Introduction
Cystic fibrosis (CF), the most common life-shortening autosomal recessive disease, is diagnosed either through newborn screening (NBS) or the traditional method of performing a sweat test after signs/symptoms appear. Although the original estimates suggest that CF affects about one in every 2000 live births (1:2000) in Western European regions¹, recent data indicate great regional variation^2,3 with a range from 1:1500 to 1:10,000. The Republic of Ireland seems to have a relatively high incidence with an estimate of approximately 1:1500 in one study⁴. Previous research has examined clinical aspects in the Irish population⁵, but no study has characterized the epidemiology of CF in Ireland based on data at the time of diagnosis.

Although a working group convened by the Chief  Medical Officer of the Department of Health and Children in 1999 recommended establishing a NBS programme for CF, primary sweat testing after appearance of signs/symptoms of the disease or a positive family history remains the established method of diagnosing CF in the Republic of Ireland. Since sites doing sweat tests are well known, we organized this study to quantify costs and new CF patients during 2001-2003. In addition, we determined the age of diagnosis using The Cystic Fibrosis Registry of Ireland.

Methods
To quantify the number of sweat tests being performed in Ireland, we surveyed all clinical centres that were performing sweat tests. Each centre received a questionnaire that requested information on the total number of sweat tests performed during 2001, 2002, and 2003, and all responded fully. These sites were also asked to report the sweat test method used and the outcomes. The questionnaire summaries were used to compute the national number of sweat tests performed and how many of these were positive (thus diagnosing CF). Finally, the above data allowed us to compute the incidence of CF in Ireland for 2001-2003 using the mid-2002 population data collected from the Irish Central Statistics Office⁶. In collaboration with the University of Wisconsin, diagnostic data from the Cystic Fibrosis Registry of Ireland were analyzed in a similar fashion to previously published U.S. Registry studies⁷.

The data from the collected questionnaires also supplied some of the necessary information to estimate the cost of performing sweat tests for CF diagnoses nationally. All of the costs were computed in Euro during 2003 using expenses for both labour (per hour) and materials obtained, from three of the major sweat testing centres in Dublin: Our Lady’s Children’s Hospital (Crumlin), The National Children’s Hospital (Tallaght), and The Children’s University Hospital (Temple Street). We only considered direct costs in this estimation and calculated method-specific costs because we learned that three different procedures are being used for sweat testing in Ireland (Table 1). Both the Wescor electrolyte (chloride and/or sodium) method and the Wescor conductivity method required approximately one hour to complete, while the filter paper chloride method involved approximately one hour and 45 minutes to complete. In order to calculate the annual total cost for diagnosing CF, we computed the cost per test at each site based on the method multiplied by the total number of tests.

The final objective of this study was to determine the gender- specific average age of diagnosis of CF patients in Ireland. Initially, we reviewed the hospital charts of the Crumlin CF Centre at Our Lady’s Children’s Hospital which receives CF patient referrals from the entire country; the charts were then abstracted for 36 patients diagnosed in 2001, 2002, and 2003 of whom 12 had meconium ileus (MI) or intestinal obstruction of the newborn. Subsequently, we analyzed observations on de-identified data according to gender, age, and the presenting symptoms at diagnosis for Irish patients registered during 2004-2006.

Results
There were a total of 17 hospitals performing sweat tests. The number of annual tests per centre ranged from 3 to 368. Four centres responsible for 48% of the Irish sweat tests used filter paper/electrolyte analysis, five (30%) used the Wescor collection system with electrolyte analysis, and the remaining eight did Wescor/conductivity testing (22% of annual sweat tests). As shown in Table 1, the 3-year annual average of the number of annual sweat tests was 2025, and the 3-year annual average number of positive sweat tests was 44. Thus, there was an average of 45 negative sweat tests for each CF patient diagnosed (range, 40 to 56). Based on these data, the incidence of CF in Ireland was 1:1353 during 2001-2003 (95% CI=1:1045-1916).

The costs computed for sweat tests by three methods are shown in Table 2. The direct costs data indicate that the national expense for performing 1,823 sweat tests in 2001 was €106,229. The cost per sweat test was €58.27 and the direct cost per patient diagnosed was €2,414. Using the 3-year average data, we calculated €117,160 total, €57.86 per sweat test and €2663 per patient diagnosed from 2001-2003.

