In two large health systems, 75% of patients lacked sufficient data in the electronic health record to meet criteria to exit cervical cancer screening at age 65 years.
Abstract
OBJECTIVE:
To describe the proportion of patients with documentation of meeting cervical cancer screening exit criteria in the medical record and to estimate the incidence of cancer and high-grade precancerous lesions after age 65 years detected by the end of the study period, by screening exit eligibility.
METHODS:
We conducted a retrospective analysis of patients who turned 66 years old between 2010 and 2019 at two large health care systems located in Dallas, Texas, and Boston, Massachusetts, from the METRICS (MultilEvel opTimization of the ceRvIcal Cancer Screening process in diverse Settings & populations) Research Center, part of the PROSPR II (Population-based Research to Optimize the Screening Process) consortium. The primary outcomes were 1) the percentage of patients who were eligible for screening exit at age 66 years after adequate documented negative screening history (at least three consecutive negative cytology test results or two consecutive negative human papillomavirus [HPV] test or co-test results [cytology+HPV testing] within the prior 10 years) and 2) the worst pathology or cytology outcome documented in the cohort after age 65 years.
RESULTS:
Among 42,393 people who turned 66 years old while in the METRICS cohort, approximately 75.7% (n=32,094) were not eligible to exit screening at age 66 years, primarily because too few screening tests were documented (n=29,354, 91.5% of those ineligible). Furthermore, only 4,037 patients at both sites had 10 years of prior observation within the health care system, meaning data from current and prior health care delivery sites would be needed to assess exit eligibility for more than 90% of the cohort. Patients remained in the cohort after their 66th birthday on average for 3.8 years (interquartile range 2.0–6.5 years), and most (79.3%) had no subsequent evaluation after their 66th birthday. Among those ineligible to exit due to insufficient screening history, 83.7% were not subsequently screened. Of the 16.3% who were screened, 90 (1.9%) were diagnosed with cancer or high-grade precancerous lesions, including 34 (0.7%) who were diagnosed with cervical cancer. By contrast, among patients eligible to exit with sufficient screening history, 34.3% had at least one subsequent cytology or pathology result or both, and high-grade precancerous lesions or cervical cancer were diagnosed in only 0.6% (n=16).
CONCLUSION:
Data documented in the medical record are often insufficient to meet guidelines for exiting cervical cancer screening.
Although rates of cervical cancer have been declining over the past several decades, approximately 20% of cervical cancer cases in the United States diagnosed from 2017 to 2021 occurred in people older than age 65 years.1 Current U.S. guidelines recommend discontinuing cervical cancer testing at age 65 years for individuals at average risk who have established an adequate negative screening history in the prior 10 years (at least three consecutive negative cytology test results or two consecutive negative human papillomavirus [HPV] test or co-test results [cytology+HPV testing]).2 Additionally, guidelines stipulate that people are eligible to exit screening only if they have never been diagnosed with cervical cancer or with a high-grade precancerous lesion in the past 25 years and if they are not immunosuppressed (eg, due to infection with human immunodeficiency virus [HIV]).3,4
Current guidelines for exiting screening are based on the rarity of cervical cancer diagnosed after age 65 years among patients who had regular screening with multiple negative cytology and HPV test results.5–7 However, the feasibility of implementing these guidelines in clinical practice is relatively unknown. Given substantial patient movement among different health care systems in the United States and poor patient recall of prior abnormal screening results,8–10 it may be particularly challenging for clinicians and patients to ascertain cervical abnormalities over a more than 25-year period. Prior research examining this topic primarily used national billing and claims databases, which have limited data available from electronic health records (EHRs) detailing cytology and pathology reports necessary for clinicians to make a recommendation about exiting screening.11 As such, the present study was undertaken to gain a better understanding of the information available to clinicians within the EHR to make decisions regarding continuing compared with discontinuing screening in two large health care systems in the United States. There were two main goals of this study: 1) to describe the proportion of patients eligible to exit screening based on current guidelines and the reasons patients were ineligible to exit, and 2) to enumerate patients who continue to test for cervical cancer after age 65 years and the worst outcomes from tests received after age 65 years by screening exit eligibility status.
METHODS
This was a retrospective analysis of the METRICS (MultilEvel optimization of the ceRvIcal Cancer Screening process in diverse Settings & populations) Research Center, part of the PROSPR II (Population-based Research to Optimize the Screening Process) consortium.12 Two health care systems contributed to data presented here: Parkland Health, a publicly funded, integrated safety-net health care system for underinsured and uninsured residents in Dallas County, Texas, with academic oversight from the University of Texas Southwestern Medical Center; and Mass General Brigham, an integrated health care delivery system in the Boston area with two academic medical centers and their affiliated primary care networks. The IRBs at each site approved all study activities.
The METRICS cohort has been previously described.13 Briefly, all female patients aged 18–89 years seen at least once in a primary care or women's health clinic (Mass General Brigham or Parkland Health) or HIV clinic (Parkland Health) between January 1, 2010 and December 31, 2019 were included in the cohort. For this analysis, we included patients at Mass General Brigham and Parkland Health who turned 66 years old during the cohort period. Electronic health records, administrative data, and central cancer registries were used to identify demographic information, all cytology and HPV tests and procedures that occurred within the health care system, and cancer diagnoses within the geographic catchment area. Data sources and collection methods at Parkland Health have been previously described and are similar at Mass General Brigham.14
Demographic characteristics were described overall and by site. Race and ethnicity and Yost quintile15 were identified at cohort entry, and health insurance, comorbidity score, and body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) were captured during the calendar year in which the cohort member turned 64 years old, as previously described.13 Race and ethnicity was included in our descriptive data given known disparities in cervical cancer screening and outcomes among racial and ethnic minority patients. Additionally, we reported the observable years in the health care system, defined as time from the first documented primary care encounter within the health care system (up to 3 years before cohort entry) to the patient's 66th birthday and the number of primary care encounters per year in the cohort before the 66th birthday. We identified the subset of cohort members with 10 or more observable years in the health care system before turning age 66 (ie, sufficient time in the health care system to have had two screens to inform eligibility for screening exit). Statistical significance of differences in demographic characteristics by observable years in the health care system was determined using the χ2 test.
To assess our first study goal of describing eligibility for screening exit and reasons for ineligibility, we classified cohort members’ eligibility for screening exit on their 66th birthday (Fig. 1) as follows:
Ineligible due to being under surveillance for a prior abnormality or high-risk condition.
