Pap Smear Versus Speculum Examination: Can We Teach Providers to Educate Patients?

Introduction

Low-income women who receive gynecological care in urgent care settings may believe that problem-specific pelvic examinations include routine cervical cytological screening. ^1,2 This is a concern because these women many not have routine cervical cancer screening later. The problem of patient misunderstanding has been noted as an action item by the American Cancer Society. In recent years, there has been a call by the American Cancer Society and other organizations to educate women that a pelvic examination does not equal a Papanicolau (Pap) test. ³

The American Cancer Society speculates that the largest gain in reducing cervical cancer incidence and mortality would be achieved by increasing screening among women who have not been screened or who have not been screened regularly. ^4,5 Populations at most risk for not being screened regularly are older women, women of low income or low education, and women who are uninsured. ⁶ Almost 20% of women with cervical cancer are diagnosed when they are >65 years of age. ¹ Cervical cancer occurs most often in Hispanic women, with a rate almost twice that of non-Hispanic white women. ⁷ African American women develop cervical cancer about 50% more often than non-Hispanic white women. ⁷ These groups of patients often do not have an identified primary care provider and choose to seek care through emergency centers, which are often busy and lack the personnel to properly educate all patients about routine health maintenance.

Educating patients about primary care services and procedures in the acute care setting has been demonstrated to be effective. Several studies have evaluated educational interventions in the emergency department for such conditions as acute asthma exacerbation prevention ⁸ and coronary artery disease (CAD) risk factors, ⁹ and a small study has addressed misconceptions about pelvic examinations and Pap testing. ¹⁰ All report that educating patients in an acute care setting about preventive services is a worthwhile endeavor to improve use of health maintenance services.

The objective of this project was to test whether provider-based training in focused patient communication can improve patient knowledge about Pap screening as part of the speculum examination in an urban urgent care setting. Pap tests are still the most commonly used form of cervical cytological screening performed. ¹¹

Materials and Methods

This interventional before-and-after survey study was conducted in an urgent care clinic at an urban public hospital in the Midwest. The Women's Visit Center (WVC) provides urgent care in a hospital clinic setting for women with any complaint, including abnormal discharge, new onset pelvic pain, heavy vaginal bleeding, spontaneous abortion, or suspected ectopic pregnancy and realizes approximately 15,000 annual visits. Most of the gynecological visits to the WVC include a speculum examination; very few include a Pap test. The care provided is urgent and episodic in nature, and all women are instructed to receive follow-up and routine care in the outpatient clinic setting. The patient population is ethnically diverse and economically disadvantaged. Women are evaluated and treated by resident physicians under the supervision of faculty physicians. The governing Institutional Review Board approved this study.

All women seeking urgent care at the WVC were offered an exit survey to determine their knowledge about both the examination they received and Pap smears in general. Consecutive women were given a survey regardless of the complaint upon presentation. Women were excluded only if they actually received a Pap smear during that visit or spoke a language other than English or Spanish. Briefly, the survey asked what a Pap smear test was for, how often it should be performed, and if the patient thought she had received a Pap smear at the time of her visit.

Subjects were asked to complete the anonymous surveys in English or Spanish. Spanish translations were accomplished by certified translators and verified by translation-reverse translation, with consensus used to resolve differences in wording. Subjects were able to complete the survey in their examination room in private before leaving from their visit. No identifying information was recorded on the survey form. The survey is provided in the Appendix. Because the surveys were anonymous and not analyzed until later, we were unable to correct any misperceptions at that acute care visit. However, all women were given a follow-up visit in the clinic for routine care where education is also provided.

We first assessed baseline Pap smear knowledge in a convenience sample of WVC patients. After completion of the baseline preintervention period, an educational session was provided to the resident physicians who work in the WVC as the intervention. Provided by the lead author, the education instructed the residents how to educate patients in the clinical setting, modeled after the One-Minute Preceptor methodology, ¹² as follows:

All residents were instructed to provide the education verbally at every patient encounter that involved a speculum examination. The script was printed on cards for them and readily available in the WVC. Spanish interpreters or Spanish-speaking residents gave all the instructions to subjects who preferred information in Spanish. We then administered the same exit survey to consecutive women coming to the WVC in the time after the intervention.

