Washington:

Republican presidential candidate Donald Trump on Wednesday stood by debunked claims that immigrants in Ohio were eating pets, telling Latino voters during a town hall he was “just saying what was reported.”

Trump in recent weeks has amplified a false claim that has gone viral that Haitian immigrants in Springfield, Ohio, were stealing residents’ pets or taking wildlife from parks for food.

There have been no credible reports of Haitians eating pets, and officials in Ohio – including Republicans – have repeatedly said the story is untrue.

At a town hall hosted by Univision, an undecided Latino Republican voter from Arizona, a battleground state, asked Trump whether he truly believed that immigrants were eating pets.

“I was just saying what was reported. All I do is report,” Trump replied during the event held in Miami. “I was there, I’m going to be there and we’re going to take a look.”

Trump added that “newspapers” had also reported on the claim, without naming any or providing any details.

Trump, who has not yet traveled to Springfield, has previously said he would conduct mass deportations of Haitian immigrants from the Ohio city, even though the majority of them are in the U.S. legally.

The city has faced bomb threats since Trump began repeating the false accusations about Haitians.

In the final weeks before the Nov. 5 election, Trump is increasingly resorting to darker and more violent language about illegal immigration, an issue that opinion polls show resonates with many voters, especially Republicans.

He is competing against Democrat Kamala Harris for key votes from the growing Latino population. Latino voters have typically backed Democrats, but the Trump campaign is hoping to win over more Latino voters, especially men, on the back of economic discontent.

Harris led Trump by eight percentage points – 47% to 39% – among Hispanic voters in Reuters/Ipsos polling conducted between Sept. 11 and Oct. 7.

(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)