Today I was torn between attending a mobile health presentation and a mobile libraries brainstorming session. I chose the mobile health one, and was privileged to hear Kay Connelly lead off the Bartels Health Informatics Speaker Series with a presentation on the long drawn out research process involved in designing an effective mobile device application for a chronically ill low-literacy population.
SI launches Bartels Health Informatics Speaker Series: http://blog.si.umich.edu/2009/10/29/si-launches-bartels-health-informatics-speaker-series/
A couple of us were live-tweeting the event, and had folks watching the hashtag stream from the UK, Massachusetts, and North Carolina. Here is the Twitter stream, courtesy of my new fave Twitter tool, What the Hashtag.
Transcript from November 16, 2009
11:11 am pfanderson: WARNING: Live tweeting from Kay Connelly presentation on mobile health application for chronically ill low-literacy population #mobhlth
11:15 am litebulb11: Kay Connelly-mobile health apps for chronically ill low literacy populations #mobhlth
11:17 am pfanderson: hemodialysis patients is population, highly restricted diet. Kidneys clean toxins fr body. 80% can’t adhere #mobhlth
11:18 am pfanderson: Patients told to make a paper diary of what they eat/drink, but can’t read/write. Ahem. #mobhlth
11:19 am SeerGenius: RT @pfanderson: Patients told to make a paper diary of what they eat/drink, but can’t read/write. Ahem. #mobhlth
11:20 am pfanderson: Application based on resrch fr informatics in diabetes, forthcoming bk. Dietary Intake Monitoring Application. Icons for foods #mobhlth
11:21 am litebulb11: Bar code scanner and icons on mobile device to help monitor dietary intake #mobhlth
11:22 am pfanderson: Neat – using barcode scanner in mobile device for some foods. #mobhlth Iterative design – patient feedback, redesign.
11:23 am pfanderson: First question: can target population use PDAs, can they press buttons, play games. Coordination & “pincer” strength, error rate #mobhlth
11:24 am pfanderson: 2nd question: visual acuity. On mobile devices, what size of icons is most viewable / pt prefs. #mobhlth Large icons preferred by elderly
11:26 am pfanderson: Question 3: could patients use a voice recorder? No problem. Yay!!! Elderly needed two hands for this task. #mobhlth
11:28 am pfanderson: Question 4: SDIO Barcode scanners vs bluetooth device pen scanner. Surprise finding: women cdnt use pen because of nails #mobhlth
11:30 am pfanderson: Patients had never realized that most prepackaged foods have barcodes. Had trouble finding barcodes on foods. #mobhlth
11:30 am pfanderson: They made a “game” BarcodeEd to help folks practice finding and scanning barcode foods. #mobhlth
11:32 am pfanderson: Patients were incredibly protective of their PDAs, lived in unsafe environments, high theft. #mobhlth
11:33 am pfanderson: Scanning, even w/o feedback, changed behavior. 60% foods not in open source database. Pts ate more than they thought (LOL) #mobhlth
11:34 am litebulb11: Low literate populations had difficulty reading actual food item brand names, had to describe ‘cereal with leprechaun and rainbow’ #mobhlth
11:34 am pfanderson: Interesting. Pts cdnt voice ID foods eaten because they cdnt rd labels. Voice input was stream of consciousness, unstructured #mobhlth Oops
11:36 am pfanderson: Participants preferred voice input, but performed better with scanning and they didn’t like using it. #mobhlth Turns out they wanted phone
11:37 am amcunningham: interested in health inequalities/ health literacy? follow #mobhlth today
11:37 am pfanderson: She mentioned a website that IDs foods & dietary input from uploaded pics. I want to know what it was!! #mobhlth
11:40 am pfanderson: Patients sorting food cards noticed what they can/can’t eat, ignored side dishes. Awareness was key in training 4 app. #mobhlth
11:41 am pfanderson: For organization of food input, patients preferred a combination of input types – Time of Day & Food Group #mobhlth
11:41 am MarkOneinFour: @amcunningham do you know the context of #mobhlth?
