The Robert Wood Johnson Foundation awarded the University of California, Irvine, (UCI) Department of Medicine a grant to study the question of where do patients prefer to receive care when they have a minor illness or surgery. The study was led by Drs. Dana Mukamel, Dara Sorkin and Alpesh Amin and conducted during 2016-12017.
To address this question we developed a survey that was administered to all UCI employees. The survey presented respondents with 12 scenarios, each describing a different clinical situation, time of day, out of pocket cost, wait time and travel time. Respondents were asked to consider each scenario and indicate which one of 8 care settings they prefer under the given situation: their doctor’s office, the emergency room, urgent care center, retail clinic, a virtual physician they could access through their smart phone or computer, calling a nurse on a nurse advice line, paying for a physician to come to their home, or adopting a “wait and see” strategy.
The survey also asked respondents about their age, gender, education, income, health status, family composition, insurance status, and employment.
Responses were analyzed statistically to provide 12 models predicting where individuals will choose to receive care given the clinical condition, time of day, the cost they are facing, the time they have to spend and their individual characteristics. These models are embedded in this web site and drive the predictions you can obtain from the site.
The population included all employees of the University of California, Irvine campus (UCI). UCI was chosen for the study because it employs individuals in all occupations, including, but not limited to, medical services (the hospital and clinics and their associated services, ranging from clinical such as laboratories, to cafeteria and hospitality), teachers and students, police officers, housing services, transportation, landscaping and janitorial services, electrical and building maintenance, factory workers, clerical workers, retail shops and more. Its employees include individuals form all educational and income strata, all level of socio-economic and demographic groups, including individuals for whom English is not the first language. All 21,037 UCI employees received invitations to participate in the survey. Of those 5,451 responded resulting in a response rate of 26%. 5,202 (95%) completed at least one clinical scenario and 4,855 (93%) of those had complete data for all variables and were included in our final models.
The survey included several questions about socio-demographic and economic status; questions about health status; and a preference elicitation module which asked respondents to indicate which of eight care-settings they would choose in each scenario. Each scenario was presented separately with pictures of the eight care-settings. Each care-setting picture also included information about the out-of-pocket costs (OPC), wait time, and travel time the respondent should associate with the setting. The values for these parameters were based on a survey of providers in the UCI vicinity. These parameters were fixed for all the scenarios each respondent was presented, but randomized across respondents. Once developed, the survey was tested with pilot subjects similar in their background to our target population, and revised based on their feedback.
Survey invitations were sent to all employees using their UCI email addresses. The emails came from the UCI Vice Chancellor for Health Affairs, explaining the purpose of the study and its importance, as well as promising confidentiality. Survey respondents were eligible to participate in a sweepstake with prizes ranging from $100 to $575. To increase the response rate the Deans of the Schools and the Directors of non-academic departments were asked to send follow-up emails to their employees as well. Respondents also received up to 10 email reminders. Those employees without regular access to computers (e.g. janitorial staff) were invited to attend several sessions in which they could access the survey on iPads. The survey was open between September 27 and December 7, 2016.