Parental Attitudes Toward Varicella Vaccination FREE

DESPITE THE 1995 recommendation,¹ immunization against varicella remains far from universal. Results of a 1997 national survey indicated that only 26% of US children between the ages of 19 and 35 months had received the varicella vaccine.² To design interventions aimed at increasing the proportion of young children who receive vaccination against varicella, it is important to elucidate reasons for this high rate of underimmunization.

In general, reasons for underimmunization can be divided into "systems" barriers, provider policies and beliefs, and parental characteristics and beliefs.³ There are significant barriers to varicella immunization: payment for the vaccine was not initially covered by Medicaid in some locations,⁴ immunization against varicella is not required for school entry in all states,⁵ and the necessity of storing the vaccine at −15°C makes handling cumbersome.¹ In addition, physicians have not uniformly embraced use of the vaccine. In a previous study,⁶ we found that fewer than 50% of Washington State pediatricians recommended universal varicella immunization. Adherence to the recommended schedule for providing the vaccination was influenced by previous clinical experience with a serious complication of varicella infection, perceptions about the seriousness of naturally acquired infection, and beliefs about the cost-effectiveness of providing the immunization.

Given the barriers to immunization and lack of physician endorsement, parental health beliefs might be a major influence on whether a child receives the varicella vaccine. Overall, results of recent studies indicate that parents have generally positive beliefs about immunizations.^7- 8 There are, however, few data on parental beliefs specifically regarding varicella vaccination. One small survey of 34 parents who elected not to immunize their children against varicella found that reasons for not choosing the vaccine were related to concerns about efficacy, long-term consequences of immunization, and side effects.⁴

To further evaluate parental attitudes regarding the varicella vaccine, we conducted a survey of parents of children followed by private pediatricians in the Seattle, Wash, area. The survey was designed using a health beliefs model.⁹ The use of a parental health beliefs model may help to explain utilization of preventive services. With this model, the decision by the parent to use the varicella vaccine, or not, is divided into 5 components: the perceived susceptibility of the child to varicella, the severity of the infection, the perceived benefit of the immunization in preventing disease, barriers to receiving the vaccine, and perceived self-efficacy (the perception that the parent can promote his or her child's health despite obstacles).

In addition, we were interested in the effect of their child's pediatrician on these parental beliefs, and the influence of previous experience with the vaccine or natural infection on attitudes about varicella immunization. Prior to conducting the study, we postulated that parents would have positive beliefs about the vaccine and that pediatricians' beliefs would significantly influence parental judgments.

The study was conducted by the Puget Sound Pediatric Research Network, a regional practice-based research group in the Seattle area. Approximately 40 practicing pediatricians from 9 offices are members of the Puget Sound Pediatric Research Network. For this study, data were collected in 8 of the 9 offices from October 1997 through March 1998.

In each participating practice, parents were asked to complete an anonymous survey on attitudes about the varicella immunization at the time of an office visit. The final sample consisted of 598 completed surveys; 92.3% of respondents were mothers. Among parents who completed the survey, 68.8% indicated that they had graduated from college, and 99.0% were at least high school graduates.

The survey focused on 10 health belief statements about the varicella vaccine (Table 1). Topics covered in the statements included potential benefits of the immunization, susceptibility of unvaccinated children, severity of varicella, concerns about side effects and length of immunity, and general statements about immunizations. Parents indicated their level of agreement with each statement using a 6-point Likert scale, with possible responses ranging from "completely agree" to "completely disagree." Parents were also asked how much their decision on whether to vaccinate their child was influenced by their pediatrician's opinion. Responses to this question ranged from "not at all" to "strongly"; a response of "do not know pediatrician's opinion" was also included. Finally, parents were asked to indicate how much money they would be willing to spend out-of-pocket on the varicella vaccine. Other items on the survey included birth date of child being seen in the office and name of this child's primary pediatrician, whether the child had previously received the varicella immunization, whether any of the respondent's children had had varicella in the past, and parental educational level.

