Improving the Utilization of Human Papillomavirus and Cervical Cytology Co-testing for Cervical Cancer Screening in an Obstetrics and Gynecology Resident Clinic

Abstract

Human Papillomavirus (HPV) testing in combination with cervical cytology (HPV co-testing) has been recommended for cervical cancer screening for women 30 to 65 years of age. In several studies, HPV co-testing increased sensitivity for detecting high grade dysplasia and resulted in cost-savings. This retrospective cohort study assessed the prevalence of HPV co-testing in an obstetrics and gynecology resident clinic before and after a brief educational intervention which was designed to reinforce current cervical cancer screening recommendations. The intervention consisted of a short presentation that was given to all residents and medical assistants in October 2011. The proportion of women age 30–65 years of age who had cervical cancer screening with HPV co-testing as compared to cervical cytology alone was compared before and after the intervention using chi-square tests. The goal of the intervention was to increase the percentage of patients receiving co-testing from 0.5% to 7.8%. Each arm (pre- and post-intervention) required 130 subjects to achieve 80% power with a significance of P = .05. No significant differences in demographics including age, insurance type, and cytology were noted. HPV co-testing increased from 0% to 55% (P < .001). Of the 72 subjects who had co-testing, 58 (80%) will not need cervical cancer screening for another 5 years. HPV co-testing represents an underutilized cervical cancer screening modality for women 30 years and older. This brief educational intervention, adaptable to any clinical setting, significatnly increased co-testing at the clinical site.

Introduction

An increased risk of cervical cancer following infection with high-risk human papillomavirus (HPV) subtypes has been well documented.¹ In February 2003, the Food and Drug Administration approved the use of HPV testing in association with routine cervical cytology (HPV co-testing) as a screening modality for cervical cancer. In March 2003, the American Cancer Society (ACS), the American Society for Colposcopy and Cervical Pathology (ASCCP), and the American Society for Clinical Pathology (ASCP) released cervical cancer screening guidelines which recommended HPV co-testing every five years as the preferred cervical cancer screening tool for women 30 to 65 years of age.² When compared to cytology alone, HPV co-testing resulted in improved sensitivity (95% to 99%) for detecting high grade cervical dysplasia.³ Additionally, although the up-front cost of HPV co-testing is higher, health care costs are overall significantly decreased because of a longer screening interval for those with negative tests (5 years).⁴

In 2011, few providers at the University of Hawai‘i obstetrics and gynecology resident clinic had incorporated HPV co-testing as the recommended screening modality for cervical cancer. This mirrored what was occurring nationally; a 2010 survey of primary care physicians revealed that 19% of providers still recommended screening every three years even when women had normal cytology and a negative HPV test.⁵ To promote adherence to cervical cancer screening recommendations, a ten-minute educational intervention for resident providers and clinic staff was designed. The purpose of this study was to assess the efficacy of this educational intervention in increasing co-testing rates for women 30 to 65 years of age in an outpatient obstetrics and gynecology resident clinic.

Methods

This study was deemed exempt by the Institutional Review Board of Hawai‘i Pacific Health. A ten-minute educational intervention consisting of eight slides was designed. The slides briefly reviewed the data supporting the evidence and indications for HPV co-testing. Examples of the co-testing order form and the ordering procedure were included. The presentation concluded with a five-minute question and answer session and printed copies of the presentation were available upon request. No HPV co-testing supports were added into the existing electronic medical record.

A retrospective cohort study was performed to compare HPV co-testing before and after the intervention. We examined the medical records of all women between the ages of 30 and 65 years who were screened for cervical cancer at the Kapi‘olani Medical Center Women's Outpatient Clinic. This clinic serves as the primary clinic for the University of Hawai‘i Obstetrics and Gynecology resident physicians. The study period spanned from July 2011 through January 2012, which included the three-month period before and after the intervention. The principal investigator gave the presentation to all residents and medical assistants in the clinic in October 2012. Collected data included whether HPV co-testing was performed, cervical cytology and HPV results, and demographic information including patient age and insurance type.

The primary outcome was the difference in HPV co-testing before and after the educational intervention. At baseline we estimated approximately 0.5% of eligible patients were co-tested for HPV. Based on HPV co-testing data from John Hopkins University,⁶ an increase to 7.8% was determined to represent a clinically meaningful difference. A sample size of 130 subjects in each arm, 260 subjects total, was necessary to achieve 80% power with an alpha of 0.05. To achieve this sample size with our clinic volume, a six month data collection period spanning three-months before the intervention and three months after the intervention was required.

Chi-squared tests were used to determine the significance of association for categorical variables. Continuous variables were compared with a t-test. Sample size calculations were performed with Power and Sample Size (PASS) 2008 (NCSS LLC. Kaysville, Utah). All analyses were performed with Statistical Package for the Social Sciences (SPSS) version 16.0 (Chicago, Illinois). If differences in demographic characteristics were noted between patients presenting before and after the intervention, multiple logistic regression was planned.

Results

Two hundred thirty eight patients were eligible for HPV co-testing during the pre-intervention period and 131 were eligible during the post-intervention period. The two study groups were similar with respect to age and insurance type. Median age was 37 years in both groups (Table 1). No statistical difference was noted between groups with respect to cervical cytology diagnoses. A negative result was the most common cervical cytology finding in both groups (91%) (Table 2). Following the intervention, HPV co-testing significantly increased from 0% prior to the intervention to 55% after the intervention (P < .0001) (Table 3). Of the 72 women who had HPV co-testing, 58 had a negative result for both cervical cytology and HPV testing. Thus, no further cervical cancer screening will be needed for five years for these 58 women.

