Potential Strategies to Achieve Universal Influenza Vaccination for Children: Provider Attitudes in Two States

Abstract

Objective:

Childhood influenza vaccination rates remain suboptimal. Provider perceptions on strategies to achieve universal vaccination are needed. We assessed the perceptions and attitudes of primary care providers across 2 states regarding 2 strategies to potentially bolster rates: centralized reminder/recall (C-R/R), such as reminder/recall (R/R) notices from state immunization registries, and influenza vaccination by complementary community vaccinators (CCVs), such as retail pharmacies, schools, and health departments.

Methods:

We sent a mailed survey to a representative sample of providers across Colorado and New York. Questions addressed R/R activities for influenza vaccine, preferences and attitudes about the health department sending C-R/R notices for influenza vaccine, and attitudes about CCVs. Bivariate analyses assessed provider perceptions and compared perceptions by state.

Results:

The overall response rate was 56% (n = 590/1052). Twenty-two percent of providers in Colorado and 33% in New York performed practice-based R/R for all patients during the 2015–16 influenza season. Eighty-one percent of providers in both states preferred the health department or had no preference for who sent C-R/R notices for influenza vaccine to their patients; most preferred to include their practice names on C-R/R messages. Many providers in both Colorado (75%) and New York (46%, P < .001) agreed that their patients like the option of having CCVs where children can receive influenza vaccine. Some providers expressed concerns regarding potential loss of income and/or difficulty documenting receipt of influenza vaccine at CCVs.

Conclusions:

Most providers support C-R/R, and many support CCVs to increase influenza vaccination rates. Collaborations between traditional primary care providers and CCVs might boost coverage.

The Advisory Committee on Immunization Practices recommends that all people ≥6 months old receive influenza vaccination each year.¹ Vaccinating children protects those who are vaccinated^2–4 and reduces the burden of influenza illness in the entire population through herd immunity.⁵ However, despite current recommendations, the vaccination rate for children 6 months to 18 years of age during the 2015–16 season was 59%, well below the Healthy People 2020 objectives of 70% for children.⁶ The task of vaccinating all children within the relatively short time frame between when the vaccine is available and when influenza becomes widespread is a daunting task that will likely require innovative and collaborative approaches.^7–9 Two potential strategies to boost coverage involve conducting reminder/recall (R/R) centrally and vaccination by complementary community-based providers.

Centralized reminder/recall (C-R/R), in which public health departments send out reminders to entire counties or regions using data available within an immunization information system (IIS), is an innovative method that has previously been shown to increase childhood vaccinations. C-R/R messages can include the name of the child’s primary care practice and provide education about the importance of vaccinations. This approach has been shown to be more effective and cost-effective for increasing childhood immunization rates than efforts to incentivize providers to conduct R/R at the practice level^10–16 and substantially reduces the burden on providers, as messages are sent centrally and not by the practice. C-R/R using an IIS for influenza vaccine has been conducted with some success among urban low-income children,¹⁴ among children with high-risk conditions,¹⁵ and for pandemic influenza among Medicaid-eligible children with chronic conditions.¹⁶ However, few studies have conducted C-R/R using IISs among healthy children of all ages and income groups. Although providers have been supportive of C-R/R for childhood vaccines,^17,18 it is unknown whether they will be as supportive of C-R/R for influenza vaccination because of the unique challenges influenza vaccine delivery poses, including timing of vaccine receipt, potential vaccine hesitancy among parents, and varying availability of vaccine during the season.

Another potential approach to increasing influenza vaccine delivery that has been touted among other age groups¹⁹ is the inclusion of CCVs, such as public health departments, schools, and retail pharmacies, in the vaccine delivery effort. Although most childhood influenza vaccines in the United States are administered by primary care providers,²⁰ CCVs also provide vaccinations for children and could potentially be key partners. The extent of involvement of CCVs varies according to state laws; for example, pharmacies in Colorado can vaccinate children with influenza vaccine, but in New York they could not do so until legislated by the state in 2018.²¹ Provider attitudes and preferences regarding sharing responsibility for the delivery of influenza vaccine with other CCVs has not been well studied. Providers may see CCVs as a way of alleviating pressure on their practice to deliver high volumes of influenza vaccine during the short time between availability of vaccine and the influenza season. Alternatively, providers may view them as competitors for patients or as impeding children’s use of the medical home.

