How to subset a remote tbl in r?

Question

I have loaded a sql table to a tbl using dplyr (rr is my db connection) :

s_log=rr%>%tbl("s_log")

then extracted three columns and put them in a new tbl :

id_date_amount=s_log%>%select(id,created_at,amount)

when i run head (id_date_amount) it works properly:

      id              created_at                amount
1   34101           2016-07-20 10:41:23          19750
2   11426           2016-07-20 10:38:15          19750
3   26694           2016-07-20 10:38:18          49750
4   25656           2016-07-20 10:42:05          49750
5   23987           2016-07-20 10:40:31          19750
6   24564           2016-07-20 10:38:35          19750

now , i need to filter the id_date_amount in a way that it only contains the past 21 days:

filtered_ADP=subset(id_date_payment,as.Date('2016-08-22')- as.Date(created_at) > 0 & as.Date('2016-08-22')- as.Date(created_at)<=21)

i get the following error:

Error in as.Date(created_at) : object 'created_at' not found

i think that's because i don't have id_date_payment locally , but how can i shape that subset to sent it to id_date_payment and get back the results?

i tried using deployer::filter :

id_date_amount=id_date_amount %>% filter( as.Date('2016-08-22') - as.Date(created_at) > 0 & as.Date('2016-08-22')- as.Date(created_at)<=21 )  

but i get this error :

   Error in postgresqlExecStatement(conn, statement, ...) : 
  RS-DBI driver: (could not Retrieve the result : ERROR:  syntax      error at or near "AS"
LINE 3: WHERE "status" = 'success' AND AS.DATE('2016-08-22') - AS.DA...
                                   ^
)