Work in progress (paper three of the PhD)

in #science6 years ago

I’m stuck on one of my papers, so that’s what I thought I’d write about today. You get to see a little about how the science sausage is made and, hopefully, by explaining this paper to you lovely people I can figure out what exactly all this nonsense means!

As I’ve mentioned a few times now, I’m a research psychologist (i.e. not the type that asks you how stuff makes you feel). I study the acceptance, hesitancy and refusal of vaccination. I don’t do any of that squidgy biological science that goes into the actual development and evaluation of vaccines (although I do work closely with those that do), instead I look at public perceptions and beliefs towards the evidence based recommendations that are arrived at through their extensive work. If you’re interested in my personal views on vaccination I wrote a post a while back on this topic: Why I choose to fully vaccinate

The PhD

My PhD focuses on one small aspect of the Psychology of Vaccination: Vaccine information seeking behaviour.

Most people vaccinate automatically when their doctor recommends a particular vaccine. I’m interested in what happens when, instead, people actively weigh up the pros and cons of vaccination and seek out additional information to aid in the decision making process. If people seeking information what information are they looking for, what do they find and does this information have an effect on whether they vaccinate or not? This seeking can involve asking questions in a consultation with a health care professional, searching online, talking to friends and family members, watching TV news or any other way of gathering information.

I don’t personally design intervention strategies to increase vaccine uptake rates at this point in time but it’s conceivable that some of my research will eventually be used to inform vaccine communication strategy in the future (if that is I can get this damn paper finished)

The Paper

Currently I’m in the process of writing up a systematic review (see the blue box for a simple overview of systematic reviews) on vaccine information seeking and scanning behaviours.

Vaccine information seeking behaviour is when people actively put effort into collecting information on a particular vaccine or vaccination in general in order to aid in their decision making. Vaccine information scanning behaviour, on the other hand, is when people passively acquire information about vaccination during the course of their everyday life.

With this paper, we want to find out:

  • How other academics have studied these concepts in the past.

  • How often people seek additional information about vaccination.

    • Where they are seeking this information from?

    • Does this seeking influences their final decision to accept of refuse a particular vaccine?

With this in mind, we set out to find the peer-reviewed papers that could help us answer these questions. First, we went wide and found around 700 papers that might be of use to us, we whittled away at these until we were left with 34 papers that were exactly what we were looking for. After reading and making extensive notes on these papers, we found that they referenced a number of other great paper that we didn’t find in our initial search so we tracked those down and added them to our big stack of papers, and also by that point a few more had been published so we added them to the pile too. This lead to a final count of 54 papers that all touched on our topic of interest.

That last paragraph took around two and a half months, and this is quite a small review

The Findings

Prevalence

Trying to make sense of this many papers on a similar topic can be tough. Rarely do any two papers examine the same concept in exactly the same way. Each paper usually has its own spin on a topic and may not even be interested in the same outcome as you are.

We got quite luckily here, firstly 53 papers is a somewhat manageable number of papers and secondly most of the papers were surveys that examine the vaccine decision-making process for a specific vaccination (e.g. the HPV vaccine or the 2009 H1N1 influenza vaccine) or childhood vaccination in general. The rest were either interviews or focus groups related to the vaccine decision-making process or analytics of online search engine use.

Several of the survey studies reported their findings in such a way that we could pick out data indicating how many participants searched for information about a particular vaccination. From those papers we put together this forest plot:

The size of the box represents the number of participant that were in each of the study (this ranged from around 100 to 4000), their position across from left to right represents the percentage of the participants that actively searched for information about vaccination. (Note: this is likely not the final version of this image and should be taken with a handful of salt for the time being)

This was a bit of a shock for us, it seems that a big chunk of people are seeking out information about vaccination to inform their decision making. The stand out result here is for the vaccination that was available during the H1N1 influenza outbreak during 2009. It would make sense for this to generate such high levels of seeking, the news media at the time gave the outbreak a great deal of coverage. This likely prompted people to look for more information about the vaccine as a form or protection from the virus (in fact this has been supported by some of the interview papers we have in our review).

If we put the H1N1 vaccine to one side for a moment it would seem that still around 40% of people are seeking out information about vaccination. Interesting, not sure what to do with that piece of information at the time being but interesting none the less.

From these papers, we can also look into the numbers a little deeper and find out what factors predict searching for information about vaccination. I won’t go into this in much depth here but the most common factor that shows up in these studies is level of education. The higher people’s qualification level the more likely they are to seek out additional information during the vaccine decision-making process.

