There are differing legal definitions of the term “stillbirth” in different jurisdictions and also modern registration of these sad events differs from country to country. As a genealogist you encounter stillbirths e.g. as “stillborn son” or “stillborn daughter” in the birth section of church records. Details on the circumstances of the birth are in the majority of cases not available.

At least in the German-language area such stillbirths can be marked with the symbol †* in conventional family trees.

I wonder what is best practice to records stillbirths in family tree software? The child never carried a name. Should I enter "stillborn son" like in church records in the first name field? Is it a good idea to add the parent’s surname to the correspondent field to sort the stillbirth among other family members? Should I set the birth and death date to the the day the death child was born, although it might have died earlier and wasn’t born alive? How should I distinguish the entry of a stillbirth from the entry of a child that was born alive and died on the same day, without christening and thus without a name as well? With a note?

So please tell me: What is best practice to record a stillbirth as a genealogical event?

2 Answers 2


The GEDCOM standard includes STILLBORN as one of the possible values for its AGE_AT_EVENT item. It gives the following example:

2 DATE 13 MAY 1984

with the meaning that this person died at age approximately 0 days old. Although the GEDCOM example shows this on a death event, I have also seen it on the birth event.

GEDCOM also gives an example indicating that stillborn can be a qualifier as a type of birth, as in:

2 DATE 13 MAY 1984
2 TYPE Stillborn

Most software at least loosely follows the GEDCOM standard, so that it can import and export data with other programs. So check in your genealogy software to see if it has an option to select "stillborn" as either an age or a type, on either birth or death events. If your software allows it (hopefully in just one unambiguous way), then you'll be recording the stillborn event in a manner that will probably be exported correctly to GEDCOM, and will then be available to be correctly read by your program again, or by other programs that recognize that construct.

If your program has no such option, I would not create a custom event type to replace your birth or death event. Doing so will prevent giving your program use of the birth and death info for that child and other programs won't understand that custom event. "Stillborn" is not really an event but is a descriptor to be placed on a birth or death event. So if you have nothing better, then add it as a note on your birth or death event.

Only use the name field for names. If your program has an option for "unknown" or "unnamed", then use that, but don't put such information in the name field. Otherwise, leave it blank. Adding the parent's surname is allowable but that's up to you. Use notes as you suggested to provide more information.

Include both a birth and death event because your program makes use of those dates and places. Set those using the most reliable information you have, and state the basis of that information in the notes. Don't set the death date prior to the birth date. Many programs will continually report that you have an error. if you think the child might have died earlier or wasn't born alive, just add that as a note.


I don't know if there's a general best practice, but this is what I do.

I create the child as a person and add a death event with the cause of death as "stillborn". If your genealogy software doesn't have a cause of death field then a note would be fine - sometimes I use both. This is how I would distinguish them from a child that died on the same day it was born, who would either have a blank cause of death, or a cause of death other than stillborn. If I know the date, I'll set the date of death to match the date of birth (yes, I still have a birth event). I always take this approach if I know the name of the child.

If I don't know the name of the child then I usually only create them as a person if they are closely related or "interesting" in some way, for example if I have some other type of information to store about them such as where they were "born", a note about issues the mother may have had during pregnancy, etc. In this case I would use the father's surname and leave the given names blank (assuming the family was in a region/period when children would generally take their father's surname), but I would have a note stating explicitly that the child's surname is an assumption.

If I decide against creating the child as a person then I would just add a note to the family stating that there was a still birth, and stating when it occurred in relation to the other children.

As I wrote this answer another option occurred to me. The software I use (Gramps) lets me use custom event types, so I have the option of creating a Stillbirth event rather than a Birth and/or Death event. It would be very understandable when looking at a child's information, but I imagine it might cause havoc with the reports as the program may not be able to determine whether or not the child is alive. But if your software allows custom event types this may be an option for you.

  • Thank you for your helpful answer and welcome to genealogy.SE!
    – lejonet
    Nov 20, 2013 at 14:34
  • I guess the same would go for a pre-natal injury, such as a car crash. Jan 3, 2018 at 19:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.