Identifying women at risk for polycystic ovary syndrome using a mobile health application
Rodriguez, Erika Marie
MetadataShow full item record
BACKGROUND: Polycystic Ovary Syndrome (PCOS) is an endocrine disrupting disorder affecting at least 10 percent of reproductive-aged women. In many cases, women develop comorbidities such as diabetes, cardiovascular disease, and other metabolic disorders. In North America and Europe, it takes several years and multiple doctors for women to receive a diagnosis of PCOS. This results in lost time for risk-reducing interventions. Menstrual tracking applications are one potential tool to alert women of their risk for PCOS and prompt them to seek further evaluation from a medical professional. OBJECTIVE: The objective was to develop the Irregular Cycles Feature (ICF) on the mobile phone application Clue®, which generates a probability of a user’s risk for PCOS. The secondary aim was to assess the accuracy of the tool by testing the feature on virtual test subjects. METHODS: A literature review was conducted to generate a list of signs and symptoms of PCOS. Probabilities were assigned to each variable and built into a Bayesian Network. The Irregular Cycles Feature, an adaptive questionnaire, was then developed in order to detect high-risk PCOS patients. The ICF detected at risk Clue® users through self-reported menstrual cycles and answers to medical history questions. Upon completion of the questionnaire, a Result Screen is displayed to the user. The Screen is a summary of the individual’s probability of having PCOS. For each eligible user, a Doctor’s Report is also generated. This is a screen containing information regarding menstrual irregularities and a brief medical history to be used by a medical professional in order to make a final diagnosis. Both the Result Screen and Doctor’s Report disclose information about PCOS and detailed explanations for consulting a medical provider. A brief statistical validation was then performed to compare the output of the network to predictions made by a physician-scientist using a correlation coefficient, a p-value, and a Pearson’s coefficient. RESULTS: The Irregular Cycles Feature successfully predicts probability of PCOS based on eight test cases. The correlation between the network’s calculation and the assessment made by a board-certified reproductive endocrinology/infertility physician-scientist was 0.82, with a p-value of less than 0.05. The Pearson’s coefficient was calculated to be 0.69. These values indicate that the ICF made statistically significant predictions when compared to the physician-scientist. CONCLUSIONS: The ICF provides consumer-friendly ways to improve interactions between medical providers and patients. The tool can be adapted to capture other causes of menstrual irregularities and can serve as an important mechanism for drawing attention to potentially hazardous health problems. Further validation studies will be conducted to confirm the utility of the ICF with Clue® users, particularly amongst those who receive an official diagnosis from a medical professional.