Quantifying Participant Burden In Clinical Trials: Data From Prostate Cancer Rcts

doi:10.21203/rs.3.rs-1339592/v1

Download PDF

Research Article

Quantifying Participant Burden In Clinical Trials: Data From Prostate Cancer Rcts

https://doi.org/10.21203/rs.3.rs-1339592/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: The restrictions implemented due to the COVID pandemic have underscored the importance of clinical research and trial methodology, while also highlighting the potential need for reduced clinical visits or face to face interaction. This has reignited the discussion surrounding the impact of clinical research participation on trial participants. The primary aim of this study was to quantify participant burden in a sample of prostate cancer clinical trials. Secondary to this was establishing the effect of participant burden on recruitment and retention.

Methods: We identified a body of prostate cancer clinical trials in a five-year period from 2017-2021 and randomly selected 30 for inclusion. We categorised participant burden as physical, psychological, economic, familial and societal. We created a measurement tool called the Trial Burden Score and applied it to each included trial. We used the trial burden score to categorise each trial into high participant burden, medium participant burden and low participant burden. We compared the trial burden to recruitment and retention rates in each trial.

Results: 50% (n = 15) of the included trials were categorized as high participant burden, 47% (n = 14) as medium participant burden and one trial was categorized as low participant burden. The median participant burden score for each category was: physical burden 11 (range 8-16); psychological burden 14 range (11-18); economic burden 5 (range 2-6); familial burden 10 (range6-11); societal burden 0 (range 0-1). There was no association between stage of cancer diagnosis and the participant burden categories (low, medium, high). There was no correlation between trial burden score and under recruitment percentage and trial burden score and retention.

Conclusions: The burden placed on clinical trials participants is significant and requires consideration at the trial design stage. Serious thought should be given to only collecting data necessary to answer the research question while also ensuring the trial is applicable to the population it purports to serve. Limiting burden to improve participant trial experience is vital to ensure our participants have a positive experience, stay in the trial, pass on this positivity to others and are willing to participate in further trials.

Participant burden

trial methodology

randomised controlled trials

cancer trials

Participant burden is considered an integral concept in research ethics but has yet to be conceptualised(1). Common sense tells us that if you give a patient a choice of participating in a high participant burden trial or a low participant burden trial, they will choose the low participant burden. So how we define participant burden and keep it 'in check' when we are designing trials is crucial to trial success. The COVID-19 pandemic has emphasised the importance of timely, efficient, and ethical health research, with changes to trial conduct practices necessary to facilitate remote recruitment and monitoring of participants (2)(3)(4). This has had the positive impact of reducing participant burden. Cancer patients are a unique group as they often endure additional physical burdens associated with their diagnosis and related treatments as well as the standard burden of trial participation which can result in the exacerbation of physical or mental illness (5). Increases in the amount of travel to attend trial visits, the possibility of undergoing supplementary procedures and completing trial questionnaires are examples of additional burden that result in psychological, physical, and financial stress for patients that may also affect their willingness to participate in a trial(5)(6)(7)(8). The excess burden clinical research imposes on participants remains an obstacle for trial enrolment in certain patient populations (9).

Recruitment and retention are known to be problematic aspects of health research, with successful recruitment and retention of patients to clinical trials being identified as two of the most difficult aspects of the trial process (10)(11). Many trials fail to recruit within the envisioned timescale leading to an increase in cost, or fail to reach their required sample size giving the potential for underpowered studies(10)(11)(12)(13). Once recruited, retention is equally important, as participant drop-outs or incomplete data can cause problems in the analysis, interpretation, and external validity of the research (11)(14)(15). Participant burden is described by Lingler et al. as ‘the direct risks associated with research interventions or data collection procedures’ (16:46). There is a significant gap in the literature on the effect of participant burden on cancer patients and the potential impacts arduous methodological approaches have on recruitment and retention in cancel clinical trials (17). We know that the higher the level of burden participants are subjected to, the higher the associated thoughts of withdrawal from a trial (18). Additionally, a strong inverse relationship between protocol design complexity and patient recruitment and retention performance has been identified, with the perceived demands noted as a top contributing factor to the decision not to participate

