System Analysis And Design On Whis

Published Date: 02 Nov 2017

According to System Development Lifecycle (SDLC), analysis requirement is the most challenging phase of the development lifecycle. In this phase, user requirements are obtained in a number of ways: both by quantitative research methods by questionnaires and data collections or qualitative research methods by one-to-one interview in formal and informal ways and focus group discussion led by an analyst. Prototyping can also be

utilised when the high-level requirements have already in need to refine them at a lower

level. An analyst must decide what methodology should be applied and which the best suitable one for the development is. Then an analyst needs to model and abstract the investigations and findings of requirements. When functional requirements of the proposed system are obtained for the interactions between the user and the system, they are built up for every function by step- by- step view and define the scope of the system.

In order to have balanced perspectives of WHIS, four user groups are identified based on their roles as clinicians, admin staff, IT staff, and patients. Each user group has different sub-group of users respectively. Clinicians are doctors or nurses and other clinical staff. Admin staff will range from receptionists to managers. IT staff are also in different expertise like IT support, Network engineer or software specialists. The patient group consists of patients themselves or carers and family members. To secure the opinion of different user groups, both quantitative surveys and qualitative survey are conducted. Although the number of participants in focus group is small (N=50), the survey results are intensely attractive. The levels of Computer knowledge across participants are mostly similar (Q1). Five are beginners and thirty-five are intermediate users and the remaining ten are advanced users. Amazingly, all are familiar only with Windows and Office Suite as their main application for business information system. (Q2) The favourite features of the system are the user-friendly interface and except for the five beginners, all the others choose web systems (browser-based systems) are easy to understand. (Q3)When they are asked for specific information systems and GP’s medical information systems, only the participants who work in the NHS are aware of the system although all participants know their health records are stored by their GPs. (Q4) As a means of communication or access to GP’s system electronically, only five of the advanced users know how communicate or access. (Q5). Regarding the problems encountered using the health systems, all users had login problems, and half of all (25 users) answered they are not satisfied with the performance owing to the slow speed. Nearly a quarter of users (12 users) also have experienced virus and other security problems. (Q6).As to how often they have problems 50% chose SOMETIMES and 50% chose FREQUENTLY.

When a clinical system is inter-operated or integrated to other systems, 75% responded with do not KNOW and in the other 25% said "NOT SURE" (Q8).In terms of their opinions or comments as free text, most said they wanted to know more in depth knowledge about information systems and that they also want more privacy. (Q9).From this mini survey, health information system are not familiar with the users except who are directly using these systems for their daily work. They do not understand even similar health information systems except their systems and how they are interoperated or integrated to other systems. The survey was performed on all types of user groups at a small practice about twenty staff members (clinician, administrative staff, IT Staff) and a few patients. The survey was also distributed to post-graduate students of a class who are not directly related to the medical field or the computing field. Some volunteers participated from different user groups. Although the number of participants was small, the broad varieties of participants make the survey reasonably comprehensive. In age –range, most are in terms of middle aged persons (30-50) but three elder people (over-sixty) also participated.

For the qualitative research part of the survey, interviews were performed. Most clinicians feel that information systems are good but they think IT is dominated as opposed to the field to their main clinical domain. Most clinicians said information systems like Choose and Book are only for administrative purpose and they do not wish to participate it. A practice manager said the Information Governance toolkit is tremendously crucial for the health information system. A patient also asked more information about health information system and he used to call his GP frequently as a mode of communication.

In contemporary society, online services are maturing such as NHS Direct and EMIS ACCESS have become popular and even NHS Direct has recently developed a specialised app (applications) for mobile phone users. Though surveys and interviews, it can be clearly proven that although health information systems are used for daily healthcare, web-enabled health information system is still in the realisation stages and all interviewees confessed that they are unaware of web-enabled health information system like Health Space, Google Health, or Microsoft Health Vault.

In designing and developing web-enabled health information system, socio-technical factors should be considered. Instead of human-to-human interactions, human-computer interactions have taken over and people increasingly feel that they are ignored by their service providers or separated by the digital divide and lack of digital status.

