Newborn screening programs are vital to finding serious diseases in infants. Screening in the United States averages 4 million children a year for as many as 58 conditions in some states. But it’s not a perfect system, Rina Shaikh-Lesko writes in the October issue of Clinical Laboratory News. “States have different procedures for how babies’ blood gets collected and transported from hospitals to state labs, different testing methods and turnaround times, and different approaches for interpreting results, including setting cutoffs,” she writes.
A series of reports in the Milwaukee Journal Sentinel have documented the lack of uniformity in state-run newborn screening programs. In early January, CLN Stat
the Journal Sentinel's report about variable newborn screening cutoffs among states and how this could adversely affect children’s health. Shaikh-Lesko’s article offers a thorough summary of the Journal Sentinel's findings, highlighting a case in Wisconsin in which a high cutoff led to a child’s belated diagnosis for propionic acidemia.
In setting thresholds for newborn conditions states must juggle science, time constraints, and money issues. “Higher thresholds lessen the risk of missing true positive cases, but setting a cutoff too low also causes problems. If too many false positives turn up, limited resources are wasted on follow-up testing for infants who aren’t truly sick while raising families’ anxieties,” Shaikh-Lesko writes.
State public health laboratories are taking proactive steps to improve the screening process. According to Guisou Zarbalian, a senior specialist of newborn screening and genetics at Association of Public Health Laboratories (APHL), the November meeting of the Advisory Committee on Heritable Disorders in Newborns and Children will feature a report that documents such improvements. APHL’s Newborn Screening Quality Assurance/Quality Control Subcommittee is also drafting a guidance document that will serve as a resource for determining cutoffs. “The document is expected to discuss setting cutoffs for specific disease categories like amino acid disorders and to describe approaches for challenging a preliminary cutoff such as running known positives from other states,” according to Shaikh-Lesko.
APHL’s Committee on Newborn Screening and Genetics in Public Health recently polled 53 state and territory screening programs on their cutoff procedures and found that just under 50% had re-evaluated their cutoff procedures in response to problems such as too many false-positive results or missing cases. Survey respondents reported using a number of different approaches for establishing cutoffs, including Mayo Clinic database tools R4S and Collaborative Laboratory Integrated Reports (CLIR). “Rather than characterize a screening result strictly in terms of deviations from normal, R4S and CLIR analyze how the result fits within a disease range for a particular condition based on an extensive database of true positive cases,” Shaikh-Lesko writes.
Not everyone is sold on these database tools—some survey respondents seemed confused about how these systems operated or were skeptical of the technology.
Some experts have called for setting national thresholds. Others argue that this simply isn’t attainable in the laboratory setting. Pick up the October issue of CLN and learn more about other initiatives to help states evaluate their performance of newborn screening programs.