Highlights of the Canadian Perinatal Health Report 2003

In January 2004, the second “Canadian Perinatal Health Report” (2003) (1) was released by the Canadian Perinatal Surveillance System (CPSS). The CPSS collaborates with Statistics Canada, the Canadian Institute for Health Information, provincial and territorial governments, health professional organizations, advocacy groups and university-based researchers in the National Health Surveillance. This report is one of a series of reports which includes “Perinatal Health Indicators for Canada: A Resource Manual” (2000) (2), “The first Canadian Perinatal Health Report” (2000) (3), “Congenital Anomalies in Canada” (2002) (4) and the “Report on Maternal Deaths and Severe Illness” scheduled for release later in 2004. These reports fulfill part of the CPSS mandate of data collection, expert analysis and interpretation and communication of information for action.

The Perinatal Health Report focuses on the current status of 27 specific indicators of the health of mothers and infants, along with temporal trends and interprovincial/territorial comparisons. Data related to the indicators is from routinely collected vital statistics, hospital discharge databases and the National Longitudinal Survey of Children and Youth (5). It includes an appendix of vital statistics data from Ontario, which has been excluded from previous reports because of data quality issues. The report also includes, for the first time, vital statistics data from Nunavut.

DETERMINANTS OF MATERNAL, FETAL AND INFANT HEALTH

Perinatal health indicators that describe determinants of maternal, fetal and infant health are grouped into two categories: behaviours and practices, and health services.

BEHAVIOURS AND PRACTICES

Data on maternal smoking, alcohol consumption, breast-feeding and low maternal education levels were obtained from the National Longitudinal Survey on Children and Youth (5).

Between 1994/1995 and 1998/1999, maternal smoking rates decreased from 23.5% to 19.4%. Younger mothers were most likely to report smoking during pregnancy and there were somewhat higher rates in the Atlantic provinces, Quebec and the Prairie provinces, and lower rates in Ontario and British Columbia. There was a modest decline in the rate of pregnant women who reported alcohol consumption during pregnancy, from 17.4% in 1994/1995 to 14.6% in 1998/1999. Older mothers were more likely to report alcohol consumption than younger mothers.

Breastfeeding rates increased from 75.1% in 1994/1995 to 81.9% in 1998/1999, and in 1998/1999, 63.0% of Canadian mothers breastfed for three months or more. Breastfeeding rates and duration were higher in older mothers and for the more recent years, the rates ranged from a low of 64.5% in the Atlantic provinces to a high of 95.2% in British Columbia. Previous research indicated that low maternal education levels are related to poor perinatal outcomes. In 1994/1995, 17.2% of mothers had not completed high school compared with 13.4% in 1998/1999. Children with better educated mothers were less likely to be exposed to tobacco, but more likely to be exposed to alcohol prenatally. Breastfeeding initiation rates and duration were highest in more educated women.

Data on rates of birth at the extremes of maternal age were obtained from each province’s Vital Statistics Agency. Birth rates among older teenagers, those at 18 to 19 years of age, have decreased but there were smaller declines in women who are 15 to 17 years and 14 years or younger. There were striking regional differences, with the lowest rates in Quebec and British Columbia and highest rates in Nunavut. Increased rate of live births to older mothers comprised one of the striking results included in the 2003 report. Since 1991, the proportion of live births to teenage mothers has declined in all three age groups.

The rates of live births to older mothers has been steadily increasing over the past decade in Canada. While this may be associated with adverse outcomes for both mother and infant, some evidence suggests that older women with prudent health behaviours who receive good quality obstetrics care are not at increased risk of complications. In Canada, the live birth rate among women 35 to 39 years old increased from 25.6 to 30.6 per 1000 women, and in women who were 40 years or more it increased from 3.5 to 5.1 live births per 1000 women. In 2000, rates of live births to older mothers were lowest in the Yukon for women 35 to 39 years of age and highest in Nunavut for women 30 to 34 years of age.

