Immunisation attitudes, knowledge and practices of health professionals in regional NSW
National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases (NCIRS), The Children's Hospital at Westmead and the Discipline of Paediatrics and Child Health, University of Sydney, New South Wales
National Centre for Immunisation Research and Surveillance of Vaccine Preventable Diseases (NCIRS), Department of Infectious Diseases and Microbiology, The Children's Hospital at Westmead and the University of Sydney, New South Wales
Correspondence to: J. Leask, NCIRS, The Children's Hospital at Westmead, University of Sydney, Locked Bag 4001, Westmead NSW 2145, Australia. Fax: (02) 9845 1418; e-mail: JulieL3@chw.edu.au
Objective: This study investigated the immunisation knowledge, attitudes and practices among health professionals in two regional Area Health Services of NSW with low and high immunisation rates. It also compared these factors between the areas and between the health professional groups.
Methods: A self-administered questionnaire was posted in 2006 to health professionals, located within the North Coast and Hunter New England Area Health Services, whose practice could include immunisation. This included general practitioners (GPs), practice nurses, community nurses, hospital nurses and midwives.
Results: Out of 926 surveys sent, 434 were returned (47%). The great majority of the health professionals (97%) believed that vaccines were safe, effective and necessary. However, in approximately one-third of respondents, there were specific concerns about additives, immune system overload and the number of vaccines. Significantly more health professionals in the North Coast area believed that additives in vaccines may be harmful and that adding more vaccines to the schedule would make immunisation too complex. Among GPs, over half felt uncomfortable about giving more than two injections at the one visit.
Conclusions: Health professionals in this study had overall confidence in vaccines but had specific concerns about the number of vaccines given to children and vaccine content. These unfounded concerns may reduce parental confidence in immunisation.
Implications: There is value in governments and immunisation support workers continuing their efforts to maintain up-to-date knowledge among health professionals and support the delivery of appropriate and targeted information to address concerns about vaccines.
Australia currently enjoys high immunisation rates with 94% of 2-year-old being fully vaccinated.1 Most vaccines for children are publicly funded and provided free at point of care under the National Immunisation Program (NIP), and vaccine coverage is recorded and reported by the national population-based Australian Childhood Immunisation Register (ACIR). Despite high vaccination rates, pockets of opposition to vaccination exist in some communities in Australia, where lower coverage rates are known to occur. Communities with low vaccination coverage experience outbreaks of vaccine-preventable diseases.2
Concerns about vaccine safety significantly contribute to under-immunisation of children. A 2001 Australian study found that 58% of parents of children not up-to-date for immunisation cited disagreement or concerns about immunisation as their main reason.3 For advice regarding the safety of vaccines, parents look foremost to health professionals.4–6 Therefore, it is essential that health professionals are equipped to respond confidently to a wide range of parental concerns. Instances where health professional confidence in immunisation safety has declined have been followed by major reductions in immunisation rates.7
Few studies have compared attitudes between health professionals in areas of low and high immunisation rates. This study aimed to describe differences in attitudes between two regional areas in New South Wales (NSW) where, in 2005, immunisation coverage rates by two years of age differed at 86% for the North Coast (NC) and 93% for the Hunter New England (HNE).8 The second aim of the study was to identify the immunisation knowledge, attitudes and practices among health professionals who provide immunisation in regional NSW, with a focus on differences between professional groups. The goal was to improve the existing support of health professionals and ultimately to find targeted ways to improve immunisation rates, particularly in areas with sub-optimal coverage.
A particular focus of this study was to examine beliefs about the safety of the MMR vaccine in the light of an alleged but unfounded link to inflammatory bowel disease and autism, and evidence of the impact of this concern on UK health professionals.9,10 In addition, we examined the impact that the changes to the immunisation schedule prior to the survey had had on health professionals. This included the introduction of new vaccines and many schedule changes. Meningococcal C and pneumococcal conjugate vaccines and varicella zoster vaccines were recommended as part of the schedule but, for a time, not funded under the NIP.11 A funded program for hepatitis B vaccination of all neonates, rather than just those at risk, also commenced. With this change, midwives became the first provider of immunisation to infants. As a relatively new immunisation provider in Australia, our study sought information on the extent of midwives’ concerns about vaccine safety.