(19)(20). Ulrich et al. (18) classified the level of perceived ‘burden’ for a participant in a clinical trials into five categories: physical, psychological, economic, familial, and social. The primary outcome of our study was to quantify, using the Ulrich et al. categories, the participant burden in a sample of prostate cancer clinical trials. Secondary outcomes were the effect of participant burden on recruitment in prostate cancer clinical trials and the effect of participant burden on retention in prostate cancer clinical trials.

The work had four stages:

1. Identify a body of trials in a clinical area (prostate cancer) that will provide the study sample

2. Define participant burden

3. Create a measurement tool to quantify participant burden and apply it to each trial - Trial Burden Score (TBS)

4. Establish the recruitment and retention rates and compare them to the TBS

Stage 1: Identify a body of trials

The search strategy was generated by applying the PICO framework and defining study outcomes (Table 1). We searched Embase, Medline, Scopus, Web of Science and Google Scholar for a five-year period 2017-2021. An example search strategy used for Medline is included in Supplementary File 1. We also searched https://clinicaltrials.gov/ and Australian New Zealand Clinical Trials Registry (ANZCTR) for further information on the selected trials. We set ourselves a target of 30 prostate cancer clinical trials to answer our research question. We wanted a sample that was large enough to say something meaningful, but not so large that it could not be completed in a 3-month window, the timeframe allocated to complete this work as an MSc dissertation for a taught MSc Clinical Trials. Thirty trials seemed a reasonable compromise between sample size and feasibility. We anticipated that the list of potentially eligible studies identified by our search would be greater than our target of 30, meaning we would need to make a selection. We did this by random selection from the list of all eligible studies. Two hundred and twelve eligible studies were filed to the reference manager Zotero and duplicates were removed. We randomly selected 30 for inclusion. Figure 1 displays the flowchart for the study selection.

Table 1

Study Eligibility Criteria
Eligibility Criteria
Study Design	Phase II or III, Randomised Controlled Trial (RCT) in the field of prostate cancer (PC) research.
Participants	Inclusion	Exclusion
Participants	1. Must include prostate cancer patients or treat prostate cancer side effects in case of prostate cancer survivor (PSC) >18 years of age. 2. Must be a randomised controlled trial* of a CTIMP or other therapy. 3. Must have access to free full text and associated data	1. Trials for screening risk/diagnosing PC 2. Feasibility/Pilot studies without published results 3. Abstracts from conference 4. Secondary Analysis/Outcome Analysis
Intervention	Any therapeutic treatment of disease/disease side effects from diagnosis of prostate cancer/neoplasms.
Comparators	Alternative therapeutic agent, Standard of care, Placebo or Sham treatment/device.
Primary Outcomes	Determine the effect of low participant burden compared to high participant burden on recruitment to cancer clinical trials
Secondary Outcomes	Establish the effect of low participant burden compared to high participant burden on retention to cancer clinical trials.
Timing	Only trials published between 2017-2021 will be eligible for inclusion.
Setting	There is no limitation to setting applied.
Language and Publication Status	English language studies only. Only published studies with published protocols will be included.

Stage 2: Define participant burden

We used the Ulrich et al.(18) categories to define participant burden: Physical, Psychological, Economic, Familial and Societal. Each trial was screened by two people: one (AD) extracted the data from the literature, the second (FS) reviewed the process. Any discrepancies were discussed and agreed. First we read each included trial, documented the burden in each trial and categorised them under the five Ulrich et al. categories. We then justified the assumptions made. The full table with all burdens, justifications and assumptions is presented in Supplementary File 2.

Stage 3: Create a measurement tool to quantify participant burden and apply it to each trial

There is no standard tool to measure participant burden (21) so we created a measurement tool and called it the Trial Burden Score (TBS). We created a burden scale, with three values, 1, 2, or 3 - the lower the value, the lower the burden. The participant burden tool and the values for each burden are shown in Supplementary File 3.

