Validating software estimates

6854933580_2c8b688306_z

In VA studies, interviewers ask family members questions about the signs and symptoms leading up to a death that occurred in the household, and physicians or computer models can be used to classify the estimated cause of death.The advent of advanced computer techniques for solving this problem are described elsewhere [].Mexico, and much of Latin America in general, have seen a relative increase in their NCD burden in the past 20 years, with conditions such as heart disease, arthritis, and vision loss steadily increasing in terms of disability-adjusted life years [].Despite the substantial burden of NCDs around the world, it continues to be difficult to collect accurate information on their prevalence, particularly in areas that lack consistent or accessible health care.The SD study consisted of two components: data collection and model validation.The data collection portion consisted of identifying cases of different NCDs in a hospital and then conducting a questionnaire with the patient at a later date.The diagnostic criteria for a condition such as chronic obstructive pulmonary disease (COPD), for example, require medical resources such as spirometry or medical knowledge to interpret FEV1/FVC ratios and differentiate COPD from asthma based on subtle differences in clinical signs and symptoms.

Thus, given that NCDs contribute significantly to the global burden of disease, and given that the diagnosis of NCDs requires clinical expertise and medical resources, the analytic question in this study is whether self-reported signs and symptoms in a questionnaire survey can be accurately assessed by data-driven computational models in order to better measure the burden of these diseases.In part, this is due to inherent limitations in diagnosing these conditions.While information on some infectious diseases, such as HIV, malaria, and tuberculosis, can be collected through biological assays or cultures, such an equivalent does not exist for certain NCDs.Diagnosis based on self-reported signs and symptoms (“Symptomatic Diagnosis,” or SD) analyzed with computer-based algorithms may be a promising method for collecting timely and reliable information on non-communicable disease prevalence.The objective of this study was to develop and assess the performance of a symptom-based questionnaire to estimate prevalence of non-communicable diseases in low-resource areas.We assessed the performance of this instrument and analytical techniques at the individual and population levels.

You must have an account to comment. Please register or login here!