Hospital care utilization of infants born after IVF

A. Ericson1, K.G. Nygren2, P.Otterblad Olausson3 and B. Källén4,5

1 Centre for Epidemiology, National Board of Health and Welfare, SE-106 30, 2 Sophiahemmet, SE-114 86, 3 Centre for Epidemiology, National Board of Health and Welfare, SE-106 30 Stockholm and 4 Tornblad Institute, University of Lund, SE-232 65 Lund, Sweden


    Abstract
 Top
 Abstract
 Introduction
 Material and methods
 References
 
BACKGROUND: Infants born after IVF are often twins, and singleton IVF babies have an increased risk for preterm birth. Both conditions are likely to increase morbidity. We examined the frequency and duration of hospitalization required by babies born after IVF, and compared this information with all infants born in Sweden during the same time period. METHODS: We used a nationwide registration of IVF pregnancies from 1984 to 1997 and a nationwide register of all in-patient care up to the end of 1998. We identified 9056 live born infants after IVF treatment and compared them with 1 417 166 non-IVF live born infants. RESULTS: The highest odds ratio (OR ~3) was seen for neonatal hospitalization, but an increased OR (1.2–1.3) was noted for children up to 6 years of age. The OR for being hospitalized after IVF was 1.8, but when the analysis was restricted to term infants it was 1.3 and this excess was then explainable by maternal subfertility. Statistically significant increased ORs were seen for hospitalization for cerebral palsy (1.7), epilepsy (1.5), congenital malformation (1.8) or tumour (1.6), but also for asthma (1.4) or any infection (1.4). When information from the Swedish Cancer Registry was used, no excess risk for childhood cancer was found. The average number of days spent in hospital by IVF and non-IVF children was 9.5 and 3.6 respectively. CONCLUSIONS: The increased hospitalization of IVF children is, to a large extent, due to the increased incidence of multiple births. Therefore, the increased costs associated with this may be reduced by the use of single embryo transfers, with the savings in health care costs being offset against the increased number of embryo transfer cycles required to maintain the pregnancy rate.

Key words: cerebral palsy/costs/epilepsy/hospitalization/IVF


    Introduction
 Top
 Abstract
 Introduction
 Material and methods
 References
 
Infants born after IVF in Australia were more likely to require neonatal ventilator use than non-IVF infants, irrespective of whether they were born as a result of singleton or multiple pregnancies (Leslie et al., 1992Go). They were more likely to utilize the resources of neonatal intensive care units, but less likely to visit general practitioners or other health care workers during the first year of life (Leslie et al., 1998Go), a conclusion based on interviews with mothers of 95 infants born after IVF and 79 control infants born after natural conceptions. Interviews were performed at an infant age of 4 and 12 months. The increased need for neonatal care has been attributed to the well-known increased rate of preterm births and multiple births after IVF (Gissler et al., 1995Go; Bergh et al., 1999Go).

We undertook a study of IVF infants using central health registers in Sweden in order to study frequency and amount of hospitalization among these infants compared with all infants born in Sweden.


    Material and methods
 Top
 Abstract
 Introduction
 Material and methods
 References
 
Information on all infants born in Sweden after IVF procedures during the years 1984–1997 was collected from all 15 units where IVF were performed in Sweden during that period. Part of this material (up to 1995) has been published (Bergh et al., 1999Go). Each person living in Sweden gets a unique personal identification number which is extensively used in society, including all health care. We used the identification number of the mother and the date of IVF treatment and/or the date of birth of the infant to identify the relevant records in the Medical Birth Registry (Cnattingius et al., 1990Go) which gave information on the personal identification number of each infant, on maternal age, parity, smoking habits in early pregnancy and years of involuntary childlessness. The latter two variables were ascertained by interviews performed by midwives at the first antenatal visit of the pregnant woman (usually week 10–12). During the period 1984–1997, there were 1 417 166 live born infants born according to the Medical Birth Registry. A total of 9056 live born infants after IVF were identified. A total of 156 (1.7%) of them died before the end of 1998.

