Influenza vaccination during pregnancy: a qualitative study of the knowledge, attitudes, beliefs, and practices of general practitioners in Central and South-Western Sydney

Pre-natal influenza infection is associated with an increased risk for both mother and baby, including respiratory and cardio-pulmonary hospitalisation, pre-term delivery, fetal distress, and in some cases death [1‐7]. These risks are compounded when the mother has comorbidities such as asthma or diabetes, or is infected with a pandemic strain of the virus [4, 8]. The most effective strategy for preventing influenza in pregnant women is immunisation, and benefits for both the mother and the infant have been demonstrated, with maternal immunisation significantly reducing respiratory illnesses in both the women and their infants in the first six months of life [9‐11]. Influenza vaccination during any trimester of pregnancy is considered safe for both the mother and the foetus [12‐15]. It is recommended that all pregnant women in Australia receive the vaccine, [16‐18] and these guidelines have been in place for more than six years. The vaccine is available free of charge to pregnant women in Australia. However, the proportion of pregnant women who receive the vaccine is low, with coverage between 10 and 40% [19‐23].

While many factors may influence influenza vaccine uptake during pregnancy, it has been identified that pregnant women who receive a recommendation for the vaccine from a health care provider are more likely to receive the vaccine [22, 24‐27]. However, antenatal care providers do not routinely recommend influenza vaccination to their pregnant patients: three recent surveys of pregnant and post-partum women in Australia found that only 21–41% of women could recall receiving such a recommendation [21‐23]. Where a recommendation was given, approximately half were by a general practitioner [22, 23].

Antenatal care providers are more likely to recommend the influenza vaccine to pregnant women if they have good knowledge about influenza and influenza vaccination during pregnancy, positive attitudes towards influenza vaccination during pregnancy, have observed serious medical conditions due to influenza, or have personally received the influenza vaccine [22, 28‐30]. These studies, which investigated the knowledge, attitudes, and practices of antenatal care providers towards influenza vaccination during pregnancy, have mainly relied on data gathered using written survey techniques in antenatal care settings such as a hospital or clinic. No studies have been undertaken in the Australian primary health care context, and little is known about the perceptions of general practitioners who are the main provider group or avenue through whom pregnant women can access influenza vaccination. The attitudes and practices of general practitioners are therefore likely to have a significant influence on influenza vaccine uptake during pregnancy.

In early 2012, strategies to improve awareness about influenza vaccination during pregnancy among general practitioners working in central and south-western Sydney were implemented. The local director of public health with the directors of obstetrics wrote to all general practitioners in the region reminding them of the recommendation to vaccinate pregnant women for influenza, and referring them to a recent evidence-based statement [18]. They were also provided with multiple copies of a brochure designed to encourage women to be vaccinated, [31] and a reminder stamp about the vaccine was placed in patient-held antenatal care cards.

The aim of this study therefore was to investigate the knowledge, attitudes, beliefs, and practices of general practitioners in the Sydney and South-Western Sydney Local Health Districts in Australia towards influenza vaccination during pregnancy, using qualitative methodology. The qualitative approach was used to provide rich information on the perspectives of the general practitioners, to identify their practices regarding influenza vaccination during pregnancy, and their knowledge, attitudes, and beliefs that inform these practices.

A qualitative descriptive methodology [32] was used to describe general practitioners knowledge, attitudes, beliefs, and practices regarding influenza vaccination during pregnancy, and to generate knowledge that can be applied to practice.

Purposive sampling [33] was used to select general practitioners to ensure diversity in terms of the location of the general practice, the practice size, and the practitioner’s sex. A matrix was developed to map these characteristics to guide participant selection informed by a detailed list of all 666 general practices in the Sydney and South-Western Sydney Local Health Districts. The matrix fields included size of the practice (a large practice is where there are 3 or more general practitioners), type of local government area (urban and rural) and sex of general practitioners within a practice. From this matrix, 44 general practitioners were selected to be invited to participate in the study, to ensure that sex, practice size and type of local government area were evenly represented. It was estimated that twenty participants would be required to reach saturation, therefore with an estimated participation rate of less than 50%, forty-four general practitioners were selected to be invited. A low participation rate was expected, due to the high workload and competing demands faced by general practitioners which may limit their availability to participate.

