While the impact of social restrictions on sexual and romantic life early in the COVID-19 pandemic has been widely studied, little is known about impacts beyond the initial months. We analyzed responses from 2,098 British adults (aged 18-59) taking part in the Natsal-COVID study (Waves 1 and 2). Participants were recruited via a web panel and surveyed twice: four months and one year after the start of the UK's first national lockdown (July 2020 and March 2021). Changes in the prevalence and frequency of participants' physical and virtual sexual behaviors between the two surveys were analyzed using multinomial logistic regression. Changes in the quality of intimate relationships were modeled using logistic regression for the 1,407 participants in steady relationships, adjusting for age, gender, and relationship status. The reported prevalence of any sexual activity amongst the full sample increased over the study period (from 88.1% to 91.5%, aOR = 1.50, 95% CI 1.23-1.84). Increases were observed for physical (aOR = 1.41, 95% CI 1.15-1.74) and virtual (aOR = 1.20, 95% CI 1.07-1.34) activities, particularly masturbation (aOR 1.53, 95% CI 1.37-1.72). Increases were larger for men than women. The proportion of participants in steady relationships whose relationship scored as "lower quality" increased (from 23.9% to 26.9%, aOR = 1.28, 95% CI 1.10-1.49). These findings have implications for understanding sexual health needs during disasters and planning sexual health service priorities following the pandemic.