Observational study to assess pregnant women’s knowledge and behaviour to prevent toxoplasmosis, listeriosis and cytomegalovirus

Pregnancy complications associated with Toxoplasma gondii, Listeria monocytogenes and cytomegalovirus (CMV) infections are prevented by simple precautions and behaviours of pregnant women [1‐3] (Table 1). Many pregnant women are not aware of the risk and consequences of infectious diseases and are not practicing preventive strategies [3‐9]. Studies from the United States have shown that most women of childbearing age and pregnant women had a limited knowledge of methods to prevent toxoplasmosis, listeriosis and CMV infections [3, 5‐7]. Studies about listeriosis revealed that women who had never heard of listeriosis were more likely to be young, be a single mother, live in rural areas, speak a foreign language at home, or have less formal education. In addition, knowledge levels of listeriosis among pregnant women differ between ethnic groups [5, 10].

There is not much information about the frequency with which pregnant women practice preventive behaviours for toxoplasmosis, listeriosis and CMV infections. While knowledge is an important determinant to establish behavioural change, accurate knowledge may not lead to appropriate preventive behaviour. Attitudes of pregnant women towards changing their behaviour, and their perception about the likelihood of contracting the infectious disease during their pregnancy may also be important contributors to establish behavioural change [24, 25].

The objective of this survey was to assess knowledge of and risk behaviour related to toxoplasmosis, listeriosis and CMV infection in pregnant women, who are cared for by a primary care midwife in the Netherlands, and to determine which demographical characteristics are related to knowledge and risk behaviour. In addition, to determine differences in risk behaviours within sources of received information. We hypothesize that more knowledge regarding infectious disease prevention is associated with less risk behaviour in pregnant women.

To assess pregnant women’s knowledge and behaviour towards infectious diseases we developed a self-administered questionnaire for this observational survey. Survey questions were divided per infectious disease (toxoplasmosis, listeriosis and CMV infections). In total, 50 questions covered respondents’ knowledge of preventive practices in general, risk behaviour, and sources of received information. Questions were based on previous studies [3‐8, 17, 26‐28] and the questionnaire took approximately 20 minutes to complete. The first section of the questionnaire covered knowledge about preventive practices. Respondents were presented with twelve (eight true and four false) preventive practices about the three diseases and were asked to tick an applicable box, including a ‘yes’, ‘no’ and ‘don’t know’ option. Each correctly identified preventive practice contributed to the knowledge score (one point for each correct answer). The knowledge score could therefore vary between zero and six for toxoplasmosis, between zero and three for listeriosis and between zero and three for CMV. The second section focused on the actual behaviour of pregnant women to prevent specific infectious diseases during their pregnancy. Only respondents who had children aged less than five years answered the questions about risk behaviour regarding CMV infections.

This survey was embedded in the DELIVER study, which is a large scale national survey into primary care midwifery in the Netherlands [29]. The DELIVER study took place in twenty primary midwifery care practices across the country and clients were recruited from these practices. Purposive sampling was used to select the midwifery practices, using three stratification criteria: region (north, east, south, west), level of urbanisation (urban versus rural area), and practice type (dual or group practice. Data collection of the DELIVER study took place between August 2009 and March 2011. The study protocol of the DELIVER study was approved by the Medical Ethics Committee of the VU University Medical Centre Amsterdam. Client participation was voluntary and they could withdraw at any time. Privacy was guaranteed in accordance with Dutch legislation. Clients’ anonymity was maintained by using anonymous participant and practice identifiers.

Demographic and pregnancy related information including the respondents’ age, number of pregnancies experienced, nationality, level of education, language spoken at home, living in a deprived area, planned or unplanned pregnancy, single/marital status, folic acid use and smoking behaviour was obtained via linkage to the questionnaires of the DELIVER study by an anonymized participant identifier. Nationality was determined by the respondents’ reported nationality and was dichotomized (Dutch nationality versus non Dutch nationality), because of small numbers of non-Dutch participants. Highest achieved educational level was determined and defined in three groups for analysis: low level of education (medium- level secondary education or below), medium level of education (higher-level secondary education or vocational education) and high level of education (diploma level or university education). The gestational age of the respondent was dichotomized into first or second trimester versus third trimester, because of small numbers of respondents who were in their first trimester. Finally, the number of pregnancies a respondent experienced was categorized into first (primigravidae) and multiple pregnancies (multigravidae).