The study was an anonymous self-completed postal survey of immunisation health professionals from the Hunter New England and North Coast Area Health Services (AHS) in NSW. The target population for the survey was health care providers involved in immunisation. This included general practitioners (GP), practice nurses (PN), early childhood nurses, generalist community nurses, midwives, and hospital nurses from paediatrics wards, neonatal intensive care and emergency departments. Approval was gained from the Hunter, New England, and the North Coast Area Research Ethics Committees.
A range of data informed development of the survey including the findings of previous studies and the in-depth knowledge of local immunisation coordinators (MT, CC, JT, PM).10,12–14 Many of the questions were adapted from a 2002 unpublished survey of health care workers in Western Sydney, NSW and a 1998 survey of health professionals in Quebec, Canada.15,16 The survey was pre-tested with a sub-sample of 11 health professionals from each AHS.
The survey covered five major areas:
1Knowledge regarding vaccine contraindications.
2Beliefs and attitudes about safety and necessity of certain vaccines.
3Practices in recommending unfunded vaccines.
4Practices in communicating with parents about immunisation.
5Resources currently utilised.
Previous research in Western Sydney indicated that, in an area of high vaccine coverage, such as the Hunter New England, 95% of health professionals would be supportive of immunisation.15 We assessed that if only 88% of health professionals in the North Coast were to support immunisation, this 7% difference may influence coverage. To detect this difference with 80% power at a significance level of 0.05, 248 health professionals were required from each area. The sampling procedure involved stratifying by each AHS and sampling each health professional group proportionally to the population distribution of all health professionals within that area. The exception to this was midwives, who were ‘over sampled’ in each area to allow precise estimates for this group alone. After taking into account likely response rates, 507 health professionals from the NC and 419 health professionals from HNE were mailed the questionnaire. For each health professional group, participants were randomly selected from records kept by public health units. Surveys were sent in April 2006 with a follow-up mailout to non-respondents.
Completed questionnaires were reviewed and information entered into an EpiInfo database. Data cleaning, recoding and statistical analysis were performed using SAS v9.1.3. Samples were weighted to take into account non-response and disproportionate sampling of health professionals from each AHS. Where possible, response variables, including 5 point Likert scales for agreement, were collapsed into dichotomous variables. Unless otherwise stated, those reporting ‘agree strongly’ or ‘agree somewhat’ were recoded as ‘agree’ and compared with all others (including ‘disagree strongly’, ‘disagree somewhat’ and ‘unsure’). Where results for early childhood nurses and generalist community nurses were combined a new group was named ‘Community Nurses’.
Point estimates and 95% confidence intervals (CI) were calculated where appropriate. When comparing the two areas, point estimates and confidence intervals were used to determine significance. Non-overlapping confidence intervals were considered significant. When comparing dichotomous variables, an odds ratio was calculated and a chi square test used to assess statistical significance. A p value less than 0.05 was considered statistically significant. Content analysis was performed on text responses and included grouping answers into common themes.
Of 926 surveys sent, 427 were returned completed (47%). Response rates to the survey varied between the AHS and health professionals (Table 1). In comparison with population data on GPs, our survey over-represented females (43% this study versus 33% NSW medical workforce census data).17
Table 1. Survey response rates by area and health professional category.
Area response rates
Hunter New England
Health professional group response rates
Early Childhood Nurses
Generalist Community Nurses
Childhood vaccines in the National Immunisation Program at the time of the survey were considered to be safe by 414 of the 427 respondents (97%), effective by 416 (98%) and necessary by 409 (96%). No significant difference was found between HNE and NC areas. Table 2 shows these findings by professional group.
Table 2. Proportions of health professionals in regional NSW agreeing that childhood vaccines are safe, effective and necessary.