The hospitalizations of each infant in the Medical Birth Register were ascertained by record linkage with the Hospital Discharge Register. This register contains information on all hospitalizations in Sweden, including diagnoses, operations performed and dates of entrance to, and discharge from, the hospital. Each person is identified by their personal identification number, which enables linkage with other registers. The register covers the whole country from 1987 onwards. For the years 1984–1986 a few counties were missing, but only 69 IVF infants were born during these years and information loss is small and probably similar for IVF births and all births. The registry contains information on discharge events for 557 990 infants.

We studied the proportion of infants born after IVF who had been hospitalized at any age (up to a maximum of 14 years) and for various age windows. This analysis was made irrespective of the number of infants in the birth and also comparing singleton IVF infants with singleton controls and IVF twins with control twins. Analysis was made using the Mantel–Haenszel estimator procedure with stratification for year of birth, maternal age (5-year classes), parity (1–4+), smoking habits in early pregnancy (unknown, none, <10 and >=10 cigarettes per day), and years of involuntary childlessness (0–5+ years). A separate analysis was made on term infants (gestational duration >=37 weeks or if gestational duration was unknown, a birth weight of >=2500 g).

We then repeated the analyses restricted to a number of specified diagnoses: cerebral palsy [international classification of diseases (ICD)-9 343, ICD-10 G80], epilepsy (ICD-9 345, ICD-10 G40 or G41), mental retardation (ICD-9 317–319, ICD-10 F70-F79), attention deficit hyperactivity disorder (ICD-9 314, ICD-10 F90), non-psychotic disorders of psychological development (ICD-9 315, ICD-10 F80-F83) or autism and related conditions (ICD-9 299, ICD-10 F84), accidents (ICD-9 800–995, ICD-10 S00-T98), tumours (ICD-9 140–239, ICD-10 C00-D48), asthma (ICD-9 493, ICD-10 J45 or J46) after the age of 1 year, any infection, or any congenital malformation (Chapter 14 in ICD-9, Chapter 17 in ICD-10). Most of these diagnoses occur at an excess rate in children which are born preterm.

We calculated the mean number of days spent in hospital. If date of entrance and discharge were the same, duration was called 1 day. A difference was formed between the average number of days spent in hospital by IVF children and that spent by all other children.

By record linkage with the Swedish Cancer Register, we identified children who developed childhood cancer before the end of 1998. The odds ratio (OR) for developing cancer was determined, stratifying for year of birth, maternal age, parity and known period of involuntary childlessness. No stratification for smoking was made as it has been shown that smoking is a no risk factor for childhood cancer (Pershagen et al., 1992Go).

Results

Figure 1Go shows some maternal characteristics which affect the risk for a child to be hospitalized. There is an excess risk at low maternal age, high parity, smoking and involuntary childlessness. Each factor increases the risk of hospitalization independently of the other factors.



View larger version (19K):
[in this window]
[in a new window]
 
Figure 1. Maternal age, parity, smoking habits and length of involuntary childlessness as risk factors for the child being hospitalized, expressed as odds ratios (OR) with 95% confidence intervals (CI). Each variable is stratified for the other three variables and year of birth of the child.

 
Table IGo summarizes the OR for IVF infants being hospitalized versus all other infants. Various steps of stratification are demonstrated, but no large differences in ORs are seen. With stratification for involuntary childlessness, IVF children were compared with children born of women with fertility problems who either were treated in some other way or became pregnant spontaneously. The total OR does not change [1.84, 95% confidence interval (CI) 1.72–1.96], whilst the OR for singletons increases somewhat (OR = 1.65, 95% CI 1.53–1.78) and the OR for twins decreases slightly (OR = 1.12, 95% CI 0.95–1.33). For term infants, the OR is then no longer increased (OR = 0.95, 95% CI 0.86–1.05).


View this table:
[in this window]
[in a new window]
 
Table I. Odds ratio with 95% confidence intervals (CI) for being hospitalized among IVF infants compared with non-IVF infants. Stratification for year of birth. Further stratification as specified. Singletons compared with singletons, twins with twins. n refers to number of children in hospital care
 
Figure 2Go shows the OR for hospitalization according to the age of the child at hospitalization. The highest OR is seen during the first week of life but an increased OR of 1.2–1.3 is observed up to the age of 6 years.



