If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Respirology, St. Michael's Hospital, 30 Bond Street, Toronto, M5B 1W8, CanadaKeenan Research Centre, Li Ka Shing Knowledge Institute of St. Michael's Hospital, 209 Victoria Street, Toronto, Ontario, M5B 1T8, Canada
1 Institute of Health Policy, Management and Evaluation, University of Toronto, 155 College Street, Toronto, Ontario, Canada M5T 3M6.
Affiliations
Department of Respirology, St. Michael's Hospital, 30 Bond Street, Toronto, M5B 1W8, CanadaKeenan Research Centre, Li Ka Shing Knowledge Institute of St. Michael's Hospital, 209 Victoria Street, Toronto, Ontario, M5B 1T8, Canada
Meakins-Christie Laboratories, Research Institute of McGill University Health Centre, Montreal, CanadaMontreal Children's Hospital-McGill University Health Centre, Respiratory Division, Montreal, Canada
The impact of newborn screening (NBS) for cystic fibrosis (CF) on early indicators of long-term health was evaluated in the context of government-sponsored healthcare and access to current therapies.
Methods
Using data from the Canadian CF Registry between 2008 and 2013, we compared the rates of respiratory infections and markers of nutritional status in those diagnosed through NBS to those who were diagnosed clinically within the same time period using Mann–Whitney and Fischer's exact test as appropriate.
Results
The study included 303 subjects, 201 in the NBS group and 102 in the non-NBS group. NBS patients were diagnosed earlier and had their first clinic visit at a younger age. Pancreatic insufficiency was less common in NBS patients. The incidence of Pseudomonas aeruginosa and Staphylococcus aureus were lower in NBS patients. After adjusting for age at clinic visit, gender, pancreatic status, and Pseudomonas aeruginosa infection status, mean z-scores for weight-for-age and height-for-age were higher in NBS patients, with no differences in BMI-for-age.
Conclusions
NBS programs for CF lead to improved long-term health outcomes for the CF population.
There is overwhelming evidence that supports the establishment of newborn screening (NBS) programs for cystic fibrosis (CF) to improve the overall survival, long-term growth and health outcomes of people living with CF [
]. NBS programs for CF have been widely adopted in several European countries, Australia, New Zealand and the majority of North America — with the exception of Quebec (QC) and Mexico [
The Australian Paediatric Respiratory Group. Diagnosis of cystic fibrosis after newborn screening: the Australasian experience—twenty years and five million babies later: a consensus statement from the Australasian Paediatric Respiratory Group.
]. NBS most often consists of a two-step screening test using immunoreactive trypsinogen (IRT) from a blood spot procured during the first few days of life that is followed by a search for common mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, although other approaches exist. A positive CF screen results in a referral to a CF clinic to confirm the diagnosis through additional blood and sweat testing. Babies with a high risk of having CF are identified and once the diagnosis is confirmed, education and treatment is initiated, typically in the first 4 to 6 weeks of life.
The benefits of NBS for CF have been studied in other parts of the world and reported in 2005 [
]. This may be due to a combination of early diagnosis through NBS and intervention, including mucolytic therapy, aggressive treatment of infection with Pseudomonas aeruginosa (P.aeruginosa), and aggressive nutritional supplementation to prevent malnutrition. Recognition of the need for early intervention in regards to treatments such as nutritional support, early eradication of P.aeruginosa, and airway clearance has resulted in improvements in clinical factors in young CF patients, such as lung function and nutritional status [
]. It is currently unknown if these interventions could mitigate the consequences of later diagnosis and diminish the benefits of newborn screening.
In Canada, NBS for CF was first introduced in the province of Alberta (AB) in April 2007, followed by Ontario (ON) a year later in April 2008. The screening algorithms differ slightly between AB and ON [
]. In AB, the IRT cutoff is >60 μg/L or 98th percentile and in ON, it is the 96th percentile. In both cases, this is followed by genetic testing using the TM Biosciences Tag-It 39+3 mutation kit. The diagnosis of CF is typically confirmed when two disease causing mutations are identified along with a sweat chloride value >60 mmol/L. QC is the last province in Canada that has not committed to adding CF testing to their NBS program. The objective of this study is to compare health outcomes in the first 6 years of life in children diagnosed through NBS in AB and ON, with children born in QC who did not have access to NBS. This comparison will allow us to evaluate the potential effects of NBS on early clinical outcomes specifically within the Canadian context where non-screened children would have the benefit of government-sponsored access to healthcare and medications. To our knowledge, this is the first study conducted to evaluate the impact of NBS programs for CF in Canada.
