Association between radiologists' and facilities' characteristics and mammography screening detection of ductal carcinoma in situ.

PURPOSE: The aim of our study was, first, to measure association between radiologists and facilities characteristics and DCIS detection. Second, to assess whether those characteristics affect differently the likelihood of detection of DCIS versus invasive breast cancer. When applicable, we examined whether the identified characteristics were similarly associated with low-grade and high-grade DCIS detection. METHODS: This retrospective cohort study included 1,750,002 digital screening mammograms (2145 screen-detected DCIS) performed in the Quebec breast cancer screening program between 2007 and 2015 inclusively. The associations between radiologists' and facilities' characteristics and (1) the DCIS detection rate, (2) the invasive detection rate, and (3) the odds of DCIS on invasive detection were assess. For statistically significant associations in the latter analysis, analyses stratified by DCIS grade were performed. Multivariable logistic regression with generalized estimating equations estimates to account for correlation among mammograms was used. RESULTS: Compared to radiologists with recall rate between 5.0 and 9.9%, radiologists with recall rate between 15.0-19.9% and ≥ 20% reached a higher DCIS detection rate, with adjusted detection ratios of, respectively, 1.33 (95% confidence interval = 1.15-1.53) and 1.43 (95% confidence interval = 1.13-1.81). Increase in radiologist' recall rate was statistically significantly associated with an increase in detection of low/intermediate-grade DCIS (P < 0.001), while not in high-grade DCIS (P = 0.15). CONCLUSIONS: A major determinant of DCIS detection is the radiologists' recall rate. Abnormalities referred by radiologists with higher recall rates should be identified in order to understand how to decrease recall rate while keeping an optimal DCIS and invasive detection rate.

The mammography screening detection of ductal carcinoma in situ and invasive breast cancer according to women's characteristics: is it the same?

PURPOSE: Detection of ductal carcinoma in situ (DCIS) has increased with the mammography dissemination. Given the potential role of DCIS as a precursor of invasive breast cancer (IBC), we aimed to assess whether women's characteristics have a different effect on the DCIS compared to IBC detection rate. METHODS: This study included 3,609,569 screening mammograms performed from 2002 to 2015 in our organized breast cancer screening program, which actively invites women 50-69 years of age. The association between women's characteristics and the DCIS detection rate, the IBC detection rate and the odds ratio of DCIS among screen-detected cancers was assessed by logistic regression and generalized estimating equations with independent correlation matrix and sandwich estimator. RESULTS: A total of 4173 DCIS and 15,136 IBC were screen-detected. Increasing women's age, current hormone replacement therapy use and higher body mass index were less associated with the DCIS than with IBC detection rates (p value for the odds of DCIS among screen-detected cancers of, respectively, < 0.0001, 0.0244 and < 0.0001). In contrast, having a previous breast aspiration or biopsy and increasing breast density were more strongly associated with DCIS than with IBC detection rates (p value of, respectively, 0.0050 and < 0.0001). CONCLUSION: The results suggest that some women's characteristics could be playing a role in the initiation and other in the progression from in situ to invasive breast cancer. These characteristics can also affect the screening sensitivity, and this effect may differ depending on whether screen-detected cases were DCIS or IBC.

Screening Sensitivity According to Breast Cancer Location.

PURPOSE: To examine the relation between breast cancer location and screening mammogram sensitivity, and assess whether this association is modified by body mass index (BMI) or breast density. METHODS: This study is based on all interval cancers (n = 481) and a random sample of screen-detected cancers (n = 481) diagnosed in Quebec Breast Cancer Screening Program participants in 2007. Film-screening mammograms, diagnostic mammograms, and ultrasound reports (when available) were requested for these cases. The breast cancer was then localised in mediolateral oblique (MLO) and craniocaudal (CC) projections of the breast by 1 experienced radiologist. The association between cancer location and screening sensitivity was assessed by logistic regression. Adjusted sensitivity and sensitivity ratios were obtained by marginal standardisation. RESULTS: A total of 369 screen-detected and 268 interval cancers could be localised in MLO and/or CC projections. The 2-year sensitivity reached 68%. Overall, sensitivity was not statistically associated with location of the cancer. However, sensitivity seems lower in MLO posterior inferior area for women with BMI ≥ 25 kg/m2 compared to sensitivity in central area for women with lower BMI (adjusted sensitivity ratio: 0.58, 95% confidence interval [CI]: 0.17-0.98). Lower sensitivity was also observed in subareolar areas for women with breast density ≥ 50% compared to the central areas for women with lower breast density (for MLO and CC projections, adjusted sensitivity ratio and 95% CI of, respectively, 0.54 [0.13-0.96] and 0.46 [0.01-0.93]). CONCLUSIONS: Screening sensitivity seems lower in MLO posterior inferior area in women with higher BMI and in subareolar areas in women with higher breast density. When interpreting screening mammograms, radiologists need to pay special attention to these areas.