To address the potential of these 2 strategies for increasing influenza vaccination among children, we assessed and compared among primary care providers for children in Colorado and New York 1) the prevalence of practice-based R/R efforts for influenza vaccine, 2) attitudes and preferences about C-R/R for influenza vaccine, and 3) attitudes and perceptions about influenza vaccine delivery by CCVs.

Methods

This study was approved by the Colorado Multiple Institutional Review Board and by the University of California at Los Angeles Institutional Review Board as an expedited application with a waiver of written consent. This survey preceded a randomized controlled trial in these states testing the effectiveness of C-R/R using both the Colorado and New York State IISs for influenza among children. The states were chosen to be able to contrast different IIS reporting laws, experiences with C-R/R conducted by public health departments, and regulatory guidelines for CCVs.

Survey Study Population

In Colorado, information in the Colorado Immunization Information System (CIIS) was used to select a sample of primary care practices (pediatric, family medicine, community health, and rural health centers) from all urban and rural counties to receive a survey. Frontier counties (<6 people/square mile) were excluded. In rural counties, all primary care practices were included. All pediatric and community health clinics were selected in urban counties. Owing to the large numbers of urban family medicine practices, a sample of those practices was selected based on the proportion of children they served according to CIIS data. Patients in the CIIS are automatically assigned to the clinics that provided their last immunization service unless the clinic actively declines to take responsibility for the patient.

In New York, owing to the large number of primary care practices a stratified sample of all primary care practices was selected using information in the New York State Immunization Information System (NYSIIS), with the probability of inclusion being proportional to the number of children associated with the practice. In New York, patients can be assigned to multiple clinics based on where they receive immunizations.

In both states, we used the site of last immunization service to determine the proportion of children. An invitation to participate in the survey was addressed to the senior physician partner, and we requested that the person in charge of immunization decisions within the practice complete the survey.

State Differences

There are some important differences between Colorado and New York that might impact our survey. First, pharmacists in Colorado can provide influenza vaccinations to children starting at age 2; however, before 2018, pharmacists in New York could not provide influenza vaccine to those ≤ 19 years old.²¹ In addition, although public health entities in both Colorado and New York offer immunizations, the involvement of the public sector is higher in Colorado owing to the predominant role of health departments in the provision of rural health care.²² Second, in recent years, many Colorado providers have experienced C-R/R programs primarily focused on routine childhood and some adolescent immunizations, although not influenza immunization. In contrast, providers in New York have not had any experience with C-R/R conducted by the health department.

Finally, the states differ in regulations regarding reporting to the state IIS. The state of Colorado does not mandate that providers (public, private, pharmacy) enter or upload immunization data into the CIIS, although most providers participate. In addition, because Colorado has an implicit consent policy with the ability to opt out, a patient’s information may be uploaded to the CIIS through birth records automatically unless a parent actively opts out.²³ Unlike Colorado, the state of New York mandates that all health care professionals (public, private, Vaccines for Children program) send information about childhood vaccinations to the NYSIIS. New York also requires mandatory enrollment among those <19 years old with no option to opt out, thereby allowing birth records to be uploaded into the NYSIIS.²⁴

Colorado and New York State IISs

The CIIS includes client and vaccine event data through live data entry into a web-enabled application and through electronic transfers from providers, state vital statistics, and insurers. More than 99% of children <6 years old, 95% of children ages 6 to 10 years, and 79.6% of adolescents ages 11 to 17 years have ≥2 records in the CIIS.²⁵ Similarly, in the NYSIIS, data are uploaded by providers through direct entry via a web-enabled application or through electronic transfers of immunization data. The NYSIIS provides oversight for all of New York State except New York City. Eighty-nine percent of children <6 years old, 93% of children ages 6 to 10 years, and 94% of adolescents ages 11 to 17 years have ≥2 records in the NYSIIS.

