Reaching Out to Physicians

Matthew Lee, dwellSense Lead Researcher, Ph.D. Student in Human-Computer Interaction, Carnegie Mellon University

As mentioned in our previous blog post, each of our study participants has a different set of clinicians who provide care for them. At our request, our participants took out their address books to look up their doctor's information and shared it with us. Now we had the daunting task of trying to convince time-strapped physicians to spend 30-60 minutes with us to review the patient-generated data from the dwellSense system.

Our recruitment approach began with sending a letter to the physicians' offices that described the study, said it involved one of their patients, and emphasized how valuable their contributions would be to our study. We didn't hesitate to use whatever persuasive strategies we could think of, including name-dropping and some gentle groveling. We made sure to mention that the research was funded by the Robert Wood Johnson Foundation, a name that undoubtedly carried a lot of weight with clinicians. We also used the fact that I’m a student who’s working on completing my Ph.D. dissertation. The letter asked not only for the clinician to agree to an interview session to support the cause of the study, but also to support my Ph.D. dissertation. People often are willing to help students with their work in ways that they might not help a corporation or something a little less faceless. Fortunately, I didn't mind casting myself in the role of a struggling graduate student for the sake of science. 

After we sent these letters to the clinicians’ offices, we waited. Fortunately, because many of the physicians were also part of the University of Pittsburgh Medical Center, we were able to find their email addresses on the University of Pittsburgh website and send the letter to them via email. To our surprise, we received responses from four of the primary care physicians we targeted within a week of sending the letter and/or email.

For those we didn't hear back from, we followed up with a phone call. Some offices received the letter, but other offices reported that the letters were lost in the sea of paperwork that flows in and out of the office. Many asked to have the letter sent again by fax, a 1980s-era technology apparently still widely used in doctors’ offices. After faxing the letters, we followed up with the offices to make sure they received them and got on the radar of the physician. We found that the front office staffs are usually quite accommodating of our requests to make sure the letters (and presumably other paperwork) get into the hands of the physicians. 

Thus far, we have been able to interview five physicians, and we still have approximately seven physicians to follow up with to see if they would like to participate. Overall, we were delighted with the responses we have received so far. As we reflect on the reason for our success, we have come to find that certain physicians were really interested in the idea of patient-generated data, in particular, medication taking, and how it could provide a window in the everyday functioning of one their patients (with multiple chronic conditions). We also think that leveraging the fact that it is part of a Ph.D. dissertation and giving busy physicians a chance to help a student was another reason they were willing to make time for us.  Stay tuned for a follow up post about some of the interesting findings from our interviews with physicians. 

Learn more about the dwellSense project.
Read more posts from the dwellSense team.

Recruitment Issues

Anind Dey, Ph.D., dwellSense Principal Investigator, Carnegie Mellon University

Throughout our project, we have had issues with recruiting subjects for our study. Our goal was to recruit 30 elders, along with their caregivers and doctors. However, we have struggled to recruit 20 subjects. There are a number of possible reasons for this. Some are due to not being targeted enough in recruiting, and some are due to not having enough constraints. To perform our study, we require a great deal of access to subjects and their apartments in order to perform assessments, to regularly check the state of the display and sensors, and to replace batteries.

From a time and manpower perspective, we needed to recruit subjects within a small geographic area. Otherwise, the commute time to and from our participants' apartments would have been unmanageable. So, we made a choice to identify a small number of locations from which to recruit subjects. That drastically limited our subject pool, but having subjects spread around the city would have reduced the number of subjects we could have actually worked with, too. It was not easy to identify the potential locations or potential subject populations centered in an area, particularly because we weren't looking for individuals with particular disabilities or diseases. Instead, we were looking for elders who were "at risk" but not necessarily exhibiting signs of cognitive decline.

Once we found suitable locations with large potential subject pools, we faced other difficulties in recruiting. One issue was the amount of time we wanted participants to commit to being part of our study. Because we expect the personal health information we collect via sensors to provide value over time rather than immediately, and because we wanted enough time for participants to get over the novelty factor of our embedded assessment appliances and displays, we needed participants to commit for several months. Many potential subjects had a hard time making this long-term commitment.

