How can I speed up a function operating on a table?

I have a table that is roughly 100x20. I have a function that takes a set of rows from this table and returns the same set of rows with some of the variables (columns) modified. For the sake of clarity, I use dot indexing within the function to access the variables I'm modifying. I call this function many times, and the majority of the time spent on the function is spent in table.subsref and table.subasgn. How can I speed up my function?

Below is a generic function with similar behavior.

load patients
patients = table(Age,Gender,Height,Weight,Smoker,...
    'RowNames',LastName);
patients.Factor = zeros(size(patients,1),1);

rows = strcmp(patients.Gender,'Male');
patients(rows,:) = TableFcn(patients(rows,:),10);

rows = strcmp(patients.Gender,'Female');
patients(rows,:) = TableFcn(patients(rows,:),7);

function T = TableFcn(T,modulus)
    if nnz(double(T.Smoker)) >= 5
        TotalHeight = sum(T.Height);
        Factor = mod(100*T.Height/TotalHeight,modulus);
    else
        TotalWeight = sum(T.Weight);
        Factor = mod(100*T.Weight/TotalWeight,modulus);
    end

      Numbers = repmat((1:2:9).',20,1)/10;
      for ii = 1:size(T,1)
          if T.Age(ii) < 30
              T.Factor(ii) = 18*Factor(ii);
          elseif T.Age(ii) < 40
              sum(Numbers(1:T.Age(ii)))
              T.Factor(ii) = Factor(ii)*sum(Numbers(1:T.Age));
          else
              T.Factor(ii) = NaN;
          end
      end
  end

[Geoff Hayes]

Are you using profile to determine that the majority of time spent is in the table.subsref and table.subasgn methods?