Mammography Clinical Image Quality and the False Positive Rate in a Canadian Breast Cancer Screening Program.

PURPOSE: The study sought to determine if mammography quality is associated with the false positive (FP) rate in the Quebec breast cancer screening program in 2004 and 2005. METHODS: Mammography quality of a random sample of screen-film mammograms was evaluated by an expert radiologist following the criteria of the Canadian Association of Radiologists. For each screening examination, scores ranging from 1 (poor quality) to 5 (excellent quality) were attributed for positioning, compression, contrast, exposure level, sharpness, and artifacts. A final overall quality score (lower or higher) was also given. Poisson regression models with robust estimation of variance and adjusted for potential confounding factors were used to assess associations of mammography quality with the FP rate. RESULTS: Among 1,209 women without cancer, there were 104 (8.6%) FPs. Lower overall mammography quality is associated with an increase in the FP rate (risk ratio [RR], 1.4; 95% confidence interval [CI], 1.0-2.1; P = .07) but this increase was not statistically significant. Artifacts were associated with an increase in the FP rate (RR, 2.1; 95% CI, 1.3-3.3; P = .01) whereas lower quality of exposure level was related to a reduction of the FP rate (RR, 0.4; 95% CI, 0.1-1.0; P = .01). Lower quality scores for all other quality attributes were related to a nonstatistically significant increase in the FP rate of 10%-30%. CONCLUSIONS: Artifacts can have a substantial effect on the FP rate. The effect of overall mammography quality on the FP rate may also be substantial and needs to be clarified.

Clinical Image Quality and Sensitivity in an Organized Mammography Screening Program.

PURPOSE: The study sought to examine the association between clinical image quality of mammograms and screening sensitivity. METHODS: Four radiologists evaluated the clinical image quality of 374 invasive screen-detected cancers and 356 invasive interval breast cancers for which quality evaluation of screening mammograms could be assessed from cancers diagnosed among participants in the Quebec Breast Cancer Screening Program in 2007. Quality evaluation was based on the Canadian Association of Radiologists accreditation criteria, which are similar to those of the American College of Radiology. The association between clinical quality and screening sensitivity was assessed by logistic regression. Adjusted sensitivity and adjusted sensitivity ratios were obtained through marginal standardization. No institutional review board approval was required. RESULTS: A proportion of 28% (206 of 730) of screening mammograms had lower overall quality for the majority of assessments. Positioning was the quality attribute that was the most frequently deficient. The 2-year screening sensitivity reached 68%. Sensitivity of screening was not statistically associated with the overall quality (ratio of 2-year sensitivity = 1.03; 95% confidence interval: 0.93-1.15) or with any quality attributes (positioning, exposure, compression, sharpness, artifacts, contrast). Results were similar for the 1-year sensitivity. CONCLUSIONS: Although not all mammograms in the Quebec screening program met the optimum quality required by the Canadian Association of Radiologists or American College of Radiology accreditation, the screening mammograms produced in this population-based organized screening program reached a high enough level of quality so that the remaining variation in quality is too little to impair screening sensitivity.

Detection Rate, Recall Rate, and Positive Predictive Value of Digital Compared to Screen-Film Mammography in the Quebec Population-Based Breast Cancer Screening Program.

PURPOSE: The study sought to compare performance indicators of computed radiography (CR) using different plate readers, digital direct radiography (DR), and screen-film mammography (SFM) in a population-based screening program. METHODS: This analysis involved women 50-69 years of age who participated in the breast screening program of Quebec (Canada) and who had screening mammogram between January 1, 2007, and September 30, 2012. The detection rate, recall rate, and positive predictive value of CR (n = 672,125 mammograms) and DR (n = 60,023) were compared to SFM (n = 782,894) using mixed-effect logistic regression, adjusting for potential confounders. No institutional review board approval was required. RESULTS: CR was not associated with change in cancer detection rate (odds ratio [OR]: 0.95; 95% confidence interval [CI]: 0.88-1.03), but with a small increase in recall rate (OR: 1.03; 95% CI: 1.01-1.06) compared to SFM. The association of CR with recall rate varies with the CR plate reader manufacturer (P < .0001). DR was not associated with change in detection rate (OR: 1.06; 95% CI: 0.89-1.25), but with an increase in the recall rate (OR: 1.25; 95% CI: 1.19-1.30) compared to SFM. CONCLUSIONS: In our screening program, digital mammograms gave detection rates equivalent to those of SFM, but with an increase of recall rate, particularly for DR. If this situation persists, the adoption of DR may increase the adverse effects of screening with little or no benefit for women.