To calculate the Trial Burden score (TBS) for each trial, we used the data extraction tables in supplementary file 3, assigned a value between 1 and 3 to each sub-category, 1 being low burden, 2 being medium burden and 3 being high burden, and calculated the sum of the five burden categories, physical, psychological, economic, familial, and social, for each trial.

Finally, using the lowest scoring and highest scoring burden results as reference points, we created three burden categories: low burden, (TBS of <30), high burden, TBS of >40) and medium burden, TBS of 30-40.

Stage 4: Establishing the recruitment and retention scores

The recruitment percentage required for comparison was created by subtracting the actual recruitment from the sample size. A percentage difference was calculated which showed either over recruitment or under recruitment and is represented as an under-recruitment percentage in Supplementary File 4. This was compared against the TBS. The under-recruitment percentage was then plotted against the TBS to determine the effect of total burden on the surrogate recruitment rate, under-recruitment %.

The retention rate required for comparison was determined by subtracting the loss to follow up (LTFU) from the original recruitment value and dividing this result with the recruitment value to determine the retention rate in percent. LTFU in this instance included any withdrawal of consent to participate including participants that did not reach the first outcome analysis, drop-outs, and non-responders. Disease progression, clinical decision to withdraw and death were excluded from the LTFU analysis as correlation to burden would not be appropriate in these cases.

Data analysis

Data analysis was simple. To answer the question of participant burden in clinical trials, we used simple counts and proportions to calculate the number and percentage of trials in the high, medium or low participant categories. We used chi-squared analysis to investigate the association between cancer diagnosis, under recruitment and retention rate and total burden categories (low, medium, high). We conducted spearman correlation between RCT burden and under-recruitment and RCT burden and retention. The significance level applied was 95%.

Thirty prostate cancer clinical trials were analysed. The mean age was 67.9 (SD 4.2). Twenty-nine trials reported a stage of diagnosis. Of these, 62% (18) were Stage 1 or Stage 2, and 52% (15) were Stage 3 or Stage 4 (some trials included stage 2/3 or 3/4 so these totals do not add to 100%). The characteristics of the included studies are presented in Table 2, along with the participant burden (TBS for the 30 included trials. The median participant burden for each category was: physical burden 11 (range 8-16); psychological burden 14 range (11-18); economic burden 5 (range 2-6); familial burden 10 (range6-11); societal burden 0 (range 0-1). Categorising the TBS, 50% (n = 15) of the included trials were categorized as high participant burden (TBS of >40), 47% (n = 14) as medium participant burden and one trial was categorized as low burden. There was no association between stage of cancer diagnosis and the participant burden categories (low, medium, high).

Seventy-six percent (23/30) of trials met a recruitment target of at least 90%, but trial sample sizes of the included studies were relatively small (42% (14/30) recruited fewer than 100 participants) so there was little variation between each trial. Similarly, the retention rates of the trials remained high with 70% (21/30) of trials maintaining 90% retention. We conducted spearman correlations between TBS and under recruitment percentage and TBS and retention. There was no relationship evident. Neither was there a trend in under recruitment or retention rate when examined in the context of the TBS divided into tertiles.

The proportion of trials that are classed as high participant burden in this small sample of prostate cancer clinical trials should give the trial community pause for thought. That only one trial had a low participant burden is very significant. We know that trials collect a lot of data, and much of this goes unused. We don't know why trial teams collect so much data that is not subsequently used but anecdotally we know that investigators think that if they are going to the trouble of collecting data, they might as well get as much as possible – the information might get used some time. Prior studies have given some thought to the effort and cost trial teams go to collect this data (22) but it's now time to consider the effort of the participant, and the potential impact on their experience of participating in a trial with a high or medium participant burden.

