The above table illustrates the bulk of information which forms the basis for my inquiry. It presents the age of women who have given birth in Ireland from 2007 to 2015. Visually it is awful, overpopulated, and likely to be immediately dismissed rather than explored. The extremes of ’15 years and under’ and ‘All ages’ are the easiest statistics to pick out. There is so much information on display, there is a risk that everything in the middle is missed or dismissed.

Pie charts representing data from 2007 (Fig2), and from 2015 (Fig3) Produced using Google Data Studio. Hover over pie chart segments for details. (Source: CSO.ie)

The source data from the Central Statistics Office website makes the data available chronologically by age, as well as by age groupings. I have used those age groupings to compile two simple pie charts. These pie charts do not illustrate detailed changes over time. Rather, they each present a picture of data for one year. Therefore, there is no obvious way to track trends or forecast in any detailed manner the needs of the maternity system in Ireland.

The above examples of data visualisation show either too much or too little detail to gain any significant insights into the data being represented. With the ‘age’ or ‘year’ dimensions being too broad or narrow the opportunity to gain a comprehensive understanding of the data is lost, and so is cogent meaning. Expanding the values of one or both of these dimensions, utilising other visualisation methods and cross referencing with other data sets, brings context and meaning thereby increasing the chances of this data being properly interpreted.

Presenting the main table of data condensed into age groupings makes it immediately more accessible at a glance. The vital statistics are all still present, yet displayed in a manner that can be interpreted without the user being visually bombarded with a whole page of figures. The space given to each figure, as well as the ability to dynamically sort through figures by selecting either an age range or year, makes this data immediately more accessible. The interactivity of the table invites exploration.

(Fig4. Produced using Google Data Studio. Source: CSO.ie)

The data from the table can be presented in a more concise format like so:

(Fig5 Static image. Source: CSO.ie)

Looking at this visual, there are steps involved in understanding what has been presented.
1) Intuitive perception
2) Interpretation, analysis, cross reference

1) Intuitive Perception:
Data explorers should always be wary of steep inclines, declines, or trends that seem to approach the base line. This visualisation at a quick glance, without proper attention to the y-axis, may lead a casual viewer to believe that Ireland’s birth rate is approaching zero.
2) Interpretation:
This graph shows an obvious decline in birth rates.
Analysis:
A decline in birth rates has far reaching effects for society as a whole. It could indicate future economic decline and an aging population. Specifically, for maternity services the above interpretation may result in an assumption that staff numbers can be reduced in maternity units.
Cross Reference:
To facilitate more detailed understanding of figures 4 and 5, other data should be cross referenced to gain better insight into the applicability of this information. I will do so below with data relating to labour interventions, specifically Caesarean-Sections. Consideration of that data will show that any reduction in staffing is not wise because there is clearly an increasing frequency of medical interventions during delivery, as well as a rising maternal age (pregnancies are routinely classed as ‘high-risk’ if the woman is over 35). This conclusion is reinforced by the consistent outcry from the INMO regarding staffing levels, and the government’s 2016 National Maternity Strategy to prop up the maternity services in Ireland. The necessity for these strategies is partially a result of the nurses and midwives recruitment freeze in 2008 despite the fact that birth rates had risen sharply in 2007, and still increased in 2008 (as shown in figure 5 above).

Fig6 Produced through Google Sheets and Google Charts. Hover over lines for data points and details. Source: data.OECD.org

Similar caution should be exercised with this visualisation as with previous visualisations. It would appear that C-Section rates are rising directly in line with the increase in women over 35 giving birth. Though the image is quite compelling, one cannot assume that there is a direct correlation between the increase of women aged 35 and over having children, and the increase of C-Sections taking place.

In 2015, 30% of births happened by C-Section and 34.5% of births were to women over 35; however, it would be grossly irresponsible to assume that all women over 35 give birth via C-Section.