11:43 am pfanderson: Patients liked to show off medical knowledge they’d picked up, and would use jargon they didn’t actually understand 🙂 #mobhlth
11:43 am litebulb11: Patients proud of obtained medical knowledge, affects design of interface/icons try not appear ‘dumbed down’ #mobhlth
11:44 am pfanderson: Important to make interface sophisticated, not childish, but with tricky balance between what pts liked / what they understood #mobhlth
11:44 am pfanderson: @MarkOneinFour #mobhlth is presentation by Kay Connelly on mobile hlth app for chronic ill low literacy patients at UMichigan
11:47 am pfanderson: Patients could use any nontext widget, as long as they were large enough for vision / dexterity #mobhlth
11:48 am litebulb11: Project abstract and biography of presenter for mobile health app from previous presentation: http://tiny.cc/T93oe #mobhlth
11:48 am pfanderson: Patients favorite part of tool was HOME button (used as an escape/help). NOTE: TV remotes lack this functionality #mobhlth
11:49 am pfanderson: Patients tended to use it either out of home or in home, but not both. Plan: one device for each loc #mobhlth
11:50 am litebulb11: Pride affects choice, patients often preferred interface they used incorrectly #mobhlth
11:50 am pfanderson: Advantage of #mobhlth device is to not attract attention, avoid stigma. But, this population loved showing off. 🙂 They had cool toy, status
11:51 am litebulb11: Technology is a status symbol, patients showing pda off to neighbors by scanning everything ie paper towels. #mobhlth
11:51 am pfanderson: Patients would show off device by scanning non-edible items, meant muddy data to clean. #mobhlth Pts lie to caregivers to look good.
11:52 am pfanderson: Patients confided in tech geeks because they felt they would not be judged for poor food choices #mobhlth Pts want privacy, not nagging
11:53 am pfanderson: Patients *will* cheat on diet because disease is so miserable. They don’t want to be made to feel bad for cheating. #mobhlth
11:54 am pfanderson: Made it harder to use voice record, accessible only after navigation to food type failed to locate actual food. #mobhlth
11:55 am pfanderson: Organize common foods as easiest to find (ice). #mobhlth
11:57 am pfanderson: Self management and monitoring was empowering. 6 wk trial. Will results persist? Don’t know yet. Pts want to keep device. #mobhlth
11:59 am pfanderson: Future directions: add portion sizes for nonliquids (size comparisons); acceptance by higher-literacy pts? Does design change #mobhlth
11:59 am pfanderson: Can you design for lowest common denominator and find acceptance by others? #mobhlth
12:00 pm pfanderson: OK, she’s taking questions. Anything you want me to ask? #mobhlth
12:01 pm pfanderson: Aha – had star so they could add list of favorite foods. #mobhlth
12:02 pm pfanderson: Suggestion from audience. Use photos not for ID and feedback in realtime, but for validation. #mobhlth Compared data to 24hr recall
12:04 pm pfanderson: Navigation structure – patients were confuse by anything other than linear design #mobhlth Audience: cd use with children?
12:05 pm pfanderson: Brainstorming – taking these design guidelines to create authoring tool for low literacy mobile device apps. COOL!!! #mobhlth
12:07 pm pfanderson: Audience: is it harder to get info to low literacy, low SES, rural, chronic ill or combo populations? Answer: we don’t know yet. #mobhlth
12:09 pm pfanderson: Talking about assumptions of working w/ low SES population, blaming patient for illness. #mobhlth DONE, talk over.
12:17 pm inetnurse: RT @amcunningham: interested in health inequalities/ health literacy? follow #mobhlth today
12:40 pm SeerGenius: @pfanderson thanks for tweeting #mobhlth ! Lovin’ it like Mickey Deez… oops, #notgoodfood.