91, p<0.0001). For males, ADR of 25% corresponded to APC of 0.46 (95% CI 0.35-0.57); for females, ADR of 15% corresponded to APC of 0.20 (95% CI 0.13-0.27). Additional screening ADR-APC correlates are detailed in Table 2.Table 1: Endoscopist-level screening ADR and APCTable 2: Screening ADR-APC correlates (males and females combined)Conclusion: The screening ADR and APC are very strongly correlated. We propose minimum screening APC detection benchmarks of 0.50 for males and 0.20 for females.

Improving Measurement of Adenoma Detection: The Indiana University Experience

Purpose: The adenoma detection rate (ADR) is a validated measure of colonoscopy quality. However, the restriction to screening colonoscopy may be a barrier to uptake. Knowledge of the ADR in nonscreening groups, and its relation to the screening ADR, might allow expansion of the ADR to include surveillance and diagnostic indications and simplify calculations. The aims of this study were: (1) to study the correlation between ADRs for screening, surveillance, and diagnostic indications, and (2) explore the association of the screening ADR with an overall ADR inclusive of all colonoscopy indications. Methods: We performed a retrospective review of a prospectively updated database at a University hospital and associated ambulatory surgery center endoscopy units. Colonoscopies performed by 20 attending gastroenterologists on patients aged ≥ 50 between 1/1/1999 and 4/30/2012 were reviewed. ADR was defined as the proportion of colonoscopies in which at least one adenoma was detected. Pearson correlation coefficients were calculated to evaluate the association of the ADRs of the three colonoscopy indication categories (screening, surveillance, or diagnostic). The second study aim was addressed using a bootstrap analysis. We calculated the Pearson correlation coefficient of screening ADR with overall ADR, and estimated the overall ADR corresponding to screening ADRs of 15%, 20%, 25%, 30%, 40%, and 50% for patient populations with varying proportions of screening, surveillance, and diagnostic colonoscopies. Results: 21,766 colonoscopies were included; indication screening in 7,434 (34.2%), surveillance in 8,338 (38.3%), and diagnostic in 5,994 (27.5%). Endoscopist ADRs are shown in Table 1. A consistent pattern was observed, with ADR highest for surveillance, followed by screening, then diagnostic indications. ADRs for screening and surveillance, and screening and diagnostic indications, were significantly correlated (R=0.77, p<0.0001; R=0.56, p=0.01, respectively). The correlation between ADRs for surveillance and diagnostic was borderline significant (R=0.42, p=0.07). For males, screening ADR of 25% corresponded to surveillance ADR of 31.9% (95% CI: 24.8%-38.9%); for females, ADR of 15% corresponded to a surveillance ADR of 24.3% (18.3%-30.5%). The corresponding diagnostic ADRs were 17.0% (12.4%-21.6%) and 15.4% (11.5%-19.3%), respectively. The results of the bootstrap analysis exploring the association of screening ADR with the overall ADR are shown in Table 2.Table 1: ADRs for screening, surveillance, and diagnostic colonoscopiesTable 2: Screening ADR-Overall ADR associationsConclusion: ADRs for screening, surveillance, and diagnostic colonoscopy are correlated and can be used to derive a simplified overall ADR inclusive of all colonoscopy indications. Our findings provide a framework for endoscopists to calculate their ADRs in different practice settings.