Ineligible due to insufficient screening history (eg, none or too few documented screening tests received over a 10-year time period).
Eligible based on a documented cervix-removing hysterectomy with a normal prior screening history.
Eligible based on an established and adequate negative screening history in the 10 years preceding their 66th birthday.
Fig. 1. Identification of screen exit eligibility on 66th birthday. We included all METRICS (MultilEvel opTimization of the ceRvIcal Cancer Screening) cohort members who turned 66 years old from 2010 to 2019 while in the cohort. Among people in the study cohort, we first identified people who were under surveillance on their 66th birthday due to the following: a previous history of cervical cancer, no cervix after a full or total hysterectomy or trachelectomy at any age and abnormality (atypical squamous cells of undetermined significance [ASC-US] cytology or worse, human papillomavirus [HPV]–positive test result or cervical procedure, regardless of pathology outcome) from age 40–65 years, a high-grade result (high-grade squamous intraepithelial lesion cytology or worse or HPV 16/18–positive test result) from age 40–65 years, a low-grade result (ASC-US or low-grade squamous intraepithelial lesion cytology or HPV non–16/18-postive test result) from age 55–65 years, or prior human immunodeficiency virus (HIV) diagnosis. People who met any of these criteria were deemed ineligible to exit screening on their 66th birthday. Next, among people not under surveillance on their 66th birthday, we identified people without a cervix after a full or total hysterectomy or trachelectomy at any age and a further subset based on whether the person had one or more documented normal test results; people who had one or more documented normal test results were deemed eligible to exit screening on their 66th birthday, and people who did not were deemed ineligible to exit screening on their 66th birthday. Lastly, among people not under surveillance and with a cervix (ie, no documented previous cervix removal), we identified people with a sufficient screening history based on either a normal co-test result from age 61 to 65 years with one or more preceding normal co-test results from age 56 to 65 years, or a normal Pap test result from age 63 to 65 years with two or more preceding normal Pap test results from age 56 to 65 years. People who met these criteria were deemed eligible to exit screening on their 66th birthday, and people who did not meet these criteria were deemed ineligible to exit screening on their 66th birthday.
Alimena. Cervical Cancer Screening Exit Eligibility. O&G Open 2024.
“Ineligible due to being under surveillance” included people with any of the following conditions on their 66th birthday: a previous history of cervical cancer, no cervix after a full or total hysterectomy or trachelectomy at any age and abnormality (atypical squamous cells of undetermined significance cytology or worse, HPV-positive test result or cervical procedure, regardless of pathology outcome) from age 40–65 years, a high-grade result (high-grade squamous intraepithelial lesion cytology or worse, or HPV 16/18-positive test result) from age 40–65 years, a low-grade result (atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion cytology or HPV non–16/18-positive test result) from age 55–65 years, or prior HIV diagnosis. People who met any of these criteria were deemed ineligible to exit screening on their 66th birthday. Next, among people not under surveillance on their 66th birthday, we identified people without a cervix after a full or total hysterectomy or trachelectomy at any age and further subset based on whether the person had one or more documented normal test results; people who had one or more documented normal test results were deemed eligible to exit screening on their 66th birthday, and people who did not were deemed ineligible to exit screening on their 66th birthday. Lastly, among people not under surveillance and with a cervix (ie, no documented previous cervix removal), we identified people with a sufficient screening history based on either a normal co-test result from age 61–65 years with one or more preceding normal co-test results from 56 to 65 years, or a normal Pap test result from age 63–65 years with two or more preceding normal Pap test results from age 56–65 years. People who met these criteria were deemed eligible to exit screening on their 66th birthday, and people who did not meet these criteria were deemed ineligible to exit screening on their 66th birthday.
To assess our second study goal of understanding frequency of testing and testing outcomes among patients after turning 66 years old, we used the Kaplan Meier method to estimate the cumulative occurrence of the next event (cytology or HPV test, pathology, or cancer diagnosis reported in cancer registries if no preceding cytology or HPV test or procedure) after turning 66 years old for each screening exit eligibility category. Patients were censored at cohort exit due to reaching the end of the study period (December 31, 2019), death, going without a primary care or women's health clinic visit for more than 37 months, or moving out of Dallas County, Texas (for Parkland Health only). The most severe outcome (cytology, HPV, pathology, cancer registry diagnosis) after turning 66 years old was described and classified according to screening exit eligibility category.
Analyses were conducted using SAS 9.4 and R 4.0.3.
RESULTS
A total of 42,393 cohort members turned 66 years old during the study period (2010–2019, Table 1). Of these, only 4,037 (9.5%) were in the health care system for 10 or more years (ie, under observation) before turning 66 years old. Table 1 shows demographic characteristics of the overall study population, as well as the subset in the cohort for 10 or more years. Differences in demographic characteristics by study site have been described in our previous work and were comparable with prior analyses with this cohort.13 Notably, more than 90% of patients had at least one visit per year in the health care system before their 66th birthday, indicating opportunities for screening.
Table 1.
