O-88 Early life social factors associated with mortality in Jamaicans with Sickle Cell Disease

Methods: A medical chart review was conducted using census sampling of 535 SCD individuals enrolled in a birth cohort between 1973 and 1981. Early life exposures included parental occupation classified into skill levels 1 to 4 with lower levels representing more manual tasks with lower educational attainment using the International Standard Occupational Classification, union status, household structure (who the child lives with), and social amenities rating (SAR) score at birth where higher score meant access or possession of more amenities. These SAR scores were classified into tertiles for statistical analysis. Mortality was the outcome, with censored observations assigned to participants not known to have died at the end of data collection. Kaplan-Meier survival probabilities, median survival times and incidence rates per 1000 person-years(/1000PY) were estimated. Hazard ratios from Cox proportional hazards (PH) models quantified association with mortality.

Results: There were 277(51.8%) males, 302(56.5%) with haemoglobin sickle cell (HbSS) genotype, 247(61.6%) who lived with both parents and 279(59.49%) with a parent in occupation skill level 2 while 36.0% and 36.2% were classified at SARS Z-score tertiles 1 and tertile 2, respectively. The median follow up age was 28.6±15.6 years and 163 (30.5%) of participants died. Overall mortality rate was 10.7/1000PY – 16.7/1000PY and 4.0/1000PY for HbSS and HbSC genotypes, respectively. Median survival was not attained for the sample but was 42.3(95% CI: 37.2–46.6) years in those with HbSS genotype. Multivariable Cox PH model showed increased mortality associated with genotype (HbSS vs HbSC: (HR (95%CI) = 6.3(3.4–11.7), p<0.001), and household structure (Single(mother) vs both parents: (HR (95%CI) =1.6(1.1–2.2), p=0.019)).

Conclusion: Genotype remains the strongest clinical predictor of mortality, however early life factors inclusive of family structure has significant influence. Possible interventions should consider the role of early life social factors on long term outcomes.