Abstract
Objective(s) Our objective was to investigate the motivators and barriers associated with the individual or family decision to participate in cerebral palsy research. Based on this information, we offer suggestions to increase the likelihood of participation in future CP studies.
Methods A digital survey was administered to stakeholders affected by cerebral palsy across the US. Our analysis focused on variables related to personal interests, travel, and study-specific elements. Statistical tests investigated the effects of responder type, cerebral palsy type, and Gross Motor Function Classification System level on travel and study-specific element variables. Recommendations were informed by responses reflecting the majority of respondents.
Results Based on 233 responses, we found that respondents highly valued research participation (on average 88.2/100) and compensation (on average 62.3/100). Motivators included the potential for direct benefit (62.2%) and helping others (53.4%). The primary barriers to participation were schedule limitations (48.9%) and travel logistics (32.6%). Schedule limitations were especially pertinent to caregivers, while individuals with more severe cerebral palsy diagnoses reported the necessity of additional items to comfortably travel.
Conclusions Overall, we encourage the involvement of stakeholders affected by cerebral palsy in the research process. Researchers should consider offering flexible study times, accommodating locations, and compensation for time and travel expenses. We recommend a minimum compensation of $15/hour and a maximum time commitment of 4 hours/day to respect participants’ time and increase likelihood of research participation. Future studies should track how attitudes toward research change with time and experience.
Introduction
Cerebral palsy (CP) is a broad pediatric-onset diagnosis caused by a non-progressive injury to the developing brain [1]. The etiology of CP is also extremely heterogeneous, often resulting from brain injuries that occur during the early developmental period [2–5]. Although CP is considered a pediatric-onset disorder, the associated physical and behavioral presentations are present across the lifespan and may fluctuate in severity over time. As there is currently no cure for CP, research efforts are critical for advancing our understanding of the pathophysiology and most efficacious treatments. However, the diversity of this population poses a significant recruitment challenge to researchers [6]. Limited funding for CP research also puts an added burden on researchers to be efficient with study-specific elements and recruitment, especially in the United States [7]. In an attempt to address this, previous studies have surveyed individuals with CP to identify priority research areas [8–10]. Research registries have also been established to facilitate collaboration among US institutions and to improve communication between researchers and individuals with CP [11,12]. Despite these efforts, the success of CP research is dependent on the desire of individuals to participate and their ability to reasonably access the study within the limitations of their environment.
A previous study investigated the barriers to intervention-based CP research and recommended the involvement of patient populations and their families in the study development pipeline to improve recruitment [6], an approach towards community-based participatory research. However, the study did not consider facilitators to research and only evaluated a small subset of the population interested in home-based training programs. Therefore, our objective was to sample a larger and more heterogeneous cohort of stakeholders and investigate the motivators and barriers associated with the decision to participate in CP research. We aimed to inform researchers of specific stakeholder perspectives by understanding whether factors such as Gross Motor Function Classification System (GMFCS) level, CP type, or age contribute to the decision to participate. Based on a nationwide survey, we provide recommendations for investigators to increase likelihood of recruitment and participation in future CP research.
Methods
I. Recruitment
The survey, including informed consent, was created and administered using the Research Electronic Data Capture (REDCap) platform. It was approved by the Northwestern University Institutional Review Board and remained open between May 6th and July 7th, 2020. Respondents were eligible to voluntarily participate if they resided in the United States and were either 1) caregivers of minors (under 18 years of age) with a diagnosis of CP or 2) adults with a diagnosis of CP. The survey link was shared via several platforms, including the Cerebral Palsy Research Registry [11], ResearchMatch.org, department social media accounts, and emails to previous research participants.
II. Experimental Protocol: Qualitative Survey
The digital open survey was designed to collect data about motivators and barriers of participation in CP research. During survey development, we sought feedback from caregivers of minors with CP and adults with CP on clarity of questions and importance/completeness of content. The online survey (available in supporting information) contained optional questions in six different categories: demographics, personal interests, travel needs & preferences, study-specific elements, past research experience, and impact of COVID-19. The present analysis focused on the first four categories to summarize attitudes towards general CP research. Details of the survey development are described in further detail by Joshi et al. [13].
III. Survey Categories
Demographics
We collected a number of variables to describe features of the respondents, listed in Table 1. In addition to CP type and GMFCS level, we asked about elements such as the time it takes for respondents to get to medical appointments and information about current and previous medical treatments common to study inclusion or exclusion criteria.