Survey Design and Administration

Survey questions were developed by the study teams in both states based on previously developed survey instruments.^17,18 The questions were piloted by 20 primary care physicians, half in each state. R/R was defined in the survey as including “phone calls, mailings, text messages or messages via a patient portal to parents of children or adolescents who may be due or overdue for immunizations.” Question responses were either discrete variables or 4-point Likert scale responses, and the questions explored current practice-based R/R for influenza vaccine, attitudes/preferences about who should conduct R/R (practice-based or C-R/R) and how it should be done, and attitudes about influenza vaccination by CCVs.

The survey took place from February to April 2016 and was conducted using a modified Dillman methodology for mailed surveys.^26,27 All practices were sent a pre-survey information letter followed by a paper-based survey sent via certified mail within 14 days. Up to 3 additional mailed surveys were sent to nonresponders for 8 more weeks. A $10 bill was included in the first and third mailings.

Data Analysis

Survey responses were analyzed by state, and chi-square tests were used to compare between states. The Fisher exact test was used when low frequencies of a response were observed. For the Likert scale questions, the responses “strongly disagree” and “disagree” were combined before analysis. A chi-square test was also used to examine whether beliefs about IIS accuracy were associated with preferences for who performs C-R/R. Results are reported combined if responses were not significantly different between the states and by state if significant differences were found.

Results

A total of 1052 practices in both Colorado (n = 569) and New York (n = 483) were included in the sample. The response rate was 56% (n = 590/1052) overall, 62% (n = 351/569) in Colorado, and 49% (n = 239/483) in New York. Forty practices were removed from the analysis because they did not provide childhood immunizations. Table 1 summarizes the breakdown by practice type for each state. Virtually all practices provided influenza vaccine to children during the 2015–16 season. Three-quarters of the practices surveyed in Colorado reported entering influenza vaccine into the CIIS compared with all New York practices. Among those Colorado practices (n = 78) that did not upload influenza vaccine, the main reasons given were that they did not participate in the CIIS and/or they did not have time to manually enter seasonal influenza into the CIIS.

Table 1.

Practice Characteristics in Colorado and New York^*

Characteristic	Colorado	New York	Total	P Value
Total responses	56.9 (313)	43.1 (237)	100 (550)	—
Practice type
Pediatric	33.6 (105)	51.9 (123)	41 (228)	<.001
Family medicine	51.8 (162)	39.7 (94)	47 (256)	—
Community health center	14.7 (46)	8.4 (20)	12 (66)	—
Participate in IIS	74.3 (226)	100.0 (237)	—	<.001
Provide VFC vaccines	75.7 (228)	97.9 (231)	85.5 (459)	<.001
Provide flu vaccine 2015–16 season to children ≤18 years	98.4 (308)	100.0 (237)	99.1 (545)	.07
Sent out reminders to all children for flu 2015–16 season	21.8 (67)	32.8 (76)	26.5 (143)	.003
Position title of respondent
Clinician†	50.8 (133)	61.0 (125)	55.2 (258)	.005
Office manager	30.9 (81)	30.7 (63)	30.8 (144)	—
Nurse	18.3 (48)	8.3 (17)	13.9 (65)	—
Female respondents	75.4 (230)	63.4 (149)	70.2 (379)	.002
Practices with ≥50% of patient visits to children <18years	39.5 (118)	56.4 (127)	46.8 (245)	<.001
Practices with ≥50% Medicaid or CHP patients	35.7 (104)	33.9 (75)	35.0 (179)	.005
Practices with ≥50% Hispanic patients	21.2 (60)	11.2 (24)	16.9 (84)	<.001
Practices with ≥50% black patients	1.4 (4)	5.9 (13)	3.4 (17)	<.001

IIS indicates immunization information system; VFC, Vaccines for Children program; and CHP, community health program.

