mHealth refers to the use of mobile technologies within the medical and public health sectors. The challenge is to manage information collection, agregation and utilisation in a secure manner with or without a network connection.
The “Chez Soi d’abord” name comes from the “Housing first” program. This four-year programme was put in place by the interministerial delegation for housing (Diahl) and the public health laboratory EA3279 to evaluate a Canadian originated model which accompanies homeless individuals suffering from severe psychiatric problems to obtain a lodging. It is based on the idea that having a stable lodging condition is a prerequisite for working on the problems that led the individual to his/her current condition.
As part of a national project, Epiconcept put in place an information collection system based around netbook computers (small, inexpensive low-powered laptops) to collect physical and mental health information from the homeless. Every six months, for four years, the interviewers based in the four cities (Paris, Lille, Paris, Toulouse) and collect information from the individuals participating in the programme.
Both the Epiconcept team and the research team kept in mind the sensitivity of the information to be collected and the weighty consequences to the individual if he or she were to be admitted to the program or not.
The questionnaires were created using the Voozanoo 3 framework. There are an enormous number of questions (just a bit less than 1000 items) which can be asked depending on which branch of the study the individual is in and his/her stage of the study (check-up at 6, 12, 18 or 24 months). Apart from initial training, the questionnaires were made and put into Voozanoo by the Laboratory EA3279 team without the aid of Epiconcept.
When a new individual becomes part of the study, he or she is randomly assigned to one of the two branches of the study; those receiving the Chez Soi services, or those who can apply to the standard housing possibilities. For the statistical needs of the study, 200 individuals per city were initially recruited.
The data entry task is done directly on the netbook in the presence of the individual in the street. The collected data which is stored locally on the netbook is then sent to a server once the investigator has found a stable Internet connection. At the same time, while the netbook has an Internet connection, it receives a calendar of the next set of questionnaires to be filled out for the individuals the investigator is in charge of so that he or she can organise the other interview sessions for that week. The management of the Netbooks and the validation and exploitation of the centralised data is done at the server side by the Chez Soi team.
The directorate of national health system entrusted the programme classified as “experiment and research” to the research unit of the laboratory EA3279 directed by Professor Pascal Auquier. The programme is run by an interdisciplinary consortium of research teams in epidemiology, evaluative research, sociology, social psychology, anthropology and economy placed under the responsibility of the EA3279 research unit.
In order to allow the Chez Soi team to be as autonomous as possible in the creation of their bi-annual questionnaires, Epiconcept ran training session at their Marseille offices on using Microsoft Excel for that task.
Given the large volume of information to be collected in the four cities by the team of approximately 15 investigators, it was advantageous for the Chez-Soi team to enter the data directly on the computer in front of the interviewee and thus by extension, directly into a centralised database. The three principal advantages were
- large amounts of time were saved
- decreases in data entry errors
- better security assured
Also, because the population to be questioned are homeless people, the only logistically realistic thing to do was to interview where they were on the streets. There, an information system that was disconnected from the Internet, secure and centralised was required. This project is hosted on Epiconcept’s secure web-based platform certified by the ministry of health for hosting personal health data.
Multiple challenges came up during the conception of the laptop application. The randomisation step places the patient into one of the two groups of the study, the group that will benefit from the “Housing first” program or those who have access to the standard housing options (control group). The stakes for this study are thus considerable. For these patients who have significant mental problems, security, ergonomics and the question flow were the crucial points during the randomisation. The netbooks had to be as transportable as possible as well as be able to withstand the conditions of the street (summer and winter temperatures, falls, theft, and otherwise).
The project required approximately 90 days of work which included the software development and its management, general project management and the various meetings with the Chez Soi team. In addition to the training given for the correct method of creating questionnaire items, Epiconcept gave trainings to the investigators on using the netbook on the streets and also to the coordinators of the study for managing the fleet of devices and exploiting the collected data.
Voozanoo standard / custom development: 70/30
The majority of the work done was with the standard Voozanoo 3. The two customised elements of the application were firstly the transfer of data from the locally running Voozanoo on the laptop to the centralised database (even if the Wi-Fi connection was lost in midstream or the ensuing download of the calendar data was lost) and secondly, a system that interpreted the lines and columns of an MS Excel sheet into the XML representation of a questionnaire that the Voozanoo application natively works with.