Most of the information we get about how a company is running its business, in any market, comes from the company. If the information is related to financial statements, it is checked by an external audit, and then provided to the public through official channels. All of these controls are meant to make it very unlikely for a firm to commit fraud or to cook its books. In that sense the controls are successful, very few firms provide fraudulent information to the public compared with the thousands of companies that list in stock markets around the world.
Now, there is still a handful of firms that have committed fraud, and it is probable that a few firms are committing fraud right now. But, these companies go to great lengths to keep information about their fraud hidden from both the public and the authorities. All of these factors contribute to such frauds being black swan events to the outside observer.
A black swan event is an event that is highly improbable, impossible to foresee with the information available before the event (it can only be analyzed in retrospect), and it has very large impact. The classification of an event as a black swan depends on your perspective. E.g. the Enron collapse was not as unexpected to the Enron executives as it was to its investors.
You cannot foresee black swan events, but there are a few strategies that allow you to insure yourself against them. One such strategy is buying out of the money puts in the stocks where you have an investment, the idea being that in the event of a crash - due to fraud or whatever other reason - the profits in your puts would offset the loses on the stock. This strategy however suffers from time and loses a little money every day that the black swan doesn't show up, thanks to theta decay.
So while it is not possible to detect fraud before investing, or at least not feasible with the resources and information available to the average investor, it is possible to obtain some degree of protection against it, at a cost. Whether that cost is too high or not, is the million dollar question.