Highlights
-
•
On-time follow-up rates after benign and low-grade colposcopy were low.
-
•
Tobacco nonsmoking and indeterminate biopsy results were associated with timely follow-up.
-
•
Timing of follow-up did not impact the results of cotesting.
Keywords: Human papillomavirus, cervical intraepithelial neoplasia, cervical cancer, Papanicolaou test, colposcopy
Abstract
Introduction
This study was aimed to evaluate follow-up after benign or low-grade colposcopic findings and identify the factors associated with delayed care.
Methods
The authors completed a retrospective study to evaluate follow-up for patients with benign or low-grade colposcopy findings over a 4-year period at a single institution. Demographics as well as cotesting and pathology results were abstracted. Descriptive statistics and bivariate associations were completed.
Results
Of 393 patients, 20.6% had an on-time follow-up. The majority of patients (50.4%) had delayed follow-up, with a mean time to repeat evaluation of 27.6 months. There were no differences in follow-up cytology and human papillomavirus test results between those patients who followed up within 12 months and those who were evaluated later. Tobacco nonsmoking and initial biopsy with indeterminate findings were associated with on-time follow-up.
Conclusions
This study demonstrated significant delays in follow-up after colposcopy, consistent with prior work showing insufficient cervical cancer screening and prevention nationally. Improving follow-up times and rates will likely depend on addressing barriers that impact access to care.
INTRODUCTION
As part of cervical cancer prevention, the American Society for Colposcopy and Cervical Pathology guidelines recommend that patients who have an abnormal Papanicolaou (Pap) test undergo colposcopy for further assessment.1 Colposcopy includes microscopic evaluation of the cervix and may entail biopsies of the endocervix and ectocervix to identify precancerous lesions. If colposcopy identifies either benign tissue or cervical intraepithelial neoplasia (CIN) I, follow-up is recommended in 12 months with a repeat Pap test and human papillomavirus (HPV) testing.1 Although most low-grade cervical dysplasia regress or remain stable, up to 11% of patients with CIN I have been shown to progress to CIN II or beyond with conservative management.2 Therefore, it is important for patients to undergo continued surveillance as a mechanism for cancer prevention.
Previous research has shown that nearly 25% of patients do not receive cervical cancer screening in accordance with national recommendations.3 For patients who need colposcopy, the procedure is often delayed, with over half of patients with low-grade cytology HPV-positive Pap tests not undergoing colposcopy within 6 months.4 A recent study of pregnant patients in whom colposcopy was indicated showed that only 40.8% completed the recommended procedure.5 Follow-up rates after colposcopy have been reported to be as low as 54%.6 However, little is known about follow-up for patients who specifically have benign or low-grade findings on colposcopy.
The objectives of this study were to evaluate follow-up after benign or low-grade colposcopy findings and identify the factors associated with delayed or absent follow-up. The authors hypothesized that patients in this study would experience delays in care, mirroring what occurs throughout cervical dysplasia and cancer screening and management.
METHODS
Study Population
The study setting is a public safety net hospital that provides care for a largely publicly insured and underserved patient population. Patients undergoing colposcopy at the authors’ site are referred from within the authors’ care system or outside clinics.
Measures
This retrospective study was approved by the authors’ IRB (Approval Number STUDY2022-1338-MOD001). Patients who underwent colposcopy at an urban, academic colposcopy clinic between May 2015 and December 2018 and had benign, low-grade, suboptimal results (determined through biopsy or visual inspection) with a recommendation to follow up in 12 months were included. Suboptimal results included insufficient tissue samples and indeterminate dysplasia as graded by pathology. Demographic information, including age, ethnicity, race, insurance type, smoking status, solid organ transplant status, and ZIP code, were retrospectively accumulated through electronic medical record review. Race was designated as Black, White, or none of the above (which included patients indicating other, Asian American or Pacific Islander, American Indian, and unknown).
Antecedent and follow-up cytology and HPV results were reported. Cytology was reported as negative for intraepithelial lesion or malignancy, atypical squamous cells of undetermined significance, low-grade squamous intraepithelial lesion, high-grade squamous intraepithelial lesion, atypical squamous cells that cannot exclude high-grade squamous intraepithelial lesion, squamous cell carcinoma, atypical glandular cells of undetermined significance, atypical glandular cells not otherwise specified, and unsatisfactory.
