College women's knowledge, perceptions, and preventive behaviors regarding Human Papillomavirus infection and cervical cancer.

Abstract: Human Papillomavirus (HPV) infection has reached epidemic proportions in college females due in large part to high risk sexual activity. This trend indicates a rising wave of cervical cancer in the future, and the importance of regular PAP tests. This study examined HPV/cervical cancer knowledge, perceptions and sexual behaviors among 428 college women ages 18-30 using selected constructs from the Health Belief Model. Results indicate low levels of HPV/cervical cancer knowledge, perceived seriousness, and perceived susceptibility coupled with high-risk sexual behaviors. Relationships were found to exist between HPV/cervical cancer knowledge and perceived seriousness, HPV/cervical cancer knowledge and Pap test compliance, perceived susceptibility and number of sexual partners, perceived susceptibility and Pap test compliance, and perceived seriousness and Pap test compliance. Health educators are challenged with the task of increasing knowledge, perceived threat, and ultimately preventive behaviors among college women.

**********

Cervical cancer is a substantial public health issue for women in the United States (American College of Obstetricians and Gynecologists [ACOG], 1998). The American Cancer Society (1999a) estimates that approximately 16,000 new cases of cervical cancer are diagnosed every year in the United States. Infection with certain types of the Human Papillomavirus (HPV), (other than the types that cause genital warts), is the major risk factor for cervical cancer (Namkoong, 1995; National Institute of Allergy and Infectious Diseases [NIAID], 1999). Since 1980 there has been an increase in both HPV and cervical cancer (Center for Disease Control Division of Sexually Transmitted Disease Prevention, 1999). HPV infection is one of the most common of all sexually transmitted Diseases (STD) (ACOG, 1995).

Human Papillomavirus (HPV) is highly prevalent among young, sexually active adult women (Burak & Meyer, 1998). The level of sexual activity in a given population is a primary determinant of its sexually transmitted disease (STD) infection rate (Vail-Smith & White, 1992). The National Institutes of Health (1996) stated that HPV is transmitted through sexual intercourse, with peak prevalence in women of the 22-25 year age group. While many college students underestimate their risk of contracting various sexually transmitted diseases, HPV has become a common sexually transmitted infection on college campuses. College women have a greater risk of acquiring sexually transmitted diseases (STDs) than the general population because of the high-risk sexual behaviors in which they engaged (Burak & Meyer, 1997). One study round that as many as 60% of college-age women had some form of HPV, which is casually linked to cervical cancer (Alvey, 1995).

While many college women underestimate their risk of contracting various sexually transmitted diseases, HPV has become a common sexually transmitted infection on college campuses. The incidence and problems associated with contracting STD's are of great concern, especially in adolescents and young adults. A recent study showed that 40% of sexually active adolescent women had an STD (CDC, 1999). Each year approximately 3 million new cases of STDs are diagnosed in the U.S. (NIAID, 1999). One reason for file increase of STDs is risky sexual behavior including sexual contact without the use of a condom or barrier. Only 57% of sexually active students reported that they or their partner used condoms during their last sexual intercourse (CDC, 1999). These risky behaviors, a lack of knowledge and lack of preventive care including a pap test are leading to the high incidence of HPV among college women.

A lack of HPV knowledge coupled with misperceptions about susceptibility impacted college women's' attitudes and behaviors regarding cervical cancer prevention (Burak & Meyer, 1998). Vail-Smith and White (1992) found that 72% of female university students had never heard of HPV infection and were unaware of the related cervical cancer risk.

For decades the Health Belief Model had been used to explain and predict health behaviors. The Health Belief model has been applied to a variety of health issues including Acquired Immunodeficiency Syndrome (AIDS), cigarette smoking behavior, and predicting condom use (Rosenstock & Stecher, 1997). The Health Belief Model was developed in the 1950's by a group of social psychologists to explain widespread failure of people to participate in programs to prevent or detect disease (Rosenstock & Strecher, 1997). The Health Belief Model can be used to examine the relationship between people's beliefs and their health specific behaviors. The constructs of perceived threat, perceived benefits, perceived barriers and cues to action are all part of the Health Belief Model. This study was limited to examining only the perceived threat aspects of the Health Belief Model which could be directly related to HPV knowledge levels and preventive behaviors. Perceived threat included the concepts of perceived seriousness and perceived susceptibility.

