DEFINITIONS

"Longitudinal studies provide data about the same individual at different points in time allowing the researcher to track change at the individual level.
There are a number of different types of longitudinal studies, including:

• Individual level panel surveys, where samples of individuals are tracked and interviewed.
• Household panel surveys, where individuals are followed within the context of the households where they live, and information is normally collected about the whole household at each wave.
• Cohort studies, where samples from a particular age range are followed to explore their different trajectories as they age.
• Record linkage studies, administrative or census data are linked across time."

The UK Longitudinal Studies Centre: What are longitudinal studies? - <www.iser.essex.ac.uk/ulsc/longitudinal-faqs>

"Longitudinal studies are data sources that contain observations of the same research units over a period of time. Such data can be collected either through surveys, or through linkage of administrative data."
Census.ac.uk: Longitudinal Studies - <http://census.ac.uk/guides/Longitudinal.aspx>

ADVANTAGES/DISADVANTAGES

Longitudinal studies are unique in their ability to provide useful data about individual change. This type of study also allows for flexibility, meaning the focus of the study can be shifted whilst data is being collected.

One of the biggest disadvantages to using longitudinal studies is the time factor. This type of study is time consuming, which affects cohort retention and the ability to maintain a committed research team. It also means that these types of studies are costly and the financial burden is one of the biggest challenges facing longitudinal studies. 

RESOURCES

Australian Institute of Health and Welfare (2004). Longitudinal studies of ageing: Implications for future studies, AIHW cat. no. AGE 42, 206.

Gregory A, Armstong R, Grassi T, Gaut B, Weyden M. (2008). On our selection: Australian longitudinal research studies, Medical Journal of Australia 189(11/12):650-657.

Van Weel C. (2005). Longitudinal Research and data collection in primary care, Annals of Family Medicine Vol 3(1):546-51.

Najman JM, Bor W, O'Callaghan M, Williams GM, Aird R, Shuttlewood G. (2005). Cohort profile: The Mater-University of Queensland study of pregnancy (MUSP), International Journal of Epidemiology 34:992-997.

The General Practice Research Database (GPRD)
This is the world's largest computerised database of anonymised longitudinal medical records from primary care that is linked with other healthcare data. To date, it contains data from around 5 million patients from about 625 primary care practices across the UK.

The General Practice Research Network (GPRN)
This is an Australian longitudinal database involving over 1 000 GPs who supply de-identified prescribing data for more than 3 million patients.

USING LONGITUDINAL DATA

Many studies welcome and encourage the use of their datasets as a resource for further studies.

Researchers can apply to access the data through the relevant pathways set out by the organisations that hold the data. To find out the criteria for access to specific data please view the studies' website and contact the appropriate person.

ANALYSING THE DATA

There are some great statistical software packages available that can help you analyse the data sourced from longitudinal studies. These include multi-level analysis products like MLwiN and HLM as well as some of the more generic software brands like SPSS, SAS and Stata. For more information about these software products please view the links provided.

If unsure about how to proceed with data analysis, you may find the following references useful:

• Newsom JT, Jones RN, Hofer SM, (Eds). (2011). Longitudinal data analysis. A practical guide for researchers in aging, health, and social sciences, Routledge Academic.
• Arnau J, Balluerka N, Bono R, Gorostiaga A. (2010). General linear mixed model for analyzing longitudinal data in developmental research, Perceptual and Motor Skills, 110, 547-66.
• Holditch-Davis D. (2010). Potential pitfalls in collecting and analyzing longitudinal data from chronically ill populations, Newborn and Infant Nursing Reviews, 10, 10-18.