Effects of a web based decision aid on parental attitudes to MMR vaccination: a before and after study

Decision aid development

We developed the pilot MMR decision aid in the first two months of 2004. It was designed to comply, as much as possible, with the CREDIBLE criteria as defined by the Cochrane Review of Decision Aids.9 13The aid provided numerical and graphical evidence of the risks associated with the diseases, alongside the potential risks associated with the vaccine, and provided references for these estimates. We used Australian data where possible, supplemented with evidence from large international epidemiological studies, systematic reviews, and meta-analyses.

Fig 1

Screen from MMR vaccination decision aid: direction of leaning about vaccination

• Download figure
• Open in new tab
• Download powerpoint

Frequently asked questions were addressed, including information about the alleged association between autism and MMR vaccination. The aid listed some key advantages and disadvantages of MMR vaccination, and users were asked to rate the importance of these on a three point scale (“Very important to me,” “Slightly important to me,” and “Not important to me”). Users were also asked to rate their attitudes to MMR vaccination, both before using the decision aid and again afterwards, on a three point scale (“Leaning away from MMR vaccination,” “Undecided,” and “Leaning towards MMR vaccination”). After using the decision aid, users' information requirements for decision making were assessed with four questions (see table 2) on a three point scale (“Yes,” “Unsure,” and “No”).

The aid was written for a reading age of 12 years. We consulted a range of stakeholders about its content, design, and layout, including immunisation and decision aid experts, health professionals, and parents. The aid was located via the website of the Australian National Centre for Immunisation Research and Surveillance (NCIRS) at www.ncirs.usyd.edu.au/decisionaid (figs 1 and 2 show two sample screens from the aid).

Fig 2

Screen from MMR vaccination decision aid: comparison of risks

• Download figure
• Open in new tab
• Download powerpoint

Study design and outcomes

Respondents had open access to the decision aid, which was designed for sequential movement through its 21 screens but with the option to move to any section via a menu. Respondents completed online questions in a stepwise fashion using radio buttons and free text but could proceed without completing questions if they chose. At the end of the aid, respondents were asked to complete demographic details. Using a before and after design, the primary outcome was change in attitude towards the MMR vaccine.

Data analysis

Respondents' attitudes were coded as −1 for “leaning away from,” 0 for “undecided,” and 1 for “leaning towards” vaccination. We calculated respondents' mean score before and after they viewed the aid. We used Wilcoxon's matched pairs signed ranking test (by means of SPSS version 11.5) to determine whether the change in these scores was significant. We cross tabulated each respondent's “leaning” before using the aid against their “leaning” after using the aid. This allowed identification of the number of people who were less favourably or more favourably disposed to the MMR vaccine after using the aid, and the number whose opinion was unaltered.

We dichotomised respondents' perceived importance of the advantages and disadvantages of MMR vaccination as “very important” or other (“slightly important” and “not important” combined). We compared ratings for each advantage and disadvantage with respect to whether respondents were leaning towards MMR vaccination or staying undecided or leaning away from vaccination after viewing the decision aid. We also measured the extent to which people with favourable and unfavourable attitudes to MMR vaccine perceived that their information needs had been met.

Respondents' characteristics

A total of 1277 people accessed the site, with 158 completing the aid by providing responses to set questions. Of these 158 respondents, 62 (39%) supplied demographic information: 55 were parents, 52 with children aged less than 5 years; 53 were aged between 25 and 44 years; and 39 had a university degree. In addition, 42 were from Australia, with the remainder from New Zealand, then the US, UK, Canada, Europe, and Asia. Thirty nine respondents were prepared to be contacted for follow-up by email or telephone.

Attitudes to MMR before and after decision aid

Table 1 shows changes in the 158 respondents' leanings before and after using the decision aid: 48 shifted in a positive direction towards MMR vaccination, and the proportion of respondents “leaning towards” MMR vaccination increased significantly from 61 (39%) before using the aid to 87 (55%) afterwards (P < 0.0001).

Beliefs about MMR vaccination

Table 2 shows respondents' views of a list of advantages and disadvantages of MMR vaccination according to whether they reported a positive attitude towards vaccination or a negative or undecided attitude. As might be expected, participants with a positive attitude towards MMR vaccination were more likely to perceive all of the listed advantages of MMR as important to them than those who remained undecided or were “leaning away from” vaccination. They were also less likely to perceive rare but serious side effects of vaccination as being important and were less likely to have residual concerns about autism and bowel disease.

The most important perceived advantage of MMR vaccination among those with a positive attitude was protection against the serious symptoms of measles, mumps, and rubella (98% rated this as “very important,” compared with 84% of those with a negative or undecided attitude). Other advantages that most respondents with a positive attitude considered very important were the protection of other children who could not be vaccinated for medical reasons (68%, versus 25% of those with negative or undecided attitudes) and protection against common symptoms of measles, mumps, and rubella (63%, v 30%).

Participants who remained “undecided” or “leaning away from” vaccination after using the aid were more likely to consider the disadvantages of MMR vaccination as very important, particularly the chance of rare but serious complications (78%, v 57% of those “leaning towards” vaccination), residual concerns about autism and bowel disease (78%, v 27%), and fear of feeling guilty or responsible should their child have an adverse reaction to the vaccine (77%, v 50%). Other reasons that respondents cited in the free text box for not getting their child vaccinated included concern over mercury in vaccines, interest in alternative types of disease prevention, family experience of serious complications (including autism and inflammatory bowel disorders), and previous adverse reactions.

Among all respondents, perceived disadvantages of MMR vaccination included the unproved link between MMR vaccination, bowel disease, and autism (50% considered it “very important”) and the potential for feelings of guilt or responsibility if harm occurs as a result of MMR vaccination (62% “very important”).

Information needs for decision making

Respondents who were “leaning towards” MMR vaccination after using the decision aid were more likely to feel that they knew their options and that they had enough information to make a decision (table 2). Of those who remained undecided or were leaning away from vaccination, most (78%) said that they knew their options, but 52% still felt that they didn't have enough information to make a decision or remained unsure. Among all respondents, 32/147 (22%) remained negative toward vaccination despite feeling well informed.

Limitations of study

Because of the nature of the internet and the decision aid itself, users were able to access all parts of the aid without completing each section and were not compelled to provide their demographic details. Some users may have felt uncomfortable leaving such information on the website because of concerns over confidentiality. As a result, there is potential for selection bias from the low response rate. This may also have been due to the length of the aid, reflecting the difficulty of providing sufficiently comprehensive information to cover the main risks and benefits of MMR vaccine without being onerous to complete.

In addition, only 52 (84% of those providing demographic information) of the respondents reported having children of vaccination age: the other respondents' beliefs about vaccination may not be relevant and limit the generalisability of our results. Despite the problems of representativeness, online data collection represents a low cost and low maintenance way to obtain evaluative data and has promise if loss to follow-up can be minimised.

Future work

Further research is needed to clarify the effectiveness of the decision aid. Firstly, the aid could be adapted to be more interactive (such as response-dependent summary paragraphs) and to provide more in depth scientific evidence (including access to papers from medical journals, critiques of articles, and meta-analyses) that may be layered according to parents' information needs. Then, a randomised controlled trial of the online decision aid might allow the above methodological limitations to be addressed. Given the difficulty some parents have deciding whether to immunise their children with the MMR vaccine, such a study could examine changes in knowledge, decisional conflict, and anxiety. Of interest to government and immunisation advocates would be whether the aid improved the timeliness and completion of MMR vaccination among children of parents who are concerned about the safety of the vaccine.