Based on the sample of 36 patients referred in 2001-2003 to Our Lady’s Children’s Hospital in Crumlin, which is 27% of the total identified nationally, the mean age of diagnosing of CF was 20.7 months, including infants with meconium ileus (MI); however, without MI patients, the mean age was 31.0 months. The patient sample included 19 females and 17 males. The mean age of diagnosis of the male patients was 18.6 months including the patients with MI and 24.3 months excluding the MI population, while the corresponding ages for females were 22.6 and 35.7 months, respectively.

Table 3 presents a summary of observations at diagnosis on patients enrolled through 2006 in the Cystic Fibrosis Registry of Ireland. MI was present in 18%, which is similar to U.S. data⁷. The average age of diagnosis was delayed to 24.6 months, but some differences were identified when presenting signs/symptoms and gender-specific comparisons were performed according to clinical factors leading to a sweat test. Specifically, the mean age of 133 patients presenting principally with respiratory symptoms at diagnosis (56.0 months; ±7.79 ) was older (p<.0001) than the 231 presenting with later gastrointestinal symptoms (20.8 months ±2.77). In addition, as shown in Table 3, females experienced a longer delay overall, particularly when they presented with respiratory symptoms, i.e., girls were 78.9 months old and boys 33.5 on the average. This difference was confirmed in a statistical comparison of median ages of diagnosis after respiratory signs/symptoms (66 girls=24.7 months compared to 67 boys=9.99 months; p=.0202 by the Kruskal-Wallis test). However, both females (p=.0003) and males (p=.0267) showed a much older age at diagnosis when they presented principally with respiratory signs/symptoms compared to gastrointestinal manifestations; the differences in the mean values were 55.6 and 22.1 months, respectively.

Discussion
During the current decade, many countries have transformed from the traditional method of CF diagnosis to NBS⁸. Early diagnosis of CF through NBS has been underway in Northern Ireland for nearly 25 years. There are a number of reasons why screening programs have been implemented such as:

1. concerns about difficulties and delays in diagnosis⁹;
   
   
2. the morbidity⁸ or mortality¹⁰ associated with delays;
   
   
3. disparities related to geographical or demographic factors such as gender⁷; and
   
   
4. anticipated financial savings in costs for diagnosis¹¹ and/or treatment¹².

The incidence of CF which we determined in the Irish population of 1:1353 confirms the previous estimate of 1:1461 by Devaney et. al.⁴ but was calculated with more reliable methodology over 3 years. Ireland’s incidence is the highest among Western European nations³ and is much higher than North American countries². This is probably attributable to a high CFTR mutation prevalence and consanguinity. To our knowledge, no European nation has evaluated the direct costs of diagnosing CF by the traditional method, taking into account all sweat tests performed. In fact, only one study elsewhere by Lee, et. al.¹¹ examined the costs of diagnosing CF in Wisconsin with traditional methods compared to newborn screening. Irish records, however, unlike those available in the USA, made it possible to study retrospectively the population of CF patients diagnosed nationally by sweat test analysis. It is interesting to compare regions with regard to CF diagnostic expenditures. In Wisconsin¹¹ the estimated cost per newly diagnosed patient with CF was calculated at U.S. $9,952 during 2000 using the traditional method of diagnosis and $9,025 with NBS. Ranieri et al¹³ in South Australia estimated the cost of diagnosis for CF with a neonatal screening program as $4,590. In a study conducted by Scotet et al⁹ in 1998, the estimated cost of the CF newborn screening program in Brittany, France, was $6,825 per child diagnosed with CF. In The Republic of Ireland, the average direct cost per CF diagnosis with the traditional method is lower at €2,663 or $3,435, using a conversion
factor of 1.29.

There are some limitations in this assessment of diagnostic costs. First, our estimates are limited to direct costs of a sweat test, and we found a lack of standardization as a number of sites were not diagnosing CF based on the recommendations of the UK National External Quality Assurance Schemes (NEQAS) as suggested to be the standard in Ireland¹⁴. There were many sites that were using conductivity¹⁵ as an alternative method for diagnosis instead of a quantitative pilocarpine test with electrolyte analyses. Although conductivity may correlate well with sweat chloride concentrations, it has not been approved for diagnosis of CF by any organization. In addition, there may be missed cases of CF; however, previous research2 in the USA has shown that these small numbers have very little effect on estimating incidence or for cost calculations.

