Fe and iodine deficiencies are among the common forms of nutritional deficiencies globally and cause detrimental effects on mother and child. In developing countries, strategies to address these problems rarely use preventative dietary approaches. This cross-sectional study aims to ascertain whether dietary diversity score (DDS) and dietary patterns (DP) can predict the associated risk (AR) of Fe and iodine deficiencies in mother–toddler dyads during the dry/lean season in northern Ghana. Data were collected from women (15–49 years; n 118) and their toddlers (6–23 months, n 121). Dyads DDS and DP were calculated from their food intake recall. Multivariable logistic regression models were used to examine whether DDS, DP and socio-demographic factors predict higher odds of Fe and iodine deficiencies in dyads. DDS and DP predicted Fe and iodine deficiencies AR. Compared with DDS ≥4, women with DDS <4 have higher odds of Fe deficiency (ID) (adjusted OR (AOR) 1·82, 95 % CI 1·03, 3·21) and iodine deficiency (AOR 9·73, 95 % CI 3·19, 29·67). Two DP emerged in principal component analyses, ‘rural elites diet’ predicted ID risk in mothers (AOR 8·65, 95 % CI 1·76, 42·39), while ‘indigenous diet’ predicted iodine deficiency risk (AOR 11·41, 95 % CI 1·36, 95·97) in mothers only. The AR of Fe and iodine deficiencies in vulnerable populations may be predicted by diet-based measures in northern Ghana. We recommend further research into DDS and DP to predict Fe and iodine status of mother–toddler dyads in Ghana.