The 44 selected general practitioners were initially contacted by letter, and subsequently by telephone, to invite them to participate. A semi-structured interview guide, using open-ended questions, was developed and consisted of the following broad topic sections: reflections on antenatal care in general practice, the risk of influenza during pregnancy, attitudes and practices about influenza vaccination during pregnancy, perceptions of women’s attitudes towards influenza vaccination during pregnancy, and reflections on how to improve and record coverage rates of influenza vaccination during pregnancy. Within each section of the interview topic guide, more detailed questions and specific probes were prepared to allow the discussion to develop. The interview questions were piloted with two non-participating general practitioners and minor modifications were made to question expression. Semi-structured interviews were conducted in person with all consenting general practitioners in their place of practice by the one researcher (LM) in September 2012. The duration of each interview ranged from fifteen to forty-five minutes. Interviews were audio-recorded and transcribed verbatim by an independent transcriber.

Four researchers (LM, KW, AD, KH) conducted the analysis. Transcripts were read and re-read by the four researchers several times who then met to discuss the emergent concepts, themes and issues across the dataset. Consensus was reached on the main themes and categories to be included in the analysis, based on the aim of the study, and a conceptual framework was developed to capture this. The researchers undertook a closer analysis of each transcript using the conceptual framework to direct this. The qualitative research software tool N-Vivo was employed to code the transcripts using a constant comparative method [34] and each researcher generated a diagrammatic model to illustrate their coding and patterns. These models were then synthesised and guided the reporting process.

This project was approved by the Sydney Local Health District Human Research Ethics Committee (RPAH Zone).

Of the 44 general practitioners selected, 17 agreed to participate in the study: nine females and eight males; ten from small and seven from large practices; and six from Sydney Local Health District and eleven from South Western Sydney Local Health District. The reasons for non-participation were not ascertained.

Thirty main codes were identified, and grouped into three main categories: (1) general practitioners’ risk perception of influenza infection during pregnancy; (2) general practitioners’ knowledge and perceptions about influenza vaccination during pregnancy; and (3) general practitioners’ approach to promoting and providing influenza vaccination during pregnancy.

Many of the general practitioners interviewed in this study demonstrated limited knowledge about the risks of influenza infection and vaccination during pregnancy, or expressed limited confidence in the safety of the vaccine. These general practitioners may therefore either not be recommending the vaccine, or recommending it with varying levels of clarity and conviction. This may impact upon the fact that only a few general practitioners were able to report high acceptance of the vaccine among their pregnant patients. Our research findings concur with the results of previous studies which report that other antenatal care providers have varying levels of knowledge about influenza vaccination during pregnancy [22, 28‐30, 36]. This is significant, as increased levels of provider knowledge are associated with higher rates of influenza vaccination in pregnant patients [28].

This is the first study to ascertain the attitudes, perceptions, and practices of Australian general practitioners towards influenza vaccination during pregnancy, and the qualitative methods used enabled us to capture rich information about their perspectives which could not have been garnered from a written survey. Semi-structured interviews were identified as the most appropriate data collection method to minimise participant burden and maximise the participation rate, as the interview could be conducted in the general practitioner’s workplace. It was anticipated that the participation rate for focus group discussions would be low. The limitation of this study is that only 17 of the 44 general practitioners invited to participate in the study agreed to participate. A greater number of interviews may have provided greater diversity of practice types. The responses of the general practitioners may have been affected by the fact that the interviewer is a NSW public health employee, however the findings did not indicate that they were reluctant to report practices and attitudes that were contrary to current guidelines.

In our study, general practitioners appear to base their overall risk assessment on their perceptions of the risk of influenza infection during pregnancy, the benefits and risks of the influenza vaccine, as well as their personal experiences of influenza. General practitioners appear to perceive the risks associated with maternal influenza infection to be lower than the evidence suggests, while conversely, their perception of the risks associated with the vaccine seem to be higher. Based on their risk assessments, general practitioners appear to be demonstrating two decision-making biases which influence whether and how they recommend the vaccine. Firstly, omission bias (choosing to do nothing with some probability of harm over doing something with a lower probability of equivalent harm) is demonstrated with general practitioners choosing to not recommend influenza vaccination, even though the probability of harm from influenza during pregnancy is higher than the probability of harm from the vaccine [15, 37]. Omission bias is greater when practitioners anticipate that they will feel regret should an adverse outcome occur. However practitioners may anticipate a greater sense of regret if they perceive that the outcome could be the result of their own action rather than lack of action [38, 39]. The general practitioners interviewed in our study clearly indicated anticipating regret (and fearing liability) should an adverse outcome occur due to the vaccine. Omission bias has been observed in parents deciding whether or not to vaccinate their child against pertussis, [40] and in practitioners deciding whether or not to recommend hormone replacement therapy for women [41]. Secondly, ambiguity bias (avoiding an option when information about the consequences is perceived to be missing [36]) is also observed here, with some general practitioners perceiving that more time is required to demonstrate that the use of the vaccine during pregnancy is safe. Both these decision-making biases may be limiting the extent to which the general practitioners provide influenza vaccination during pregnancy, and strategies aimed at improving the frequency of influenza vaccination to pregnant women by general practitioners should address these biases.

