Based on Kaplan-Meier plots, kidney (log-rank P<0.001), liver (log-rank P=0.02), and lung (log-rank P=0.01) participants in research had better long-term patient survival than those who did not. These associations were consistent in the multivariable Cox models in which patients who did not participate in research had statistically significantly worse survival among kidney (adjusted hazard ratio [AHR]=1.14, 95% confidence interval [CI]: 1.05–1.22), liver (AHR=1.25, 95% CI: 1.05–1.49), and lung (AHR=1.45, 95% CI: 1.11–1.90) transplant recipients (Table 6). For the same organ transplant groups, the hazards for overall graft loss were also statistically significantly worse among nonparticipants. There was no marked variation in the effect of participation within subgroups of the recipient population. The improved survival among kidney, liver, and lung participants remained statistically significant excluding patients at centers with no reported participants and after adjustment for the participation rate by center.
The distribution of participation rates by transplant center aggregated across organs are displayed in Figure 1. Among the 279 transplant centers included in the study population, 29% of centers reported no participation in research and 43% of centers reported less than 1% of participants. One center reported 58% of patients participating in research studies and 13% of centers had more than 10% participation. There was a significant positive correlation between transplant center volume and the proportion of patients participating in research studies (r=0.34, P<0.001).
Participation varied by type of baseline immunosuppressive medication among each of the transplant populations (Fig. 2a–e). Among kidney transplant recipients, there was higher participation among recipients using cyclosporine, thymoglobulin, and sirolimus at baseline and reduced participation among patients on tacrolimus, steroids, mycophenalate mofetil (MMF), campath, and interleukin (IL)-2 receptor blockers (RB). Among liver transplant recipients, patients on steroids at baseline were less likely to participate than those who are not on steroids. Among heart transplant recipients, participation was greater among patients on cyclosporine and sirolimus and reduced among patients on tacrolimus, MMF, thymoglobulin, campath, and IL-2 RB. Lung participants were less likely to receive MMF or campath at baseline, and SPK transplant recipients were less likely to receive thymoglobulin or campath.
Missing levels during the period of mandatory reporting for participation (through 2005) ranged from 0.2% to 2.2%. The proportions of participation in this period were all within 1% of those reported for the full study population for each organ. Factors associated with participation were generally consistent with the exception that several factors lost statistical significance. The qualitative results of the survival models were also similar in this subset for each organ type, the AHRs for 1-year conditional patient survival limited to the mandatory reporting period for kidney, liver, and lung transplant recipients were 1.08, 1.24, and 1.46, which all remained statistically significant.
The primary findings of the study are that (a) there are systematic differences between participants and nonparticipants of research protocols for immunosuppressive medications among transplant recipients, (b) outcomes are superior among patients who participate in research protocols for kidney, liver, and lung transplant recipients, and (c) participation in research protocols is largely clustered within certain transplant centers and variable by type of immunosuppression medication. Cumulatively, this study suggests that findings from research studies may not be clearly generalizable to the broader transplant population.
Consistent with other healthcare contexts, both demographic characteristics and indicators of patient acuity were associated with patient participation in research (1, 5, 6, 26, 27). Among kidney transplant recipients, participation was associated with markers of higher socioeconomic status such as private insurance, elevated household income, and greater educational attainment. Reduced participation among pediatric and elderly patients, repeat transplant recipients, diabetics, recipients of older age donors, and sensitized patients may indicate that findings from research studies may be less applicable to patients with higher acuity or special treatment requirements. Despite studies highlighting reduced access to care among African Americans kidney transplant patients, there was not a significant difference in participation after adjustment for other characteristics (28, 29). In contrast, African Americans were less likely to participate in research among SPK transplant recipients, which also has documented access disparities (30). Interestingly, higher income was associated with a diminished likelihood to participate in research among liver transplant recipients who may reflect differences in socioeconomic status between end-organ patients or in the type of research studies between transplant populations. For both heart and liver transplant recipients, patients' distance to the transplant center was associated with reduced likelihood to participate, which is consistent with studies demonstrating that the logistical considerations may explain both participation in studies and outcomes (31, 32).
The elevated survival among kidney, liver, and lung transplant recipients who participated in research may have different implications. First, findings derived from clinical trials or other research studies may be a reflection of a healthier cohort of patients. This may be a product of intentional selection of patients more or less likely to exhibit a particular endpoint and exclusion criteria that may not be captured by variables available in this study. Another potential interpretation of these findings is that participation in research itself affords certain benefits. Superior outcomes may derive from various trials effects such as more rigorous monitoring by caregivers, free or discounted care or medications, beneficial interventions in which patients are participating in research protocols, the Hawthorne effect or changes in patient or caregiver behavior.
