Music Therapy and Autism Ranking: Limited positive evidence

Status Research

There are a number of limitations to all of the research studies published to date. For example  

  • The majority of the studies had a small number of participants - between 1 and 8. One of the studies with a greater number of participants (Kim, 2008) had a dropout rate of 5 out of 15, that is, 33%.
  • In some of the studies (such as Aldridge et al, 1995) it is not clear how many, if any, of the participants were on the autism spectrum.
  • Some of the studies (such as Kaplan and Steele, 2005) were retrospective (where the study is planned once the intervention has taken place) or were not experimental in design (where the study is planned to test specified objectives).
  • Some of the studies (such as Dezfoolian et al, 2013) had no comparison control group (a group that gets a different treatment).
  • Some of the controlled studies (such as Ghasemtabar et al, 2015) were non-randomised (so the control group may have had different characteristics to the experimental group) and non-blinded (the researchers and participants knew who got which treatment).
  • Some of the studies (such as Saperston, 1973) did not establish adequate baseline measures before the intervention began (which makes it difficult to assess the significance of any reported changes).
  • Some of the studies (such as Gattino et al, 2011) did not use a structured, manualised protocol (a set of codified instructions which ensures the intervention is delivered the way it should be).
  • Some of the studies (such as Rosenblatt et al, 2011) used music therapy alongside a wide range of other interventions (which makes it difficult to assess the effectiveness or otherwise of each intervention).

 For a comprehensive list of potential flaws in research studies, please see "Why some research studies are flawed."

19 Dec 2017
Last Review
01 Sep 2016
Next Review
01 Sep 2019