Adolescent Immunization Practices:  A National Survey of US Physicians FREE

DURING THE past decade, a series of initiatives have greatly increased immunization rates among infants and young children.¹ In contrast, until recently, little attention has been paid to immunizing adolescents. Therefore, it is not surprising that adolescent immunization rates remain quite low.² However, in 1996, the Advisory Committee on Immunization Practices (ACIP), the American Academy of Pediatrics, the American Academy of Family Physicians, and the American Medical Association (AMA) jointly proposed a new strategy to both improve the delivery of vaccination services to adolescents and to integrate recommendations for vaccination with other preventive services provided to adolescents.² This strategy emphasizes the importance of a routine office visit at 11 to 12 years of age, at which time it is recommended that adolescents receive tetanus and diphtheria toxoid and other vaccines (hepatitis B, varicella, and a second dose of measles-mumps-rubella) to prevent conditions to which adolescents still may be susceptible.

Little is known about how physician practices affect adolescent immunization. It is likely, however, that some of the same practice-related factors that influence the immunization of infants and young children also affect the provision of immunizations to adolescents. With that in mind, this survey was designed to describe physicians' adolescent immunization practices, and to determine which practice-related barriers influence adolescent immunization. Since pediatricians and family physicians together provide primary care to the majority of adolescents,³ we evaluated adolescent immunization practices of physicians from both specialties.

In September 1997, a 24-item forced-choice questionnaire, which previously had been piloted locally in Rochester, NY, and was approved by the Research Subjects Review Board of the University of Rochester, was sent to 660 general pediatricians and 820 family physicians located throughout the United States. Surveyed physicians were randomly selected from the AMA's Masterfile. More family physicians were deliberately surveyed because prior national surveys related to immunization practices reported lower response rates among family physicians than among pediatricians.^4- 7 The questionnaire included demographic items, as well as questions about physicians' adolescent immunization practices and policies, beliefs regarding school-based immunizations and perceived barriers to immunization. The AMA's Masterfile was used because it is the most extensive listing of US physicians; it includes demographic information on AMA members as well as on physicians who are not members of the AMA.

The sole inclusion criterion was that physicians reported that they routinely saw 4 or more adolescents (aged ≥11 years) per month for health supervision visits. Subjects who met this criterion were asked to complete and return the survey. Survey recipients seeing fewer adolescents were asked to indicate this on the form and return the uncompleted survey. Nonrespondents were sent additional surveys, and if they did not return them, they were contacted by telephone, at which time they were encouraged to complete the survey. Up to 5 mailings were sent to each physician.

Data from completed surveys were analyzed using descriptive statistics and χ² analysis. Independent associations were determined by a multivariate logistic regression model using all of the following independent categorical variables: physician specialty, sex, geographic region, practice setting, and years (ie, decades) since medical school graduation. In the regression analyses, the effect for each category of every independent variable was compared with the overall combined effect of the other categories of the variable. SPSS (Statistical Product and Service Solutions 6.0.1; SPSS Inc, Chicago, Ill) statistical software was used for all analyses.

Of the 1480 physicians surveyed, 1110 (75%) returned the questionnaire, including 536 pediatricians (81% response rate) and 582 family physicians (71% response rate). Of these respondents, 761 (69%) indicated that they met the inclusion criterion of seeing 4 or more adolescents per month for primary care. Among this final study sample of 761 physicians, 406 (53%) were pediatricians, and 355 (47%) were family physicians. Demographic information about these physicians is included in Table 1. All of the following results refer to these 761 physicians.

One hundred fifty-three physicians (20%) indicated that the state in which they practiced required insurance companies to fully cover the cost of all preventive care visits, while 208 (27%) indicated that the state required insurance companies to fully cover the cost of immunizations. Fifty-one percent (390 physicians) reported that they participated in the Vaccines for Children (VFC) program, which is designed to provide free immunizations for children from low-income households.

While 95% of the physicians indicated that they routinely check the immunization status of their adolescent patients at health maintenance visits, only 43% reported routinely doing so at illness-related visits, and only 47% reported doing so at follow-up visits. Similarly, while 94% of the physicians indicated that they immunized eligible adolescents at health maintenance visits, only 23% reported ever doing so at illness-related visits, and only 59% reported ever doing so at follow-up visits.