Fewer studies investigated vaccine information scanning (the passive acquisition of vaccine information). Probably because it’s much harder to measure. When studies do it seems to be more common than the seeking behaviour, for the H1N1 Vaccine and the HPV vaccine at least, which would make sense. With only around 5 studies in our review touching on this topic I think we could learn a lot from studying this process of passive information acquisition in some more detail at some point in the future.

Channels and sources

Ok, this is where I got stuck for a while earlier in the month. I think we’ve got two separate concepts being talked about in these studies that are important that I separate out for, information channels and information sources.

An information channel is the means through which information is acquired (e.g. Broadcast media, the internet or friend and family members). An information source is the specific originator of information (e.g. The World Health Organisation, Natural News or The Guardian).

Most studies don’t make this distinction. When they ask their participants where they sought out information from and give options for participants to choose from the studies just sort of mush sources and channels together. So, one of the main points I want to make with this paper is that there is a difference between the two concepts, especially when it comes to the internet. Under our definition, the internet is not a source of information, it’s a channel. Sources exist on the internet, some of them are reliable some of them are not so reliable, some of them may even be friend and family members if social media is taken into account. The internet therefore isn’t really a source of information, and treating it as such probably doesn't give us any useful findings as it will mean something vastly different depending on who is asked. It’s a minor point but some more nuance here in the future may make our measurement better (although this distinction does make for a complete pain when trying to write about the subject).

When it comes to channel/source use this is how the different vaccines compare:

This is 14 studies that asked people to report various channels/sources used when making a vaccine decision. I’ve re-categorised, taken a weighted average of each and separated them out by vaccination (HCS = Health Care System, HCP = Health Care Professional). Again, this is likely to change before the final publication comes out so don’t get overly attached to it.

Some nice little findings in here, the difference in broadcast media usage between childhood vaccines and other vaccines for instance. Also internet use for maternal vaccinations compared to other vaccines, although this finding is based on just the one study, so potentially a good indication for more research being needed here (which luckily I have coming along in my next study).

Just as with the prevalence of the vaccine information seeking behaviour, the use of many of the channels and sources of information can be predicted through the measurement of other variables. The main finding here looks to be attitude towards vaccination. How people feel about vaccination appears to relate to the use of channel/source for information acquisition. This could mean a few different things:

  1. The use of certain channels/source of information are having an effect on attitude towards vaccination

  2. Peoples attitude towards vaccination is effecting their choice of channels/sources for information acquisition

  3. Both

It’s very likely both to some degree but it’d be very interesting if we could pick out a cause and effect from this correlation at some point as I feel the second point isn’t something that has had much attention in the literature.

Impact

So what does all this searching actually mean for uptake? Well it appears to be a bit of a mixed bag from what I’ve seen so far. Here is another forest plot that I’ve been having a play around with lately (its still not as clear as I'd like it to be, much more work needed I think. Also may just drop it.

The plot is separating each study into two data points, the percentage of those that sought information and accepted the vaccination and the percentage of those that sought information and rejected vaccination (Important note: this is not the percentage of people that accepted or rejected vaccination, the vast majority are accepting the vaccine in each study).

It looks like in some studies seeking information about the vaccine relates to people being more likely to vaccinate (e.g. Health care professionals for their seasonal influenza vaccination) while in other studies seeking information about vaccinations relates to people being less likely to vaccinate (e.g. Brunson et al, childhood vaccination).

There are many other factors going on here for sure. My next step is to dig into these relationships in more depth, I have a strong suspicion the interviews and focus groups will flesh these finding out nicely.

And then at about this point the paper trails of into nothingness…

What’s next then?

Thanks for letting me write at you all for the last almost 2000 words, I know what I have to do now! It’s the same thing I have to do literally every time I get stuck on a systematic review, re-read every one of the papers this time with the gaps in the review fresh in my mind. Almost there now, I hoping to have a draft ready by the end of the month. Although I did say that last month.

---------------------------------------------

Sorry this post was more for my benefit rather than yours today, I’ll try and write something of actual interested next week.

About me:

My name is Richard, I blog under the name of @nonzerosum. I’m a PhD student at the London School of Hygiene and Tropical Medicine. I write mostly on Global Health, Effective Altruism and The Psychology of Vaccine Hesitancy. If you’d like to read more on these topics in the future follow me here on steemit or on twitter @RichClarkePsy.

