Do subjects investigated for occupational asthma through serial peak expiratory flow measurements falsify their results