The focus of sustainable lighting tends to be on reduced CO emissions and cost savings, but not on the wider environmental effects. Ironically, the introduction of energy-efficient lighting, such as light emitting diodes (LEDs), may be having a great impact on the health of wildlife. These white LEDs are generated with a high content of short-wavelength 'blue' light. While light of any kind can suppress melatonin and the physiological processes it regulates, these short wavelengths are potent suppressors of melatonin. Here, we manipulated the spectral composition of LED lights and tested their capacity to mitigate the physiological and health consequences associated with their use. We experimentally investigated the impact of white LEDs (peak wavelength 448 nm; mean irradiance 2.87 W/m ), long-wavelength shifted amber LEDs (peak wavelength 605 nm; mean irradiance 2.00 W/m ), and no lighting (irradiance from sky glow < 0.37 × 10 W/m ), on melatonin production, lipid peroxidation, and circulating antioxidant capacity in the tammar wallaby (Macropus eugenii). Night-time melatonin and oxidative status were determined at baseline and again following 10 weeks exposure to light treatments. White LED exposed wallabies had significantly suppressed nocturnal melatonin compared to no light and amber LED exposed wallabies, while there was no difference in lipid peroxidation. Antioxidant capacity declined from baseline to week 10 under all treatments. These results provide further evidence that short-wavelength light at night is a potent suppressor of nocturnal melatonin. Importantly, we also illustrate that shifting the spectral output to longer wavelengths could mitigate these negative physiological impacts.