Prospective observational cohort study to collect granular social determinants of health data for hematopoietic cell transplant patients

Abstract Background: Although patient (pt) social/financial barriers impact transplant (BMT) outcomes, comprehensive, granular social determinants of health (SDOH) data are not routinely collected across our field. Methods: We launched a single center, prospective, observational cohort study to determine BMT pts' social/financial barriers to care (clinicaltrials.gov NCT06431347). Pts were eligible if ≥18 years & received BMT financial approval. The study opened 7/2024 to allograft & 11/2024 to autograft pts. A dedicated staff member recorded detailed SDOH data, after explaining why these data are needed. Pts reporting urgent financial needs were offered social work referral. We examined multiple social (low pt/caregiver educational attainment/ English language level or low pt confidence with written medical information) & financial [household income below median of our city ($60,000) or cost-of-living/ medical expense insecurity or high-risk of financial toxicity by COST-FACIT score ≤24] barriers by pt ancestry & graft source. Results: From 7/1/24-6/30/25, 424 pts were screened for the study & most (401/424, 95%) enrolled; 11 (3%) died prior to enrollment & only 9 (2%) declined to participate (6 due to lack of interest & 3 discomfort with survey questions). Of 401 enrolled pts [232 (58%) male; median age 62 yrs, range 19-82; 240 (60%) with acute leukemia/MDS/MPN], most (394, 98%) enrolled ±45 days from BMT. 122 received auto- & 276 allo-grafts [246 (89%) HLA-matched & 30 (11%) HLA-mismatched (4 cord blood, 25 haploidentical, 1 mismatched unrelated)]. 271 had European ancestry & 129 non-European ancestry (39 African, 57 non-Black Hispanic, 23 Asian, 10 mixed non-European). In a sub-analysis of 107 participants exploring data collection acceptability & reception, most felt the data collection process was comfortable (89, 83%; with only 1 pt reporting it was uncomfortable) & low-effort (90, 84%; with none reporting high-effort), with ~60% recognizing (& none disagreeing) that the data gathered will help our care teams understand their specific social situation & that of all our pts. Overall, 54 (13%) pts reported ≥1 social barrier: 27 (8%) pts &/or their caregiver had an education level less than high school & for 8 (2%) one of the pair did not speak English, & 32 (8%) pts needed help with written medical instructions. 173 (43%) pts reported ≥1 financial barrier: 84 (21%; median household size of 2, range 1-12) had a combined household income <$60,000 in the past year [36 (9%) had <$30,000]; 76 (19%) had cost-of-living &/or medical expense insecurity [41 (10%) housing, 36 (9%) food, 30 (7%) transportation, 26 (6%) utility, 40 (10%) medication/co-pays/insurance]; & 116 (29%) had high-risk of financial toxicity based on COST-FACIT score ≤24. Importantly, different SDOH measures classified different subsets of pts as vulnerable, with 54 (13%) pts having ≥1 social barrier but only 11 (3%) with ≥2 & only 2 (<1%) all 3; similarly, 173 pts had ≥1 financial barrier but only 74 (18%) had ≥2 & only 29 (7%) all 3. By ancestry, compared with Europeans, non-European pts had greater social & financial vulnerability, with >2x the proportion having ≥1 [9% vs 22%, P<.001] or ≥2 [1% vs 7%, P<.001] social barriers & over 60% more with ≥1 [36% vs 59%, P<.001] & >2x with ≥2 [13% vs 29%, P<.001] financial barriers [including 2x as many with cost-of-living &/or medical insecurity (15% vs 28%, P=.003) or with high risk of financial toxicity based on COST-FACIT score ≤24 (22% vs 44%, P<.001)]. Non-European HLA-mismatched allograft recipients (n=10), a population with high medical needs, had high proportions with ≥1 social (40%) & ≥1 financial (80%) barriers. Overall, 80 (20%) pts reported urgent financial needs with 53 (13%) referred to social work, most of whom (37/53, 70%) received grant support &/or financial/resource counselling.Conclusions: We describe detailed SDOH data for pts across a department at a large cancer center & show the data collection process to be feasible & acceptable to pts with a very high participation rate. We demonstrate the application of these data to health outcomes research (ie describing the social/financial barriers impacting specific at-risk populations) & to pt care (by supporting identification & addressing of urgent pt social/ financial needs). Our work will support standardized SDOH data collection processes in the BMT field & across oncology, & inform pilot interventions to address social & financial barriers to care.