The impact of different definitions and reference groups on the prevalence of cognitive impairment: a study in postmenopausal breast cancer patients before the start of adjuvant systemic therapy.

OBJECTIVE: Several prospective studies into the effects of adjuvant systemic therapy on cognitive functioning suggest that a proportion of breast cancer patients show cognitive deficits already before the start of systemic therapy. Owing to, among others, methodological inconsistency, studies report different rates of this pre-treatment cognitive impairment. We examined the impact of four different criteria of cognitive impairment and two types of reference groups (a study-specific healthy reference group versus published normative data) on the prevalence of cognitive impairment.
METHODS: Two hundred and five postmenopausal breast cancer patients underwent a battery of neuropsychological tests before the start of endocrine therapy, 124 healthy subjects underwent the same tests. Proportions of cognitive impaired patients were calculated for each of four criteria for cognitive impairment, using (1) study-specific healthy controls and (2) published norms of healthy controls as reference groups.
RESULTS: The prevalence of cognitive impairment varied greatly with the strictness of the criterion, as expected, but also was dependent on the reference group used. Cognitive impairment, relative to published norms, ranged from 1% for the strictest to 36.6% for the less strict criterion, cognitive impairment relative to study-specific healthy controls, ranged from 13.7 to 45.4% for the same criteria.
CONCLUSION: This study highlights contrasting proportions of cognitive impairment by using different criteria for cognitive impairment and different reference groups. (Dis)advantages of the methods using a criterion for cognitive impairment, and of the use of published norms versus a study-specific reference group are discussed. Copyright 2009 John Wiley & Sons, Ltd.