Demographic Characteristics of PROSPR METRICS Cohort Members on Their 66th Birthday by Health Care System
| 66th Birthday 2010–2019 | 66th Birthday 2010–2019 and 10 Years Prior Observation | |||||
| Total | PH | MGB | Total | PH | MGB | |
| Cohort members [n (%)] | 42,393 | 14,597 (34.4) | 27,796 (65.6) | 4,037 | 1,278 (31.7) | 2,759 (68.3) |
| Characteristics on 66th birthday | ||||||
| Race and ethnicity* | ||||||
| Black, non-Hispanic | 7,536 (17.9) | 5,493 (37.9) | 2,043 (7.4) | 709 (17.6) | 534 (41.9) | 175 (6.4) |
| Hispanic | 7,747 (18.4) | 5,585 (38.5) | 2,162 (7.9) | 737 (18.3) | 533 (41.8) | 204 (7.4) |
| White, non-Hispanic | 23,988 (57.1) | 2,310 (15.9) | 21,678 (78.8) | 2,311 (57.5) | 134 (10.5) | 2,177 (79.2) |
| None of the above or multiple races | 2,748 (6.5) | 1,124 (7.8) | 1,624 (5.9) | 265 (6.6) | 73 (5.7) | 192 (7.0) |
| Unknown | 374 | 85 | 289 | 15 | <5 | 11 |
| Health insurance† | ||||||
| Medicaid | 5,580 (16.6) | 1,141 (10.1) | 4,439 (19.9) | 758 (19.0) | 207 (16.9) | 551 (20.0) |
| Medicare | 5,114 (15.3) | 1,548 (13.8) | 3,566 (16.0) | 333 (8.4) | 251 (20.4) | 82 (3.0) |
| Commercial | 14,923 (44.5) | 722 (6.4) | 14,201 (63.7) | 2,236 (56.1) | 113 (9.2) | 2,123 (77.0) |
| Other or uninsured | 7,913 (23.6) | 7,839 (69.7) | 74 (0.3) | 657 (16.5) | 657 (53.5) | 0 |
| Unknown | 8,863 | 3,347 | 5,516 | 53 | 50 | <5 |
| Comorbidity score‡ | ||||||
| 0–1 | 19,781 (61.0) | 6,143 (53.1) | 13,638 (65.3) | 1,717 (44.3) | 406 (32.3) | 1,311 (50.2) |
| 2 or higher | 12,667 (39.0) | 5,420 (46.9) | 7,247 (34.7) | 2,155 (55.7) | 852 (67.7) | 1,303 (49.9) |
| Unknown | 9,945 | 3,034 | 6,911 | 165 | 20 | 145 |
| BMI (kg/m2)§ | ||||||
| Lower than 18.5 | 679 (2.1) | 91 (0.9) | 588 (2.7) | 64 (1.7) | 8 (0.7) | 56 (2.1) |
| 18.5–24.9 | 8,820 (27.8) | 1,427 (14.1) | 7,393 (34.3) | 1,109 (29.0) | 164 (14.2) | 945 (35.3) |
| 25.0–29.9 | 9,365 (29.5) | 2,823 (27.9) | 6,542 (30.3) | 1,094 (28.6) | 304 (26.3) | 790 (29.5) |
| 30.0 or higher | 12,844 (40.5) | 5,794 (57.2) | 7,050 (32.7) | 1,564 (40.8) | 679 (58.8) | 885 (33.1) |
| Unknown | 10,685 | 4,462 | 6,223 | 206 | 123 | 83 |
| Yost quintile (state)‖ | ||||||
| 1 | 9,048 (22.6) | 5,590 (39.4) | 3,458 (13.4) | 1,015 (26.3) | 562 (45.5) | 453 (17.2) |
| 2 | 6,586 (16.4) | 3,683 (26.0) | 2,903 (11.2) | 603 (15.6) | 324 (26.2) | 279 (10.6) |
| 3 | 5,975 (14.9) | 2,145 (15.1) | 3,830 (14.8) | 528 (13.7) | 153 (12.4) | 375 (14.3) |
| 4 | 7,000 (17.5) | 1,878 (13.2) | 5,122 (19.8) | 625 (16.2) | 145 (11.7) | 480 (18.3) |
| 5 | 11,474 (28.63) | 895 (6.31) | 10,579 (40.9) | 1,091 (28.3) | 51 (4.1) | 1,040 (39.6) |
| Unknown | 2,310 | 406 | 1,904 | 175 | 43 | 132 |
| Observable years in health system before 66th birthday¶ | ||||||
| 0–2 | 9,967 (23.5) | 3,771 (25.8) | 6,196 (22.3) | — | — | — |
| 3–5 | 14,300 (33.7) | 5,286 (36.2) | 9,014 (32.4) | — | — | — |
| 6–9 | 14,089 (33.2) | 4,262 (29.2) | 9,827 (35.4) | — | — | — |
| 10 or more | 4,037 (9.5) | 1,278 (8.8) | 2,759 (9.9) | 4,037 (100.0) | 1,278 (100.0) | 2,759 (100.0) |
| Primary care encounters per year in cohort# | ||||||
| Fewer than 1 | 3,478 (8.2) | 1,594 (10.9) | 1,884 (6.8) | 218 (5.4) | 65 (5.1) | 153 (5.6) |
| 1–2 | 21,664 (51.1) | 7,258 (49.7) | 14,406 (51.8) | 2,411 (59.7) | 718 (56.2) | 1,693 (61.4) |
| 3–5 | 13,000 (30.7) | 4,513 (30.9) | 8,487 (30.5) | 1,174 (29.1) | 422 (33.0) | 752 (27.3) |
| 6 or more | 4,251 (10.0) | 1,232 (8.4) | 3,019 (10.9) | 234 (5.8) | 73 (5.7) | 161 (5.8) |
PROSPR, Population-based Research to Optimize the Screening Process; METRICS, MultilEvel opTimization of the ceRvIcal Cancer Screening process in diverse Settings & populations; PH, Parkland Health; MGB, Mass General Brigham; BMI, body mass index.
Data are n (column %) unless otherwise specified.
Race and ethnicity was identified at METIRCS cohort entry using the following mutually exclusive categories: Hispanic; Black, non-Hispanic; White, non-Hispanic; and Other, which included Asian, Native Hawaiian, Pacific Islander, Native American, Alaskan Native, or Other, as well as those cohort members that identified with multiple races and ethnicities described above.
Health insurance was ascertained during the calendar year in which the person turned 64 years old during the METRICS cohort period. If multiple insurance designations were observed within a calendar year, a single insurance designation was assigned in decreasing priority as follows: Medicaid, Medicare, commercial, and other or uninsured, which included other government payers, other insurance, medical assistance, and uninsured.
Comorbidity score was identified from all comorbidities diagnosed in the calendar year in which the person turned 64 years old and in the preceding years. Comorbidity score was calculated according to weights presented in Quan H, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005;43:1130–9. doi: 10.1097/01.mlr.0000182534.19832.83
Body mass index was ascertained during the calendar year in which the person turned 64 years old during the METRICS cohort period.
The Yost quintile15 is a composite socioeconomic index based on median household income, median home value, median rent, percent of the population living below the poverty line, educational achievement, percent of the population who are considered working class, and percent of the population who are unemployed based on state-level Census tract data. Yost quintile was identified based on the Census tract the cohort member lived in at METRICS cohort entry and may not reflect the Census tract in which the cohort member lived on their 66th birthday.
Time from the first documented primary care encounter within the health care system, which may have occurred up to 3 years before METRICS cohort entry, to the patient’s 66th birthday.
Calculated per years in the METRICS cohort up to 66th birthday. Qualifying primary care encounters included all in-person outpatient office visits while in the cohort that occurred either with any clinician at a primary care or women's health clinic (MGB) or with a physician (attending, resident, or fellow), nurse practitioner, or physician assistant at a community health, family practice, internal medicine, women's health, geriatrics, gynecology, or human immunodeficiency virus (HIV) clinic (PH).
