A clinical trial in Kenya confirmed that human kindness is the secret ingredient to health and mobile phones are an ideal delivery system. Well, that’s my interpretation.
Here’s the gist:
Taking your meds is essential to maintaining your health when you live with a chronic condition. People know this, but they need help doing it. They tend to tune out reminders — constant texts become more of a nag than a boost, studies show. Instead, people respond to empathy — in this study, a single word, texted once a week: “Mambo?” which means “How are you?” People who texted back that they were not feeling well received follow-up phone calls to see if clinicians could help them get back on track.
And the results? I’ll quote the study:
Patients who received SMS support had significantly improved ART adherence and rates of viral suppression compared with the control individuals. Mobile phones might be effective tools to improve patient outcome in resource-limited settings.
This research, conducted in 2008 and published in 2010, reminds me of others of a similar vintage.
In working with teens living with serious, chronic conditions, Project HealthDesign researchers found that technology is a comfort, especially if it is portable, like an ipod or cell phone. They also found that teens are less likely to take their meds if they are feeling sad. So the researchers came up with a tool that tracks the teens’ moods, monitoring the songs they listen to and the words they use in text messages to friends and family. If a teen’s mood seemed to dip, the tool sent an auto-generated medication reminder. Crucially, the teens agreed to surveillance by their clinicians because they were willing to trade privacy for better health and independence from their parents.
Sadly, it was only a pilot project. but did show how a mobile device can serve to deliver just-in-time, clinically-guided medication reminders in the context of people’s lives. Or, as I like to call such things: health care (with an emphasis on care).
What if this idea could be expanded beyond the clinical setting? What if we could break open the box that contains all the support that friends, family members, peer patients and caregivers beam to each other every day?
Some examples:
PatientsLikeMe demonstrates the power of peer education and support in HIV, epilepsy, and other condition communities. People who are able to connect with others who share their same condition are more likely to be able to manage their own care, including taking their meds on time and with regularity, understanding that a “treatment holiday” is not a good idea.
QuitNet is another community of people who are working together to stay well by not doing something (smoking). Members who are successful stick around to help newcomers because they want to give back what they received.
Neither of those powerful examples of peer-to-peer health care involve text messaging, but what if they did? What if we could marry the power of peer support with the ubiquity of cell phones?
It’s also not lost on me that these studies date back 5+ years in some cases. And yet we are still debating both concepts: text messaging for health and peer-to-peer health care. On the one hand, pioneer clinicians, patients, and caregivers are using these techniques to good effect. On the other hand, the annual mHealth summit is growing to Hulk-like proportions. But what will it take to bring these ideas to the massive middle, the mainstream, so more “real people” can benefit?
I’d love to hear what you think — please share your ideas and observations in the comments.
Meantime, if you want to geek out, I’ve included some background readings. Please add to the list of studies and resources in the comments (and remember, self-promotion is OK if you think it’s relevant!):
Segmenting Mobile Health, by Boone (Healthcare Standards, 2013)
Trial to Examine Text-message Based mHealth in Emergency Department Patients With Diabetes (TExT-MED) (PDF), by Arora, et al (Annals of Emergency Medicine, 2013)
Ask, Don’t Tell — Mobile Phones to Improve HIV Care, by Lester (NEJM, 2013)
Text messaging best practices emerge, by Rowe (Healthcare IT News, 2013)
A text message programme designed to modify patients’ illness and treatment beliefs improves self-reported adherence to asthma preventer, by Weinman et al (International Journal of Integrated Care, 2012)
Sharing health data for better outcomes on PatientsLikeMe, by Wicks et al (JMIR, 2010)
3 stories related to diabetes: Texting & Diabetes Management: Launchpad to ‘mHealth’, by Dyer (Diabetes Mine, 2010); Texting improves Type 1 diabetes adherence, by Dolan (MobiHealthNews, 2010); and an NBC story on the program.
Related posts by me:
Building a research agenda for participatory medicine
All posts related to text-messaging (also known as SMS)
All posts related to peer-to-peer health care
Image: Holding hands by Adam Foster on Flickr.
Another insightful and inspiring post, bridging the gaps between rationality and emotions, and between technology and health. I especially the quotes
and
Will look forward to geeking out on some of your related reading over the weekend.