P values were obtained using the chi-square test, with the exception of that for “provide VFC vaccines,” which was obtained by the Fisher exact test owing to low cell counts. Data are presented as % (n).

^*There were <10% missing responses, except for “position title,” which had 15.1% missing responses.

^†Clinician = physician, nurse practitioner, physician assistant, or senior physician partner.

Practice-Based R/R Efforts for Influenza

Less than a quarter of Colorado practices and about a third of New York practices reported sending influenza vaccine reminders to all children in their practices during the 2015–16 season (P = .003); 10% and 14%, respectively, sent R/R notices to a select group of children, such as patients with chronic conditions (data not shown).

Attitudes and Preferences About Centralized R/R FOR Seasonal Influenza

Table 2 demonstrates provider attitudes and preferences about C-R/R for influenza vaccine. Three quarters of Colorado practices and half of New York practices agreed that C-R/R notices should include information about other community sites where children could be immunized (P < .001). Other attitudes about C-R/R did not differ by state. The overwhelming majority of providers (90%) were in favor of the health department sending out C-R/R notices to families in their practice. Slightly more than a third were concerned about meeting demand for influenza vaccine if a C-R/R notice was sent by the health department. Most felt that a C-R/R for influenza vaccine might help generate revenue for their practice.

Table 2.

Providers’ Attitudes About Centralized Reminder/Recall for Influenza Vaccine

Statement	Colorado	New York	Total	P Value
I would be in favor of the health department sending out centralized reminder/recall notices to families in my practice with children who need seasonal influenza vaccine.				.02
Strongly agree	49.8%	59.7%	54.2%
Agree	40.6%	28.8%	35.4%	—
Disagree/strongly disagree	9.6%	11.6%	10.5%	—
It would be okay with me if a centralized reminder/recall asked families to contact my practice for seasonal influenza vaccination and also included other community locations where they could be immunized.				<.001
Strongly agree	24.8%	18.2%	21.9%
Agree	49.0%	34.2%	42.4%	—
Disagree/strongly disagree	26.2%	47.6%	35.7%	—
I would be concerned about meeting demand for influenza vaccine in my practice if a centralized reminder/recall for seasonal influenza vaccine were conducted by the health department.				.55
Strongly agree	7.6%	5.2%	6.5%
Agree	28.9%	30.4%	29.6%	—
Disagree/strongly disagree	63.6%	64.3%	63.9%	—
My practice does not have the resources to conduct reminder/recall for pediatric patients in need of seasonal influenza vaccine.				.16
Strongly agree	11.7%	10.6%	11.2%
Agree	34.8%	27.8%	31.7%	—
Disagree/strongly disagree	53.4%	61.7%	57.1%	—
A centralized reminder/recall done by the health department for seasonal influenza vaccine might help generate revenue for my practice.				.66
Strongly agree	10.0%	11.1%	10.5%
Agree	60.2%	56.2%	58.4%	—
Disagree/strongly disagree	29.8%	32.7%	31.1%	—
Centralized reminder/recall sent by the health department for seasonal influenza vaccine would particularly help children in my practice who need 2 doses of influenza vaccine.				.76
Strongly agree	35.2%	34.1%	34.7%
Agree	47.9%	46.6%	47.3%	—
Disagree/strongly disagree	16.9%	19.4%	18.0%	—
The state immunization registry does not have the most accurate information about my patients’ seasonal influenza vaccines.				<.001
Strongly agree	14.2%	7.1%	11.1%
Agree	41.8%	26.3%	35.0%	—
Disagree/strongly disagree	44.0%	66.5%	54.0%	—

P values were obtained using the chi-square test; all questions had <10% missing responses.