About three-quarters of the cohort (75.7%, n=32,094) were ineligible to exit screening at age 66 years (Table 2). Among those ineligible to exit screening, 8.5% were under surveillance for prior abnormalities or high-risk conditions, and 91.5% had too few screening tests documented in the EHR to meet exit recommendations. Most of the group under surveillance had a prior low-grade result. In terms of insufficient screening history, the most prevalent reasons were no tests and less than 10 years in the health care system (45.3%) and too few tests over the time period (46.1%).
Table 2.
Screen Exit Eligibility for PROSPR METRICS Cohort Members on Their 66th Birthday by Health Care System
| 66th Birthday 2010–2019 | 66th Birthday 2010–2019 and 10 Years Prior Observation | |||||
| Total | PH | MGB | Total | PH | MGB | |
| Cohort members | 42,393 | 14,597 | 27,796 | 4,037 | 1,278 | 2,759 |
| Screen exit eligibility on 66th birthday* | ||||||
| Ineligible | 32,094 (75.7) | 12,046 (82.5) | 20,048 (72.1) | 2,474 (61.3) | 879 (68.8) | 1,595 (57.8) |
| Under surveillance† | 2,740 (6.5) | 1,155 (7.9) | 1,585 (5.7) | 456 (11.3) | 183 (14.3) | 273 (9.9) |
| Prior cervical cancer diagnosis | 333 (12.2) | 99 (8.6) | 234 (14.8) | 40 (8.8) | 9 (4.9) | 31 (11.4) |
| Hysterectomy and abnormality (40–65 y) | 424 (15.5) | 283 (24.5) | 141 (8.9) | 59 (12.9) | 39 (21.3) | 20 (7.3) |
| High-grade result (40–65 y) | 209 (7.6) | 84 (7.3) | 125 (7.9) | 42 (9.2) | 16 (8.7) | 26 (9.5) |
| Low-grade result (55–65 y) | 1,681 (61.4) | 626 (54.2) | 1,055 (66.6) | 296 (64.9) | 107 (58.5) | 189 (69.2) |
| HIV-positive | 93 (3.4) | 63 (5.5) | 30 (1.9) | 19 (4.2) | 12 (6.6) | 7 (2.6) |
| Insufficient screening history‡ | 29,354 (69.2) | 10,891 (74.6) | 18,463 (66.4) | 2,018 (50.0) | 696 (54.5) | 1,322 (47.9) |
| Hysterectomy and no tests | 1,884 (6.4) | 116 (1.1) | 1,768 (9.6) | 199 (9.9) | 8 (1.2) | 191 (14.5) |
| No tests, 10 or more years in system | 645 (2.2) | 240 (2.2) | 405 (2.2) | 645 (32.0) | 240 (34.5) | 405 (30.6) |
| No tests, less than 10 years in system | 13,296 (45.3) | 6,225 (57.2) | 7,071 (38.3) | 0 | 0 | 0 |
| Too few normal test results or poorly timed | 13,529 (46.1) | 4,310 (39.6) | 9,219 (49.9) | 1,174 (58.2) | 448 (64.4) | 726 (54.9) |
| Eligible | 10,299 (24.3) | 2,551 (17.5) | 7,748 (27.9) | 1,563 (38.7) | 399 (31.2) | 1,164 (42.2) |
| Hysterectomy and no abnormality§ | 2,705 (6.4) | 1,687 (11.6) | 1,018 (3.7) | 379 (9.4) | 228 (17.8) | 151 (5.5) |
| Sufficient screening history‖ | 7,594 (17.9) | 864 (5.9) | 6,730 (24.2) | 1,184 (29.3) | 171 (13.4) | 1,013 (36.7) |
| 2 normal co-test or HPV test results | 1,960 (25.8) | 173 (20.0) | 1,787 (26.6) | 509 (43.0) | 32 (18.7) | 477 (47.1) |
| 3 normal cytology results | 5,634 (74.2) | 691 (80.0) | 4,943 (73.5) | 675 (57.0) | 139 (81.3) | 536 (52.9) |
PROSPR, Population-based Research to Optimize the Screening Process; METRICS, MultilEvel opTimization of the ceRvIcal Cancer Screening process in diverse Settings & populations; PH, Parkland Health; MGB, Mass General Brigham; HIV, human immunodeficiency virus; HPV, human papillomavirus.
Data are n or n (column %).
Eligibility categories and subcategories report column percent of total. Reasons for belonging to each subcategory report column percent of subcategory.
Subgroups included people with the following history, applied in descending priority if multiple: prior cervical cancer diagnosis (squamous cell carcinoma, adenocarcinoma, adenosquamous carcinoma, cervical cancer not otherwise specified, and cervical cancer of the cervix of unknown origin on pathology) at any age, prior radical hysterectomy or hysterectomy after abnormal cytology (atypical squamous cells of undetermined significance [ASC-US] or worse or HPV-positive, regardless of strain), cervical pathology when 40–65 years old, or a combination of these; prior high-grade (high-grade squamous intraepithelial lesion, cervical intraepithelial neoplasia [CIN] 2–3, CIN 3 or carcinoma in situ, and adenocarcinoma in situ) cytology or pathology when 40–65 years old, prior low-grade (CIN 1, CIN 1–2, low-grade squamous intraepithelial lesion, or ASC-US) cytology or pathology or HPV-positive result when 55–65 years old, prior HIV diagnosis, or a combination of these.
Subgroups included people with documented hysterectomy and no documented cytology, HPV test, or pathology; no documented cytology, HPV test, or pathology and either at least 10 years of observation by the health care system or fewer than 10 years of observation by the health care system; or people with a documented normal cytology evaluation history that either did not include two HPV-negative test results or negative co-test results in the 10 years preceding their 66th birthday, with one test occurring within 5 years of their 66th birthday, or three normal cytology results preceding their 66th birthday, with one test occurring within 3 years of their 66th birthday.
Subgroups included people with a documented total hysterectomy, trachelectomy, or hysterectomy not otherwise specified and at least one prior cytology with a normal result. Those with normal cytology results may or may not have had an HPV test; if HPV test was completed, the result was negative.
Subgroups included people with either two HPV-negative test results or normal HPV-negative co-test results in the 10 years preceding their 66th birthday, with one test occurring within 5 years of their 66th birthday, or three normal cytology results preceding their 66th birthday, with one test occurring within 3 years of their 66th birthday.
