Staff Update
Spring 2008
Participant Retention in Clinical Research: A Major Challenge
Linda Gonder-Frederick, Ph.D.
The Problem of Participant Retention and Adherence
In any type of clinical research, but especially longitudinal studies, participant retention poses a major challenge and issues of dropouts and non-adherence can become frustrating. In weight loss and lifestyle change research, the average dropout rate is 32% (Davis and Addis, 1999), but this can be even higher depending on the participant population, the length of the study, and the demands of the protocol. In most clinical research, each participant represents a significant amount of time, effort and other resources, so high dropout rates are quite costly. High dropout/non-adherence rates are not only frustrating and costly, but they also pose a risk to the interpretation and validity of research findings. This problem is extremely important in translational research, such as the ICAN project, where the goal is to create effective programs that can be widely employed and are useful to large numbers of people across a broad geographical range.
Because of the threat to scientific integrity and validity, recent research has focused on the dropout problem, with two major questions: 1) What are the characteristics of participants who are more likely to dropout or become non-adherent?
2) What strategies are most helpful in improving rates of participant retention and adherence? This paper will review some of the findings from this growing area of research and the implications of these findings for researchers who are designing and conducting long-term clinical research projects.
Characteristics of Participant Dropouts
Several studies have investigated demographic characteristics of participants who tend to dropout of research protocols. One characteristic predictive of dropout appears to be age, with younger participants (< 50 years old) at significantly higher risk than older participants (Glasgow et al.; Honas et al, 2003). One explanation for this finding is that younger participants face numerous family demands and obligations which may not affect older individuals. In addition, older participants may be experiencing declines in health which might motivate them to continue in research protocols in hopes of achieving health benefits. Regardless of the reason, the implication of these findings is that it is important for researchers to find ways to improve the rates of study completion in younger participants. Studies have also found that females and divorced participants are more likely to dropout, which is again attributed to family and caretaking demands (Janson et al, 2001; Honas et al, 2003). However, others have reported that single participants (El-Khorazaty et al, 2007) and fathers (Moser et al, 2000) have a higher dropout rate.
Some studies have reported that minorities are also more likely to drop out (Janson et al., 2001) which may be related to SES and increased life stress. However, minorities are also less likely to consent to participate in clinical trials, which has been attributed to lack of trust in the health care system in general (Kennedy et al, 2005). Higher dropout rates in unemployed participants, as well as those with lower education levels have also been noted (El-Khorazaty et al, 2007). When interpreting all of the above reports, it is important to note than some studies have found no demographic variables that predicted dropout (Blanton et al, 2006).
In addition to demographic factors, psychological and behavioral characteristics appear to predict higher dropout risk. Substance use and abuse has been cited as a risk factor for dropout by several studies (El-Khorazaty et al, 2007; Hansen et al 1985)). Moser (2000) found higher dropout rates in participants who scored higher on measures of depression, hostility, psychological distress, and negative attitudes about health care. Characteristics typically viewed as positive may also be associated with higher dropouts. Glasgow et al. (2007) found that participants who scored higher on measures of self-efficacy had lower retention rates, perhaps due to their ability to resolve issues that motivated them to join the study on their own.
Causes of Participant Dropout
Several studies have interviewed participants who dropped out concerning the reasons for their failure to complete the protocol. The reasons cited most frequently tend to reflect issues related to competing life demands, logistical problems, demands of the study, and lack of motivation/commitment. By far, the most commonly cited reasons are stress related to family care responsibilities and interference with work (Jansen et al, 2001; Parra-Medina et al, 2004). Another common reason is lack of time, as well as complicated and cumbersome record-keeping and paperwork associated with a study. A logistical barrier frequently cited is difficulty with transportation and inconvenience of study site location, including distance and parking. Other logistical issues include the timing of appointments with study staff, and the need for flexibility in times and dates available for meetings and data collection. (Jansen et al, 2001; Tansey et al, 2007).
Positive reinforcement and participant motivation also play a major role in decisions concerning whether to complete or drop out of research protocols. Dropouts occur when participants' perceived time and effort invested outweigh the perceived benefits of being in a study (Berger & Neumark, 2007). Reasons participants give for completing research protocols also reflect incentive and motivation, and include remuneration, a commitment to finish, and a belief that the study is important (Jansen et al, 2001). All of these factors that determine whether participants remain in protocols or dropout obviously have implications for developing strategies that enhance the likelihood of study completion.
