A brief educational intervention increases providers' human papillomavirus vaccine knowledge

Abstract

Recommendation by a healthcare provider is critical to increase human papillomavirus (HPV) vaccine uptake in the US. However, current deficits in providers' knowledge of HPV and its vaccine are not fully understood and interventions to amend knowledge gaps are untested. To determine whether attending a structured presentation could increase provider knowledge of the HPV vaccine, we assessed knowledge levels of physicians, non-physician healthcare workers, and medical students before and after attending a 30-minute lecture held between October 2012 and June 2014. Paired t-test and McNemar's test were used to compare knowledge scores and the proportion of correct responses for each question, respectively. Multiple linear regression analyses were performed to examine correlates of baseline knowledge and change in knowledge scores post-intervention. A total of 427 participants, including 75 physicians, 208 medical students, and 144 nurses or other healthcare workers, attended one of 16 presentations and responded to both pre-test and post-test surveys. Baseline knowledge was low among all groups, with scores higher among older participants and physicians/medical students. On average, knowledge scores significantly improved from 8 to 15 after the presentation (maximum possible score 16) (P < .001), irrespective of specialty, race/ethnicity, gender, and age. Although lower at baseline, knowledge scores of younger participants and non-physician healthcare workers (e.g., nurses, physician assistants (PAs), nursing students) improved the most of all groups. We conclude that a brief, structured presentation increased HPV knowledge among a variety of healthcare workers, even when their baseline knowledge was low.

Keywords:

• educational intervention
• healthcare providers
• HPV vaccine
• human papillomavirus (HPV)
• vaccine uptake

Abbreviations:

• HPV, Human papillomavirus
• UTMB, University of Texas Medical Branch
• PA, physician assistant

Introduction

Healthcare provider recommendation is the single best predictor of adolescent vaccination against HPV,^1-5 yet more than one-third of parents of eligible adolescent girls do not receive a recommendation to vaccinate their child from their doctor.⁵ Moreover, less than a quarter of providers always recommend this vaccine to male adolescent patients.^6,7 Lack of recommendation by providers could be related to their own understanding of HPV and the vaccine that prevents it, as providers with more knowledge of these topics have been found to recommend or administer the vaccine more often or report higher intentions to do so.^7-9 Conversely, deficits in HPV knowledge may be associated with lower rates of vaccine recommendation. For example, Perkins and Clark¹⁰ found in a qualitative study of 31 providers that only 2 knew that HPV vaccination offered any cancer prevention benefits to males and few offered the vaccine to boys.

If provider knowledge is related to recommendation, then educating providers on HPV epidemiology and vaccine-related issues could play an important role in improving the vaccine's uptake among adolescent children in the US. Provider education has been shown to improve clinic behaviors with other immunizations.¹¹ However, studies specifically exploring HPV-related educational interventions for providers are limited. In fact, a recent editorial asserts that the only study testing a provider-focused intervention explored decision support, not education.¹² The objective of this study was to fill this gap in the literature by assessing whether a brief structured presentation on HPV and the HPV vaccine could increase baseline knowledge among physicians, other healthcare workers, and medical students.

Results

A total of 427 clinicians, including practicing physicians (n = 75), medical students (n = 208), and other healthcare workers (n = 141; unknown 3), attended the lecture and returned both pre-test and post-test surveys. The specialties of the 75 practicing physicians were pediatricians (n = 39), obstetricians/gynecologists (n = 18), family medicine (n = 16) and unknown specialty (n = 2). Most of the respondents were less than 30 y old (74.2%), female (67.0%), white (46.6%), and enrolled in medical school (48.7%) (Table 1). Nearly 70% of the medical students were in their last 6 months of their 3rd year at the time of the lecture.