Providers were asked their overall preference for how a C-R/R for seasonal influenza among all children should be conducted taking into account feasibility, cost, privacy, and medical responsibility issues. Eighty-one percent of respondents preferred that the health department send out C-R/R notices or had no preference for how it was conducted (the Figure presents state differences). Less than a quarter (19%) preferred that their practice send out reminders to their patients for seasonal influenza rather than the health department. Among those who approved of C-R/R notices being sent by the health department, 95% of providers in both states preferred to have their practice’s name and phone number included along with the health department information on the R/R (data not shown).

Figure.

Overall preferences for who conducts reminder/recall (P = .72 by chi-square test; <10% missing values).

State differences were seen with respect to the timing of R/R notices, with 90% of New York providers preferring reminders be sent in the early flu season (August–October) compared with 78% of Colorado providers (P < .001). Providers in both states preferred that C-R/R notices include 1 (41%) or 2 (45%) reminders, and a minority preferred 3 or more reminders (10%) (data not shown). More than a third of the practices had no preference with regard to reminders being sent all at one time or staggered in waves to patients in their practice. A third of the practices said they would prefer to have C-R/R messages sent staggered in waves, and a small proportion (7%) did not want the health department to send C-R/R notices for influenza vaccine to anyone (data not shown).

Finally, there were state differences in attitudes about the perceived accuracy of seasonal influenza data in the IIS. Fifty-six percent of the providers in Colorado agreed that the state IIS does not have the most accurate information about seasonal influenza vaccine versus only 33% of New York providers (P < .001). There were no observed state differences when comparing perceived inaccuracies in the IIS to preferences for how C-R/R was performed. In Colorado, 76%, 83%, and 83% (strongly agree, agree, somewhat/strongly disagree regarding inaccuracies in the CIIS, respectively) preferred that C-R/R be performed by the health department or had no preference for who sent C-R/R notices (P = .57). Similarly, in New York, 80%, 79%, and 80%, respectively, preferred that C-R/R be performed by the health department or had no preference for who sent C-R/R notices (P = .99).

Attitudes About Influenza Vaccine Delivery by CCVs

As shown in Table 3, three quarters of providers in Colorado and almost half of providers in New York reported that families would like the option of being vaccinated at other locations (P < .001). A majority thought having CCVs would make it difficult to estimate how much vaccine to order (P < .001) and that their practice would lose income if children receive influenza vaccine elsewhere, although this perception was more notable in New York (P = .002). Providers felt that children receiving influenza vaccine outside the practice would create vaccine documentation problems. Although almost all providers thought that families would prefer to receive influenza vaccine at their practice, less than half in both states agreed that the availability of CCV reduced the burden on their practice to immunize all children. Roughly two thirds of practices in both states agreed that influenza vaccine delivery was profitable at their practice.

Table 3.

Attitudes About Complementary Community Vaccinators

Statement	Colorado	New York	Total	P Value
Families at my practice like the option of having multiple locations besides my practice where their children can receive seasonal influenza vaccine.				<.001
Strongly agree	23.9%	11.2%	18.9%
Agree	51.1%	34.8%	44.7%	—
Disagree/strongly disagree	25.0%	53.9%	36.3%	—
It would be hard to estimate how much seasonal influenza vaccine to order if my pediatric patients were to receive seasonal influenza vaccine elsewhere.				<.001
Strongly agree	36.0%	55.9%	44.7%
Agree	43.2%	29.7%	37.2%	—
Disagree/strongly disagree	20.9%	14.4%	18.0%	—
I prefer that all seasonal influenza vaccine be delivered at sites outside of my practice such as public health clinics, pharmacies, retail sites, or schools.				.58
Strongly agree	2.8%	2.2%	2.5%
Agree	2.1%	3.5%	2.7%	—
Disagree/strongly disagree	95.0%	94.3%	94.7%	—
When my patients receive their seasonal influenza vaccine at a site other than my practice it creates a problem for my practice to document receipt of the vaccine.				.15
Strongly agree	34.9%	42.7%	38.4%
Agree	34.6%	32.8%	33.8%	—
Disagree/strongly disagree	30.4%	24.6%	27.8%	—
Having multiple seasonal influenza delivery sites outside of my practice would reduce the burden on my practice to immunize as many of my pediatric patients as possible.				.13
Strongly agree	12.7%	7.8%	10.5%
Agree	31.0%	28.7%	30.0%	—
Disagree/strongly disagree	56.3%	63.5%	59.5%	—
Seasonal influenza vaccine delivery is profitable for my practice.				.57
Strongly agree	25.0%	21.4%	23.3%
Agree	41.9%	45.9%	43.8%	—
Disagree/strongly disagree	33.1%	32.7%	32.9%	—
My practice loses income if children in my practice receive seasonal influenza vaccine at sites other than my practice.				.002
Strongly agree	27.1%	34.2%	30.3%
Agree	30.4%	38.7%	34.1%	—
Disagree/strongly disagree	42.5%	27.0%	35.6%	—

