Mays, Rose M. PhD, RN*; Zimet, Gregory D. PhD†
Efforts are underway to develop vaccines for a number of sexually transmitted infections (STI), including N. gonorrhea, C trachomatis, herpes simplex virus Type 2 (the principal cause of genital herpes), human papillomavirus (HPV), and human immunodeficiency virus (HIV).1–4 The health and economic costs of these infections are substantial, suggesting that successful STI immunization programs are likely to prove cost-effective.5 However, the success of these programs will depend on adequate rates of acceptance among those most at risk. Experiences with established vaccines, including hepatitis B, influenza, and varicella, suggest that the availability of vaccination does not ensure widespread uptake.6–8 Therefore, as progress is made toward successful clinical trials and licensing, the likely extent to which STI immunization will be accepted and the potential determinants of acceptance need to be evaluated.9
Ideally, an effective STI immunization program should include young adolescents, preferably before they become sexually active.5,9 STI immunization strategies targeted toward early adolescents and preadolescents will place most of the decision-making burden on parents and healthcare providers. A study of hepatitis B vaccination of adolescents found that the best predictor of parental acceptance was the parent’s belief that vaccination was regarded as important by the provider.10 This finding suggests that the success of these immunization programs will depend to a large extent on health care providers recognizing the importance of STI immunization for adolescents and preadolescents. In addition, health care providers must be willing and able to discuss STI vaccination with patients and parents of patients. However, research suggests that primary health care providers are often reluctant to discuss sexuality-related issues with adolescents and their parents.11–14
Although several studies have examined determinants of acceptability of hypothetical STI vaccines, including individuals’ health beliefs15–19 and attitudes about vaccine characteristics,20–23 we have found no published research on the views of health care providers. Given that there are increasing numbers of nurse practitioners involved in direct primary care24 and that nursing interventions have demonstrated effectiveness in improving vaccine rates,25–26 it is important to examine the attitudes of these health care providers. The objective of this study was to evaluate how the age and gender of the adolescent patient, type of infection prevented, and endorsement by a professional organization would influence the willingness of nurse practitioners to recommend vaccination to parents of adolescents. Our hypotheses were as follows: 1) nurse practitioners would be more willing to recommend vaccination for older adolescents because prior research suggests that health care providers are more likely to address sexuality issues with older adolescents;14 2) there would be greater willingness to recommend STI vaccination for adolescent females because of the disproportionate burden of STI on young women compared to young men;27 3) there would be greater willingness to recommend non-STI vaccines compared to STI vaccines because of the documented reluctance of health care providers to discuss sexuality with their adolescent patients;11–14 and 4) endorsement by a professional organization would enhance willingness to recommend vaccination, because such an endorsement would create the sense that recommending immunization is a normative behavior. Prior research suggests that the practices of health care providers are influenced by perceived social norms.28
Materials and Methods
Sample and Study Design
Questionnaire data were collected from convenience samples of nurses attending professional conferences in 2 Midwestern states. Participants were told that the questionnaire concerned their attitudes about vaccination for adolescents. The data were collected via anonymous, self-administered written questionnaires that took 5–10 minutes to complete. Questionnaires were filled out in the meeting. The study was approved by the university’s Institutional Review Board. Given the low-risk nature of the study, the requirement for written informed consent was waived; however, confidentiality was maintained at all times.
Independent variable included gender, age, and years in practice. Additionally, each participant was asked to indicate what proportion of time in her/his practice was spent working with adolescents. Response options were: 0%, 1% to 25%, 26% to 50%, 51% to 75%, and 76% to 100%. Given the unequal distribution of responses across these categories, work with adolescents was dichotomized as 0% to 25% and 26% to 100%. Participants were also asked to indicate whether their work involved principally primary care or specialty care.
To assess vaccine acceptability, the dependent variable, participants were presented with 13 items. Each item described a hypothetical vaccine scenario uniquely defined along the following 4 dimensions: infection targeted (HIV, genital herpes, or mononucleosis); patient age (11, 14, or 17 years of age); patient gender (male or female), and endorsement of vaccine by the American Academy of Pediatrics (AAP; yes or no). For example, the first item read as follows: “Scenario 1: Your patient is an 11-year-old boy. The AAP has not specifically recommended that all adolescents get a genital herpes vaccine.” See Table 1 for descriptions of the 13 vaccine scenarios. Participants were asked to rate their willingness to recommend to a parent that his or her child receive each vaccine. An 11-point response format was used, ranging from 0 (“I would never recommend this vaccine to the parent”) to 100 (“I would definitely recommend this vaccine to the parent”) in intervals of 10. Vaccine acceptability in general was evaluated by creating a scale score based on the mean value across all 13 items (coefficient alpha = 0.95).
