2.1 Data Analysis

Each of the components of the needs assessment listed above was analyzed and reported both separately and collectively. A summary of findings and recommendations based on these components is included in this report. The interviews were structured and analyzed using a grounded theory approach. The constant comparative method was used to refine the interview guide and begin analysis as the interviews progressed. Data were transcribed from interviewer notes and recollections into a data table, with 1 column for each person interviewed. Each question in the interview constituted a row in the table, with the responses filled in for each interviewee. The rows of questions were grouped into blocks of questions with common subject matter. Each of these blocks of questions/responses was synthesized down to an archetypal response to each question. The interview responses were then compared in order to derive significant themes and issues. Demographic groups were compared for similar and dissimilar data. Interview summarization included archetypal responses and key verbatim responses.

Continue Reading

The results of the survey are presented descriptively and graphically to support the findings and recommendations. The responses in total are contained in the appendix. Initial analysis suggested a number of questions to be answered by sub-analysis. These questions were answered by standard statistical testing using commercially available statistics software. Correlations and relationships between demographic variables, confidence, knowledge gaps, practice gaps, and barriers to best practices were explored.

A number of tests were done using Pearson product-moment correlation. These tests were used to determine whether certain responses varied with each other, suggesting a relationship. An overall master analysis utilized triangulation to compare data from different components and develop overall findings and recommendations.

We used repeat sampling to ensure a data set that is representative of the target audience and results that are generalizable to that audience. All data‐collection techniques and processes adhered to best practices around human research subject protection. Data are stored on a locked and password‐protected computer, accessible only by authorized study researchers. No private health information was collected. Personal information was collected only for purposes of reimbursing interviewees for their time. This information was separated (de‐identified) from the study data. All study data were reported in aggregate fashion only. All data that have been collected will be de‐ identified and made available on request for all legitimate educational purposes.

3.  Results

3.1   Demographics

In-depth interviews were completed with 36 clinicians: 28 physicians and 8 nurses/nurse practitioners. Fifteen of the clinicians were from academic medical centers and 21 were from community settings. The profession of the interviewees appears in Table 1.

Table 1.  Interviewees by profession.

In-depth Interview Participation


Number of Participants

Medical Oncologist


Radiation Oncologist


Surgical Oncologist




Oncology Nurse Practitioner/Nurse


Several themes were identified during the interviews that aided with survey development and the analysis of results. Each of these themes is discussed below.

There is a desire for more guidance on treatment for geriatric patients with comorbidities. Treatment guidelines do not specifically identify where or how adjustments need to be made for patients based on frailty. Clinicians approach therapy differently with these patients, some “starting low and going slow,” while others use full dose and adjust when side effects occur. In particular, familiarity and dosing were the concerns in using new treatments for geriatric patients with cancer.

Oncologists rarely use available tools to assess a patient’s functional, cognitive, and support status. This results in no baseline measurement and the inability to document a change in these metrics. The clinicians indicated “we all do it,” when referring to conducting a functional assessment, but when questioned, admit that they do not use objective tools to assess the patient.

Compliance is an issue in geriatric patients. This is due to a variety of factors, including both intentional and unintentional causes. This was cited as one of the major problems that exists in treating these patients.

Communication with primary care is an issue. Many of the clinicians reported that better communication with primary care would improve the care for geriatric patients with cancer.

Goals of care are the same for all patients regardless of age. However, with geriatric patients with cancer, quality of life may play more into the patient’s treatment decision. The clinicians also reported being more likely to accept an older patient’s refusal of treatment, and option for palliative care.