Personal Interests
All subsequent variable names are italicized in text and described in Table 2. Personal interest factors were considered intrinsic to the respondent. Respondents were asked about their perception of research importance, how highly they value research participation, and compensation importance, all on a scale of 0-100. Open-ended questions in the survey requested comment on personal goals, motivators, and barriers for participation in CP research. To gauge specific research interests, respondents were asked what study types and body functions were of high interest to them.
Travel Needs & Preferences
Respondents reported whether additional arrangements for childcare would be required, whether time off work would be required, and what additional travel needs would be required when leaving the house. To understand travel preferences, respondents were asked about their typical mode of transportation to medical appointments. We also asked participants to identify their perceived maximum travel time for indirect benefit study, maximum travel time for direct benefit study, and overall willingness to make an extended overnight trip for research participation. The importance of travel reimbursement for local study and travel reimbursement for distant study was also evaluated, both on a scale from 0-100.
Study-specific Elements
There were a number of variables related to explicit design of the study, which have the potential to be modified by the researcher. Respondents were asked about their most preferred study locations and their preferred time of year for study participation. Respondents’ desired compensation amount was evaluated per hour of study participation. Respondents were also asked about the maximum time commitment that was reasonable for one day of participation, the maximum study visits they would be willing to commit, and whether they would consider participating in a longitudinal study.
IV. Data and Statistical Analysis
IBM SPSS Statistics version 26 (IBM Corp., Armonk, NY, USA) was used to perform all analyses on the survey responses. Participants with missing data for a given question were excluded from analysis pertaining to that question. To summarize responses, descriptive analyses were first completed, with all percentages reported relative to the number of respondents for each question. We defined 50% as the threshold to describe the majority of survey respondents. Statistical analyses were only performed on travel and study-specific variables, as researchers can directly use this information to modify study methods during the developmental pipeline. Q-Q plots were created for quantitative variables to assess normality. Kruskal-Wallis tests, Chi-squared tests, or Kaplan-Meier survival analyses were performed on variables hypothesized to be dependent on three factors: Responder type (2 levels: adult, caregiver), CP type (4 levels: hemiplegia, diplegia, quadriplegia, other), and GMFCS level (5 levels: I, II, III, IV, V). A p-value < 0.05 was considered significant. Post hoc analyses were used to determine significant pairwise comparisons, where p-values were corrected for multiple comparisons using Bonferroni corrections. Further analyses were run to test specific hypotheses. For open-ended questions pertaining to personal interests, AH reviewed all responses, identified common themes, and categorized each response accordingly. Categorizations were reviewed and approved by KMZ and VG and summarized semi-quantitatively. For travel preferences, Wilcoxon signed-rank tests were run to determine differences between travel time for indirect benefit and travel time for direct benefit and between importance of travel reimbursement for local study and travel reimbursement for distant study.
Results
I. Demographics
In total, 255 individuals were consented and 233 at least partially completed the survey. Respondent demographics are listed in Table 1. The survey population closely matches US census data in terms of sex, ethnicity, and race [13–15]. The majority of participants reported no previous research experience (53.9%), a proximity to medical appointments of 1 hour or less (75.1%), and previous or ongoing physical/occupational therapy treatment for their arms (53.2%) or legs (65.7%).
II. Personal Interests
All subsequent variables and associated p-values are listed in Table 2. Respondents reported high mean scores for research importance (93.8/100), value research participation (88.2/100), and compensation importance (62.3/100). Open-ended questions revealed that the biggest personal motivators for CP research were personal benefit (62.2% of respondents) and helping others (53.4%) (Figure 1A), while the biggest personal barrier was schedule limitations (48.9%) (Figure 1B). With regard to research interests, the most popular study types were physical or occupational therapy treatments (90.8%), activity monitoring (79.1%), imaging of muscle/bone (72.4%), survey or online (71.5%), robotic games (68.9%), imaging of the brain (68.4%), and new treatments (64.9%). The most popular body functions of interest were the legs/feet (79.9%), muscles (79.5%), movement/fitness (79.0%), brain/nerves (76.0%), arms/hands (62.9%), and pain (50.7%) (Figure 2).
Percentage of respondents who indicated (A) motivators and (B) barriers for participating in research relating to the categories shown. Indented categories are subcategories of the parent category. (C) Representative quotes indicating goals for participating in research.
Body functions of research interest to respondents. Each of these options was offered as a checkbox for respondents to indicate if they would be interested in participating in a study that focused on these body regions/functions. Percentages are out of n = 233 respondents.