The associations of provider age and number of years in practice to the 13-item vaccine acceptability measure were evaluated with Pearson product-moment correlations. The relationship of percent of adolescents in the practice (0%–25%; 26%–100%) to vaccine acceptability was assessed via t-test.
Full-profile, ratings-based conjoint analysis was used to evaluate the effects of the vaccine scenario dimensions on willingness to recommend vaccination. This analytic technique, which is based on regression analysis, is often used in marketing research to evaluate how a product’s characteristics influence the product’s acceptability.29 Conjoint analysis is a method specifically designed to account for multiple observations from each participant. The term “full-profile” refers to the fact that each scenario is described along all dimensions. The term “ratings-based” refers to the fact that participants are asked to respond to each scenario on a rating-scale, rather than using a ranking approach. As noted above, in this study there were 4 dimensions evaluated (infection targeted, child age, child gender, and AAP endorsement of vaccine), each with either 2 or 3 attribute levels (e.g., infection could be defined as genital herpes, HIV, or mononucleosis). A full factorial design involving every possible combination of characteristics would have required subjects to rate 36 separate vaccine scenarios, which would have been too repetitious and time-intensive. Therefore, a fractional factorial design with a representative subset of 9 vaccines was developed before questionnaire development through the use of the SPSS Conjoint procedure.30 Using a fractional factorial design made the ratings task more feasible, but limited the analysis to the main effects of each dimension. The conjoint procedure was also used to define 4 additional vaccine scenarios as hold-out items that could be used to assess model fit.
The analysis of the 9 vaccine items resulted in the development of a conjoint model, which provided information on each Subject’s preference profile (part-worth utilities) about attributes within each dimension (e.g., preference for a vaccine endorsed by the AAP over a vaccine not endorsed by the AAP). The relative ranges of the part-worth utilities across the 4 dimensions were used to calculate importance scores, which reflect the extent to which each dimension contributes to the vaccine ratings. The sum of importance scores across dimensions always equals 100. The assessment of model fit was accomplished by computation of a Kendall’s tau correlation between the actual rank-order of the 4 hold-out items and the rank-order predicted by the conjoint model.
Of 229 nurses completing the survey, 5 (2%) failed to respond to one or more of the 13 vaccine scenarios, leaving a useable sample size of 224. Respondents were 96% female and ranged from 25 to 68 years of age (Mean = 44.3, standard deviation (SD) = 9.1). As a group they had from less than 1 year to 46 years of experience working as a nurse practitioner (Mean = 18.5, SD = 10.6). Eighty-seven percent indicated that they principally worked in primary care settings. Forty-nine percent of the nurses reported that they spent over 25% of their practice time working with adolescents and 51% reported that they spent 25% of their time or less.
Across the 13 scenarios, overall willingness to recommend vaccination ranged from 0 to 100 (Mean = 72.0, SD = 19.8). The least acceptable vaccine scenario (“Your patient is an 11 year old boy. The AAP has not specifically recommended that all adolescents get a genital herpes vaccine”) received a mean score of 52.3 (SD = 27.3), whereas the most acceptable vaccine scenario (“Your patient is a 17 year old boy. The AAP has specifically recommended that all adolescents get an HIV vaccine”) received a mean score of 88.8 (SD = 20.1). Table 1 lists the mean scores for all 13 vaccine scenarios described.
Respondent age and years in practice were unrelated to overall willingness to recommend vaccination. However, practitioners who spent more than 25% of their time working with adolescents indicated greater willingness to recommend vaccination (mean = 75.4) than did those who spent 25% or less of their time (mean = 68.7; t = 2.5, P < 0.02).
Twenty-six participants gave all 9 of the key conjoint vaccine scenarios the same score. Three of these 26 respondents gave ratings of zero across all vaccines, 7 gave ratings of 50, 1 gave ratings of 90, and 15 gave ratings of 100. Given that this subgroup of participants demonstrated no variability in preferences across the vaccine scenarios, they were necessarily excluded from the conjoint analysis. The overall conjoint model for the remaining 198 respondents, developed from the 9 vaccine scenarios, demonstrated an excellent fit, with a Kendall’s Tau of 0.94 for predicting the rank-order of the 4 holdout items. A Kendall’s Tau was also calculated for each subject to assess how well each Subject’s conjoint model predicted the rank-order of the hold-out items. The median Kendall’s Tau across all subjects was 0.71, again indicating an acceptable fit for the conjoint model.