---------------------------------------------

References

I’ll give you some of the references, seeing as you’ve been very good to read this far, but there’s no way you’re having them all. I’m afraid that would be level of academic rigour that I wrote this post to avoid today!

Brunson EK. How parents make decisions about their children’s vaccinations. Vaccine [Internet]. 2013;31(46):5466–70. Available from: http://dx.doi.org/10.1016/j.vaccine.2013.08.104

Bults M, Beaujean DJMA, Richardus JH, Steenbergen JE van, Voeten HA. Pandemic influenza A (H1N1) vaccination in The Netherlands: Parental reasoning underlying child vaccination choices. Vaccine [Internet]. 2011;29(37):6226–35. Available from: http://dx.doi.org/10.1016/j.vaccine.2011.06.075

Cheung EKH, Lee S, Lee SS. Pattern of exposure to information and its impact on seasonal influenza vaccination uptake in nurses. J Hosp Infect [Internet]. 2017;97(4):376–83. Available from: https://doi.org/10.1016/j.jhin.2017.08.005

Harmsen IA, Doorman GG, Mollema L, Ruiter RAC, Kok G, Melker HE De. Parental information-seeking behaviour in childhood vaccinations. 2013;1–10.

Jung M, Lin L, Viswanath K. Associations between health communication behaviors , neighborhood social capital , vaccine knowledge , and parents ’ H1N1 vaccination of their children. Vaccine [Internet]. 2013;31(42):4860–6. Available from: http://dx.doi.org/10.1016/j.vaccine.2013.07.068

Kim S, Real K. A profile of inactive information seekers on influenza prevention: a survey of health care workers in Central Kentucky. Heal Inf Libr J [Internet]. 2016;33(3):n/a-n/a. Available from: http://doi.wiley.com/10.1111/hir.12132

Manika D, Ball JG, Stout PA. Factors Associated with the Persuasiveness of Direct-to-Consumer Advertising on HPV Vaccination Among Young Women Factors Associated with the Persuasiveness of Direct-to-Consumer Advertising on HPV Vaccination Among Young Women. J Health Commun. 2014;19(11):1232–47.

Stevens CF, Caughy MO, Lee SC, Wendy P, Tiro JA. Does language moderate the influence of information scanning and seeking on HPV knowledge and vaccine awareness and initation among Hispanics? Ethn Dis. 2013;23(1):95–102.

Walter D, Böhmer MM, Reiter S, Krause G, Wichmann O. Risk perception and information-seeking behaviour during the 2009/10 influenza A (H1N1) pdm09 pandemic in Germany. Eurosurveillance. 2012;17(13):1–8.

Weiner JL, Fisher AM, Nowak GJ, Basket MM, Gellin BG. Childhood immunizations: First-time expectant mothers’ knowledge, beliefs, intentions, and behaviors. Am J Prev Med [Internet]. 2015;49(6):S426–34. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0749379715003529

Wheeler M, Buttenheim AM. Parental vaccine concerns, information source, and choice of alternative immunization schedules. Hum Vaccin Immunother [Internet]. 2013;9(8):1782–9. Available from: http://www.tandfonline.com/doi/abs/10.4161/hv.25959

Sort:  

Just a suggestion, but look at using ANOM (Analysis of Means) charts for proportions instead of the forest plots. ANOM is a nice (and better) alternative to ANOVA that permits multiple comparisons to overall system performance. I use the techniques in Nelson's book: The Analysis of Means: A Graphical Method for Comparing Means, Rates, and Proportions. Since I am sometimes dealing with over 100 comparators on a graph, I also sometimes format the plot as a funnel plot. I generated a file of critical values for several different levels of alpha, since the tables in the books stop at a small number of comparators. In any case, I can send you a Tableau Public workbook that steps through all of this and contains all of the limit calculations (based on the normal approximation). I had to go back through several decades of papers to work out all of the details for the limits, but the bulk of the technique is in the book above. You can also use the R package ANOM.

Ok cool, thank you for this! I’m certain that this is some truly spectacular advice you have give here. But I'm two pints into my Friday evening so I'm afraid all I can give in reply at the moment is my gratitude. This looks like some good options for to look into on Monday. I've not yet ventured into R much yet but I it's probably best to learn how to do this properly now while I have the time. Thanks again, i really appreciate your input.

Two pints on a Friday evening is definitely not a statistical outlier!

It is here in the UK, I'm well below the mean right now!

Hey @toddrjohnson
Here's a tip for your valuable feedback! @Utopian-io loves and incentivises informative comments.

