| Objective: To evaluate the impact of frailty on respiratory function in a community-dwelling elderly. Method: 51 community-dwelling elderly were evaluated (mean age of 73±6 years), being 29 men (56.7%) and 22 women (43.3%). We collect the following variables: sociodemographic characteristics, frailty phenotype, pulmonary function test and assessment of the respiratory muscles using an analog manometer. The statistical analysis was performed using the Kolmogorov and Smirnov tests, one-way ANOVA, Paired Student's t-test and Pearson correlation coefficient (p<0.05). Results: There were no statistically significant between-group differences among the frail group (FG=9.8%), pre-frail group (PG=47.1%) and non-frail group (NG=43.1%), in relation to anthropometric, demographic and spirometric data. Regarding to the maximum inspiratory and expiratory pressures (MIP and MEP), statistically significant between-group differences were observed among the three groups, being these pressures significantly lower in the FG and PG compared to the NG. With regards to the obtained and predicted values, the FG and PG showed statistically significant difference (p=0.004). The PG showed positive correlations between the MIP and MEP with the values of hand grip strength (r=0.7). The NG showed positive correlation between the MEP and the values of physical activity level (r=0.7). Conclusions: The study demonstrated that maximal respiratory pressures may decrease according to the frailty condition among the nonfrail, pre-frail and frail elderly. Furthermore, it also indicated a positive correlation between inspiratory muscle strength, expiratory muscle strength and hand grip strength in pre-frail elderly. Further investigation with regards to prevention or intervention programs that incorporate actions to minimize the loss of respiratory function are necessary in order to reverse or prevent the progression of the frailty condition.