The conjoint analysis resulted in the following findings. Hypothesis 1 was supported, in that there was a preference for recommending vaccination for 17-year-olds and a relative reluctance to recommend vaccination for 11-year-olds (part-worth utilities = 4.6 and −5.7, respectively). Hypothesis 2 was not supported, in that there was little difference in preference for recommending vaccination for male compared to female adolescents (part-worth utilities = 0.6 and −.6, respectively). Hypothesis 3 was not supported, in that there was a preference for HIV vaccination and a relative disinclination to recommend mononucleosis vaccination (part-worth utilities = 2.6 and −2.4, respectively). Finally, hypothesis 4 was supported, in that there was a strong preference for AAP-endorsed vaccines over those not endorsed by the AAP (part-worth utilities = 9.8 and −9.8, respectively). Figure 1 shows the part-worth utility values across the dimensions. Translating the part-worth utilities into importance scores indicated that recommendation by the AAP was most influential in vaccine ratings (importance score = 54.2), followed by age of patient (importance score = 28.4) and infection targeted by vaccination (importance score = 13.8). As might be expected by the negligible range in part-worth utilities, gender of the adolescent patient had little influence on vaccine ratings (importance score = 3.6).
AAP recommendation had the highest importance score for 87 participants, with all of them favoring an AAP-endorsed vaccine. Age of the adolescent patient had the highest importance score for 46 participants, with 78% of these individuals indicating a linear increase in acceptability from 11 to 17 years of age. Infection targeted had the highest importance score for 61 respondents, with 58% favoring an HIV vaccine, 32% favoring a genital herpes vaccine, and 10% favoring a mononucleosis vaccine. Gender of the adolescent patient had the highest importance score for only 2 of the participants, with one indicating a strong preference for vaccinating males and the other for vaccinating females.
The purpose of this study was to evaluate factors related to the willingness of nurse practitioners to recommend STI vaccination to parents of adolescent patients. We were particularly interested in how patient age and gender, type of infection prevented, and endorsement by a professional organization might influence the inclination of nurse practitioners to recommend vaccination.
We found support for 2 of our 4 hypotheses. As predicted, endorsement by the AAP strongly influenced acceptability of vaccination. This result suggests that endorsements by organizations such as the AAP, the American Academy of Family Physicians (AAFP), and the Advisory Committee on Immunization Practices (ACIP) will be important for successful implementation of new immunization programs, including those involving STI vaccines. This finding is consistent with prior research indicating that perceived social norms, such as recommendations by professional organizations, significantly influenced physicians’ intentions to educate adolescent patients about sexually transmitted diseases.28 In addition, our findings are consistent with research reports indicating that the recommendations for universal hepatitis B immunization of infants put forth in 1991 and 1992 by the ACIP, AAP, and AAFP had an effect over time on physicians’ immunization practices.31–33
We also found that nurse practitioners’ willingness to recommend vaccination increased with increasing age of the patient. This finding held for both the STI vaccines (i.e., genital herpes and HIV) and the non-STI vaccine (i.e., mononucleosis). The relative reluctance on the part of the study participants to recommend vaccines to parents of 11-year-olds is an issue of some importance, particularly with STI immunization. Recent data from the nationally administered Youth Risk Behavior Survey indicated that 34.4% of in-school 9th graders had already engaged in sexual intercourse.34 In addition, a study of chlamydia prevalence among urban adolescents receiving clinical services found that 14-year-olds had a disease prevalence of 27.5%.35 It is clear from these data that in order to have the most protective population-wide effect, STI immunization should be given to adolescents before initiation of sexual activity, ideally no later than 11 to 12 years of age.
We expected a disinclination to endorse STI vaccination due to the potential for discomfort around discussing sexuality issues with parents of adolescent patients. However, our findings indicated the opposite pattern, with an HIV vaccine most recommended, followed by genital herpes, with mononucleosis coming in last. It is possible that the nurse practitioners were responding to the relative severity of each infection and attending less to the modes of transmission. In future research it will be important to control for perceived severity of infection in order to ascertain the potential effect of sexual transmissibility on providers’ willingness to recommend vaccination.
