DOING A PILOT STUDY: WHY IS IT ESSENTIAL?
4. Data entry and analysis
A basic descriptive statistical analysis was done. Out of the 20 subjects, 12 were females and all were from the Malay ethnic group. Six subjects had insomnia. Two of them experienced sleepiness during the daytime but did not meet the criteria for ‘excessive daytime sleepiness’ based on the Epworth Sleepiness Scale.9 None of them had anxiety or depression. No difficulties were encountered in conducting the analysis.
This pilot study has demonstrated that the study protocol is feasible. The project did not appear to be too disruptive to the clinic or have a significant impact on staff time. It also seemed to be acceptable to patients in the waiting room.
It was possible to recruit patients based on the inclusion criteria of the study, although by chance only Malay patients were included. In order for the sample to be more representative of primary care attendees and other ethnic groups, both government and private GP clinics of different geographical locations (north, central, south) as well as doctors from different ethnic groups will be invited to participate in the full study.
This study has demonstrated the effectiveness of a pilot study in identifying flaws in a questionnaire that after appropriate amendments can then be utilised in a full study. It has also provided a better understanding of how to implement the survey; in this case, the research assistants were required to provide patients with occasional help with the questionnaire and to check item completion. The use of more than one method of data collection, i.e. interviews and self-administered questionnaires is called "triangulation".
Data entry was not problematic, although in the final study a statistician will be involved with data analysis. Pilot studies can also be used to help calculate sample size by providing data about the likely responses to questionnaire items.The steps in this pilot study demonstrate the benefits and the methods of this important phase of a research project. While studies using different designs, such as experimental
ones, will involve pilot testing of other instruments and protocols, the general principles apply. The lessons learned in this particular study can be applied elsewhere (see Box 1).
The decision to conduct a pilot study prior to embarking on the main research project can be a difficult one for researchers. Sometimes it is tempting to omit this step, especially if the main study has been reasonably well planned. Constraints of time and a rush to get on with the main study are common reasons for passing over pilot work. However, this approach is risky, as no matter how thoughtfully a study has been planned, there are likely to be unforeseen difficulties, as this paper has shown. The investment in time and resources is generally worthwhile.
We wish to thank the Malaysian Medical Association Foundation (MMAF FD RES/05) and the International Medical University of Malaysia (IMU 098/2005) for their funding of the project and to the GP clinics, doctors and staffs who participated in this pilot study.
- Stewart PW. Small or pilot study, GCRC protocols which propose "pilot studies". Cincinnati Children's Hospital Medical Center. [HTML]
- Lancaster GA, Dodd S, Williamson PR. Design and analysis of pilot studies: recommendations for good practice. J Eval Clin Pract. 2004;10(2):307-12 [PubMed]
- Kraemer HC, Mintz J, Noda A, Tinklenberg J, Yesavage JA. Caution regarding the use of pilot studies to guide power calculations for study proposals. Arch Gen Psychiatry. 2006;63(5):484-9 [PubMed]
- Simon S. Steve attempt to teach statistics. Children’s Mercy Hospital and Clinic. [HTML]
- Shochat T, Umphress J, Israel AG, Ancoli-Israel S. Insomnia in primary care patients. Sleep. 1999;22 Suppl 2:S359-65 [PubMed]
- Terzano MG, Parrino L, Cirignotta F, Ferini-Strambi L, Gigli G, Rudelli G et al. Studio Morfeo: Insomnia in primary care, a survey conducted on the Italian population. Sleep Med. 2004;5:67-75 [PubMed]
- Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361-70 [PubMed]
- Khoo EM, Mathers N. Somatisation disorder in primary care attenders: a cross cultural experience. 17th World Conference of Family Doctors in Orlando, USA in October 2004. [abstract]
- Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14(6):540-5 [PubMed]