We also wanted to maximize the number of assessments we could make for each participant. This meant we were looking for subjects who regularly perform the three activities we monitor: coffee making, making/receiving phone calls, and taking medicine. Some potential subjects didn't work out because they drank tea instead of coffee, or made instant coffee rather than using a coffee maker. These activities, though perfectly reasonable, made them inappropriate for our study. Others had digital phone lines that were not compatible with our phone sensing hardware, which works with analog phone lines. Furthermore, some took pills from the original pill bottles rather than from pill boxes. We were afraid that changing someone's habits (from using pill bottles to using pill boxes) would impact their behavior too much and introduce errors in actions, so we removed these individuals from the potential subject pool. In the end, we chose to accept subjects if they used pill boxes and satisfied the requirement for either the coffee making or the phone use.

Further complications arose from the nature of the research itself. When they heard our description of the study, some subjects declined to participate. When asked why, it was clear that participating in such a study and using such a system that tracks and represents health information served as another reminder of their declining physical and/or cognitive health. The last thing they wanted was a "constant" notification of how they were performing everyday activities.

We are continuing to recruit subjects, but this is a slower process than we had expected. We’ve definitely learned a lot about recruiting for this type of study.

Learn more about the dwellSense project.
Read more posts from the dwellSense team.

Creative Commons photo by jcestnik.

Reflections from the First Two Months of Data Collection

Karen Cheng, Ph.D., Estrellita Co-Principal Investigator, University of California, Irvine

About two months ago, we did our first baseline interview. So far, four parents have started in the intervention group (and have received mobile devices with our app) and two parents in the comparison, or control, group. Two more participants are scheduled to start the study by the end of this week.

So far, our intervention participants have given us very positive feedback that they enjoy using the Estrellita system. The two families with the sickest babies seem to be the most consistent in recording their observations of daily living (ODLs), and one of those families even reported using their baby’s ODL data during an appointment with a specialist.

Not surprisingly, most of our participants identify themselves as early adopters of new technology — even if they cannot afford (and therefore do not have) the newest technology.  They report being drawn to this study for the opportunity to use the latest technology.

One interesting discovery was that parents were concerned about weighing their babies when the weather became colder. We were not seeing weight data come in regularly for most of the babies, and we gently reminded parents to let us know if they were having any trouble with the scales. One parent finally responded, explaining that her baby had been sick and she was reluctant to put him naked on a cold scale. An EDAC case manager contacted the parent and explained how to zero out the scale with a blanket and then put the baby on the scale. Problem solved, and weight data came in the next day!

As difficult as recruiting has been, scheduling the interviews has been just as difficult and maybe even more difficult. I spent much of last Friday afternoon in my car in the parking structure of CHOC Children's Hospital, waiting for a participant to return from an unexpected errand for our scheduled interview. My colleague, Karen Tang, spent all of Friday afternoon in a parking lot in another part of Orange County, waiting for a different participant. That participant missed her scheduled interview as well.

I look forward to our upcoming mid-study interviews, and hope to learn more about parents’ experiences with the Estrellita system and their overall strategies for managing their children’s health information in the midst of such hectic lives.

Learn more about the Estrellita (formerly FitBaby) project.
Read more posts from the Estrellita (formerly FitBaby) team.

Creative Commons photo by biblicone.

If I’d Only Known Then What I Know Now (A Letter to Myself)

Katherine Kim, M.P.H., MBA, iN Touch Principal Investigator, San Francisco State University

Dear Kathy,

Three years from now you are going to be finishing the iN Touch project. You’ll have done some things well, made plenty of mistakes, and learned so much. I’m writing this letter to share some of those lessons.

1. You and Christina made the right decision to apply for the Project HealthDesign grant, even though you were worried because those grants from the Robert Wood Johnson Foundation are so competitive.


2. When you start hitting road bumps, lean on your team. Christina, Hali, Judy, Cam-tu, Holly, Ed, Janelle, and Yulia will amaze you with their creativity, optimism, gumption and all-around smarts.


