Simple Summary: The COVID-19 pandemic has contributed to elevated rates of mental health problems and distress among the U.S. population . Pets may be an important source of social support to combat social isolation . This cross-sectional study used latent profile analysis to identify subgroups of U.S. pet owners based on their perceived mental health symptoms prior to and after the onset of the pandemic . Latent transition analysis was used to determine the stability of subgroup membership and examine the effect of attachment to pets on transition probabilities . Five subgroups were identified: low symptoms, mild symptoms, moderate symptoms, high symptoms, and severe symptoms . Evidence of moderation was found, X2 (16) = 41.47, p = 0.04 . Specifically, results indicated that attachment to pets functioned as a protective factor for individuals exhibiting moderate and high levels of mental health symptoms, as above average attachment to pets was associated with greater odds of transitioning to a less severe symptom profile . However, individuals with severe symptom profiles and high attachment to pets fared worst in the context of COVID-19 restrictions . This study has important implications for future research investigating the role of pets on mental health and for those providing services to pet owners during the COVID-19 pandemic . Abstract: This cross-sectional study examined whether, and to what extent, attachment to pets was associated with changes in latent patterns of adults ’ perceived mental health symptoms during the COVID-19 pandemic (n = 1942). We used latent transition analysis to determine the stability of subgroup membership pre- and post-COVID and the effect of attachment to pets on transition probabilities . Mental health before COVID-19 was measured retrospectively . Five subgroups were identified: low symptoms, mild symptoms, moderate symptoms, high symptoms, and severe symptoms . Among individuals in the moderate and high symptoms subgroups, those who reported high attachment to pets generally had greater odds of transitioning to a less severe symptom profile (OR = 2.12) over time than those with low attachment to pets (OR = 1.39). However, those who had a severe symptom profile and high attachment to pets had lower odds of transitioning to a less severe symptom profile (OR = 0.30) and higher odds of maintaining a severe symptom profile (OR = 3.33) than those with low attachment to pets . These findings suggest that the protective and risk effects of attachment to pets differ based on individuals ’ psychological symptom patterns across multiple indicators . We discuss the implications of these findings for research, policy, and practice.