The primary study outcome was on-time follow-up within a year of colposcopy. Secondary outcomes included the results of follow-up testing and associations between demographic variables, follow-up testing, and time to follow-up.
Statistical Analysis
Descriptive statistics were calculated as mean and SD. Bivariate associations of demographic variables with follow-up were evaluated using Student’s t-test and chi-square tests. To determine the association of predictive variables with time to follow-up, a multivariable Cox regression model was performed. Sensitivity analyses were also completed with a more liberal cutoff of 14 months rather than 12 months. Bonferroni correction was completed for multiple variables. Time to follow-up was measured in months, and right censoring occurred for those who did not have a follow-up within the study period. All tests of significance were 2 sided, and a p=0.05 was considered significant. SPSS (version 29.0) (IBM Corp., 2022, Armonk, NY) was used for analysis.
RESULTS
A total of 393 individuals were included, of whom 81 (20.6%) followed up within 12 months. The majority of patients (50.4%) had delayed follow-up after 12 months. Table 1 illustrates participant demographics. The majority of study patients self-identified as Black (61.6%) and were publicly insured (76.1%). Most patients’ (71.3%) residential ZIP code was within 10 miles of the authors’ institution, roughly approximating the limits of local public transit.
Table 1.
Demographic Information
| Characteristics | Total (N=393) | Follow-up within 12 months (n=81) | Follow-up >12 months (n=198) | Lost to follow-up (n=114) | X2 or t-test, df, p-value |
|---|---|---|---|---|---|
| Months to follow-up, mean (SD) | 9.7 (2.6) | 27.6 (15.7) | — | — | |
| Race, n (%) | 14.98, 4, <0.01 (0.005) | ||||
| White | 43 (10.9) | 13 (16.0) | 11 (5.6) | 19 (16.7) | |
| Black | 241 (61.2) | 52 (64.2) | 129 (65.2) | 60 (52.6) | |
| None of the above | 110 (27.9) | 16 (19.8) | 58 (29.3) | 36 (30.7) | |
| Ethnicity, n (%) | 7.07, 4, 0.13 | ||||
| Hispanic | 90 (22.9) | 12 (14.8) | 46 (23.2) | 32 (28.1) | |
| Non-Hispanic | 286 (72.7) | 67 (82.7) | 144 (72.7) | 75 (65.8) | |
| Unknown | 17 (4.3) | 2 (2.5) | 8 (4.0) | 7 (6.1) | |
| Insurance, n (%) | 2.09, 2, 0.35 | ||||
| Public | 298 (75.8) | 64 (79.0) | 144 (72.7) | 90 (78.9) | |
| Private | 95 (24.2) | 17 (21.0) | 54 (27.3) | 24 (21.1) | |
| Distance, n (%) | 1.91, 2, 0.38 | ||||
| 10.0 miles or closer | 280 (71.3) | 61 (75.3) | 143 (72.2) | 76 (66.7) | |
| 10.1 miles or further | 113 (28.8) | 20 (24.7) | 55 (27.8) | 38 (33.3) | |
| Antecedent Pap results, n (%) | 6.25, 10, 0.79 | ||||
| NILM | 39 (9.9) | 8 (9.9) | 18 (9.1) | 13 (11.4) | |
| ASCUS | 243 (61.8) | 48 (59.3) | 125 (63.1) | 70 (61.4) | |
| AGUS or AGC-NOS | 7 (1.8) | 3 (3.7) | 2 (1.0) | 2 (1.8) | |
| LSIL | 72 (18.3) | 13 (16.0) | 37 (18.7) | 22 (19.3) | |
| ASC-H | 24 (6.1) | 8 (9.9) | 11 (5.6) | 5 (4.4) | |
| Unsatisfactory | 8 (2.0) | 1 (1.2) | 5 (2.5) | 2 (1.8) | |
| Antecedent HPV, n (%) | 2.69, 4, 0.61 | ||||
| Negative | 52 (13.2) | 7 (10.0) | 32 (17.1) | 13 (12.5) | |
| HPV 16 or 18 positive | 126 (32.0) | 26 (37.1) | 64 (34.2) | 36 (34.6) | |
| Other HPV positive | 184 (46.8) | 38 (53.5) | 91 (48.7) | 55 (52.9) | |
| Smoking, n (%) | 7.09, 2, <0.05 (0.029) | ||||
| Yes | 33 (8.4) | 2 (2.5) | 16 (8.1) | 15 (13.2) | |
| No | 360 (91.6) | 79 (97.5) | 182 (91.9) | 99 (86.8) | |
| Colposcopy pathology, n (%) | 9.89, 6, 0.13 | ||||
| None | 53 (13.5) | 13 (16.0) | 21 (10.6) | 19 (16.7) | |
| Benign | 160 (40.7) | 29 (35.8) | 86 (43.4) | 45 (39.5) | |
| LSIL | 173 (44.0) | 35 (43.2) | 90 (45.5) | 48 (42.1) | |
| Suboptimal for evaluation | 7 (1.8) | 4 (4.9) | 1 (0.5) | 2 (1.8) |
Note: Boldface indicates statistical significance (p<0.05).
