Optimizing the Dutch newborn screening for congenital hypothyroidism by incorporating amino acids and acylcarnitines in a machine learning-based model

Objective

Congenital hypothyroidism (CH) is an inborn thyroid hormone (TH) deficiency mostly caused by thyroidal (primary CH) or hypothalamic/pituitary (central CH) disturbances. Most CH newborn screening (NBS) programs are thyroid-stimulating-hormone (TSH) based, thereby only detecting primary CH. The Dutch NBS is based on measuring total thyroxine (T4) from dried blood spots, aiming to detect primary and central CH at the cost of more false-positive referrals (FPRs) (positive predictive value (PPV) of 21% in 2007-2017). An artificial PPV of 26% was yielded when using a machine learning-based model on the adjusted dataset described based on the Dutch CH NBS. Recently, amino acids (AAs) and acylcarnitines (ACs) have been shown to be associated with TH concentration. We therefore aimed to investigate whether AAs and ACs measured during NBS can contribute to better performance of the CH screening in the Netherlands by using a revised machine learning-based model.

Methods

Dutch NBS data between 2007 and 2017 (CH screening results, AAs and ACs) from 1079 FPRs, 515 newborns with primary (431) and central CH (84) and data from 1842 healthy controls were used. A random forest model including these data was developed.

Results

The random forest model with an artificial sensitivity of 100% yielded a PPV of 48% and AUROC of 0.99. Besides T4 and TSH, tyrosine, and succinylacetone were the main parameters contributing to the model's performance.

Conclusions

The PPV improved significantly (26-48%) by adding several AAs and ACs to our machine learning-based model, suggesting that adding these parameters benefits the current algorithm.

Machine learning to improve false-positive results in the Dutch newborn screening for congenital hypothyroidism

OBJECTIVE

The Dutch Congenital hypothyroidism (CH) Newborn Screening (NBS) algorithm for thyroidal and central congenital hypothyroidism (CH-T and CH-C, respectively) is primarily based on determination of thyroxine (T4) concentrations in dried blood spots, followed by thyroid-stimulating hormone (TSH) and thyroxine-binding globulin (TBG) measurements enabling detection of both CH-T and CH-C, with a positive predictive value (PPV) of 21%. A calculated T4/TBG ratio serves as an indirect measure for free T4. The aim of this study is to investigate whether machine learning techniques can help to improve the PPV of the algorithm without missing the positive cases that should have been detected with the current algorithm.

DESIGN & METHODS

NBS data and parameters of CH patients and false-positive referrals in the period 2007-2017 and of a healthy reference population were included in the study. A random forest model was trained and tested using a stratified split and improved using synthetic minority oversampling technique (SMOTE). NBS data of 4668 newborns were included, containing 458 CH-T and 82 CH-C patients, 2332 false-positive referrals and 1670 healthy newborns.

RESULTS

Variables determining identification of CH were (in order of importance) TSH, T4/TBG ratio, gestational age, TBG, T4 and age at NBS sampling. In a Receiver-Operating Characteristic (ROC) analysis on the test set, current sensitivity could be maintained, while increasing the PPV to 26%.

CONCLUSIONS

Machine learning techniques have the potential to improve the PPV of the Dutch CH NBS. However, improved detection of currently missed cases is only possible with new, better predictors of especially CH-C and a better registration and inclusion of these cases in future models.

Parents’ views on accepting, declining, and expanding newborn bloodspot screening

Introduction

The goal of newborn bloodspot screening (NBS) is the early detection of treatable disorders in newborns to offer early intervention. Worldwide, the number of conditions screened for is expanding, which might affect public acceptance. In the Netherlands, participation is high (>99%), but little is known about how parents perceive NBS. This study assessed parents’ views on accepting, declining and expanding NBS.

Methods

A total of 804 of 6051 (13%) invited parents who participated in NBS in the Netherlands during the last two weeks of December 2019, and 48 of 1162 (4%) invited parents who declined participation in NBS in 2019 and 2020, completed a questionnaire.

Results

The most important reason for parents to participate in NBS was to prevent health complaints, whereas the most important reason to decline NBS was parents’ viewpoint on life and the belief that the heel prick would be painful for the child. Compared to NBS participants, respondents who declined NBS were more actively religious, considered alternative medicine or lifestyle more important, were less inclined to vaccinate their child for infectious diseases, and reported more doubt about NBS participation (all differences p < .001). Informed choice was lower among respondents who declined NBS (44%) compared to participants in NBS (83%, p < .001), mostly due to insufficient knowledge. Of the NBS participants, 95% were positive about NBS expansion. Most NBS participants agreed to include conditions that could unintentionally reveal a diagnosis in the mother instead of the child (86%) or a condition that may not cause symptoms until later in the child’s life (84%).