Among the 10,299 eligible to exit, 26.3% had a prior hysterectomy, at least one negative screen, and no cervical abnormality documented previously (Table 2), though there were significant differences in this proportion by site (Parkland Health 66.1%, Mass General Brigham, 13.1%). The remaining 73.7% of the population eligible to exit had sufficient screening history documented over the prior 10 years and no history of an abnormality or high-risk condition meriting surveillance (Table 2).
Subgroup analysis of the 4,037 with 10 or more years of prior observation showed that 61.3% were ineligible to exit screening at age 66 years, among whom 18.4% were ineligible due to being under surveillance and 81.6% lacked sufficient screening history to meet exit recommendations. Among the 1,563 people eligible to exit, 24.2% (Parkland Health 57.1%, Mass General Brigham 13.0%) had a prior hysterectomy and at least one prior negative screen, and 75.8% (Parkland Health 42.9%, Mass General Brigham 87.0%) had a sufficient screening history.
Figure 2 illustrates when patients received the next event (ie, cytology, HPV test, pathology, or cancer registry diagnosis) over the decade after their 66th birthday stratified by eligibility status. Among patients ineligible to exit, those under surveillance more frequently had cervical results compared with those with insufficient screening history (Fig. 2; overall, 51.4% vs 22.7%). Although the relative patterns were similar across sites, patients at Parkland Health had fewer events overall compared with patients at Mass General Brigham (17.6% vs 32.1%).
Fig. 2. Next event after 66th birthday by exit eligibility. Next event (cytology, human papillomavirus test, pathology, or cancer registry diagnosis) after 66th birthday among the total cohort (top panel), with at-risk counts reported (bottom panel), based on screen exit eligibility as follows: ineligible due to being under surveillance (pink line), ineligible due to insufficient screening history (orange line), eligible after hysterectomy and no documented abnormality (blue line), and eligible after sufficient screening history (green line). Cohort members were censored at cohort exit (indicated by vertical hashing).
Alimena. Cervical Cancer Screening Exit Eligibility. O&G Open 2024.
Table 3 reports the most severe outcome documented in the EHR or cancer registry after a patient's 66th birthday. Among people ineligible to exit due to needing surveillance, most (60.4%) had no subsequent evaluation after their 66th birthday. Of the 39.6% who were evaluated, 76 (7.0%) of women were diagnosed with cancer or a high-grade precancerous lesion, including seven (0.7%) cervical cancers. Among people ineligible to exit due to insufficient screening history, 83.7% were not subsequently screened; of the 16.3% who were screened, 90 (1.9%) women were diagnosed with cancer or a high-grade precancerous lesion, including 34 (0.7%) diagnosed with cervical cancer. Among people eligible to exit due to prior hysterectomy, 11.7% had a cytology or pathology result or both after their 66th birthday; no patients in this category were subsequently diagnosed with cancer, and five (1.6%) were diagnosed with a high-grade precancerous lesion. Finally, among people eligible to exit with a sufficient screening history, 34.3% had at least one subsequent cytology or pathology result or both, and high-grade precancerous lesions or cervical cancer were diagnosed in 0.6% (n=16; n=6 [0.2%] were cancer).
Table 3.
Most Severe Outcome After 66th Birthday for PROSPR METRICS Cohort Members by Exit Eligibility and Health Care System
| 66th Birthday 2010–2019 | |||
| Total | PH | MGB | |
| Cohort members* | 42,393 | 14,597 | 27,796 |
| Most severe outcome after 66th birthday | |||
| Ineligible | 32,094 | 12,046 | 20,048 |
| Under surveillance† | 2,740 | 1,155 | 1,585 |
| No evaluation | 1,655 (60.4) | 832 (72.0) | 823 (51.9) |
| At least 1 evaluation after 66th birthday | 1,085 (39.6) | 323 (28.0) | 762 (48.1) |
| Cancer or high-grade result‡ | 76 (7.0) | 29 (9.0) | 47 (6.2) |
| Low-grade result§ | 283 (26.1) | 97 (30.0) | 186 (24.4) |
| Normal result | 714 (65.8) | 187 (57.9) | 527 (69.2) |
| Unknown result | 12 (1.1) | 10 (3.1) | <5 (0.3) |
| Insufficient screening history‖ | 29,354 | 10,891 | 18,463 |
| No evaluation | 24,572 (83.7) | 9,910 (91.0) | 14,662 (79.4) |
| At least 1 evaluation after 66th birthday | 4,782 (16.3) | 981 (9.0) | 3,801 (20.6) |
| Cancer or high-grade result‡ | 90 (1.9) | 31 (3.2) | 59 (1.6) |
| Low-grade result§ | 241 (5.0) | 79 (8.1) | 162 (4.3) |
| Normal result | 4,397 (92.0) | 846 (86.2) | 3,551 (93.4) |
| Unknown result | 54 (1.1) | 25 (2.6) | 29 (0.8) |
| Eligible | 10,299 | 2,551 | 7,748 |
| Hysterectomy and no abnormality¶ | 2,705 | 1,687 | 1,018 |
| No evaluation | 2,388 (88.3) | 1,597 (94.7) | 791 (77.7) |
| At least 1 evaluation after 66th birthday | 317 (11.7) | 90 (5.3) | 227 (22.3) |
| Cancer or high-grade result‡ | 5 (1.6) | 0 | 5 (2.2) |
| Low-grade result§ | 18 (5.7) | 11 (12.2) | 7 (3.1) |
| Normal result | 285 (89.9) | 74 (82.2) | 211 (93.0) |
| Unknown result | 9 (2.8) | 5 (5.6) | <5 (1.8) |
| Sufficient screening history# | 7,594 | 864 | 6,730 |
| No evaluation | 4,993 (65.8) | 685 (79.3) | 4,308 (64.0) |
| At least 1 evaluation after 66th birthday | 2,601 (34.3) | 179 (20.7) | 2,422 (36.0) |
| Cancer or high-grade result‡ | 16 (0.6) | <5 (0.6) | 15 (0.6) |
| Low-grade result§ | 98 (3.8) | 9 (5.0) | 89 (3.7) |
| Normal result | 2,467 (94.9) | 167 (93.3) | 2,300 (95.0) |
| Unknown result | 20 (0.8) | <5 (1.1) | 18 (0.7) |
PROSPR, Population-based Research to Optimize the Screening Process; METRICS, MultilEvel opTimization of the ceRvIcal Cancer Screening process in diverse Settings & populations; PH, Parkland Health; MGB, Mass General Brigham.