AGC-NOS, atypical glandular cells not otherwise specified; AGUS, atypical glandular cells of undetermined significance; ASC-H, atypical squamous cells that cannot exclude high-grade squamous intraepithelial lesion; ASCUS, atypical squamous cells of undetermined significance; df, difference; HPV, huma papillomavirus; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy.
Of patients with delayed follow-up, the mean time to follow-up was 27.6 (SD=15.7) months. A total of 114 (29.0%) patients were lost to follow-up during the study period. For patients with no documented follow-up, 46.0% continued to receive care within the authors’ system in the first 12 months after colposcopy, and 24.8% of patients had documented care within the authors’ system >12 months after colposcopy.
Smoking was associated with follow-up behavior; individuals who did not smoke tobacco were more likely to return within 12 months and less likely to be lost to follow-up. (X2[2]=7.09, p<0.05). Follow-up behavior also differed across self-reported race categories, in which after Bonferroni-corrected posthoc analyses, White individuals were less likely to follow up beyond the 12-month suggested window (X2[2]=14.98, p<0.01). Sensitivity analyses that used a more liberal definition of on time at 14 months rather than 12 months, allowing for 27 more individuals to be considered on time, showed parallel results.
Age, insurance status, distance from the hospital, and antecedent Pap and HPV results were not associated with the time to follow-up (Table 1). Among the 279 patients who underwent follow-up surveillance, 143 patients (51.44%) had normal cytology, and 17 (6.12%) had high-grade cytology (Table 2). There were no differences in follow-up cytology and HPV test results between those patients who followed up within 12 months and those who were evaluated later.
Table 2.
Follow-Up Outcomes
| Follow-up outcomes | Follow-up within 12 months | Follow-up >12 months | X2 or t-test, df, p-value |
|---|---|---|---|
| Follow-up pathology, n (%) | 5.38, 5, 0.37 | ||
| NILM | 42 (51.9) | 102 (51.3) | |
| ASCUS | 28 (35.4) | 54 (27.1) | |
| LSIL | 4 (5.1) | 19 (9.5) | |
| AGC-NOS | 1 (1.3) | 1 (0.5) | |
| High-grade (ASC-H, HSIL, suspect squamous cell carcinoma) | 4 (4.9) | 13 (6.6) | |
| Unsatisfactory | 1 (1.3) | 10 (5.0) | |
| Follow-up HPV, n (%) | 1.11, 2, 0.57 | ||
| Negative | 37 (51.4) | 104 (58.1) | |
| HPV 16 or 18 positive | 5 (6.9) | 13 (7.3) | |
| Other HPV positive | 30 (41.7) | 62 (34.6) |
AGC-NOS, atypical glandular cells not otherwise specified; ASC-H, atypical squamous cells that cannot exclude high-grade squamous intraepithelial lesion; ASCUS, atypical squamous cells of undetermined significance; df, difference; HPV, huma papillomavirus; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy.
In the baseline Cox Regression survival analysis in Table 3, individuals who did not identify their race as Black or White were shown to have significantly lower odds of on-time follow-up (hazard ratio [HR]=0.44, 95% CI=0.21, 0.91) (Table 3). In the adjusted model that controlled for race, smoking, insurance, residential distance from the hospital, and antecedent testing result, none of the above race (HR=0.36, 95% CI=0.16, 0.80) and smoking cigarettes (HR=0.19, 95% CI=0.05, 0.80) were associated with a lower likelihood of on-time follow-up, whereas having a suboptimal colposcopy pathology result for evaluation was associated with higher odds of follow-up within 12 months (HR=4.22, 95% CI=1.28, 13.90).
Table 3.