HEALTH SERVICES

Data describing health services were obtained through the Canadian Institute of Health Information, Discharge Abstract Database (DAD), Quebec’s Système de maintenance et d’exploitation des données pour l’étude de la clientèle hospitalière, and Nova Scotia and Manitoba hospitalization databases.

Rates of medical labour induction increased from 12.9% to 19.7% between 1991/1992 and 2000/2001 but there is some evidence from a recent medical record reabstraction study that these data may not be reliably captured. Labour induction rates in Western Canada appear to be relatively low in comparison to the rest of the country. In spite of efforts to reduce them, cesarean section rates that had decreased from 18.2% of deliveries in 1991/1992 to 17.5% in 1994/1995, increased to 21.2 % in 2000/2001. The increase in primary cesarean sections was larger than the increase in repeat cesarean sections, and it was mainly in women age 25 years and older, so it was associated with an increasing percentage of first births to women in older age groups. There are substantial regional variations in rates in Canadian provinces and territories, with a low of 8.1% in Nunavut and a high of 25.8% in New Brunswick.

Rates of operative vaginal deliveries, those assisted by means of vacuum extraction or forceps, remained stable overall at 16.3% of hospital vaginal deliveries. However, the proportion due to vacuum extraction has increased while the proportion due to forceps has reciprocally decreased. A similar pattern was noted for rates of perineal trauma, defined as either episiotomy or a perineal laceration. Episiotomy rates fell dramatically, from 49.1% in 1991/1992 to 23.8% in 2000/2001, while first and second degree lacerations increased from 33.0% to 49.7% in the same time period. Third and fourth degree laceration rates were unchanged over the past decade.

The proportion of mothers who stayed in hospital for less than two days following vaginal birth increased dramatically from 3.7% in 1991/1992 to 21.5% in 1998/1999, and subsequently, it has decreased to 19.8% in 2000/2001. The trend for a hospital stay of less than four days among mothers with cesarean section births differed in that it has progressively increased from 2.7% in 1991/1992 to 35.7% in 2000/2001. In 2000/2001, women delivering in Nunavut and Alberta were discharged from hospital after childbirth sooner than women delivering in other provinces and territories. Rates of early neonatal discharge from hospital after birth have been of major interest to physicians. Data were presented for two birth weight categories, greater than 2500 g and 1000 g to 2499 g. Gestational age data has not been recorded on the DAD so possible analyses and interpretation of data in this area are limited. Major differences were in the heavier birth weight category where rates of early discharge increased markedly from 4.3% in 1991/1992 to 24.5% in 1998/1999 and then decreased slightly to 22.3% in 2000/2001. Currently, the mean length of stay of 1000 g to 2499 g babies is 8.5 days (SD 6.8) and 2.4 days (SD 1.7) for babies weighing more than 2500 g. The rate of neonatal hospital readmission after discharge following birth is of particular interest to paediatricians. Overall, rates have increased from 1.9% hospital live births in 1991/1992 to 3.2% of hospital live births in 1999/2000. The leading identified causes were neonatal jaundice, feeding problems, sepsis, dehydration and inadequate weight gain.

MATERNAL, FETAL AND INFANT HEALTH OUTCOMES

Maternal health outcomes such as maternal mortality, induced abortions, ectopic pregnancy and maternal readmission data were included in the report but will not be discussed in detail in this commentary. Information on fetal and infant health outcomes was based on Vital Statistics data, including the live birth, stillbirth and death files, as well the Canadian Institute of Health Information DAD, the Canadian Congenital Anomalies Surveillance System, Quebec’s Système de maintenance et d’exploitation des données pour l’étude de la clientèle hospitalière, and Nova Scotia and Manitoba hospitalization databases.

Ontario data were not included in most analyses using Vital Statistics information because of concerns regarding data quality in the early to mid 1990s; some of these data were analyzed and presented separately as an appendix in the Perinatal Health Report.

