We conducted three iterations of focus groups for pediatricians and mothers in Seattle, WA; Cincinnati, OH; and Rochester, NY (total of 9 physician and 9 parent focus groups). A standard protocol, moderator guides, and pre-/post- test comparisons of acceptability of the rotavirus vaccine were uniformly applied.

The initial focus groups were held in July 2010, approximately 4 months following the first public announcement of PCV in a rotavirus vaccine, and approximately 2 months following the FDA recommendation to resume/continue rotavirus vaccinations. Two further focus group iterations occurred in August and September 2010. All focus group iterations included core discussions on rotavirus disease and rotavirus vaccines, followed by information on PCV. Co-investigators developed communication materials and messages prior to the focus group sessions and obtained parent and pediatrician feedback on these materials during the focus group discussions. The moderator's guide and communication materials were then revised iteratively based upon comments and data obtained from the previous focus groups until conceptual insights were exhausted. The focus groups lasted approximately 90 minutes. Although each moderator was independent and experienced in leading focus groups, participant responses may be affected by the manner in which information is conveyed to the focus groups.

Each focus group consisted of up to 8 unique participants. Physician participation was restricted to actively practicing, board-certified/-eligible, non-military, primary care pediatricians who typically administered >4 rotavirus vaccine doses per week. In each focus group, only one pediatrician participated from any given office. Participants were excluded if the pediatrician or his/her immediate family member worked in vaccine development, marketing, research, or regulation. Recruitment of pediatricians occurred through community pediatrician list serves, at events (e.g., grand rounds), and other administrative meetings, and the study team followed up with interested participants for a determination of eligibility.

Mothers at least18 years of age who reported having normally developing children between 6 months and 4 years of age were eligible for participation if they 1) were fluent and literate in English, 2) agreed to routine vaccination for the index child, and 3) at least partially influenced health-care decisions for their child. The ineligibility of mothers with an infant <6 months of age was intended to avoid the focus group discussions from influencing a mother's decision to have a current infant vaccinated or not. Potential participants were excluded if they or any immediate family member worked in vaccine development, marketing, research, or regulation; if they participated in any market study within the previous 6 months; or if they worked in the healthcare field. For each focus group, recruiters attempted to enroll at least 2 mothers with the educational attainment of a high school diploma or less. While these eligibility and exclusionary criteria for mothers were consistently applied across participating sites, the sampling frame and solicitation methods for mothers differed by site. In Seattle, mothers were recruited from a university-based primary care clinic, but in Cincinnati and Rochester a marketing telephone list was used to contact households for eligibility.

Focus group questions were open-ended, non-sensitive, and designed to maintain participant privacy. They included the following domains: a) perceptions of rotavirus disease, b) perceptions of rotavirus vaccination, c) attitudes toward the detection of PCV material in rotavirus vaccines, and, d) attitudes toward communication materials on these topics.

After each round of focus groups at each site, the audiotaped discussions were transcribed verbatim and then analyzed by use of an inductive coding technique for qualitative data [17,18]. Each investigator independently read each focus group transcripts specific to their site(s) and abstracted key themes regarding rotavirus disease, rotavirus vaccine, attitudes regarding the PCV finding, and communications. Investigators then met to discuss their independent analyses and negotiate a final list of themes. Quotations used in this article were excerpted from transcripts and are representative of the category to which they have been assigned. Group dynamics inherently affect focus group results and this was taken into consideration during our analysis.

We asked attitudinal questions at the beginning and end of each focus group to compare whether or not the focus group discussion had influenced or changed subject perceptions of rotavirus vaccines. These questions included attitudes and opinions regarding rotavirus vaccination (i.e., their acceptability, understanding and barriers to rotavirus vaccination). Changes in pre- and post-focus group questionnaire responses were assessed using the Wilcoxon signed rank test for non-normal distributions. For these comparisons, the estimated aggregated study power was >90% for pediatricians and mothers.

Prior to focus group enrollment, the CDC Human Research Protection Office determined this evaluation to constitute a public health non-research project, and the institutional review boards at each participating institution (Seattle Children's Hospital, University of Rochester, New York, School of Medicine and Dentistry, Cincinnati Children's Hospital and Medical Center) declared the project to be exempt. Potential participants consented to participate at screening and if they decided not to participate, they were thanked for their time and not contacted further. Permission to audiotape was obtained during screening and again at the outset of the group.

We conducted focus groups in three different regions of the United States (the West, Midwest, and Northeast). Table 1 describes the characteristics of pediatricians (n = 45) (Table 1) and mothers (n = 58) (Table 2) participating in focus groups. Pediatricians had an average 15.4 years of experience (median = 15, range = 1-39) following residency, and most (62%) belonged to a private practice group. About one-third of the children enrolled by their pediatric offices were publicly insured and eligible for the Vaccines for Children Program, [19] a federally-funded entitlement program providing vaccines at no cost to socioeconomically disadvantaged children. Most pediatricians solely administered RotaTeq™ (76%), compared with 4% who solely used Rotarix^® and 20% who reported using both vaccines.

More than 80% of pediatricians from all sites either strongly or somewhat agreed to the statement, "Before rotavirus vaccine was available, rotavirus was the most common cause of severe infectious diarrhea in children <2 years old in the US." About three-quarters (76%) strongly or somewhat agreed that rotavirus was responsible for more annual hospitalizations than influenza during the pre-vaccine era.

The average age for participating mothers was 33.2 years (median = 33.5, range = 21-44), and 76% had achieved more than a high school education. Most mothers (61%) reported being privately insured, compared with 33% having public insurance/Medicaid. The average household had 2.3 children, and most mothers (79%) had heard of rotavirus before the focus group.

