Diversifying Electronic Cohort Research at UCSF

A community-engaged contest to select and support a diverse "eCohort" at UCSF

The Talking About Prostate Cancer (TAP) Study

Proposal Status: 

Research questions

Approximately 1 in 3 of prostate cancer diagnoses are classified as “low risk” and “very low risk” for disease progression and mortality. Active surveillance, that is, closely monitoring the course of disease, is a recommended management option for all men with very low risk and most men with low risk cancer. In the U.S., more than 100,000 men a year are estimated to be candidates for active surveillance, but only 40% of eligible men receive active surveillance. Despite the fact that prostate cancer is the most commonly diagnosed cancer in men among all major racial/ethnic groups, there is very little known about the determinants of active surveillance and outcomes across racial/ethnic groups.

The research questions for this study are:

1)    What are provider experiences with and perceptions of racial/ethnic differences in patients’ preferences regarding active surveillance?

2)    What are patient perspectives and ethnic/cultural/psychosocial beliefs regarding active surveillance?

3)    What is the influence of patient and contextual (family, social networks, institutional/provider, neighborhood) factors on use of active surveillance?

4)    What are patient-reported outcomes (PROs), including quality of life and physical, physiological, and psychosocial well-being, at various time points after prostate cancer diagnosis?

To address these questions, the Talking About Prostate Cancer (TAP) study will establish a diverse population-based cohort of low risk prostate cancer patients that can be followed for future research inquiries, including examination of disease progression.This study, conducted within a range of medical care settings, will enable us to understand the factors that drive differences in active surveillance referral and utilization. In identifying vulnerable sub-populations, study results will inform culturally-targeted clinic-level policies and practices regarding patient communication and counseling, and ultimately ensuring that all men, regardless of race/ethnicity or socioeconomic status (SES), have the opportunity to be fully informed about their treatment options and experience optimal wellness as a survivor.

An online engagement and data collection platform would allow this study to recruit and retain the diverse participants that are needed to answer the above research questions. Making data collection more efficient, streamlined, and culturally relevant will promote engagement from participants who are typically underrepresented in research projects. Finally, an online platform would facilitate the development of a diverse population-based cohort for future research studies.

The impact of the project on minority communities will be to: understand the unique factors that affect decision-making about treating prostate cancer; inform culturally-targeted clinic-level policies and practices directed towards improving patient communication and counseling; and ultimately, to support all men diagnosed with prostate cancer to make decisions about their treatment that will reflect the same positive outcomes across all racial/ethnic groups.

Study sample

This study will include a survey component among both prostate cancer patients and physicians in the Bay Area. We propose to use the Eureka platform only for the patient component of this study, thus, only the methods pertaining to patient participants are described here. Eligible patients will be identified through the Greater Bay Area Cancer Registry (GBACR), a part of the statewide California Cancer Registry and the NCI Surveillance, Epidemiology, End Results (SEER) Program since 1973. Using an early case ascertainment method to identify cases within 6-9 months after diagnosis, eligible patients are those newly-diagnosed with very low risk or low risk first primary invasive prostate cancer, ages 40-79 years and residing in one of the nine Greater Bay Area counties at diagnosis. This age range was selected because it represents 95% of all cases, and treatment decisions are less confounded by comorbidities and life expectancies. We will recruit all non-White ethnic groups and a 20% sample of NH Whites. Of approximately 4,700 cases of invasive prostate cancer reported each year in the catchment region, more than 40% (N approximately 1,900) have very low or low risk disease. These patients include both UCSF patients and non-UCSF patients. We plan to recruit patients who speak English, Spanish, Chinese (Mandarin and Cantonese), and Tagalog.


The main survey constructs are listed in the Table. Tumor characteristics will be ascertained from the GBACR. Most domains will be based on established items, adapted accordingly based on a qualitative phase of the study, which will inform the survey development.