View larger version (11K):
[in this window]
[in a new window]
 
Figure 2. Odds ratio (OR) with 95% confidence interval (CI) for being hospitalized among IVF children versus non-IVF children according to age at hospitalization. Stratification for maternal age, parity, smoking habits and length of involuntary childlessness.

 
Table IIGo presents ORs for specific causes of hospitalization, irrespective of plurality. As can be seen, ORs are increased for all conditions except mental retardation (based on only five cases), developmental disturbance and accidents, which were not significantly affected. An excess risk for being hospitalized for tumours (based on 43 children) made us search the Swedish Cancer Registry for childhood cancers. No increased risk for childhood cancer could be demonstrated. Only 11 cases were found (three acute lymphoblastic leukaemia, two histiocytosis, two sarcomas, two CNS neoplasms, one retinal neoplasm and one hepatic carcinoma). The expected number of cancers was 12.5 after stratification for year of birth, maternal age, parity and length of involuntary childlessness. No increased risk for childhood cancer could thus be demonstrated [relative risk = 0.88, 95% CI 0.44–1.58].


View this table:
[in this window]
[in a new window]
 
Table II. Odds ratio with 95% confidence interval (CI) for being hospitalized with specific diagnoses, IVF children versus non-IVF children. Stratification for year of birth, maternal age, parity and smoking
 
We scrutinized the records of 32 children who had been in hospital care because of a tumour diagnosis, but who did not appear in the Cancer Registry. There were two children with Letterer–Siwe's syndrome, which means four cases of histiocytosis including the two listed above from the Cancer Registry. The expected number of infants with any diagnosis of histiocytosis, stratified for year of birth, was 0.35. The 95% CI, based on the Poisson distribution of 4, is 1.1–10.2. The remaining children had benign conditions.

Table IIIGo shows the average number of days spent in hospitals (calculated for all children born) in IVF and non-IVF children. It can be seen that IVF children have, on average, spent 6 more days in hospital than non-IVF children. When singleton IVF children are compared with singleton non-IVF children, an excess period of hospitalization is still seen amounting to 3 days. A similar comparison between IVF twins and non-IVF twins shows no excess hospitalization, but non-IVF twins were hospitalized 2–3 times longer than non-IVF singletons.


View this table:
[in this window]
[in a new window]
 
Table III. Average number of days spent in hospital (calculated on all children) among IVF and non-IVF children according to year of birth, 1987–1997. All children, singletons and twins
 
For children born in 1991–1993 (and thus on average followed for 6 years), the average length of hospital stay in IVF singletons is 5.6 days and for IVF twins 13.0 days, a difference of 7.4 days. The average length of hospital stay for IVF children irrespective of twinning is 10.7 days and for non-IVF children it is 3.5 days, a difference of 7.2 days. For singleton children, the corresponding times are 5.6 and 3.4 respectively, a difference of only 2.2 days.

Among infants born in 1991–1993 (and thus on average 6.5 years old at the end of follow-up), IVF children spent an average of 10.7 days in hospital, while non-IVF children spent an average of 3.5 days, a difference of 7.2 days. In 1996, ~1500 IVF children were born in Sweden. Their excess hospital care before the age of 6 years can thus be estimated to be ~10 800 days, with an average cost of 5000 SEK (~US$500) per hospital day (Statistisk Årsbok för Landsting, 1999); this means an excess cost of 54x106 SEK.

Discussion

This study confirms and extends previously published observations that the main increased utilization of health care by IVF children occurs during the first period of life (Leslie et al., 1998Go), even though we found an increased use up to the age of 6 years. The increase is mainly due to the occurrence of twins, but IVF singletons are also more often hospitalized than non-IVF singletons, perhaps partly reflecting the increased frequency of preterm births. When the analysis is restricted to term infants, there is still an apparent increase, but this disappears when the length of involuntary childlessness (when known) is taken into consideration. There may be a parental factor involved—parents of children who were born after a long period of involuntary childlessness may be more concerned over child morbidity and may seek medical advice more easily than other parents.

When the risk for hospitalization is compared between IVF and non-IVF children, the excess risk is higher when singletons rather than twins are studied. The latter excess risk is probably lower because one is comparing IVF twins, which are mainly dizygotic, with non-IVF twins, which in a large proportion are monozygotic (and with a higher disease risk).

