PUBH 8500 Walden University Wk 7 Gender Age and Medical Comorbidities Responses
Stephanie Minbiole-Snider
RE: Discussion 1 - Week 7
COLLAPSE
Nikpouraghdam, et al., (2020) identified the following variables in their study examining the epidemiological characteristics of COVID-19:
Independent variables: age, gender, medical comorbidities
Dependent variable: final outcome (death or survival)
Confounders: none were identified by the authors but could include cigarette or other drug use that affects the lungs or other comorbid disease not examined by the authors such as history of stroke. Weather, political, or religious ideologies that impact decisions about medical care, and sociodemographic data could also impact viral susceptibility.
The research question was to characterize the epidemiological factors of patients with COVID-19 in Iran that impacted mortality. Multiple logistic regression was used for this study to assess those factors that had an impact on patient mortality, which is a dichotomous variable (death or survival), such as age, gender and disease comorbidities.
The main results of the study identified which variables were more likely to cause mortality in a patient. Those variables that were statistically significant included male gender (OR=1.45, 95% CI: 1.08 – 1.96), underlying disease increased mortality by 53% (OR=1.53, 95% CI: 1.04 – 2.24) and each year of advanced age resulted in odds of death increasing by 5% (OR=1.05, 95% CI: 1.04 – 1.06). The interpretation of these results stated there was a significant mortality risk to those COVID-19 patients who were older males with underlying medical comorbidities and therefore these individuals should take advanced precautionary measures.
Limitations of these study included the time period in which the disease had time to run its course. With the disease being new to the medical community in March of 2020, there had not been time to allow the disease to take its course with the full population. One theory at that point could have been that the disease impacts those most at risk (i.e., elderly with underlying medical conditions) before taking hold of younger populations. Additionally, this study focused on patients at one hospital in one country but, as a worldwide disease, identifying patterns of mortality throughout the world, as the disease had time to progress, would allow for more accurate patterns of morbidity to understand who is fully at risk (Lippold, Laske, Hogeterp, Duke, Grünhage, & Reuter, 2020). Lastly, by conducting further research as time progresses, researchers could then identify mutated viral patterns which could also impact morbidity and mortality in various forms (Sharif & Dey, 2021).
References
Lippold, J. V., Laske, J. I., Hogeterp, S. A., Duke, É., Grünhage, T., & Reuter, M. (2020). The role of personality, political attitudes and socio-demographic characteristics in explaining individual differences in fear of coronavirus: A comparison over time and across countries. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.552305
Nikpouraghdam, M., Jalali Farahani, A., Alishiri, G. H., Heydari, S., Ebrahimnia, M., Samadinia, H., … & Bagheri, M. (2020). Epidemiological characteristics of Coronavirus Disease 2019 (COVID-19) patients in Iran: A single center study. Journal of Clinical Virology, 127, 104378. https://doi.org/10.1016/j.jcv.2020.104378
Sharif, N., & Dey, S. K. (2021). Impact of population density and weather on COVID-19 pandemic and SARS-COV-2mutation frequency in Bangladesh. Epidemiology and Infection, 149. https://doi.org/10.1017/s0950268821000029
3 days ago
Enica Saffold
Multiple Logistic Regression
COLLAPSE
Multiple Logistic Regression
The association between symptoms of depression during pregnancy and low birth weight: A prospective study.
Variables
Independent Variable: (Continuous or Categorical): Neonatal outcomes of women; newborn gender, preterm birth, low birth rate, macrosomia, small gestational age, and large gestational age.
Dependent Variable: (Continuous): no antenatal depression (EPDS score < 12) or antenatal depression (EPDS > 12)
Confounders: age, pregnancy BMI, neonatal gender, degree of education, high-risk women and parity
Research Question
This study aims to investigate whether antenatal depression symptoms are a risk factor for adverse birth outcomes in Chinese populations, especially for low birth weight.
Purpose of Multiple Logistic Regression?
Multiple logistic regression can be used to find the value of the dependent variable at a particular value of the independent variables (Pacifico, 2021). The purpose of this statistical test is used to estimate the odds ratio or the 95% confidence interval for the risk of low birth weight by comparing the Edinburgh postnatal depression scale (EPDS). The risks of adverse outcomes in pregnant women with antenatal depression were determined by multivariate logistic regression and represented as odds (OR) and 95% confidence interval.