2. Materials and methods
2.1 Data source
There are 42 accredited Canadian CF clinics found in all provinces with the exception of Prince Edward Island and the three territories. All CF clinics receive annual incentive grants from Cystic Fibrosis Canada (CFC) to support and enhance their clinical services that are contingent on submitting patient data to the Canadian CF Registry (CCFR). The CCFR is a patient registry and has been in existence since the early 1970s and currently contains over 110,000 annual records on more than 7100 unique individuals with CF who have attended a CF clinic in Canada. It is estimated that the majority of the Canadian CF population is represented within the CCFR, giving a comprehensive picture of the CF population in Canada.
The CCFR data undergo routine validation checks to ensure that they are free of errors. Discrepancies are resolved by directly contacting the reporting CF clinic so that the raw data can be reviewed and unusual data can be confirmed or revised. All individuals within the CCFR have provided informed consent for having their data collected and used for research purposes. This study was approved by the Research Ethics Board of Montreal Children's Hospital and the CCFR oversight committee.
Patient data from the CCFR collected between January 1, 2008 and December 31, 2013 were used in this study. During this time period, the CCFR only captured data on patients with a confirmed diagnosis of CF based on current guidelines [
]. Longitudinal clinical measurements such as height and weight, were recorded from the first stable visit of the year while other variables such as microbiology, pancreatic status, genotype, number of hospitalizations, length of hospital stay, and other CF-complications reflect events that occurred within the calendar year. Lung function is typically measured reliably at 6 years of age and older; therefore, lung function measurements were not included as an outcome for this study.
2.2 Study cohort
The study included two groups of patients followed from 2008 to 2013: (1) infants diagnosed through NBS in AB and ON identified as the NBS group; and (2) the non-NBS group comprised of children diagnosed with CF in QC. Subjects were followed for a maximum of 6 years to reflect the starting date of the NBS program for CF in both AB and ON. In the CCFR, the upper limit of 6 years for diagnosis covers more than 80% of Canadian CF patients, and those diagnosed after 6 years of age are more likely to have milder forms and thus may not benefit from NBS. Children in QC diagnosed prenatally, due to an affected sibling, or presenting with meconium ileus were excluded, as these circumstances would increase the likelihood of an early diagnosis and they would be unlikely to benefit from an NBS program. With respect to microbiology, once a subject was found to have a positive respiratory sample for a bacteria, they were categorized as positive for all subsequent years in the analysis. Subjects who never had a specimen culture taken were removed to avoid misrepresenting their infection status. We categorized patients into three genotype classifications: homozygous for F508del, heterozygous for F508del and other.
We examined multiple health outcomes of early clinical factors that have been well-established as key indicators of long-term health outcomes in CF [
]. The primary outcomes of this study were markers of nutritional status, specifically weight-for-age, height-for-age and BMI-for-age measurements. World Health Organization (WHO) growth charts were used for all calculations [
Development of a WHO growth reference for school-aged children and adolescents.
Bull World Health Organ.2007; 85 ([http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2636412&tool=pmcentrez&rendertype=abstract. Accessed August 16, 2015]): 660-667
]. Secondary outcomes included the incidence rate of infection with P.aeruginosa and S.aureus; the number of hospitalizations; and length of hospital stay.
2.3 Statistical analysis
Demographic and baseline characteristics between the NBS and non-NBS groups were compared using the Mann–Whitney test for continuous variables and the Fisher's exact test for categorical variables. We used generalized estimating equations (GEE) model and generalized linear mixed models (GLMM) to evaluate differences in health outcomes over time between the two study groups, as these models account for the correlated longitudinal nature of the data. In order to account for overdispersion of hospitalization data, we used a negative binomial distribution to model this outcome. Additional covariates included in the models were: age at the time of the clinic visit; gender; pancreatic status; P.aeruginosa and S.aureus infection status; CF clinic location; and genotype.
In order to rationalize our decision to combine the populations from AB and ON, a sensitivity analysis was conducted to investigate potential effects between the three provinces. The same regression models comparing outcomes used in the main analysis were done across three subgroups (AB vs. ON, AB vs. QC and ON vs. QC) instead of by NBS status.