Given the increased HPV risk and low HPV knowledge levels of college females, the purpose of this study was to assess college women's knowledge, perceived seriousness, perceived susceptibility and preventive behaviors regarding HPV infection and cervical cancer at a large Midwestern university. Further, the study was designed so that relationships could be examined between these variables.

METHODS

The participants in the study were full time, undergraduate and graduate college women between the ages of 18 and 30 studying on the main campus of a large Midwestern University during autumn quarter, 1999. Following Human Subjects Committee approval, 1,000 of the approximately 15,000 registered female students were randomly selected by the Registrar's computer and mailed survey packets. The list was generated three weeks into the quarter in order to include students who had registered late or changed address.

After a comprehensive review of the literature, the researchers developed a 40-item instrument to assess HPV and cervical cancer knowledge, perceptions and preventive behaviors. The knowledge portion of the questionnaire consisted of 15 multiple-choice items. Each question had one correct response. The total number of correct questions provided the knowledge score for this instrument. The possible range of scores was from 0 to 15.

Fifteen questions were included to assess perceived susceptibility and perceived seriousness of cervical cancer. These questions used a five point Likert-type scale ranging from strongly agree to strongly disagree. Each potential response was assigned a point value from one to five. Six of the 15 questions related to perceived seriousness and had a possible subtotal score range from 6 to 30. Nine questions related to perceived seriousness and had a potential score which ranges from 9 to 45.

The last ten questions focused on individual sexual behaviors and demographics. These questions were multiple choice and fill in the blank. They requested information concerning age, college level, marital status, sexual activity, contraceptive behavior, risk factors and past history of Pap tests. Three of the questions specifically addressed preventive behaviors including number of sexual partners, condom use, and yearly Pap tests.

Stability reliability of the instrument was established through test-retest procedures with ten college females over a two-week period. Test-retest correlation coefficients were .90 for knowledge, .95 for perceptions, and .90 for behaviors indicating a high level of reliability over rime.

Content validity of the instrument was determined by using consensual validity via a panel of experts. These members represented several health professionals including two gynecologists, two professors of health education, and a medical professional from the Breast and Cervical Cancer program. These individuals were selected based on their knowledge and experience regarding HPV and cervical cancer. The panel was instructed to take the survey and make comments, suggestions, additions or deletions directly on the research tool. They were also asked to examine and comment on the scoring system. Revisions were made based on the panel's input and the instrument was deemed ready for use. Following successful pilot testing and approval from the graduate committee and the University's Human Subjects committee the survey was administered 5 weeks into Fall Quarter.

Survey packets were mailed to 1,000 female students. The mailed packets included a cover letter, consent form, survey, and return self-addressed postage-paid envelope. The detailed cover letter explained the importance of the study, and thanked students for participating. It also assured students that their responses would remain confidential. Students who returned their completed questionnaires by the deadline were entered in a raffle to win $100. Reminder postcards urging participants to return the questionnaire were sent to all subjects two weeks after the initial mailing in a further effort to increase the return rate.

RESULTS

Four hundred and twenty-eight students (n = 428) returned completed surveys for a 42.8% response rate. Overall the mean age of participants was 21.5 years with an age range of 18-30. Almost 80% of the participants were traditional college students between the ages of 18 and 24. There was a fairly even distribution of subjects based on academic level with the largest number of subjects from the junior class (n = 99, 23.1%) and the smallest number from the freshman class (n = 59, 13.8 %). Ninety-one percent of the sample students were single (n = 391), while only 9% were married (n = 37).

