• Category:
  • Document type:
  • Level:
  • Page:
  • Words:

Qualitative Evaluation of Mobile Diabetes Information System

Qualitative Evaluation of Mobile Diabetes Information System


Information system experts increasingly favor qualitative research methods in evaluating the causality in use case scenarios, system performance, possible user behaviour control and varied system execution contexts. The healthcare sector also rely on information systems to improve patient care, health administration and education, maintenance of medical records and improvement of patient diagnosis. Diabetes management can be improved by developing an integrated information system that incorporates varied players in healthcare, including the patient, to reduce treatment response times and effect treatment adjustment. Such an information system can then be evaluated using qualitative research methods to ascertain the extent that it meets the expected healthcare goals initially intended. This report presents a qualitative research evaluation of a diabetes information system using a methodological approach anchored on the evaluation objectives. It starts by presenting the background of the diabetes information system to be evaluated.

System Background

The diabetes information system is a self-monitoring system for blood glucose levels in patients. Data collected from the patient is used to inform decisions to continue or change treatment during visits to diabetes specialist. The biometric data collected is then transmitted securely over mobile phones to some central server. These data can then be interpreted and used to develop a treatment plan that is then sent to the patient to improve diagnosis. The system has patient stations, medical control center, physician stations, medical administration stations and system maintenance stations. The medical control has a master station – that processes client requests and updates a database — and a system information database that stores information on a health database. Such information may include medications, operations records, user account information and health measurements. Reliability and high availability are key system requirement for the health information database (Lee & Baskerville, 2003).

The patient station is meant to acquire data from the patient and provide a portal for operation. Information like blood glucose levels, blood pressure levels and weight measurements can be recorded in such stations. Bluetooth technology allows sending this information to the medical control center. Physician stations enable physicians to access system functionalities such as acquiring patient’s medical history, managing patients’ accounts and medical treatment schemes and suggesting treatment options. Medical administration stations are meant for medical administrators to edit and update medication protocols, manage accounts for physicians and supervise patient therapy procedures. Notably, the administrator oversees the patients’ accounts information. A system maintenance station allows the performance of maintenance procedures either remotely or locally. Such procedures include checking user operation records, system operation status, and managing medical administrator accounts.

.Kaplan & Maxwell, 2005). Quantitative evaluation of information systems within the healthcare context may concentrate on issues such as error rates, rate of use, physician behaviour changes, benefits and costs, user satisfaction, timeliness, and ease of information retrieval. Such measurements are discrete, objective, stable and relatively independent (Berg, 2001)Qualitative research methods for evaluating information systems such as the one described above are normally inductive in nature. They probe the information, the user, the organization and the information technology used. They help in understanding the user perspectives, execution contexts, causality in processes, organizational influence on information system design, and varied use case scenarios (

. They help in understanding the meaning of system events, their associated events and processes in their execution contexts. Social, political, organizational, and cultural concerns associated with the system, the system development processes and the users’ perspective on all these issues are captured in qualitative evaluations.Mingers, 2001) Contrary, qualitative research methods applied when evaluating information systems inherently examine the dynamics involved in the processes executed instead of the individual notable characteristics of the system (

. Kjeldskov & Graham, 2003). To evaluate the designed diabetes information system described above using qualitative research methods, we will administer online questionnaires to every user. This will create an understanding on the impact that the diabetes self-monitoring system has on the diagnosis and treatment of patients, and diabetes management. Interviews, open-end questionnaires, and observations by the system users may help determine the incremental value of the diabetes information system on patient care and disease management (Friedman & Wyatt, 2006)It is important to decide upfront the qualitative method that can be used to evaluate an information system. Every method should be chosen based on the value they add and the answers they are likely to capture (

. For purposes of our qualitative system evaluation, we will seek to know the users’ perception of the mobile diabetes information system; the contexts under which process change is accepted; and the context that the information system was developed for. The study intends to capture the user explanations about system events, processes, and the outcomes of such processes in relation to the mobile diabetes information system. The questions probed by this qualitative evaluation of the mobile diabetes information system include:Petter, DeLone & McLean, 2008)The research questions and the intended evaluation goals should also be established before the actual evaluation process (

  • What is happening in every process and segment of the information system?