III. Travel Needs
Most respondents needed time off work (62.7%) but did not need additional arrangements for childcare to engage in research (66.1%). There was a significant main effect of responder type on the latter, where caregivers of minors with CP needed childcare more than adults with CP. When leaving their homes, the majority of respondents had additional travel needs such as time (59.2%), transportation items such as a wheelchair or stroller (58.0%), and snacks (52.9%) (Figure 3A).
(A) Summary of additional travel needs required to participate in CP research. The most prevalent categories were time, transportation, and snacks. (B) Summary of the most common transportation methods. Driving was the most cited mode of transportation. *6.8% of respondents selected other transportation modes.
Specific additional travel needs varied significantly based on GMFCS level, CP type, and responder type. All significant pairwise comparisons are reported in the supplemental section (S1 Table). There was a significant main effect of GMFCS level on breathing items, seizure items, feeding items, medications, toileting items, and transportation items. In summary, individuals classified as GMFCS level V reported needing these items more to comfortably travel. There was a significant main effect of CP type on time, feeding items, toileting items, and transportation items. In general, individuals affected by quadriplegia reported needing these items more to comfortably travel. Finally, there was a significant main effect of responder type on transition, seizure items, feeding items, snacks, and toileting items. Adults with CP had more concerns about transition to a new environment than caregivers of minors with CP. However, caregivers needed seizure items, feeding items, snacks, and toileting items for their children more than adults with CP did for themselves.
IV. Travel Preferences
The most common mode of transportation was by car, whether the individual drives (63.2%) or gets a ride from a family member (35.0%) (Figure 3B). There was a significant main effect of GMFCS level on transportation methods including drive self, public transit, and other, though there were no significant pairwise comparisons. There was a significant main effect of CP type on drive self, where respondents (both caregivers and adults with CP) affected by hemiplegia preferred to drive themselves more than those affected by diplegia or quadriplegia. There was a significant main effect of responder type on drive self, public transit, ride service, and other. Caregivers of minors with CP preferred to drive themselves more than adults with CP, whereas adults preferred public transit, ride services, or other methods of transportation.
The mean response for travel time for indirect benefit study was 3.53 hours, which was significantly lower (z = -6.857, p < 0.001) than the mean response for travel time for direct benefit study at 4.13 hours. Approximately half (49.8%) of respondents were willing to make an overnight trip for research participation. There was a significant main effect of responder type on the importance of travel reimbursement for a local study, where adults with CP thought reimbursement was more important than caregivers of minors with CP. The overall mean score for this variable (49.1/100) was significantly lower (z = -9.901, p < 0.001) than the mean score for the importance of travel reimbursement for a distant study (75.3/100).
Study-specific Elements
A current clinic (88.7%), home (84.7%), lab (73.0%), and new clinic (59.9%) were the most preferred locations for research participation (Figure 4A). There was a significant main effect of GMFCS level on park, where individuals who are GMFCS level I were more likely to select this location than all other levels. There was a significant main effect of CP type on new clinic and lab, where individuals affected by hemiplegia were more likely to select these locations over those affected by quadriplegia. There was a significant main effect of responder type on current clinic, park, and school, where caregivers of minors with CP were more likely to select these locations over adults with CP.
(A) Summary of preferred locations for CP studies. *5.4% of respondents chose other locations. (B) Survival analysis for maximum time commitment by responder type. Less than 30% of caregivers and adults remain at 4 and 7 hours, respectively. (C) Summary of preferred time of year for participation by responder type.
The majority of respondents were flexible to participate in research at any time of year, except for parents on weekdays during the school year (Figure 4C). There was a significant main effect of responder type on weekdays, where adults reported more availability compared to caregivers. There was also a significant main effect of responder type for summer break, winter break, and other school holidays, where caregivers were more willing to engage in research during these times of year than adults. Respondents indicated a mean compensation amount of $16.69/hour for participation in research. The average maximum time commitment was 3.83 hours/day and the average maximum study visits was 4.28 visits. Survival analyses on maximum time commitment yielded significant main effects of CP type (log rank χ2 (1) = 11.9, p = 0.001), with no significant pairwise comparisons, and responder type (log rank χ2 (3) = 13.4, p = 0.004). Most notably, caregiver interest dropped from 54.1% to 24.1% at a maximum time commitment greater than 4 hours (Figure 4B). Finally, the vast majority of respondents were willing to participate in a longitudinal study (95.5%).