Results indicated that a minority of patients participate in research for immunosuppressive medications. Sources of this low participation rate requires further study but importantly results indicate that participation is largely clustered by transplant center. Almost one third of centers report having no participants and 13% of centers report at least 10% of recipients in research protocols. This variation was in part explained by center volume, with a significant association between larger centers and participation. If in fact the superior outcomes associated with participation are explained by enhanced levels and access to care, these findings may explain one of the mechanisms by which large centers generally have better outcomes (33, 34). Sponsors of research may be attracted to centers with a larger volume of patients as well as centers with better outcomes. Wide variations in participation at different centers may be salient to decision-making for prospective transplant patients interested in research (35). Center clustering of participants may question whether research studies completed at selected centers are necessarily applicable to smaller centers or those that do not have research protocols. Results also raise interesting questions about the impact of ongoing research and evaluated quality of care between centers. Differences in participation may not only be an indicative of better quality of care at centers with active research but could also be viewed as a potential source of bias for comparing center outcomes with and without active research programs (36–38).
There is substantial variation in participation in research based on the type of baseline immunosuppressive medication for recipients. This is not surprising given that medications have been available at different stages of approval over the study period through the Federal Drug Administration and that industry sponsors have variable number of active research protocols. Differences in participation by baseline medications may be important to interpret outcomes from observational studies. For instance, studies that compare medications that typically include both participants and nonparticipants may affect results comparing regimens. Further validation of outcomes by medication may also be important to understand the potential external validity for each individual therapy in a research versus a nonresearch context.
There are several important limitations that may impact inferences from our study. First, the measure of participation may be imprecise. It is likely that documentation of participation in research is a sensitive metric (i.e., patients who were noted to participate actually did); however, it may be less specific and patients who participated may not have been noted. This may tend to dilute results with respect to comparisons of patient characteristics and outcomes. It is not as likely that there is a systematic bias in reporting for certain patient characteristics, but the overall proportion of participants may be underreported. Also, based on reporting at 6 and 12 months, we had to exclude patients who had graft loss within 1 year of transplantation. This may impact outcomes in either direction as early events may be more disproportionally distributed among participants and nonparticipants. Participation and outcomes are also likely affected by underlying characteristics of patients who are not available for risk adjustment. For example, patients with non-codified comorbidities may be less likely to be considered viable for participation and the degree of this effect is not known. Finally, there is a wide variety of types of “research studies” (e.g., investigator-initiated or phase I-IV trials) and no documentation available for this study for why patients did not participate in research or providers failed to enroll select patients. These questions require further study with more detailed information.
Each of the differences evaluated in this study (patient characteristics, survival outcomes, center variation, and immunosuppressive therapies) suggests some lack of external validity of research studies as they would be applied to the broader transplant population. Some of these differences may be relatively minor and do not necessarily invalidate the qualitative generalizability of studies. However, it is important to recognize that patient groups with lower participation in studies may not benefit from research findings to the same degree as those populations that more commonly participate (39). Prospectively, well-designed studies that have broad inclusion criteria may be important to this field to understand the external validity of interventions and to minimize the risk for study results that are not applicable in practice (10, 40). Further research is also needed to determine whether differences in participation are based on consent processes, justified or invalid inclusion, and exclusion criteria, and derived from provider or patient preferences or patients' access to care.
In summary, participation in research protocols among transplant recipients varies significantly by patient characteristics, transplant center, and individual medications. Long-term outcomes are superior among participants versus nonparticipants among kidney, liver and lung transplant recipients. Further understanding of the mechanism for differential participation, whether findings suggest disparities in access to care, and the implications for the interpretation and implementation of research findings into practice is needed.
MATERIALS AND METHODS
We used data from the Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donor, wait-listed candidates, and transplant recipients in the United States, submitted by the members of the Organ Procurement and Transplantation Network (OPTN). The Health Resources and Services Administration, US Department of Health and Human Services, provides oversight to the activities of the OPTN and SRTR contractors.
The study population consisted of recipients of a solitary kidney, liver, heart, or lung transplant and recipients of SPK transplants between the years 2000 and 2008. Patients were excluded if they were recipients of multiple organ transplants other than SPK. To classify patients as participants or nonparticipants in research protocols, we used information from OPTN recipient follow-up forms. Recipient forms include a field indicating whether patients have participated in research protocols for immunosuppressive medications during the applicable follow-up period. For the purpose of this study, we classified patients as participants if they were denoted as such at the 6-month or the 1-year follow-up period. To appropriately compare the likelihood of participation and outcomes between patients who did or did not participate in research, we excluded patients with less than 1-year follow-up or graft loss before 1-year or missing information in both follow-up periods.
