It is often very difficult to confidently assign an attack of vivax malaria as being due to either relapse (from hypnozoites) or recrudescence (from blood). This was especially true in Allied Prisoners of War (POW) many of who recorded >12 malaria attacks / year especially on the Thai-Burma railway. Published and unpublished historical records were examined to determine if this dichotomy could be distinguished using extant data. Even at the beginning of railway construction in 1942, vivax malaria predominated 2:1 over falciparum. Surveys done during construction in 1943 indicated that 70% of new infections were due to P vivax. Once POW had been repatriated to Singapore, 92% of all infections were from P vivax at about 6 episodes / person-year. Clinical attacks were often recorded as reoccurring within 10-15 days of previous treatment which was usually 2g of quinine daily for 10 days when stocks were available. Wartime supply issues likely negatively impacted on the quantity and quality of quinine from the Javanese plantations. Antimalarial drugs (mainly quinine) were traded as a barter item in the cashless prisoner economy. Only limited quantities of pamaquine (8-aminoquinoline) were available. Malaria-induced anemia (<9 gm Hb) occurred in ≈10% of POW necessitating a major blood transfusion effort of >3000 units during the last year of the war; >80% of all POW blood transfusions were given for malaria anemia. Recurrent malaria was common in former POW until 1950 when it largely disappeared. It seems likely that a combination of sub-optimal therapy (suppression without cure, possibly limiting immunity) was superimposed on a debilitated population with very high infection rates. Successive relapses were possibly triggered by the fever and hemolysis in the preceding clinical attack, leading to an apparent unending series of infections until all hypnozoites were expended.