In this fourth installment of our series on oncology quality studies we will talk about conducting the study according to the planned methodology or quality measures and how to organize the data collection process.
The methodology is the roadmap, or strategy, that describes the way the study is to be conducted and identifies the methods to be used during the process. The methods define the way the data is collected and how specific the results need to be for calculation and analysis. The methodology also specifies who will participate in the study, how, when and where data will be collected.
Various decisions and considerations need to be addressed in the methodology including the population or sample that at the core of the problem. For example, in a study the population might be defined as all male colon cancer patients between the ages of 18 and 69 years of age with Stage III non-resectable tumors who are receiving chemotherapy as their first line of treatment. In this example the population is specific and includes only the patients that are relevant to the identified problem.
All the steps to be completed in the study must be spelled out in the methodology. This includes what data will be collected, when, and how it is collected, who is collecting it and how the data will be grouped, categorized, and used for analysis. The study plan should describe the expected scenarios and actions to be taken to ensure that the team has thoroughly reviewed all the data and walked through the various situations or decisions that were pre-identified or discovered during the data collection process.
After the roadmap is complete, it is time to collect the data. This is a critical step in gathering information needed to objectively answer the question(s)s raised in the problem statement. Every quality study includes the collection of some type of data – whether it is from the literature, medical record, Cancer Registry case abstracts, manual or electronic processes, flow diagrams, or even directly from patients or study subjects themselves.
There are diverse means for collecting study data. Team members may observe, code, or take note of certain actions, thresholds, workflows, or processes, administer tests or questionnaires, skills inventories, conduct interviews, online surveys, or review official documentation such as the medical record. For example, the study team may collect certain variables related to physical characteristics of the sample population at the time of their initial consult or at a pivotal visit, such as height, weight, race, family, or past medical history. In another example the team may conduct a physical walk through to identify the steps taken as part of a process. Or they may identify dates to calculate the duration of time elapsed between two distinct data points, such as date of screening mammography and date of diagnostic mammography.
At a minimum, two sets of data are collected for each quality study. The first is collected at the beginning of the data collection period and the second is collected at the end. Depending on the type and scope of the study and how it is defined in the methodology, additional data collection points may be necessary. For example, if a weekly huddle is held to determine chemotherapy patients at high risk
of needing emergent care over the weekend is the primary data source, then weekly data collection for a specified number of weeks may be appropriate.
As data is gathered it should always be recorded in a tangible format either in print or electronically. The tools used may include, a spreadsheet, data collection form, tally sheet, tracking log, flow diagrams, or databases.
When the study team is satisfied that they have collected sufficient data to meet the specifications outlined in the methodology, and the volume of data accurately represents the sample population, then they are ready to move to the next phase.
In part five of this series, we will look at data analysis and steps that are taken to use the data collected to make informed decisions or steps to be taken to resolve the problem.
To see other installments in this series, click on the link below:
Leave a Reply
Your email address will not be published. Required fields are marked *