Missing reports of participation ranged from 5.8% among SPK transplant recipients to 8.6% among lung transplant recipients. This proportion of missing levels increased over time likely based on the fact that the field for reporting participation changed from mandatory to optional after 2005. Therefore, as a part of sensitivity analyses to assess whether missing reports changed the qualitative results of the study, we repeated analyses restricted to the period of mandatory reporting.
Using census data and recipients' primary reported zip code, we added two additional patient characteristics to data collected in OPTN forms. We included estimated household income based on zip code level information and distance to the transplant center based on patients' permanent residence to the zip code of the center calculated by the Statistical Analysis Software (SAS) function zipcitydistance. We used multivariable logistic models to assess the likelihood of participation among recipients for each individual solid organ type. For model covariates, we included both demographic characteristics of donors and recipients consistent across organs along with patients' primary diagnoses for each form of end-stage organ disease and additional variables that were considered relevant markers of patient acuity. The Hosmer-Lemeshow test was used to test goodness of fit of these models. We generated Kaplan-Meier plots and multivariable Cox proportional hazard models for overall graft loss (defined as graft loss or death) and patient death, both conditioned at 1-year overall graft survival. Proportionality of the Cox models was assessed by visual inspection of the residual plots and testing the interaction with follow-up time for the primary explanatory variables. To ascertain whether the effect of participation was affected by participation rates at individual centers, we also used models adjusted for the proportion of participants within each center. Certain immunosuppressive medications were grouped: all formulations of cyclosporine (generic or brand name) were combined and IL-2 RB (simulect or zenapax) were placed in the same group. All analyses were conducted in SAS (version 9.2; SAS Institute Inc., Cary, NC).
This study was approved by the Cleveland Clinic Institutional Review Board.
1. Britton A, McKee M, Black N, et al. Threats to applicability of randomised trials: Exclusions and selective participation
. J Health Serv Res Policy
1999; 4: 112.
2. Lesko LM, Dermatis H, Penman D, et al. Patients', parents', and oncologists' perceptions of informed consent
for bone marrow transplantation. Med Pediatr Oncol
1989; 17: 181.
3. Stocking CB, Hougham GW, Danner DD, et al. Speaking of research advance directives: Planning for future research participation
2006; 66: 1361.
4. Wade J, Donovan JL, Lane JA, et al. It's not just what you say, it's also how you say it: Opening the ‘black box’ of informed consent
appointments in randomised controlled trials. Soc Sci Med
2009; 68: 2018.
5. Chou P, Kuo HS, Chen CH, et al. Characteristics of non-participants and reasons for non-participation
in a population survey in Kin-Hu, Kinmen. Eur J Epidemiol
1997; 13: 195.
6. Jaskiw GE, Blumer TE, Gutierrez-Esteinou R, et al. Comparison of inpatients with major mental illness who do and do not consent to low-risk research. Psychiatry Res
2003; 119: 183.
7. Hall JK. Exclusion of pregnant women from research protocols: Unethical and illegal. IRB
1995; 17: 1.
8. Kattan MW, Vickers AJ. Incorporating predictions of individual patient risk in clinical trials
. Urol Oncol
2004; 22: 348.
9. Zink S, Wertlieb S, Kimberly L. Informed consent
. Prog Transplant
2005; 15: 371.
10. Oude RK, Opmeer BC, Logtenberg SL, et al. IMproving PArticipation
of patients in Clinical Trials
—Rationale and design of IMPACT. BMC Med Res Methodol
2010; 10: 85.
11. Joseph G, Dohan D. Diversity of participants in clinical trials
in an academic medical center: The role of the ‘Good Study Patient?’ Cancer
2009; 115: 608.
12. Goode PS, Fitzgerald MP, Richter HE, et al. Enhancing participation
of older women in surgical trials. J Am Coll Surg
2008; 207: 303.
13. Joseph G, Dohan D. Recruiting minorities where they receive care: Institutional barriers to cancer clinical trials
recruitment in a safety-net hospital. Contemp Clin Trials
2009; 30: 552.
14. Karlawish J, Cary MS, Rubright J, et al. How redesigning AD clinical trials
might increase study partners' willingness to participate. Neurology
2008; 71: 1883.
15. Marcantonio ER, Aneja J, Jones RN, et al. Maximizing clinical research participation
in vulnerable older persons: Identification of barriers and motivators. J Am Geriatr Soc
2008; 56: 1522.