Clinical image quality in daily practice of breast cancer mammography screening.

OBJECTIVE: To assess the quality of screening mammograms performed in daily practice in the Quebec Breast Cancer Screening Program. SUBJECTS AND METHODS: Clinical image quality of a random subsample of 197 screening mammograms performed in 2004-2005 was independently evaluated by 2 radiologists based on the criteria by Canadian Association of Radiologists (CAR). When disagreement occurred for overall judgement or positioning score, the mammograms were reviewed by a third radiologist. Cohen's kappas for interrater agreement were computed. Multivariable robust Poisson regression models were used to study associations of overall quality and positioning with body mass index (BMI) and breast density. RESULTS: The CAR criteria were not satisfied for 49.7% of the mammograms. Positioning was the quality attribute most often deficient, with 37.2% of mammograms failing positioning. Interrater agreement ranged from slight (kappa = 0.02 for compression and sharpness) to fair (kappa = 0.30 for exposure). For overall quality, women with a BMI ≥ 30 kg/m(2) had a failure proportion of 67.5% compared with 34.9% for women with a BMI<25 kg/m(2) (risk ratio 2.1 [95% confidence interval, 1.5-3.0]). For positioning, women with a BMI ≥ 30 kg/m(2) had a failure proportion of 53.8% compared with 27.9% for women with a BMI < 25 kg/m(2) (risk ratio 1.9 [95% confidence interval, 1.2-3.1]). Effects of breast density on image quality differed among radiologists. CONCLUSION: Despite measures to ensure high-quality imaging, including CAR accreditation, approximately half of this random sample of screening mammograms failed the CAR quality standards. It would be important to define quality targets for screening mammograms carried out in daily practice to interpret such observations.

Technologists' Characteristics and Quality of Positioning in Daily Practice in a Canadian Breast Cancer Screening Program.

RATIONALE AND OBJECTIVES: This study evaluates to what extent technologists' experience, training, or practice in mammography are associated with screening mammography positioning quality. MATERIALS AND METHODS: Positioning quality of a random sample of 1278 mammograms drawn from the 394,190 screening examinations performed in 2004-2005 in the Breast Cancer Screening Program of Quebec (Canada) was evaluated by an expert radiologist. Information on technologists' experience, training, and practice was obtained by mailed questionnaire. Multivariable Poisson regression models with robust estimation of variance were used to assess the association of technologists' characteristics with higher positioning quality. RESULTS: Of 254 randomly selected technologists, 220 (86.6%) completed the questionnaire. Participating technologists did 89.2% of available sampled mammograms (1088 of 1220), of which 45.9% were of higher positioning quality. Technologists who, in addition to mandatory training, followed at least 15 hours of hands-on training in positioning performed higher positioning quality (adjusted ratio = 1.3, 95%CI = 1.1-1.5) than technologists with no such additional training. Technologists providing at least 15 hours of continued medical education also performed higher positioning quality (adjusted ratio = 1.3, 95%CI = 1.1-1.5) than those who provided less than 15 hours of continued medical education. Being involved in film development and proportion of mammograms performed that are screening compared to diagnostic were also associated with positioning quality, although the latter association was less clear. CONCLUSIONS: Extra hands-on training in positioning could further improve screening mammography positioning quality in the screening program because many technologists did not have such additional training.

Volume of screening mammography and performance in the Quebec population-based Breast Cancer Screening Program.