Logistical Strategies that Improve Participant Retention
In general, developing protocols that make research participation as convenient as possible appears to be one of the most effective strategies. Providing transportation or transportation vouchers to participants has been associated with higher than average retention rates (> 80%) in several studies (Parra-Medina et al, 2004; Tansey et al, 2007). A pilot study of a peer-led exercise program offered in a local church achieved a retention rate of 90% (Kennedy et al, 2005). Offering participants flexible appointment/meeting hours, including evening and weekend slots, can increase retention rates to above 80% (Tansey et al, 2007). Other logistical strategies that increase convenience and retention are collecting data with home visits and providing child care during meetings and refreshments (Berger & Neumark, 2007; Tansey et al, 2007. Retention and adherence can also be enhanced by alleviating the burden of remembering study meetings by providing calendars, schedules, and pre-appointment reminders to participants.
Of course strategies such as the ones described above take an enormous amount of resources, including staff time, and the cost of creating an "ideal" protocol in terms of participant convenience is typically prohibitively high. However, as awareness of the importance of participant retention in clinical research continues to grow, it may become more common for research projects to increase their requested budgets to support the strategies needed to reduce dropout.
Strategies to Increase Participant Motivation
Because of the critical role of "perceived benefits" in research participation, most studies provide some type of incentive, including monetary payment. Numerous types of token incentives are also utilized including, t-shirts, refrigerator magnets, coffee cups, lunch bags, and pedometers (Berger & Neumark, 2007; El-Khorazaty et al, 2007; Parra-Medina et al, 2004; Tansey et al, 2007). These tokens not only act to enhance incentive, but can also serve as reminders to participants that they are in a project or methods for strengthen the participants' sense of "identity" with the project. Medical insurance programs are just beginning to explore the role of incentives in improving health care and reducing illness related to lifestyle factors, but research supports the potential of this strategy. For example, one recent study found that offering annual membership in a local gym for free was more important for retention in a weight loss program than any of the patient education or social support provided to participants (Yancey et al, 2006).
Frequent personal contact is another essential strategy to maintain participant motivation. Some methods include monthly phone calls, sending birthday and holiday cards, thank you letters, and periodical newsletters. Most of the authors cited above would agree that strategies such as these also serve to create a positive relationship between participants and the study, which is a key factor. In fact, the degree to which there is a positive relationship between study participants and staff may one of the most important determinants of motivation to complete protocols (Tansey et al., 2007; Blanton et al, 2006). In addition, strategies can be created to remind participants of the value of being in a study, or the personal benefits of the study to them, to maintain level of motivation or increase motivation when it begins to wane.
Recent research has also looked at the effect of using Motivational Interviewing Techniques (MIT) to increase participant motivation and reduce dropout rates in lifestyle research protocols. In one study, MIT was employed at group orientation meetings to address ambivalence about making changes in diet and exercise, participating in a randomized controlled trial, and unrealistic weight loss expectations (Goldberg and Kiernan, 2005). Over 95% of these participants completed a protocol involving clinic visits over an 18-month period. Another study involving an 18-month weight loss program randomized female participants to either a MIT or control group (West et al, 2007), and found a 98% retention rate in the MIT group, who also lost significantly more weight. However, MIT did not appear to be as beneficial for African-American women compared to Caucasians, and the reason for this difference was unclear.
The Importance of Data Tracking Systems
The challenge of participation retention in clinical research has also caused scientists to devote more time and attention to data tracking systems, whose role in identifying and resolving problems cannot be overemphasized. Improvements in data tracking increase the ability to identify problems in participant contact and adherence before dropouts occur (El-Khorazaty et al, 2007). This, in turn, assists research staff in responding to problems in a timely manner, including arranging for make-up sessions when appointments are missed (Parra-Medina et al, 2004). Systems also need to provide ways to contact participants who become difficult to reach or lost to follow-up. One helpful tool is to obtain multiple ways to contact individual participants (e.g., home and work phone numbers, phone numbers of family members and friends), and to update these contacts periodically, especially in long-term studies. Again, creating optimal data tracking systems and procedures requires that adequate staff time and resources can be devoted to this important task, which often presents budgetary challenges.
The Emerging Model of Participant Retention in Clinical Research
One of the central "take home" messages of this literature review is that dropout rates and non-adherence are pervasive problems in clinical research. Even studies that use multiple techniques (e.g. incentive payments, token gifts, frequent contact) may still have low retention rates. Nonetheless, it is highly recommended that studies use multiple methods to enhance participant retention, and that these methods address multiple barriers and facilitators of research participation, such as motivation, convenience, and data tracking (Davis, Broome & Cox, 2002).
It is now widely recognized that high dropout rates in clinical trials are a major threat both to the ability to conduct critical research and to generalize results to broader populations. Many researchers believe that this problem will only worsen with certain trends in contemporary society such as longer working hours, demanding schedules outside of work, increased stress and sleep deprivation. On a positive note, this problem is forcing researchers to directly address the need to develop measures that make it feasible for people from all walks of life to participate in clinical trials. It also increases awareness for the need to respect the time and effort volunteers invest in clinical research, as well as the need to increase cultural sensitivity.