While the majority of the general practitioners reported intermittently recommending the vaccine, they reported varied levels of acceptance of the vaccine by their pregnant patients, which may be related to the quality, style, and consistency of the recommendation they give. This may in turn be related to our finding that these practitioners often lack confidence in their knowledge of the evidence, or in the safety of the vaccine for pregnant women. Many of the general practitioners reported that they are adopting the principles of patient informed choice – they are reluctant to provide a strong opinion or recommendation to their patients, often due to their own lack of confidence in the safety of the vaccine and their fear the consequences of liability should something go wrong. Those interviewed stated that they believe it is the patient’s decision whether they receive the vaccine or not, and pregnant women are therefore expected to decide for themselves. However, an informed decision requires relevant good quality information, [42] and it appears that pregnant women may not receive clear information from their general practitioner on the risks of influenza, and the benefits and risks of the vaccine. A qualitative survey of post-partum women in Switzerland indicated that they perceived that the information they received about vaccination from a provider, when given, lacked unequivocal advice and a sure recommendation [43]. Significant challenges have been reported regarding patient informed decision making - even when good information is provided patients face difficulties, especially when making complex decisions or when they are required to weigh benefits and risks [44]. In relation to influenza vaccination during pregnancy, women have indicated that they feel it is the physician who should explain the choice of whether to be vaccinated or not, and if the message was clear and unequivocal then they would follow the recommendation [42]. Even women who have safety concerns about the vaccine still indicate that they would accept it if the provider recommended it [23].

These interviews were undertaken approximately six months after the implementation of a number of strategies aimed to increase awareness of maternal influenza vaccination during pregnancy among general practitioners, including a letter and a brochure, however only a small number of interviewees recalled the information contained in these documents. A survey of post-partum women in the same region conducted concurrently with this study found the rate of vaccine uptake among pregnant women to be 25% during the 2012 influenza season, and that only 30% of women could recall receiving a recommendation for the vaccine from an antenatal health care provider during their pregnancy [45]. This suggests that the strategies undertaken were not sufficient to increase rates of provider recommendation and vaccine uptake to acceptable levels. Different strategies may be required that more clearly address the risk perceptions of general practitioners identified in this study, and as many interviewees raised the issue of requiring more time to increase their confidence, it may be expected that changes in practice regarding this issue may not occur quickly.

We have not located any research that has investigated how to improve rates of vaccination recommendation in general practitioners across a region; however a number of studies conducted at single antenatal clinic sites have demonstrated that it is possible to increase provider knowledge about influenza vaccination during pregnancy, rates of provider recommendation or patient acceptance of the vaccine through strategies targeting providers, pregnant women, or both. Strategies targeting women include displaying posters in the clinics and providing patients with information brochures, and strategies targeting providers include provider education programs, reminder stamps in patient’s files, e-mail reminder to providers, and making the vaccine available in the clinic [21, 46, 47]. In an obstetric hospital in Melbourne the implementation of a combination of these strategies saw influenza vaccination coverage increase from 30 to 40%, and rates of provider recommendation increasing from 37% to 62.5% [21]. The inclusion of an electronic best practice alert about influenza vaccination in each patient’s electronic medical record in an antenatal clinic in Wisconsin saw vaccination coverage improve from 42% to 61% and provider recommendation rates improve from 49 to 89% [48]. This strategy could translate well to the general practice context in Australia, where electronic decision support for cardiovascular disease management in general practice shows promising potential to increase provider adherence to best practice guidelines [49]. A number of general practitioners in this study reported that influenza vaccination during pregnancy is not always their key priority and they often forget to mention it, and some suggested an electronic reminder could assist with this. Others also recommended that if women were more aware and knowledgeable about the vaccination it would make their job easier to convince them, and suggested that a public awareness campaign could assist with this.

General practitioners have varied knowledge, attitudes and beliefs about influenza vaccination during pregnancy and this is likely to influence both the number of recommendations given to pregnant women, and the conviction with which these recommendations are made. This study indicated that general practitioners have a low perception of risk of influenza during pregnancy, and have considerable concerns about vaccine safety and potential liability, and additional research could further investigate these issues. Providing clear information and communication to general practitioners on the current recommendations regarding influenza vaccination during pregnancy, and the evidence upon which they are made, may increase provider knowledge and confidence in recommending the vaccine, which is likely to improve influenza vaccine coverage among Australian pregnant women.