Table. TAP Study patient survey constructs

Independent variables

Outcomes – dependent variables

  • Tumor characteristics
  • Individual risk factors
  • Demographic information
  • Social relationships
  • Social context
  • Institutional context
  • Decisional   & psychological factors
  • Treatment - active surveillance, definitive treatment   (radiation, radical prostatectomy)
  • Patient-reported outcomes
  • Prostate-specific quality of life and symptoms
  • Disease progression
  • Recurrence


We will conduct a baseline survey within 2-3 months of diagnosis when the men are making their decision regarding their initial treatment. Follow-up surveys will be administered at 12- and 24-months after the baseline survey (approximately 15- and 27-months after diagnosis), selected to capture relevant clinical events in the trajectory of treatment and surveillance, such as  PSA measurement at 6 months and biopsy at 12 months following diagnosis, and to capture changes in PROs.

We currently have a study website available in 3 languages (English, Spanish, Chinese) for patient recruitment (tapstudy.ucsf.edu). The Eureka platform would be an ideal solution for tailoring our website with culturally-relevant content that would encourage participation, easily incorporate electronic consent for this minimal-risk protocol, and provide a stable self-administered survey platform in 4 languages (English, Chinese, Spanish, Tagalog). The Eureka platform would also enable us to offer links to suveys via computers or apps and could be programmed to notify participants when surveys are ready and notify study staff of important data collection issues, completion of surveys, and non-response flags. Simple, tested, stable platforms for this type of research are critical to engaging diverse populations and retaining their participation in the future.


This research will be led by a multi-disciplinary team of collaborators with a productive history of research in prostate cancer, especially focused on disparities. The Principal Investigator, Dr. Scarlett Lin Gomez, Professor in the Department of Epidemiology and Biostatistics (DEB) and Member of the HDFCCC, is a cancer epidemiologist at UCSF with experience in disparities research focusing on multi-dimensional social and cultural determinants of health. Dr. Iona Cheng, Associate Professor in DEB and Member of the HDFCCC, is a molecular epidemiologist whose research focuses on understanding the role of genetic, molecular, lifestyle, and neighborhood factors in relation to cancer risk and outcomes, with much of her research focused on prostate cancer. Dr. Shariff-Marco, Associate Professor in DEB and Member of the HDFCCC, is trained in social and behavioral sciences and conducts epidemiologic and mixed methods research aimed at addressing the role of social determinants of health, including neighborhood determinants, as it relates to cancer disparities in racial/ethnic groups. Dr. Janet Shim is Professor in the Department of Social and Behavioral Sciences at UCSF, and brings expertise in sociology, psychology, qualitative and mixed methods research. Dr. Daphne Lichtensztajn is an epidemiologist with M.D. training, and is well-versed in cancer registry data, particularly clinical and diagnostic data on prostate cancer.  Additional study staff at UCSF include Mindy DeRouen, Laura Allen, Debora Oh, and Mei Chin Kuo.

Co-Investigators include urologic oncologists, Drs. James Brooks, Benjamin Chung, and John Leppert, from Stanford School of Medicine, and Drs. Peter Carroll and Matthew Cooperberg from UCSF; Dr. Robert Haile from Cedars-Sinai Medical Center; Weiva Sieh from Mt. Sinai; Qian Lu from MD Anderson; and Dominick Frosch from Palo Alto Medical Foundation Research Institute.

We have assembled a racially/ethnically-diverse 4-member patient advisory group to help with input on our survey measures, recruitment plans and website design. We will work with HDFCCC to engage with community partners to inform communities about the research, encourage participation, and distribute findings that could impact men’s experience of prostate cancer in the future (e.g., information sessions tailored to issues learned from the results of this study). 


This study is funded through the National Cancer Institute (R01CA211141) from August 2017-July 2022. The direct cost budget per year is about $500,000 (minus a 17% budget reduction applied by the NCI). The timing for the e-cohort initiative fits in well with our study as we are beginning the qualitative phase of the study and will begin the patient cohort recruitment in early 2019. Per our timeline, we will begin developing the survey for the patient cohort in upcoming months.


Thank you for taking the time to add your proposal. Our selection committee has UCSF and non-UCSF community-based members and non-medical staff. While your proposal is well written, a short, accessible summary will make it easier for everyone to engage. Could you share a 150-word plain-language summary of your ideas?