For the analysis, the Likert scale responses to each health belief statement were converted to an ordinal scale between 1 and 6. A score of 6 corresponded to a response of "completely agree" for positively worded statements about the varicella vaccine, or a response of "completely disagree" for negatively worded statements. Thus, higher scores denoted positive health beliefs regarding the varicella immunization. This scoring system allows for comparison of level of agreement with different health belief statements.

A composite varicella vaccine health belief score was computed from each parent's responses on the survey by dividing the sum of scores for individual statements by the number of statements for which the parent indicated a level of agreement (because not every parent responded to every statement). Thus, composite health belief scores also ranged from 1 to 6. Composite health belief scores of parents with several different dichotomous characteristics were compared using Mann-Whitney tests. These characteristics included previous experience with a child with varicella (yes or no), varicella vaccine status of child being seen (previously received immunization or not), age of child being seen (<18 months old or ≥18 months old), high level of parental education (at least college graduate or less education), willingness to pay for varicella vaccine out-of-pocket (≥$50 vs <$50), and level of influence by pediatrician in making decision about the vaccine (influence quantified as "quite a bit" or "strongly" vs other responses).

Puget Sound Pediatric Research Network pediatricians had previously participated in a survey on physician attitudes regarding the varicella vaccine.⁶ Each respondent indicated whether he or she followed the American Academy of Pediatrics recommendation for varicella vaccination, or had another policy. These responses were linked to the parental surveys based on whom parents identified as their child's primary pediatrician. Using this schema, composite health belief scores of parents whose child received care from a pediatrician who recommended universal varicella vaccination were compared with health belief scores of parents whose child's pediatrician had some other varicella immunization policy. Differences in health belief scores between these 2 groups were assessed with Mann-Whitney tests. In addition, parents were placed in 1 of 4 groups based on the degree of influence by their child's pediatrician on their decision to immunize their child against varicella and this pediatrician's varicella vaccine policy. Differences in composite health belief scores between parents in different groups were assessed with analysis of variance. Finally, regression analysis was used to determine whether level of physician influence and/or pediatrician varicella immunization policy were independently associated with composite health belief scores.

The study was approved by the Children's Hospital and Regional Medical Center Institutional Review Board and the University of Washington Human Subjects Committee.

The median age of the 598 children being seen on the day that their parents completed the varicella vaccine survey was 16.8 months (range, 2 days to 16 years 10 months); 45.5% of the children were younger than 18 months. A total of 224 parents responded that at least 1 of their children had previously contracted varicella. However, among 380 study children older than 12 months, only 83 had received the varicella vaccine (22.7% of those whose parents responded either yes or no; 12 parents were unsure whether their child had received the immunization, and 3 parents did not respond to the question). In addition, 124 patients who were older than 12 months had previously had varicella. Thus, among eligible children, the rate of vaccination against varicella was 32.0%.

Responses to individual health belief statements regarding the varicella vaccine are summarized in Table 2. Parents agreed that receiving all recommended vaccines was important to their child's health (median score, 6.0; 88.4% of respondents agreed or completely agreed with this statement). Generally, parents agreed that their children were likely to get varicella if not immunized (median score, 5.0), and were not overly concerned about the number of shots that their children were receiving (median score, 5.0). However, parents did not agree that the vaccine was worthwhile if the only benefit was in reducing time lost from work; 57.4% of respondents disagreed or completely disagreed with this statement.

The distribution of composite varicella vaccine health belief scores among the 598 responding parents is displayed in Figure 1. Health belief scores were widely distributed between 1 and 6; the median score was 4.1. Comparisons of composite health belief scores in parents with differing characteristics are shown in Table 3. Parents whose children were younger or had previously received the varicella vaccine had significantly more positive health beliefs than the appropriate comparison groups. In addition, parents who indicated their pediatrician influenced their opinions about the immunization "quite a bit" or "strongly" had more positive beliefs about the vaccine than those who stated that their pediatrician's opinion was less influential (or unknown). Health beliefs of parents who had at least a college education were similar to those with lower levels of education. Finally, parents who had experienced a child with varicella had significantly less positive health beliefs about the immunization than those who had not had a child with varicella.