Cox Regression Predicting Time to Follow-Up Across Baseline and Model Adjusted for Initial Pap, Colposcopy, and HPV Results
| Characteristics | Model 1, baseline HR (95%CI) |
Model 2, adjusted HR (95% CI) |
|---|---|---|
| Race | ||
| White | ref | ref |
| Black | 0.67 (0.36, 1.23) | 0.52 (0.26, 1.01) |
| None of the above | 0.44 (0.21, 0.91) | 0.36 (0.16, 0.80) |
| Smoking | ||
| No | ref | ref |
| Yes | 0.25 (0.06, 1.00) | 0.19 (0.05, 0.80) |
| Insurance | ||
| Private | — | ref |
| Public | — | 1.72 (0.93, 3.18) |
| Distance >10 miles | — | 0.66 (0.37, 1.17) |
| Pap result | ||
| NILM | — | ref |
| ASCUS | — | 0.90 (0.41, 1.96) |
| AGUS or AGC-NOS | — | 1.69 (0.20, 14.35) |
| LSIL | — | 0.78 (0.31, 2.07) |
| ASC-H | — | 1.43 (0.37, 5.60) |
| Unsatisfactory | — | 0.98 (0.11, 8.45) |
| Colposcopy pathology | ||
| None | — | ref |
| Benign | — | 0.53 (0.26, 1.10) |
| LSIL | — | 0.68 (0.34, 1.35) |
| Suboptimal for evaluation | — | 4.22 (1.28, 13.90) |
| HPV | ||
| Negative | — | ref |
| High risk HPV | — | 1.73 (0.72, 4.12) |
| Other HPV | — | 1.97 (0.85, 4.56) |
Note: Boldface indicates significant to at least (p<0.05).
AGC-NOS, atypical glandular cells not otherwise specified; AGUS, atypical glandular cells of undetermined significance; ASC-H, atypical squamous cells that cannot exclude high-grade squamous intraepithelial lesion; ASCUS, atypical squamous cells of undetermined significance; HPV, huma papillomavirus; HR, hazard ratio; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy.
DISCUSSION
In this study, the time to follow-up after benign or low-grade colposcopy was generally not consistent with the recommended guideline of 12 months. For the patients who ultimately followed up, the average time to do so was 22.39 months. These results echo other data illustrating gaps in the management of cervical dysplasia.3, 4, 5, 6
Individuals with pathology that was suboptimal for evaluation were more likely to follow up. These results may be driven by patient perception about indeterminate findings; however, the relatively small size of this group (1.8% of the overall population) limits generalizability.
Tobacco smoking is a known risk factor for development of cervical cancer.7 In this study, nonsmokers had higher rates of timely follow-up. This result may represent an area for targeted counseling or intervention focused on smoking cessation, education about the link between tobacco and cervical cancer, and assessment of specific barriers to follow-up. In univariate analyses, White patients were found to have lower than expected rates of delayed follow-up. In multivariate analyses, this difference was no longer present; instead, individuals of other race categories showed comparatively lower odds of on-time follow-up. These results may reflect barriers in access to care not fully captured within the study, including gaps in employment or insurance and scheduling limitations.
The majority of patients followed up but did so in a delayed fashion. Reassuringly, the results did not demonstrate an association between time to repeat evaluation and follow-up Pap test, cytology, or HPV results. This finding is consistent with the natural history of low-grade dysplasia; however, the risk of disease progression persists, and ways to improve timeliness should be addressed.2 At the time of this study, patients at the authors’ institution were unable to schedule cervical cancer screening 1 year in advance. Counseling, when accompanied by a scheduled appointment, may impact patient behaviors around follow-up. In addition, although patients were notified of their recommended follow-up through electronic medical record, telephone, or letter, there were no formal reminder systems to increase compliance. Furthermore, the type of patient contact was not able to be abstracted as a variable.
A large portion of the patient population was lost to follow-up during the study period, although many of them remained engaged in care at the authors’ institution. Likely, this group represents a heterogeneous group of patients who may have truly not completed follow-up, may have relocated, or may have established care elsewhere in the city. Since completion of this study, the authors’ healthcare system has transitioned to an electronic medical record that more easily allows providers to see results outside of the system, although not all health systems are included. Medical record–based interventions have been studied but need to be evaluated on a larger scale to see whether they lead to improved adherence to screening paradigms.8
All Pap and HPV testing in this study were completed by providers in the office. However, self-collection of HPV in the office is now approved as a screening option. Early studies of at-home self-sampling have shown promise in reaching patients who are due or overdue for cervical cancer screening.9 This new modality may have future applications for patients who have barriers to follow-up after colposcopy.