From October to December 2010 we asked all participants in the DELIVER study, who were still pregnant at the time of survey completion, to fill in the additional questionnaire about toxoplasmosis, listeriosis and CMV infections. The questionnaire was sent to the home addresses of pregnant women with a return envelope. One reminder was sent to all non-responders after one month. Respondents were excluded from this survey if the link with the questionnaire of the DELIVER study that included demographic data could not be made.

A total of 1,663 respondents who participated in the DELIVER study were invited between October 2010 and December 2010 to complete the questionnaire on infectious diseases. In total 1,123 (67.5%) respondents completed the questionnaire. Of these, we excluded 26 questionnaires which could not be linked to the demographic data. Data of 1,097 respondents were included in the analyses, representing a net response rate of 66.0%.

Respondents were more likely to have the Dutch nationality (96.9%) than the general Dutch population (93%), and were more likely to have a high education (56.9%) than the general Dutch population (17.1%) [30, 31]. Other demographic characteristics of the study population are presented in Table 2.

Our observational survey regarding knowledge and risk behaviour related to toxoplasmosis, listeriosis and CMV infections during pregnancy showed that - although there was limited knowledge about specific practices to prevent each of these infections - the majority of respondents reported they practiced appropriate behaviour to prevent toxoplasmosis and listeriosis, but not to prevent CMV.

A strength of this study is its large sample size (n = 1,097) and high response rate (66.0%). This study observed both knowledge about preventive practices and the actual risk behaviour of pregnant women regarding preventable infectious diseases, which has not been conducted by many other studies. However, the study population was not representative for all pregnant women in the Netherlands. The study population included mainly higher educated women and women without the Dutch nationality were under-represented [30, 31]. This could be partly due to the fact that the questionnaire was not presented in other languages than Dutch. It is also possible that women who did not know the answers did not return the questionnaire, which may have resulted in an overestimation of knowledge levels. In addition, the behavioural questions may have contributed to social desirability bias, which may have affected the frequency of of reported risk behaviour. Another potential weakness of this study is that only respondents who had children aged less than five years answered the questions about preventive behaviours regarding CMV infections. By doing this we could have missed important information from other respondents who are also at risk for CMV infections, like women who work in a children-day care setting or other health care setting. It needs also to be mentioned that we considered risk behaviour to be present if a respondent reported one of the included risk behaviours at least once during their current pregnancy. This may indicate that the many respondents who had undertaken a risk behaviour were actually not frequent risk takers, but had a single exposure. In addition, as this study had a cross-sectional design, the interpretation of associations should remain with caution.

This study indicates that health care professionals play an important role in informing women about preventable infectious diseases as many respondents reported having received information about toxoplasmosis, listeriosis or CMV from their health care provider. Books, magazines and the Internet were also important sources of information for the respondents. Most pregnant women in the Netherlands receive a brochure entitled “Pregnant” during their first antenatal care visit, which is developed by several Dutch organizations involved in mother and child care. This brochure includes information on listeria and toxoplasmosis, but not on CMV infections, which could partly explain the general lack of knowledge of the respondents about CMV. Confirmed by other studies, this study revealed lower median knowledge scores for preventive practices for toxoplasmosis among respondents who had less formal education, had an unplanned pregnancy, were single, had smoked during pregnancy and had never worked as a health care professional. For listeriosis lower level of median knowledge score were seen among respondents who were younger than 25 years, had less formal education, had an unplanned pregnancy, were single, experienced their first pregnancy, had smoked during pregnancy and had never worked as a health care professional [5, 7, 10]. In addition to these two infectious diseases, this study revealed that respondents with the Dutch nationality and respondents who spoke Dutch at home, had less knowledge about CMV preventive practices. Women with a non-Dutch nationality may have had more knowledge, and thus be more aware, of CMV infections, because the maternal and congenital CMV prevalence is higher among immigrants than among native mothers [22, 32]. Counter to our hypothesis and confirmed by another study [5], disease specific knowledge was not necessarily associated with preventive behaviour during pregnancy, regarding toxoplasmosis and listeriosis. And conversely, a lack of knowledge was not always associated with engaging in risk behaviour. Infection with toxoplasmosis during pregnancy is highly associated with eating raw or undercooked meat [12]. And while only half of the respondents demonstrated knowledge of this relationship, the majority indicated that they avoided the behaviour. These results are comparable with alcohol consumption during pregnancy. Many women know they should not drink alcohol during pregnancy, but they do not exactly know the effects of alcohol on the foetus [33]. Contrarily, although there was an overall good understanding that toxoplasmosis could be prevented by gardening with gloves, only one fifth of the respondents did garden without gloves during their pregnancy. For six out of the eight included risk behaviours, there were almost no differences in reported risk behaviours between respondents who were or were not aware of the practices to prevent infectious diseases.