The variation of participant age appeared to have a slight effect on the burden score with 58.8% (10/17) of trials in the over 70’s being identified as high burden. This coincided with a high perceived familial burden, which supports the idea of perceived negative effect of trial participation on caregivers of elderly participants. Physical burden had the strongest correlation with trial burden, the lowest physical burden score was recorded for the trial that had the lowest overall TBS score. On the corollary, all trials considered high burden had higher than the average (>10.5) physical burden scores.

Recruitment and retention have been widely documented as one of the most difficult aspects in clinical research, where slow recruitments and high loss to follow up rates often affect a trial’s financial and logistical feasibility and its external validity (10),(15),(23). In this review over 65% (20/30) of the trials met their recruitment targets or had less than 5% (acceptable variance) between the pre-determined sample size and actual recruitment, while 70% (21/30) of trials maintained 90% retention. While we did not statistically show any trend associated with recruitment or retention difficulties and high medium or low participant burden, most likely because recruitment and retention was in relative terms pretty high in these trials, our sample consisted of the people that consented to participate in the RCTs. We have no way of knowing if the burden of the trial dissuaded any other potential participants. This would need further investigation, and could be done effectively in a SWAT (Study Within A Trial).

Participants in trials with diseases such as cancer, which can be life-limiting and have limited treatment options, may have a higher tolerance for higher burden in trials, as their perception of risk of a trial could be outweighed by their motivation to receive potentially life-saving treatments and may account for the higher-than-average recruitment and retention rates. This was seen to be a relative trait at all stages of diagnosis, as there was no variance between diagnosis stage and recruitment or retention rates. As this was a retrospective study with limited knowledge of the inner workings of each trial other than the reported methodologies, the potential use of strategies to reduce/mitigate attrition rates such as undocumented conversations between caregivers/trialists with participants to encourage and support them during the trial process may also have been applied to some or all trials in this sample and these strategy benefits cannot be understated, as both frequency and quality of patient contact have previously been identified as a workable retention strategy(24),(25). It is possible that this was even more important in high participant burden trials.

The concept of participant burden has yet to be fully defined. The data categories identified for this review are not by any means definitive. If the methodology outlined in this study were to be applied to trials that are currently in progress and a survey to participants was an available option, information such as marital status, number of children, employment status and financial situations (availability of pensions or savings) could all have been introduced to the categories previously identified to give a more comprehensive view of the potential burdens placed on participants and their families during trial participation. Further research, using a mixed-methods approach, should be conducted to quantify the level of participation burden on cancer patients during clinical trials.

Due to the retrospective nature of this study, access to information was restricted to that reported in a sample of prostate cancer clinical trials in peer reviewed journals. This confined the amount of extractable information that could be utilised to create the burden sub-categories. For example, data was not available for distance to clinic visit for all studies and this has previously identified as a major barrier to initial participant recruitment (26),(27). Societal and economic burden both had the lowest burden scores, which on observation can be associated with the inconsistency of the number of sub-categories per data category. Due to restricted data availability, these burden categories could not be expanded or developed, which lead to an under-representation of the potential burden effect of each category within each trial.

We have highlighted an area of trial methodology that needs further investigation. Though our sample is small, we have at least pointed to a problem within trials of unnecessarily burdening trial participants. The fact that these participants are cancer patients, who endure treatments that make them severely ill, is of note. Our study would also benefit from investigation in RCTs of clinical specialties other than prostate cancer. While we have previously considered trial burden from a data perspective, we have not adequately considered the burden from the participant's perspective and how it may impact recruitment and retention. Reducing participant burden through more efficient data collection methods which are acceptable to participants should be considered. Limiting burden to improve participant trial experience is vital to ensure our participants have a positive experience, stay in the trial, pass on this positivity to others and are willing to participate in further trials.

Ethics approval and consent to participate

This was a retrospective study of existing publicly available research therefore ethical approval was not required.

Consent for publication

Not applicable

Availability of data and materials

The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

AD and KG have no competing interests. FS will take up the role of Associate Editor of Trials in April 2022

Funding

There was no funding received for this research.

