Hello,
I have data from 9 time points from ~150 hospitals. At each time point hospitals reported counts of hypothermia cases among NICU admissions. The dataset has a lot of missing data due to either hospitals not reporting for a period. For presenting this data to our stakeholders I have been asked to create control charts, specifically pcharts showing the project-level hypothermia rates across the 9 timepoints with lower and upper control limits that include all available data. I am new to control charts, but I do understand my data violates some rules. Specifically, I do not have enough timepoints for calculate proper control limits and more importantly my data is nested within hospitals. If I do not account for the nesting when estimating standard deviations I will get tiny SDs as the denominators for my rates are quite large. Can someone please advise me on how to more appropriately calculate my SD that will be used to calculate the control limits? And how to adjust the control limits at each timepoint to reflect the variation in subgroup size at each time point.
thank you
I have data from 9 time points from ~150 hospitals. At each time point hospitals reported counts of hypothermia cases among NICU admissions. The dataset has a lot of missing data due to either hospitals not reporting for a period. For presenting this data to our stakeholders I have been asked to create control charts, specifically pcharts showing the project-level hypothermia rates across the 9 timepoints with lower and upper control limits that include all available data. I am new to control charts, but I do understand my data violates some rules. Specifically, I do not have enough timepoints for calculate proper control limits and more importantly my data is nested within hospitals. If I do not account for the nesting when estimating standard deviations I will get tiny SDs as the denominators for my rates are quite large. Can someone please advise me on how to more appropriately calculate my SD that will be used to calculate the control limits? And how to adjust the control limits at each timepoint to reflect the variation in subgroup size at each time point.
thank you