We expected the nurse practitioners to more strongly recommend STI immunization to parents of young women compared to parents of young men, due to the greater disease burden experienced by women. However, our results indicated no gender-based preference for vaccination, perhaps reflecting the respondents’ recognition that vaccination of young men may prove to be an important strategy for the protection of young women.
Those nurse practitioners who reported spending more time with adolescents in their clinical practices were more willing to recommend immunization. This finding may indicate that health care providers who work more intensively with adolescents have a better understanding of adolescents’ health needs, as well as more awareness about sexuality and adolescent development. As immunization programs targeted at young adolescents are developed, it will be important to ensure that providers are equipped to address the unique health care needs of this group. This issue will be particularly relevant to the many health care providers who see relatively few adolescents.
This study has several limitations, suggesting that the results should be interpreted with some caution. First of all, the sample of nurse practitioners was a non-random, convenience sample, indicating that the results cannot necessarily be generalized to all nurse practitioners. A second limitation is that individuals’ responses to hypothetical scenarios will not perfectly predict their actual behaviors. This problem is inherent in any study of acceptability of products not yet available. However, marketing research studies and research based on the Theory of Reasoned Action suggest that intention to engage in a particular behavior is a reasonable, if imperfect, predictor of carrying out the behavior.36–37
This study focused on particular sets of vaccine, provider, and adolescent patient characteristics. In the future it will be important to explore other factors, such as providers’ beliefs about immunization in general, their attitudes about adolescent sexuality, and their ability to communicate with parents about sensitive topics. In addition, it will clearly be important to examine vaccine acceptability among other providers, including physicians and physician assistants who provide primary health care to adolescents.
In summary, endorsement of vaccination by a professional organization, older age of the adolescent patient, and more clinical involvement with adolescents were associated with greater willingness to recommend vaccination to parents of adolescents. Sexual transmissibility of the targeted infection and gender of the patient were unrelated to nurse practitioners’ intentions to endorse immunization. Of concern, however, was the relative reluctance of these providers to recommend immunization for younger adolescents, the age group most likely to benefit from STI vaccines. Before these vaccines are made available, it will be important to ensure that health-care providers are comfortable with, and skilled at, discussing the benefits of STI vaccines with parents and their young adolescent children.
1.Koutsky LA, Ault KA, Wheeler, et al. A controlled trial of a human papillomavirus type 16 vaccine. N Engl J Med 2002; 347:1645–1651.
2.Stanberry LR, Bernstein DI. Sex Transm Dis. Vaccines, Prevention and Control
. San Diego: Academic Press; 2000.
3.Stanberry LR, Spruance SL, Cunningham AL, et al. Glycoprotein-D-Adjuvant vaccine to prevent genital herpes. N Engl J Med 2002; 347:1652–1661.
4.Whitley RJ, Roizman B. Herpes simplex viruses: Is a vaccine tenable? J Clin Invest 2002; 110:145–151.
5.Institute of Medicine. Vaccines for the 21st Century
. Washington, DC: National Academy Press; 2000.
6.Barie PS, Dellinger EP, Dougherty SH, et al. Assessment of hepatitis B virus immunization status among North Am surgeons. Arch Surg 1994; 129:27–32.
7.Ehresmann KR, Mills WA, Loewenson PR, et al. Attitudes and practices regarding varicella vaccination among physicians in Minnesota: Implications for public health and provider education. Am J Public Health 2000; 90:1917–1920.
8.Fedson DS. Adult immunization: Summary of the National Vaccine Advisory Committee report. JAMA 1994; 272:1133–1137.
9.Zimet G, Mays RM, Fortenberry JD. Vaccines against sexually transmitted infections: Promise and problems of the magic bullets for prevention and control. Sex Transm Dis 2000; 27:49–52.
10.Rosenthal SL, Lewis LM, Succop PA, et al. College students’ attitudes regarding vaccination to prevent genital herpes. Sex Transm Dis, 1999; 26:438–443.
11.Maheux B, Haley N, Rivard M, et al. STD risk assessment and risk-reduction counseling by recently trained family physicians. Acad Med 1995; 70:726–728.
12.Millstein SG, Igra V, Gans J. Delivery of STD/HIV preventive services to adolescents by primary care physicians. J Adolesc Health 1996; 19:249–257.