An analysis of the behavioral questions on the survey provided a behavioral profile of the college women participants (See Table 1). The majority of the students were currently involved in a sexual relationship (n = 263, 61.3%), with 25.2% reporting two or more sexual partners in the past year. Condom use was not high among this group with 35.4% of participants rarely or never using condoms (n = 152) and only 15.6% reporting always using condoms (n = 67). Fifty percent of respondents used oral contraceptives, perhaps explaining the low usage rate of condoms. Twenty-two percent of participants smoked cigarettes, and 23% had never had a Pap test.

A knowledge score was calculated for each subject. The knowledge score was the total number of correct responses to the 15 multiple choice knowledge questions. The mean knowledge score was 6.8 correct responses with a standard deviation of 3.6.

Based on the Health Belief Model, fifteen questions were used to assess perceived threat of HPV/ cervical cancer. Perceived threat was comprised of two sub-scores; perceived susceptibility (six questions) and perceived seriousness (nine questions). All of these questions used a rive-point Likert Scale with higher scores indicating stronger perceptions of susceptibility and seriousness. The mean perceived susceptibility score was 17.4, (range of 0-30 points) with standard deviation of 2.4 points. The mean perceived seriousness was 24.8 points (range 0- 45 points) with standard deviation of 5.6 points.

Based on the descriptive analysis of the above data, this population of college students participated in high-risk behaviors for HPV infection, had fairly low HPV- knowledge levels, did not perceive HPV infection to be especially serious, and did not perceive themselves to be particularly susceptible to HPV infection.

Relationships between HPV/cervical cancer knowledge and perceived seriousness, perceived susceptibility and preventive behaviors were examined (See Table 2). h had been hypothesized that as knowledge increased, perceived seriousness and perceived susceptibility also would increase. Using Pearson Correlation coefficients, no significant relationship was found to exist between HPV/cervical cancer knowledge and perceived susceptibility to HPV/cervical cancer (r = .020, p = .680). A significant negative correlation was round to exist between HPV/cervical cancer knowledge and perceived seriousness of HPV/cervical cancer (r = -0.242, p = .000), indicating that as knowledge increased, perceived seriousness decreased.

No significant relationships were found between HPV/cervical cancer knowledge and either number of sexual partners (r = .0468, p = .389) or condom use (r = .0305, p =.575). Using an ANOVA, to determine whether the mean differences existed in knowledge scores between women having pap tests within the past year, or having a pap test but hOt within the past year, or never having a pap test. It was determined that HPV/cervical cancer knowledge levels did differ between the three categories of women seeking pap tests (F (175) = 10.0, p = .000). Post hoc analysis using the Tukey test demonstrated significant knowledge differences between those who received Pap tests in the past year; and both those who had had a Pap test but not in the past year (p = .002) and those who had never had a Pap test (p = .000). That is, those who had had a Pap test in the past year were significantly (p < .001) more knowledgeable concerning HPV/cervical cancer.

Perceived seriousness of HPV/cervical cancer was examined in relation to the three preventive behaviors. It had been hypothesized that as perceived seriousness increased, preventive behaviors would increase. Using Pearson Correlation coefficients, no significant associations were observed between HPV/cervical cancer seriousness and either number of sexual partners (r = .0996, p = .066) or condom use (r = -.0408, p = .452). Using an ANOVA, it was determined that perceived HPV/cervical cancer seriousness did differ between women who had had a Pap test in the past year, women who had had a Pap test but not in the past year, and women who had never had a Pap test (F(19) = 3.4, p = .034). Post hoc analysis using the Tukey test demonstrated significant perceived seriousness differences existed between respondents receiving Pap tests in the past year and those who had had a Pap test but not in the past year (p = .040). That is those who had had a Pap test in the past year perceived HPV/cervical cancer to be more serious than those who did not have Pap tests in the past year.