Meanwhile, if you’re open to other reading recommendations, here’s a relevant excerpt from a book I just finished, A General Theory of Love, by Thomas Lewis, Fari Amini, Richard Lannon (all MDs). The book emphasizes the importance of limbic resonance, regulation and revision, and the importance of empathy and love in many dimensions of healthy human development and interactions.
The excerpt is from page 222:
Oops – substitute “limbically” for “limbic ally” above …
Looks like I munged up the blockquote tags, too … note to self: don’t comment on blog posts before finishing first coffee of the day.
Fixed! And grateful 🙂
This is one of my favorite posts on your blog. I think it succinctly summarizes three very important elements of successful mHealth interventions:
– it has to engage and motivate
– it can be easy
– it can’t completely substitute for the human touch (although it can approximate it)
As a clinician and mHealth researcher, I am sometimes disappointed by the push for the “next big thing” – apps, mobile sensors, etc etc – and the concomitant lack of attention to basic, simple behavioral change theories.
After all, we know what works: making people feel heard, helping them to care, and giving them self-efficacy. This is the ideal basis of “health care” and “health” (and what most of us went into medicine to provide!).
We know that SMS is not in itself a panacea. But it can help us deliver and extend interventions in a more low-cost/scaleable manner. (SMS programs will still work even when the next iOS come out ;).)
Re: the peer-to-peer portion — I expect that a lot of this will have to be led from outside of the clinical/research world. Because those of us who have to go through IRBs are limited by (understandable, but often overwrought) concerns about pt safety and ethics. We would need to *guarantee* that participants have anonymity, that mental health safeguards are in place, that noone is disseminating erroneous information, that HIPAA is not being breached, etc…..
So we need patient groups to take the lead here (for the most part).
(As an aside: we also know that clinicians are resistant to change. So interventions that can be easily incorporated into our existing workflow are more likely to have high uptake. Automated SMS interventions are way less work for a provider than, say, a gazillion QS data points…. )
Interested in others’ thoughts. And Susannah, thank you as always for your thoughtful curation and commentary.
Susannah – thanks so much for bringing to the forefront (again) that when all is said and done, how we foster the human connection should be on top of our priority list when building digital interventions. Don’t get me wrong – I am an automated tailored decision support digital intervention person but ignoring the social component only reduces the efficacy of what we do. There is emerging evidence that digital interventions with a human component outperform those without and when you really look at the data from certain trials the key ingredient is often the social component (or supportive accountability as termed by David Mohr and his group) rather than the digital piece alone. In our own research, during times of crisis like a relapse, 96% of individuals in drug treatment were interested in having a friend alerted and 78% percent wanted a counselor alerted. While this social component might not be as important for reminding yourself to drink more water and some people early in the change process might want more anonymity, it certainly is crucial when we are dealing with life threatening diseases. Mobile is a wonderful way to foster social connection – while also offering continuous tailored assessment and feedback to keep us on track between these connections.
Thanks, Fred!
One of the nicest compliments I’ve received recently was from someone who saw me speak and later thanked me for not trying to sell the audience anything. Not that there’s anything wrong with sales! But sometimes we need to step back and say, Wait, if we are focused on improving care, not pushing the Next Big Thing, we may need to go back to basics (echoing what Megan & Michael & Steph have also commented).
I can’t wait to hear more about what you’re presenting at the mHealth Summit! Where can I sign up for updates on what you’re working on?
The presentation at the mhealth summit is not on this but will someday start the blog I have been saying I would start for many years in a few years ;-). Look forward to seeing you!
That is great. It just shows you that simple things are are often the best. I love ‘mambo?’ Thanks for sharing this.
Susannah – Brilliant post (again). I’m reminded of advice given (on the same day) by a community organizer and a professor at MIT’s Media Lab: It’s not about the technology. It’s about how technology can facilitate human goals.
I suspect Megan is correct, that some of this research will have to come from outside academia proper – more reason to deepen meaningful collaboration between the people who are interested in evaluating these ideas and those who are interested in using them 🙂
Yes! The emphasis on empathy is key and too often missing from tech driven healthcare solutions. Most fantastic part of this post: “People who are able to connect with others who share their same condition are more likely to be able to manage their own care, including taking their meds on time and with regularity, understanding that a “treatment holiday” is not a good idea.”