Another positive change is a moving away from the older human research perspective, which assumed that volunteer subjects "owed" it to a project to complete all data collection once they had signed an informed consent form. Instead, the current model of participant retention actually looks more like a "customer-service" approach (Roberg, Carlson-Dakes & Anderson, 2006). In this approach, it is the responsibility of the research project to make its protocol appealing enough to attract volunteers. Once volunteers have consented, it is the project's responsibility to "keep the customer happy" as much as possible. This requires an increased awareness that volunteers have choices concerning whether or not to participate in clinical research, and that they deserve respect and gratitude when they donate their time and energy. The customer-service approach also places a higher value on making studies more convenient for volunteers, by developing protocols that are not logistically burdensome or too complex and demanding. Finally, incentives and benefits are of utmost importance in this approach which recognizes that, just like a business depends on customer satisfaction, the success of clinical research depends on providing participants with an experience they perceive as rewarding.
References
Berger A, Neumark D. Enhancing recruitment and retention in randomized clinical management. Oncology Nursing Forum, 2007, 34, e18.
Blanton S, Morris D, Prettyman M, McCulloch K, Redond S, Light K, Wolf S. Lessons learned in participant recruitment and retention: The EXCITE trial. Physical Therapy, 2006, 86, 1520-1523.
Davis L, Broome M, Cox R. Maximizing retention in community-based clinical trials. Journal of Nursing Scholarship, 2002, 34, 47-53.
Davis M, Addis, M. Predictors of attrition from behavioral medicine treatments. Annals of Behavioral Medicine, 1999, 21, 339-349.
El-Khorazaty M N, Johnson A, Kiely M, El-Mohandes A, Subramanian S, Laryea A, Murray K, Thornberry J, Joseph J. Recruitment and retention of low-income minority women in a behavioral intervention to reduce smoking, depression, and intimate partner violence during pregnancy. BMC Public Health, 2007, 7, 233.
Glasgow R, Nelson C, Kearney K, Reid R, Ritzwoller D, Strecher V, Couper M, Green B, Wildenhaus K. Reach, engagement, and retention in an Internet-based weight loss program in a multi-site randomized controlled trial. Journal of Medical Internet Research, 2007, 9, e11.
Goldberg J, Kiernan M. Innovative techniques to address retention in a behavioral weight-loss trial. Health Education Research, 2004, 20, 439-447.
Hansen W, Collins L, Malotte C, Johnson C, Fielding J. Attrition in prevention research. Journal of Behavioral Medicine, 2004, 8, 261-275.
Honas J, Early J, Frederickson D, O'Brien M. Predictors of attrition in a large clinic-based weight-loss program. Obesity Research, 2003, 11, 888 - 894.
Janson S, Alioto M, Boushy H. Attrition and retention of ethnically diverse subjects in a multicenter randomized controlled research trial. Controlled Clinical Trials, 2001, 22, 236S-43S.
Kennedy B, Paeratakul S, Champane C, Ryan D, Harsha D, McGee B, Deyhim J, Forsythe W, Bogie M. A pilot church-based weight loss program for African American adults utilizing church members as health educators: A comparison of individual and group intervention. Ethnicity and Disease, 2005, 15, 373-378.
Moser D, Dracup K, Doering L. Factors differentiating dropouts from completers in a longitudinal multicenter clinical trial. Nursing Research, 2000, 49, 109 - 116.
Parra-Medina D, Antonio A, Smith S, et al. Successful recruitment and retention strategies for a randomized weight management trial for people with diabetes living in rural, medically underserved counties of South Carolina: The POWER Study. Journal of the American Dietetic Association, 2004, 104, 70-75.
Roberg K, Carlson-Dakes K, Anderson B. The key to successful recruitment and retention in clinical research programs. Monitor, 2006, June, 13-16.
Tansey C, Matte A, Needham D, Herridge M. Review of retention strategies in longitudinal studies and application to follow-up of ICU survivors. Intensive Care Medicine, 2007, 33, 2051-2057.
West D, Dillilo V, Bursac Z, Gore S, Greene P. Motivational interviewing improves weight loss in women with type 2 diabetes. Diabetes Care, 2007, 30, 1081-1087.
Yancey A, McCarthy W, Harrison G, Wong W, Seigel J. Challenges in improving fitness: Results of a community-based, randomized, controlled lifestyle change intervention. Journal of Women's Health, 2006, 15, 412-429.
HIC# 11143