Socio-technical thinking is defined that systems design should be a process taking both social and technical factors into the functionality and usage of computer-based systems. The rationale for adopting socio-technical approaches to systems design is that the failure to do so can increase the risks. As a result systems will not make their expected contribution to the goals of the organisation. Systems often meet their technical ‘requirements’ but they are considered to be a ‘failure’. The reason is that they cannot deliver the expected support for the real work in the organisation. The source of the problem is that techno-centric approaches to systems design without proper considering the complex relationships between the organisation and the people enacting business processes and the system that supports these processes (Norman, 1993; Goguen, 1999).Therefore, the system should be approached by the soft-system method(i.e. Human-centric) rather than hard-system (technical-centric). In general, the larger the scales of a new technology project in health care, the greater its chances of failure. This is because healthcare information systems are complex; they raise unique technical, administrative, and security challenges; and introducing new technologies into a complex system requires extensive changes in individual roles, relationships, and business processes—the so-called "socio-technical" aspects of change. [i] The design of the socio-technical requirements for ensuring appropriate Knowledge Management, i.e. the design of the necessary people and technology requirements for facilitating Knowledge Management in a targeted organisation. In spite of this transformative power, WHIS has not been able to equally and successfully ‘penetrate’ the clinician–patient relationship and replaced certain paper forms during encounters. The standardisation of healthcare provision seems to facilitate the deployment of information technologies. Clinicians are now better equipped to align effectively their activities under rigid protocols and well-defined frameworks of practice and vice versa, the configuration of the system has been perceived to be limited in encompassing some of the unique caring aspects those clinicians understand being part of the essence of their practice culture and which the medium of paper still support.

To understand the user’s requirements and current situations including socio-technical effects, Rich picture is supported an analyst to get a better understanding and communication with users. The common issue and focus of the proposed system can be figured out by drawing as well as evaluating the rich picture. It can express the unstructured problems like not in real-time and different format for activities recording, report sending, etc. of real world by identifying the key actors such as clinicians, medical Centre managers, identifying the key issues and areas of conflict (e.g. Finance, data entry and data format) affecting this environment, and suggesting reasons for these and explaining the main focus of the system that the organisations require.

Key actor groups, GPs and their primary care teams are working in the operational level by directly contacting with patients who are benefactors. These key actor groups have problems and difficulties concerned with day-to-day basic issues. They always want to respond the working environment with the bottom-up approach. They want their autonomy and accountability by their way of undertaking the actions to the working environment. One Centre manager thinks what the Centre managers need for the new proposed system. Even he wants a local finance system – so it is no point waiting for central finance. Such an issue is the current controversial issue of NHS reforms and "Health Bill". Another Centre manager thinks the data quality and Information Governance are crucial for his Centre. Therefore, it is clearly deduced that each Centre managers has their own perspectives, desire, skill, knowledge and usage of the system in their own understanding and interpretations. They do not want to give up their own autonomy and the way of working. In the real world, each health centres runs in an autonomous way. As a result, there is a wide range of activities that are handled differently depending on which centre is responsible for entering the data. It is happened like that too many cooks spoil the broth. This is increasingly causing problems, as there is inconsistency and confusion about how things are done. In a number of cases, this has led to wrong information being given out. Reasonably, data is extensively more and more vital in modern working environments. Lacking of a good standard and quality, searching for the required data is looking for the needle in a haystack.

In contrast, consistent and uniform workflow and dataflow is making well-organised

working environment to get the best quality and productivity targets by meeting user

requirements and budget and time constraints to become the proper professional

organisation. Additionally, the profile of the organisation is raised by significant

features and values. Introducing the proposed WHIS, it can solve these divergent ways of working styles and data recording out in efficient and effective manners.

The whole Rich picture is summarised by CATWOE Model. For a typical web-enabled health information system, CATWOE model is identified as Customers are patients who can get the benefit from the system, Actors are different user groups including patients, Transformation Process is input patient’s data such as blood pressure or height and weight is processed and transformed to output as hypertension or obesity, World View as the relationships and interactions between the users or actors, Ownership is the NHS organisation and Environment Constraints are finance, IT literacy and Mindset.