Many pediatricians had negative memories of rotavirus infections and had observed a high burden of childhood disease from rotavirus gastroenteritis, as represented by common remarks included in Table 3. Those pediatricians who had the opportunity to observe rotavirus both during and after the pre-vaccine era understood firsthand the success of rotavirus vaccines in preventing severe pediatric acute gastroenteritis (Table 4).

Overall, pediatricians reported that the mothers of their patients rarely hesitated to accept rotavirus vaccination. Specifically noted were reports that, "[Mothers] are willing to accept [rotavirus vaccine] when you say the virus gives horrible vomiting and diarrhea and it can last two weeks." As the vaccines are orally administered, pediatrician participants stated that many mothers do not count it among the injections recommended in the vaccination schedule.

At the time of the focus groups, most participating pediatricians had some prior knowledge that DNA material from PCV had been detected in rotavirus vaccines. When given detailed information on this topic, many seemed to intuitively measure the theoretical risks of PCV against the benefits they personally observed from rotavirus vaccination. Nearly every participating pediatrician expressed an opinion that the detection of DNA or DNA fragments from PCV in rotavirus vaccines was not clinically important, repeatedly using the term "non-issue." Particularly influential to this appraisal was that the published, large phase III randomized clinical trials that found both vaccines to be highly effective and safe were actually conducted with DNA material from PCV already in both vaccines. Respondents reported that, "Now I've learned that the particles have been there from day one, so if you believe the safety data on the vaccine itself then you believe that [the vaccine] is safe." Furthermore, many pediatricians found reassuring that PCV is not known to cause illness in humans, stating, "... [humans are] literally exposed to thousands of viruses every day and don't become ill." Since PCV material is likely consumed by many people in pork products without adverse consequences, many observed that, "...we are exposed to [PCV] in the pork we eat, and we're fine. So this is not our only exposure. It's not a new virus to us."

Given their general dismissal of any theoretical risk from DNA material from PCV in rotavirus vaccines, many pediatricians indicated that they would not devote time to explain the PCV finding, although many agreed that they would provide information to mothers who specifically queried them on the topic. Their general reluctance was rooted in three stated reasons: 1) DNA material from PCV in rotavirus vaccines did not constitute any risk, so the point was both moot and unduly alarming to the mothers, 2) discussion would be lengthy (and potentially non-reimbursable) and would compete with discussion of many other important clinically relevant topics, and, 3) many mothers would not fully understand the issue even with further elaboration.

While more than three-quarters of the participating mothers had at least some college education, many expressed a general lack of scientific and technical understanding of viruses, DNA, how vaccines are manufactured, and how vaccines work. Approximately one-fifth (21%) of the mothers reported unfamiliarity with the disease "rotavirus" and nearly none had prior knowledge of the PCV finding. One mother remarked, "I guess what all of this brings up for me is that I don't really understand what's in vaccines... it makes me feel like I want to understand more what a vaccine actually is, that this is what's happening... that there's a virus that somehow got in there that people didn't know about." Simplified information regarding PCV material in rotavirus vaccines seemed frightening, while detailed information was frequently overwhelming. "I don't know what the porcine circovirus is, so to me, this statement sounds like there's another virus present in the rotavirus vaccine. So, that would concern me." Stating that the vaccine was simply "safe" without providing evidence often brought further suspicion, exemplified by the comment, "When they say something is safe... well, it's the classic hedge. They don't have any evidence that it's not safe, which just means that they may not have studied it enough."

Mothers largely supported the ideal of transparency in being provided information regarding all vaccine risks and benefits. However, mothers often remarked that if they did not understand a medical or technical topic regarding their children's health they commonly relied on their pediatrician's guidance. Nonetheless, a few mothers mentioned that the detection of PCV material in rotavirus vaccines eroded their confidence in the vaccines (Table 5).

Many mothers and pediatricians expressed alarm that those who abstain from pork consumption for religious or personal reasons may have unsubstantiated fears that PCV is pig material (rather than the fact that PCV is a virus commonly found in pigs). Concern was most directed at whether parents of Jewish or Muslim faiths would reject the vaccines due to this misunderstanding. Some participants were similarly concerned that, due to its name, Porcine circovirus was somehow related to H1N1 influenza (a.k.a. swine flu) or illnesses of animal origin (e.g., bovine spongiform encephalopathy, a.k.a. mad cow disease) (Table 6).

Pediatricians and some mothers cautioned that using the same acronym, PCV, for both Porcine circovirus and pneumococcal conjugate vaccine could cause confusion.

Following the focus group discussion of rotavirus, rotavirus vaccines and PCV, pediatricians showed no statistically significant changes in their recommendation of rotavirus vaccines or in perceived barriers to vaccination. Nearly all pediatricians (98%) reported no change in their opinion to recommend the vaccines (P = 1.00). Also, no appreciable net change was observed among pediatricians who reported that parental concern with vaccine safety would be a barrier to rotavirus vaccination (26% reported vaccine safety to be an increased barrier, whereas 23% reported it to be a diminished barrier; P = 0.75). (Tables 7 and 8)

Mothers, however, had more negative perceptions regarding the importance and safety of the vaccines and whether a subsequent baby would receive rotavirus vaccines. Roughly half of the participating mothers continued to feel confident that rotavirus vaccines are safe, while 37% changed their opinion to disagree with this statement. (P < 0.01) Following the focus group discussion, 16 of 52 mothers somewhat or strongly disagreed with the statement "If I had another baby, I would have that baby get rotavirus vaccine.", compared with 4 of 52 mothers stating this during the pre-focus group questionnaire. This indicates that 12 (23%) of 52 participating mothers changed their opinion on this issue following the focus group discussion. (P < 0.01) No significant change was observed regarding parental confidence that rotavirus vaccine prevents severe diarrhea (Tables 7 and 8).