Fig7 (Source: data.gov.ie) and Fig8 (Source: data.gov.ie) illustrate the ratio of labour interventions to spontaneous delivery in 2014 and 2015. Static images produced using Excel Charts.

Data that specifies all types of labour interventions utilised in delivery wards in Ireland is only available for 2014 and 2015. From this data we see that labour intervention rates and spontaneous delivery rates were maintained at 46% and 54% respectively in 2014 and 2015. From figure 6 we know that C-Section rates are increasing. Further analysis of the intervention statistics, and specifically identifying which interventions are increasing or decreasing, will further indicate exactly what type of staffing increases need to be made; for example, surgeons/anaesthetists in the case of C-Sections.

The argument could be made that the frequency of interventions is decreasing in line with birth rates (labour based interventions remain at 46%), and so there is no reason for increasing staffing levels. However, when the actual frequency or number of labour based interventions, C-Sections in particular, are displayed, this argument is negated.

Fig9 Produced using Excel Charts. Source: data.OECD.org

Despite falling birth rates, the data indicating the percentage of C-Sections, as well as that same data represented in actual numbers shows that C-Sections are:

• increasing in frequency
• increasing as a percentage of the total number of births
• increasingly being chosen as the preferred method of labour intervention.

Due to the lack of data over a number of years, it is hard to build a picture of trends in labour based interventions. Using only the 2014 and 2015 data from data.gov.ie we can see a snapshot of how the percentage and frequency of spontaneous delivery, as well as delivery interventions have changed over these two years.

Fig10 Produced using Google Sheets and Charts. Click on chart for details. Source: data.gov.ie

Fig11 Produced using Google Sheets and Charts. Click on chart for details. Source: data.gov.ie

There are only small variations in the data between 2014 and 2015, but what is definitively consistent is the decline in spontaneous delivery, and the rise in labour interventions, the most popular being the Caesarean Section, as maternal age increases. In both 2014 and 2015, the percentage of women over forty giving birth via C-Section actually surpasses the percentage of women in that same age bracket giving birth without interventions.

Medical interventions save lives, and I acknowledge that this survey of data takes a reductive view of the immensely detailed services offered by maternity units, measuring only a minute number of variables.  I set out to provide an overview of the maternity service in Ireland, and the changing landscape of women giving birth. As society changes, so must the services that the state provides. Data regarding trends in national birth rates, ages of women giving birth, and types of labour interventions utilised, can be used to predict specific staff requirements and client needs across the maternity service in Ireland.

Sources and References

Bo Jacobsson, MD, PhD, Lars Ladfors, MD, PhD, and Ian Milsom, MD, PhD, 2004. Advanced Maternal Age and Adverse Perinatal Outcome: http://www.gums.ac.ir/Upload/Modules/Contents/asset68/Advanced%20Maternal%20Age%20and%20Adverse%20Perinatal%20Outcome.pdf. (Last Accessed 16 April 2018).

Central Statistics Office, ‘VSA04: Births by Age of Mother, Sex of Child, Year and Statistic’. http://www.cso.ie/px/pxeirestat/Statire/SelectVarVal/Define.asp?Maintable=VSA04&Planguage=0. (Last Accessed on 16 April 2018).

Data.gov.ie, ‘Table 4.7b: Method of Delivery by Maternal Age, Total Births, 2014’. https://data.gov.ie/dataset/perinatal-statistics-report-2014-method-of-delivery-by-maternal-age-total-births-2014. (Last Accessed 16 April 2018).

Data.gov.ie, ‘Table 4.7b: Method of Delivery by Maternal Age, Total Births, 2015’. https://data.gov.ie/dataset/perinatal-statistics-report-2015-method-of-delivery-by-maternal-age-total-births-2015. (Last Accessed 16 April 2018).

OECD (2018), ‘Caesarean sections (indicator)’. doi: 10.1787/adc3c39f-en. https://data.oecd.org/healthcare/caesarean-sections.htm. (Last Accessed on 16 April 2018).