When reviewing the age limits for pilots, ICAO consulted with all member states and the Aerospace Medical Association (AsMA), the International Academy of Aviation and Space Medicine (IAASM), the International Air Transport Association (IATA), and the International Federation of Airline Pilots' Associations (IFALPA). It's interesting that IFALP was the only one in favour of continuing with an arbitrary age 60 restriction and indeed the AMA commented that there was insufficient medical evidence to support restriction of pilot certification based on age alone.
The following link shows the replies received by ICAO before deciding to make a recommendation to review the age upwards to 65 (whilst commenting that it is still arbitrary and due to pressure from the IFALPA):
Review of Replies on Proposal to Increase Pilot Retirement Age
The final conclusions are quite enlightening:
The Secretariat agrees with United Kingdom and United States that statistical evidence cannot be used to predict with certainty the medical future of the individual pilot. Such evidence only shows what risk group the pilot belongs to. An applicant, whose medical examination indicates that he/she belongs to the group of young and healthy persons, can safely be certificated although it is impossible to predict the medical future of the individual applicant. All that can be said is that he/she belongs to a statistically defined risk group where the probability - on average - of any incapacitating event is very low. For many years now, statistical grouping of applicants for medical certificates has been considered a reasonable way of evaluating and assessing those who do not fully meet the medical requirements for certification. Although it is impossible to predict the future for an applicant with a certain medical condition, it is possible, often with a high degree of certainty, to predict the outcome for a group of such applicants. Many Contracting States, accepting this viewpoint, have for decades certificated applicants with medical conditions that place the individual pilot in a group that has a higher than normal risk but still - on average - no greater than one percent probability per year for developing an in-flight incapacitating event (this is often referred to as "The 1% Rule" - when the 1% Rule is applied to airline pilots, the licence is limited to multi-crew operations). The experience accumulated in these States supports continued application of the 1% rule. Worldwide experience of fatal accidents indicates that physical incapacitation in two-pilot aircraft operations poses little risk to flight safety since the second pilot takes control and operates the aircraft to a safe landing. In one Contracting State, no accidents arose from 127 reported in-flight incapacitations in 19 million flying hours over a ten-year period. Since 1980, as far as the Secretariat is aware, there is no recorded fatal accident in the world with a two-pilot airliner that has had cardiovascular incapacitation cited as a contributory cause despite the occurrence of many such incapacitations during this period.
If certification of pilots, based on their membership of statistically defined risk groups, is acceptable for younger pilots with certain medical conditions, it is also acceptable for healthy pilots who because of their age belong in a higher risk group. The comments from Azerbaijan, Finland, Libya and Mexico all express confidence in the safety of a higher age limit for airline pilots and even that older pilots may enhance flight safety. Gabon points out that the rate of ageing varies from one person to another. The Secretariat agrees fully. This may be considered an argument for having more comprehensive and more frequent medical examinations of older pilots, as proposed by Argentina. The Secretariat has found no evidence that more frequent or more comprehensive examinations are required for the sake of flight safety and agrees with United Kingdom that the content of additional tests, if indeed necessary, still has to be determined. Australia mentions that national law does not permit age discrimination. Legal systems usually provide for safety exceptions to the general principles of non-discrimination. As the upper age limit is a safety specification, it may be assumed that it could make exception to the anti-discrimination laws and principles that may exist in some States. Lesotho points out the fact, most likely the case in many States, that available national data do not provide statistical information on which an upper age limit can be based. The Secretariat, however, bases its opinion on data compiled from sixty-four States, reflecting accumulated experience with well over 3 000 older pilots, totalling at least 15 000 pilot-years. These data indicate that a higher upper age limit is compatible with safe flying.