The preterm birth rate, defined as the number of live births at less than 37 completed weeks or less than 259 days of gestation, as a proportion of all live births, has been progressively increasing in recent years. From 6.6 per 100 live births in 1991, it has increased to 7.6% per 100 live births in 2000. Rates of preterm birth are much higher in multiple births than in singletons; more than half of all twin deliveries and almost all triplet births were preterm. Nevertheless, 80% of preterm births were single births in 2000. Regional variations were generally not striking. The multiple birth rate has continued to increase steadily over time from 2.1% of total births in 1991 to 2.7% of total births in 2000. This trend is similar across most provinces and territories, and is linked to increased births to older mothers and increased use of fertility treatments and assisted conception.

The post-term birth rate was defined as the number of births that occur at a gestation of 42 or more completed weeks or 294 days or more as a proportion of live births. It has decreased dramatically from 4.4% in 1990 to 1.2% in 2000. Some of this change may be artefactual since more frequent use of ultrasound dating of pregnancies tends to shift the gestational age distribution slightly to earlier gestational ages, but a good portion is due to changes in health care practices. Rates of post-term birth varied among Canadian provinces and territories. Fetal growth restriction is associated with increased fetal and infant morbidity and mortality. Rates of small for gestational age births, defined as birthweight below the 10th percentile using the Canadian reference for birth weight and gestational age (6), decreased from 10.7% in 1991 to 7.9% in 2000. A number of factors are associated with this decrease, including more frequent use of ultrasound-assisted dating, a decrease in the maternal smoking rate, and possibly, changes in various sociodemo-graphic factors. Large for gestational age births, those above the 90th percentile, have increased, possibly in relation to the same types of factors. The large for gestational age rate increased from 9.5% in 1991 to 12.0% in 2000.

The fetal mortality rate, defined as the number of stillbirths per 1000 total births, has fluctuated over the past 10 years, but overall changes are small. This was calculated both including and excluding births of infants weighing less than 500 g. There are concerns regarding the completeness of recording of live births and stillbirths at the lower birth weight ranges. Infant mortality rates, defined as the number of deaths of infants in the first year of life per 1000 live births, is a key health measure in all societies. Overall, the rates have decreased from 6.5 per 1000 live births in 1991 to 5.1 per 1000 live births in 2000. Both the neonatal and postneonatal components have decreased, particularly the latter, which decreased from 2.5 per 1000 live births in 1991 to 1.7 per 1000 live births in 2000. Congenital anomalies, immaturity and sudden infant death syndrome remain the leading causes of infant mortality. Overall infant mortality and postneonatal death rates varied among provinces and territories but neonatal death rates were more uniform.

Neonatal morbidity data are notoriously difficult to obtain and interpret, but in this report, two examples of severe morbid conditions, respiratory distress syndrome and sepsis, were analyzed. Respiratory distress syndrome rates decreased slightly in the early 1990s, but have since been stable. The 2000/2001 rate was 11.6 per 1000 live births. The rate of sepsis, on the other hand, has increased progressively from 16.9 per 1000 live births in 1991/1992 to 24.8 per 1000 live births in 2000/2001. Sepsis rates were abstracted from the hospital discharge databases, and it is not possible at this time to provide more detailed analysis regarding suspected versus confirmed sepsis or regarding the exact type of infection.

The prevalence of congenital anomalies has fluctuated over the past decade. Rates of some anomalies such as Down syndrome have not changed (14.3 per 10,000 total births in 1991 and 14.2 per 10,000 total births in 1999), but others such as neural tube defects have changed dramatically, from 10.0 per 10,000 total births in 1991 to 5.6 per 10,000 total births in 1999. Several factors may have contributed to this decline including increased use of vitamin supplements and prenatal diagnosis with termination of affected pregnancies. For the years 1997 to 1999 combined, Prince Edward Island, Yukon and Northwest territories had the lowest prevalence of neural tube defects with no reported cases, and Newfoundland had the highest with a rate of 9.7 per 10,000 total births.

To review the report in full, please visit <www.hc-sc.gc.ca/pphb-dgspsp/publicat/cphr-rspc03/index.html>.