Gender Concordance of Doctor and Patient During Total Body Skin Exam

Open Access
- Author:
- Kassabian, Michael
- Graduate Program:
- Industrial Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 26, 2022
- Committee Members:
- Ling Rothrock, Thesis Advisor/Co-Advisor
Scarlett Rae Miller, Committee Member
Steven Landry, Program Head/Chair - Keywords:
- Doctor-Patient Gender Interaction
Skin Exam
Gender Bias
Healthcare - Abstract:
- In a previous study out of the Penn State School of Medicine, 10 participants (5 faculty, 5 residents) were recorded performing Total Body Skin Exams (TBSE) on two healthy patients. These exams were analyzed for body segment order, duration of exam, and movements of both examiners and patients. From observations of the 20 resulting TBSEs, a standard procedure was developed that would have examiners start at the top of the patient and work their way down, examining the face, arms, torso, and legs before moving to the back and repeating the process. The idea was that the specific body segment order and short transition distances between segments would result in a more accurate, more efficient exam. While often unconscious, gender bias in healthcare can have a large impact on a patient’s quality of life. As such, it is important to determine what aspects of healthcare can be susceptible to these biases and the best way to prevent them. In this thesis, the effects of a training protocol on total body skin exams are analyzed to determine if the performance of a skin exam could be impacted by gender bias. Specifically, doctor-patient pairs were grouped based on gender agreement between the two, and exam time and accuracy were recorded. It was hypothesized that examiners untrained in this new protocol would perform better exams on those patients that had their same gender than those of another gender, but that no difference would be seen between the male and female patients of those examiners trained in the new exam protocol. To test this new protocol, 31 participants were recruited to examine two model patients. Participants were then randomly assigned to either the control group (n=15) or the experimental group (n=16). Those in the experimental group were trained on the previously developed TBSE protocol, and those in the control group were not. All participants were instructed to complete a skin exam of the whole body. Examinations were recorded with special eye-ware (Tobii Glasses) that contain cameras and eye tracking technology. Together the footage of the exam and eye gaze data show where on the patient’s body participants were looking at any given time during the exam. Participants were analyzed on how many body segments were examined and for how long to get a total exam completeness proportion, and a total exam time. The low sample size resulted in some low powered analyses. This reduces the strength of any conclusions drawn, but interesting relationships can still be gleaned from the data that could be used in further studies. From the data collected, it was determined that female examiners in the control group performed significantly longer exams on female patients than on male patients but performed significantly more complete exams on male patients than the female patients. This could indicate a slight bias towards male patients. No other demographic of examiner showed a significant difference in exam time or completeness between male and female patients. This seemed to indicate that female examiners in the trained group were more consistent between patients than those in the control group. Overall, these results show that participation in a training protocol was associated with reduced differences between male and female patients. That is to say that there was no significant difference shown between male and female patients of experimental group examiners where there was in the control group. More research needs to be done to improve the effects of a small sample size such as low power, but there are promising results so far. If more subjects are included in future studies, the conclusions made here can be verified. Consistent TSBE results for the trained group of examiners is a good thing even if the original inconsistencies did not arise from gender bias. This method of implementing protocols for simple examinations could then be applied to other areas of healthcare while more long term, proven gender bias mitigation techniques are employed. It is unclear if any actual gender bias was improved here but providing consistent results across patients benefits everyone.