- Background The overall implementation rate for outpatient comprehensive geriatric assessment (OCGAU) recommendations ranges from 48.6 to 71%. The purpose of the study was to identify factors that reduce the implementation rate of geriatric recommendations. Methods The medical records of patients who were assessed in the comprehensive geriatric assessment unit over an 8 year study period were surveyed. Data collected included patient's characteristics (socio-demographic, functional, cognitive, and affective condition, co-morbidity), number of recommendations, the identity of the geriatrician, and data related to the primary physician (age, sex, seniority, number of patients referred for geriatric assessment). Results Three thousand four hundred thirty-four recommendations were made for 488 patients (mean age 83.6 ± 0.6 years) of which 1,634 (47.6%) were implemented by their primary physician. In univariate analyses patients with an implementation rate < 25%, compared to patients with implementation rate ≥75%, had a higher Charlson Comorbidity Index Total Score (CCITS) (2.5 ± 1.9 vs. 1.8 ± 1.7, P < 0.05), a lower Barthel Index (82.8 ± 16.2 vs. 87.0 ± 15.3, P < 0.05), and a lower Instrumental Activity of Daily Living score (7.2 ± 3.5 vs. 8.2 ± 3.7, P < 0.05). There were no differences between these groups in other patient characteristics or the number of recommendations made during the assessment. Similarly, there were no differences in the identity of the geriatrician or the primary physician's characteristics. In the multivariate analysis only higher CCITS was associated with a lower rate of recommendation implementation by primary physicians. Conclusions There is a need to increase the implementation rate by primary physicians by increasing and strengthening the link with them and by further training in the field of geriatrics medicine. Trial registration The Helsinki committee of the Meir Medical Center approved the study (Approval #024/2015 [k]).