  • Why is action happening within the process?

  • What causes the action to happen?

  • What so the system users think the action happening within the processes are meant to achieve?

  • What is the user response to every action within the processes in the information system? Why do they respond in a particular way?

Open-end questionnaires would show the different perspectives help by the medical administrators, patients, physicians and system maintenance persons. A sample questionnaire administered to the system users in the evaluation of the mobile diabetes information system is presented below.

Qualitative Evaluation Questionnaire

Mobile Diabetes Information System



  1. Do you find it easy to navigate the mobile diabetes information system using your mobile phone? Which operations are hard to complete? Explain.

  2. What do you understand the term hyperglycaemic to mean?

  3. What do you think the data on blood pressure, blood glucose and BMI measurements is used for?

  4. What is the importance of the ‘Latest ADAP’ figures on one’s diabetic condition? Indicate whether lower or higher levels of the figures indicate stabilization and wellness in relation to diabetes.


1. Do you find the system enhancing you in delivering your tasks? If so, how?

2. What merit do you see in accessing the system remotely?

3. Do you find the system comprehensive enough in aiding the patient diagnosis and medical prescription process? How can it be improved?

4. Do you think the patient data collected need to be processed automatically to give more direct information on the specific diabetic status of a patient. If so, which particular information need further automatic analysis?

5. Are the current medication protocols convenient enough to work with? Which areas need to be improved?


  1. The input of the patient in determining the time for taking medical actions leaves much of the medication in the hands of the patient, depending on their lifestyles. The doctor also has some say on the timelines for taking diabetic medication. What is your take on the time aspect of diabetic management?

  2. How has the mobile diabetes information system enhanced your work? How can it be improved?

  3. Which step do you take when a medical prescription runs out of stock amidst urgent pending requests?

  4. Is the information system helpful in tracking the medicines in stock? How can this aspect be improved?

  5. What occasions may necessitate the change of an initial medical actions specified for a particular patient? How best can this be handled?


1. What is the optimal uptime for the master station?

2. What is the optimal uptime for the health information database?

3. It is anticipated that the system would handle multiple requests from users at the same time. What measures are in place to ensure high system reliability, fast performance and high availability?

4. Which user permissions need to be changed (enhanced or reduced) to improve system performance and process integrity? Explain.

Physicians would give varied views on their perception about the mobile diabetes information system. Some would see that the system improves the speed and ease of tracking medical operations, provides adequate information for educating patients on diabetes control and management. Patients may give positive responses indicating enhanced transparency from the ability to check their measurement records, ability to manage and update their own user accounts, invaluable education on diabetes management, and provisions that allow them to browse history of the diabetic medications given to them over a period. Medics may praise the enhanced way of managing medicines through online prescriptions. The mobile diabetes information system indicates the medicine available in stock, informing immediate prescription options for medics. This enhances their work.

. Dobrica & Niemelä, 2002). The system maintenance persons may indicate optimal system design that uses minimum instructions to execute operations, thereby achieving optimal use of system resources. These perspectives are the majority summative opinions drawn from the user responses in the open – ended questionnaires. People are likely to give varied responses informed by their own independent assessment of the mobile diabetes system. The minority opinions may also suggest ways of improving the quality of the mobile diabetes information system. The varied opinions indicate the non-uniformity in perceptions of the users about the system, contrary to the initial thought that might have been held by the system designers (Poissant, Pereira, Tamblyn & Kawasumi, 2005)Medical administrators can possibly give responses to the effect that the information allows for better management of medicine information, maintenance of medication protocols, and supervision of patient therapy procedures. They also have the ability to define the actions that a patient would take under specific diabetic conditions depending on the read measurements such as weight, glucose and blood pressure levels (

