So, inrush is mostly determined by the amount of input capacitance on the power supply (which also determines hold-up time, the amount of time it keeps operating after power is lost). Inrush is really very loosely specified by the ATX standard, which is probably why it varies so much.
Other than reducing the cap size, the way it's usually controlled is with a thermistor on the input. The thermistor starts at high resistance and drops as it gets warmer. Higher inrush means you need to size/rate components that see that inrush accordingly. But, just using a larger thermistor has some downsides too. You can also get fancy and put in a by-pass around the thermistor once the cap is charged, but that's extra cost.
These things get designed by finding the most cost effective balance of parts that does the job. Which is why I'd be comfortable recommending a supply from a reputable maker despite higher inrush. There I trust they took all the relevant factors into account, and if the higher inrush meant a fraction of a watt power savings, or a few cents off their BOM cost, I'm fine with that. It's always possible they got something wrong, but they've earned the benefit of the doubt. On a cheapo unit it's a potential red-flag as they may have just cut corners to keep the cost down.