Discussion

Providers’ perspectives on strategies to remind families to immunize all children for influenza are important to understand given the challenges of accomplishing universal influenza vaccination within a limited time span. Given the fact that less than a third of providers in both states reported sending reminders of any kind for influenza vaccine to all patients within their practice, C-R/R notices sent by the health departments using IIS data could reach many more children than reminders sent by some practices. Providers in both states were generally supportive of C-R/R notices being sent by the health department for influenza vaccine, and the majority of them preferred a collaboration that would list practice names. In addition, providers in both Colorado and New York recognize that having CCVs may benefit families and somewhat reduce the burden on their practices; however, they also had concerns that vaccination by these community vaccinators could complicate vaccine ordering decisions or adversely impact practice finances.

R/R strategies have long been recommended by the Community Preventive Services Task Force as a means to increase childhood, adolescent, and adult immunization rates.²⁸ Specifically, practice-based R/R approaches continue to be encouraged; however, there are many obstacles to providers conducting R/R.²⁹ With increasingly automatic communication becoming available based on electronic health record systems to remind patients of upcoming appointments, one would think that R/R at the practice level would be easier than ever before. Our survey findings, however, suggest that there are still other significant barriers preventing practices from sending their own reminders. Centralized R/R has been proven effective for improving immunization rates in a number of settings and populations and for different immunizations. Although several studies have performed C-R/R for influenza vaccine for high-risk groups, few have assessed C-R/R for influenza vaccine for all healthy children through age 18 years.^14–16 Our results suggest that C-R/R using IISs, in collaboration with practices, is an alternative strategy that would generally be supported by providers for improving influenza immunization rates among healthy children.

Provider perceptions about C-R/R were quite similar in both states, despite differences in IIS reporting laws (New York, mandatory reporting; Colorado, voluntary reporting), provider exposure to C-R/R (greater in Colorado), and greater availability of CCVs in Colorado compared to New York. A prior article evaluated Colorado providers’ preferences for C-R/R for different age groups and vaccines, including influenza; compared with routine childhood and adolescent immunizations, practices were accepting but less positive about C-R/R for influenza vaccine.¹⁷ The current study goes beyond prior data to examine in more depth attitudes in contrasting states regarding C-R/R for influenza vaccine, as well as preferences for specific aspects of C-R/R. Providers in both states preferred a collaborative approach that included the provider’s name. Providers in Colorado are likely to have experienced C-R/R by the health department for other immunizations as part of previous research studies,^10,11 and their attitudes were generally similar, although a bit more positive than providers in New York who have not experienced C-R/R efforts. This suggests that participation in C-R/R did not cause any ill feelings toward such a program. Overall, providers in both states were strongly in favor of C-R/R.