Responses from preintervention and postintervention surveys were compared using SPSS version 15 (SPSS, Inc., Chicago, IL). Differences in knowledge between the English-speaking and Spanish-speaking subgroups were also compared within survey groups. Dichotomous variables were compared using chi-square tests. Continuous variables were compared using a t test. A logistic regression model was constructed to asses for predictors of correct knowledge. The logistic regression model controlled for age, whether the survey was completed in Spanish or English, gravidity, age at first Pap test, whether the subjects thought they had received education at that visit, and whether they were in the postintervention group. A p value <0.05 was required for statistical significance. We assumed that 40% would have correct knowledge about whether they had a Pap smear before the intervention. In order to accurately estimate the proportion of women who knew this, using the formula: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}\begin{align*}{\rm n} = ({\rm z}^2 {\rm pq}){\rm e}^2\end{align*}\end{document}

and given values for z = 1.96 and e = 5%, we determined we would need at least 368 surveys in the preintervention group. Based on that assumption, in order to have adequate statistical power to demonstrate a 30% improvement in the proportion of women answering the question correctly, we determined we would need at least 42 women in the postintervention group.

Results

The preintervention sample had a total of 383 patients, and the postintervention sample had 130 patients who completed the survey. The mean age of patients surveyed was 29.2 (±9.3) years of age. The mean reported age of first Pap smear was 16.7 (±3.9) years of age, and the mean gravidity was 2.6 (±1.9) pregnancies per patient. Table 1 shows the characteristics of the survey respondents in the preintervention and postintervention phases. There were no statistically significant differences in average age, gravidity, age of first Pap smear, or English vs. Spanish speakers between the two groups.

In the bivariate analysis (Table 2), there was an increase in the number of patients who correctly knew that they had not received a Pap smear from the preintervention group to the postintervention group (n = 221 of 383, 57.7% vs. n = 97 of 130, 70.7%, respectively) (odds ratio [OR] 2.15, 95% confidence interval [CI] 1.35-3.44, p = 0.013). There was no difference in knowledge about Pap screening interval (OR 1.00, 95% CI 0.64-1.58). Knowledge of what a Pap smear screens for was higher in the postintervention group (n = 45 of 383, 11.7% vs. n = 30 of 130, 23.1%, OR 2.28, 95% CI 1.31-3.88, p = 0.002).

We conducted several subgroup analyses to better understand the groups most needing targeted intervention. Most women who knew the correct use of the Pap smear also correctly reported that they did not receive a Pap smear at that visit (n = 62/75, 82.7%) compared with only 250 of 438 (57.1%) of those women who did not know the use of a Pap smear correctly reporting if they had one at the visit (p<0.001).

Spanish-speaking patients were nearly three times more likely to know what a Pap smear screened for compared with English-speaking patients in the preintervention group (n = 37 of 53, 30.2% vs. n = 29 of 330, 8.8%, respectively, p<0.001), although there was no difference in the postintervention phase (n = 4 of 13, 30.8% vs. n = 26 of 117, 22.2%, respectively, p = 0.49). The improvement seen in English-speaking patients postintervention was statistically significant (p<0.001). There were no differences noted between English and Spanish respondents in knowledge about the Pap screening interval. Respondents completing the survey in English had a significant improvement in knowledge about whether they received a Pap smear at that visit (n = 183 of 330, 55.5% preintervention vs. n = 84 of 117, 71.8% postintervention, p = 0.002), whereas respondents submitting the survey in Spanish did not note improvement in correct responses to that question (n = 38 of 53, 71.7% preintervention vs. n = 7 of 13, 53.8% postintervention, p = 0.22).

Table 3 shows the results of the logistic regression controlling for age and other potential confounders. As correctly identifying the proper Pap screening interval had no significant bivariate associations, that outcome is not included. Responding to the survey in Spanish predicted a higher chance of knowing what a Pap smear is used for. Being in the postintervention group predicted accurate knowledge about the purpose of the Pap smear and also correctly knowing if they received a Pap smear during the visit. Surprisingly, women who noted they had education about Pap smears that day in clinic less frequently knew if they had a Pap smear that day.