Data are n or n (column %).
Median (interquartile range) time in the cohort from 66th birthday to cohort exit was 3.8 (2.0–6.5) years overall, 5.9 (3.9–8.0) years among cohort members with at least one cytology or pathology result or both after their 66th birthday and 3.2 (1.8–5.8) years among cohort members who exited before cytology, HPV test, or pathology. Among cohort members with 10 or more prior years of observation, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 1.9 (1.3–2.5) years overall, 2.3 (1.7–2.9) years among cohort members with at least one cytology or pathology result or after their 66th birthday, and 1.8 (1.3–2.5) years among cohort members who exited before cytology, HPV test, or pathology. Median (interquartile range) age at cervical cancer diagnosis was 68 (67–71) years.
Among cohort members ineligible for screen exit on their 66th birthday due to surveillance, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 3.5 (1.9–5.8) years overall, 4.9 (3.2–7.1) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 2.5 (1.5–4.6) years among cohort members who exited before cytology, HPV test, or pathology. Among cohort members with 10 or more prior years of observation, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 1.9 (1.3–2.4) years overall, 2.3 (1.7–2.9) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 1.7 (1.2–2.3) years among cohort members who exited before cytology, HPV test, or pathology. Median (interquartile range) age at cervical cancer diagnosis was 68 (68–69) years.
Cancers included squamous cell carcinoma, adenocarcinoma, adenosquamous carcinoma, cervical cancer not otherwise specified, and cervical cancer of the cervix of unknown origin on pathology. High-grade results included cervical cancer, high-grade squamous intraepithelial lesion, cervical intraepithelial neoplasia (CIN) 2–3, CIN 3 or carcinoma in situ, and adenocarcinoma in situ pathology or cytology.
Low-grade results included CIN 1, CIN 1–2, low-grade squamous intraepithelial lesion, or atypical squamous cells of undetermined significance cytology or pathology or HPV-positive test result.
Among cohort members ineligible for screen exit due to insufficient screening history, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 3.8 (2.0–6.6) years overall, 6.2 (4.1–8.4) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 3.3 (1.8–6.0) years among cohort members who exited before cytology, HPV test, or pathology. Among cohort members with 10 or more prior years of observation, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 1.9 (1.3–2.5) years overall, 2.3 (1.8–2.9) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 1.8 (1.3–2.5) years among cohort members who exited before cytology, HPV test, or pathology. Median (interquartile range) age at cervical cancer diagnosis was 68.5 (67–71) years.
Among cohort members eligible for screen exit after hysterectomy, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 3.5 (1.9–6.2) years overall, 6.2 (4.3–8.8) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 3.2 (1.8–5.7) years among cohort members who exited before cytology, HPV test, or pathology. Among cohort members with 10 or more prior years of observation, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 1.7 (1.2–2.4) years overall, 2.6 (1.3–3.2) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 1.7 (1.2–2.3) years among cohort members who exited before cytology, HPV test, or pathology.
Among cohort members eligible for screen exit after sufficient screening history, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 4.3 (2.4–6.4) years overall, 5.7 (3.9–7.6) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 3.5 (2.0–5.5) years among cohort members who exited before cytology, HPV test, or pathology. Among cohort members with 10 or more prior years of observation, median (interquartile range) time in the cohort from 66th birthday to cohort exit was 2.0 (1.4–2.6) years overall, 2.3 (1.7–2.9) years among cohort members with at least one cytology or pathology result or both after their 66th birthday, and 1.9 (1.3–2.5) years among cohort members who exited before cytology, HPV test, or pathology. Median (interquartile range) age at cervical cancer diagnosis was 67.5 (67–69) years.
DISCUSSION
Approximately 20% of cervical cancers are diagnosed after age 65 years, and older people with cervical cancer present at later stages, have worse survival, and are more often diagnosed with aggressive histologic types of cervical cancer compared with younger patients.16 One possible explanation is that patients stop screening before they have met the eligibility criteria for discontinuation. Many clinicians may rely on patient self-report of prior screening when making decisions to end screening17; however, this is likely inaccurate and is not concordant with guidelines, which require records of prior testing.2 It is therefore important to investigate whether current guidelines for screening cessation after age 65 years are feasible to implement with data available to clinicians in the EHR. In two large health care systems, we found that 75.7% of patients may not be eligible to exit cervical cancer screening at age 66 years, despite more than 90% having at least one primary care visit per years in cohort preceding their 66th birthday. Most of these ineligible patients lacked sufficient documentation to meet guideline recommendations to stop testing due to too few cervical cancer tests before their 66th birthday (91.5%). In a country lacking a registry or broadly integrated EHR to capture receipt of cervical cancer screening nationwide, our data suggest that clinicians often lack the formal documentation over a sufficient time period to make screening exit recommendations based on guidelines. Concerningly, 0.7% of those who may have inappropriately exited screening after their 66th birthday were subsequently diagnosed with cervical cancer over an average time period of 4 years.
Although our study is similar to prior research published on the subject of cessation of cervical cancer screening after age 65 years,11,18 it is unique in reporting cancer outcomes and in examining structured EHR data with details about cytology and pathology results from two health care systems with varied populations. Prior claims-based data suggest that up to 65% of patients may lack documentation necessary to exit screening11; in our cohort, 69% of patients had insufficient screening history. Similarly, another study suggested that 23.6% of women aged 66–70 years who responded to the National Health Interview Survey had not had a cervical cancer screening test within the prior 5 years.18 Taking these studies together, the data suggest that anywhere from 65 to 76% of patients likely lack sufficient testing to exit screening after age 65 years. Within our cohort, only 9.5% of patients had 10 or more observable years of data in the EHR before their 66th birthday. Even among these patients, sufficient documentation of normal test results to exit screening were absent from the EHR half of the time. Thus, patients who maintained a continuous, long-term relationship with a health care institution often still did not meet current criteria to exit screening. This suggests that either testing is occurring elsewhere and is not readily available in structured EHR fields, or more likely that testing is not happening at all.11 We may have noted slightly higher rates of incomplete documentation compared with prior claims-based analyses, because, if patients move health care systems but retain the same insurance payer, claims-based data may not reflect the move and may overestimate data available to subsequent clinicians. However, when patients move health care systems, their records often fail to follow them to the next treating physician.