I just wrote a blog post about this…I call it the “modify factor” http://habifhealth.com/2013/11/15/the-modify-factor-designing-engagement/
Wonderful synchronicity of thoughts! Everyone, please read Steph’s post because it complements this one very well (and closes the loop by suggesting actions you can take if you want to get started NOW incorporating these principles into your work).
Another of your great discussion posts Susannah. As you know we do a lot of work in the area, and we have worked with clients in structuring the most effective type of communication. In general we have found the following:
1) The more specific, the better
2) The more interactive the better
3) The more personal the better
4) The more the response provides a relevant call to action, the better
5) The more there is a tie to peer group, the better.
More specifically, and example regarding medication compliance. This was with a mental health client who had patients suffering from depression management. The initial messages that went out were essentially “Don’t forget to take your medication” Moderately effective for a short while but quickly became white noise. when switched to an interactive question ” Did you take your medication? Yes or No?” It was deemed a little more effective but found patients were responding how they thought they should versus how what they were actually doing. It was when we changed the question to being more of a proactive mode versus reactive, such as “How effective was your medication today” That the engagement jumped and the value of the responses became very strong. Then when we added the ability for the peer group to me notified when they responded negatively, that the client found dramatic change. and on an on-going basis.
And this has been replicated repeatedly. So its not just a message but how and that it should be crafted in a manner that satisfies the 5 criteria I noted above.
Thanks again for initiating the discussion.
Howard
Beautiful. Thank you!
One of the examples I have in mind, but haven’t yet found the cite, is a smoking cessation program that asks the person interested in quitting to write a series of “this is why I’m doing this” encouragement texts in advance, at the start of the process. Then, when they hit a crave, they can text a shortcode and get back one of those personal notes, like a letter to your future self. The program also allowed people to assign that task to a loved one, like a child or a spouse. Like Fred’s research finds — and yours, too — we know that the crave or dip in motivation will hit, so why not anticipate and design for it? Love that.
If anyone knows what study or program I’m talking about, please post a link! Otherwise I’ll keep looking through my outboard memories (this blog, my notebooks).
Sounds like an interesting study. While I’m able to find references to several studies involving personalized text messages to support smoking cessation, I can’t find any that report on subjects composing messages they will later receive.
This study seems to be the most often cited:
Caroline Free, et al., Smoking cessation support delivered via mobile phone text messaging (txt2stop): a single-blind, randomised trial. Lancet. 2011 June 30. 378(9785):49-55.
The study is cited in a number of other peer-reviewed studies, as well as a Jan 2013 MobiHealthNews article on Agile Health’s Kick Buts mobile application, an app which purportedly includes additional features somewhat related to what you describe:
The only study I could find involving personalized text messages used to promote health written by the subjects themselves is in the realm of weight management:
Ben S. Gerber, et al. Mobile Phone Text Messaging to Promote Healthy Behaviors and Weight Loss Maintenance: A Feasibility Study. Health Informatics J. 2009 Mar. 15(1):17-25.
Perhaps the study you have in mind may be listed in the “cited by” list for one or both of these studies.
Thanks, Joe! These are great finds. I remember hearing about the personalized text messages at a conference, so it’s possible that it’s unpublished research — the ones you list are pretty excellent replacements.
Yes – so crucial – the motivating the future self. http://www.youtube.com/watch?v=QVUrMXJ3wZ8. One of the most interesting finding we had was that in my old company (its gone now), we guided people to write messages to themselves as you know. What ended up happening is that people had great difficulty writing messages and sometimes when they did they were mean. We have a paper coming out where we asked hundreds of people to write messages to themselves for a goal and if they were not meeting the goal, then for someone else with the same goal. What happened was if they were having trouble, people who wrote messages to themselves were significantly more negative and mean than when they were writing messages to another person who was not meeting their goal. I guess that is why others are so important. So they can be positive when we can’t be ourselves. As you can see I am completely loving this thread! Thanks again.
So interesting, Fred. Similar results with teens – they are much better at identifying strategies for others, than for themselves.
The problem is: it gets complicated to do the programming to allow ppl to send themselves self-written msgs (at least in my limited experience) …. ? Is this one of those cases where tech is difficult to match to our concept of the ideal?
Interesting findings Fred, I look forward to reading it when published. In that regard, what we have found is that case/care managers are asking these questions of their patients as well as family and friends, crafting simple but effective motivation messages from these insights. And because of the very positive feedback being received by the case/care managers, they get all the more jazzed to be more creative, working with the patient at all times. Similar to what you found Fred, to be effective, this cannot be a solitary exercise and involvement with loved ones/peers is a very powerful “tool”.