Five hundred seventy-eight physicians (76%) indicated that their practice had an established policy or protocol for the immunization of adolescents at specific ages, while 157 (21%) did not have such a policy or protocol, and 26 (3%) either did not know or did not complete that item on the questionnaire. Only 149 physicians (21%) indicated that their practice had a system to track and recall adolescents who were behind on immunizations. Of these physicians, 88 (59%) indicated that their practice used systematic chart reviews, 39 (26%) used a practice-based computerized tracking system, 36 (24%) used "tickler" files, and 23 (15%) used a county or regional immunization registry. Twenty percent of physicians who reported that their practice had an immunization tracking and recall system noted that the system was not regularly used to recall adolescents who were found to be underimmunized.

The percentages of physicians who reported vaccinating eligible adolescents with specific vaccines ranged from a high of 97% for tetanus toxoid and 96% for a second dose of measles-mumps-rubella vaccine, to 84% for hepatitis B vaccine, and just 61% for varicella vaccine. The primary reason cited by physicians who did not routinely immunize eligible adolescents was that their adolescent patients did not have insurance coverage for immunization with specific vaccines. Most of those physicians also did not participate in the VFC program.

Based on self-reports, as determined by both bivariate and multivariate analyses, pediatricians were significantly more likely than family physicians to practice in accordance with ACIP recommendations (Table 2 and Table 3). In particular, pediatricians were more likely than family physicians to offer adolescents ACIP-recommended immunizations. Immunization practices also varied by geographic region (Table 3). Thus, physicians from the northeastern and western United States were significantly less likely to report that they referred adolescents out of their practice for immunizations than physicians from the South and the Midwest. Physicians from the Northeast were also consistently the most likely to report that they offered each ACIP-recommended immunization.

Twenty-seven percent of physicians reported referring some or all of their adolescent patients outside their practice for immunizations, while 31% referred young children outside their practice for immunizations. Ninety-nine percent of the physicians who acknowledged that they referred adolescent patients for immunizations referred to public health clinics; in addition, 27% sometimes referred patients to school-based clinics and 24% sometimes referred them to community health centers. The primary reasons cited for referring adolescents outside the practice for immunizations were: (1) the cost to patients for immunizations (69%); (2) the high purchase price of vaccines for physicians (34%); (3) patient or parent request for a referral outside the practice (32%); (4) insufficient reimbursement from private insurance plans for immunization (28%); and (5) insufficient Medicaid reimbursement for immunization (25%).

Physicians were asked several questions about school-based immunization programs. Thirty-five percent indicated that school-based immunization programs were available to at least some of their patients, while 40% indicated that, to their knowledge, such programs were not available to any of their patients; 25% were unsure. While 84% of physicians preferred that adolescent immunizations be administered at their practices, most also found public health clinics (83%), schools (71%), and teen clinics (63%) to be acceptable alternative immunization sites.

The survey addressed specific physician attitudes about school-based immunization programs for adolescents. Although 58% believed that teens were less likely to return to their physician's office for health maintenance visits if they were immunized in school, 65% nevertheless indicated that they supported or strongly supported such programs for adolescents, compared with only 15% who opposed school-based adolescent immunization programs. However, 75% of respondents indicated that primary care physicians would be unlikely to receive notification that their adolescent patients had been immunized in school.

The measles outbreaks in various parts of the United States in the late 1980s and the early 1990s resulted in a renewed national focus on childhood immunization. Consequently, the nation concentrated on preschool immunizations through national goals,⁸ standards for managed care organizations,⁹ efforts to minimize missed opportunities to immunize,^10- 12 reminder-recall,^13- 17immunization registries,¹⁸ and other interventions targeting the medical home. However, until recently, adolescent immunization was not emphasized.

Adolescents often fail to receive preventive health care services, which results in significant unmet health needs.¹⁹ In particular, many do not receive recommended immunizations.^8- 20 As a result, adolescent immunization rates lag significantly behind immunization rates for younger children. This study provides new information about the effect that provider practices and attitudes have on adolescent immunization, building on a framework acquired through childhood immunization studies.

Most physicians who reported that they did not vaccinate adolescents cited lack of health insurance coverage for immunizations as one of the primary reasons. Of note, however, most of these physicians also reported that they did not participate in the VFC program. Many others reported that they did not routinely immunize adolescents because at the time of the survey, the VFC program did not provide immunizations for adolescents. In addition, most physicians who reported that they did not routinely immunize adolescents also practiced in states that did not require insurance companies to fully cover the cost of all recommended immunizations through so-called "first-dollar" laws. Hence, a concerted effort to increase physician participation in the VFC program—which has since been expanded to include all routine adolescent immunizations—and to urge states to require insurance companies to fully cover the cost of childhood and adolescent immunizations should prompt more physicians to routinely administer ACIP-recommended immunizations to all of their eligible adolescent patients.

