Wisdom of the crowds harnessed for pain prevention

The Kipuriihi tool developed at the University of Oulu assesses the effectiveness of treatments for lower back pain using wisdom of the crowds. In April, Kipuriihi will be introduced in Canada at the world's premier international conference of Human-Computer Interaction.

Low back pain is the most common global health problem. "The harm caused by low back pain has increased by approximately 50% over the past 25 years, and at any given time half a billion people are suffering from it," says Jaro Karppinen, Professor of Physiatry at the University of Oulu.

There are a massive number of treatment methods available for this massive problem. That these have not all been assessed is an acute problem. Randomised studies are published on medical treatments for back pain of course, but they do not cover alternative treatments or the general public's tricks for treating pain.

"There are an abundance of treatments that are difficult to research with randomised set-ups. It would be advisable to also bring forth the views of back pain patients on the effectiveness of the various treatment forms," the professor says.

The Kipuriihi tool created by Karppinen and Simo Hosio, Postdoctoral researcher of Ubiquitous Computing at the University of Oulu, aims to meet this challenge. It gathers assessments by both the general public and doctors on forms of treatment and combines these into a general assessment for which the user can select the criteria.

The forms of treatment include both treatments given by medical professionals and self-care treatments by patients. According to Hosio, the latter of these is of great importance because in nearly 90% of cases people seek treatment methods independently online.

"We cannot reach these people with randomised studies which only cover those sufferers, who have sought help from a doctor. Google tends to lead people to discussion forums where one must independently determine the reliability of given treatment options.

The wisdom of the crowds theory, which was developed in the early 1900s, has been adopted to support the reliability of the Kipuriihi tool. Hosio sums up the core idea in the following way:

"If we gather a large enough group of people from different backgrounds and they give their independent assessment on a topic, the sum of the assessments is on average smarter than an individual's assessment. The collective voice is close to the truth."

The new aspect we've added is dressing the old theory up into interactive form, as a web interface, which the user can utilise to assess forms of treatment and to search for wisdom of the crowds assessments on these.

"By using valid mathematical methods, we can determine a large set of tested treatments from thousands of opinions," Hosio states.


Treatment therapies can be proposed on a website open to everyone

As wisdom of the crowds requires a large mass of assessments, the development of the Kipuriihi tool began three years go with data collection. A total of nearly ten thousand assessments were submitted by hundreds of members of the general public and 70 doctors. These assessments covered around one hundred forms of treatment.

One of the conditions for participation was that respondents responded independently separate from one another. In this way, we can avoid the speaking over one another type of phenomenon common on discussion forums, which allows the most prominent and noisiest participants to mould opinions.

Setting the research question for data collection and fine-tuning the wisdom of the crowds platform were complex tasks, but the end result is easy to use: Anyone can go on the kipuriihi.org/tutki web page and select whether they want to search for treatments assessed by professionals (i.e. doctors and physical therapists), the general public (laymen) or both. After this, the user can adjust the perspective to anything they like; variables include cost, speed of impact, permanence of impact and the effectiveness of the treatment.

For example, professionals named pharmacotherapy, exercise and stretching as the cheapest and fastest treatment forms, while the crowd's suggestions for the best methods were "don't get stuck", the constructive rest position and stretching. Whereas suggestions by professionals represent official medicine, the crowd's methods are in vernacular and common sense, and can in principle be anything at all.

Although the actual data collection stage has already ended, users can continue to suggest and assess forms of treatment.


Methods used by the general public are unrestricted as these are assessed by the wisdom of the crowds

"The question arises of how medical treatments fare in comparison to the crowd's methods. However, no such analysis has yet been conducted," Jaro Karppinen says. "At this point in time, it is most important to get the patients' voices heard and facilitate a comparison assessment from patients for patients."

Crowd assessments are necessary from the perspective of medical science when we want to know how patients feel about treatments, whether they are successful or whether information campaigns are effective. For doctors, Kipuriihi is predominantly a monitoring tool, a means to hear patients' voices.

"In the future, it would be beneficial if Kipuriihi users could also see the opinions of medical professionals on treatment methods proposed by the general public," Karppinen reflects, but he also believes that the wisdom of the crowds in itself will weed out ineffective nonsense.

"No restrictions as such should be specified for methods listed by the general public in Kipuriihi. A treatment methods such as unicorn therapy could prove very interesting due to how it will be assessed by peers," Hosio states.

"Now, citizens have not assessed the treatment types articulated by doctors as any better than those articulated by citizens themselves. This indicates that lower back pain is a challenging issue. There are no evident treatment methods that everyone feels are effective."

The next step, if the project can get more funding, could be to monitor results. "A questionnaire could be sent for example once a month on the treatment methods listed in Kipuriihi to people's mobile phones, and after half a year the application could indicate how many people felt the treatment method was effective," Hosio envisions.


Wisdom of the crowds platform could be developed into a public health portal

Kipuriihi is the result of a tradition of research: Crowdsourcing has long been researched at the University of Oulu. The wisdom of the crowds is a form of crowdsourcing. The tool in itself was created during a three year long Academy of Finland-funded wisdom of the crowds project, which aims to bring the wisdom of the crowds to support decision-making.

The wisdom of the crowds platform created by Hosio can be applied in many areas: it has been used for example in a racism study and to seek help for overweight people.

"It could also be applied to all common illnesses," says Jaro Karppinen. He hopes that Kipuriihi will end up being used on an official website. Hosio tosses out a follow-up idea.

"A crowd-sourced public health portal!"

There is no reason why the quality of the platform would be an issue. Kipuriihi has already been accepted to the ACM Conference on Human Factors in Computing Systems (CHI), which will be held in April.


AUTHOR: Jarno Mällinen

Last updated: 24.4.2018