Qualitative work has shown that health literacy skills may be predictive of a patient’s understanding of cervical cancer screening.10 Interventions to improve health literacy and a patient’s understanding of follow-up needs may ultimately be key to ensuring appropriate surveillance.
Limitations
The study strengths include the sample size and multi-year follow-up after participant inclusion. Limitations were the retrospective design, which limits the ability to fully capture how patients were counseled about their results. In addition, the lost-to-follow-up group is likely heterogeneous and difficult to fully characterize.
CONCLUSIONS
Patients who underwent low-grade colposcopy at the authors’ institution had overall low rates of subsequent on-time follow-up. Improving follow-up times and rates will likely depend on addressing barriers that impact access to care and improving health literacy.
Acknowledgments
ACKNOWLEDGMENTS
The data that support the findings of this study are available from the corresponding author upon request.
Funding: None.
Declaration of interest: None.
CRediT AUTHOR STATEMENT
Lara D. Slesnick: Conceptualization, Methodology, Investigation, Writing - original draft. Elana Gordy: Data curation, Investigation. Quetzal A. Class: Conceptualization, Formal analysis, Writing – original draft, Writing – review & editing. Priyanka Gokhale: Conceptualization, Methodology, Validation, Writing – original draft, Writing – review & editing.
REFERENCES
- 1.Perkins R.B., Guido R.S., Castle P.E., 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(2):102–131. doi: 10.1097/LGT.0000000000000525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Loopik D.L., Bentley H.A., Eijgenraam M.N., IntHout J., Bekkers R.L.M., Bentley JR. The natural history of cervical intraepithelial neoplasia grades 1, 2, and 3: a systematic review and meta-analysis. J Low Genit Tract Dis. 2021;25(3):221–231. doi: 10.1097/LGT.0000000000000604. [DOI] [PubMed] [Google Scholar]
- 3.Lee Y.W., Morgan J.R., Fiascone S., Perkins RB. Underscreening, overscreening, and guideline-adherent cervical cancer screening in a national cohort. Gynecol Oncol. 2022;167(2):181–188. doi: 10.1016/j.ygyno.2022.09.012. [DOI] [PubMed] [Google Scholar]
- 4.Perkins R.B., Adcock R., Benard V., et al. Clinical follow-up practices after cervical cancer screening by co-testing: a population-based study of adherence to U.S. guideline recommendations. Prev Med. 2021;153 doi: 10.1016/j.ypmed.2021.106770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Roof K.A., Wichmann H.K., Carlton L.J., et al. Factors associated with loss to follow up after abnormal cervical cancer screening in pregnancy. Gynecol Oncol. 2024;185:46–50. doi: 10.1016/j.ygyno.2024.01.037. [DOI] [PubMed] [Google Scholar]
- 6.Chase D.M., Osann K., Sepina N., Wenzel L., Tewari KS. The challenge of follow-up in a low-income colposcopy clinic: characteristics associated with noncompliance in high-risk populations. J Low Genit Tract Dis. 2012;16(4):345–351. doi: 10.1097/LGT.0b013e318249640f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Castle PE. How does tobacco smoke contribute to cervical carcinogenesis? J Virol. 2008;82(12):6084–6085. doi: 10.1128/JVI.00103-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dupuis E.A., White H.F., Newman D., Sobieraj J.E., Gokhale M., Freund KM. Tracking abnormal cervical cancer screening: evaluation of an EMR-based intervention. J Gen Intern Med. 2010;25(6):575–580. doi: 10.1007/s11606-010-1287-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Winer R.L., Lin J., Anderson M.L., et al. Strategies to increase cervical cancer screening with mailed human papillomavirus self-sampling kits: a randomized clinical trial. JAMA. 2023;330(20):1971–1981. doi: 10.1001/jama.2023.21471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kulkarni A., Glynn S., Gamble C.R., et al. Understanding perceived barriers to colposcopy follow-up among underserved women at an urban teaching hospital: a qualitative study. J Low Genit Tract Dis. 2023;27(1):87–92. doi: 10.1097/LGT.0000000000000700. [DOI] [PubMed] [Google Scholar]