The ideal measurement or experimental test of a claim would have several properties. First, the test would distinguish between the claim being true and the claim being false -- the claim being true would give different results from the claim being false. (This should go without saying. Any kind of test must satisfy these results; otherwise it's not a test.) Second, there should be wide ranges for both the true and false results, to allow for random fluctuations, false positives and negatives, and so forth. Third, there should be a wide gap between the true and false result, to ensure that they are distinguishable -- no borderline cases, no interpretive debate.
In other words, there should be three regions: the "true" result, the "false" result, and a "no man's land". If a "no man's land" result occurs, something was very wrong with the test, although boundaries can be blurred without harming the test.
Christian Science and its practitioners make strong claims about the efficacy: "Divine Love always has met and always will meet every human need" (Science and Health quote seen on most Christian Science churches), "Christian Science was based on the law of God that never fails", "Don’t you as the practitioner doubt or question the healing work. Christian Science never fails." Death is illusion, and raising the dead and preventing death are among the claims made in Christian Science.
Therefore, consider adults who became Christian Scientist before Mary Baker Eddy's death in 1910, and take the percentage of those who are still alive now. Christian Science is true if that number is greater than 50%, and false if that number is less than 1%. Between 1% and 50% is the no man's land, and should not occur.
The result of the test is obvious. It's also obvious that believers in Christian Science will reject the conclusion, through reams of double-talk that never address the issue.
My own observation of everything they say is that Christian Scientists are incompetent in evaluating their success, and even determining whether a healing occurred. (Notice the irony of the date I chose.)