1) Enrollment of subjects. There was a significant lag time between the initial training (December 2002-February 2003) of the visiting nurses selected to be in the study and the actual enrollment of subjects (April 2003). Because of this delay, we had to repeat some of the training and orientation sessions. The orientation sessions were essentially "buy-in" meetings to confirm the willingness of each nurse to participate in the study. To accomplish these sessions, it was necessary to schedule and often re-schedule the training classes and orientation sessions to accommodate the busy work day of the VNA nurses, in our study called Visiting Nurse Investigators (VNIs) --who were spread out across Northern Virginia, DC, and Prince George's and Montgomery counties in Maryland--as well as holidays, sick leave, delays due to weather and traffic, etc. Planning, scheduling, and completing all these meetings simply took longer than expected as the activities associated with participating in this study were "add-ons" to the regular work schedule for the VNIs. These delays were further compounded by some VNIs in groups A, B, and C leaving the VNA. These individuals were replaced with randomly selected backup nurses who then had to go through all the standard orientation and training sessions. In addition, although the agency continued to have a large volume of wound care cases, these cases may not have entered the study either because they did not meet the study criteria or because they did not live in the VNIs' geographic unit and were therefore not assigned to the VNI;

2) Technical issues. The technical issues turned out to be more numerous and, frequently, more difficult to solve than we had anticipated. Although the cameras we used (Nikon Coolpix 4500), were relatively easy to use -- especially in the automatic mode -- not everyone was equally gadget tolerant. While some visiting nurses were taking high-quality images after only a few training sessions, others required multiple sessions and, even then, their images were not always satisfactory without further technical adjustments and problem solving. Another task that took more time to accomplish than anticipated included uploading the images to the laptop. This required modifications of the software program and ensuring that images were identified and stored in the proper file. Finally, transmitting the images from laptops to the WCSs computers took more time than anticipated (the visiting nurses had to log on to the MedStar LAN in a MedStar office) and performing the consults in a timely manner was often difficult for the busy VNIs; and 3) Human factors. Some of these ‘human factors' have already been alluded to in the obstacles described above so only a few additional ones will be mentioned here. A very prominent issue was the fact that the visiting nurses were being asked to play a research role in addition to their usual role as a home health nurse. Thus, they were wearing two hats, instead of their customary one. In this new role, they were the researchers' surrogates and carried the main burden of collecting the data and making the study successful: they needed to explain the study, answer questions, obtain informed consent, and carry out all the tasks required of them by the protocol depending on which group they belonged to. These tasks were often made more difficult due to the elderly nature of the population, the potential subjects' inability to talk to the primary researchers, and the requirement that they read a 7-page consent and a 2-page HIPAA form.

The research team deployed a number of strategies to surmount these barriers. For example, some of the delays described above -- especially the turnover in nurses -- were anticipated. As a result, when investigators did the original random selection of nurses to be in the study, they purposely selected six or more per group instead of just four. Therefore, when a selected nurse dropped out of the study because he/she was leaving the VNA, we had a replacement nurse available to take his/her place. Another strategy, implemented at the beginning of the study, to make participation in the project more attractive was to offer financial compensation to the "pay- per-visit nurses" whose productivity would be affected by their participation. In addition, because of the importance of the study to the VNA, the Agency agreed to absorb the cost of the salaried nurses (especially in groups A and B) whose productivity would also be adversely affected during the course of the project. In addition, the VNA offered low cost incentives or ‘prizes' to the nurses who enrolled the most subjects per month and held special ‘thank you' receptions for all participating nurses and staff. These small gestures helped improve moral and maintain interest in the study. Another very effective strategy to overcome some of the technical and human factor barriers alike, involved having the engineer on the project, Ms. Donal Lauderdale, accompany each of the Group A nurses during a home visit. This permitted the engineer to make on-the-spot observations, do instant troubleshooting, and on-site training with the cameras and laptops.