Objective To compare the effects of different pre⁃hospital analgesia on the prognosis of elderly patients with hip fracture after surgery. Methods A total of 180 elderly patients with hip fractures were enrolled. According to the random number ta⁃ ble method, they were divided into three groups (n=60): a non⁃steroidal analgesia group (group N), a fascia iliaca block group (group F), and a lumbar plexus block group (group L). Before admission to the emergency department, the three groups underwent oral non⁃steroi⁃ dal analgesia, ultrasound⁃guided iliac fascia block analgesia, and lumbar plexus block, respectively. Their Visual Analogue Scale (VAS) scores were assessed before pain treatment (T0), after pain treatment (T1), in the early morning of the surgery day (T2), 24 h after surgery (T3), and 72 h after surgery (T4). The levels of serum interleukin (IL)⁃6, tumor necrosis factor (TNF)⁃α and substance P were measured at T2, T3, and T4. The intensive care unit (ICU) admission rate, the length of hospitalization stay, the mortality of 1 and 6 months after operation, and complications were recorded. Patients were followed up by telephone 1, 3, and 6 months after surgery. Ac⁃ cording to the modified version of Barthel score sheet, their abilities of daily living were evaluated. Results The incidence of respi⁃ ratory failure and heart failure was higher in groups F and L than group N (P<0.05). The serum IL⁃6 concentration of the three groups first increased along with time and then decreased, while TNF⁃α concentration gradually decreased along with time. Groups F and L produced lower serum IL⁃6 concentrations than group N at T2 (P<0.05), and lower serum IL⁃6 and TNF⁃α concentrations than group N at T3 and T4 (P<0.05). The serum substance P concentration was lower in groups F and L than group N at T2, T3, and T4 (P<0.05), while the serum P substance concentration was lower in group L than group F at T2 and T3 (P<0.05). The serum substance P index of group N and group L decreased first and then increased with time, while that of group F decreased gradually with time; The rest VAS pain score and exercise VAS pain score of the three groups decreased significantly at T1, and then changed with time. The VAS scores at rest and during movement were lower in groups F and L than those in group N at T1 and T4 (P<0.05); group L produced a higher VAS score dur⁃ ing movement than groups F and N at T0 (P<0.05), and a lower VAS score during movement than group F at T1 (P<0.05). Groups F and L presented a lower incidence of at least one postoperative complication than group N (P<0.05), and a lower incidence of delirium than group N (P<0.05). The length of hospitalization stay in group F was shorter than that in groups N and L (P<0.05). The difference in oth⁃ er indicators was not statistically significant (P>0.05). Conclusions Among the three prehospital analgesia methods, prehospital re⁃ gional block analgesia can significantly relieve perioperative pain at rest and during movement, reduce perioperative inflammation and the incidence of delirium, and shorten the length of hospitalization stay to some certain extent; lumbar plexus nerve block is not superi⁃ or to iliac fascia block; the effects of prehospital regional block analgesia on long⁃term prognosis needs further study.