Lately Dogsbite org and Anti Pit Bull Zealots have been running with this study like it is the word of God. Here is the study
Characteristics of 1616 Consecutive Dog Bite Injuries at a Single Institution
July 2016 | By Michael S. Golinko, MD, MA, Brian Arslanian, MD, and Joseph K. Williams, MD, FAAP
A retrospective review of 1616 dog bite victims over a 4-year period from a pediatric level 1 trauma center in the Southeast. Pit bull bites were implicated in half of all surgeries performed.
Study highlights
In this study of 1616 consecutive dog bite injuries treated at Children’s Healthcare of Atlanta, the only pediatric level 1 trauma center in the state of Georgia, there were 46 breeds of dog identified. The 3 most prevalent were 38.5% pit bull (also identified as Staffordshire bull terrier, American Staffordshire terrier, or bull terrier), 13% mixed breeds, and 8.1% labradors. Of the mixed breeds (n=66), 11 were pit bull mixes, 12 labrador mixes, and 4 labrador/pit bull mixes.
Features and morbidity of patients with the most severe injuries. In all 5.5% of patients underwent surgery; of these, 68.5% involved the head/neck region. Of the breeds identified, 50% involved pit bulls. With regard to breed, operative intervention was most strongly associated with a pit bull injury. Pit bull breeds were also more likely to bite in multiple anatomical locations (2.5 times as likely as other breeds to bite in multiple anatomical locations).
The lone mortality involved a 5 day-old girl attacked on the head by the family’s pit bull. The child underwent emergency craniotomy. Her postoperative course was complicated by acute respiratory distress syndrome, neurogenic pulmonary edema, and transfusion-related acute lung injury. Despite maximal ventilator support, she was persistently hypoxic and succumbed on postoperative day 3.
Pit bull injuries. Our data confirm what detractors of the breed and child advocates suggest—that, with rare exceptions, children and pit bulls do not mix well. Of the 8 studies listed in Table 5, 6 report pit bulls as the most prevalent breed, and in many cases, they inflicted the most severe injuries.
Our data were consistent with others, in that an operative intervention was more than 3 times as likely to be associated with a pit bull injury than with any other breed. Half of the operations performed on children in this study as well as the only mortality resulted from a pit bull injury.
The study’s authors agree with Bini et al, that attacks by pit bulls result in a higher injury severity score, lower Glascow coma score, higher risk of death, and higher hospital charges than attacks by any other breed.
The study’s authors recommend strong consideration to avoidance of any interaction between pit bull breeds and young children, particularly infants.
Here is the problems with this study
The study lumps the Bull Terrier, The Staffordshire Bull Terrier, And the Am Staff together as one breed with the American Pit Bull Terrier.
http://www.akc.org/dog-breeds/staffordshire-bull-terrier/
http://www.akc.org/dog-breeds/bull-terrier/
http://www.akc.org/dog-breeds/american-staffordshire-terrier/
https://www.ukcdogs.com/american-pit-bull-terrier
They claim mixed breed dogs accounted for 13% of the dogs involved in the attacks. If I’m reading it correctly the total number of mixed breed dogs is 66. 66 is only 4% of 1616 not 13%. . 5.5% required surgery and 50% of those were associated with pit bull attacks. 5.5% of 1616 is 88 so 44 pit bulls were involved in attacks requiring surgery. 44 is about 3% of 1616
If 38.5% of 1616 were pit bulls that comes to 622 pit bull related attacks. 44 required surgery which is 7% of the total attacks attributed to pit bulls. So 93% of pit bull related attacks only required a “band aid” as the BSL Activist put it.
When Dr. Golinko was interviewed about how he was certain what breeds were responsible he claimed the dog belonged to the family or friends of the family. Well there is a HUGE problem with that answer he gave.
Comparison of adoption agency breed identification and DNA breed identification of dogs.
Governmental and other agencies may require dog caregivers (owners) to provide breed identification of their dogs. This study compares breed identification by adoption agencies with identification by DNA analysis in 20 dogs of unknown parentage. Of the 20 dogs who had been adopted from 17 different locations, the study identified 16 dogs as having (or probably having) 1 or 2 specific breed(s) in their ancestry. DNA analysis of these dogs indicated that 25% (4/16) did in fact contain genetic evidence of an adoption agency’s identified breed as one of the predominant breeds in a dog’s ancestry. DNA analysis did not detect all specified breeds in 14 of these dogs. That is, 87.5% of the dogs identified by an adoption agency as having specific breeds in their ancestry did not have all of those breeds detected by DNA analysis. The discrepancies between opinions of adoption agencies and identification by DNA analysis suggest that it would be worthwhile to reevaluate the reliability of breed identification as well as the justification of current public and private policies pertaining to specific dog breeds.
https://www.ncbi.nlm.nih.gov/pubmed/20183478
And when viewing the full study the Limitations and Bias verify the study i just posted
Certain variables such as breed of dog could not be independently verified. Also you will see it says : The authors acknowledge that etiology of a dog bite is complex and multifactorial depending not only on the canines characteristics but also on owner training, child behavior and the specific conditions when the bite occurred. And it states: There may be a reporting bias for “typically biting breeds” such as pit bulls.:
And it gets worse. Where did he obtain his percentages for breeds involved in fatalities that is listed under the CDC statistics.? 
The CDC does not track by breed that pit bull fatality percentage comes from Dogsbite org. Thusforth that percentage is not only incorrect it is also flashy and irrelevant as dogsbite claims Pit Bulls make up 6% of the dog population in the United States. In the United States there is an estimated 89.7 million dogs. 6% of 89.7 million equals 5,382,000 pit bulls. And if we entertained dogsbite org’s numbers they claim 22 fatalities in 2016. The 78% mentioned equals 22. And 22 is 0.000004% of 5,382,000.
Next up the age and locations of injury on the body. That is also a flashy yet irrelevant point in regards to pit bulls. The younger the child the shorter in height which is why children over 12 were more commonly injured in lower body as opposed to younger children injured in neck face and head. Using the example of the five day old child is also flashy yet irrelevant a five day old child is extremely fragile. What would be a minor injury for an adult can be fatal to a baby especially a 5 day old baby. Any breed can kill a baby. A Black Lab Puppy killed a baby
http://m.newson6.com/story.aspx?story=8753165&catId=112042
A Pomeranian killed a baby
http://amarillo.com/stories/100900/usn_pet.shtml#.We3jlnMpAwg
Injury and fatality rates involving dog attacks will always be higher and more severe with younger children REGARDLESS of breed. The CDC doesn’t track breeds anymore for numerous reasons one being
As for the claim that a pit bull bite/attack/mauling is somehow unique or worse than other bite/attack/maulings of all other breeds that has been proven false repeatedly this study from Ireland proves
As well as france
http://www.journalvetbehavior.com/article/S1558-7878(17)30149-1/fulltext
And it says in the study they sided with Bini. I repeat Bini.The same person who was behind the Mortality and Maimings study that has been debunked. See the peer review study here that debunks it
And also Jim Crosby broke down his study as well.
http://jimcrosby.canineaggressionissueswithjimcrosby.com/2011/05/response-to-injury-study.html?m=1
And also Bini’s sources of study were Youtube Videos and Dogsbite org. The Golinko study is biased, flawed and lacks validity. This study is quack science
Dogs R Dogs 