Conclusion

Most participants made an informed decision to participate in NBS and are positive about screening for more conditions. Insights into parents’ views on (non-)participation and expansion of NBS can help to ensure that NBS suits the population needs while safeguarding ethical principles for screening.

The Methodological Quality and Challenges in Conducting Economic Evaluations of Newborn Screening: A Scoping Review

INTRODUCTION

Cost-effectiveness (CEA) and cost-utility analyses (CUA) have become popular types of economic evaluations (EE) used for evidence-based decision-making in healthcare resource allocation. Newborn screening programs (NBS) can have significant clinical benefits for society, and cost-effectiveness analysis may help to select the optimal strategy among different screening programs, including the no-screening option, on different conditions. These economic analyses of NBS, however, are hindered by several methodological challenges. This study explored the methodological quality in recent NBS economic evaluations and analyzed the main challenges and strategies adopted by researchers to deal with them.

METHODS

A scoping review was conducted according to PRISMA methodology to identify CEAs and CUAs of NBS. The methodological quality of the retrieved studies was assessed quantitatively using a specific guideline for the quality assessment of NBS economic evaluations, by calculating a general score for each EE. Challenges in the studies were then explored using thematic analysis as a qualitative synthesis approach.

RESULTS

Thirty-five studies met the inclusion criteria. The quantitative analysis showed that the methodological quality of NBS economic evaluations was heterogeneous. Lack of clear description of items related to results, discussion, and discounting were the most frequent flaws. Methodological challenges in performing EEs of neonatal screenings include the adoption of a long time horizon, the use of quality-adjusted life years as health outcome measure, and the assessment of costs beyond the screening interventions.

CONCLUSIONS

The results of this review can support future economic evaluation research, aiding researchers to develop a methodological guidance to perform EEs aimed at producing solid results to inform decisions for resource allocation in neonatal screening.

Critical evaluation of the newborn screening for congenital hypothyroidism in the Netherlands

OBJECTIVE

Congenital hypothyroidism (CH) is defined as thyroid hormone deficiency at birth due to disorders of the thyroid gland (thyroidal CH, CH-T), or the hypothalamus or pituitary (central CH, CH-C). The Dutch Newborn Screening (NBS) strategy is primarily based on determination of thyroxine (T4) concentrations in dried blood spots followed, if necessary, by thyroid-stimulating hormone (TSH) and thyroxine-binding globulin (TBG) measurement enabling detection of both CH-T and CH-C. A calculated T4/TBG ratio serves as an indirect measure for free T4. A T4/TBG ratio ≤ 17 in a second heel puncture is suggestive of CH-C.

DESIGN AND METHODS

In the present study, we evaluated 11 years of Dutch CH NBS using a database of referred cases by assessing the contribution of each criterion in the unique stepwise T4-TSH-TBG NBS algorithm.

RESULTS

Between 2007 and the end of 2017, 1 963 465 newborns were screened in the Netherlands. Use of the stepwise algorithm led to 3044 referrals and the identification of 612 CH cases, consisting of 496 CH-T, 86 CH-C, and 30 CH of unknown origin diagnoses. We detected 62.8% of CH-C cases by the T4/TBG ratio in the second heel puncture. The positive predictive value (PPV) of the stepwise T4-TSH-TBG NBS algorithm was 21.0%.

CONCLUSION

This evaluation shows that the Dutch stepwise T4-TSH-TBG NBS algorithm with a calculated T4/TBG ratio is of great value for the detection of both CH-T and CH-C in the Netherlands, at the cost of a lower PPV compared to TSH-based NBS strategies.

Evaluating progress towards triple elimination of mother-to-child transmission of HIV, syphilis and hepatitis B in the Netherlands

Background

In 2014 the World Health Organisation (WHO) established validation criteria for elimination of mother-to-child transmission (EMTCT) of HIV and syphilis. Additionally, the WHO set targets to eliminate hepatitis, including hepatitis B (HBV). We evaluated to what extent the Netherlands has achieved the combined WHO criteria for EMTCT of HIV, syphilis and HBV.

