The Effect of Education on Age-Related Changes in Three Cognitive Domains: A Cross-Sectional Study in Primary Care

Abstract

The present study aims to investigate the protective effect of formal education on age-related changes in different cognitive domains with the hypothesis that it may attenuate the rate of decline. Individuals aged 50 years or older attending primary care physicians without known brain disease (431 participants, mostly [60.3%] female with 66.3 [±9.1] years of age and 7.7 [±4.1] years of education, on average), were evaluated with a neuropsychological battery including 28 cognitive measures. Cognitive domains identified by factor analysis were subject to repeated multiple regression analyses to determine the variance explained by age and education controlling for gender, depressive symptoms, and vascular risk factors. The slope of the regression equation was compared between two educational groups with an average of 4 years and 11 years of education, respectively. Factors identified corresponded to processing ability (Factor 1), memory (Factor 2), and acquired knowledge (Factor 3). Although education improved performance in Factors 1 and 3, it did not change the slope of age-related decline in any factor. This study suggests that in culturally heterogeneous groups, small increments in education enhance cognition but do not modify the rate of decline of executive functioning with age. These results contradict some clinical findings and need to be confirmed in longitudinal studies.

Key words:

• cognitive reserve
• education in aging
• screening battery

ACKNOWLEDGEMENTS

The authors thank Fundação Calouste Gulbenkian for sponsoring the present study (Project 0488). The authors are indebted to all participants and to GPs and Health Center Directors who collaborated in this study, namely Drs. Teresa Costa, Teresa Mota, Elisabete da Fonseca, Luís Afonso, Renato Graça, Cristina Galamba, Helena Febra (Centro de Saúde da Lapa); Cecília Cabral, Eugénio Oliveira, Paula Freitas, Neto Nogueira, Edite Branco, Helena Ferreira (Centro de Saúde de Alcântara); Luísa Romeiro, Rosário Braz, Teresa Libório, Óscar Miranda, Teresa Campos, Áurea Farinha, Isabel Santos, Nave Ferreira, Cristina Bastos, Rita Lourenço, Judite Viana, Manuel Rosmaninho, Luísa Costa, Isabel Santos, Bernardino Costa, Luísa Teixeira (Centro de Saúde de Oeiras); Elisabete Serra, João Reis, Maria José Heleno, Rui Cóias, Maria João Mendes, Sónia Pereira, Carla Coimbra (Centro de Saúde de Paço D'Arcos); Ana Paula Granadeiro, Rosa Oliveira, Analila Cruz, Vítor Cardoso (Centro de Saúde Moita); João Belbut, Manuela Ribeiro, José Luís Gomes, Susete Gomes, Maria José Rosa (Centro de Saúde do Barreiro); Jaime Torre, Miguel Santos, Luís Pinto, Paula Dias, Raquel Caetano (Centro de Saúde do Lavradio); Manuela Cruz (Centro de Saúde de Benfica); Graça Carneiro, Ana Maria Ferreira, Pedro Silva, Elvira Nunes, Paula Atalaia (Centro de Saúde de Alvalade); Maria José Galha, Emília Soares, Rosário Martins, Rogério Costa, Paula Costa, Teresa Neto, João Costa, Vitória Amaral, Carmo Velez, Luísa Santana (Centro de Saúde de Évora); and Fátima Portugal, Maria João Palma, Maria José Luís (Centro de Saúde de Sete Rios).

The authors also thank Profs. Maria Amália Silveira Botelho and Manuela Guerreiro and Dr. Sofia Madureira for their contribution in the discussion and organization of the evaluation battery.

Initial concepts and framework developed by Isabel Pavão Martins. Acquisition of subjects and data by Carolina Maruta, Claudia Silva, Pedro Rodrigues, Catarina Chester, Sandra Ginó, Vanda Freitas, and Sara Freitas. Data analysis and interpretation by António Gouveia de Oliveira, Isabel Pavão Martins, and Carolina Maruta. Preparation of manuscript by Isabel Pavão Martins, Carolina Maruta, and António Gouveia.

The sponsor (Fundação Calouste Gulbenkian) had no participation in any of the scientific steps of the study or in writing the article.

Notes

SD = standard deviation; CI = Confidence Interval; SDFR = Short-delay free recall; SDCR = Short-delay cued recall; LDFR = Long-delay free recall; LDCR = Long-delay cued recall; M = mean; WMS-III = Wechsler Memory Scale-Third edition; WASI = Wechsler Abbreviated Scale of Intelligence; MMSE = Mini-mental state examination; CVLT = 9 = 9-item version of the California Verbal Learning Test; F = female; M = male; T = t-test; X ²  = Chi-square test; ns = not significant at p < .05.

*highest level of significance.

CVLT-9 = 9-item version of the California Verbal Learning Test; SDFR = Short-delay free recall; SDCR = Short-delay cued recall; LDFR = Long-delay free recall; LDCR = Long-delay cued recall; WMS-III = Weschler Memory Scale-Third Edition; WASI = Weschler Abbreviated Scale of Intelligence.

Note: Regression analysis: Factor 1 was positively associated with educational level but negatively associated with age and depression. Factor 2 was associated with female gender but negatively associated with age. Factor 3 was positively associated with education level.

CVLT = California Verbal Learning Test 9-item version; SDFR = Short-Delay Free Recall; SDCR = Short-Delay Cued Recall; LDFR = Long-Delay Free Recall; WASI = Wechsler Abbreviated Scale of Intelligence; WMS–III = Wechsler Memory Scale-Third Edition.