Growing Up in New Zealand study director Susan Morton is tracking the development of almost 7000 children born seven years ago in the Upper North Island. Like the world-famous Dunedin Study, this research will show how childhood factors influence outcomes later in life.

1 The babies in this study are now reaching 7 years old. What have been your most surprising findings so far?

We knew New Zealand's population had changed since the Dunedin Study began in 1972 - that's one of the reasons we set this study up - but we weren't quite prepared for the diversity we found. One in three of our children have a parent born outside New Zealand.

About 90 different languages are spoken in their homes and 10 per cent of our interviews need an interpreter. The traditional household has changed from a generation ago. One in four of our children now live in extended family situations. Only 5 per cent live in a house with a sole parent - more commonly they live with other adults who are family, friends or flatmates. Almost half of our children live in rental properties, the majority private rentals, compared with one third in the past, and those who are renting generally stay renting. Half have moved house at least once by the age of 2 and many have moved often. Support services are often delivered geographically so perhaps we need to rethink that for this generation.

2 The Dunedin Study has been hugely influential for researchers and policy-makers worldwide. What have you done differently?


The Dunedin Study started from birth. We've started in pregnancy, which we now know is such an important foundation for child development. We've also included fathers from the beginning. Our cohort of 7000 is seven times larger than Dunedin's. Because we engage with 16 government agencies, we have a very strong focus on helping shape public policy. As soon as we find something that could make a difference to the wellbeing of our children we try to get the story out - like when we found that one in eight mums experience symptoms of depression in late pregnancy.

3 How many members of your original cohort are still with the study seven years on?

We've retained contact with 98 per cent of the families and have complete information from 92 per cent which is amazing and a credit to our interviewers. Overseas longitudinal studies often lost 20 to 25 per cent of their cohort by the time the children were 2. A key factor has been recognising that families often have very hectic lives and we need to make it as easy as possible for them to engage. We see them in their own homes on any day of the week, at any time of the day, with their preference in interviewer.

4 How did you ensure that your cohort reflected the current generation of Kiwi kids?

Recruiting pregnant mums was a challenge because there is no register of currently pregnant women. So we recruited our cohort based on the characteristics of the last five years of births. We tried to reach all eligible mothers through midwives, antenatal clinics, Plunket, media campaigns, community events and spaces. When we were able to check back with the birth records from the year our cohort were born we found we had close alignment especially in ethnic and social diversity.

5 Was it hard to include families who tend to avoid engaging with authorities?

We've actually got a good range of high-risk families. Often these are silent groups whose voices are not heard at the policy table. The main reason I think they talk to us is because the study is anonymous. We've promised we won't pass their individual information on to other agencies like Winz or IRD. So they share very sensitive information with us such as their drug use, jail time, relationship problems, sources of income - legal and illegal - and even residency status. Also they know their stories can help make things better.

6 What do you do when you can see families need help?


That's a huge ethical challenge. Technically we can't step in because we're an observational study and we're not in a position to provide direct support. But we don't want to do nothing because that doesn't feel respectful of what people are sharing. If there was an emergency you'd do something, but that's extremely rare. In all our interviews we provide a list of support services to contact for help. Where risky situations like violence or depression are exposed we'll check in with families to make sure they can connect with appropriate help.

7 But wouldn't that intervention skew your results?

Possibly. Most longitudinal studies have looked at the "Hawthorne Effect" - if the fact you're following families can actually alter outcomes - but most found any effect is negligible. I think that in the rare occasions where our presence has resulted in a family changing behaviour, that's a small price to pay for such a valuable collection of stories.

8 What do you know about our most vulnerable children?

We found that almost 10 per cent of our cohort is being exposed to real hardship right from before birth. These children experience a cluster of four out of 12 markers of disadvantage including poverty, rental accommodation, having a sole parent, a mother who hasn't completed her education or a mother who smokes or drinks during pregnancy.

9 How do you ensure you're gathering the right information?

We try to collect information that will be useful to as many people as possible. Part of the challenge is anticipating what might be valuable in 10 years' time. We did ask about the impact of unpredicted events like the Global Financial Crisis and the Christchurch earthquakes.

10 Can you point to policy changes that have come about as a result of your findings?

We've been reliably informed that evidence from our families given to the select committee on paid parental leave was incredibly important in achieving change. Information we've collected on the ability of families to keep rental homes safe with things like working smoke alarms and fenced driveways will feed into new policies on the responsibilities of landlords, to be rolled out over the next couple of years.

11 Why does it take years between when information is collected to when it's released?

It's called the "longitudinal lag". It takes us about two years from the time we start collecting to when it's released, which is about as quick as you can safely do it. Our children were born over a year from April 25, 2009 to March 25, 2010 so it takes a year to visit all their families. They provide thousands of responses which become millions of data items. It takes another year to "clean" those answers and combine them with direct observations to create information that accurately reflects their stories.

12 How did you become the director of this study?

I trained in paediatrics as a mature student and when my children were young moved to London to do a PhD based on the UK longitudinal studies. When I heard New Zealand might set up a study, I returned home really enthused. The lead agency, the Ministry of Social Development, called for proposals. Our University of Auckland team was one of three to submit and eventually find success.

Also, Why Am I? series, TV One, Tuesdays at 9.30pm.