Thanks Fred,
Howard
I love your breakdown of what you’ve learned, Howard. We aren’t working on mHealth apps but do have related learning (particularly in a project, Range, that is about finding available food for youth). The question we keep finding ourselves asking is how do we narrow the universe of information in a way that provides actionable information to communities.
We’re looking at giving the reminders to other community members (people who interact often with you) to help with the social connection. I haven’t clicked through the links yet but do wonder how that part of social — support through caretakers — plays into mHealth apps.
Hi Marnie,
Best to look at this as not “Apps” but as communication, whether through an app or a platform or other means. The social element is a big driver for individuals to get more involved with others and themselves. What we have found, like in any case, if you have a non ending stream of messages or updates to everybody, it becomes noise in the background and ignored. People want to help each other and be informed but usually only when needed. And similarly people like the idea of sharing information but not necessarily all information. So by parsing the outreach to certain parameters, driven by by the patient, you get a much more robust involvement/engagement of all parties.
Not sure if I answered your question so please feel free to contact me directly to discuss further.
Howard
Howard – thanks so much. You have been doing great work in this area for years and love to see your engagement findings – completely gels with what we have found.
Thanks Fred, I look forward to your mHealth presentation in a couple of weeks.
Howard
Love how you decribed this and the examples. We definitely have developed an emotional relationship with our devices and it probably stems from the fact that they connect us to loved ones. My iPhone told me I had a busy day coming up and it felt like it was looking out for me.
Thanks, Anne!
You might like the work that Nathan Jurgenson, Alice Marwick, danah boyd (and others…) are doing to better understand the relationships we have with each other & our tech (the betwixt & between).
Thanks to a tweet from @alicetiara, for example, I got to read this
letter to the editor which sums up how many of us navigate friendships these days:
Well compiled thoughts Susannah! I’m a huge fan of technology. BUT, if it removes the human element of emotion and connection, the value of it is short lived.
In your opinion, what can big-box HIS and other tech developers do to include opportunities for human kindness in their apps?
Keep up the good work!
– @tedouglasjr
Great question!
Anyone?
I’m going to meditate on it, maybe come up with outside-of-health-care examples to share and I’d love to hear what others think.
“Human kindness in their apps.” Now there is a design spec. A universal design spec for Health 2.0 writ large IMHO. Human kindness is the province of, well, humans. The conduits have to be sensitive to how it is expressed, received, how trust builds over time to make it real for the humans so connected, as is possible in online patient forums at PatientsLikeMe and elsewhere. But you can’t increase the speed of that process with an IT injection, like say another gig of data transport.
Really it gets down to a question of value, in terms of what outcomes any digital product or process embodies. Take a PTSD mobile app. Maybe a med-taking reminder is helpful, but because sufferers can experience flair-ups of grim and dangerous dimensions, having a group to connect with Right Now, a group where a known measure of human kindness exists, among peers, gives a simple display of text a powerful therapeutic quality.
So by all means, human kindness. Let’s run it through all these conduits.
Susannah,
What a lovely summation of the relationship between technology and emotion: human kindness is the secret ingredient to health and mobile phones are an ideal delivery system. Your article highlights the powerful complement between these two unlikely allies and the enormous benefit their relationship potentially brings to the patient and their community.
The practical application of technology is sometimes lost in the ‘wow’ factor of technology, creating more noise than value. Task lists and reminders alone become less effective and almost a deterrent to treatments, just one more sound bite, one more frustration factor in the management of an illness.
A long standing social network enthusiast and motivated by a dear friend’s health scare, I founded Curatio to develop a peer support platform that gives people the support, guidance and motivation they need for improved recovery and health empowerment. Our mission is to enable personalized care by taking advantage of what technology has to offer – engagement of the community.
Our goal is for the technology – an app – to have the smarts to engage the support network when most needed by the patient. This is done by the technology comprehending cues to a patient’s emotional status and activity level. By enabling contextual understanding of mood within the app to guide and encourage health management, ensures a better opportunity for treatment adherence. The technology is a driver with a very clear purpose within the eco-system of personalized care.