(a) Weighted percentages
(b) Total for each group vary due to missing data
Practice Nurse n=81
General Practitioner n=95
Hospital Nurses n=74
Generalist Community Nurses n=39
Early Childhood Nurses n=30
Respondents rated their level of agreement with a number of commonly held concerns about vaccines (Figure 1). Significant proportions agreed that children get too many vaccines in the first two years of life. Consistently more health professionals in the NC area compared with HNE agreed with these statements of concern; however, these differences were only statistically significant regarding the statement that additives in vaccines may be harmful (23%, [95% CI: 16.9-28.7%] versus 10%, [95% CI: 6.3-14.0%]). When examining these results by health professional category, hospital nurses were more likely to agree or be unsure about all the statements than those in general practice. In addition, midwives were significantly more likely than all other respondents to agree that the additives in vaccines may be harmful (53% versus 36% [OR: 2.0, 95% CI: 1.3-3.0]) and that concurrent immunisation might overload the immune system (56% versus 28% [OR: 3.3, 95% CI 2.1-5.1].
Respondents were asked about the importance of vaccinating young children against specific diseases. Vaccination against polio was most strongly supported (95% agreed completely) followed by pertussis (94%), measles (92%), pneumococcal disease (85%), hepatitis B (84%), meningococcal type C disease (80%), and varicella (79%).
When examining the attitudes of specific health professionals, both early childhood nurses and practice nurses were significantly more likely than GPs to agree completely about the importance of vaccinating young children against varicella (early childhood nurses 94% versus GPs 73% [OR: 5.7, 95% CI: 1.2-26.1] and practice nurses 89% versus GPs 73% [OR: 3.0, 95% CI: 1.4-6.6]). Similarly, practice nurses were significantly more likely than GPs to agree completely about the importance of vaccinating young children against meningococcal C (PNs 98% versus GPs 77% [OR: 2.4, 95% CI: 1.1-5.3]) and hepatitis B (PNs 92% versus GPs 80% [OR: 3.0, 95% CI: 1.2-7.4]).
There was general support for the neonatal hepatitis B vaccine for newborns with overall 83% agreeing it was necessary and 85% agreeing it was safe. While 88% of midwives felt the vaccine was necessary, slightly fewer (81%) believed it was safe and 35% felt it was given too young. GPs had the least support for the vaccine, with only 77% believing the newborn dose to be necessary for disease control.
Immunisation knowledge and beliefs
With regard to true and false contraindications to vaccination, based on the NHMRC Australian Immunisation Handbook 8th edition, the majority of GPs and PNs correctly believed that vaccines could still be given with antibiotic use (86%); temperature of 37°C and runny nose (90%); family history of convulsions (87%); and child's mother being pregnant (89%); and should not be given when the child had a temperature of 38°C and felt unwell (95%). There were no significant differences in these results between areas.
Table 3 shows beliefs about an association between MMR vaccine, inflammatory bowel disease and autism. Only 59% of respondents believed there was no association, with little difference between the two areas. Overall, hospital nurses and midwives had the most uncertainty with 57% and 40% respectively, being unsure. Community nurses were the group most likely to believe in an association with 21% responding “yes”. Of all respondents, only 7% correctly answered that there was an association between MMR vaccine and a rare bleeding disorder, thombocytopenic purpura, despite the established link being listed in the Australian Immunisation Handbook 8th edition.11
Table 3. Beliefs about MMR vaccine among NSW regional health professionals.
Infammatory bowel disease
Idiopathic thrombocytopenic purpura
(a) Weighted percentages
Beliefs about an
association between MMR
vaccine and each condition
Responding to concerned parents and conscientious objectors
Respondents were given a list of all current vaccines on the schedule and asked which ones raised the most questions and concerns from parents and what these concerns were. Of the eight vaccines listed (Table 4), respondents noted the MMR vaccine as most frequently raised by parents, with the majority of concerns about a link to autism. For hepatitis B vaccine, the most commonly reported questions were about necessity and the view that the vaccine was given “too young”. The question of vaccine necessity was the most commonly reported issue regarding varicella, pneumococcal and polio vaccines.
Table 4. Specifc vaccine concerns or questions raised by parents according to NSW regional health professionals.