Perceived HPV/cervical cancer susceptibility was also examined in relation to the preventive behaviors. It had been hypothesized that as perceived susceptibility increased, preventive behaviors would increase. Using Pearson Correlation coefficients, a positive relationship was observed between perceived susceptibility and number of sexual partners (r = .2075, p = .000). Although the correlation was small, as numbers of sexual partners increased, perceived HPV/cervical cancer susceptibility also increased. No significant relationship ",vas observed, however, between HPV/cervical cancer susceptibility and condom use (r = -.0232, p = .669). Using an ANOVA test, it was determined that perceived HPV/cervical cancer seriousness differed significantly between women who had had a Pap test in the past year, women who had had a Pap test in the past, but not in the past year, and women who had never had a Pap test (F(19) = 3.4, p = .034). Post hoc analysis using the Tukey test demonstrated significant perceived susceptibility differences between those who had had a Pap test in the past year and those who had had a Pap test but not in the past year (p = .040). Those who had had a Pap test in the past year had higher perceived HPV/cervical cancer susceptibility than those who had a Pap test in the past, but not in the past year.

Next, as shown in Table 3, multiple linear regression was used to predict the combined effects of HPV/ cervical cancer knowledge, perceived seriousness and perceived susceptibility on number of sexual partners in the past year (R = .397, [R.sup.2] =.158 F = 2.337, p = .000). HPV/cervical cancer knowledge, perceived seriousness and perceived susceptibility accounted for 15.8% of the variance in total number of sexual partners.

Multiple linear regression also was used to determine the combined effects of HPV/cervical cancer knowledge, perceived seriousness and perceived susceptibility on condom use (R = .450, [R.sup.2] = .203, F = 3.161, p = .000). As shown in Table 3 HPV/cervical cancer knowledge, perceived seriousness, and perceived susceptibility accounted for 20.3% of the total variance regarding HPV/cervical cancer in condom use.

DISCUSSION

College women participating in this study demonstrated low levels of knowledge concerning HPV/ cervical cancer while exhibiting high-risk sexual behaviors. This is consistent with the work of Vail-Smith and White (1992) that surveyed 263 sexually active college women and concluded that these women lacked awareness of HPV, were at risk for contracting the infection, and were not practicing preventive behaviors to reduce their risk. Study results also are consistent with the work of Burak and Meyer (1998); documenting that college woman who had an STD were not inclined to seek a Pap test. They proposed that this might represent a lack of knowledge about the association between sexually transmitted HPV and cervical cancer.

These findings have important implications for high school health education programs, college level health education programs and student wellness center personnel. Accurate information concerning HPV and cervical cancer is either not being stressed in health education programs or it is not being internalized and accepted by students. In either case, health educators must seek new opportunities to present this important information in creative and innovative ways to young people. STD prevention education must actively encourage behaviors that prevent HPV. Health educators working with the campus and community can develop such initiatives. Perhaps an HPV awareness week could be initiated on campus or within an entire community. Articles could be written in student papers, and posters displayed throughout campus. Perhaps by working with the local health department a speaker could be found to present a case study of HPV/ cervical cancer. Another idea might be to offer reduced rates for PAP smears for a given time period and connect the promotion for this offer with HPV information. HPV should be made the focus of these activities and not simply tie it in with all sexually transmitted diseases.

The present study also found that there was no relationship between HPV/cervical cancer knowledge and perceived susceptibility, while a negative correlation was round to exist between HPV/cervical cancer knowledge and perceived seriousness of HPV/cervical cancer. In other words, as knowledge increased, perceived seriousness decreased. This counter initiative finding may suggest that as women became more knowledgeable about HPV, they learned how they could prevent the disease and thus viewed the potential consequences to be less serious.

HPV/cervical cancer knowledge among this sample of college women was not related to either condom use or number of sexual partners. It was, however, related to obtaining PAP smears. Women reporting higher HPV/cervical cancer knowledge levels were more likely to obtain PAP smears. This reinforces the importance of providing good HPV/cervical cancer information to students.

