It’s been two months since the first cases of Omicron were identified in the UK. In the weeks after it emerged, modelling scenarios presented to Sage suggested we were facing a situation even worse than last winter, with potentially tens of thousands of hospital admissions and thousands of deaths a day. Thankfully, as we now know, these scenarios did not materialise. The numbers of admissions and deaths peaked below the level expected in even the best-case scenarios. Pressure on hospitals remained very high, but in most cases, the situation was better than feared. The editor of the Lancet, Richard Horton, this week described scientists’ response to Omicron as “a case study in error”. He attributed this to an “over-reliance on mathematical modelling and too little emphasis on the experience of health workers on the frontlines of care”, with insufficient attention paid to the views of South African doctors. “Following the science” in relation to restrictions was always a misnomer, as “the science” was often uncertain, leading to models that inevitably provided a very wide range of scenarios with different levels of restrictions. Many people are now understandably concerned about the potential negative impact of ending plan B restrictions. But again, it is important to look at what’s actually happened to see how effective these restrictions have been – and whether they’ve led to better outcomes. Plan B restrictions have effectively been in place in Wales and Scotland since July (in England, the government announced the move to plan B on 8 December). A comparison of death rates since 19 July from ONS data shows that England has actually had the lowest death rates. (It is not possible to directly compare case and hospitalisation rates due to differences in how they are measured, but these are both closely linked to death rates.) Many have long assumed that more and earlier restrictions lead to better outcomes than voluntary behaviour changes. This assumption is largely based on evidence from the pre-vaccination era, when the countries that locked down quickly against Covid experienced far fewer deaths and hospitalisations. But over the last few months, the real-world evidence no longer appears to support this assumption: now, the main determinant of hospitalisation and death rates is the level of immunity in a population, through both vaccination and natural infection. This is especially the case among older and higher-risk groups. This is why England – where 98% of over 15s have some immunity to Covid-19 – appears to have fared relatively well since July compared with other European countries, despite having fewer restrictions. Behaviour change and compliance to rules also play a role, of course. There is evidence that household mixing in England rose and fell along with perceptions of risk rather than necessarily because of the rules in place at the time. And we have now seen that it is voluntary behaviour changes over the last few weeks, such as reducing contacts, that has led to the same reduction in admissions and deaths that the models showed would be produced by a return to step 1 of the roadmap. At the other extreme, lockdown sceptics now say that because lockdowns and other non-pharmaceutical interventions such as social distancing are no longer needed, they were never needed – even though before the vaccine programme, Covid-19 overwhelmed the NHS and the health service was not able to provide all of its services. Even during this wave, the pressure on the NHS remains very high – particularly due to staff absence – and many of my colleagues on the frontline are physically and mentally exhausted. That is why it’s so important for everyone to continue to follow the public health guidance – including wearing masks – even after mandatory restrictions end. Medical and public health interventions are usually judged by the criteria: do they have a clinically significant benefit? Does that benefit outweigh any harms? And are they the best use of resources – or would spending money on something else produce greater benefits? Of course, during the first wave this evidence was lacking – which is why lockdown measures were justified. Even in the second wave, there was sufficient evidence to show that the benefits of lockdown outweighed the costs – especially with the arrival of vaccines, when lockdowns were not just delaying admissions and deaths but actually preventing them. Vaccines have since transformed the situation. The cost-benefit equation has changed, and the benefits of lockdown restrictions do not outweigh the harms. It’s important to remind ourselves of we are trying to achieve – not just a reduction in harms from Covid, but also from the measures used to control it. This is best measured by quality-adjusted life years (QALYs), not deaths – a measurement that takes into account disability too, and has been used in the most comprehensive analyses of the impact of Covid and restrictions. The key is to implement restrictions that have the most favourable cost-benefit and are most acceptable to the public to ensure these benefits are actually realised. So it makes sense to now focus on interventions that cause the least harm and that people are most likely to comply with. The use of lateral flow tests is one example; so is improved ventilation in schools and workplaces. Of course we must not ignore the needs and concerns of those who are most vulnerable to Covid, including those with long Covid and especially people for whom vaccines don’t provide good protection. Fortunately there are things clinically vulnerable patients can do to protect themselves – including wearing well-fitted FFP2 or FFP3 masks, which have been shown in hospitals to greatly reduce the risk of infection. The arrival of antiviral drugs will also provide additional protection from hospital admission and death. We need to establish the effectiveness of any potentially harmful restrictions before they are brought in again, and ensure their benefits outweigh the harms. Most importantly, although restrictions are ending, Covid is far from over. With our new freedoms we all have the responsibility to continue to protect others and be considerate of those who are more vulnerable to the virus. Raghib Ali is a senior clinical research associate at the MRC epidemiology unit at the University of Cambridge