Childhood vaccine raised by parent
Number of respondents nominating concerns about vaccine raised by parents
In their discussions with parents, most respondents felt confident answering questions about immunisation. Community nurses reported significantly more confidence (99%) than hospital nurses and midwives combined (77%) [OR: 23.0, 95% CI 3.6-146.7]. GPs and PNs combined also reported significantly more confidence than hospital nurses and midwives combined (91%) [OR: 3.2, 95% CI 1.7-5.8]. When parents refused immunisation for their child, only 2% of GPs would refuse to treat the child in the future and 18% would never sign the form required by vaccination conscientious objectors to receive government family assistance payments.
The vaccine schedule
One-third of all respondents felt that adding more vaccines to the schedule would make immunisation too complex for health professionals. This was reported significantly less often in HNE (28%, 95% CI: 22–34%) when compared with NC (43%, 95% CI: 36–50%).
GPs and PNs were asked about their practices and feelings about giving more than two injections at the one visit. The majority said they would give more than two injections if needed (89% for GPs and 91% for PNs), but almost half (46%) of all GPs and one-third (35%) of PNs reported feeling uncomfortable doing so.
During the time that certain vaccines were recommended as part of the Australian Standard Vaccination Schedule but were not publicly funded under the National Immunisation Program, only around half of all GPs would “mostly or always” recommend them (meningococcal C vaccine [57%], pneumococcal vaccine [46%], and varicella vaccine [47%]).
Use of resources
The utilisation of immunisation resources is shown in Figure 2. The Australian Immunisation Handbook 8th edition in hardcopy was rated as useful by 93% of health professionals, while the accompanying CD and online version were less often reported as useful. The second most useful resource was a booklet for consumers, Understanding Childhood Immunisation and telephone advice from the public health unit (PHU), whereas websites were the least used resource.
When examining the popularity of resources by health professional group, 94% of GPs and 95% of PNs found newsletters from Divisions of General Practice (DGP) a useful source of information on immunisation. Between 86% and 100% of community nurses found the local immunisation coordinator, PHU telephone advice, and the Understanding Childhood Immunisation booklet useful.
The highest rate of accreditation training occurred among community nurses, with 93% reporting having undergone or undergoing the immunisation accreditation course. This was followed by practice nurses (69%), midwives (39%) and hospital nurses (26%). Nurses who received accreditation training reported significantly more confidence in answering parental questions about immunisation (p<0.001), compared with nurses who did not receive such training. They were also more likely to answer correctly that a family history of convulsions was not a contraindication to vaccination (p<0.001) but were not more likely to believe that immunisation was safe, effective or necessary (p=0.16, p=0.10 and p= 0.44, respectively).
While GPs wanted more workshops and/or seminars (43%), inservice and/or updates were the most popular form of continuing education for PNs (58%), community nurses (51%), midwives (74%) and hospital nurses (70%).
In this study, health professionals in the region with relatively low immunisation rates (North Coast) were slightly but consistently less confident in vaccine safety compared to the region with higher immunisation rates (Hunter New England). While it is possible the differences in confidence levels may partly explain the differences in immunisation coverage, other differences between the areas may also contribute, including transience, and social and economic disadvantage, in addition to health care utilisation and access. The slight attitudinal differences between areas indicate that health professionals are situated in their community contexts and both reflect and influence community opinion about immunisation. Indeed, an overall figure of 3% of health professionals in this study disagreed that vaccines on the ASVS were safe and this is the same small but significant proportion of parents who actively refuse immunisation in Australia.3
With respect to the study's overall findings on attitudes, the vast majority of health professionals felt vaccines to be safe, effective and necessary. However, despite these attitudes, a significant proportion held unwarranted concerns about the content, timing and number of vaccines given to children, and a smaller proportion believed that there were suitable alternatives to vaccination. These findings are similar to a previous study of health professionals in Western Sydney and to those of studies with parents.6,15 Overall, health professionals tended to have greater support for older, more established vaccines, again similar to studies of parents.18,19 GP respondents, in particular, had limited support for the varicella vaccine which could indicate the need for further education about program rationale.