As a process evaluation, we asked if the woman believed she had received Pap smear education at the visit. Only 66.3% (n = 81 of 128) of women in the postintervention group believed they were educated at the visit about Pap smears compared with 33.1% (n = 123 of 372) (p < 0.001) in the preintervention group.

Discussion

This study demonstrates that training providers to give targeted patient education in an acute gynecological care setting may improve patient knowledge about Pap tests. Nearly 70% in both groups knew the correct screening interval for Pap tests. The survey also revealed that this patient population has very limited knowledge of what a Pap smear screens for, and they often believe that any pelvic examination includes a Pap smear. These findings are consistent with those of other studies. ^10,13 With increased patient awareness of what a Pap smear screens for and understanding that they are not having a Pap smear performed in the urgent care setting, women may be more likely to follow up in clinic or an early detection program, thereby decreasing the morbidity and mortality of cervical cancer.

It was disappointing that after the intervention, only 23% of the women knew the correct use of a Pap test. This rate is similar to the 26% rate reported by Lyons et al. ¹⁰ Most who missed the question marked multiple answers, including the correct one. This may demonstrate either a lack of knowledge or a flaw in the survey methodology, with patients feeling a need to mark multiple options. We chose to count only women who marked the correct option and no incorrect ones as having the correct knowledge in order to limit the bias caused by guessing by women who marked more than one option. In future knowledge surveys, we plan to allow a Yes/No choice for each option in order to reduce the impact of this design issue. Although the intervention was designed to differentiate a Pap test from a pelvic examination and to define the purpose of the Pap test, clearly other factors were present that could have produced misperceptions and affected knowledge and survey responses. Truly dissecting the cognitive and informational barriers to correct knowledge acquisition is essential to further progress in this field.

There are several potential areas where improvements in educating patients can be made. Interpreters were used for patient care but were not used by the patients to take the surveys. The fact that there was no improvement in the knowledge of women taking the survey in Spanish after the educational intervention may mean that the education did not translate well in the clinic setting; perhaps the interpreters could benefit from the same One-Minute Preceptor training session given to the resident physicians. The very act of patient instruction on the Pap smear may also have had unintentional consequences on patient understanding. Patients may have believed that all the discussion about Pap smears meant that they did have a Pap smear that day. One way to correct this unintentional effect might be to reiterate specifically that a Pap smear was not performed and give the patient verbal and written instructions on where to follow up for routine health maintenance and Pap smear. Additionally, we did not ask when the patient had her last cervical cytology screening. We may have been surveying a population who did not obtain Pap tests regularly and may have lower baseline knowledge.

The strengths of our study include its focus on a population at risk for inadequate screening and adequate power to demonstrate important differences in patient knowledge after provider-focused intervention. This study has limitations common to surveys and intervention studies without randomization. The survey tool was not validated, and the ability of survey responses to predict actual screening behavior is unknown. Despite our use of logistic regression analysis, the nonrandomized design raises the possibility that improvement occurred for reasons other than the intervention. There is a potential that the preintervention and postintervention groups differed in a way we did not measure, which would threaten the validity of our findings. The separate subject samples may have received different routine instructions from other members of the patient care team. In addition to the residents, perhaps the nurses, staff, and interpreters increased their patient educational efforts, thus further improving the knowledge of the postintervention group. We did not assess if the resident actually did read the script. We also did not pilot test the script for understandability. Querying the WVC nurses present in the examination room would be a way to assess if the resident actually gave the correct education. Future studies will include methods to ensure that the resident gives the scripted education, such as a checkbox. This will help us assess if failures may be a result of ineffective intervention delivery and if the information is being presented in the best manner for the patient. Our study did not assess if the subjects actually changed behavior, followed up for routine Pap testing, or had long-term retention of the information taught to them during the visit. Future studies are needed to explore whether educating patients leads to increased follow-up in clinic for screening and whether the intervention needs to be reinforced over time. As there are multiple factors that may impact follow-up screening behaviors, such as health literacy, family history, socioeconomic status, and transportation, future research will be needed to assess the impact of education while controlling for these other variables.

Conclusions

This study demonstrates that training residents to educate women about Pap smears in the acute gynecological care setting may improve patient knowledge in the short term. Enabling providers to communicate brief, focused education to their patients has the potential to improve screening behavior in an indigent population that seeks care disproportionately from urgent and acute care settings.