13.Schuster MA, Bell RM, Petersen LP, et al. (1996) Communication between adolescents and physicians about sexual behavior and risk prevention. Arch Pediatr & Adolesc Med 1996; 150:906–913.
14.Torkko KC, Gershman K, Crane LA, et al. Testing for chlamydia and sexual history taking in adolescent females: Results from a statewide survey of Colorado primary care providers. Pediatrics 2000; 106:e32.
15.Gagnon M-P, Godin G. Young adults and HIV vaccine: Determinants of the intention of getting immunized. Can J Public Health 2000; 91:432–434.
16.Lazcano-Ponce E, Rivera L, Arillo-Santillan E, et al. Acceptability of a human papillomavirus (HPV) trial vaccine among mothers of adolescents in Cuernavaca, Mexico. Arch Med Res 2001; 32:243–247.
17.Liau A, Zimet GD. Undergraduates’ perception of HIV immunisation. Attitudes and behaviors as determining factors. Int J STD AIDS 2000; 11:445–450.
18.Rosenthal SL, Kottenhahn RK, Biro FM, et al. Hepatitis B vaccine acceptance among adolescents and their parents. J Adolesc Health 1995; 17:248–254.
19.Zimet GD, Mays RM, Winston Y, et al. Acceptability of HPV immunization. J Womens Health Gend Based Med 2000; 9:47–50.
20.Lewis, LM, Stanberry LR, Rosenthal SL, et al. Attitudes regarding vaccinations of STDs and other diseases. Internat J STD AIDS 2000; 11:170–172.
21.Liau A, Zimet GD. The acceptability of HIV immunization: Examining vaccine characteristics as determining factors. AIDS Care 2001; 13:643–650.
22.Zimet GD, Fortenberry JD, Blythe MJ. Adolescents’ attitudes about HIV immunization. J Pediatr Psychol 1999; 24:67–75.
23.Zimet GD, Blythe MJ, Fortenberry JD. Vaccine characteristics and acceptability of HIV immunization among adolescents. Internat J STD AIDS 2000; 11,143–149.
24.Cooper RA, Laud P, Dietrich CL. Current and projected workforce of nonphysician clinicians. JAMA 1998; 280:788–794.
25.Christy C, McConnochie KM, Zernik N, et al. Impact of an algorithm-guided nurse intervention on the use of immunization opportunities. Arch Pediatr & Adolesc Med 1997; 151:384–391.
26.Rhew DC, Glassman PA, Goetz MB. Improving pneumococcal vaccine rates: Nurse protocols versus clinical reminders. J Gen Intern Med 1999; 14:351–356.
27.Eng TR, Butler WT. The Hidden Epidemic: Confronting Sexually Transmitted Diseases
. Washington, DC: National Academy Press; 1997.
28.Millstein SG. Utility of the theories of reasoned action and planned behavior for predicting physician behavior: A prospective analysis. Health Psychol 1996; 15:398–402.
29.Churchill GA. Marketing Research: Methodological Foundations
. Chicago: Dryden Press; 1991.
30.SPSS Inc. SPSS Conjoint 8.0
. Chicago: SPSS; 1997.
31.Freed GL, Bordley WC, Clark SJ, et al. Universal hepatitis B immunization of infants: Reactions of pediatricians and family physicians over time. Pediatrics 1994; 93:747–751.
32.Kim SC, Sinai LN, Casey R, et al. Universal hepatitis B immunization. Pediatrics 1995; 95:764–766.
33.Kraus DM, Campbell MM, Marcinak JF. Evaluation of universal hepatitis B immunization practices of Illinois pediatricians. Arch Pediatr & Adolesc Med 1994; 148:936–942.
34.Centers for Disease Control, Prevention. Trends in sexual risk behaviors among high school students—United States, 1991–2001. MMWR 2002; 51:856–859.
35.Burstein GR, Gaydos CA, Diener-West M, et al. Incident chlamydia trachomatis infections among inner-city adolescent females. JAMA 1998; 280:521–526.
36.Jamieson L, Bass F. Adjusting stated purchase intentions measures to predict trial purchase of new products. J Mark Res 1989; 26:336–345.
37.Terry D, Gallois C, McCamish M. The theory of reasoned action and health care behaviour. In: Terry DJ, Gallois C, and McCamish M, eds. The Theory of Reasoned Action: Its Application to AIDS-preventive Behaviour.
Oxford: Pergamon; 1993:1–27.