Polypharmacy (≥5 drugs) increases the risk of discrepancies between patient- and general practitioner (GP)-reported drugs, leading to adverse outcomes. This explorative analysis assesses the agreement between patient- and GP-reported drugs under the influence of a paper-based patient portfolio in a pilot cluster randomized controlled trial (cRCT). Complete data were available for 68 patients aged 65 or older (26 were female), with multimorbidity, polypharmacy, and at least one hospitalization in the past year. Agreement was assessed for drug name and strength level. Differences between the intervention and control group (IG/CG) and comparisons between two time points (six-month interval) stratified according to gender were analyzed using Wilcoxon and Mann–Whitney U tests (α = 5%). To evaluate the reasons for discrepancies, the agreement of active pharmaceutical ingredients (APIs) and anatomical therapeutic chemical (ATC) groups was analyzed. At baseline, the agreement was 72.1% for the IG and 73.9% for the CG. Inclusion of the reported drug strength reduced the agreement in both groups (IG 66.7%, CG 60.0%). Agreement for the IG decreased statistically significantly after six months (−5.4%). ATC groups B, C, and H had the highest agreement, while N, R, and Z had the lowest. Large discrepancies in the drugs reported, due to the APIs and the corresponding ATC group, were observed.