While studies have repeatedly failed to confirm a link between MMR vaccine, inflammatory bowel disease and autism, our study found a number of health professionals are still unsure about the association.20 That the majority were unaware of an established link between MMR vaccine and thrombocytopenic purpura demonstrated a lack of factual knowledge, or easy access to factual knowledge, a finding consistent with a similar UK study.10 This may indicate a propensity for ongoing media coverage to influence their beliefs and highlights the need for continued education. In addition, timely and targeted information would be of value for health professionals when the media raises their concerns and those of parents.21
The views of midwives regarding vaccination are important. As the health professionals now at the first vaccine encounter for parents, there is the potential for midwives’ attitudes to have an impact on parent vaccine acceptance throughout the rest of childhood. Also, completion of the hepatitis B vaccination series is more likely among those receiving the first dose at birth.22 Midwives in this study had overall confidence in the neonatal hepatitis B vaccine's safety, although more than one-third believed that the vaccine was given too young. In addition, their concern about concurrent immunisation has implications for how midwives would support additions to the infant schedule. Given the low response rate among midwives, further research is needed to determine the extent of such views.
Our study indicated good levels of knowledge about vaccine contraindications and indicated an improvement in knowledge since Herceg et al.'s survey in 1997.13 The persistence of the belief that a family history of convulsions contraindicated vaccination is likely to be a lingering effect of the disquiet over the safety of the no-longer used whole cell pertussis vaccine.23
A recent study found that immunisation providers in general practice are delaying the meningococcal C vaccination at 12 months of age, possibly to avoid giving three injections at the one visit.24 Our finding that nearly a half of GPs were uncomfortable giving more than two injections at the one visit indicates where some of this delay might originate, although the preferences of parents will also be a factor.14
This study is limited due to the low GP response rate but it has raised an important area for closer consideration: while the introduction of multivalent vaccines, particularly in the first year of life, has the potential to alleviate this problem temporarily, new vaccines in the pipeline may make multiple injections a persistent issue for health professionals.
The existence of recommendations for the use of vaccines not funded under the NIP showed that vaccine funding significantly affects the propensity for a GP to suggest vaccine/s to a child. The anticipated availability of new vaccines is likely to continue to challenge policy makers, funding mechanisms, and health professionals in this regard.
The Australian Immunisation Handbook continues to be the most important resource for health professionals. However, this survey shows that local support workers, such as immunisation coordinators, public health units and divisions of general practice, are an important conduit for, and source of, information. In addition, immunisation accreditation training for nurses improved knowledge about one contraindication and confidence in addressing parents’ concerns but did not change attitudes to vaccination.
While this study had adequate response rates for nurses, there was a low response rate among GPs and midwives. This limits the extent to which the findings for these professionals can be generalised and further information is needed. It is possible that non-respondents were less engaged with, or less supportive of, immunisation. Hence, this study may over-estimate confidence in vaccines. However, a previous study with high response rates indicated similar findings.15 Our study may also have limited generalisability to other Australian health professionals involved in immunisation, as it was undertaken in regional NSW. Application of these study methods to other areas is currently occurring.
Health professionals influence parents to immunise or not immunise their children.4–6,25 Therefore, there is value in governments and immunisation support workers continuing their efforts to maintain up-to-date knowledge and support the delivery of targeted information to health professionals.
Current investment in the dissemination of recommendations via avenues such as the Australian Immunisation Handbook and local support personnel should continue. The introduction of more vaccines to the childhood schedule will continue to challenge health professionals in terms of coverage and complexity. This impact needs to be considered in making decisions about future vaccine programs.
As vaccine-preventable diseases have become less common in Australia, the focus increasingly has moved to vaccine safety and necessity. Increased effort in ensuring health professionals are better informed about the benefits and safety of immunisation and are kept up-to-date with changes will help to maintain the currently high immunisation coverage needed to protect Australians against potentially serious vaccine-preventable diseases.
We are grateful to Glenda Lawrence, Nick Wood, Sean Gibney, Peter Eisenberg and Sue Campbell-Lloyd who provided input into the survey content, Christine Staples, Penny Cox and Kylie Parsons who assisted in survey mailouts, Melissa Helferty who undertook data entry and cleaning, Cornelis Biesheuvel and Jennifer Peat who provided statistical advice, and four peer reviewers who provided helpful feedback.
NCIRS is supported by The Australian Government Department of Health and Ageing, The NSW Department of Health and The Children's Hospital at Westmead.