Recruiting an underserved, difficult to reach population into a cancer trial: Strategies from the Restore-2 Rehabilitation Trial for gay and bisexual prostate cancer patients

BACKGROUND/AIMS

Sexual minorities are small and under-researched populations that are at disproportionate risk for cancer and poor cancer outcomes. Described as a "hidden population," the principal research challenge has been to develop effective methods to identify and recruit such cancer patients into cancer studies. Online recruitment strategies, as well as targeted clinic recruitment using patient-entered sexual orientation and gender identity data from electronic medical records have potential to transform recruitment, but studies testing the effects of how to recruit using these have not been published.

METHODS

In 2019, we conducted a naturalistic, three-arm, stratified prospective study to compare three recruitment strategies: (a) clinic based recruitment of prostate cancer patients from gay health and urology clinics; (b) directly from the gay community; and (c) online recruitment (through cancer support, sex/dating, and social sites). For each strategy, we estimated time, workload, and direct costs involved. To study how recruitment strategy may affect sampling, we tested for retention rates, demographic and outcome differences across sites. Using these methods, we successfully recruited 401 gay and bisexual prostate cancer patients into a randomized, controlled, 24-month trial testing an online sexual and urinary rehabilitation curriculum tailored for this population.

RESULTS

There were seven key results. First, it is possible to recruit substantial numbers of sexual minority men into prostate cancer studies provided online recruitment methods are used. Second, we observed big differences in dropout during study onboarding by recruitment source. Third, within online recruitment, the online sex/dating application (app) was the most successful and efficient, followed by the cancer support site, and then the social networking site. Fourth, while clinics were the cheapest source of recruitment, they were time intensive and low in yield. Fifth, the cancer support site and sex/dating app recruits differed by several characteristics, with the former being more rehabilitation-focused while the latter were younger and more sexually active. Sixth, we found almost no differences in outcomes across the three online recruitment sites. Seventh, because retention in online studies has been a concern, we confirm very low attrition at 3- and 6 months into the trial.

CONCLUSION

For sexual minority cancer research, more research on how to use sexual orientation and gender identity electronic medical record data for clinic-based recruitment is needed. For other small or hard-to-reach populations, researchers should compare and publish online versus offline recruitment strategies.

Creation and Psychometric Validation of the Sexual Minorities and Prostate Cancer Scale (SMACS) in Sexual Minority Patients-The Restore-2 Study

BACKGROUND

Existing measures of sexual functioning in prostate cancer survivors focus primarily on erectile function and do not adequately measure the experiences of sexual minority men.

AIM

To develop and psychometrically evaluate a new scale to measure sexual functioning among sexual minority men with prostate cancer.

METHODS

Sexual minority prostate cancer patients (n = 401) completed an online battery of urinary and sexual functioning tests in 2019, including a new 37-item instrument about their sexual functioning post-treatment for prostate cancer.

OUTCOMES

We used confirmatory factor analysis to determine the construct validity of a new scale including five subscales: a four-factor model for all participants (n = 401) evaluated Sexual Satisfaction, Sexual Confidence, Frequency of Sexual Problems, and Urinary Incontinence in Sex. A single-factor model completed only by participants who had attempted or desired receptive anal sex (n = 255) was evaluated in the fifth subscale: Problematic Receptive Anal Sex. To evaluate criterion validity, we calculated the intercorrelations between each Sexual Minorities and Prostate Cancer Scale (SMACS) subscale and four related scales: the Expanded Prostate Cancer Index Composite-50 (EPIC), the Functional Assessment of Cancer Therapy-Prostate, the Brief Symptom Inventory-18, and the International Consultation on incontinence questionnaire. Cronbach's alphas were calculated to measure internal consistency (ie, reliability).

RESULTS

Cronbach's alpha values ranged from 0.64 to 0.89. Loadings (0.479-0.926) and model fit indices were strong (Root Mean Square Error of Approximation: 0.085, Standardized root mean squared residual: 0.063, comparative fit index: 0.927, Tucker-Lewis Index: 0.907). For criterion validity, Sexual Satisfaction, Sexual Confidence, and Frequency of Sexual Problems were moderately correlated with EPIC function and bother scores (r = 0.50-0.72) and Urinary incontinence in sex correlated moderately with EPIC Urinary Function and International Consultation on incontinence questionnaire scores (0.45-0.56).

CLINICAL IMPLICATIONS

The SMACS can be used by clinicians and researchers to comprehensively measure sexual functioning in sexual minority men, in conjunction with existing scales.

STRENGTHS AND LIMITATIONS

This new scale is validated in a large, geographically diverse cohort of sexual minority cancer survivors and fills an important gap in existing measures of sexual functioning. Limitations include a lack of a validation sample.

CONCLUSION

The SMACS is a valid and reliable new scale that measures sexual minority men's experience of urinary incontinence in sex, problematic receptive anal sex, and sexual distress. Polter EJ, Kohli N, Rosser BRS, et al. Creation and Psychometric Validation of the Sexual Minorities and Prostate Cancer Scale (SMACS) in Sexual Minority Patients-The Restore-2 Study. J Sex Med 2022;19:529-540.

Detecting, preventing, and responding to "fraudsters" in internet research: ethics and tradeoffs

Internet-based health research is increasing, and often offers financial incentives but fraudulent behavior by participants can result. Specifically, eligible or ineligible individuals may enter the study multiple times and receive undeserved financial compensation. We review past experiences and approaches to this problem and propose several new strategies. Researchers can detect and prevent Internet research fraud in four broad ways: (1) through the questionnaire/instrument (e.g., including certain questions in survey; and software for administering survey); (2) through participants' non-questionnaire data and seeking external validation (e.g., checking data for same email addresses, usernames, passwords, and/or fake addresses or phone numbers; (3) through computer information, (e.g., IP addresses and cookies), and 4) through study design (e.g., avoid lump sum compensation and interviewing participants). These approaches each have pros and cons, and raise ethical, legal, and logistical questions, given that ethical tensions can emerge between preserving the integrity of research vs. protecting the privacy and confidentiality of study respondents. While past discussions concerning the ethics of online research have tended to focus on the participants' ability to trust the researchers, needs now arise to examine researchers' abilities to trust the participants. This analysis has several critical implications for future practice, policy, and research.

Detecting, preventing, and responding to "fraudsters" in internet research: ethics and tradeoffs

Internet-based health research is increasing, and often offers financial incentives but fraudulent behavior by participants can result. Specifically, eligible or ineligible individuals may enter the study multiple times and receive undeserved financial compensation. We review past experiences and approaches to this problem and propose several new strategies. Researchers can detect and prevent Internet research fraud in four broad ways: (1) through the questionnaire/instrument (e.g., including certain questions in survey; and software for administering survey); (2) through participants' non-questionnaire data and seeking external validation (e.g., checking data for same email addresses, usernames, passwords, and/or fake addresses or phone numbers; (3) through computer information, (e.g., IP addresses and cookies), and 4) through study design (e.g., avoid lump sum compensation and interviewing participants). These approaches each have pros and cons, and raise ethical, legal, and logistical questions, given that ethical tensions can emerge between preserving the integrity of research vs. protecting the privacy and confidentiality of study respondents. While past discussions concerning the ethics of online research have tended to focus on the participants' ability to trust the researchers, needs now arise to examine researchers' abilities to trust the participants. This analysis has several critical implications for future practice, policy, and research.