The status of neonatal screening in China,2013

Abstract

Objectives

To investigate the status of neonatal screening in China in 2013.

Method

All Chinese neonatal screening laboratories were asked to submit information about the number of newborns screened and for which diseases, the rate of acceptable blood spots, and the time between sample collection and delivery to laboratories and reporting.

Results

Of the 202 laboratories, 194 (96%) submitted data. In 2013, 73.6% (142/193) of laboratories testing for phenylketonuria (PKU), 72.7% (141/194) for congenital hypothyroidism (CH), 58.9% (43/73) for glucose-6-phosphate dehydrogenase (G6PD) deficiency and 53.1% (34/64) for congenital adrenal hyperplasia (CAH) screened more than 30,000 newborns. The incidences of PKU, CH, G6PD deficiency, and CAH were 1:12189, 1:2281, 1:44 and 1:6084, respectively. The estimated coverage of screening for PKU, CH, G6PD deficiency, and CAH were 86.3 ∼ 87.5%, 87.9 ∼ 89.1%, 24.0 ∼ 25.0% and 18.9% ∼ 19.9%, respectively. The rates of acceptable blood spots were ≥98% in 161 (87.0%) of 185 laboratories. More than half the laboratories could not achieve a time from collection to reporting of 5 days.

Conclusion

Neonatal screening has developed rapidly in China but the biggest challenge is still to increase coverage and expand the screening disease panel across the entire country. More government support is needed to make neonatal screening more efficient.

Introduction

Neonatal screening for disorders such as phenylketonuria (PKU), congenital hypothyroidism (CH), glucose-6-phosphate dehydrogenase (G6PD) deficiency and congenital adrenal hyperplasia (CAH) is routine in most developed and some developing countries. ¹ Neonatal screening in China started with a pilot study in 1981, and in the following two decades was extended to almost all provinces. ² Despite this, overall coverage was under 10% before 2001 and only 40% by 2007.^3,4 Data on key indicators of quality management, such as the rate of acceptable neonatal screening blood spots, the time intervals between samples being collected and received and reports being issued by laboratories has never been collated. We aimed to gather this data from all neonatal screening laboratories in China, calculate the incidence and coverage, and compare this with studies from other countries.

Methods

This survey covered all neonatal screening laboratories in China. Questionnaires were sent to the laboratories by the external quality assessment network reporting system. Laboratory management staff were asked to complete the questionnaires, and verify the data accuracy with a second staff member before submitting the results to the National Centre for Clinical Laboratories (NCCL). The questionnaire covered: the number of newborns screened, diagnostic standard, variety of disease, human resources, charges, the costs of reagents, the rates of acceptable blood spots, the minimums, maximums and averages of time from sample collection and receipt by laboratories, and the minimums, maximums and averages of time from sample receipt to reports being sent out by laboratories. Data from the Maternal and Child Health Information System and the National Bureau of Statistics ⁴ , showed that in China there were about 16,400,000 livebirths in 2013. ⁵

The rate of acceptable blood spots for each laboratory was calculated as: (the number of qualifying blood spot filter papers)/(total number of newborns screened in that laboratory) × 100%. A qualifying blood spot filter paper was required to have: 1) at least three blood spots, diameters of each spot greater than 8 mm; 2) natural infiltration of blood drops, blood spots on the front and back of filter paper; 3) no contamination; 4) no oozing ring.

Data were analyzed using SPSS 13.0. The overall incidence of neonatal screening disorders was calculated by: (total number of screened cases reported)/(sum of numbers of newborns screened). The parameters of routine statistics, including average, median, minimum, maximum and 95% confidence interval (CI) of average were calculated. Because eight of the 202 neonatal screening laboratories did not submit data, the range of the number of overall newborns screened in China in 2013 was estimated by: total number of newborns screened in the 194 laboratories + 8 times the 95% CI of average.

Results

Of the 202 screening laboratories in China in 2013, 194 (96.0%) submitted data to NCCL. All provinces except Tibet have screened for PKU and CH since 2001. In Hong Kong screening only included CH and G6PD deficiency. Of the 194 responding screening laboratories, 193 provided screening for PKU, 194 for CH, 73 for G6PD deficiency, and 64 for CAH (see table 1). There were 13,687,196 neonates screened for PKU, 13,938,427 for CH, 3,619,917 for G6PD deficiency and 2,834,792 for CAH in 2013. About 50% of laboratories screened more than 50,000 newborns. The percentages of laboratories screening more than 30,000 newborns for PKU, CH, G6PD deficiency, and CAH were 73.6% (142/193), 72.7% (141/194), 58.9% (43/73) and 53.1% (34/64), respectively.

Table 1.

Newborn screening in China, 2013.