3. The IRB approval process will take twice as long as you expect. The phased IRB that Barbara Massoudi suggests will be the way to go. Submit the application design and testing IRB right away, then work on the intervention IRB which will probably undergo some changes based on the testing. Just-in-time IRB will not be soon enough.


4. Recruitment will take twice as long as you expect. Prototype your recruitment process with the clinicians and staff. Don’t be discouraged by the disconnected phone numbers and no-shows. Remember that life is complicated and messy for the patients. Start recruiting as early as possible.


5. Over-recruit. I just said life is complicated and messy for patients. Even if they want to finish the study, a third of them won’t be able to. Don’t feel bad, this is typical for health and lifestyle studies and those studies with younger and minority participants have even higher dropout rates. You’ll need to recruit more to compensate.


6. Developing attractive, useful reports of ODL data will be harder than you expect. The clinicians won’t know exactly what they want so you’ll have to start out with a mock-up. Recruit a visualization/design expert next time.  Oh yeah, the hospital, clinic, and school will all be on different browsers and versions and they won’t want to update just for you. So make sure to test the reports on multiple browsers.


7. Use multiple communication modes for reminders and scheduling. Paper notes still work!  Deliver them to the classroom to remind youths to come to their appointments. Some of them will want reminders to record ODLs through their iPod Touches, others via text messages to their phones. Forget about email – they never check email.


8. Be ready to educate and evangelize. Hospital, clinic, and school staff members haven’t heard of observations of daily living (ODLs) before and they’re not sure how they feel about mobile technology. You are not just asking them to partner with you on a project, you are asking them to change how they work and engage with patients.


9. The youths who participate in the study will amaze you. They are motivated and excited to use the technology and new strategies to meet their health goals. They’ll teach you a lot about improving health behaviors.

Have a great time with the iN Touch project.

Kathy

The Challenges of Locating Parents of Preterm Infants

Karen Cheng, Ph.D., Estrellita Co-Principal Investigator, University of California, Irvine

In the best of circumstances, it requires a whole lot of patience to recruit parents of preterm infants (born at <32 weeks gestational age). The parents are stressed, sleep-deprived, and may be caring for other children at home. We try to catch the parents in the Neonatal Intensive Care Unit (NICU) when they are visiting their babies. If we cannot find them there, we try to reach them by phone after their babies have been discharged from the NICU. We call and leave messages. We call and call again. For parents with stable jobs and stable home environments and who have active cell phones, we do eventually reach, well, some of them.

To complicate matters, we have been finding that a number of life factors make it even more difficult to reach and recruit these parents. There is the possibility of post-partum depression, for example, in the case of the mom whose voicemail inbox has been full for over a month. Some parents may have unstable home environments, as in the case of parents who do not have their own phone number and must be contacted through family members and friends; they may or may not live with these family members or friends, making it very difficult to pin down when they might be available at that number. And still other parents simply can no longer be reached at the numbers they gave the hospital upon discharge: in one case, the parent’s phone number stopped accepting calls within a few days of discharge.

Our colleagues at the Early Developmental Assessment Center (EDAC) face these challenges every day. EDAC and other high-risk infant follow-up programs across the country face overall attrition rates between 10-25%; rates are estimated to be nearly 50% for people of lower socioeconomic status. One key problem is the inability to reach parents to remind them to come in for appointments.

It’s a problem that appears deceivingly simple: contacting parents to tell them about the study or to remind them of their babies’ EDAC appointments. Yet it is surprisingly challenging and in the case of the EDAC appointment, it may have adverse effects on the babies’ health and development. Dini Baker, R.N., CHOC EDAC manager and Estrellita clinical partner, jokes about how nice it would be to give every baby a GPS tracking device upon leaving the NICU. I wonder if it might make it easier to contact the parents if U.S. cell phone plans allowed incoming texts or phone calls to be free – or if incoming texts or phone calls from certain phone numbers were free. For the parents whom we are able to contact and enroll in the study, our hope is that recording and sharing their babies’ ODLs with EDAC nurse case managers will increase their communication with EDAC and their likelihood of attending EDAC appointments, even after the study is over.

