Retrospective Research and Its Use in BPD

Doctor getting file in medical records room
Medical records can be used in retrospective research. Medioimages/Photodisc/Getty Images

One way to obtain scientific data on medical conditions like borderline personality disorder is through retrospective research — when scientists look backward to form a conclusion. 

Let's gain a better understanding of retrospective research through two examples of retrospective studies in borderline personality disorder.

What Is Retrospective Research?

Retrospective research is a method in which the factors related to the development of a particular outcome — like an illness or disorder — are studied after the outcome has already occurred.

This means that data is studied after it was collected for a reason other than for research. This data may include a whole host of sources including:

  • Doctor and nursing notes
  • Emergency room reports
  • Hospital records, like admission and discharge paperwork
  • Lab or imaging test results

Data can also be analyzed from a person's memory or recall of past events — like memories of trauma or abuse during childhood

Example of Retrospective Research in Borderline Personality Disorder

An example of retrospective research in borderline personality disorder (BPD) is a 2007 study in CNS Spectrums. In this study, the charts of 13 females with borderline personality disorder treated with an anti-seizure medication called Lamictal (lamotrigine) from the year 2003-2004 were reviewed. These patients all suffered from affective instability — or intense mood shifts — due to their BPD. The review of the charts showed that for most of the women, lamotrigine was effective in treating their mood shifts.

Another example is an older study in the American Journal of Psychiatry which interviewed both people with and without borderline personality disorder about major childhood trauma. Over 80 percent of the people with borderline personality provided a history of childhood trauma, including physical abuse, sexual abuse, and domestic violence. This was significantly more than people without borderline personality disorder — suggesting that childhood trauma is a potential trigger for the development of BPD. 


One drawback of this method — mostly if the data source is a person's memory — is something called recall bias. That is, the information that is recalled by participants may be biased by their current condition. In the above example, it may be that participants with BPD, who experience very intense emotions, are more likely to interpret past events as traumatic than people without BPD.

In addition, your research data is reliant on someone else's notes or data collection — which may be incomplete, scattered, and not always contain the information the researcher needs. 

Selection bias may be another drawback of retrospective studies. Selection bias means that the study population has already been chosen and is not randomized in a retrospective trial. For instance, in the above 2007 study in CNS Spectrums, there was no randomization of which women with BPD received Lamictal (lamotrigine) and which women did not. These biases can diminish the validity or accuracy of the study's conclusion.

What Does This Mean for Me?

Retrospective research can provide key information about a variety of health conditions. That being said, they may be more prone to certain biases or limitations that have to be taken into consideration when interpreting findings. If you are intrigued by a particular retrospective study, speak with your doctor to see how it applies to your health care. 

Was this page helpful?
Article Sources
  • Herman JL, Perry JC & van der Kolk BA. Childhood trauma in borderline personality disorder. Am J Psychiatry. 1989 Apr;146(4):490-5.
  • Pannucci CJ & Wilkins EG. Identifying and Avoiding Bias in Research. Plast Reconstr Surg. 2010 Aug;126(2):619-25.
  • Weinstein W & Jamison KL. Retrospective case review of lamotrigine use for effective instability of borderline personality disorder. CNS Spect. 2007 Mar;12(3):207-10.