16. Brown RF, Butow PN, Boyle F, et al. Seeking informed consent
to cancer clinical trials
: Evaluating the efficacy of doctor communication skills training. Psychooncology
2007; 16: 507.
17. Payne JK, Hendrix CC. Clinical trial recruitment challenges with older adults with cancer. Appl Nurs Res
2010; 23: 233.
18. Stryker JE, Wray RJ, Emmons KM, et al. Understanding the decisions of cancer clinical trial participants to enter research studies: Factors associated with informed consent
, patient satisfaction, and decisional regret. Patient Educ Couns
2006; 63: 104.
19. Israni AK, Halpern SD, McFadden C, et al. Willingness of dialysis patients to participate in a randomized controlled trial of daily dialysis. Kidney Int
2004; 65: 990.
20. Mills E, Wilson K, Rachlis B, et al. Barriers to participation
in HIV drug trials: A systematic review. Lancet Infect Dis
2006; 6: 32.
21. Mills EJ, Jha P. Participation
in HIV vaccine trials: Listening to participant & community concerns. Indian J Med Res
2006; 124: 608.
22. Henderson GE, Churchill LR, Davis AM, et al. Clinical trials
and medical care: Defining the therapeutic misconception. PLoS Med
2007; 4: e324.
23. Lidz CW, Appelbaum PS, Grisso T, et al. Therapeutic misconception and the appreciation of risks in clinical trials
. Soc Sci Med
2004; 58: 1689.
24. Wendler D, Krohmal B, Emanuel EJ, et al. Why patients continue to participate in clinical research. Arch Intern Med
2008; 168: 1294.
25. Brennan TV, Fuller TF, Vincenti F, et al. Living donor kidney transplant recipients and clinical trials
profiles and impact on post-transplant care. Am J Transplant
2006; 6: 2429.
26. Creel AH, Losina E, Mandl LA, et al. An assessment of willingness to participate in a randomized trial of arthroscopic knee surgery in patients with osteoarthritis. Contemp Clin Trials
2005; 26: 169.
27. Bjork T, Clinton D, Norring C. Reasons for non-participation
in follow-up research on eating disorders. Eat Weight Disord
2006; 11: 147.
28. Alexander GC, Sehgal AR. Why hemodialysis patients fail to complete the transplantation process. Am J Kidney Dis
2001; 37: 321.
29. Young CJ, Gaston RS. Renal transplantation in black Americans. N Engl J Med
2000; 343: 1545.
30. Isaacs RB, Lobo PI, Nock SL, et al. Racial disparities in access to simultaneous pancreas-kidney transplantation in the United States. Am J Kidney Dis
2000; 36: 526.
31. Tonelli M, Klarenbach S, Rose C, et al. Access to kidney transplantation among remote- and rural-dwelling patients with kidney failure in the United States. JAMA
2009; 301: 1681.
32. Axelrod DA, Guidinger MK, Finlayson S, et al. Rates of solid-organ wait-listing, transplantation, and survival among residents of rural and urban areas. JAMA
2008; 299: 202.
33. Axelrod DA, Guidinger MK, McCullough KP, et al. Association of center volume with outcome after liver and kidney transplantation. Am J Transplant
2004; 4: 920.
34. Shuhaiber JH, Moore J, Dyke DB. The effect of transplant center volume on survival after heart transplantation: A multicenter study. J Thorac Cardiovasc Surg
2010; 139: 1064.
35. Schold JD, Harman JS, Chumbler NR, et al. The pivotal impact of center characteristics on survival of candidates listed for deceased donor kidney transplantation. Med Care
2009; 47: 146.
36. Schold JD, Srinivas TR, Howard RJ, et al. The association of candidate mortality rates with kidney transplant outcomes and center performance evaluations. Transplantation
2008; 85: 1.
37. Schold JD, Srinivas TR, Poggio ED, et al. Hidden selection bias deriving from donor organ characteristics does not affect performance evaluations of kidney transplant centers. Med Care
2010; 48: 907.
38. Weinhandl ED, Snyder JJ, Israni AK, et al. Effect of comorbidity adjustment on CMS criteria for kidney transplant center performance. Am J Transplant
2009; 9: 506.
39. Ioannidis JP. Some main problems eroding the credibility and relevance of randomized trials. Bull NYU Hosp Jt Dis
2008; 66: 135.
40. Ioannidis JP. Why most published research findings are false. PLoS Med
2005; 2: e124.
Keywords:© 2011 Lippincott Williams & Wilkins, Inc.
Clinical trials; Participation; Informed consent; Graft survival; Immunosuppression