Nature or nurture in low back pain

Clinical research into the management of low back pain has shown that the current available treatments offer, at best, only moderate effects. Our Spinal Research Group at the University of Sydney has been one of the pioneers in the field and most of these discouraging results have been produced by high quality randomized controlled trials and systematic reviews conducted by our group. Not surprisingly we thought that we needed a change in the course of direction of the spinal management boat.

We realized that a step backwards would be necessary if we wanted to move forward with the design and implementation of interventions for back pain. Stepping back, in this scenario, and in a research group that conducts mainly clinical research, meant that we would have to re-visit some epidemiologic aspects of the condition. From these aspects we thought the most important one would be the investigation of causal factors in back pain.

The systematic review recently published in the European Journal of Pain investigated factors that were associated with, or consistent with, a cause of back pain arising from studies that used twins. The use of twins in back pain research is particularly important because this design allows for a cleaner and more precise estimate of what causes back pain because genetics can be controlled for. In a twin design approach a pair of twins is used in the analysis.  When one twin has the condition and the co-twin does not we  have the opportunity to use an analysis called twin case-control. This design circumvents the common problem in back pain studies where heterogeneous groups of back pain patients are investigated and the genetic component is disregarded.

The results of this systematic review showed that the genetic component affects 21% to 67% of back pain depending on the type of the condition. Interestingly, genetic influences are higher for more chronic and disabling back pain. Twin studies also showed that smoking and obesity are also associated with back pain even after the genetic component is controlled for.

It does appear that some types of back pain (chronic disabling) have a higher genetic causation component while others (acute or less disabling) have a higher environment component. These findings have important implications for the understanding, prevention and management of back pain. Reducing the occurrence of back pain by controlling causal component factors is a promising field but we need to identify, for different types of back pain, which ones are dominant and necessary factors for the condition to be established.

This study has been used as the framework for a current NHMRC grant proposal submitted by our group aimed at investigating physical activity as a protective or causal factor for back pain using twins registered at the Australian Twin Registry. We are expecting to use the twin design in our future studies as a step forward in the understanding of what causes back pain and how we can optimize our management of this highly prevalent and costly condition.

About Paulo Ferreira

Paulo FerreiraPaulo Ferreira is a Senior Lecturer and the leader of the Spinal Research Group at the Faculty of Health Sciences, University of Sydney, Australia.


Ferreira, P., Beckenkamp, P., Maher, C., Hopper, J., & Ferreira, M. (2013). Nature or nurture in low back pain? Results of a systematic review of studies based on twin samples Eur J Pain DOI: 10.1002/j.1532-2149.2012.00277.x


  1. I have always wondered how studies can be performed for environmental causes of back pain and the genetic component is always ignored. I look forward to seeing some of the research using twin studies. Hopefully these studies will point people away from back surgeries and more toward preventative maintenance like daily spinal multifidi strengthening. Good post, thanks.

  2. Evanthis Raftopoulos says:

    Hi Paulo, I understand that twin studies suggest a possible causal relationship between genetics and stucture/anatomy eg
    However , the pain experience is context dependent, complex , dynamic, and multidimensional. Do twin studies demonstrate a valid way of establishing a causal relationship between genetics and low back pain? Correlation does not mean causation comes to mind.

    Evanthis Raftopoulos, PT

  3. Mine is a similar question to the one above by Greg. How can we separate genetics and modeling in assessing movement patterns? Isn’t it possible for two people, even twins, to have the same scoliotic spines and yet be very different where pain, athletics and even respiration are concerned?
    Thanks so much,

  4. Hi Paulo,

    I’m interested to know to what extent you can separate genetic influences from familial environmental circumstances. It is often observed that movement characteristics are ‘modelled’ by children, and in a typical familial situation children are exposed throughout childhood to very similar environmental conditions. I’m aware of some twin studies that attempt to control for this by using twins who have grown up in disparate households.
    Thanks Greg