The potential to treat diabetes by increasing beta-cell mass is driving a major effort to identify beta-cell mitogens. Demonstration of mitogen activity in human beta cells is frequently performed in ex vivo assays. However, reported disparities in the efficacy of beta-cell mitogens led us to investigate the sources of this variability. We studied 35 male (23) and female (12) human islet batches covering a range of donor ages and BMI. Islets were kept intact or dispersed into single cells and cultured in the presence of harmine, glucose, or heparin-binding epidermal growth factor-like growth factor (HB-EGF), and subsequently analyzed by immunohistochemistry or flow cytometry. Proliferating cells were identified by double labeling with EdU and Ki67 and glucagon, c-peptide or Nkx6.1, and cytokeratin-19 to respectively label alpha, beta, and ductal cells. Harmine and HB-EGF stimulated human beta-cell proliferation, but the effect of glucose was dependent on the assay and the donor. Harmine potently stimulated alpha-cell proliferation and both harmine and HB-EGF increased proliferation of insulin- and glucagon-negative cells, including cytokeratin 19-positive cells. Given the abundance of non-beta cells in human islet preparations, our results suggest that assessment of beta-cell mitogens requires complementary approaches and rigorous identification of cell identity using multiple markers.