There is little in the literature that assesses primary care providers’ attitudes and preferences regarding the role of CCVs in influenza vaccination. One previous article assessed how acceptable collaborative delivery methods for universal influenza immunization were to pediatricians nationally.⁷ Our study, unlike the previous one that included only pediatricians, sought the viewpoints of family medicine practitioners and community health centers. Both the current study and the previous national study found that providers recognize the potential benefit of CCVs to increase vaccination rates, yet both noted some logistic and financial concerns. Our 2-state study showed moderate state differences, which are likely explained by how immunizations are delivered by the state. Providers in Colorado, where more non–primary care immunization delivery opportunities exist via public health entities and pharmacies, were more accepting of CCVs and had fewer concerns with CCV vaccination affecting their bottom line, perhaps because they have experienced sharing the task of influenza delivery. Providers in New York appeared to be more cautious about CCVs, perhaps owing to their having less exposure to CCVs. Overall, however, providers across both states were generally accepting of CCVs. Collaborative efforts to engage primary care providers with CCVs might help overcome some reticence among providers and raise vaccination rates. Approaches that support public health entities and state IISs collaborating with medical establishments and CCVs to increase childhood influenza vaccination, perhaps with joint funding, are in line with the Institute of Medicine’s report about the necessity of public-private collaborations to improve population health.³⁰

One CCV worth noting because of its availability in the majority of states is the prescriptive authority given to pharmacists to immunize children of various age groups.³¹ Allowing pharmacists to immunize children improves access for patients and families because of their convenient locations and expanded hours. In addition, large pharmacy retail stores often receive influenza vaccine supplies before medical establishments and have the resources to advertise and provide incentives (eg, gift cards) to patients receiving influenza vaccinations at their locations. However, there has been some resistance to the involvement of retail-based clinics in the delivery of childhood vaccines.^32–34 Studies are needed to assess the potential impact of pharmacies on increasing influenza rates at the population level.

In states where other CCVs can be involved in vaccine delivery for children, it is intriguing to consider potential enhancements in C-R/R. For example, with collaboration between primary care practices and health departments, a C-R/R could refer patients back to their primary care provider or other community locations, depending on the availability of vaccine stocks of both private vaccine and the Vaccines for Children program. A targeted C-R/R could direct healthy school-aged children and adolescents to school-located vaccination clinics or pharmacies but young children or those with high-risk medical conditions to their medical home.⁷ These more nuanced approaches would require not only collaboration but also increased data accuracy and completeness within the IISs. More than half of states, like New York, mandate that providers report immunization data to the state’s IIS, which enables more complete and accurate IIS data.³⁵ However, in Colorado, where some pharmacies vaccinate but do not report to the CIIS, documentation problems could grow for both primary care providers and CCVs. In sum, more policies and strategies that increase all types of provider participation within IISs are necessary to ensure that centralized R/R for influenza or other vaccines can work most effectively.³⁵

There are several strengths and limitations to this study. This survey appears to be the largest recent study of provider R/R practices and preferences for influenza vaccination. The response rate for this survey is relatively high for a provider survey, although it was only 49% in New York. As with all surveys, respondents’ attitudes may differ from those who did not respond. Our sample is from 2 states, so it may not be nationally representative. Additionally, it is possible that providers’ attitudes might be swayed by factors we did not measure, such as how providers upload data into the IISs. Our results did not show that IIS accuracy influenced provider preferences for who performs C-R/R. We did not ask providers to report how they uploaded data to their IISs. Finally, self-reported practices may not always reflect actual behavior.

Achieving universal influenza immunization for children annually remains a major challenge for the US health care system. Given the low percentage of providers that actually conduct R/R for their patients for influenza vaccine and the generally high level of provider support for centralized R/R for influenza vaccination, using IISs to perform C-R/R for influenza appears to be a good strategy, especially if it can be shown to be as effective as it has been for childhood vaccines. In addition to primary care providers, the assistance of CCVs including pharmacies, visiting nursing services, and schools may be necessary to bring children in for immunization and to ensure vaccine delivery. Increased collaborative efforts among providers, state/regional IISs, and CCVs may be an important step toward achieving our national goal of raising influenza vaccination rates among children.

WHAT’S NEW?

Achieving influenza vaccination of all children is a monumental task. Health care provider opinions on strategies to accomplish this task are not well understood. Two strategies are explored with providers in 2 states: centralized reminder/recall and complementary community vaccinators.