How many EMA-workshops are needed to collect a representative sample of events in a hospital ward?

The effect modifier assessment (EMA) method (Edwards & Winkel, 2016) is a method for assessing the impact of an intervention and modifiers on a desired outcome e.g. improved work environment. The EMA-method captures events (a change in work) in a ward and for each event asses 1) impact on work environment and 2) if the event was part of the intervention or not. The EMA-method rely on the EMA-workshop – a structured group interview method inspired by the chronicle workshop (Limborg & Hvenegaard, 2011) to collect data. However, healthcare organizations are complex and staff carry out many different and diverse tasks. This poses a problem when using the EMA-method and raises the research question of this abstract: How many EMA-workshops are needed to generate a representative collection of events in a ward? Methods Six EMA-workshops each with a full surgical team of six people was conducted in a heart surgery ward with 150 employees (Edwards & Teewes, 2015) . The collected events from all workshops was analyzed and grouped into themes. Data was considered representative when the next workshop did not produce any new themes (saturation). Results In the test case, most if not all employees focused on surgery. The ward was organized in three specialties: Heart surgery, Lung surgery and Child heart surgery. Events differed between specialties and therefor it was expected that saturation would be reached after minimum three workshops. The heart center is comparable to other surgical units and will exhibit a lower degree of variation in work tasks and processes than a medical ward. The general recommendation is that each EMA-workshop include participants from all relevant occupational groups for a specialization. A specialization should be understood broadly as they may not be formally defined but have developed over time e.g. a specific patient group. The number of needed EMA -workshops depend on the number of specializations and researchers should uncover the extent and number of specializations before deciding the number of workshops. The number of EMA-workshops is recommended to be at least equal to the number of specialties relevant to the study.