Use the "WHERE" clauses for filteration, and "GROUP BY" for summarization.

You will need to share your SQL schema and objectives with a SQL professional to get answers.

Alternatively, you can do an"OFF TOPIC" (OT) post here with that info, and try your luck.

Yeah, that's is what SQL is made for.

First, some comments on the SQL itself:

Remove the quotes. Without spaces or case sensitivity, there's no point. Worse, if the query is run in a case sensitive environment, it will fail. Quotes preserve case, but databases usually uppercase all names. So, unless the name was created with quotes and something other than uppercase, you probably want to avoid using them. There's also the clutter they create.

Keep like columns together. For ease of reading, keep columns from the same table together. Otherwise, it can get hard to identify which columns come from which table. As the order the columns are used in is how you use the variable in your code, there's no reason to order the columns in any specific way. So, i'd go for ease of reading.

Alias normally. Why people use coded names for table is beyond me. Regardless, they do it. But if you're going to alias the names in the query, use words that make sense and are easily identified.

Qualify all column names. When there is more than one table in the query, make sure all names are qualified. This avoids errors, and helps the reader understand what is going on.

Just say no to inner joins. ANSI SQL Join syntax is just plain wrong. While many disagree with my opinion, i have to mention how hard that query is to discern. Moved into the WHERE clause, makes it so much cleaner and obvious.

Taking those notes into consideration, the query can be rewritten as follows:

SELECT TOP 500
    Service.Voucher_Number,
        Service.Service_Date_From, 
        Service.Procedure_Code,
    --
        Patient.Patient_Number, 
        Patient.Patient_First_Name, 
        Patient.Patient_Last_Name, 
    --
    Payment.Transaction_Type,
        Payment.Transfer_To_Carrier_Abbr
FROM
    Ntier_Training.PM.vwGenSvcInfo        Service,
    Ntier_Training.PM.vwGenPatInfo        Patient,
    Ntier_Training.PM.vwGenSvcPmtInfo    Payment
WHERE
    Patient.Patient_ID = Service.Patient_ID
  AND    Payment.Service_ID = Service.Service_ID
ORDER BY
    Service.Voucher_Number;
[download]

Isn't that easier to read and understand? :)

As for the aggregation, you probably want to use GROUP BY. Without knowing exactly what is required, it is hard to suggest a specific query.

You could use a hash of arrays

my %pats_hoa=();             
while (my $row = $sth->fetchrow_hashref){
  my $voucher   = delete $row->{'Voucher_Number'};
  my $proc_code = delete $row->{'Procedure_Code'};
  push @{$pats_hoa{$voucher}{$proc_code}},$row;
}

for my $voucher (sort keys %pats_hoa){
  for my $proc (sort keys %{$pats_hoa{$voucher}}){
    for my $rec (@{$pats_hoa{$voucher}{$proc}}){
      if ($rec->{Transaction_Type} eq '???'){
        # do something;    
      }       
    }
  }
}
[download]