TY - JOUR
T1 - Variation in Screening Abnormality Rates and Follow-Up of Breast, Cervical and Colorectal Cancer Screening within the PROSPR Consortium
AU - Tosteson, Anna N A
AU - Beaber, Elisabeth F.
AU - Tiro, Jasmin
AU - Kim, Jane
AU - McCarthy, Anne Marie
AU - Quinn, Virginia P.
AU - Doria-Rose, V. Paul
AU - Wheeler, Cosette M.
AU - Barlow, William E.
AU - Bronson, Mackenzie
AU - Garcia, Michael
AU - Corley, Douglas A.
AU - Haas, Jennifer S.
AU - Halm, Ethan A.
AU - Kamineni, Aruna
AU - Rutter, Carolyn M.
AU - Tosteson, Tor D.
AU - Trentham-Dietz, Amy
AU - Weaver, Donald L.
AU - On Behalf Of The Prospr Consortium, Behalf Of The Prospr Consortium
N1 - Publisher Copyright:
© 2015, Society of General Internal Medicine.
PY - 2016/4/1
Y1 - 2016/4/1
N2 - Background: Primary care providers and health systems have prominent roles in guiding effective cancer screening. Objective: To characterize variation in screening abnormality rates and timely initial follow-up for common cancer screening tests. Design: Population-based cohort undergoing screening in 2011, 2012, or 2013 at seven research centers comprising the National Cancer Institute-sponsored Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) consortium. Participants: Adults undergoing mammography with or without digital breast tomosynthesis (n = 97,683 ages 40–75 years), fecal occult blood or fecal immunochemical tests (n = 759,553 ages 50–75 years), or Papanicolaou with or without human papillomavirus tests (n = 167,330 ages 21–65 years). Intervention: Breast, colorectal, or cervical cancer screening. Main Measures: Abnormality rates per 1000 screens; percentage with timely initial follow-up (within 90 days, except 9-month window for BI-RADS 3). Primary care clinic-level variation in percentage with screening abnormality and percentage with timely initial follow-up. Key Results: There were 10,248/97,683 (104.9 per 1000) abnormal breast cancer screens, 35,847/759,553 (47.2 per 1000) FOBT/FIT-positive colorectal cancer screens, and 13,266/167,330 (79.3 per 1000) abnormal cervical cancer screens. The percentage with timely follow-up was 93.2 to 96.7 % for breast centers, 46.8 to 68.7 % for colorectal centers, and 46.6 % for the cervical cancer screening center (low-grade squamous intraepithelial lesions or higher). The primary care clinic variation (25th to 75th percentile) was smaller for the percentage with an abnormal screen (breast, 8.5–10.3 %; colorectal, 3.0–4.8 %; cervical, 6.3–9.9 %) than for the percentage with follow-up within 90 days (breast, 90.2–95.8 %; colorectal, 43.4–52.0 %; cervical, 29.6–61.4 %). Conclusions: Variation in both the rate of screening abnormalities and their initial follow-up was evident across organ sites and primary care clinics. This highlights an opportunity for improving the delivery of cancer screening through focused study of patient, provider, clinic, and health system characteristics associated with timely follow-up of screening abnormalities.
AB - Background: Primary care providers and health systems have prominent roles in guiding effective cancer screening. Objective: To characterize variation in screening abnormality rates and timely initial follow-up for common cancer screening tests. Design: Population-based cohort undergoing screening in 2011, 2012, or 2013 at seven research centers comprising the National Cancer Institute-sponsored Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) consortium. Participants: Adults undergoing mammography with or without digital breast tomosynthesis (n = 97,683 ages 40–75 years), fecal occult blood or fecal immunochemical tests (n = 759,553 ages 50–75 years), or Papanicolaou with or without human papillomavirus tests (n = 167,330 ages 21–65 years). Intervention: Breast, colorectal, or cervical cancer screening. Main Measures: Abnormality rates per 1000 screens; percentage with timely initial follow-up (within 90 days, except 9-month window for BI-RADS 3). Primary care clinic-level variation in percentage with screening abnormality and percentage with timely initial follow-up. Key Results: There were 10,248/97,683 (104.9 per 1000) abnormal breast cancer screens, 35,847/759,553 (47.2 per 1000) FOBT/FIT-positive colorectal cancer screens, and 13,266/167,330 (79.3 per 1000) abnormal cervical cancer screens. The percentage with timely follow-up was 93.2 to 96.7 % for breast centers, 46.8 to 68.7 % for colorectal centers, and 46.6 % for the cervical cancer screening center (low-grade squamous intraepithelial lesions or higher). The primary care clinic variation (25th to 75th percentile) was smaller for the percentage with an abnormal screen (breast, 8.5–10.3 %; colorectal, 3.0–4.8 %; cervical, 6.3–9.9 %) than for the percentage with follow-up within 90 days (breast, 90.2–95.8 %; colorectal, 43.4–52.0 %; cervical, 29.6–61.4 %). Conclusions: Variation in both the rate of screening abnormalities and their initial follow-up was evident across organ sites and primary care clinics. This highlights an opportunity for improving the delivery of cancer screening through focused study of patient, provider, clinic, and health system characteristics associated with timely follow-up of screening abnormalities.
KW - breast cancer screening
KW - cervical cancer screening
KW - colorectal cancer screening
KW - practice variation
UR - http://www.scopus.com/inward/record.url?scp=84949511675&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84949511675&partnerID=8YFLogxK
U2 - 10.1007/s11606-015-3552-7
DO - 10.1007/s11606-015-3552-7
M3 - Article
C2 - 26658934
AN - SCOPUS:84949511675
SN - 0884-8734
VL - 31
SP - 372
EP - 379
JO - Journal of general internal medicine
JF - Journal of general internal medicine
IS - 4
ER -