When asked what the most that they would spend out-of-pocket for their child to receive the varicella vaccine, 56.3% of respondents indicated that the maximal amount was $25 or less, and 81.5% indicated that the most they would spend was $50 or less. Conversely, 7.8% of responding parents indicated that they would be willing to spend $100 or more. Parents who responded that they were willing to pay more than $50 for the vaccine had significantly more positive health beliefs about the varicella immunization than those who would spend $50 or less out-of-pocket (median scores, 4.5 and 4.1, respectively; P<.001).

A total of 30 different physicians were identified as primary pediatricians for the children whose parents completed surveys. One provider, identified as the primary pediatrician for one child, did not complete the physician survey on varicella vaccine. Among the remaining 29 pediatricians, 11 (38%) adhered to the American Academy of Pediatrics recommendation for universal immunization against varicella. Parents whose child's pediatrician recommended universal immunization against varicella had significantly more positive health beliefs than those whose child's pediatrician had another varicella vaccine policy (P<.001). These health beliefs were also modified based on how much parents indicated that the pediatrician influenced their decision to immunize their child against varicella (Table 4). Overall, the distribution of health belief scores in the 4 groups was significantly different (P<.001). Both level of influence of the pediatrician and the varicella vaccine policy of the child's physician were independently associated with composite health belief scores (P<.001 and P = .002, respectively, using regression analysis and adjusting for each characteristic).

Results of a 1993 survey indicated that parents of children followed by private pediatricians in the Seattle area had positive health beliefs regarding immunizations.⁸ Using techniques similar to those employed for the survey on varicella vaccine, the overall health belief score for immunizations was 4.6 (on a scale of 1-6). This composite score was decreased by very low scores for statements regarding the advisability of providing immunizations to an ill child. For the majority of the remaining health belief statements, mean scores were greater than 5, with 6 being the highest score possible. Since this survey, the immunization rate in the United States has increased.^2,10 Given this, the results of the survey on parental attitudes regarding the varicella vaccine were surprising; we had expected that parents would have quite positive health beliefs. Instead, we found that parental enthusiasm about the immunization was rather weak.

It appears that the ubiquitous nature of varicella in childhood may play a role in decreasing the perceived necessity for varicella vaccine. A mean score of 3.7 for the statement, "My child may not need the chickenpox vaccine because chickenpox is usually a minor illness," indicates that, overall, respondents are ambivalent about the seriousness of varicella. The finding that parents reporting previous experience with chickenpox in their own child(ren) had less positive health beliefs about varicella vaccine suggests that the disease is seen by many parents as benign, or perhaps as a rite of passage. This may explain the relatively low health belief score (mean, 3.6) for the statement regarding the risks vs benefits of the varicella vaccine. Even for a very safe vaccine, concerns about side effects are magnified when there is a common perception that the illness prevented by the immunization is generally minor.

Responses to the survey indicate that parents were concerned about the value of the varicella vaccine if the immunity provided is not lifelong. Whether the parents were responding to the statement about lifelong immunity as a hypothetical situation or had previous knowledge about debate among health professionals over this issue is unknown. In our survey of Washington State pediatricians, concern over the possible lack of lifelong immunity was associated with not recommending universal varicella vaccination.⁶

Perhaps the most striking finding of the survey was that parents placed little value on the potential of childhood immunization against varicella in preventing time lost from work. The statement, "The chickenpox vaccine is worthwhile even if the only benefit is that, in preventing chickenpox in my child(ren), I and/or my spouse will not have to miss work," was the lowest rated health belief statement in the survey. In a study by Lieu et al,¹¹ it was estimated that the mean value of work loss time for parents caring for a child with varicella was $183. Only after the inclusion of the savings from preventing work loss was varicella vaccine found to be cost-effective. However, our results suggest that these potential savings may not be a motivating factor for parents in their decision to immunize their children against varicella. Further, promoting the potential of varicella vaccine to reduce parental work loss may not be an effective strategy to raise the rate of immunization.

