The investigation was performed as part of a master’s thesis in the Master of Health Business Administration (MHBA) study program. 347 pages, 240 illustrations, 28 tables, 92 annexes and 1,085 footnotes covered the development of the criteria and evaluation of the HIS [60]. A total of 1,102 criteria were developed and tested in practice successfully. Therefore, only main results can be presented. Detailed information is available from the corresponding author. Criteria were divided into the categories task-related requirements (Table 1) with concentration on important medical processes in all departments and cross-functional requirements (Table 2) with a more global approach and technical point of view. 959 task-related requirements focused on patient care, medical history, organization, resource planning, management, research and teaching. 143 task-spanning requirements focused on management and operation of the information system, integration, architecture, data protection, data security and user interface. Detailed evaluation criteria are provided as supplementary material for task-related (51 pages) and cross-functional requirements (9 pages).
For both categories groups with tasks and subtasks were derived to structure the findings. Not all tasks were sub-classified if not useful for achieving efficiency. Criteria were weighted indirectly via the points scale. Separate weighting factors were not used. Instead, weighting was based on the number of criteria per task and the subdivision into subtasks.
Totals in Tables 1 and 2 were based on the degree to which assessment criteria were met using a 4-level rating scale for calculating the point sums:
- 0: not fulfilled at all (maximum inefficiency)
- 1: only marginally fulfilled with clear weaknesses
- 2: medium, only partially fulfilled
- 3: completely fulfilled and efficient
Table 1 ist not public. Please contact Dr. med. Stefan Wagner, MHBA.
Optimal planning, collaboration and minimization of documentation load proved to be crucial. Particularly in the operating room with influence on upstream and downstream services, not only added value but also costs are the greatest. Thus, in a new group, criteria were introduced in special areas of anesthesia and surgery. Accordingly, new sub-items were anesthesia documentation and preparation for operations, intensive care as well as documentation, planning and management of operating rooms. Operating rooms, anesthesia and intensive care were assigned most criteria per single subtask.
Table 2 ist not public. Please contact Dr. med. Stefan Wagner, MHBA.
Evaluations of local and maximum possible implementation, target status of total efficiency and percentage of achievement were compared and aggregated at level of tasks and subtasks. Detailed solutions for optimizing identified weaknesses were presented in the master thesis [60]. With 959 task-related criteria, the local implementation achieved 2,285 out of 2,877 points (79 %), the maximum possible functionality 2,843 points (99 %). 20 percent improvement was achieved by the parameterized HIS. For 143 cross-functional criteria, on-site implementation achieved 379 of 429 points (88 %), the most comprehensive one 426 points (99 %). An increase of 11 percent was realized. Usability was important because of significant influence on the time required to complete tasks. In the practical example the score was 80 %. Compared to the overall amount of functions, only a small number was licensed. 33 additional third-party systems were identified. Some selected results of the detailed findings are summarized.
At a glance, it became clear how efficiency could be increased through further modules, addons and adjustments. Addons were analyzed in detail with regard to possible use, previous solutions and contribution to achievement of the objective. Most addons (41) were offered for the user group of doctors. In terms of product type, 32 addons were available for medical documentation. The most common application area for 31 addons was the ward, followed by specialist departments (29). Of 43 target-oriented addons, only 17 were installed (40 %). Instead, inefficient third-party software and paper forms were often used. With 26 additional addons the discovered weaknesses could be eliminated.
In the subtasks of medical and nursing admission, there was potential for optimization with regard to reduction of free text entries, still prevailing, in favor of fully structured documentation. Implementation scored better in admission of doctors (83 %) than in the one by nurses (58 %). Data entry forms for anamnesis should be adapted to requirements of doctors and nursing staff. Users must have mutual access to administrative, medical and nursing documentation. Lack of a uniform cockpit for emergency department to control patient flows and triage instead of parallel use of HIS and third-party software was one weakness. A waiting list of appointments was maintained manually with nearly identical information. With the emergency room cockpit addon functionalities could be replaced eliminating redundancy.
Due to third-party software integration information sheets for operations, examinations and anesthesia procedures, provided with personal data, could be printed. Synergisms due to almost identical questions in anesthesiologic and surgical clarification were not used. Digital signatures and automated pre-completion of questionnaires based on medical history and medication plans were not applied either. Documentation of blood products was time-consuming on paper and via third-party software in the laboratory. Also missing was a cockpit for consultation hours and outpatient clinics, which efficiently supports appointment management, control of patient flows and process-accompanying documentation.
Lack of decision support with rule engines for doctors and nurses was another clear weakness. Only 40 percent was reached in decision making, consent and education, 45 percent in access to decision-relevant knowledge. Digital knowledge resources, standards and reference works were available. However, there was deficiency of process-oriented integration. HIS and resources were unconnected. Access during processes could only be realized with increased effort of manual activities. Doctors could not use clinical pathways to control processes or define pathways in workflow designers. The biggest weakness was the missing workflow engine for controlling work lists and treatment paths.
Speech recognition simplified writing of doctor’s letters and creation of reports. But dictation and speech recognition were redundantly implemented as solutions for operating room reports and pathology findings to HIS-integrated functions. Restriction to one solution would contribute to efficiency as well as usage for other documents of medical history.
While scores and scales were largely calculated automatically (cf. intensive care unit), interlinking of drug databases and specialist information with prescriptions, automated dose calculation and plausibility checks had not yet been implemented. Third-party systems overlapped as drug databases and thus represented redundant functionality. The Radiology Information System (RIS) was integrated supporting efficiency. A strength of the HIS was its support of diagnostic requirements eliminating numerous telephone calls and paper forms. In radiotherapy, irradiation planning was not done in the RIS, but in third-party applications. The unavailability of electronic anesthesia protocols with import of vital and ventilation parameters from medical devices via interfaces led to high inefficiency. In intensive care with only 28 % achievement of the target status, inefficient paper forms were used similar to anesthesia protocols. Anesthesia and surgical documentation were not linked. Therefore, parallel entries of identical information were necessary, e.g. anesthesia personnel, surgeon and operating times.
Diversity and inconsistency meant that information had to be searched for manually. End users could never be sure they had all the relevant information. This was especially the case because data was hidden in past visits. Available form designers could enable adaptations. The absence of a medical knowledge database for decision support in addition to media discontinuous documentation of anesthesia and intensive care was of particular importance. Paper documentation must be converted into digital documentation packages.