We’re continuing to wrestle with and brainstorm solutions for this problem. If you have any suggestions, we would love to hear them!

Learn more about the Estrellita (formerly FitBaby) project.
Read more posts from the Estrellita (formerly FitBaby) team.

 
Tweet

Identifying Potential Participants from an EMR, Part 2: Now That’s More Like It!

Stephen Rothemich, M.D., M.S., BreathEasy Co-Principal Investigator and Clinical Lead

Last November I wrote (okay, whined) about the challenges to selecting a pool of potential subjects for the design phase of the BreathEasy project. At that time, the limited data in our newly implemented EMR made it necessary to use medication lists to find patients. It was both complicated (I needed an EMR data programmer to do the query) and messy (I did a lot of chart reviews to winnow a list of 271 down to 151).

What a difference a few months has made! We needed to identify a slightly different patient pool for the evaluation phase. In the interim, clinicians at our two study practices, including me, have been diligently inputting medical conditions into the EMR’s problem list. With that additional data, this time we simply looked for patients with asthma on the problem list, a prescription for an asthma controller medicine prescription in the last 12 months (a proxy for moderate to severe asthma) and residence in an urban zip code. A second advantage was that had I learned to use the EMR data query tool. As a result I could do the query myself, come up with a list of 83 patients that I was confident met our selection criteria, and avoid hours of chart reviews. After just a few days of calls, we've already recruited a quarter of the30 participants needed. If necessary, it will be simple for me to expand the pool a bit by replacing the maintenance medication criteria with a prescription for a steroid taper (used to treat asthma exacerbations, so another reasonable proxy for more than mild asthma).

This time it was actually fun. Bottom line: Better data plus a data retrieval tool I could use myself equals better results with a lot less effort.

Learn more about BreathEasy.
Read more posts from the BreathEasy team.

The Cohort Selection Process for Crohnology.MD

Elena Rodriguez Fisher, R.N., FNP-C, Crohnology.MD Clinical Partner

Would you like an iPad?

Offering an iPad in exchange for participating in a study that aims to improve patient-provider communication instantly gets people’s attention. But even without the iPads, recruiting 30 study participants with moderate to severe Crohn’s Disease is far from difficult.

Patients with Crohn’s may require immunosuppressive medications to help control their disease and many have had complications that have necessitated surgery. As a result, Crohn’s patients tend to be highly motivated and are often up for trying anything that may limit the symptoms of the disease or improve their quality of life. The harmful effects of assorted medications, therapies, surgeries, and impositions on personal and professional pursuits tend to foster this openness toward new approaches if there is a chance their health may be improved.

The selection criteria for our study require that participants must:

1. Currently be under the care of a health care provider at the University of California, San Francisco Center for Colitis and Crohn’s Disease;
   
   
2. Possess the means and be willing to communicate with their provider electronically;
   
   
3. Be willing to attend clinical follow-up visit(s) with their provider during the study period;
   
   
4. Be open to participating in focus groups, completing surveys and using technology to collect and integrate observations of daily living (ODLs);
   
   
5. Be able to read, write and speak English; and
   
   
6. Be eighteen years old or older at the time the study begins.  

Potential participants received informational handouts at scheduled office visits or after contacting their providers via email. These handouts instructed them to contact the nurse practitioner at the Center for Colitis and Crohn’s Disease via email if they were interested in enrolling in the study. Most candidates quickly responded to express interest. When discussing the study, many potential participants even disclosed that they already tracked their symptoms using paper or electronic diaries, spread sheets, or timelines.

Most candidates were selected because they have a history of emailing their health care provider two or more times a year or because they expressed an interest in doing so. Most of the candidates who declined to participate in the study cited the time commitment or their current life demands as the primary reasons for refusing to participate. The selection process occurred over a 2 1/2 month  period, and the study enrollment goal was reached prior to study initiation.

Study participants are genuinely excited to be involved; though, as one might expect, they do frequently ask, “When do we get our iPads?”

