Jeroen van de Pol

56 Chapter 2.2 Strengths and limitations The characteristics of the community pharmacists who participated in this study are largely comparable to known characteristics of all community pharmacists in The Netherlands: 33% male compared to 46% nationally, 66% resident pharmacists compared to 72% nationally, 0.97 full-time equivalent (FTE) vs 0.89 FTE nationally [26]. Using a smartphone application to gather work-sampling data is considered user-friendly and an efficient way to attract participants to generate more data. The self-reporting aspect limits the “Hawthorne” effect, as participants feel less scrutinized than when being observed. This will limit behavioral changes in this study [27]. In previous research, it has been found that sensitivity analysis showed no major differences between responses given within an hour versus responses after more than an hour [11]. It has also been stated earlier that pharmacists have better understanding of their (current) activities than observers would have [28]. However, this type of research methodology also comes with disadvantages. Participants may provide socially desirable answers in using both the smartphone application, but also when providing background information (e.g. wanting to spend more time on CPS). However, it is expected that this effect is limited due to the provision of insights into time utilization and benchmark data. Also, using a smartphone application meant that participants had to keep their phone with them as much as possible, which could be undesirable in daily practice. Thereby, only pharmacists with good and coordinated workstreams may have participated, because only they may have had the time to participate. This would generate recruitment bias. A total of 65 initial participants did not provide background information which could be a result of this. However, participating pharmacists were provided with information about their own time utilization. This might have encouraged pharmacists who are struggling with their time utilization to also participate [11]. Also, the group of 65 excluded participants contained participants who may not have had the intention to fully participate in the study. Within this group a relatively high proportion downloaded the smartphone application but only used it a limited number of times. They installed the application and registered a few activities, but dropped out early. Participating pharmacists were provided with information about their own time utilization, including a benchmark of other pharmacists. We expect that this will have encouraged pharmacists who are struggling with their time utilization to participate [11]. Moreover, feedback and benchmarking is likely to stimulate honest reporting.