Prospective Studies of Proteinuria and Dyspnea as Potential Predictors of All Cause and Chronic-Disease Mortality in a Rural Bangladesh Population

This dissertation describes the background, setting, set-up and analysis of several 11-year prospective longitudinal studies with exposures of Proteinuria or Dyspnea and the primary outcome of all-cause mortality. Cause-specific mortality was also obtained for each exposed/unexposed group to determine whether exposures are at all related to mortality outcome. These studies came out of the Health Effects of Arsenic Longitudinal Study (HEALS). The objectives of this dissertation are to: 1) assess the reproducibility of dyspnea as determined by questionnaire in evaluating for the presence or absence of dyspnea; 2) examine the association between arsenic exposure and dyspnea cross-sectionally since one of the longitudinal studies proposed, evaluating the symptom of dyspnea as a predictor of mortality, is embedded in an ongoing study evaluating the effects of chronic exposure to arsenic in well water; 3) review the worlds relevant literature on the potential for dyspnea, a symptom, to be a predictor of all-cause mortality; 4) try to determine whether dyspnea, a symptom, is a predictor of all-cause and cause-specific mortality in the developing country of Bangladesh; 5) try to determine whether simple dipstick proteinuria, is a predictor of all-cause and cause-specific mortality in rural Bangladesh. A methodologic study was done on a small subgroup of subjects to determine whether dyspnea determined by simple questionnaire was reproducible. If the presence or absence of dyspnea on questionnaire occurred by chance, then using dyspnea as the exposure variable would not be valid. The results of this study revealed that: 1) dyspnea as determined by questionnaire was reproducible ie the same response occurs when the same question on dyspnea was asked at a later time and disguised by being buried in a list of questions; 2) the reproducibility of the response was greater than 90%, independent of whether dyspnea was present or absent on the initial response. A second preliminary cross-sectional evaluation was done to determine whether the exposure variables of proteinuria or dyspnea were associated with arsenic exposure at baseline since the primary overall focus of HEALS is related to arsenic exposure. No definitive association for proteinuria and arsenic was found. However, an analysis and study found a strong dose-response relationship between arsenic well water concentration (exposure) and the presence of dyspnea, independent of smoking. A weak dose-response relationship was also found between smoking and dyspnea. Clearly, both arsenic exposure and smoking are two of a number of important variables that need to be controlled for in these prospective studies. In addition, dyspnea was found in a longitudinal study to be associated with all-cause and cause-specific mortality, diseases most related to the heart and lungs. Finally, dipstick proteinuria at the 1+ level (not trace) was found to be a predictor of all-cause and cardiovascular disease mortality in rural Bangladesh. Further discussion is on the implications of the study findings including the concepts of dipstick proteinuria in screening and dyspnea in screening and directions of future research.