Rich Picture is easier and less technical way of describing circumstances and analysis of

the current complex working environment in an organisation, which are unstructured problems. Most rich pictures seem naive but in fact, they are artless representing an utter natural way of innocent and simplified thinking to the real world. The relationships and conflicts between actors and constraints of the real world environment can be evaluated by rich picture. In rich picture, each user groups as key actors having their own views and responses to the working environment and proposed system. When drawn rich picture, the essence and goal of rich picture can be summarised by root definition, which concisely describes the system relevant to the problems. It also defines the essential elements of the relevant system being modelled rather like a mission statement. In this case, the root definition, primary task relating to basic tasks and structures, is defined as "a system to improve the activities by means of standard formats and common user interface on web pages in order to produce records and reports to help decision making properly". In issued based root definition, relating to qualitative concerns, or concerns of judgment, we can define as "a system operated to implement a quality service for providing healthcare by devising and operating procedures to delight its customers (patients) in order to improve its activities and to develop good standard for these activities". The root definition can deduce the real world situations by system thinking. It leads to conceptual models and then can compare this model with the rich picture already drawn.

Fig 1: Rich Picture of proposed WHIS

Drawing of Rich Picture also makes better understanding and communication between users or stakeholders and a system analyst. Normally, the system analyst is always biased on the technical aspect and little or none on human aspects or organizational aspects. Soft System Methodology drives the situations to consider by soft approach. It introduces the human activities system by combining social systems with natural systems and technical systems or design systems. Ontological Approach to Systems Analysis is the branch of metaphysics dealing with the nature of Being (e.g. "There is the system") looking for a technical solution with understanding the functionality, data requirements, systems behaviour while Epistemological Approach is the theory of the method or grounds of knowledge (e.g. "I can consider any collective set as a system") looking at the social, moral, ethical, human aspects of systems development with understanding people’s aspirations, needs, fears, emotion, etc. It can be figured out from rich picture that the technical aspects are different data formats, inaccurate records and reports by missing quality targets of in time and budget. On the other hand, there are human activities of communications and relationships between actors each other. As exposed earlier, there are conflicts of interest between them. The rich picture of the proposed system can make the outcome for how to get accurate and reliable data in time by standard format and common user interface (e.g. website of the information system with the central database PDS Spine) to share the patient details easily and securely. Since proposed web-enabled health information system is the prototype model, it can obtain the business models plus technical and financial models for future projects and extensions of the current project. In addition, it would be good if this new system could work out for the best ways in working by analysing the current conditions and problems in the real world. Hard systems are technology based, easy to predict and little complex and on the other hand soft systems are human based, hard to predict and complex.

Briefly, Rich Picture can demonstrate to abstract relevant information from the real world settings and to understand the complex environment within the organisations.

WHIS is concerned with different environments and different professions. In other words, WHIS belongs to different domains. Even the user types are varied. Although clinicians were used to be the main users, patients are now involved more and more. Different worlds with a different point of views, and different opinions are concerned. Clinicians are bound to clinical governance. Clinicians are also using information systems so information governance is also involved. Since clinicians are working under organisations, organisational governance or corporate governance is also involved again. Furthermore, everything is concreted with finance, especially in the public sector organisations, so financial management is involved in Financial Governance. Clinical governance is the system through which NHS organisations are accountable for continuously improving the quality of their services and safeguarding high standards of care, by creating an environment in which clinical excellence will flourishing guidelines are also the standard in clincher domain. Care Quality Commission is established as the watchdog for clinical domain.

In information technology domain, data and information management is exceedingly crucial. Data quality, data accuracy, data flow and data control are the main factors of the information domain. Data quality and data governance should never be considered a onetime project. A quality culture must be established as an on-going, continuous processional quality and information governance is a continuous process and No organisation can tackle them all at once. Data quality is totally supportive and decisive role in quality of health care when clinicians can provide more accurate diagnosis and treatment. Information Governance is also covered the information security, information assurance, at protection and confidentiality. It also contains legal regulations such as Data Protection Act 1998, Computer Misuse Act 1990, and Medical Record Acts.

Corporate governance or organisation governance is also vital part of the web-enabled health information system. Most systems are exceptionally good in technology but not good enough in approaching organisation culture and behaviour. Leadership, accountability, and transparency are the key factors of organisational governance. Freedom of Information Act 2000 exposes and discloses how an organisation is behaved. Organisational structure and organisational workflow can lead to the Role Based Access Control (RBAC) model.

Last one is the financial governance since finance is controlled everything and care quality and targets are compensated by finical incentives. Moreover, NHS organisations are public sector organisations so financial governance is also beneficial. For any projects or developments, financial sponsors are the key actors and decision makers to continue or scrap the Project. Because of the global crisis and public sector cut, NPfIT is also under fire and value and money is balanced now. Recent National Audit Office report in May 2011 stated that NPfIT is not worthy for spending.