Learn more about the Chronology.MD project.
Read more posts from the Chronology.MD team.

Incentives for Research Participants in Personal Health Information Technology Studies

Katherine Kim, M.P.H., MBA, iN Touch Principal Investigator, San Francisco State University
Holly Logan, iN Touch Project Coordinator, San Francisco State University

I was recently asked whether we were using incentives to motivate our participants to use iN Touch. In designing our incentive strategy we combined our team’s experience with youth and the advice of the Project HealthDesign National Advisory Committee. We wanted to promote participation in research but we didn’t want incentives to be an external motivator for health goal attainment. We also wanted to minimize participants’ disappointment if they were selected for the control group.

Our strategy is to provide iPod Touches to all intervention group participants, which they will keep upon completion of the project. Control group participants are randomly selected to receive either an iPod Touch or an iPod Shuffle, but are guaranteed to receive one of these items. In addition, we offer a $10 gift card from Target or iTunes for each focus group or interview. The 13- to 18-year-olds prefer iTunes and the 18- to 24-year-olds usually select Target. We do not provide any incentives for use of the application, collection of observations of daily living (ODLs) or visits with a health coach or clinician.

We wondered how other research projects were approaching the use of incentives so we did a little research. AHRQ recently published an excellent report on the impact of consumer health informatics applications.¹ That was a good starting point. We conducted a quick review of 12 randomized clinical trials that showed a positive relationship between the use of personal health information technology and either health outcomes or intermediate outcomes. This is what we found:

Three papers didn’t report on incentives, four reported giving no incentives, and five gave incentives. Here is a summary of the five papers that reported on incentives:

• An Internet self-help application for test anxiety for college students provided a $5-10 incentive for filling out pre and post surveys.²
• An Internet weight loss intervention for African-American adolescent girls provided a subsidy for the purchase of a computer and compensation for time for completing each of four assessments (adolescents received a gift valued at ~$10, parents received $30).³
• A computer-aided behavioral weight loss program for adult women offered a pedometer and $9 per appointment.⁴
• A computer-based health information support system for HIV/AIDS patients offered an incentive for each of three surveys (no amount stated).⁵
• A comparison of computer-tailored counseling and email counseling for overweight/obese adults offered $25 for first visit, $50 for second visit and discounts on meal replacement drinks.⁶

This mini-review showed that detailed incentive strategies were inconsistently described in clinical trials of personal health information technology. We’d love to know more about what others in the field are doing and what the impact of incentives are on behavior or adoption of technology.

References

1. Gibbons MC, W. R., Samal L, Lehmann CU, Dickersin K, Lehmann HP, Aboumatar H, Finkelstein J, Shelton E, Sharma R, Bass EB. (2009). Impact of Consumer Health Informatics Applications. Evidence Report/Technology Assessment No. 188. (Prepared by Johns Hopkins University Evidence-based Practice Center under contract No. HHSA 290-2007-10061-I). Rockville, MD: Agency for Healthcare Research and Quality.

2. Orbach G, Lindsay S, Grey S. A randomized placebo-controlled trial of a self-help Internet-based intervention for test anxiety. Behav Res Ther 2007; 45(3):483-96. 134. March S, Spence SH, Donovan CL. The Efficacy of an Internet-Based Cognitive-Behavioral Therapy Intervention for Child Anxiety Disorders. Journal of Pediatric Psychology 2008.

3. Williamson DA, Walden HM, White MA et al. Two-year internet-based randomized controlled trial for weight loss in African- American girls. Obesity (Silver Spring) 2006; 14(7):1231-43.

4. Adachi Y, Sato C, Yamatsu K, Ito S, Adachi K, Yamagami T. A randomized controlled trial on the long-term effects of a 1-month behavioral weight control program assisted by computer tailored advice. Behaviour Research and Therapy 2007; 45(3):459-70.

5. Gustafson DH, Hawkins R, Boberg E et al. Impact of a patient-centered, computerbased health information/support system. American Journal of Preventitive Medicine 1999; 16(1):1-9.

