The global burden and threat of non-communicable diseases constitute one of the major challenges for development in the twenty-first century which undermines social and economic development throughout the world. A huge proportion of deaths due to non-communicable diseases, including cancer, occur in the low and middle income countries. Cancer is among the top five causes of mortality in Kenya. Its estimated annual incidence is close to 37000 and unfortunately close to 80% of the patients is diagnosed at an advanced stage. The economic impact of cancer is felt at an individual and most importantly a population level. Shortening the interval between onset of symptoms and initiation of treatment would greatly improve patient outcomes and also reduce the cost of more aggressive treatment. Understanding the pathways to care is critical for interventions to be made in a timely manner. An assessment of patients‟ perspectives guides strategic decision making to meet expectations and effectively manage health care performance. The main aim of this study was to identify the factors associated with late cancer diagnosis among patients presenting for treatment at Kenyatta National Hospital. A descriptive cross-sectional design was used for this study. The study was undertaken at Kenyatta National Hospital‟s Cancer Treatment Center. Standardized questionnaires were administered to all eligible and consenting patients with either stage three or four of cancer presenting for treatment at the cancer treatment center. Patients‟ records were also reviewed to ascertain the diagnosis, stage and type of cancer. The data collected was summarized and analyzed using Microsoft Excel and MINITAB 14. Female respondents were more than the male respondents. The mean age of the respondents was 49 years; 87% of the respondents were below 65 years old. 71% had attained only primary education, 10% had attained tertiary education. Most of the respondents (74%) resided in rural areas and 84% of them were in the informal sector. 87% had no pre-existing illnesses and only 23% were smokers. 65% of them had no prior knowledge on cancer, 71% were unaware of measures to reduce the risk of cancer and only 19% had been previously screened for cancer. 78% of the participants had experienced their symptoms for more than 6 months before the diagnosis of cancer was made. 58% of the participants initially sought help for their symptoms in private hospitals and 45% had their diagnosis made in private hospitals. 55% had their diagnosis made in secondary and tertiary public hospitals. 61% waited for more than 6 months before iv getting their diagnosis of cancer and a similar proportion had to wait for more than 3 months before initiation of treatment. Long waiting time for specialized treatment was identified as the main reason for delay. The findings of this study corroborate those of other similar studies. The success of any cancer control programs will require well-coordinated demand-side and supply-side efforts to address the barriers to early detection and treatment of cancer.