Table 1. Characteristics of participants and impact of presentation on HPV vaccine knowledge score (n = 427)

Characteristics		Pre-test score	Post-test score	Mean difference (95% CI)	P value
Overall	427	8.44	14.72	6.29 (6.01 to 6.56)	<.001
Age categories
<30	317 (74.2)	8.15	14.95	6.80 (6.51 to 7.09)	<.001
30–49	62 (14.5)	8.89	14.24	5.35 (4.51 to 6.20)	<.001
50 and above	41 (9.6)	9.70	13.83	4.12 (3.15 to 5.09)	<.001
Unknown	7 (1.6)	10.00	14.00	4.00 (0.84–7.16)	.021
Gender
Female	286 (67.0)	8.45	14.79	6.34 (6.00 to 6.69)	<.001
Male	135 (31.6)	8.39	14.61	6.22 (5.72 to 6.70)	<.001
Unknown	6 (1.4)	8.50	13.5	5.00 (0.45–9.55)	.037
Race/ethnicity
White	199 (46.6)	8.48	15.12	6.64 (6.23 to 7.04)	<.001
African American	40 (9.4)	7.73	13.68	5.95 (4.93 to 6.97)	<.001
Hispanic	69 (16.2)	8.81	14.33	5.52 (4.78 to 6.26)	<.001
Asian American	73 (17.1)	8.58	14.68	6.11 (5.52 to 6.70)	<.001
Other	26 (6.1)	8.00	14.23	6.23 (5.07 to 7.38)	<.001
Unknown	20 (4.7)	7.75	14.00	6.25 (4.58–7.92)	<.001
Participants
MD/DO	75 (17.6)	9.48	14.51	5.03 (4.39 to 5.66)	<.001
Medical student	208 (48.7)	8.58	14.88	6.31 (5.99 to 6.62)	<.001
Other	141 (33.0)	7.63	14.57	6.93 (6.35 to 7.53)	<.001
Unknown	3 (0.7)	10.33	16.00	—

Maximum possible score = 16

^a P values were based on paired-t tests.

^b Includes 39 pediatricians, 18 Ob-Gyn physicians, 16 family physicians and 2 with unknown specialty.

^c E.g., registered nurse, nurse practitioner, nurse student, physician assistant (PA), PA student, medical assistant, certified nurse midwife.

^d Statistical analysis was not performed due to small number of observation in this category.

The mean baseline HPV knowledge score was 8.4 (SD 2.5, range 2–15) (maximum possible 16). After adjusting for demographic characteristics, older respondents, MD/DOs, and medical students were observed to have higher baseline knowledge scores compared to their counterparts (Table 2). African-American respondents had lower adjusted baseline knowledge scores compared to whites (P = .001), while female respondents had slightly higher adjusted baseline knowledge scores than male respondents (P = .023) (Table 2). After the educational intervention, the mean score went up to 14.7 (SD 1.9, range 7–16) with significant improvement across all age groups, genders, races/ethnicities, and specialties (Table 1). Younger participants and non-physician healthcare workers such as nurses, nursing students, physician assistants (PAs), and PA students scored lower on the pre-test, but had the same level of understanding as physicians after the educational intervention. These participants exhibited the largest improvement in knowledge scores (Table 2).

Table 2. Correlates of participants' HPV vaccine-related knowledge before and after the structured presentation

	Knowledge score at baseline		Changes in knowledge score after presentation
Characteristics	Regression coefficient (95% CI)	P value	Regression coefficient (95% CI)	P value
Age categories
<30	Reference		Reference
30–49	1.01 (0.29 to 1.73)	.006	−1.41 (−2.24 to −0.58)	.001
50 and above	2.41 (1.54 to 3.28)	<.001	−3.32 (−4.32 to −2.32)	<.001
Gender
Female	Reference		Reference
Male	−0.62 (−1.15 to −0.09)	.023	0.36 (−0.25 to 0.97)	.248
Race/ethnicity
White	Reference		Reference
African American	−1.35 (−2.17to −0.53)	.001	−0.19 (−1.13 to 0.76)	.698
Hispanic	0.15 (−0.50 to 0.81)	.646	−0.98 (−1.74 to −0.23)	.010
Asian American	−0.25 (−0.91 to 0.40)	.448	−0.28 (−1.04 to 0.47)	.462
Other	−0.40 (−1.40 to 0.60)	.434	0.24 (−0.91 to 1.38)	.687
Participants
MD/DO	Reference		Reference
Medical student	0.04 (−0.70 to 0.78)	.920	0.19 (−0.66 to 1.04)	.656
Other	−1.58 (−2.60 to −1.17)	<.001	1.74 (0.92 to 2.56)	<.001