WHIS has totally occupied in two distinct domains, clinical domain, and information domain. Four different types of user groups are also representing their different status. Furthermore; technology is also changing the reputations of users’ perspectives. For example, clinicians feel they are time-consuming on information systems rather than clinical features. They also think more administrative works are put on their shoulders and they need to learn new computing technologies and web technologies. IT staff feel that their roles are diminished and there is no role in boardrooms and any other higher roles.

Furthermore they think their profession is not as recognised as clinicians or other professions. Patients feel that they are ignored and they are facing the machine or machine-based mechanism. Digital divide and financial status also make more gaps between vulnerable groups and advantaged groups. Original intentions of WHIS are helping and promoting the vulnerable groups. Oppositely these vulnerable people feel like that they are degraded and more vulnerable than before because of their defects. In reality, WHIS can provide personalised information related to their medical details as directly extracted from their medical records.

When considering analysing and designing an information system, especially WHIS, four world framework is considered. Natural world consists of natural things including human beings. Material world consists of artificial products produced by human. Technical world is composed of methods, technologies, frameworks, and rules. Social world consists of human communication and interactions.

Socio-technical aspect is also famous for WHIS because the web has changed daily lifestyles. Enid Mumford (Mumford 1993; 1997) developed Socio-technical approach in IT contexts, with an emphasis on techniques that enable users, such as nurses and secretaries, to participate in helping to design the computer-based systems with which they work.

NHS itself has changed organisational culture and faces new technology changes with challenges and careful approaches. Web-enabled health information system in NHS must investigate how socio-technical features have effected, how they are tied together, what new culture may be emerged. Socio-technical issues can even prevent the human errors and errors of reactions to system. All stakeholders’ views and concerns are considered depending on their roles to interaction with web-enabled health information system. For example, clinician’s functions are different from admin staff while patients’ are also in different. Since the web is shared, linked, open and heterogeneous, evolving of socio-technical issues on web-enabled health information system are recursively composed of other infrastructures, platforms, applications and IT capabilities and controlled by emergent, distributed and episodic forms of control. Evolutionary dynamics of web systems are nonlinear, path- dependent and dominant by network effects and unbounded user and designer learning. When developing web systems, too many concentrations on the technical details and little on the human aspects should be avoided. Misconception is that a technical solution is the complete solution to the problems although socio-technical effects by human to systems are the most critical. Developers and analysts have strong technical skills but they have inadequate ‘human’ skills and little understanding of the ‘flexibility’ of human based systems. They must think the users’ perspectives and perceptions because human flexibility is a plus and a minus in the systems development.

Seeing technological procedural aspects are more valuable than socio-technical aspects can lead to the system failures and in some cases, it can be the fatal errors of the whole system. Most data driven systems development are concentrating on the ‘formal’ parts of the systems and most tools have come about through developing technical solutions

instead of how to use the data by users and what are the data sources created by human.

Very significant example of current complex IT project is NPfIT. In this project, many potential problems could be avoided if those closest to healthcare delivery are able to have a greater say in shaping NPfIT implementations to specific local requirements. Furthermore, examples of the successful application of socio-technical methods illustrated how this can be achieved in the diverse contexts encompassed by the NHS, whilst also sustaining core national requirements. [ii] In 2011, Health Secretary Andrew Lansley also proposed the NHS reform but socio-technical issues with end-user centered approach are ignored and later it is amended by listening campaigns to understand the socio-technical issues. There are a lot of consultations and responses from different professional bodies and stakeholders. Multi-disciplinary, multi-stakeholder forum like NHS future Forum were hold and recommendations and amendments to the original program were made.

Another tool and technique used for analysis and design phase is UML Use Case. UML (Unified Modelling Language) is a standardised modelling language consisting of an integrated set of diagrams helping programmers and analysts to model and abstract the

analysed requirements by specifications, visualisations, simulations, testing, and

documentation. Among different UML techniques, Use Case diagrams are the preferred

modelling technique for this dissertation. Among the tools to setup for UML Use Case

diagrams, Microsoft Visio is the best tool to express and describe the UML Use Case

diagrams. Use case documentation with primary scenario and alternative paths and secondary scenarios can be lead to the good system design of the proposed system.