However, the results of the survey provide insight on interventions that might be beneficial. For example, respondents had positive health beliefs about the ability of the vaccine to prevent complications from varicella, even if they occur in only 1 to 2 of 1000 children with the illness. Perhaps this benefit of the vaccine should be more strongly emphasized in educational materials. We also found that parents of children younger than 18 months had more positive beliefs about the varicella immunization than parents of older children. This finding suggests that it might be more effective to advocate vaccination in infants than to promote immunization in older children.

The interaction between how greatly parents are influenced by their child's pediatrician, and the varicella vaccine policy of that pediatrician, is interesting. Parents who indicated that their pediatrician's opinion about the vaccine significantly influenced their decision to use the immunization had more positive health beliefs than those who responded that they were less influenced. Since being strongly influenced by the pediatrician was independently associated with composite health belief score after adjusting for physician policy on varicella vaccine, this finding may be more indicative of a general belief in the health care system than a response to any specific information from the pediatrician. However, the converse was also true; regardless of the reported level of influence by the pediatrician, health belief scores were significantly more positive among parents whose child's pediatrician recommended universal immunization against varicella. This suggests that education by physicians might contribute to more positive health beliefs about varicella vaccine. This also underscores the importance of aggressive education to convince pediatricians of the necessity of adopting universal varicella immunization if they are to play a central role in convincing parents.

There are several limitations to this study. Health beliefs about varicella vaccine among parents in the Seattle area may not be representative of the United States as a whole, especially given the high educational level of the population included in the survey. Although this area has an overall immunization rate that is higher than the many metropolitan areas in the country, only 13% of children aged 19 to 35 months in King County, Washington, had received the varicella vaccine.² In addition, since the survey has not been validated, it is not certain if the results accurately reflect health beliefs regarding the varicella immunization. However, composite health belief scores were associated with the maximal amount of money that parents indicated they were will to spend out-of-pocket for the vaccine, providing a measure of validation. Finally, the meaning of our finding that parents of children who had previously received the varicella vaccine had more positive health beliefs than those who had not immunized their child is unclear. Since the immunization preceded the survey, it may be that the process of vaccination promoted positive health beliefs rather than vice versa.

Despite these limitations, our findings suggest that parents have significant reservations about the varicella vaccine. Efforts to change these perceptions may be a key component of interventions designed to increase the proportion of children in the United States who are immunized against varicella.

Editor's Note: It's nice to see that even though Americans work more hours than any other industrialized nation, at least the mothers don't think maintaining that pace (not missing work) is worth an immunization for their children.—Catherine D. DeAngelis, MD

Accepted for publication September 10, 1999.

This study was supported by an Immunization Special Projects Award from the Ambulatory Pediatric Association, McLean, Va, and by a National Research Service Award (5-T32-PE1002) from the National Institutes of Health, Bethesda, Md, to the University of Washington School of Medicine.

This study was presented, in part, at the 1999 Annual Meeting of the Pediatric Academic Societies, San Francisco, Calif, May 1, 1999.

Members of the Puget Sound Pediatric Research Network who participated in this study are as follows: Kathryn DelBecarro, MD, Ann Champoux, MD, Ruth Conn, MD, Ruth Crosby, MD, Steven Dassel, MD, Michele Fang, MD, Richard Joslin, MD, Bob Fukura, MD, Kathryn Koelemay, MD, Carol Lengyel, MD, Ouri Malliris, MD, Ted McMahon, MD, Helen Matthews, MD, Zaiga Phillips, MD, Peggy Sarjeant, MD, Jeffrey Scott, MD, Donald Shifrin, MD, Steve Schlafer, MD, Donna Smith, MD, Jeffrey Steele, MD, Bob Telzrow, MD, Frits vanPaasschen, MD, Lee Vincent, MD, Margaret Wheeler, MD, Cathy Wilkinson, MD, Roberta Winch, MD, Agnes Wong, MD, Cheryl Wright-Wilson, MD, and Kyle Yasuda MD.

We thank Virginia Hawkins-Gerde for her help in conducting the study.

Reprints: James A. Taylor, MD, Department of Pediatrics, University of Washington, 146 N Canal St, Suite 300, Box 358853, Seattle, WA 98103-8652 (e-mail: uncjat@u.washington.edu).