BackgroundThe use of the electronic medical record (EMR) system in recruitment in clinical trials has the potential for providing a very reliable and cost-effective recruiting methodology which may improve participant recruitment in clinical trials. We examined a recruitment approach centered on the use of the EMR, as well as other traditional methods, in the Lifestyle Intervention for Treatment of Diabetes (LIFT Diabetes) trial.MethodsLIFT Diabetes is a randomized controlled trial designed to investigate the effects of two contrasting interventions on cardiovascular disease risk: a community-based intensive lifestyle program aimed at achieving weight loss and a clinic-based enhanced diabetes self-management program. Eligible participants were overweight/obese (body mass index, BMI ≥25 kg/m2) patients with type 2 diabetes who were aged 21 years or older. Recruitment strategies included the use of the EMR system (primary), direct referrals, media advertisements, and community screenings.ResultsA total of 1102 telephone screens were conducted, resulting in randomization of 260 participants (61.5 % from EMR, mean age 56.3 years, 66.2 % women, 48.1 % non-Hispanic blacks) over a 21-month period, with a yield of 23.6 %. Recruitment yields differed by recruitment method, with referrals having the highest yield (27.5 %). A history of cardiovascular disease was the main health reason for exclusion from the study (16.5 %). An additional 8.9 % were excluded for BMI <25 kg/m2 (<27 kg/m2 for insulin users), 5.4 % could not exercise, 5.2 % had an HbA1c >11 %, and 34.9 % were excluded for other non-medical reasons. Exclusion criteria did not appear to differentially affect enrollment in terms of race or ethnicity.ConclusionsFuture clinical studies should tailor their recruitment strategies based on the participant demographics of interest. Efficient methods such as using the EMR system and referrals should be prioritized over labor-intensive, low-yielding methods such as community screenings and mass mailings.Trial registrationClinicalTrials.gov: NCT01806727. Registered on 5 March 2013.Electronic supplementary materialThe online version of this article (doi:10.1186/s13063-016-1631-7) contains supplementary material, which is available to authorized users.

Introduction To support claims that RCTs can reduce health disparities (i.e., are translational), it is imperative that methodologies exist to evaluate the tenability of external validity in RCTs when probabilistic sampling of participants is not employed. Typically, attempts at establishing post hoc external validity are limited to a few comparisons across convenience variables, which must be available in both sample and population. A Type 2 diabetes RCT was used as an example of a method that uses GIS to assess external validity in the absence of a priori probabilistic community-wide diabetes risk sampling strategy. Methods GIS and 2009–2013 county death certificate records and 2013–2014 electronic medical records were used to identify community-wide diabetes prevalence. Color-coded diabetes density maps provided visual representation of these densities. Chi-square goodness of fit statistic/analysis tested the degree to which distribution of RCT participants varied across density classes compared to what would be expected given simple random sampling of the county population. Analyses were conducted in 2016. Results Diabetes prevalence areas as represented by death certificate and electronic medical records were distributed similarly. The simple random sample model was not a good fit for death certificate record (chi-square, 17.63; p=0.0001) and electronic medical record data (chi-square, 28.92; p<0.0001). Generally, RCT participants were oversampled in high–diabetes density areas. Conclusions Location is a highly reliable “principal variable” associated with health disparities. It serves as a directly measurable proxy for high-risk underserved communities, thus offering an effective and practical approach for examining external validity of RCTs.

Objective: Professionally delivered intensive lifestyle weight loss interventions (LWLs) have been shown to improve weight status and CVD risk factors among adults with diabetes in rigorously conducted trials. We report the 12-month results of a trial in which a LWL was delivered by community health workers (CHWs), among predominantly minority and/or lower socioeconomic status participants. Research Design and Methods: We recruited overweight or obese adults with diabetes and without CVD primarily via review of electronic medical records and physician referrals. The study featured two 12-month interventions: (1) LWL delivered by trained CHWs in community settings which involved weekly group sessions and 3 individual sessions with an interventionist, or (2) diabetes self-management education (DSM) comprised of 12 monthly group sessions delivered at a primary care clinic. The main outcome was change in United Kingdom Prospective Diabetes Study (UKPDS) estimated 5-year CVD risk at 12 months; secondary outcomes included weight loss, UKPDS risk score components (hemoglobin A1c, blood pressure, lipids) and use of medications affecting these components. Results: We screened 1102 and randomized 260 adults (age range 23-83, mean 55.9, 67% female, 48% black, 52% < college degree, mean A1c 7.6%, 37.3% A1c<7%, mean BMI 37.7 kg/m 2 ). Baseline demographic and CVD risk factors were equally distributed between the 2 groups of 130 participants each. At 12 months, 92.3% of LWL and 98% of DSM participants were retained. There was not a significant difference in 5-year UKPDS estimated CVD risk by arm (increased from 5.1% to 5.9% in LWL, and 5.5% to 6.1% in DSM, group comparison p=0.61). There was no evidence of effect modification by the pre-specified parameters of gender, race/ethnicity or baseline BMI. Follow-up A1c, SBP, and total cholesterol did not significantly differ by arm. Weight loss was greater in LWL compared to DSM participants (mean of 3.2% vs 1.1% of total body weight, p=0.02). At least 5% weight loss was achieved by 30.8% of LWL and 18.1% of DSM participants (p=0.02). Among LWL participants, 11.7% discontinued their diabetes medication compared to only 1.6% of DSM participants (p=0.005); 87.5% of those who discontinued medication had achieved an A1c<7% at follow-up, compared to 36.8% who continued using diabetes medication. Hypertension and lipid lowering medication use did not differ by arm. Conclusions: In a diverse sample of adults with diabetes, a CHW-delivered LWL intervention produced greater weight loss compared to a monthly DSM education, but did not appreciably alter 5-year UKPDS-estimated CVD risk, A1c, or CVD risk factors. The average weight loss achieved (3.2%) may not have been sufficient to produce meaningful changes in metabolic functioning and CVD risk factors, however, some participants in both interventions achieved improved glycemic control without use of medications.