6. Tate DF, Jackvony EH, Wing RR. A randomized trial comparing human e-mail counseling, computer-automated tailored counseling, and no counseling in an Internet weight loss program. Archives of Internal Medicine 2006; 166(15):1620-5

Learn more about the iN Touch project.
Read more posts from the iN Touch team.

Usability Testing: What We Should Have Known

Rita Sembajwe, Health Informatics Specialist, BreathEasy Project Manager, RTI International
Sarah Cook, Survey Methodologist, BreathEasy Usability Testing Lead, RTI International

We just completed usability testing for both our Android smartphone application for patients and our clinician dashboard. Overall, it was a success and we received great feedback on the look, feel and functionality of both technologies. Equally as important, we learned a lot about what we would need to consider the next time around as we recruit for and carry out usability testing sessions.

When we were recruiting for usability testing sessions, we needed to be more prepared for patient no-shows.  Usability testing is dependent on the turnout and participation of the relevant testing audience. You might think to plan for one or two no-shows, but what could you do to prepare for a 50 percent turnout rate? How could you encourage the participants to show up?

Although our patient usability testing session slots filled up quickly when we were recruiting, we had trouble getting at least half of our patients to turn up for the actual sessions. At the last minute, we managed to reschedule and interview a few, but the effort was an unexpected undertaking. In retrospect, we now recognize several factors we needed to consider:

• Our patient population may have higher odds for no-shows due to possible depression and anxiety co-morbidities that influence their behavior.
• There may be challenges with public transportation and finding the session locations.
• Participants may not understand exactly what usability testing entails.

We now realize that there are a few things we could do proactively to increase attendance rates in the future. We could:

• Follow up with participants more frequently after they agree to participate in the sessions.
• Hold sessions in the most convenient locations.
• Increase the incentives for participation.
• Identify back-up participants in case we need additional people at the last minute.

The usability testing session for the smartphone application also taught us that we must schedule time for a crash course on how to use the smartphone device prior to or at the beginning of the usability testing sessions. Such planning would ensure that the usability testing sessions focus more on testing the application and less on learning how to use a smartphone.

Finally, we believe it would benefit us to incorporate a review of the patient smartphone application into the clinician dashboard testing session, preferably toward the end. From our interactions with clinicians, it is clear that they would appreciate seeing the technology their patients would be using to submit ODL data, and that it would offer them a greater understanding of how the patient-generated data translates into feedback for them on the clinician dashboard. 

We learned many lessons — all worth consideration. We plan to prepare for all of these the next time around!

Recruiting is About More Than Recruitment

Holly Logan, iN Touch Research Assistant, San Francisco State University

Before we started recruitment and enrollment for iN Touch, I had imagined it would be a straightforward process. I would meet a candidate, describe the project, schedule a visit and enroll her in the study in a timely manner. I figured that once I turned the new study participant over to the health coach, my job as a recruiter would be complete.

Boy was I wrong. I did not realize then how much the experience would impact me. Because we are enrolling youth, we must get parental consent before a participant joins the study. Then, we meet with each participant to complete the study intake questionnaire and measurements. This means that the process often requires multiple phone calls and visits. I have come to know these individuals better than I had anticipated and have learned a lot about their lives.

One individual, Yolanda*, has cancelled three enrollment appointments so far. At first, I was annoyed. But then I realized that she has to work many shifts each week, and is often called in to work at a moment’s notice. Even though Yolanda has expressed a desire to participate in our study, she has little control over many of the pressures in her life.

Her story has made we wonder how it is that two people, or even youths who are the same age, can live such different lives. How do I find a balance between detachment and empathy if a participant chooses to open up about a challenging personal situation? I have found that recruitment can be difficult, emotional and rewarding all at the same time, which makes it important to learn how to deal with different issues that may come up. Each day has been a learning process for our whole research team. The discussions we have as a team have been crucial to making sure we all get the support we need from our colleagues. iN Touch is changing in an important way not only the lives of the teens and young adults who participate in the study, but also those of us working on the study.

*Yolanda is not the participant’s real name.