Methods

Data of HIV, syphilis and HBV infections among pregnant women and children (born in the Netherlands with congenital infection) for 2009–2015, and data required to validate the WHO criteria were collected from multiple sources: the antenatal screening registry, the HIV monitoring foundation database, the Perinatal Registry of the Netherlands, the national reference laboratory for congenital syphilis, and national HBV notification data.

Results

Screening coverage among pregnant women was > 99% for all years, and prevalence of HIV, syphilis and HBV was very low. In 2015, prevalence of HIV, syphilis and HBV was 0.06, 0.06 and 0.29%, respectively. No infections among children born in the Netherlands were reported in 2015 for all three diseases, and in previous years only sporadic cases were observed In 2015, treatment of HIV positive pregnant women was 100% and HBV vaccination of children from HBV positive mothers was > 99%. For syphilis, comprehensive data was lacking to validate WHO criteria.

Conclusions

In the Netherlands, prevalence of maternal HIV, syphilis and HBV is low and congenital infections are extremely rare. All minimum WHO criteria for validation of EMTCT are met for HIV and HBV, but for syphilis more data are needed to prove elimination.

Sensitivity of fetal RHD screening for safe guidance of targeted anti-D immunoglobulin prophylaxis: prospective cohort study of a nationwide programme in the Netherlands

OBJECTIVE

 To determine the accuracy of non-invasive fetal testing for the RHD gene in week 27 of pregnancy as part of an antenatal screening programme to restrict anti-D immunoglobulin use to women carrying a child positive for RHD DESIGN:  Prospectively monitoring of fetal RHD testing accuracy compared with serological cord blood typing on introduction of the test. Fetal RHD testing was performed with a duplex real time quantitative polymerase chain reaction, with cell-free fetal DNA isolated from 1 mL of maternal plasma The study period was between 4 July 2011 and 7 October 2012. The proportion of women participating in screening was determined.

SETTING

 Nationwide screening programme, the Netherlands. Tests are performed in a centralised setting.

PARTICIPANTS

 25 789 RhD negative pregnant women.

MAIN OUTCOME MEASURES

 Sensitivity, specificity, false negative rate, and false positive rate of fetal RHD testing compared with serological cord blood typing; proportion of technical failures; and compliance to the screening programme.

RESULTS

 A fetal RHD test result and serological cord blood result were available for 25 789 pregnancies. Sensitivity for detection of fetal RHD was 99.94% (95% confidence interval 99.89% to 99.97%) and specificity was 97.74% (97.43% to 98.02%). Nine false negative results for fetal RHD testing were registered (0.03%, 95% confidence interval 0.01% to 0.06%). In two cases these were due to technical failures. False positive fetal RHD testing results were registered for 225 samples (0.87%, 0.76% to 0.99%). Weak RhD expression was shown in 22 of these cases, justifying anti-D immunoglobulin use. The negative and positive predictive values were 99.91% (95% confidence interval 99.82% to 99.95%) and 98.60% (98.40% to 98.77%), respectively. More than 98% of the women participated in the screening programme.

CONCLUSIONS

 Fetal RHD testing in week 27 of pregnancy as part of a national antenatal screening programme is highly reliable and can be used to target both antenatal and postnatal anti-D immunoglobulin use.

Burden of Sickness Absence Due to Chronic Disease in the Dutch Workforce from 2007 to 2011

INTRODUCTION

Chronic diseases are associated with productivity loss costs due to sickness absence. It is not always clear, however, which chronic diseases in particular are involved with how many sickness days and associated costs.

OBJECTIVE

To determine the prevalence, additional days of sickness absence, and associated costs of chronic diseases among the Dutch working population from 2007 to 2011.

METHODS

Prevalence of chronic diseases and additional days of sickness absence were derived from the Netherlands Working Conditions Survey (NWCS) from 2007 to 2011. The cost of each sickness absence day was based on linked personal income data. We used multiple regression analysis to derive the unconfounded additional days of sickness absence due to each chronic disease.

RESULTS

Annually, approximately 37 % of the Dutch working population reported some type of chronic physical or psychological disease. No clinically relevant changes in prevalence of specific chronic diseases were observed in the studied period, nor in the number of additional sickness absence days or associated costs. The national financial burden due to sickness absence associated with chronic musculoskeletal disorders amounted to €1.3 billion annually.