Overall, our data suggest that relying on the EHR to make these important clinical decisions may be neither feasible nor practical, and that this complexity may harm patients. There were more than three times as many cases of cervical cancer and high-grade dysplasia detected among people ineligible for screening exit due to insufficient history compared with those eligible for screening exit after sufficient history. Life expectancy is increasing, and hysterectomy rates are declining, meaning that we expect to see an increase in the size of the population older than age 65 years who remain at risk for cervical cancer in the coming years.19 Additionally, the proportion of non-White women older than age 65 years is expected to increase due to changing racial demographics in the United States.19 If we do not carefully evaluate our current system for exiting cervical cancer screening, this could result in worsening age-related and race-related disparities in cervical cancer rates and outcomes. Although primary HPV screening and HPV self-testing may help with access to screening, we need better systems in place to monitor patients nationally for self-testing to work. Alternatively, we could devise a different model for screening cessation while awaiting a change in national tracking of results. Another proposal may involve increasing the cutoff age to one that accounts for estimated life expectancy, similar to breast and colorectal cancer screening guidelines20 and similar to age cutoffs for cervical cancer testing used in other countries.21 This type of system would also ensure we are continuing to evaluate women for types of cervical cancer that are non-HPV related (such as gastric-type adenocarcinomas, for example) and less reliably detected by Pap and HPV testing. These types of cancers are more prevalent in older patients and could be captured through ongoing assessment of symptoms (eg, discharge, pain, bleeding) after age 65 years, with subsequent diagnostic evaluation. Additional research should explore a simple and more pragmatic approach to exiting screening or improved systems for tracking patients to ensure we are following patients at risk for cervical cancer appropriately.
Prior research based on claims data showed that Medicare paid $83.5 million for cytology, HPV testing, co-testing, colposcopy, and cervical procedures in people aged 65 years and older in 2019 alone.22 It is difficult to know what percentage of this expenditure was spent for necessary compared with unnecessary testing after age 65 years.23 Our data suggest that a notable proportion of this testing might be unnecessary, with 12% of those who could exit screening due to prior hysterectomy and 34% of those who could exit screening due to adequate negative recent screening having some form of evaluation after their 66th birthday. Whether this testing occurred because the patient was experiencing symptoms was not available for analysis. By contrast, our data indicate that a minority of patients who were ineligible to exit screening received further testing within a median of 4.9–6.2 years, with 60% of those under surveillance and 84% of those with insufficient screening history potentially inappropriately exiting screening. In other words, although some Medicare spending may be inappropriate in the population older than age 65 years, the greater issue may be that many patients with Medicare insurance are not getting necessary testing based on current guidelines, due to inappropriately exiting screening. Furthermore, the low numbers of patients eligible to exit screening highlights the trend in underscreening among patients aged 55 to 65 years; thus, interventions to promote screening in this age group may be critical to facilitate detection of precancerous lesions and cancers.
In terms of study limitations, our analysis examined structured fields in the EHR; thus, it is possible that clinicians may have had more screening data available at the point-of-care. For example, patients may have transferred records from external institutions with those stored as scanned documents. Also, some EHRs have portals and regional health information exchanges allowing clinicians to review care received in other facilities.24,25 Even when this information is available via portals, clinicians may not have the time or technological ability to search for this information. However, patients from Parkland Health are unlikely to receive care in other health care systems given the limited availability of other facilities that participate in Medicaid and other public payer programs typically used by safety-net systems.12 Similarly, institutional data from Mass General Brigham indicate that fewer than 5% of primary care patients have no record of care in the Mass General Brigham system within 5 years, suggesting a high degree of patient retention (unpublished). It is still possible that some patients may have migrated health systems after age 66 years, meaning there could be other severe results we are not capturing in our analysis of outcomes after age 66 years. We attempted to mitigate this by including data from regional cancer registries, but it is possible that we are underestimating the number of high-grade precancerous or cancerous results after age 66 years.
In conclusion, although current guidelines for screening exit result in low rates of cervical cancer diagnosed after age 65 years when implemented correctly,2 the complexity of the guidelines and the lack of longitudinal data available to clinicians may make it difficult to execute these guidelines in practice. Most patients did not have sufficient information documented in the EHR to exit screening. There were more cancers observed among people who lacked sufficient testing to exit screening than among those who established a sufficient history. Ideally, screening guidelines would account for the absence of documented patient history. Our research supports the need for better data exchange among health care systems to allow clinicians to accurately identify people at high-risk for cervical cancer after age 65 years. Additionally, future studies should consider alternative approaches to measure cervical cancer risk to enable more practical screening exit guidelines.
Footnotes
Financial Disclosure The authors did not report any potential conflicts of interest.
This work was supported by National Cancer Institute grant UM1CA221940 awarded to Jennifer S. Haas, Jasmin A. Tiro, and Aruna Kamineni. Jennifer S. Haas received funding from the American Cancer Society grant CRP-22-080-01-CTPS. The views expressed here are those of the authors only and do not necessarily represent the views of the National Cancer Institute or National Institutes of Health. The authors thank the participating METRICS sites for the data provided for this study. A comprehensive list of METRICS investigators and research staff is available at the PROSPR METRICS research site (https://utsouthwestern.edu/labs/prospr-metrics/about/team.html).
The data in this study are from the NCI PROSPR II research consortium and are available upon request after agreements are completed. Additional details are provided at https://healthcaredelivery.cancer.gov/prospr/datashare.html.
Each author has confirmed compliance with the journal's requirements for authorship.
Peer reviews and author correspondence are available at http://links.lww.com/AOG/D849.
