March 1, 2001	Data Managers receive new data call manual.

April 13, 2001	Deadline for submitting data.

October 2000		Please return the Fall 2000 MDS Update Report forms.
(deadline=Nov.1)	NACC will use your comments in considering changes
			to the data elements/variables, which must be 
			finalized soon.

early Mar. 2001	Look over the new data call manual.
			Call/e-mail NACC with any questions, big or small.

Mar.1-Apr.13, 2001
			Expanded use of FAQ page to share Q&A with
			all Data Managers via NACC's web site.

			Technical computing problems?  If you've spent
			more than 1 hour on one problem, call us.
			Coding questions? Call us, even for small things.

April 13, 2001	Please submit your data by the deadline.
			If the deadline was a problem in 2000, please 
			plan ahead and get an early start.

NACC's first data call went reasonably smoothly. Several Centers submitted their data early, for which we were very grateful. It allowed us to verify that the data submission process worked correctly, and to iron out any problems before the majority of Centers submitted their data. "Thank you"s go to Ruth Cha (Mayo), Robin Guariglia (Oregon), Scott Mobley (UC- Irvine) and Kathy Foster (UC San Diego).

Thank you to all the data managers and their staffs for their efforts to meet the deadline.

The green data call manual was long, 60 pages, and we received some positive comments on it, which were appreciated. However, few people returned the feedback sheet in the front of the manual. Next year's manual will probably be similar, so if you have any comments or suggestions, please fill out the feedback sheet and fax it to us, or just e-mail us.

I was surprised we didn't get more calls during the data call. Several people did call and we were able to clear up most questions quickly. For example, in one case, Duane recreated on his PC what the data manager was doing and was able to isolate the problem. This took 15 minutes on the phone to solve a problem the data manager had spent a day struggling with. We would like to give you this guideline:

	If a problem with NACC's software or with submitting
	data takes more than 1 hour, call us.

The quality of communications was good between NACC and several Centers where there were questions or assistance was needed. We would like to increase the quantity of communication, fulfilling our role to help you. Remember that good communications will lead to good data. If you have even a small doubt about how to code something, call us. You might make the correct assumption, but another Center might have the same question and make the wrong assumption. We need to know what your questions are, even on little things.

We had an FAQ page (Frequently Asked Questions) on NACC's web site. We didn't get many questions, so it was of limited use. We want to expand it next year. If you have a problem and let NACC know about it, even if you solved the problem yourself, we can put it on the FAQ page, helping some of the other data managers who come up against the same thing. By communicating with NACC you can allow us to be a conduit of help to the other 29 Centers.

We just sent you these report forms a few weeks ago. We need your input to improve the next data call, especially regarding problems you had locating data and mapping data to the MDS codes. We're not expecting that the MDS data can be made perfect, we're just aiming for improvement.

One good example is the NPSYCH variable. Many Centers had a large number of "No" or "Missing" answers. Upon investigating, we found there were different interpretations of what a neuropsych test was. Our instructions in the data element dictionary did not specify that this variable did not pertain to the Mini-Mental State Exam (MMSE). Therefore, different Centers coded NPSYCH differently and now the MDS data for this variable are less than pure. Good communications and closer contact between NACC and the Centers might have caught this problem earlier on.

Another example is the overuse of "Other", such as happened with the NPTHPRIM and NPTHSECN variables for neuropath diagnoses. We intended that the codes we provided should have covered almost all cases, with a few rare ones requiring an "Other" code. The only way we can improve this variable is by you letting us know what those "Others" represent in your Center's data for those subjects. Then we can add new codes or provide more specific mapping instructions for existing codes.

As much input as you can give us will be greatly appreciated, and should result in clearer coding instructions and better data next time.