Discussion
The purpose of this study was to determine the motivators and barriers involved in the decision to participate in CP research studies. We administered a survey to gain more insight on stakeholder perspectives, including their personal interest in research, travel needs and preferences for study participation, and study-specific elements. These results can be extrapolated into recommendations for future CP research studies to maximize participant recruitment and expedite new knowledge about CP.
Our study is one of the first to identify the personal and practical factors that influence research participation. Respondents overwhelmingly supported research and valued their own participation in research, in a wide range of topics (Figure 2A). The research areas of interest identified by our survey respondents are largely consistent with previous reports of CP research priorities [8,16]. Survey responses indicated that most individuals were motivated to participate by the potential for personal benefit and helping others. Where relevant, these two elements should be highlighted in recruitment materials and results should be disseminated to participants in a format that is best suited to their learning preferences (e.g. a copy of a manuscript or poster, a one-page summary, or a short video). Schedule limitations were the most prevalent barriers to research participation, especially as caregivers of minors with CP were likely to need additional childcare arrangements. Travel limitations were also a highly cited barrier to research participation. To minimize these barriers, researchers should offer flexible study times, particularly during the weekends and summer break, and/or utilize local study locations close to the home or clinics where participants are receiving care.
When scheduling participants with CP for a research study located outside of their home, travel needs for participants should be considered. To accommodate the additional time required for participants and/or their families to reach the study location, researchers should be flexible with appointment times. Indeed, utilizing flexible study protocols has previously been identified as a recommendation to improve recruitment in CP research [6]. Researchers should also consider having snacks available for participants, particularly for minors with CP. Unsurprisingly, participants classified as GMFCS level V or with a diagnosis of quadriplegia reported requiring more items in order to comfortably travel. Researchers working with these inclusion criteria might consider the home as a study location to mitigate the barrier of travel burden.
Respondents clearly noted the importance of compensation, as providing compensation to participants is consistent with appreciation of their time. In a previous study where caregivers of minors with CP were consulted on the design of randomized control trials, there was a strong preference for coverage of all treatment costs [17]. Caregivers noted that they may not be able to cover the costs themselves, the participants would be offering their time, and the study benefits were unknown [17]. From our survey responses, a minimum compensation of $15/hour and a maximum time commitment of 4 hours/day were interpreted to be respectful of the time and commitment to research participation.
In addition to direct payment, compensation for other costs associated with travel should be considered. If a study session requires a longer duration, additional compensation can include paying for a meal. Because most respondents were willing to travel long distances for studies with and without direct benefit, researchers should also consider offering travel reimbursement. This is especially important because travel limitations were a highly cited barrier to research participation. As the most preferred mode of transportation involves driving, suggestions for reimbursement include gas and parking. Flexibility around transport mode could also include fare coverage for adult participants who prefer public transit or ride services. For participants willing to make an overnight trip for a research study, researchers should consider compensating for lodging and overnight parking.
Our survey-based recommendations are centered around maximizing stakeholder participation in CP research studies. One limitation of our sample was that survey respondents were self-selected and may be biased towards research participation. Their responses may inflate measures of research importance and resource allocation (e.g. time and money), while underestimating obstacles to research participation. However, these individuals may also be more likely to respond to participant requests and therefore would be more representative of future study samples. Another limitation is that survey responses were self-reported at one time point. Future research should determine whether attitudes towards research shift over time or are dependent on the depth of previous research experience. Finally, researchers should be informed about and supported in the engagement of the community. This can span a spectrum of involvement, such as one-time consultations to provide feedback on study-specific elements, formation of Community Advisory Boards, or the inclusion of community stakeholders as project investigators.
We assessed the motivators and barriers to research participation from the perspectives of caregivers of minors with CP and adults with CP. By identifying these stakeholder attitudes and utilizing the information to design study protocols, individuals with CP and their families become true partners in the research that aims to benefit them. Researchers can best accommodate the needs of participants with CP by opting for flexible study locations, scheduling, and compensation options. We recognize that this will not be feasible for all studies but encourage researchers to consider even the smallest gestures to reduce the burden of participation.
Data Availability
Data will be available at Northwestern University's Arch Institutional Repository upon acceptance to a peer-reviewed journal.
Acknowledgements
The authors would like to thank Divya Joshi, Nayo Hill, and Heidi Roth for their help in the design of survey questions and recruitment of participants.
Footnotes
Conflicts of Interest: There are no conflicts of interest.
Abbreviations
- CP
- cerebral palsy
- GMFCS
- Gross Motor Function Classification System
- REDcap
- Research Electronic Data Capture
