OBJECTIVE To use noninvasive approaches to assess stress in New York City (NYC) carriage horses during the course of their daily routine to determine whether use of these horses affected their well-being. DESIGN Prospective case control study. ANIMALS 13 (5 mares and 8 geldings) stabled working NYC carriage horses and 5 pastured (nonworking) NYC carriage horses (1 mare and 4 geldings). PROCEDURES Samples for determination of fecal glucocorticoid and salivary cortisol concentrations were collected on 3 successive days from 10, 8, and 9 working carriage horses during rest (time 1), preparation for work (time 2), and return to the stable (time 3) and at 1 hour after work (time 4). Infrared thermography (IRT) measurements were made to determine maximum temperature of the medial canthus at each time point. Fecal samples were also collected from 5 pastured carriage horses for determination of glucocorticoid concentrations. RESULTS No difference was found in mean ± SE fecal glucocorticoid concentrations between pastured (22.1 ± 9.8 ng/g) and working (19.5 ± 4.2 ng/g) carriage horses. A significant difference was found in salivary cortisol concentrations of working carriage horses between time 3 (0.96 ± 0.06 ng/mL) and time 4 (0.77 ± 0.07 ng/mL). The IRT measurement at time 2 (35.5 ± 0.64°C [95.9 ± 1.2°F]) was significantly lower than that at time 3 (36.2 ± 0.64°C [97.1 ± 1.2°F]). No other differences in IRT measurements were found. CONCLUSIONS AND CLINICAL RELEVANCE These working NYC carriage horses did not have physiologic responses indicative of a negative welfare status.