REFERENCES
- 1.National Cancer Institute Surveillance, Epidemiology, and End Results Program. Cancer stat facts: cervical cancer. Accessed August 1, 2024. https://seer.cancer.gov/statfacts/html/cervix.html [Google Scholar]
- 2.Perkins RB, Guido RS, Castle PE, Chelmow D, Einstein MH, Garcia F, et al. 2019 ASCCP risk-based management consensus guidelines for abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis 2020;24:102–31. doi: 10.1097/LGT.0000000000000525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.American College of Obstetricians and Gynecologists. Updated cervical cancer screening guidelines. Accessed July 19, 2024. https://www.acog.org/clinical/clinical-guidance/practice-advisory/articles/2021/04/updated-cervical-cancer-screening-guidelines [Google Scholar]
- 4.Perkins RB, Guido RL, Saraiya M, Sawaya GF, Wentzensen N, Schiffman M, et al. Summary of current guidelines for cervical cancer screening and management of abnormal test results: 2016-2020. J Womens Health (Larchmt) 2021;30:5–13. doi: 10.1089/jwh.2020.8918 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Castle PE, Kinney WK, Xue X, Cheung LC, Gage JC, Zhao FH, et al. Effect of several negative rounds of human papillomavirus and cytology co-testing on safety against cervical cancer: an observational cohort study. Ann Intern Med 2018;168:20–9. doi: 10.7326/M17-1609 [DOI] [PubMed] [Google Scholar]
- 6.Castañón A, Landy R, Cuzick J, Sasieni P. Cervical screening at age 50-64 years and the risk of cervical cancer at age 65 years and older: population-based case control study. PLoS Med 2014;11:e1001585. doi: 10.1371/journal.pmed.1001585 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dinkelspiel H, Fetterman B, Poitras N, Kinney W, Cox JT, Lorey T, et al. Screening history preceding a diagnosis of cervical cancer in women age 65 and older. Gynecol Oncol 2012;126:203–6. doi: 10.1016/j.ygyno.2012.04.037 [DOI] [PubMed] [Google Scholar]
- 8.Kern LM, Ringel JB, Rajan M, Casalino LP, Colantonio LD, Pinheiro LC, et al. Ambulatory care fragmentation, emergency department visits, and race: a nationwide cohort study in the U.S. J Gen Intern Med 2023;38:873–80. doi: 10.1007/s11606-022-07888-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pham HH, Schrag D, O'Malley AS, Wu B, Bach PB. Care patterns in Medicare and their implications for pay for performance. N Engl J Med 2007;356:1130–9. doi: 10.1056/NEJMsa063979 [DOI] [PubMed] [Google Scholar]
- 10.Higashi RT, Tiro JA, Winer RL, Ornelas IJ, Bravo P, Quirk L, et al. Understanding the effect of new U.S. cervical cancer screening guidelines and modalities on patients' comprehension and reporting of their cervical cancer screening behavior. Prev Med Rep 2023;32:102169. doi: 10.1016/j.pmedr.2023.102169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mills JM, Morgan JR, Dhaliwal A, Perkins RB. Eligibility for cervical cancer screening exit: comparison of a national and safety net cohort. Gynecol Oncol 2021;162:308–14. doi: 10.1016/j.ygyno.2021.05.035 [DOI] [PubMed] [Google Scholar]
- 12.Beaber EF, Kamineni A, Burnett-Hartman AN, Hixon B, Kobrin SC, Li CI, et al. Evaluating and improving cancer screening process quality in a multilevel context: the PROSPR II consortium design and research agenda. Cancer Epidemiol Biomarkers Prev 2022;31:1521–31. doi: 10.1158/1055-9965.EPI-22-0100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Feldman S, Lykken JM, Haas JS, Werner CL, Kobrin SC, Tiro JA, et al. Factors associated with timely colposcopy following an abnormal cervical cancer test result. Prev Med 2022;164:107307. doi: 10.1016/j.ypmed.2022.107307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kamineni A, Tiro JA, Beaber EF, Silverberg MJ, Wheeler CM, Chao CR, et al. Cervical cancer screening research in the PROSPR I consortium: rationale, methods and baseline findings from a US cohort. Int J Cancer 2019;144:1460–73. doi: 10.1002/ijc.31940 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Yost K, Perkins C, Cohen R, Morris C, Wright W. Socioeconomic status and breast cancer incidence in California for different race/ethnic groups. Cancer Causes Control 2001;12:703–11. doi: 10.1023/a:1011240019516 [DOI] [PubMed] [Google Scholar]
- 16.Cooley JJP, Maguire FB, Morris CR, Parikh-Patel A, Abrahão R, Chen HA, et al. Cervical cancer stage at diagnosis and survival among women ≥65 years in California. Cancer Epidemiol Biomarkers Prev 2023;32:91–7. doi; 10.1158/1055-9965.EPI-22-0793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Holt HK, Flores R, James JE, Waters C, Kaplan CP, Peterson CE, et al. A qualitative study of primary care clinician's approach to ending cervical cancer screening in older women in the United States. Prev Med Rep 2023;36:102500. doi: 10.1016/j.pmedr.2023.102500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.White MC, Shoemaker ML, Benard VB. Cervical cancer screening and incidence by age: unmet needs near and after the stopping age for screening. Am J Prev Med 2017;53:392–5. doi: 10.1016/j.amepre.2017.02.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dilley S, Huh W, Blechter B, Rositch AF. It's time to re-evaluate cervical Cancer screening after age 65. Gynecol Oncol 2021;162:200–2. doi: 10.1016/j.ygyno.2021.04.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wilt TJ, Harris RP, Qaseem A, High Value Care Task Force of the American College of Physicians. Screening for cancer: advice for high-value care from the American College of Physicians. Ann Intern Med 2015;162:718–25. doi: 10.7326/M14-2326 [DOI] [PubMed] [Google Scholar]
- 21.Keltto N, Leivonen A, Pankakoski M, Sarkeala T, Heinävaara S, Anttila A. Cervical testing beyond the screening target age - a register-based cohort study from Finland. Gynecol Oncol 2021;162:315–21. doi: 10.1016/j.ygyno.2021.05.019 [DOI] [PubMed] [Google Scholar]
- 22.Qin J, Holt HK, Richards TB, Saraiya M, Sawaya GF. Use trends and recent expenditures for cervical cancer screening-associated services in Medicare fee-for-service beneficiaries older than 65 years. JAMA Intern Med 2023;183:11–20. doi: 10.1001/jamainternmed.2022.5261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Feldman S, Alimena S. Cervical cancer prevention and cost for women older than 65 years in the US-are we spending too much or too little? JAMA Intern Med 2023;183:20–1. doi: 10.1001/jamainternmed.2022.5251 [DOI] [PubMed] [Google Scholar]
- 24.Devine EB, Totten AM, Gorman P, Eden KB, Kassakian S, Woods S, et al. Health information exchange use (1990-2015): a systematic review. EGEMS (Wash DC) 2017;5:27. doi: 10.5334/egems.249 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Payne TH, Lovis C, Gutteridge C, Pagliari C, Natarajan S, Yong C, et al. Status of health information exchange: a comparison of six countries. J Glob Health 2019;9:0204279. doi: 10.7189/jogh.09.020427 [DOI] [PMC free article] [PubMed] [Google Scholar]