^a E.g., registered nurse, nurse practitioner, nurse student, physician assistant (PA), PA student, medical assistant, certified nurse midwife.

Depending on the question asked, 82.4% to 99.1% of participants responded correctly in the post-test survey compared to 18.0% to 99.3% correct response measured at baseline (Table 3). Analysis by provider type and question topics showed that MD/DOs more frequently provided correct responses to the pre-test survey questions on issues such as the necessity of HPV testing before administering the HPV vaccine, vaccination in those breastfeeding, how to continue the HPV vaccine series for those incompletely vaccinated, and the correct dosing intervals (Table 3). Both MD/DOs and medical students had more correct baseline responses than other healthcare workers to questions about the types of cancer prevented by the HPV vaccine. Medical students had more correct post-test responses to the question on whether the HPV vaccine prevents vaginal cancer, and they, along with non-physician healthcare workers, had more correct post-test responses on the race/ethnic group with the highest cervical cancer incidence in the US. HPV related knowledge and knowledge change did not differ by specialty among physician providers.

Table 3. Proportion of correct responses to multiple choice and true-false questions administered before and after the structured presentation

	Overall (n = 427)			By Provider Type
True-False Statements and Questions	Pre-test n = 427	Post-test n = 427	P value	Pre-test			Post-test
				MD/DO n = 75	Medical Student n = 208	Other n = 141	MD/DO n = 75	Medical Student n = 208	Other n = 141
Sexually active women who want the vaccine should be tested for HPV before receiving the first dose.	71.4	98.8	<.001	85.3	72.1	63.8	98.7	98.6	99.3
Gardasil should not be administered to breastfeeding patients.	67.7	96.7	<.001	92.0	75.0	44.7	98.7	95.7	97.2
If a patient received her 1st HPV vaccination >1 y ago but never returned for follow-up, she should start the series again and get 3 additional injections.	48.1	96.7	<.001	69.3	36.1	55.3	97.3	96.2	97.2
What is the recommended age for the HPV vaccine?	67.7	98.1	<.001	74.7	67.8	63.1	97.3	97.6	99.3
What are the recommended dosing intervals for the 3 doses of Gardasil?	53.2	95.1	<.001	69.3	44.2	57.5	93.3	94.7	96.5
How many types of HPV are associated with cervical cancer?	23.7	83.8	<.001	22.7	27.9	17.0	81.3	88.9	77.3
What percent of cervical cancer in the US is caused by the 2 high-risk HPV types (16, 18) that Gardasil protects against?	56.9	91.6	<.001	62.7	57.7	52.5	88.0	89.4	96.5
Which race/ethnicity has the highest incidence of cervical cancer in the US?	18.0	94.2	<.001	18.7	14.4	22.0	81.3	97.6	95.7
What proportion of 13–17 y old females was fully vaccinated in the US in 2010?	74.7	94.4	<.001	72.0	74.5	75.9	94.7	91.8	97.9
What is the HPV immunization goal for females to ensure maximum herd immunity?	18.5	91.8	<.001	16.0	17.3	22.0	93.3	92.8	89.4
The HPV vaccine prevents cervical cancer.	99.3	99.1	.250	100.0	99.5	98.6	100.0	98.6	99.3
The HPV vaccine prevents cancer of vulva.	35.4	84.1	<.001	44.0	37.5	27.0	81.3	87.5	80.1
The HPV vaccine prevents vaginal cancer.	31.9	82.4	<.001	32.0	31.7	31.9	77.3	87.0	78.6
The HPV vaccine prevents ovarian cancer.	89.5	93.4	<.001	97.3	94.7	77.3	97.3	92.3	92.9
The HPV vaccine prevents penile cancer	40.1	85.3	<.001	49.3	47.1	24.1	85.3	87.5	81.6
The HPV vaccine prevents anal cancer.	47.1	86.7	<.001	42.7	60.1	30.5	85.3	92.3	78.7

^a Three participants whose specialty were not known were excluded from this analysis.