For all analyses, p < 0.05 (two-tailed) was used as the criterion for statistical significance. Statistical analysis was performed using R software (version 3.1.2) with the following packages: geepack (version 1.2.0) and glmmADMB (version 0.8.0). Asymptotic 95% confidence intervals were calculated and added to the figures where appropriate.
3. Results
The study included 303 subjects: 201 in the NBS group (66 from AB and 135 from ON) and 102 in the non-NBS group (QC). There were no deaths reported in our study cohort and none have had a transplant.
There was no evidence of a statistical difference between NBS and non-NBS patients in terms of age (p = 0.10) and gender (p = 0.12). Children in the NBS group were diagnosed earlier with CF (p < 0.001) and had their first clinic visit recorded at a younger age (p < 0.001). Three children in the NBS group were diagnosed after the age of 1 year as their initial screens were positive but classified as categories B and C (high IRT levels detected but only one or no CF mutations found) [
]. Positive sweat tests appeared in subsequent follow-up testing resulting in a delayed diagnosis date. Non-NBS patients were also more likely to be pancreatic insufficient (p < 0.001) and attended fewer clinics (p < 0.001) (Table 1).
Table 1Baseline statistics of the NBS and non-NBS groups. All comparisons were calculated using the Fisher's exact test, except † which used the Mann–Whitney test.
When first diagnosed, those in the NBS group were taller and more likely to have a healthy weight as measured by their first z-scores for height-for-age and BMI-for-age (p < 0.001). The NBS group showed significantly better overall growth in height and weight as seen with their most recent z-scores for weight-for-age and height-for-age (p < 0.001) (Table 2).
Table 2The median and range of the z-scores for the health outcomes for the first and most recent visit (Nth) per subject. All comparisons were calculated using the Mann–Whitney test.
The NBS group had fewer and shorter hospitalizations (p < 0.001) due to pulmonary exacerbations (Fig. 1). Patients in the non-NBS group had a higher rate of P.aeruginosa than the NBS group (p < 0.001) and less than 30% of the NBS group had evidence of P.aeruginosa infection compared to more than 60% of the non-NBS group (p < 0.001) (Fig. 2). However, the prevalence of S.aureus is slightly higher in the non-NBS group compared to the NBS group (77.5% vs. 63.2% respectively, p = 0.013).
Fig. 1Mean number of hospital days (top) and hospitalizations (bottom) by reporting year for the NBS and non-NBS groups with 95% confidence interval bars.
Fig. 2Percentage of the first occurrence of colonization of P.aeruginosa and S.aureus with 95% confidence interval bars between the NBS and non-NBS groups for increasing age ranges.
Across the 6-year time period, we found higher growth rates in weight and height in the NBS group and a trend towards fewer hospitalizations and lower infection rates of P.aeruginosa (Table 3, Figs. 3 and 4). We also found that clinic location, genotype and the infection status of S.aureus had no effect on the health outcomes tested. Similar results can be found using only the homozygous F508del cohort (data not shown).
Table 3Results comparing the health outcome over time between the NBS and non-NBS groups using the statistical model specified. All visits per subject were included in the calculations. The estimate describes the average difference between the groups. CI: confidence interval. All comparisons were calculated using the GEE model, except † which used GLMM.
Fig. 3Cumulative percentage with 95% confidence interval bars of the first occurrence of colonization of P.aeruginosa and S.aureus between the NBS and non-NBS groups for increasing age ranges.
Fig. 4Regression curves (LOESS) for weight-for-age, height-for-age and BMI-for-age z-scores between the NBS and non-NBS groups. The NBS and non-NBS numbers at the bottom of the graph represent the sample sizes for each group for the age range indicated between the age markers along the x-axis.
We found no significant differences in our sensitivity analysis comparing health outcomes between AB vs. ON which validates our use of combining the two datasets (data not shown).
4. Discussion
Our data demonstrate that NBS resulted in better nutritional parameters, lower prevalence of infection with P.aeruginosa and S.aureus, and fewer hospitalizations in Canadian CF patients. Our results show that children diagnosed through the presence of symptoms are at an important disadvantage, compared to those diagnosed through NBS. The higher incidences of smaller stature and increased hospitalizations are likely to have long-term consequences in terms of need for more therapies and earlier recourse to lung transplantation [
] are aggressively treated as they can cause irreversible lung damage and our results indicate that those in the non-NBS group have higher incidences of P.aeruginosa suggesting that timely diagnosis and treatment may play a pivotal role in the long-term health outlook of people living with CF. While S.aureus is common in CF [
Relationship between nutritional status and lung function in cystic fibrosis: cross sectional and longitudinal analyses from the German CF quality assurance (CFQA) project.
