To explore the influencing factors of diagnosis time in infants with congenital hypothyroidism (CH).

A total of 11 689 live-born infants born in all midwifery institutions in Chengdu from January 1, 2016, to December 31, 2021, with positive or suspicious positive CH in initial thyroid stimulating hormone (TSH) screening and successfully recalled for re-examination were selected as the study subjects. Among them, 679 children with confirmed CH were divided into early-diagnosed CH group (n=467, diagnosed as CH ≤ 30 d after birth) and late-diagnosed CH group (n=212, diagnosed as CH > 30 d after birth). Chi-square test and Wilcoxon rank sum test were used to compare the gender, full-term delivery, birth season, location area, ownership and accreditation level of the birth hospital, whether the second blood spot sample for re-examination was collected at the birth hospital, as well as the turnaround time of dried blood spots on filter paper (DBS) before test, turnaround time of test report of DBS, and recall time for children with suspicious positive CH in CH initial screening. Among 11 689 infants, the receiver operating characteristic (ROC) curve of initial screening value of serum TSH was plotted to predict the risk of CH, and the area under curve (AUC) and Youden index were calculated to analyze the predictive value of initial screening value of serum TSH for the occurrence of CH. The procedures followed in this study complied with the ethical standards established by the Ethics Committee of Chengdu Women′s and Children′s Central Hospital, University of Electronic Science and Technology of China, and have been approved by the Ethics Committee [Approval No. 2021(41)].

① There were no significant differences in gender, full-term delivery, birth season, the ownership and accreditation level of birth hospital between early- and late-diagnosed CH group (P>0.05). The differences in location area of birth hospital, and whether the second blood spot sample for re-examination was collected at the birth hospital between two groups were statistically significant (χ²=26.39, 125.80; P<0.001). ②The turnaround time of DBS before test, turnaround time of test report of DBS, and recall time for children with suspicious positive CH in CH initial screening was 11 d (9 d, 14 d), 6 d (5 d, 7 d), and 3 d (1 d, 7 d) respectively in early-diagnosed CH group, all of which were shorter than those of late-diagnosed CH group at 14 d (11 d, 18 d), 7 d (6 d, 7 d), 11 d (8 d, 15 d), and all the differences were statistically significant (Z=-8.02, P<0.001; Z=-2.61, P=0.009; Z=-15.35, P<0.001). ③Among the 11 689 infants, result of ROC curve of initial screening value of serum TSH to predict CH in infants showed that the AUC was 0.871 (95%CI: 0.825-0.886, P<0.001). Based on the principal of maximum Youden index, the optimal cut-off value of initial screening value of serum TSH for predicting CH in infants was 14.7 μIU/mL, and the sensitivity, specificity and Youden index were 69.1%, 93.4% and 62.5%, respectively.

Infants with CH born in suburban hospital, with a longer recall time for suspicious positive CH in CH initial screening, and with a longer turnaround time of DBS before test may cause delay in the diagnosis of CH. Chengdu municipal health administrations should take corresponding measures for the above factors and shorten the time for diagnosing CH children. It is suggested to take the initial screening value of serum TSH 14.7 μIU/mL as the positive cut-off value for CH screening may be more conducive to shortening the time for diagnosing children with CH.