H5N1 did not transmit easily between humans in the wild...

Hat tip to @Laurie_Garrett and CIDRAP

Despite wearing next to no personal protective equipment (5% of 419 contacts used a mask, face shield, gown or gloves) and coming into contact with sick or dead poultry (12% of contacts), 85/87 household members and 332 "less close" contacts of 23 influenza A(H5N1) virus cases did not show any significant sign of antibodies to the virus, a study published in PLOS|ONE by Bai and colleagues noted.

Only 2 (0.4% of all contacts tested) were defined as infected by H5N1  during the study period of 2005-2008, on mainland China. 

The study used both haemagglutination test (antibodies in the patients sera bind horse red blood cells together giving a distinctive pattern) and micro-neutralization (presence of specific antibodies in a sample prevents a lab stock of virus from infecting a cell line-amount of virus can be determine by making dilutions of  the sample and comparing to a sample with no antibodies to the virus). When they had a single serum, the authors used:

• A neutralizing antibody cut-off titre of ≥40 children (<14-years of age)  with a haemagglutination titre ≥40.
• A neutralizing antibody cut-off titre ≥80 for those aged 15-59-years with a haemagglutination titre ≥40.

For acute and convalescent sera pairs positivity to H5N1 was defined as:

• A ≥4-fold rise in neutralizing antibody titre between acute and convalescent sera
• Convalescent sera needed a neutralizing titre of ≥40 for children and ≥80 for adults, or a haemagglutination titre ≥40

There were a few more positives below these cut-offs.

While genetically altered H5N1 can be made to spread among ferrets in the lab, it seems that some years ago in the wild, H5N1 had a ways to go before it could spread efficiently between humans. That's a good thing.