Disease	Incidence	Number of laboratories	Total number of newborns screened	Annual workload of individual laboratories			95% CI
				Median	Minimum	Maximum
PKU	1:12,189	193	13,687,196	51,423	882	1,030,748	58,477 ∼ 83,360
CH	1:2,281	194	13,938,427	51,180	882	1,030,748	58,709 ∼ 84,986
G6PD deficiency	1:44	73	3,619,917	38,728	2,768	208,444	39,635 ∼ 59,541
CAH	1:6,084	64	2,834,792	33,781	528	182,182	34,019 ∼ 54,568

There were 1,123 neonates diagnosed with PKU, 6,112 with CH, 83,808 with G6PD deficiency, and 466 with CAH. (For incidence rates, see table 1). In 2013, screening coverage, calculated as the ratio of the number of infants screened to the number of live births expressed as a percentage, for PKU, CH, G6PD deficiency, and CAH was 86.3 ∼ 87.5%, 87.9 ∼ 89.1%, 24.0 ∼ 25.0% and 18.9% ∼ 19.9%, respectively. Rates of acceptable neonatal screening blood spots were submitted by 185 laboratories. Acceptable rates were higher than 99% in 122 (68.6%) laboratories. In 34 (18.4%) laboratories, acceptable rates were between 98% and 99%, and in 20 (10.8%) they were between 90% and 98%. Only four (2.2%) laboratories had acceptable rates below 90%.

The time between samples being collected and delivered to laboratories (duration α) and between samples being received and reports sent out (duration β) were calculated: 136 (73.5%) laboratories reported duration α maximums of more than 5 days, and 92 (49.7%) reported maximums of more than 5 days for duration β; 65 (35.1%) laboratories reported averages for duration α of more than 5 days, and 19 (10.3%) reported averages of more than 5 days for duration β.

Discussion

The World Health Organization (WHO) has suggested that the number of newborns screened in a laboratory should exceed 30,000 to assure efficiency and reduce costs.^6,7 Although most laboratories in China met this target in 2013, some did not reach even 1,000. Screening coverage was lower than 90% in 2013.

The incidence of PKU varies widely in different human populations. Turkey has the highest documented rate (1 in 2,600 births), while countries such as Finland and Japan have fewer than 1 in 100,000 live births.^8,9 In our study the incidence was 1 in 12,189 live births, slightly lower than previous studies of Gu (1 in 11,763) ¹⁰ and Zhan (1 in 11,572) ¹¹ in China.

The incidence of CH in the United States is around 1 in 3,600 births ¹² ; our study showed an incidence of 1 in 2,281 live births, slightly lower than previous studies of Gu (1 in 2,050) ¹⁰ and Zhan (1 in 2,047) ¹¹ , but higher than the incidence generally accepted by the world newborn screening community.

G6PD deficiency is more common throughout Africa, Asia, the Mediterranean, and the Middle East. According to WHO, 7.5% (1:14) of people have one or two genes for G6PD deficiency and 2.9% (1:35) are G6PD deficient. ¹³ This study showed an incidence of 2.3% (1:44), lower than the WHO findings. This might be due to the higher incidence in the south of China but much lower incidence in the north.

CAH incidence varies according to ethnicity and geographic area. The highest rates occur in the Yupic Eskimos of Alaska (1:280 births) ¹⁴ . High rates are also been reported in Brazil (1:7500) and the Philippines (1:7000). ¹⁵ In the United States, the incidence of CAH was lower in African Americans than in the white population (1:42,000 vs 1:15,500). ¹⁶ In our study the incidence of CAH in China was 1 in 6084 live births, a comparatively high rate. The screening coverage of CAH was only 18.9% ∼ 19.9% in 2013, and attention should be focused on increasing this and enabling early intervention for CAH cases.

Before 2001, screening coverage in China was about 10%, and only for PKU and CH. ³ This increased in 2006 and 2007 to 31.27% and 39.96%, respectively. ² In 2013, the screening coverage of PKU and CH had exceeded 85%, but is still significantly below that in other developed countries.^17–19

In many underdeveloped regions of China, maternity wards are a considerable distance away from neonatal screening laboratories, and transportation is problematic. Because of this, the acceptability rate of blood spots is very important in saving time and for early diagnosis and intervention for cases. In 161 of 185 (87.0%) screening laboratories, the acceptability rate of blood spots was above 98%. In laboratories with lower acceptability rates, staff training and using more advanced blood collection methods may improve performance. The national newborn screening technique guidelines issued by the Chinese Ministry of Health in 2004 suggested that the durations between sample collection and delivery to laboratories, and between sample receipt and report issuing should be within five days, to assure the quality of neonatal screening. In our study more than half of screening laboratories could not achieve this standard. The improvement needed will require not only the efforts of staffs of laboratories, but perhaps more importantly, government investment.

The biggest challenge to neonatal screening in China remains increasing coverage and the number of diseases screened. Laboratories with lower acceptance rates for blood spots also need to improve, and the durations between sample collection, delivery to laboratories, sample receipt and issuing of reports should be controlled and decreased to less than five days. Improvements in managing the morbidity and mortality associated with inborn errors of metabolism should take precedence over socio-economic and technology development. More government support is still needed to make neonatal screening more efficient. Neonatal screening should be improved with a satisfactory control system, including wider screening coverage, shorter time to report, and screening for a greater range of diseases.