. Bowen, 2009). Researchers can see how individual system users of the mobile diabetes information system construe its features, their roles and jobs, and their system interaction. This allows them to figure out the research hypotheses and theories associated with the information system. Qualitative evaluation methods also enable researchers to interpret any quantitative data that may be collected from the system in a sensible way. This is where the identification of process causality comes in. Capturing the point of view of actual system users is also important in assessing the level of acceptance or rejection of the information system. It indicates the changes in the business process occasioned by the introduction of the new information system (Boudreau, Gefen & Straub, 2001)Most importantly, qualitative evaluation research methods are favored for their usefulness in cases where the evaluators cannot figure out or formulate appropriate hypotheses early in advance (

. The qualitative responses give direct face validity, making the results of the evaluation believable by the system users, and other medical organizations seeking to install a similar self-monitoring information system for diabetes management. The fact that different persons can identify with the responses captured from qualitative research methods make them reliable and useful to other people. Results from this qualitative study can be applied in other contexts by developing a theory that extends beyond the specific mobile diabetes information system evaluated. This contrasts the statistical inference methods that rely on sample evaluations. Peffers, Tuunanen, Rothenberger & Chatterjee, 2007)Explicit information on cultural, human and contextual factors that determined the user acceptance levels can also be identified. Questionnaires administered to system maintenance people can help to divulge information on the circumstances under which the mobile diabetes information system was developed, deployed, and maintained (

List of References

Berg, M. (2001). Implementing information systems in health care organizations: myths and challenges. International journal of medical informatics64(2), 143-156.

Boudreau, M. C., Gefen, D., & Straub, D. W. (2001). Validation in information systems research: a state-of-the-art assessment. Mis Quarterly, 1-16.

Bowen, G. A. (2009). Document analysis as a qualitative research method. Qualitative research journal9(2), 27-40.

Dobrica, L., & Niemelä, E. (2002). A survey on software architecture analysis methods. Software Engineering, IEEE Transactions on28(7), 638-653.

Friedman, C. P., & Wyatt, J. (2006). Evaluation methods in biomedical informatics. Springer Science & Business Media.

Kaplan, B., & Maxwell, J. A. (2005). Qualitative research methods for evaluating computer information systems. In Evaluating the organizational impact of healthcare information systems (pp. 30-55). Springer New York.

Kjeldskov, J., & Graham, C. (2003). A review of mobile HCI research methods. In Human-computer interaction with mobile devices and services(pp. 317-335). Springer Berlin Heidelberg.

Lee, A. S., & Baskerville, R. L. (2003). Generalizing generalizability in information systems research. Information systems research14(3), 221-243.

Mingers, J. (2001). Combining IS research methods: towards a pluralist methodology. Information systems research12(3), 240-259.

Petter, S., DeLone, W., & McLean, E. (2008). Measuring information systems success: models, dimensions, measures, and interrelationships. European journal of information systems17(3), 236-263.

Peffers, K., Tuunanen, T., Rothenberger, M. A., & Chatterjee, S. (2007). A design science research methodology for information systems research. Journal of management information systems24(3), 45-77.

Poissant, L., Pereira, J., Tamblyn, R., & Kawasumi, Y. (2005). The impact of electronic health records on time efficiency of physicians and nurses: a systematic review. Journal of the American Medical Informatics Association, 12(5), 505-516.

Database Integration

Database Integration


The mobile diabetes information system enhances the element of self-monitoring in the overall management of diabetes. The power to monitor body changes such as BMI measurements, glucose levels and blood pressure levels can be done remotely by the patient. The recorded information is then sent securely to the central servers that hold medical records of patients. These servers are the master stations described in the system architecture presented for the mobile diabetes information system. These servers then update the health information database in the medical control center. It is important to integrate the information in the database containing the diabetic patients’ information with other electronic patient records accessible by the medical personnel to enhance the treatment process. The integration would enable quick and real time access of the patient information. A patient in remote location with a sudden surge in glucose levels can also access immediate medical intervention that would improve the situation. This discussion highlights how the generated database containing information gathered from diabetic patients can be integrated with the other electronic patient records to allow access for the healthcare professionals.

