TY - JOUR
T1 - Determinants of 25(OH)D sufficiency in obese minority children
T2 - Selecting outcome measures and analytic approaches
AU - Zhou, Ping
AU - Schechter, Clyde
AU - Cai, Ziyong
AU - Markowitz, Morri
PY - 2011/6
Y1 - 2011/6
N2 - Objective: To highlight complexities in defining vitamin D sufficiency in children. Study design: Serum 25-(OH) vitamin D [25(OH)D] levels from 140 healthy obese children age 6 to 21 years living in the inner city were compared with multiple health outcome measures, including bone biomarkers and cardiovascular risk factors. Several statistical analytic approaches were used, including Pearson correlation, analysis of covariance (ANCOVA), and "hockey stick" regression modeling. Results: Potential threshold levels for vitamin D sufficiency varied by outcome variable and analytic approach. Only systolic blood pressure (SBP) was significantly correlated with 25(OH)D (r = -0.261; P = .038). ANCOVA revealed that SBP and triglyceride levels were statistically significant in the test groups [25(OH)D <10, <15 and <20 ng/mL] compared with the reference group [25(OH)D >25 ng/mL]. ANCOVA also showed that only children with severe vitamin D deficiency [25(OH)D <10 ng/mL] had significantly higher parathyroid hormone levels (Δ = 15; P = .0334). Hockey stick model regression analyses found evidence of a threshold level in SBP, with a 25(OH)D breakpoint of 27 ng/mL, along with a 25(OH)D breakpoint of 18 ng/mL for triglycerides, but no relationship between 25(OH)D and parathyroid hormone. Conclusions: Defining vitamin D sufficiency should take into account different vitamin D-related health outcome measures and analytic methodologies.
AB - Objective: To highlight complexities in defining vitamin D sufficiency in children. Study design: Serum 25-(OH) vitamin D [25(OH)D] levels from 140 healthy obese children age 6 to 21 years living in the inner city were compared with multiple health outcome measures, including bone biomarkers and cardiovascular risk factors. Several statistical analytic approaches were used, including Pearson correlation, analysis of covariance (ANCOVA), and "hockey stick" regression modeling. Results: Potential threshold levels for vitamin D sufficiency varied by outcome variable and analytic approach. Only systolic blood pressure (SBP) was significantly correlated with 25(OH)D (r = -0.261; P = .038). ANCOVA revealed that SBP and triglyceride levels were statistically significant in the test groups [25(OH)D <10, <15 and <20 ng/mL] compared with the reference group [25(OH)D >25 ng/mL]. ANCOVA also showed that only children with severe vitamin D deficiency [25(OH)D <10 ng/mL] had significantly higher parathyroid hormone levels (Δ = 15; P = .0334). Hockey stick model regression analyses found evidence of a threshold level in SBP, with a 25(OH)D breakpoint of 27 ng/mL, along with a 25(OH)D breakpoint of 18 ng/mL for triglycerides, but no relationship between 25(OH)D and parathyroid hormone. Conclusions: Defining vitamin D sufficiency should take into account different vitamin D-related health outcome measures and analytic methodologies.
UR - http://www.scopus.com/inward/record.url?scp=79956054932&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79956054932&partnerID=8YFLogxK
U2 - 10.1016/j.jpeds.2010.11.034
DO - 10.1016/j.jpeds.2010.11.034
M3 - Article
C2 - 21232761
AN - SCOPUS:79956054932
SN - 0022-3476
VL - 158
SP - 930-934.e1
JO - Journal of Pediatrics
JF - Journal of Pediatrics
IS - 6
ER -