The excess risk of hospitalization among IVF children was found for many different conditions. The highest OR was found for congenital malformations followed by cerebral palsy, but there are also statistically significantly increased ORs for conditions like asthma and infections. For such conditions, the possibility of an effect of increased parental concern must be considered, but prematurity may play a direct role. An increased OR for tumours (based on 43 children) was not verified when the Cancer Register was used (only 11 children were identified and no increased risk). The possibility of an increased cancer risk in IVF children has been discussed based on case reports, but has never been proved. A study of 332 IVF children found no cancer case, but the power to detect a risk increase is low (Lerner-Geva et al., 2000Go). The overuse of hospital care for benign tumours may again be an expression of parental concern. There are, however, four children with histiocytosis against the expected number of 0.35, but this finding could be due to mass significance. Support or dismissal from other studies is needed.

The figures on the number of days of hospitalization can be used for estimating the `extra' cost of paediatric care of IVF children and notably of those born as twins. Most interesting is the impact of twinning after IVF, because this is to a large extent avoidable by the transfer of only one embryo. An IVF twin born in 1991–1993 used as an average 13 days of hospital care against 5.6 days for a singleton IVF child, a difference of 7.4 days, corresponding to ~37 000 SEK. To this should be added the costs outside the hospital care cost, specifically associated with cerebral palsy and other severe handicaps. These costs can be compared with the extra cost associated with additional transfers because more transfers will be needed in order to achieve a pregnancy. Cost estimates and comparisons differ between different countries. In Sweden, the savings in hospital care cost would possibly pay for one or two extra transfers.

In conclusion, our study shows that IVF children have an increased use of in-patient care, not only in the neonatal period but for some years afterwards. By reducing the rate of twin births by using one-embryo transfers a substantial saving in hospital cost could be achieved, which could balance the extra number of embryo transfers needed in order to obtain a pregnancy.


    Notes
 
5 To whom correspondence should be addressed at: Tornblad Institute, Biskopsgatan 7, SE-223 62 Lund, Sweden. E-mail: embryol{at}embryol.lu.se Back


    References
 Top
 Abstract
 Introduction
 Material and methods
 References
 
Bergh, T., Ericson, A., Hillensjö, T., Nygren, K-G. and Wennerholm, U.-B. (1999) Deliveries and children born after IVF-treatment in Sweden 1982–1995—a retrospective cohort study. Lancet, 354, 1579–1585.[ISI][Medline]

Cnattingius, S., Ericson, A., Gunnarskog, J. and Källén, B. (1990) A quality study of a medical birth registry. Scand. J. Soc. Med., 18, 143–148.[ISI][Medline]

Gissler, M., Silverio, M.M. and Hemminki, E. (1995) In-vitro fertilization pregnancies and perinatal health in Finland 1991–1993. Hum. Reprod., 10, 1856–1861.[Abstract]

Lerner-Geva, L., Toren, A., Chetrit, A., Modan, B., Mandel, M., Rechavi, G. and Dor, J. (2000) The risk for cancer among children of women who underwent in vitro fertilization. Cancer, 88, 2845–2847.[ISI][Medline]

Leslie, G.I., Bowen, J.R., Arnold, J.D. and Saunders, D.M. (1992) In-vitro fertilisation and neonatal ventilator use in a tertiary centre. Med. J. Aust., 157, 165–167.[ISI][Medline]

Leslie, G.I., Gibson, F.L., McMahon, C., Tennant, C. and Saunders, D.M. (1998) Infants conceived using in-vitro fertilization do not overutilize health care resources after the neonatal period. Hum. Reprod., 13, 2055–2059, 1998.[Abstract]

Pershagen, G., Ericson, A. and Otterblad Olausson, P. (1992). Maternal smoking in pregnancy: does it increase the risk of childhood cancer? Int. J. Epidemiol., 21, 1–5.[Abstract]

Statistisk Årsbok för Landsting (1999) Statistical Year-book for County Councils, Table II:20, Stockholm.

Submitted on August 23, 2001; accepted on December 6, 2001.