Thorax.2002; 57 ([http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1746376&tool=pmcentrez&rendertype=abstract. Accessed July 20, 2015]): 596-601
], and thus patients who are diagnosed later, end up missing out on the possibility of early intervention to prevent lung damage and improve prognosis [
]. This is important, as novel therapies aimed at correcting the fundamental genetic defect in CF, while demonstrating remarkable improvements in health outcomes, may not reverse pre-existing lung damage [
], these were mostly done prior to the widespread availability of therapies for young children, including mucolytics (human recombinant DNase, hypertonic saline), aggressive approaches to the eradication and treatment of P.aeruginosa (inhaled aminoglycosides and other antimicrobials), and a better appreciation of the importance of aggressive nutritional support. It could be argued that such interventions could mitigate the effects of making symptom-based diagnosis, and thus diminish the benefits of newborn screening. In the present study, where there is access to mucolytic therapy and a focus on P.aeruginosa eradications, NBS had significant positive impact by reducing hospitalizations, prevalence of pathogenic bacteria, and improving nutritional status. In other words, current management strategies do not correct for the lack of NBS.
It should be emphasized that the NBS program was introduced 6 years ago in certain provinces within Canada and as a result, the data reflect many children of differing ages and that there are few children in the data who are over 5 years of age. This attrition of our dataset limits the ability to investigate the long-term impact of the NBS program and future studies are needed. Previous literature has demonstrated that non-screened patients had weight catch-up but not height catch-up in the first 6 years of life [
]. It is likely that those who were not screened profited from catch-up growth following diagnosis and treatment, but that such growth was more successful in adding weight compared to height, leaving the non-screened patients stunted, resulting in a relatively normal BMI. Another possible limitation is that pancreatic insufficiency is determined clinically by the treating physician and entered as a yes/no in the CCFR. We recognize that this may lead to an overestimation of pancreatic insufficiency status due to misclassification [
The strengths of our study include the longitudinal nature of our dataset and the fact that CCFR data captures essentially the entire Canadian CF population with similar access to care in the current context of recommended practice. Such use of registry data is currently viewed as highly valuable for healthcare planning [
]. Although we included important clinical outcomes in our study, we did not have access to information on illnesses and patterns of healthcare use prior to the diagnosis of CF which is a limitation of the dataset.
In conclusion, patients diagnosed through NBS in Canada demonstrate better nutritional status, fewer hospitalizations, and less carriage of P.aeruginosa in the 5 years following diagnosis, compared to those patients diagnosed clinically. This is despite all children having access to current recognized management strategies to optimize these outcomes. These therapies cannot make up for the absence of NBS.
Conflict of interest statement
No authors have any conflicts of interest to declare in relation to this manuscript.
Acknowledgments
We would like to thank the CF healthcare community for their work with the CCFR. We also wish to extend our deepest gratitude and appreciation to Canadian CF patients for allowing their data to be collected. This study would not have been possible without their support and participation. We would also like to thank CF Canada for their in-kind support.
References
McKay K.
Wilcken B.
Newborn screening for cystic fibrosis offers an advantage over symptomatic diagnosis for the long term benefit of patients: the motion for.
The Australian Paediatric Respiratory Group. Diagnosis of cystic fibrosis after newborn screening: the Australasian experience—twenty years and five million babies later: a consensus statement from the Australasian Paediatric Respiratory Group.
Development of a WHO growth reference for school-aged children and adolescents.
Bull World Health Organ.2007; 85 ([http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2636412&tool=pmcentrez&rendertype=abstract. Accessed August 16, 2015]): 660-667
Relationship between nutritional status and lung function in cystic fibrosis: cross sectional and longitudinal analyses from the German CF quality assurance (CFQA) project.
Thorax.2002; 57 ([http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1746376&tool=pmcentrez&rendertype=abstract. Accessed July 20, 2015]): 596-601
30018-2&id=gr1.jpg){kind=link}
30018-2&id=gr2.jpg){kind=link}
30018-2&id=gr3.jpg){kind=link}
30018-2&id=gr4.jpg){kind=link}