KEEP a tomato cool in a refrigerator and it will stay fresh far longer than it would at room temperature. Accidentally freeze it, though, and you will reduce it to a disgusting mush.

A similar problem plagues the storage of vaccines. About six in ten of those procured by UNICEF, the UN’s children’s fund, must be stored at a temperature between 2°C and 8°C. Generally, the focus of efforts to do this is on the top end of the range, with the establishment of “cold chains”, the links of which are refrigerators on the journey from factory to clinic, to stop vaccines overheating. Less effort is put into making sure a vaccine never gets too cold. But a vial of vaccine that has been accidentally frozen, and then thawed, may lose its potency as surely as one that has been warmed up.

A study published this week in Vaccine, by Celina Hanson of UNICEF and her colleagues, suggests that the overchilling of vaccines is alarmingly common. Dr Hanson and her team reviewed research that measured how often vaccines were exposed to temperatures below the lower limit. They combed through papers published between 2006 and 2015, and found 21 relevant studies conducted in 18 countries. Though not a representative global sweep, the studies in question covered both rich countries and poor ones, from several continents. Among the places they examined were America, China, India and a number of African states.

Intriguingly, the problem of overchilling was worse in the rich world than the poor. The papers Dr Hanson looked at reported that, on average, 38% of vaccine shipments in rich countries and 19% of those in poor countries had experienced temperatures that were too low. Regardless of a country’s wealth, about a third of its vaccine-storage units, which ranged from small refrigerators to huge cold rooms, were chillier than was safe.

Routine monitoring weeds out some frozen vaccines. Nurses in poor countries use a “shake and look” test to spot tell-tale crystals, for example. But the share of compromised vials that goes undetected and ends up in ineffective jabs is unknown. Studies that examine the consequences further down the line are rare, but those that exist suggest freezing matters.

According to one such study, which was conducted in America and published in 2011, places with a higher proportion of refrigerators with temperatures below zero also had higher rates of pertussis (whooping cough). A ten-year-old piece of research from Mongolia, where temperatures in winter can be as low as -55°C, found that children vaccinated against hepatitis B in winter months were more than twice as likely to be diagnosed subsequently with that disease than were those vaccinated in other months.

Another problem of vaccine distribution, “stock-outs”, is also the subject of a paper in this week’s Vaccine. Patrick Lydon of the World Health Organisation (WHO) and his colleagues analysed data from 194 countries that had been submitted to the WHO and UNICEF between 2011 and 2015. In an average year, a third of these countries had at least one vaccine out of stock at national level for a month or longer. Stock-outs were most common in sub-Saharan Africa, where bungled procurement and tracking of vaccines is common. But they were far from rare in Europe, as well. In an average year one European country in six reported a stock-out.

Mr Lydon and his colleagues did not collect data on how many children missed jabs as a result of stock-outs, so the consequence of such laxity is unknown. But childhood vaccination is important. The WHO calculates that vaccines already prevent between 2m and 3m deaths a year, but that this figure would rise by a further 1.5m if all children received the recommended jabs. That careless handling and careless stock-management are making this goal harder to achieve is a scandal.