Perceived seriousness of HPV/cervical cancer and perceived susceptibility to HPV/cervical cancer were examined in relation to the preventive behaviors of condom use and number of sexual partners. No significant relationships were round related to condom use. Perceived susceptibility to HPV/cervical cancer was found to be related to the number of sexual partners. As number of sexual partners increased, perceived susceptibility to HPV/cervical cancer increased. This is a positive finding indicating that at least some of the college women at greater risk due to their sexual behaviors were cognizant of increased HPV/cervical cancer risk. The challenge for health educators is then to translate this knowledge into the practice of low-risk behavior.

Both perceived seriousness of HPV/cervical cancer and perceived susceptibility to HPV/cervical cancer were related to Pap test compliance. Those perceiving HPV/cervical cancer to be more serious were more likely to have received a Pap test in the past year. Those who felt more susceptibility to HPV/cervical cancer were also more likely to have received a Pap test in the past year. HPV/cervical cancer information stressing the seriousness of these conditions along with individual susceptibility may increase Pap smear compliance.

A significant amount of variance in condom use and number of sexual partners can be explained by the combined effects of HPV/cervical cancer knowledge, perceived seriousness and perceived susceptibility:. Almost 16% of the variance in sexual partners and 20% of the variance in condom use can be explained by these three factors. This strongly supports health education interventions designed to impact HPV/cervical cancer knowledge, perceived seriousness and perceived susceptibility.

The results of this study need to be interpreted with caution. First, this study only involved women from one large Midwestern University. Replication studies need to be conducted at universities of different sizes and in different geographic areas. Second, the response rate for the current study was 42.8 percent. While adequate for a randomized mail survey; a higher response rate would increase confidence in the results. Future attempts to study HPV/cervical cancer should try different follow-up and incentive techniques to increase response rates, or perhaps try a telephone interview technique.

Another factor to consider when interpreting these results is that this population exhibited low levels of knowledge, perceived seriousness and perceived susceptibility. Results of this study did not demonstrate clear relationships among college women's knowledge, perceptions, and preventive behaviors. The absence of relationships between these variables and the behavioral risk factors may be due, in part, to the low number of individuals with higher levels of knowledge, perceived seriousness and perceived susceptibility. Without a good distribution of knowledge scores, this study was essentially examining small differences in knowledge among essentially uninformed subjects.

Further HPV/cervical cancer research needs to be conducted among women of the same age group that are not part of a college or university community. With fewer education opportunities related to HPV/cervical cancer, these women may have even less knowledge and understanding of HPV/cervical cancer. It may be possible to access these women through worksite settings, beauty parlors, or faith communities.

In conclusion, the results of dais study and of prior studies on college women have provided similar findings. The lack of knowledge linking HPV infection with cervical cancer, combined with the high level of risk apparent among young women, indicates that high school and college health educators needed to intensify their efforts in HPV education, and must actively encourage behaviors that prevent HPV. Education curricula dealing with STDs should be examined and revised to include appropriate information and emphasis on HPV infection. Incidence statistics, risk factors and symptoms of HPV should be provided from health educators as well as information on how to prevent HPV infection. Health educators have the potential to impact this serious, potentially fatal, disease chain.

Table 1. Health Related Behaviors (N = 428)

Characteristics                                N     %

Sexual experience
  Current sexual relationship                 263   61.3
  No current sexual relationship               86   20.0
  Never had a sexual relationship              78   18.2

Number of sexual partners/past year
  0                                            94   21.9
  1                                           226   52.7
  2 or more                                   108   25.2

Condom Use
  Always (100%)                                67   15.6
  Usually (76-99%)                             79   18.4
  Sometimes (51-75%)                           29    6.8
  Occasionally (26-50%)                        20    4.7
  Rarely (1-25%)                               58   13.5
  Never (0%)                                   94   21.9

Oral Contraceptive Use
  Yes                                         215   50.1
  No                                          208   48.5

Smoke Cigarettes
  Yes                                          94   21.9
  No                                          334   77.9

Last Pap test
  Never                                        99   23.1
  Within Past Year                            286   66.7
  Have had a Pap test, but not in past year    43   10.0