CONCLUSIONS

Chronic diseases result in substantial productivity loss due to sickness absence. Given the ageing population, the proposed increase in the state pension age and an increase in sedentary lifestyle and obesity, the prevalence of chronic diseases may be expected to rise. Coordinated efforts to maintain and improve the health of the working population are necessary to minimize socioeconomic consequences.

Estimating adolescent risk for hearing loss based on data from a large school-based survey

OBJECTIVES

We estimated whether and to what extent a group of adolescents were at risk of developing permanent hearing loss as a result of voluntary exposure to high-volume music, and we assessed whether such exposure was associated with hearing-related symptoms.

METHODS

In 2007, 1512 adolescents (aged 12-19 years) in Dutch secondary schools completed questionnaires about their music-listening behavior and whether they experienced hearing-related symptoms after listening to high-volume music. We used their self-reported data in conjunction with published average sound levels of music players, discotheques, and pop concerts to estimate their noise exposure, and we compared that exposure to our own "loosened" (i.e., less strict) version of current European safety standards for occupational noise exposure.

RESULTS

About half of the adolescents exceeded safety standards for occupational noise exposure. About one third of the respondents exceeded safety standards solely as a result of listening to MP3 players. Hearing symptoms that occurred after using an MP3 player or going to a discotheque were associated with exposure to high-volume music.

CONCLUSIONS

Adolescents often exceeded current occupational safety standards for noise exposure, highlighting the need for specific safety standards for leisure-time noise exposure.

Adolescents and MP3 players: too many risks, too few precautions

OBJECTIVE

The goal was to assess risky and protective listening behaviors of adolescent users of MP3 players and the association of these behaviors with demographic characteristics and frequency of use.

METHODS

In 2007, 1687 adolescents (12-19 years of age) in 68 classes in 15 Dutch secondary schools were invited to complete questionnaires about their music-listening behaviors.

RESULTS

. Ninety percent of participants reported listening to music through earphones on MP3 players; 32.8% were frequent users, 48.0% used high volume settings, and only 6.8% always or nearly always used a noise-limiter. Frequent users were >4 times more likely to listen to high-volume music than were infrequent users, and adolescents in practical prevocational schools were more than twice as likely to listen to high-volume music as were those attending preuniversity education.

CONCLUSIONS

When using MP3 players, adolescents are very likely to engage in risky listening behaviors and are unlikely to seek protection. Frequent MP3 player use is an indicator of other risky listening behaviors, such as listening at high volumes and failing to use noise-limiters.

Strategies for the prevention of MP3-induced hearing loss among adolescents: expert opinions from a Delphi study

OBJECTIVE

To identify parties involved in the prevention of MP3-induced hearing loss among adolescents and potentially effective prevention strategies and interventions.

METHODS

Thirty experts in fields such as scientific research, medical practice, community health professions, education, youth work, music entertainment, and enforcement authorities participated in a qualitative, electronic, 3-round, Web-based Delphi study.

RESULTS

Multiple parties involved in the prevention of MP3-induced hearing loss among adolescents were identified; the most relevant are the adolescents themselves, their parents, manufacturers of MP3 players and earphones, and the authorities. The experts did not expect that adolescents in general would perform the necessary protective behaviors to prevent MP3-induced hearing loss. Two environmental health protection measures were identified as both relevant and feasible to be implemented (ie, authorities encourage manufacturers to produce safer products, and public health campaigns will be held to improve knowledge of the risks of high-volume music, possible protective measures, and consequences of hearing loss).

CONCLUSIONS

Authorities, the music industry in general, and especially manufacturers of MP3 players and earphones should recognize their responsibility and create a safer MP3-listening environment by taking measures to protect today's youth from the dangers of listening to high-volume music on MP3 players.

Young people's exposure to loud music: a summary of the literature

OBJECTIVES

This descriptive summary of the literature provides an overview of the available studies (published before October 2006) on sociodemographic, psychosocial, and other correlates of risk and protective behaviors for hearing loss in young people aged 12 to 25 years.

METHODS

Publications were identified by a structured search in PubMed, PsycINFO, and Web of Science, and by scrutinizing the reference lists of relevant articles. The protection motivation theory was used as the theoretical framework for categorizing the psychosocial correlates.

RESULTS

Thirty-three papers were included that identified several sociodemographic and psychosocial correlates, such as age, gender, school level, ethnicity, music preference, physical activity, social influence, and free supply of hearing protection.

CONCLUSIONS

For the development of effective interventions we recommend theory-based longitudinal studies among those frequently exposed to loud music to assess these correlates in greater depth.