In Indonesia, endoscopy services are limited and studies about gastric mucosal status by using pepsinogens (PGs) are rare. We measured PG levels, and calculated the best cutoff and predictive values for discriminating gastric mucosal status among ethnic groups in Indonesia. We collected gastric biopsy specimens and sera from 233 patients with dyspepsia living in three Indonesian islands. When ≥5.5 U/mL was used as the best cutoff value of Helicobacter pylori antibody titer, 8.6% (20 of 233) were positive for H. pylori infection. PG I and II levels were higher among smokers, and PG I was higher in alcohol drinkers than in their counterparts. PG II level was significantly higher, whereas PG I/II ratios were lower in H. pylori-positive than in H. pylori-negative patients. PG I/II ratios showed a significant inverse correlation with the inflammation and atrophy scores of the antrum. The best cutoff values of PG I/II were 4.05 and 3.55 for discriminating chronic and atrophic gastritis, respectively. PG I, PG II, and PG I/II ratios were significantly lower in subjects from Bangli than in those from Makassar and Surabaya, and concordant with the ABC group distribution; however, group D (H. pylori negative/PG positive) was the lowest in subjects from Bangli. In conclusion, validation of indirect methods is necessary before their application. We confirmed that serum PG level is a useful biomarker determining chronic gastritis, but a modest sensitivity for atrophic gastritis in Indonesia. The ABC method should be used with caution in areas with a low prevalence of H. pylori.