Personal or Family Diagnosis
  Yes                                          67   15.6
  No                                          361   84.1

Table 2. Correlation's

Pearson Correlation        Knowledge   Susceptibility   Severity

Knowledge
  Pearson Correlation        1.000           .020         -.239 *
  Significance                .              .680          .000
  N                        424            420           422

Perceived Susceptibility
  Pearson Correlation         .020          1.000         -.014
  Significance                .680           .             .774
  N                        420            423           421

Perceived Severity
  Pearson Correlation        -.239 *        -.014         1.000
  Significance                .000           .774          .
  N                        422            421           425

* Correlation is significant at the.01 level (2-tailed)

Table 3. Partial Correlation's related to Preventive Behavior.

                            Sexual                  Yearly
Partial Correlation        Partners   Condom Use   Pap test

Knowledge
  Partial Correlation       .0468        .0305      .0095
  Significance              .389         .575       .861

Perceived Susceptibility
  Partial Correlation       .2075        .0232      .0052
  Significance              .000         .669       .924

Perceived Severity
  Partial Correlation       .0996       -.0408      .0473
  Significance              .066         .452       .384

Total
  Multiple Linear           .397         .450       .388
  Significance              .000         .000       .000

REFERENCES

Alvey, J. (1995). Genital warts and contagious cancers. Jefferson, NC: McFarland and Company, Inc. Publishers.

American Cancer Society. (1999). Cervical cancer- Overview. [On-Line]. Available: http://www3.cancer.org/cancerinfo/main_cont.asp?st=wi&ct=8

The American College of Obstetricians and Gynecologists. (1995). Human Papillomavirus (HPV) infection. [Brochure]. Washington, D.C.: ACOG Patient Education.

The American College of Obstetricians and Gynecologists. (1998). ACOG Clinical Review, 3, 1-15.

Burak, L.J. & Meyer, M. (1998). Factors influencing college women's gynecological screening behaviors and intentions. Journal of Health Education, 29(6), 365-370.

Center for Disease Control-Division of Sexually Transmitted Diseases Prevention. (1999). The challenge of STD prevention in the U.S. [On-line]. Available http://www.cdc.gov/nchstp/dstd/STD_Prevention_in_the United_States.htm.

National Institute of Allergy and Infectious Disease. (1999). Human Papillomavirus and genital warts, NIAID fact sheet. [On-line]. Available: http://www.niaid.nih.giv/factsheets/stdhpv.htm.

National Institutes of Health. (1996). Cervical cancer consensus statement. April 1-3;14(1); 1-38.

Namkoong, S.E. (1995). Clinical application of HPV typing in cervical cancer. International Journal of Gynecology & Obstetrics, 49, S59-S67.

Rosenstock, I.M., Strecher, V.J. (1997). The Health Belief Model. In K. Glanz, F.M. Lewis, & B.K. Rimer (Eds.), Health behavior and health education; Theory, research and practice (pp. 41 - 59 ). San Fransico, CA: Jossey-Bass, Inc.

Vail-Smith, K. & White, D.M. (1992). Risk level, knowledge, and preventive behavior of Human Papillomavirus among sexually active college women. Journal of American College Health, 40, 227-230.

HEALTH EDUCATION RESPONSIBILITY AND COMPETENCY ADDRESSED

Responsibility I-Assessing Individual and Community Needs for Health Education

Competency B-Distinguish between behaviors that foster and those that hinder well-being

Sub-competency 1-Investigate physical, social, emotional, and intellectual factors influencing health behavior

Kimberly Ingledue, M.Ed. is a Health Promotion Coordinator at Florida Atlantic University. Randall Cottrell, D.Ed. is a Professor of Health Promotion and Education and Amy Bernard, Ph.D. is an Associate Professor of Health Promotion and Education at The University of Cincinnati. Address all correspondence to Kimberly Ingledue, M.Ed., 140 SW 117th Ave, Apartment #303, Pembroke Pines, FL 33025; PHONE: [phone]; E-MAIL: [e-mail].