BACKGROUND: In the Quebec Breast Cancer Screening Program (Programme quebecois de depistage du cancer du sein [PQDCS]), radiologists' and facilities' volumes of screening mammography vary considerably. We examined the relation of screening-mammography volume to rates of breast cancer detection and false-positive readings in the PQDCS. METHODS: The study population included 307,314 asymptomatic women aged 50-69 years screened during 1998-2000. Breast cancer detection rates were analyzed by comparing all women with screening-detected breast cancer (n = 1709) and a 10% random sample of those without (n = 30,560). False-positive rates were analyzed by comparing the 3159 women with false-positive readings and the 27,401 others in the 10% random sample. Characteristics of participants, radiologists and facilities were obtained from the PQDCS information system. Data were analyzed by means of logistic regression. RESULTS: The rate of breast cancer detection appeared to be unrelated to the radiologist's screening-mammography volume but increased with the facility's screening-mammography volume. The breast cancer detection rate ratio for facilities performing 4000 or more screenings per year, compared with those performing fewer than 2000, was 1.28 (95% confidence interval [CI] 1.07-1.52). In contrast, the frequency of false-positive readings was unrelated to the facility's screening volume but was inversely related to the radiologist's screening volume: the rate ratio for readers of 1500 or more screenings per year compared with those reading fewer than 250 was 0.53 (95% CI 0.35-0.79). INTERPRETATION: Radiologists' and facilities' caseloads showed independent and complementary associations with performance of screening mammography in the PQDCS. Radiologists who worked in larger facilities and read more screening mammograms had higher breast cancer detection rates while maintaining lower false-positive rates.

Radiologist interpretive volume and breast cancer screening accuracy in a Canadian organized screening program.

BACKGROUND: To strengthen evidence on which radiologist mammography interpretive volume requirements can be based, we assessed the relation of volume to accuracy in the Quebec Breast Cancer Screening Program. METHODS: Annual interpretive volume (total, screening, and diagnostic) for all 340 radiologists who interpreted 1315327 screening examinations in the period from 2000 to 2006 was obtained using provincial databases. The association of volume to sensitivity, false-positive rate, and accuracy (sensitivity/false-positive rate) was assessed by multivariable Poisson regression with robust error variance. All statistical tests were two-sided. RESULTS: Radiologists consistently interpreting less than 500 mammograms annually experienced a 58% reduction in accuracy (adjusted accuracy ratio = 0.42; 95% confidence interval [CI] = 0.24 to 0.74) compared with those who consistently interpreted at least 500 mammograms annually. Moreover, accuracy increased progressively as total annual volume increased (P trend = .0005). Radiologists interpreting at least 4000 mammograms annually experienced a 32% increase in accuracy (adjusted accuracy ratio = 1.32; 95% CI = 1.13 to 1.54) compared with those interpreting 500 to 999 mammograms annually. This increase in accuracy is attributable to a reduction in false-positive rate as total volume increased (P trend = .001). Sensitivity changed little with total volume (P trend = .68). Gains in accuracy were greater up to approximately 3000 mammograms interpreted annually. CONCLUSIONS: The minimum annual volume of 500 mammograms required in North America is justified; radiologist accuracy may be compromised if interpretive volume is consistently less than this requirement. Raising interpretive volume may help to reduce the frequency of false positives without loss of sensitivity. Possible gains in accuracy may be greater with increases in volume of up to approximately 3000 mammograms interpreted annually.

Contribution and performance of mobile units in an organized mammography screening program.

BACKGROUND: The aims of this study were to evaluate the contribution of mobile mammography units to participation rate and to compare their performance to fixed screening centres within the organized mammography screening program of Quebec, Canada. METHODS: The study is based on all screening mammograms carried out in women aged 50-69 who participated in the Québec program from 2002 to 2010. Performance was measured by screening sensitivity, false-positive rate (1-specificity), positive likelihood ratio as well as abnormal call rate, detection rate, interval cancer rate, positive predictive value, and tumour characteristics. Poisson regression models with robust variance estimation were used to take into account the multi-level structure of the data. All models were adjusted for characteristics related to women. RESULTS: During the 2002-2010 period, 2,292,592 screening mammograms were performed, of which 42,279 (1.8%) were in mobile units. In regions serviced exclusively by mobile units, the participation rate reached an average of 63.4% during the 2006-2010 period compared to 54.7% for the entire study population. Estimated sensitivity was similar to that of fixed sites (rate ratio = 0.98 [0.84-1.15]) while the false-positive rate was lower (rate ratio = 0.76 [0.57-1.02]) although this difference was of marginal statistical significance (p=0.07). CONCLUSIONS: In this program, mobile mammography units allowed regions lacking a fixed centre to attain participation rates slightly higher than those in the rest of Quebec, without loss of sensitivity and with some gain in the false-positive rate.