Main Results
All within their 2nd trimester, one thousand five hundred women were recruited from the antenatal clinic at the Shenzhen Nanshan Maternity & Child Healthcare Hospital in China. Information on pregnancy outcomes was collected within the database, including outcomes of interest such as low birth rate, preterm birth, and early gestational age. The prevalence of antenatal depression was 19.1%. Women aged 25 and younger were more likely to have antenatal depression (p<0.05). Li et al. found that pregnant women with symptoms of depression had a higher LBW rate than non-depressed women, even though other adverse birth outcomes did not differ among the two groups. Offspring born women with antenatal depression were more likely to be low birth weight than offspring born women without antenatal depression (OR= 2.39, Cl:1.17-4.89)
Interpretation
Women with higher EPDS scores (antenatal depression) were more likely to have low birth weight infants. After adjusting pre-pregnancy BMI, neonate gender, maternal age, degree of education, high-risk women, and parity, this correlation. Xi et al. state that there have been inconclusive results between antenatal depression and low birth weight in previous studies. Again due to the disparity in different sample sizes, adjustments of the confounders, and depression symptoms measurements. Although adverse obstetrical outcomes can have long-term negative significant impacts, it can be helpful to screen depression at the first obstetric visit that predicts the risk of adverse pregnancy outcomes. Negger et al. (2006) found that women who are depressed during pregnancy also found a positive correlation with increased risk of giving birth to low birth weight (OR=1.4, CI: 1.1-1.7).
Limitations
There were a couple of limitations that were discussed in this study. The first was that depression was only analyzed in the second trimester. Therefore researchers failed to observe the trend emotions throughout the pregnancy. As a result, women whose symptoms occurred in the third trimester might be misclassified into the non-depression group. As a result, the effect of antenatal depression on low birth weight can be underestimated.
References
Li, X., Gao, R., Dai, X., Liu, H., Zhang, J., Liu, X., Si, D., Deng, T., & Xia, W. (2020). The association between symptoms of depression during pregnancy and low birth weight: A prospective study. BMC Pregnancy and Childbirth, 20(1), 1-7.
Neggers, Y., Goldenberg, R., Cliver, S., & Hauth, J. (2006). The relationship between psychosocial profile, health practices, and pregnancy outcomes. Acta Obstetricia et Gynecologica Scandinavica, 85(3), 277-285.
Pacifico, A. (2021). Robust open Bayesian analysis: Overfitting, model uncertainty, and endogeneity issues in multiple regression models. Econometric Reviews, 40(2), 148-176. doi:10.1080/07474938.2020.1770996
Be sure that the variable view in SPSS has the correct information on the 2 new variables. (10 Points)
Simple Binary Logistic Regression (30 Points)
Use Hypertension as the dependent variable and Chole_Cat as the independent variable in the first model. Report the Odds Ratio and significance of the Odds Ratio for the relationship between the dependent and independent variables. (10 Points)
Use Hypertension as the dependent variable and Serum Cholesterol (the original variable) as the independent variable in the second model. Report the Odds Ratio and significance of the Odds Ratio for the relationship between the dependent and independent variables. (10 Points)
How does the level of measurement for the independent variable affect the outcome (include the OR and its significance in your response)? How does the level of measurement of the independent variable change your interpretation of the Odds Ratio? (10 Points)
Multivariate Logistic Regression (50 Points)
Run a multivariate binary logistic regression model using SPSS and Hypertension as the dependent variable, Chole_Cat, Age_Cat, Obese, and Sex as the Covariates. Include the output in your submission. (10 Points)
Identify the Odds Ratio and the significance of the Odds Ratio for each of the covariates. How has the relationship between Chole_Cat and Hypertension changed with the addition of the other variables (compare to the output from # 2a)? (15 Points)
Test the assumption that the model fits the data using the Hosmer-Lemeshow Goodness of Fit test. Interpret the Chi Square statistic given in the output of this test and state what it means in terms of the assumptions needed to use logistic regression with this data. (10 Points)
Rerun the logistic regression model from #3a and use the save function to create the following new variables: Predicted Probabilities, Deviance Residuals, and Cook’s Distance. Evaluate the model using these saved variables and the following Scatter Plots. (15 Points)
Create a Scatter Plot of the Deviance Residuals (DEV) and the variable ID: Are there any outliers? What does this mean when evaluating your model?
Create a Scatter Plot of Cook’s Distance (COO) and the variable ID: Are there any influential cases? What does this mean when evaluating your model?
Create a Scatter Plot of Deviance (DEV) and the Predicted Probabilities (PRE). Discuss whether anything in this scatterplot could cause you some concern in terms of your model.