Active surveillance is an accepted management option for all men with very low risk and most men with low risk prostate cancer, particularly as definitive treatments including radiotherapy and radical prostatectomy are associated with significant treatment-associated problems including bowel, urinary, and sexual dysfunction. However, about less than half of eligible men receive active surveillance, with variations by race/ethnicity and socioeconomic status. Conducted within a population-based setting with an inclusive range of medical care settings, the Talking About Prostate (TAP) cancer study will: 1) Examine active surveillance perspectives of clinicians who treat diverse prostate cancer patients; 2) Examine treatment decision-making processes by race/ethnicity and socioeconomic status; and 3) Compare patient-reported outcomes (such as physical and mental quality of life) by race/ethnicity and socioeconomic status, and contributions of factors such as family and social networks. Results will enable understanding of contributing factors at multiple levels that drive the gaps in active surveillance.

Hello - I am working with Dr. Gomez on this proposal. We added the below in the "Summary" section of the form, but I am not seeing it here. So I am adding in this comment as well. Please let me know if we should submit otherwise. Thanks!


The Talking About Prostate (TAP) cancer study will investigate:

1) Clinician perspectives on racial/ethnic differences in patients’ preferences about active surveillance.

2) Patient perspectives and ethnic/cultural/psychosocial beliefs about active surveillance.

3) The influence of patient, family, social networks, etc., on use of active surveillance.

4) Outcomes such as quality of life and well-being after prostate cancer diagnosis. 

This study will enable us to better understand differences in active surveillance usage. Study results will inform clinic-level interventions, and ultimately help all men be well-informed about their treatment options.

Thank you Laura. The Summary section shows on the top-level opportunity page where all the proposals are listed. It's perfectly fine to have it here too though. 

Exciting and important project!  One of things I'm a little unclear on is whether you really need Eureka for this.  It sounds like you will be doing some in-person work anyway, and already have an online presence?  What will Eureka actually add to your project?  How would you do the study if you don't get the Diverse eCohorts support and don't otherwise engage with Eureka?

Hello - I am working with Dr. Gomez on this study and am responding as she is on vacation this week. 

While we have a study website, we do not have a platform for data collection. We think Eureka would facilitate recruitment and retention of participants improving response rates over the longitudinal aspect of the study (beyond baseline, we are interested in collecting data at 12 and 24 months).  Eureka would also facilitate the development of a diverse population-based cohort for future research studies.

If we do not get support from Eureka, we are likely to design an online survey using the REDCap system.

Will you be deliberately recruiting any UCSF patients (e.g., via clinic or EHR-based querying), or just from the community/GBACR?

Hello - I am working with Dr. Gomez on this study and am responding as she is on vacation this week. 

While these participants may include UCSF patients, we have proposed to use case listings from GBACR as the primary recruitment method.

thank you for your submission. a few questions,

1) aside from providing comments here, how else do you see the community advisors helping you with your study?

2) how do you intend to promote the study among the latin@ community, especially among the diverse latinx male population? Immigrants - monolingual spanish / english only men?  uninsured and immigrant males are at sometimes a difficult population to reach.

3) what N (number) are you looking for in your study and do you have priority target populations? this can help customize 

4) do you have plans to share research findings with the community stakeholders doing health work in the communities where patients are being recruited?

5) what considerations are you taking for understanding the regional and/or cultural differences within the populations - linguistic preferences for questions, - are questions culturally informed? have they been pilot tested with target group you hope to recruit?

Note on translation -- standard protocol should be used - translation - back translation process =  too often, translations are not vetted appropriately - staff are asked to translate but requires a thorough process


Thank you for these questions. Below please find response to each question.

1) In addition to input from our Patient Advisors (one representing each racial/ethnic group), we envision engaging community advisors to help with ensuring that our outreach materials and approaches will be culturally appropriate and acceptable.

2) Potentially eligible patients will be identified through the Greater Bay Area Cancer Registry, thus, we will not need to promote the study to a general population. We will ensure that our recruitment materials are culturally and linguistically appropriate and acceptable so as to maximize response rates. All materials and scripts have been translated and independently back-translated into Spanish, Chinese, and Tagalog, so we will be able to target monolingual men in these languages. We will be able to compare some demographic and clinical characteristics of respondents to non-respondents using cancer registry data. If we find that our response rates are biased in not adequately representing certain underserved populations, we will seek guidance from our patient and community advisors about alternative approaches to optimize response rates in these underserved groups.