Respondents with a higher educational level, who had the Dutch nationality, who did not take folic acid in their first trimester or ever worked in a children day-care setting had greater odds to report a risk behaviour for toxoplasmosis. Respondents who were in their third phase of pregnancy had higher odds to report a risk behaviour for listeriosis, but lower odds to report a risk behaviour for CMV. These factors indicate that health care professionals involved in mother and child care should give more attention to these women with regard to infectious disease prevention. Behaviour change depends on a range of factors, including the perceptions of the threat [34]. It is possible that women with a higher education are more aware of the low risk of contracting an infectious disease during pregnancy. Pregnant women receive large amounts of information during their first prenatal visit, including methods to prevent infectious diseases and as the amount of information increases, cognitive shortcuts could arise. This may imply that it is more effective for health care professionals to inform pregnant women briefly about behaviours and lifestyle habits they should adopt or avoid and that it may not be necessary to give information on specific infectious diseases. In addition, it could be helpful to repeat some information on preventive practices during a later stage in pregnancy, than only during one of the first prenatal visits. Some preventive methods need more emphasis in prenatal health care, because these were not well adopted by pregnant women in this current study. These preventive methods concern hygienic behaviours in general (e.g. not sharing utensils or cups with children and hand washing after diaper change) to prevent CMV infection, washing or peeling raw fruits and vegetables to prevent toxoplasmosis and properly reheating ready to eat foods to prevent listeriosis.

Concerning CMV, respondents who had less knowledge about the preventive practices did not report more often risk behaviour than respondents with knowledge about the preventive practices. There was a general lack of knowledge, illustrated by the fact that only one eighth of the respondents had ever read, seen or heard anything about CMV and the majority of respondents did not adopt methods to prevent CMV infections. Women within certain professions such as children-day-care workers have a 5 to 25 fold higher risk of acquiring a primo CMV infection during their pregnancy compared to women not in contact with young children [35, 36]. However, this study did not find differences in behaviour between respondents who ever worked in a health care or in a children day-care setting and respondents who did not. An explanation for the general lack of knowledge and lack of adopting behaviours towards CMV infection prevention is that health care workers pay little attention to CMV infection [21, 27, 35]. Another study showed that Dutch doctors involved in mother and child care had suboptimal knowledge on CMV themselves, and they seemed to underestimate the prior risk for a child with congenital CMV infection in their practice [27]. However, it may be important for health care professionals to give information on CMV prevention to pregnant women in the future as a study in France showed that simple information on basic hygiene measures given to women at the beginning of their pregnancy could significantly reduce the incidence of maternal infection during pregnancy [36, 37]. In addition, a recent study showed a congenital CMV birth prevalence rate of 0.54% in the Netherlands [22], which is higher than the birth prevalence rate of 0.09% [32] showed in an earlier study and on which many Dutch professional educational materials and guidelines are based [22].

Pregnant women seemed to appropriately avoid risk behaviour without exactly knowing why they avoid it. This could reflect the use of cognitive shortcuts, where complex tasks are reduced to simpler operations which allows people to make rapid, efficient, but sometimes irrational choices [38, 39].

Some studies suggest that written education is less effective to establish behavioural change than when health care providers inform clients orally about correct behaviour [40‐42]. However, this study did not find any difference in the occurrence of risk behaviour between those we received information on the infectious disease from their health care professional and those who received information through other sources. It would be interesting to investigate what kind of information health care professionals involved in mother and child care give orally and what kind of information they give in written materials about preventable infectious diseases.

In conclusion, a substantial part of the pregnant women in our study had never heard of listeriosis, toxoplasmosis or CMV, or did not know how to avoid these infections during pregnancy. However, many pregnant women were appropriately avoiding risk behaviours, without knowing what they are avoiding. It remains important that health care providers continue advising pregnant women about behaviours and lifestyle habits which can prevent infectious diseases, but it may be less important to inform pregnant women about specific infectious diseases. In addition, other sources of information about the prevention of infectious diseases in pregnancy must be complete and adequate. In general, more attention towards CMV infection prevention is necessary to improve the knowledge and adoption of the behaviours to prevent CMV infections among pregnant women.