^b P values were based on McNemar's chi square or exact test.

^c Statistical comparisons among 3 provider groups were made using chi-square test or Fisher's exact test.

^d E.g., registered nurse, nurse practitioner, nurse student, physician assistant (PA), PA student, medical assistant, certified nurse midwife.

^e P <.017 was considered as statistically significant to adjust for multiple comparison as the comparison was made among 3 groups.

^f No Statistical difference was observed among pediatricians, family physicians and obstetricians/gynecologists.

Discussion

We observed that physicians, medical students, and other healthcare workers had inadequate baseline knowledge of HPV epidemiology and the HPV vaccine. For example, only one-third knew that the vaccine protects against vulvar and vaginal cancers and only one-half knew that it protects against anal cancer, even though the vaccine is indicated to protect against these diseases.

These findings are in agreement with several prior studies that have identified inadequate levels of knowledge about HPV disease and prevention among providers in the US and abroad.^13–17 For example, a 2008 survey of Texas physicians found knowledge gaps in 4 basic facts about HPV.⁹ Our study confirms that significant knowledge gaps are still common nearly a decade after its approval. Moreover, studies have shown that knowledge gaps are more common among pediatricians than obstetrician-gynecologists and family medicine physicians.¹⁸ This is concerning given that pediatricians are more likely than other specialists to care for 11 and 12 y olds, the age group recommended by the Centers for Disease Control and Prevention for HPV vaccination.

We also observed that 80% of physicians did not know that Hispanic women have the highest incidence of cervical cancer in the US, even though the study was conducted in an area with many Hispanic community members. This agrees with a prior study which showed that even providers serving communities at high risk for HPV-related cancers have limited knowledge of HPV epidemiology.¹⁹ This misperception could limit HPV vaccine uptake among Hispanic women who should be vigorously targeted for vaccination due to their increased risk for cervical cancer. Healthcare providers have the power to greatly impact this population as provider recommendation has been shown to be a key determinant of HPV vaccination among Hispanic adolescents with Spanish-speaking parents.^20,21

In our study, baseline knowledge levels were lower for non-physician healthcare workers than physicians. However, the former showed greater improvement in scores and their post-test knowledge scores were similar to those of physicians. Thus, educational interventions of this type not only improve HPV-vaccine knowledge among physicians, but also bring other healthcare workers' knowledge to the same level as physicians and medical students. Nurses may be a particularly important type of non-physician healthcare worker to educate about HPV prevention as studies have shown that parents view nurses as an important source of vaccine-related information and vaccination programs that rely on nurses can improve vaccine coverage.^22-25

It is concerning that many providers in our study were ill-informed about how to administer the HPV vaccine; over 40% did not know the recommended dosing intervals or how to complete the series if the patient failed to adhere to the recommended schedule. Furthermore, almost one-third did not even know the recommended age for this vaccine. This demonstrates that additional educational interventions need to be developed to better inform all types of healthcare providers.