It was possible to discover the gender difference among Irish CF patients because of the development of The Cystic Fibrosis Registry of Ireland. Data accrued and analyzed annually have continued to reveal this male/female disparity. These results confirm for the first time the disparity reported by Lai et al⁷ in age of diagnosis between male and female CF patients in the United States. Moreover, we have been able to extend those findings by demonstrating that it is specifically CF females presenting with respiratory symptoms who show the greatest delays. It is somewhat alarming these girls are on the average 78.9 months old at diagnosis, 45 months older then boys, particularly when one considers the worse prognosis of females after adolescence¹⁶. Because the survival of most CF patients ultimately depends on the severity of lung disease, it is also disconcerting that those with respiratory manifestations are diagnosed at a much older age than patients with gastrointestinal signs, especially when the onset of such symptoms seems to occur at a similar age⁷. This greater delay might be attributable to the non-specificity of the respiratory symptoms and/or medical practice habits.

The question arises as to why there would be such a strikingly inordinate delay in diagnosis in girls with CF. Lai et al⁷ after presenting unequivocal evidence of similarly evolving respiratory signs/symptoms, discussed physicians’ practice patterns and a potential “unconscious bias” related to cultural attitudes and “child-rearing practices.” Whatever the cause, there is no doubt that diagnosis through NBS eliminates the gender-related disparity^7,9. This diagnostic method also resolves the disparity related to the gastrointestinal versus respiratory symptoms onset because the lungs of CF patients are normal at birth. Therefore, although the costs of CF diagnoses by traditional methods are relatively modest in Ireland, the high incidence of CF, long diagnostic delays in general and especially for girls, the likelihood of preventable deaths¹⁰ and excessive therapeutic costs¹², make it important to implement a national NBS programme without delay. The nutritional benefits of early diagnosis through NBS are well established, as are the unique opportunities for enhanced pulmonary care, while the only risks of harm are manageable and preventable⁸.

References

1. Boat TF, Welsh MJ and Beaudet AL. Cystic fibrosis In the metabolic basis of inherited disease, McGraw-Hill, New York, 1989, 2649- 2680.
2. Kosorok MR, Wei W, Farrell PM. The incidence of cystic fibrosis. Stat Med 1996;15:449-62.
3. Southern KW, Munck A, Pollitt R, on behalf o the ECFS CF Neonatal Screening Working Group. A survey of newborn screening for cystic fibrosis in Europe. J Cyst Fibros 2007;6:57-65.
4. Devaney J, Glennon M, Farrell G, et al. Cystic fibrosis mutation frequencies in an Irish population. Clin Genet 2003: 63: 121-125.
5. Cashman SM, Patino A, Delgado MG, Byrne L, Denham B, DeArce M. The Irish cystic fibrosis database. J Med Genet, 1995;32:92-975.
6. Central Statistics Office, Ireland. www.indexmundi.com/Ireland/.
7. Lai HC, Kosorok MR, Laxova A, Makholm LM, Farrell PM. Delayed diagnosis of US females with cystic fibrosis. Am J Epidemiol 2002;156:165-73.
8. Farrell PM. Improving the health of patients with cystic fibrosis through newborn screening. Advances in Pediatrics 2000;47:79- 115.
9. Scotet V, Braekeleer M, Roussey M, et al. Neonatal screening for cystic fibrosis in Brittany, France: assessment of 10 years’ experience and impact on prenatal diagnosis. Lancet 2000;356:789-94.
10. Grosse SD, Rosenfeld M, Devin OJ, Lai HJ, Farrell PM. Potential impact of newborn screening for cystic fibrosis on child survival: A systematic review and analysis. J Pediatr 2006;362-366.
11. Lee DS, Rosenberg MA, Peterson A, et al. Analysis of the costs of diagnosing cystic fibrosis with a newborn screening program. J Pediatr. 2003;142:617-623.
12. Sims EJ, Mugford M, Clark A, et al., on behalf of the UK Cystic Fibrosis Database Steering Committee. Economic implications of newborn screening for cystic fibrosis: a cost of illness retrospective cohort study. Lancet 2007;369:1187-95.
13. Ranieri E, Lewis BD, Gerace RL, et al. Neonatal screening for cystic fibrosis using immunoreactive trypsinogen and direct gene analysis: four years’ experience. Br Med J 1994; 308:1469-72.
14. Baumer JH. Evidence based guidelines for the performance of the sweat test for the investigation of cystic fibrosis in the UK. Arch Dis Child; Dec 2003; 88, 12; 1126-1127.
15. Lezana JL. Vargas MH. Karam-Bechara J. Aldana RS. Furuya ME. Sweat conductivity and chloride titration for cystic fibrosis diagnosis in 3834 subjects. J Cystic Fibrosis. 2003;2:1-7.
16. Rosenfeld M, Davis R, FitzSimmons S, et al. Gender gap in cystic fibrosis mortality. Am J Epidemiol 1997;145:794-803.