I would love to continue the dialogue. Bringing together communities, taking advantage of technology, building social networks, and the behavioural aspect of apps to enable community has become an area of particular passion for me!
Thank you, everyone, for continuing to add so much insight to the conversation! I just made a quick list of the conditions and applications we have discussed so far:
medication understanding & adherence for HIV, epilepsy, and chronic conditions among teens
smoking cessation
asthma
diabetes
recovery
weight loss
food availability for youth
PTSD
health management in general.
And this is only the beginning, only what has occurred to us or what we are working on at present. Pretty impressive.
Thanks for this, Susannah. I get geeked out over mHealth in general, and there is a ton of interest in mobile phone behavior change interventions.
One big project that the US government (I believe DHHS primarily) has invested a lot in rigorously evaluating is text4baby, as you may know:
https://www.text4baby.org/index.php/about/data-and-evaluation
When I was pregnant, I actually was a participant in one of the studies evaluating text4baby, and I’m curious to see what the investigators find. I’d receive 2-3 text messages a week along the lines of, “Call your doc if you have bleeding in your second or third trimester” and so forth. The messages were a combination of informational and supportive (“Great job if you’re breastfeeding!”). Personally, being a reproductive health research nerd myself, I felt I was getting a torrent of texts and started to tune them out, but that just underscores the importance of targeted interventions; text4baby isn’t intended to show impact among reproductive health nerds!
Anyway, here’s a 2013 PlOS review paper summarizing health behavior change mobile phone interventions – including the Kenya trial you mentioned.
http://www.ncbi.nlm.nih.gov/pubmed/23349621
Since this blog is my outboard memory, I’ll add a link to a study that relates to this conversation:
Study: Text messaging reduces pain medicine requirements during surgery
http://www.rti.org/newsroom/news.cfm?obj=B81A4EAA-CC8E-4267-57867B04EE2DC8FD
Patients who text messaged a stranger just before minor surgeries required less supplemental pain relief than patients receiving standard therapy or distraction techniques, according to a recently published study conducted by researchers at RTI International, Cornell University and LaSalle Hospital (Montreal, QC).
The study, published in Pain Medicine and funded by Cornell University, showed that mobile phones provide new opportunities for social support, improving patient comfort and reducing the need for pain relief during minor surgeries and in other clinical settings at a very low cost.
“These findings suggest that the simple act of communicating with a companion or stranger reduces the need for supplemental anesthesia in a way that surpasses usual perioperative care during surgery,” said Jamie Guillory Ph.D., digital media health research scientist RTI who conducted the study while at Cornell. “This is significant as the physical presence of a social support companion is often not feasible during many minor surgery procedures.”
By applying a text-based intervention, this study is an extension of existing research on the impact of social support on pain perceptions and the need for narcotic pain relief.
Researchers recruited 98 patients receiving general anesthesia for minor surgeries in Montreal, Quebec between January and March 2012. They randomly assigned patients to text message with a companion, text message with a stranger, play a mobile phone game for distraction, or receive surgery (i.e. do nothing).
While both texting conditions reduced the need for pain management better than standard surgery, only texting a stranger reduced it beyond the distraction method of playing a mobile phone game. The researchers believe that is because the conversations with strangers were more emotionally positive, focusing on topics and values personally relevant to the patient.
“Consistent with this finding, previous research shows that engaging in activities that reinforce a person’s core values helps people to endure a pain tolerance task longer,” Guillory said.
Conversations with companions primarily focused on the surgery, the body and negative emotions, which suggest anxious feelings in the companion about the surgery, may have resulted in shared anxiety between the patient and companion, limiting the positive effects of social support.
“Although at first it seems counterintuitive that text messaging with a stranger was more effective than with a companion, it’s the content of the conversation that makes the difference in reducing patients’ need for pain relief during surgery,” Guillory said.
Love this, thanks a mil for sharing! Exciting! It brings me back to a prototype (as part of my academic research) I worked on in 2011, called Morphine Drip. The purpose of Morphine Drip was 1) to persuade people recovering from painful injuries or surgeries to use mobile calming solutions before distress levels become too high 2) to decrease anxiety (increase calm) of users through technology based calming triggers
3) to increase self-efficacy by empowering people to develop effective daily coping habits. The basic design idea was that a user in a state of distress can opt for a “morphine drip” aka pain management solution by receiving a mobile-tech based calming message. Feels good to know we were working in the right direction 😉