Our study has both strengths and limitations. Use of a rolling accrual design was the main strength of this study as it allowed to reach a fairly large number of providers with this educational tool and test their knowledge before and after the lecture. While we observed substantial improvement in HPV-vaccine knowledge among healthcare workers immediately after completing the intervention, we do not know actual retention of information over time. Another study, however, found that significant improvement of provider knowledge lasts for at least 90 d following a continuing medical education program on lifestyle medicine.²⁶ We also do not know whether the improvement in knowledge we observed led to a change in provider attitude or behavior (recommending the HPV vaccine). Information on intention to recommend this vaccine would have been useful in this regard. However, we included limited items in the pre/post-test questionnaire to comply with the limited time available before and after the lecture session for the questionnaire. Also, psychometric properties of the items we used in the short questionnaire were not measured systematically. Finally, our study was based on one academic institution located in the southeast Texas. Thus, we do not know whether our findings are generalizable to other academic institutions or other types of clinics that deliver the HPV vaccine. However, the information generated in this study will not be outdated even when the Guardasil-9 vaccine is available for use as we examined participants' knowledge of general information about HPV and the vaccine.

In conclusion, structured presentations on HPV epidemiology and the HPV vaccine can improve knowledge among healthcare providers, which may improve the quality or quantity of their counseling regarding this vaccine. Future studies on the impact of this type of educational intervention on HPV vaccine uptake would shed more light on its effectiveness.

Materials and Methods

We used a quasi-experimental pre-test/post-test design to examine the ability of structured presentations to improve provider knowledge of the HPV vaccine. Between October 2012 and June 2014, we conducted 16 presentations on HPV and the HPV vaccine for physicians, nurses, and allied health professionals working in pediatrics, obstetrics and gynecology, and family medicine departments at The University of Texas Medical Branch at Galveston (UTMB). These lectures were targeted at providers of these 3 departments because they provide care to potential candidates for this vaccine in their clinics i.e., adolescent children, young adult women. Allied health professionals, especially nurses, were targeted because they are instrumental in HPV vaccination²⁷ and considered important patient educators in UTMB's clinics which care for uninsured, underinsured and underserved populations. In an effort to continue advancing medical education regarding the HPV vaccine and related issues, the first author (ABB) also presented lectures to third-year medical students as part of their medical rotations and to students enrolled in UTMB's physician assistant (PA) program.

Before each lecture, we informed attendees that they would be asked to complete an anonymous, voluntary, 1-page, pre-test survey that assessed their baseline knowledge with multiple choice and true/false questions. The survey, developed by the first and second author (ABB and RM), assessed provider knowledge regarding HPV vaccine uptake rates in the US, the vaccine's role in preventing cervical cancer and other cancers, the burden of cervical cancer by race/ethnicity, and the benefits of vaccination. After the 30-minute lecture, participants were asked to complete an identical post-test survey. Each pre-test and post-test survey took approximately 5–10 minutes to complete. One attendee's pre-test was excluded due to lack of a corresponding post-test. Participants were given the option to choose a small gift such as a reusable lunch bag or plastic cup from a nearby gift basket as compensation for their participation.

Statistical analysis

Bivariate comparisons were performed using paired t-tests, McNemar's tests, chi-square tests, or Fisher's exact tests, as appropriate. Multiple linear regression analyses were conducted to examine correlates of baseline HPV knowledge scores as well as changes in scores after the presentation. Analyses were performed using STATA 12 (Stata Corporation, College Station, TX). All procedures were approved by the institutional review board of UTMB.

Disclosure of Potential Conflicts of Interest

R.E. Rupp received research funding from Merck Sharp & Dohme Corp. (Whitehouse, NJ) for participating in a multicenter trial of a nine-valent HPV vaccine (2014-ongoing). The other authors report no conflicts of interest.

Funding

Support for this study (A.B. Berenson, M. Rahman, R.E. Rupp, K.O. Sarpong) was provided by a prevention grant from the Cancer Prevention & Research Institute of Texas (CPRIT, awarded to A.B. Berenson). J.M. Hirth is supported by a research career development award (K12HD052023: Building Interdisciplinary Research Careers in Women's Health Program-BIRCWH, A.B. Berenson is Principal Investigator/Program Director) from the National Institute of Allergy and Infectious Diseases (NIAID), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), and the Office of the Director (OD), National Institutes of Health (NIH). The content is solely the responsibility of the authors and does not necessarily represent the official views of CPRIT, NIAID, NICHD, OD, or the NIH.