Author’s contributions: AD undertook this research project in part-fulfilment of her MSc Clinical Trials. FS conceptualised the research project, supervised the research and commented on multiple drafts. KG contributed to the data analysis and write up of the manuscript. All authors read and approved the final manuscript.

Acknowledgements

None

Authors' Information

AD works in the biochemistry lab. of Cork University Hospital. She recently completed a taught MSc in Clinical Trials with University College Cork. KG is a final year student of BSc Public Health Sciences at University College Cork and a RA working with FS at the HRB Clinical Research Facility at UCC. FS is Director of Education and a Senior Lecturer in Patient Focused Research and Epidemiology at the HRB Clinical Research Facility and School of Public Health at University College Cork. She is Programme Director for the MSc Clinical Trials (online) and UCC PI for the HRB Trials Methodology Research Network.

Kabacińska K, Sharma N, Kaye J, Mattek N, Kuzeljevic B, Robillard JM. Investigating the concept of participant burden in aging technology research. BMC Geriatr. 2020 Feb 12;20(1):50.
Cagnazzo C, Besse M-G, Manfellotto D, Minghetti P, Cazzaniga S, Cottini L, et al. Lessons learned from COVID-19 for clinical research operations in Italy: what have we learned and what can we apply in the future? Tumori [Internet]. 2020 Dec 9 [cited 2021 Mar 28]; Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7726625/.
Lara Gongoria AB, Fontes Jardim DL, Bastos DA. Oncology Clinical Trials During the COVID-19 Pandemic. Oncol Williston Park N. 2020 Jul 15;34(7):265–9.
Goldsack JC, Izmailova ES, Menetski JP, Hoffmann SC, Groenen PMA, Wagner JA. Remote digital monitoring in clinical trials in the time of COVID-19. Nat Rev Drug Discov. 2020 Jun;19(6):378–9.
Naidoo N, Nguyen VT, Ravaud P, Young B, Amiel P, Schanté D, et al. The research burden of randomized controlled trial participation: a systematic thematic synthesis of qualitative evidence. BMC Med [Internet]. 2020 Jan 20 [cited 2021 Apr 5];18. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6970283/.
Nipp RD, Powell E, Chabner B, Moy B. Recognizing the Financial Burden of Cancer Patients in Clinical Trials. The Oncologist. 2015 Jun;20(6):572–5.
Kimmelman J, Resnik DB, Peppercorn J, Ratain MJ. Burdensome Research Procedures in Trials: Why Less Is More. JNCI J Natl Cancer Inst [Internet]. 2017 Feb 17 [cited 2021 Feb 28];109(4). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5756064/.
Townsley CA, Selby R, Siu LL. Systematic review of barriers to the recruitment of older patients with cancer onto clinical trials. J Clin Oncol Off J Am Soc Clin Oncol. 2005 May 1;23(13):3112–24.
Chino F, Zafar SY. Financial Toxicity and Equitable Access to Clinical Trials. Am Soc Clin Oncol Educ Book Am Soc Clin Oncol Annu Meet. 2019 Jan;39:11–8.
Kadam RA, Borde SU, Madas SA, Salvi SS, Limaye SS. Challenges in recruitment and retention of clinical trial subjects. Perspect Clin Res. 2016;7(3):137–43.
Page SJ, Persch AC. Recruitment, Retention, and Blinding in Clinical Trials. Am J Occup Ther. 2013;67(2):154–61.
McDonald AM, Knight RC, Campbell MK, Entwistle VA, Grant AM, Cook JA, et al. What influences recruitment to randomised controlled trials? A review of trials funded by two UK funding agencies. Trials. 2006 Dec;7(1):9.
Treweek S, Mitchell E, Pitkethly M, Cook J, Kjeldstrøm M, Johansen M, et al. Strategies to improve recruitment to randomised controlled trials. Cochrane Database Syst Rev [Internet]. 2010 [cited 2021 Mar 27];(4). Available from: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.MR000013.pub5/full.