.Parent & Spaccapietra, 1998)The database containing the diabetic patients data can be integrated by using applications developed purposefully to locate data in a database, retrieve it and integrate it with the existing electronic health records. A query language such as SQL may be used. Data warehousing can also be used to pull data from both the diabetic patients’ database and the electronic health records. The data is then transformed into a common format. It is then loaded in a new integrated database. Any subsequent query would give an integrated overview of the patient data for health professionals to use (

. Batini, Lenzerini & Navathe, 1986) The process of extracting data, transforming it and finally loading it into some single schema is important when seeking to integrate heterogeneous data. In this case, the self-monitoring mobile diabetes information system is a new system that is being developed after an initial hospital information that stored electronic health records existed before. The possibility of non-compatibility of the new diabetic patients’ database with the early existing systems for electronic health records is quite high. The problem with using data warehouses as an approach to database integration is that in a system like a mobile diabetes information system, the data in the diabetic patients’ database would be updated frequently when different patients submit data frequently. This situation would mean that the process of extracting, transforming and loading data is done repeatedly in order to keep the data synchronized (

. In such cases, there needs to be direct mappings between the original schemas in individual databases and the virtual schema. The Global As View and the Local As View approaches may be used to achieve schema mapping. The individual diabetic patients’ database and the electronic heath records are perceived as materialized views by the integration software that creates the needed mediated schema. Hull & Zhou, 1996)Another approach to apply in integrating databases can be the use of mediated schemas. In this model, data is retrieved from both the electronic health records and the diabetic patients’ databases directly. The mediated schema creates a virtual single database from which unified queries can be sent to the individual databases for data retrieval by using wrappers (

. Commercial data mining software that can be used include DataWatch, Paxata and Alteryx. Noy, 2004). The mediated schema models the possible queries that users may seek from all the individual databases. Wrappers that would mask the original query results got from the disparate databases systems are then built. The masked results would be meant for the integration software used. When a system user submits a database query, the database integration software changes the query into the corresponding queries for the individual incompatible databases. The results of the query is also combined into a form recognized by the integration software and then presented to the user as the final response (Rahm & Bernstein, 2001)In actual sense, the databases are not really integrated, they still hold their original disparities. Only their views and queries appear integrated, yet they are a result of mediation created by wrapper programs (

In conclusion, disparate database systems still pose a challenge in the development of unified data sources in information systems. It is still an area under much research. Varied database systems such as Oracle and DB2 may be quite difficult to integrate due to their varied vendor specifications. Still, databases from a single vendor may also pose challenges when columns representing the same database elements are named differently or take in different data types. For instance, organizational earnings may be represented as ‘profit’ in two decimal places in one database and as ‘income’ with no decimal place in a second database, creating disparity.


(4), 323-364.18 ,ACM computing surveys (CSUR) Batini, C., Lenzerini, M., & Navathe, S. B. (1986). A comparative analysis of methodologies for database schema integration.

(Vol. 25, No. 2, pp. 481-492). ACM. A framework for supporting data integration using the materialized and virtual approaches Hull, R., & Zhou, G. (1996).

(4), 65-70.33 ,ACM Sigmod Record Noy, N. F. (2004). Semantic integration: a survey of ontology-based approaches.

(5es), 166-178.41 ,Communications of the ACM Parent, C., & Spaccapietra, S. (1998). Issues and approaches of database integration.

(4), 334-350.10 ,the VLDB Journal Rahm, E., & Bernstein, P. A. (2001). A survey of approaches to automatic schema matching.