Table 1.

Pre- and Post-Intervention Patient Demographics

Patient Demographics	Pre-Intervention	Post-Intervention	P-value
Age (mean)	37.92 (n=248)	37.13 (n=131)	.37
Insurance Type
Medicaid and Medicare	62.9% (156/248)	53.4% (70/131)	.11
Private	30.6% (76/248)	35.1% (46/131)
No Insurance	6.5% (16/248)	11.5% (15/131)

Table 2.

Number and Percentage of Cervical Diagnoses by Type during the Pre- and Post-intervention Periods

Cytology result	Pre-Intervention	Post-Intervention	P-value
NEG	91.1% (226/248)	90.8% (119/131)	0.39
ASCUS	5.2% (13/248)	7.6% (10/131)
LSIL	2.0% (5/248)	1.5% (2/131)
HSIL	1.6% (4/248)	0% (0/131)

NEG = negative result. ASCUS= Atypical squamous cells of unknown significance. LSIL = Low-grade squamous intraepithelial lesion. HSIL = High-grade squamous intraepithelial lesion.

Table 3.

Number of Individuals Receiving HPV Co-testing during the Pre- and Post-intervention Periods

Co-Testing Performed	Pre-Intervention	Post-Intervention	P-value	OR (95% CI)
Yes	0% (0/248)	55% (72/131)	< .001	301.4 (41.0–2213.6)
No	100% (248/248)	45% (59/131)

OR=odds ratio. CI=confidence interval

Discussion

HPV co-testing is an effective means of cervical cancer screening in women between 30 and 65 years of age. The usual diagnostic procedure following an abnormal screening test involves performing a colposcopy which involves examining the cervix through a magnifying scope. The most common treatment procedure of high grade cervical dysplasia is a loop electrode excisional procedure (LEEP) which removes a portion of the cervix with a heated wire device. Both procedures are invasive and often cause some patient discomfort, and the LEEP can increase the risk of preterm delivery.⁷

HPV infection is critical in the natural history of cervical cancer.¹ After the release of the 2012 ASCCP guidelines, the recognition of the significance of HPV infection in the detection of cervical cancer increased among women's health care providers. More physicians recommended more frequent follow-up in patients with a positive HPV test compared to patients with a negative HPV test.⁸ A subsequent decrease in the number of colposcopy and LEEP procedures occurred.^9,10 Despite fewer procedures, an increase in cervical cancer was not noted. Instead, HPV testing became a successful adjunct to cervical cancer screening; a decrease in the incidence of cervical cancer was noted after the advent of HPV vaccination.¹¹ A need still exists to increase both the implementation of these practices among physicians, as well as the speed at which these practices are adopted. This study demonstrated that a simple intervention was successful in increasing HPV co-testing among a select group of providers.

Several factors contributed to the success of the study. First, the intervention was simple, short (one session), and inexpensive. These qualities mean that the intervention can be easily reproduced and disseminated to other women's health care providers to increase HPV co-testing at multiple health access points. Second, the intervention was done in a select group of providers and staff practicing in the same clinic. This allowed the quick administration of the intervention. The practice pattern in the clinic changed rapidly since all providers and staff were introduced to HPV co-testing concepts in a short period of time. Although this intervention could be implemented on a larger scale with an increased number of physicians in different settings, a longer time period may be necessary to achieve similar results.

Several limitations should be noted. The study intervention was conducted in a group of resident physicians and it is unclear whether the results are generalizable to other providers. Due to a relatively short time in practice, residents may not be as attached to practice patterns as attending physicians who may have longstanding practice patterns. Resident physicians are accustomed to discussing and adjusting practices based on information they provide to each other. The observed effect may have been due to this type of inter-resident discussion rather than the actual study intervention itself. Based on information provided in the intervention, residents would have been able to determine that part of the intent of the study was to increase HPV co-testing. Hence, it is possible that social desirability bias may have affected the results. Because the study took place in two sequential three-month time periods, it is unclear whether providers would have incorporated HPV co-testing regardless of the intervention. Additionally, the post-intervention study period was short and it is unknown whether the increase in HPV co-testing will be sustained. Olomu, et al, demonstrated that quality improvement measures implemented by physicians resulted in only a percentage of providers continuing the desired practice one year post-intervention suggesting that additional intervention measures should be performed at regular intervals.¹² Further study of this group of clinicians would be helpful to determine if the improved screening behavior was continued, or if additional interventions are warranted.

Due to rapidly emerging scientific data regarding the natural history of HPV and cervical dysplasia, the 2012 ASCCP guidelines will likely not be the last cervical screening guideline revision. Recently HPV testing alone has been offered as an alternative primary screening test to HPV co-testing and may soon replace HPV co-testing as the recommended primary screening test.¹³ Determining the best method to disseminate information rapidly into the larger medical community consisting of many independent practices would be a recommended area of future study. One possible alternative, publication in major medical journals, does seem to be effective in dictating physician behavior if given enough time.¹⁴ Other possible options include using the electronic medical record or other electronic health applications. Previous work by White and Kenton in 2013 suggested that the electronic medical record could be effective at increasing physician compliance with changing screening recommendations.¹⁵

In conclusion, a simple educational intervention can be a cost-efficient, effective method to improve the use of HPV co-testing in a resident clinic setting. Additional work is required to determine whether the improved screening behavior will be sustained in this population of physicians, as well as determining the best method for communicating ever-changing and complex guidelines to the medical community at-large.

Conflict of Interest

None of the authors identify a conflict of interest.