Contributing on Utopian
Learn how to contribute on our website.

Want to chat? Join us on Discord https://discord.gg/h52nFrV.

Vote for Utopian Witness!

Hey, cool look at that! So happy to see you get rewarded for your advice here. This platform is crazy good at incentivising a great comment section!

Yes, that was a nice surprise and quite helpful. I also got a huge boost from @postpromoter on one of my recent posts as part of their curation efforts. That has finally pushed me above 15 SP with no more SP delegated to me. I'm also glad to see your post getting lots of support too.

That's excellent, congratulations, really glad to hear that!

Yes, this one was a big surprise. I didn't try to push it with any paid promotion or anything. It just goes to show the #steemSTEM effect in action.

I've read about #steemstem and started to use the tag, buts it's not clear from their posts if I have to do something more like I did to apply for #steemitbloggers.

Hi there @toddrjohnson & @nonzerosum (thank you for the information provided :))

You do not have to apply to the @steemtem, you need to write STEM (Science, Technology, Engineering, Mathematics) and follow certain quality guidelines

I invite you to read this link for more information

In case of any questions, feel free to join us @steemstem on discord or to take a look at the FAQs

Greetings,

Katerina

Mentor_3.png

Thanks for the pointer to the detailed information. If you need more curators let me know. I've been a researcher and teacher for several decades.

We would be delighted to meet you on discord too : )

In (the unlikely) case you do not know what discord is, it is a chatting platform that allows users to communicate, send direct messages etc

You may find more info here

Thank you again for taking the time to reply!

Katerina

I've not come across the steemitbloggers tag before but the main requirements for the stem tag is 1. Write something write something on science, technology, engineering or maths. 2. Make it accessible and interesting. 3. Reference your sources used. 4. Don't use copyrighted images. 5 tag it #steemstem. Also interact with the community, which you already do a lot of anyway.

Their @steemstem account has linked to lots of resources and style guides in the past but what I've listed here are the basics. I try to only use it when I've got something that fits closely with their existing content as I don't want to overload them with irrelevant content to sift through.

Well, that's the problem. People are looking for the reasons to reject vaccination. So there is the "ecological niche" for false info. And currently, there is no mechanism to censor the sources of false news.

And what scares me is the fact that the truth is not important as belief.
It means that we can be well informed with the false news and make terrible but still very democratic and "free will" decisions

Yeah I think that's very likely the case, I've a survey that I'm going to write up after this that's looking like it might support something along those lines. That said it does work both ways, those that want to be reassured about the vaccination also look for the information that they want to see. This is why I find these results fairly encouraging, especially when it comes to the HPV vaccine. With so many people searching for information you'd expect all the anti-vaccination websites to have more of an effect, but with searching like this it's likely only those that wouldn't vaccinate in the first place that are paying them much attention.

Trusted and trustworthiness of sources also play a big part here. That's the way to avoid fake news. Although, of course all scepticism goes out the window when information back up your pre existing beliefs. What an interesting century this is going to be!

For the free will aspect I have very little comfort to give there. It's probably best not to think about it too much.

I noticed something interesting while fighting against creationists. There are the same arguments, over and over again, even after being debunked 1000 times (like Nebraska man). Do we know what is the "silver bullet" that can kill false ideas?

Yes, making sure you can replace the myth with a fact.

Just debunking a misconception is never enough, something needs to be put in the hole where the myth once was. I swear I could have done more good by studying the causes of autism rather that the work I do now!

Also the usual be nice and try and act like you've not heard the same shit 1000 times before. It's not easy, I'm still not very good at it and always end up going back to Phil Plait’s lecture don't be a dick to remind me self why I can't just lose my shit.

It does work it just might be a little bit down the line at a point where you're not there to see it.

Thanks for letting me write at you all for the last almost 2000 words, I know what I have to do now!

We call this 'rubber duck programming'.

Also, I can't quite put my thumb on it, but there's some relationship here that makes me think that reading up on Xerox Parc's work on information foraging (pdf link) would be helpful.

Ha, yes thats it exactly!

Thanks, there is a lot going on in that pdf, some of it is definitely stuff I'll look into. I've gone through a fair bit of information science and user studies now. I landed on the comprehensive model of information seeking for much of my work here in the end, but it’d be great to build other similar theories in at some point.

Hi

Thank you for your comment on my latest blog related to AI and fake/real news. I appreciate it.