3) We aim to recruit a total of 931 men (anticipating a response rate of 60% from all eligible patients (N=1552) from the Greater Bay Area Cancer Registry), including 160 African Americans, 265 Latinos, 234 Asian Americans and Pacific Islanders, and 272 non-Hispanic Whites. We will be contacting all eligible non-White men, and a 10% sample of non-Hispanic White men. Our priority target populations are non-White populations.

4) We plan to share our study findings with the urology community by publishing our results in peer-reviewed scientific journals, and presenting findings at scientific conferences. In addition, we will develop a newsletter summarizing our results and distributing this, in addition to the scientific articles, to all urologists in the Greater Bay Area.

5) We are currently in the process of developing our baseline survey. To the extent possible, we will use established and validated tools (particularly those that have been validated in non-English languages), such as EPIC and PROMISE for assessing patient-reported outcomes, and from the PhenX toolkit (https://www.phenxtoolkit.org/). Most survey items used for assessing sociodemographic domains (race/ethnicity, socioeconomic status, immigration, acculturation, language, sexual orientation and gender identity, etc) will be adapted from our prior study, STARS, in which these items were cognitively tested in 6 languages (https://cancerregistry.ucsf.edu/stars-study). Further, we will work with our patient and community advisors to ensure that the survey questions are culturally and linguistically appropriate and acceptable to the target populations.

Re: translations, we use a strict forward- and back-translation process for all of our patient materials. The back-translations are conducted by different individuals from those conducting the forward translations. For Tagalog, given the conversational nature of the spoken and written language among Pilipino Americans, we have taken the additional step of consulting with a native Pilipino American to review and adjust the translated version of materials.

Thank you,

Scarlett Lin Gomez

What are the plans for outcome dissemination to the respective communities?

Are there plans to translate the findings into interventions?

Good to see that the study website is translated in Chinese, but some of the content is somewhat difficult to understand to native Chinese-speakers due to literal translation.  Are there any plan to polish the translation?  By the way, has the Chinese verison of the website been focus group tested? 

What are the plans for promoting the study website among immigrant (Chinese, Filipino) communities?  



Dear Angela,

Thank you for your question. Please see below for our responses.

1) Outcome dissemination:

In addition to dissemination of study findings to scientific and medical audiences (e.g., urologists), we will disseminate results to lay communities via presentations at community forums. As with our prior studies, we are open and willing to share our results as opportunities arise.

2) Translation into interventions:

We absolutely inted for the finding from our study to translate into meaningful and effective improvements in the appropriate uptake of active surveillance, or at least to ensure that all men, regardless of race/ethnicity, socioeconomic status, language, etc, have the opportunity to be adequately informed about their options for active surveillance. The identification of influential multilevel factors including patient-provider communications, institutional characteristics, and family and social networks will inform the development of intervention strategies to ensure that all subpopulations among the more than 2 million survivors have the opportunity to be fully informed about their treatment options and to experience optimal wellness as a survivor. For example, study results may inform culturally-targeted clinic-level policies and practices directed towards improving patient communication and counseling.

3) Website translations:

The website was translated into Chinese by a study staff member who is a native speaker and has done several front and back Chinese-language translations for us on previous studies. The translated content was then back-translated by a different native speaker staff. The method used was somewhat literal as our understanding was that translations should be as similar to English content as possible, for IRB considerations. That said, our goal with the website is to be as accessible to monolingual Chinese-speaking participants as possible. To that end, we are currently working on a more conversational version. While the Chinese-language website has not been focus group tested, we will ask participants in our one-on-one qualitative interviews to comment on its content.

4) Promoting the study website:

Potentially eligible patients will be identified through the Greater Bay Area Cancer Registry, thus, we will not need to promote the study to a general population.

Thanks for this proposal.  In addition to disseminating resuts as opportunties arise, if community partners are interested in partnering to disseminate findings, for example, by developing education materials that could be shared with and by lay health workers in impacted communities, would your study team have the interest or capacity to do this without more support than this grant could provide?

Dear Roberto,

Thank you for this idea. Yes, we absolutely would be interested in partnering with community partners to disseminate findings.