Many opportunities to immunize eligible adolescents are missed. For instance, less than half of the physicians reported that they checked the immunization status of adolescent patients at illness-related visits, and only about a quarter reported ever immunizing adolescents at such visits. Clearly then, more needs to be done to educate physicians that there are few true contraindications for immunization, while also persuading them to check their adolescent patients' immunization histories at each encounter, and urging them to immunize when indicated instead of waiting. This is particularly important because many adolescents are not seen for health supervision, and hence may only come to the physician's office when ill.^19,21

While most physicians noted that their practices have an established policy or protocol for the immunization of adolescents at particular ages, very few indicated that their practice had any type of system to track and recall adolescents who are behind on immunizations. Tracking systems were also lacking for younger children. However, the immunization registries that are currently being developed for preschool children may eventually benefit adolescent immunization efforts. If these registries can be expanded to include adolescent immunization information, the use of tracking and recall systems for adolescents may become more widespread. However, practices need not wait for registries before implementing their own tracking and reminder-recall systems for adolescents. While reminder-recall systems have not been studied for the adolescent population per se, tracking combined with reminder and recall has been shown to be highly effective in improving immunization rates among both preschool children and adults; hence, such systems are likely to be effective for adolescents as well.²²

In this survey, immunization rates for varicella vaccine lagged behind rates for other routine immunizations. While overall, 61% of responding physicians reported that they vaccinated eligible adolescents with varicella vaccine, the percentage varied significantly by geographic region and by physician specialty. Notably, many respondents preferred that their adolescent patients contract wild chickenpox rather than receive this vaccine, despite the high complication rates that occur among adolescents who develop wild disease.

Throughout the past few years, with the expansion of the set of immunizations recommended for adolescents, school-based immunization programs have been developed in several North American cities.^23- 25 In addition, several states have begun implementing adolescent immunization requirements, often in tandem with expanded school-based immunization efforts.²⁶ Most school-based immunization programs are aimed at children attending middle school (ages 11-13 years), an age group with relatively good school attendance. To date, school-based immunization programs have primarily targeted hepatitis B immunization, and have had varied success rates, with the percentages of adolescents completing the hepatitis B series ranging from a low of 6% in one community to a high of 83% in another.²⁵ Our survey found that physicians who provided comprehensive health care to adolescents generally supported school-based immunization programs. However, only 35% of physicians were aware of school-based immunization programs that were available to any of their adolescent patients. Most also expressed reservations about potential discontinuity of care should adolescents utilize school-based immunization programs. School-based immunization programs could minimize such reservations by encouraging teens to visit their primary care physician for health supervision, and by routinely notifying physicians when their patients have been immunized in school.

Since the content of training programs, educational experiences, and continuing medical education courses offered to pediatricians and family physicians differ, it is useful to compare immunization practices to highlight specialty-specific areas to which improvements may be targeted. In this survey, pediatricians were significantly more likely than family physicians to have policies that promote adolescent immunization. This mirrors the findings of previous studies that found similar specialty-specific differences regarding childhood immunization practices.^{5- 6,11,27- 28} Therefore, while all physicians treating patients in this age group should be urged to emphasize adolescent immunization, particular consideration should be placed on educating family physicians about the importance of adolescent immunization, while promoting practices that increase their adolescent immunization rates. This might best be done in close collaboration with the American Academy of Family Physicians, as well as state and local medical societies.

We did not determine whether the reported practices of the study participants reflected their actual clinical practices, nor were we able to confirm whether survey respondents had immunization practices and attitudes that differed significantly from those of nonrespondents. However, the high response rate reduces the potential for response bias. The national scope of this survey of a randomly chosen cohort of pediatricians and family physicians is particularly valuable in that it sheds new light on factors influencing adolescent immunization practices nationwide.

The results of this survey suggest several means of increasing adolescent immunization rates including:

We can conjecture that, as these interventions are implemented in the hopes of improving adolescent immunization rates, they will prove to be as effective as were similar interventions that successfully improved infant immunization rates. However, further research is needed to elucidate the true effect of each of these proposed interventions on overall adolescent immunization rates.

Accepted for publication December 19, 2000.

This project was supported by contract 200-90-0869 from the Centers for Disease Control and Prevention.

Presented in part at the 1998 Annual Meeting of the Pediatric Academic Societies, New Orleans, La, May 3, 1998, and the 32nd National Immunization Conference, Atlanta, Ga, July 22, 1998.

Reprints not available from the author.

Corresponding author: Stanley J. Schaffer, MD, Division of General Pediatrics, Box 777, University of Rochester Medical Center, 601 Elmwood Ave, Rochester, NY 14642 (e-mail: Stanley_Schaffer@urmc.rochester.edu).