National Research Council (US) Panel on Handling Missing Data in Clinical Trials. The Prevention and Treatment of Missing Data in Clinical Trials [Internet]. Washington (DC): National Academies Press (US); 2010 [cited 2021 Mar 28]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK209904/.
Gul RB, Ali PA. Clinical trials: the challenge of recruitment and retention of participants. J Clin Nurs. 2010;19(1–2):227–33.
Lingler JH, Schmidt K, Gentry A, Hu L, Terhorst L. Perceived Research Burden Assessment (PeRBA): Instrument Development and Psychometric Evaluation. J Empir Res Hum Res Ethics JERHRE. 2014 Oct;9(4):46–9.
Treweek S, Lockhart P, Pitkethly M, Cook JA, Kjeldstrøm M, Johansen M, et al. Methods to improve recruitment to randomised controlled trials: Cochrane systematic review and meta-analysis. BMJ Open. 2013;3(2):e002360.
Ulrich CM, Zhou Q, Ratcliffe SJ, Knafl K, Wallen GR, Richmond T, et al. Development and Preliminary Testing of the Perceived Benefit and Burden Scales for Cancer Clinical Trial Participation. J Empir Res Hum Res Ethics JERHRE. 2018 Jul;13(3):230–8.
Cameron D, Willoughby C, Messer D, Lux M, Aitken M, Getz K. Assessing Participation Burden in Clinical Trials: Introducing the Patient Friction Coefficient. Clin Ther. 2020 Aug 1;42(8):e150–9.
Anderson A, Borfitz D, Getz K. Global Public Attitudes About Clinical Research and Patient Experiences With Clinical Trials. JAMA Netw Open [Internet]. 2018 Oct 5 [cited 2021 Apr 24];1(6). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6324429/.
Lingler JH, Schmidt KL, Gentry AL, Hu L, Terhorst LA. A New Measure of Research Participant Burden: Brief Report. J Empir Res Hum Res Ethics JERHRE. 2014 Oct;9(4):46–9.
Crowley E, Treweek S, Banister K, Breeman S, Constable L, Cotton S, et al. Using systematic data categorisation to quantify the types of data collected in clinical trials: the DataCat project. Trials. 2020 Jun 16;21(1):535.
Todd A, Martin C, Morgan J, Herbert E, Bradburn M, Burton M, et al. Age specific recruitment and retention to a large multicentre observational breast cancer trial in older women: The Age Gap Trial. J Geriatr Oncol [Internet]. 2020 Oct 27 [cited 2021 May 23];0(0). Available from: https://www.geriatriconcology.net/article/S1879-4068(20)30487-2/abstract.
Chhatre S, Jefferson A, Cook R, Meeker CR, Kim JH, Hartz KM, et al. Patient-centered recruitment and retention for a randomized controlled study. Trials. 2018 Mar 27;19(1):205.
Kearney A, Daykin A, Shaw ARG, Lane AJ, Blazeby JM, Clarke M, et al. Identifying research priorities for effective retention strategies in clinical trials. Trials. 2017 Aug 31;18(1):406.
Bodart S, Byrom B, Crescioni M, Eremenco S, Flood E. ePRO Consortium. Perceived Burden of Completion of Patient-Reported Outcome Measures in Clinical Trials: Results of a Preliminary Study. Ther Innov Regul Sci. 2019 May 1;53(3):318–23.
Billi JE, Pai C-W, Spahlinger DA. The effect of distance to primary care physician on health care utilization and disease burden. Health Care Manage Rev. 2007 Mar;32(1):22–9.

Table 2 is available in the Supplemental Files section.

Download PDF

Version 1

posted

You are reading this latest preprint version

Quantifying Participant Burden In Clinical Trials: Data From Prostate Cancer Rcts

Status:

Version 1

Abstract

Figures

Background

Methods

Stage 1: Identify a body of trials

Stage 4: Establishing the recruitment and retention scores

Data analysis

Results

Discussion

Conclusions

Declarations

References

Table

Supplementary Files

Status:

Version 1