Thanks for submitting this great proposal. My questions are:

1) Diverse eCohorts does not have Tagalog language capabilities. Will you have the support to create the Eureka platform for that language?

2) Management of low-risk prostate cancer patients varies, by individual patient preference and by physician practice patterns. In addition, since minorities may get care from facilities that may not have the highest quality of care.  Other than collecting data, is there a plan to educate patients or monitor their treatment and give feedback when the situation is sub-optimal?

Dear Dr. Nguyen,

Thank you for these great questions.

1. Yes, we would have the Tagalog language capabilities on our team to provide the translated versions for the portal. We have been using a language translation service in addition to Tagalog-speaking staff for the Tagalog translations of our study materials. We are in the process of recruiting a Tagalog-speaking interviewer who would serve as an additional translator for study materials.

2. This is an excellent question! Our study is currently funded as an observational study, and thus, there are currently no plans within this funded project to incorporate an intervention component. If we encounter patients who are not receiving standard of care and/or indicate to us as a part of the study that they are in need of support (logistical, psychosocial, treatment-related, etc), we will refer them to a list of local resources, and if needed, consult with the IRB office regarding whether additional interventions may be needed.

Moreover, we do intend to apply our study findings toward efforts to inform both providers and future patients about active surveillance as a treatment option for low-risk prostate cancer. In addition to the efforts targeted towards providers as described previously in a response to a prior reviewer, we will work with community organizations to disseminate our results to the lay communities. If the results seem to lend themselves to an intervention needing to be tested, we would seek to partner with a collaborator to develop a study to test this intervention.


The prospect of results influencing policy is very interesting.  Does the research team envision those policies as strictly clinical?  What expertise will the team leverage, either in-house or via partnership, toward shaping and disseminating policy proposals?

Dear Roberto,

This is an interesting question! While we do currently anticipate that the most effective channels to apply our results towards improving uptake of active surveillance is likely through clinical practice, especially among underserved patients, I think you raise an interesting idea regarding translations via policy proposals. I believe there are likely policy expertise in-house or via partnership, with whom we could partner to target these disparities through proposing policy changes. One potential external partner is the American Cancer Society, as well as the American Society for Clinical Oncology (ASCO). As a member of the ASCO Health Disparities Committee, this would be an ideal venue for proposing and drafting policies and guidelines.


I deeply appreciate the research questions, and see them as driving potential change if as my colleagues spoke about the results (of which I am making certain assumptions) can impact multiple systemic changes as well as advance attitudes towards intervetion, prevention and treatment options/active surveillance.

My concerns:

1) With the on-line surveys as typically your participants age range will be 55+ and the comfort with online platforms. How do you imagine getting around this?

2) Given this is such a deeply personal and emotionally laden subject matter - how do you anticipate getting a rich data sample?



Dear Monique,

Thank you for these comments. Our thoughts are as follows.

1) Based on recent population surveys, it is certainly he case that internet usage is more prevalent with younger age. However, surveys such as those from the PEW show that a substantial and increasing proportion of older populations do access the internet and related technologies. In our own prior research with Asian Americans, we found that among those older than age  50, more than two-thirds reported accessing the internet and being receptive to participating in health research online. We also found in two other pilot studies that older adults were receptive to and successfully participated in health surveys using ecological momentary assessment, receiving prompts via text or email to a series of short online surveys. We have conducted several large scale studies involving the use of internet surveys and have found that older populations do seem receptive to this modality. That said, we will also offer other modalities for completing the surveys, including by phone or in person with an interviewer, and via paper/pencil.

2) Potential eligible patients will be identified through our region’s population-based Cancer Registry aN’s we plan to oversample racial/ethnic minority groups, to ensure adequate representation of diverse populations. Our staff who will engage with the participants are from the same target racial/ethnic groups and can speak Spanish, Cantonese, Mandarin, or Tagalog. In developing our survey items, we will start with established measures that have been used in prior research, but will also develop new measures or adapt existing measures based on findIngs from qualitative research with patients and providers. With these approaches, we hope to be able to collect a rich source of data from a diverse population that can be used to meaningfully inform clinical guidelines or other means to reduce disparities in active surveillance use.

Thank you,


Commenting is closed.