# Attribute Control Charts in Health Care

SPC for Excel Software

SPC Training

SPC Consulting

Ordering Information

Thanks so much for reading our publication. We hope you find it informative and useful. Happy charting and may the data always support your position.

Sincerely,

Dr. Bill McNeese
BPI Consulting, LLC

#### Connect with Us

• DaleWMay 3, 2017

The data in Table 1 seems odd, but not because the last two months might be out of control on a P chart. The oddity is no significant correlation between Unplanned Readmissions and number Discharged.  No correlation begs the question of what utility there is in control charting their ratio.A C Chart on Discharged count shows three VERY low months and two VERY high months, with the rest of the months more or less under control.  Thus the real mystery here is the 5 extreme outlier months from a normal or Poisson distribution in no obvious pattern for the Discharge counts.  Unplanned Readmissions are under control on a C Chart (or U chart at 0.53 per day) until the final month, when both variables for once go high together.

• billMay 4, 2017

I am not sure I would expect a correlation between discharged and unplanned readmissions.  If it is in control, there is just random variation in the process.  Using different charts does give you different answers.

• Amy RobertoFebruary 18, 2021

To meet this condition of a U chart (The counts are independent of each other, and the likelihood of a count is proportional to the size of the area of opportunity (e.g., the probability of a ADE occurring is the not related to an area in the hospital).), I think the better solution would be to stratify the data into units/areas where the probability is the same, rather than put messy data on an XmR chart.

• AnonymousAugust 28, 2021

Thanks Bill, I always value your posts. The other thing you can do with the XmR chart, is set the control limits on the 12 month in 2009 (rationale being you wish to look as data from 2009 compared to 2010) and also use Weaker Rules sets to look for data patterns, The second out of control point (5) is a weaker signal, 2 out of 3 points beyond 2 sigma

Regards Ian

• billAugust 28, 2021

Thanks for the kind comment.  Yes, you can set the control limits for one year – that is done quite often just for the reason you said – to compare this year to that baseline year.