Just checked your profile just to realize that we seem to share a number of interests :)

In particular that we both share a similar passion towards cryptocurrencies and blockchain technology :)

big fat upvote on the way! :)

ps. are you still writting for @originalworks contest sometimes?

I will follow you closely :)
Take care, Piotr

Hi Piotr, yes I've got far more into it since I started blogging here. I'd like to do more for the @originalworks contest in the future, I learnt a lot from the FarmaTrust one that I took part in. I'm waiting for another healthcare project to come along so that it crosses over with my other interests. Thanks, I will do similar for your work also!

Im glad to see that you're so very responsive @nonzerosum

Check out @mediaworks. They also create interesting contest. They may even be related to @originalworks.

All the best
Piotr

Great post and good luck with your research. Related to your older post. All you have to do is spend a bit of time in the developing world and see people with Polio to realise that vaccines for some of the big problems (e.g. polio) are almost non-negotiable. Totally preventable and some groups, like the International Rotary club, are doing good things to reduce polio all over the world. Preaching to the converted? Likely, but basically saying that I support your cause.

Thank you! 😀 Yes but that's the issue, many people don't get to see the amazing affect the vaccine has for a population (vaccines being a victim of their own success. The Rotary Club are amazing! I really should write an update to the polio eradication program, we're at the last mile with that now, very exciting times! Great, yes, very encouraging see no anti-vaccination comments here.

You are most welcome. Glad that you stand behind the Rotary Club as well. Nice! Please do when timer permits write up a polio eradication update. Would love to hear where it stands. Thanks again. I'm now following you. Cheers.

So... gathering more information tends to make people more likely to... neither get or not get vaccinated?

There doesn't appear to be a simple answer to this question. I went into this research thinking that the more people searched (especially online) for information the less likely they were to vaccinate (because of all the anti-vaccination misinformation that I kept hearing about). This appears to be the case some of the time but majority of people are searching and then vaccinating so it might actually be the case that looking for more information is beneficial for uptake (maybe reassuring doubts or even change minds) therefore maybe we should have more faith in people's ability to do their own research. That said each decision around each vaccine appears to be different, and also people background and previously held beliefs feed into it a lot, so a lot left still to unpack!

The exception seems to be childhood vaccination (though only one study shows this and its not statistically significant). Perhaps anti-vax propaganda is most prevalent around childhood vaccines as opposed to HPV or other vaccines.

Great article @nonzerosum. Sometimes talking (or writing) about research is the best way to see it clearly. I agree you need a strong conclusion to end on, but your data up to this point is intriguing. Two other things that occur to me:

  • You may state it clearly in the paper, but the total accept/reject statistics is something a reviewer might ask for. It can get tossed in the supplemental as its not critical to the big picture

  • Your figures are well designed and clear. That's better than most reviews so good job on that :)

Hi @tking77798, thank you for getting in touch and for your very kind words! Yes perhaps however the HPV has really been targeted heavily by these groups so I’m cautious to run with that theory (although don’t as of yet have another to take its place right now).

the total accept/reject statistics is something a reviewer might ask for

Yes I do have a habit of getting bogged down in the theory side and forget the very practical aspects such as this. I’m willing to bet this’ll be a comment if I don’t include it somewhere, thanks for the reminder!

Am I right in thinking you’re an academic too? Most of your post are likely wildly away from my discipline but we should talk GMO’s at some point. Lots of overlap with the hesitancy we experience with vaccines.

Well, I used to be an academic. After I got my PhD in Biochemistry, I sold my soul and became an industry research scientist. Steemit is a great valve for scratching that academic itch though.

I'm thinking of writing another GMO post, but I've just got to find the right angle. I'm open to suggestions if you have any.

Good luck with this review and getting your PhD!

Thank you for the nice reply.

wow, you've been doing a systematic review. hats off to you, sir.

Thank you, that is absolutely the correct response to give anyone foolish enough to attempt a systematic review. Also pity, pity works too!

Only intelligent people can work out systematic reviews that deliver the highest level of research believability, even higher than that of RCTs.

Hi @nonzerosum!

Your post was upvoted by utopian.io in cooperation with steemstem - supporting knowledge, innovation and technological advancement on the Steem Blockchain.

Contribute to Open Source with utopian.io

Learn how to contribute on our website and join the new open source economy.

Want to chat? Join the Utopian Community on Discord https://discord.gg/h52nFrV

Coin Marketplace

STEEM 0.36
TRX 0.12
JST 0.039
BTC 70223.87
ETH 3561.28
USDT 1.00
SBD 4.73