All outputs of the analysis stage are transferred to design stage. For the design stage, ERD (Entity Relationship Diagram) can describe the relationship between entities with their degree of relationship as one-to-on or one-to-many. Moreover, it can expose the primary keys, the unique data field that is related by other data fields. In other words, all data in the real world are organised to relationship tables and then they are turned again to databases.

Without data, nothing can happen and no information, no database, no information system will exist and thus nothing at all. To overcome such disaster, data are systematically organised into databases, in turn, databases are the atoms of the modern

Information systems and all applications are embedded in these systems. Database systems are evolved since 1950s and early 1960s. In 1970s, relational model and relational database were born. Subsequently in 1980s, distributed and parallel databases and object –orient database were researched and widely used in 90’s along with the explosive growth of World Wide Web. Up to now, databases are widely used for all purposes and data and database are decisive core items in the modern age. When data are assembled, classified in different manners and put in order to form a database system, Database Management Systems (DBMSs) are complementary for these databases. In other words, structure, formats and management of interrelated data and databases are combined into computer programs to form a DBMS. "Database is an asset of logically related data, and a description of the data, designed for the information needs of an organisation." Databases and subcomponents such as tables and columns can be created, edited, and managed by a DBMS. Technological point of view, DBMSs are the programs managing data internally to convert general raw data into useful information by different procedures, principles, protocols and structures including tables, forms, reports and queries.

DBMSs are programs themselves and they can process the input data into user required

output by different manner including running SQL (Structured Queries Language)

statements. DBMS can build up and transform the database systems. Database systems

are originated and formulated by different models including relational model. Most of the

databases systems currently use in daily life scenario are relational database models and

Relational database model is constituted of databases with subcomponents tables.

(For example, Patients are registered to Medical Centre, Medical Centre has staff like clinicians, admin staff, and IT staff and then entity relationships are formed.).

Primary key is defined for this relationship. For example, each person should have a unique ID number. For a patient, NHS number is a unique ID number and even on the PDS Spine, it is regarded as primary keys. A new registered patient cannot have a NHS number. Furthermore, NHS number is only used in England and (CHI in Scotland or Health and Care Number in Northern Ireland) so it is not good enough for a global ideal system. Instead of complex NHS number system, NHS England is using ten –digit NHS number. For the global ideal system or generic WHIS, the global unique ID is needed. Another thing to consider is needed the machine auto- generated number or customized coded ID number. To overcome this dilemma, regional top-level domain code and one more digit number for global population can be used. i.e. 11-digit number.

Another solution is applying metadata analysis, which allows the systems to understand the exact nature of the information in all data sources, organise it successfully, and ensure that data quality initiatives run smoothly. Meta data analysis can organise the data logically and group the related data for data management programs accurately. Furthermore, it can Exclude irrelevant data and organise data and prioritise data in order to begin more in depth data profiling and data analysis. [iii] Metadata analysis helps to understand where good data is and how to bring it into a data management project. Another data quality control method is following CUI (Common User Interface) rules.

When patient details are entered, Common User Interface rule must be followed. Healthcare professionals utilize a variety of clinical applications developed by different application providers, who have their own approach to user interface design. This results in an inherent level of inconsistency between user interfaces that have the potential for risk to the patient. A healthcare professional might have difficulty identifying the same patient in different applications, and this could result in them administering the wrong care to that patient. When creating the data or entering the data, Family name must be all in UPPER CASE to differentiate from the first name and coma(,) is inserted between them. (e.g. TUN, Aung)When displaying date, the year should be four digits so there will be no confusions and no errors. In 2000, the millennium bug was emerged by two-digit year format. Even current Microsoft Office software up to Office 2003 versions have two- digit year errors Example, a patient born before 1930 cannot be put as 29 or 28 and it will display as 2029 instead of 1929. For that reason, data format for input data is necessary and all systems have standard formats.

When patients are registered, family histories are also taken. To be convenient, inherited disease groups should be organised as a database and that the database can be represented as a combo box in user interface. Therefore, data accuracy is achieved by user-friendly interface and they can be converted to READ Code or SNOMED-CT easily.

When designing for web-enabled health information system, network deployment diagram is also needed. Modern software systems are based on object-oriented approach, so classes and objects are also defined. Since WHIS has n-tier architecture, client side, server side and middle tiers must be included. In other words, front-end, back-end and data- end are defined for each tier as presentation layer, business layer and integration layers.

In summary, analysis and design can illustrate what the proposed system should be included with functionalities, user-requirements and technological requirements.