As the genetics tests become more economical than celiac blood tests, and are more accurate than celiac blood tests, then it makes sense to start with genetic HLA testing.
Read the full article here from International Journal of Celiac Disease, 2017.
Here is an excerpt: My Bold
3. A Step Wise Serology/Genetic Approach
CD patients negative for any of these HLA alleles are very rare. Therefore, the absence of both HLA-DQ2 and HLA-DQ8 heterodimer makes diagnosis of celiac disease very unlikely (sensitivity >96 %). HLA typing of patients has been included as a useful test to exclude celiac disease in the ESPGHAN guidelines for CD diagnosis. [8, 9] HLA typing confers a high negative predictive value: patients with a negative HLA (i.e. neither DQ2 nor DQ8) will not develop CD; and a suggested strategy to avoid repeated CD screening would be to first perform an HLA test. 
Targeting the HLA risk first, rather than tracking positive serology, would be a reasonable step-up approach, probably cost effective and time saving: in the past, HLA typing has been expensive and time-consuming, but new single nucleotide polymorphisms techniques  and other combined home-made procedures  have recently been reported as very cost-effective and work-time saving for HLA-DQ2 and DQ8 genotyping in CD screening................
In general population, the preferred test to screen for CD is the measurement of IgA TTG [Link here to a critique of the IgA TTG test]along with total serum IgA to avoid false-negative results due to selective IgA deficiency. Positive serology would lead to endoscopic small intestinal biopsies . These serological tests, based on TTG associated to endomysial and deamidated gliadin peptides antibodies are recognized as performant screening tools. 
However, in asymptomatic members of a high-risk group, like those presenting RA, it seems reasonable to test first for negative result of HLA-DQ2/DQ8 in order to exclude CD, so that further serologic testing would be unnecessary . Performing HLA genetic typing seems cost effective and could avoid subsequent fiberoscopies and biopsies .................recent studies emerging from the South Hemisphere confer solid arguments to such strategies  as CD is reported to be strongly associated with HLA-DQ2 in these regions .
It is my opinion and the opinion of these authors that the genetic tests for HLA provides a flexible, cost-effective methodology that could be applied to protocols to diagnose celiac disease and to obtain accurate estimates of the prevalence of CD in large cohort studies. The ESPGHAN guidelines for CD diagnosis are worth a look if you are interested in clinical matters.
To Your Health
Dr. Barbara (TM)