Nursing in the Information Age: Status Quo and Future of ICT use in German Hospitals

Transcription

1 MEDINFO 2004 M. Fieschi et al. (Eds) Amsterdam: IOS Press 2004 IMIA. All rights reserved Nursing in the Information Age: Status Quo and Future of ICT use in German Hospitals Ursula Hübner, Björn Sellemann Healthcare Informatics, Faculty of Business Management and Social Sciences, University of Applied Sciences Osnabrück, Germany Ursula Hübner, Björn Sellemann Abstract Hospital information systems (HIS) should give support to nurses in their clinical and managerial duties. Though there are statistical data on the current use of HIS systems we know only little about the numbers of nursing modules implemented. We therefore conducted a nationwide survey in Germany (n = 2182) on the current state and future plans of HIS modules including nursing applications (response rate of 27.6 %). The findings show that management applications (84 % accounting) are still more frequent than clinical applications, in particular clinical patient record systems (19 %). What applied for HIS modules in general held also true for nursing on a lower level. Whereas 51 %of the hospitals had rostering systems in place only 6 % used care planning software. Priorities and plans for the future reveal no change in the rank order of systems. We argue that in order for clinical documentation and planning systems to catch up they must be immediately rewarding for the clinicians in their daily need for information. Keywords hospital information systems, nursing, patient care, hospital management Introduction Hospital information systems 1 should serve by definition all professional groups within a hospital to support state of the art patient care and advanced hospital management [1]. Though requirements and recommendations for good practice date back to the sixties/seventies [2, 3] there is consensus that the progress made within the last decades had not yet led to a fully satisfactory level of information and communciation technology (ICT) infrastructure and systems in hospitals [4]. There are centres of excellence in which many of the desiderata have come into place [4] and seamless information flow is possible but many hospitals still suffer from insufficient, delayed, erroneous or simply missing information due to inappropriate information systems. Though there is a trend towards clinical computing (reflected by e.g. [5]) the deficiencies mentioned still hold true notably for 1. In the recent literature the focus of research has shifted from hospital to health information systems. As we expected only very few health information systems to be operational at the time of our study and as the users in the study were certainly more familiar with the term hospital information systems we prefered it to the wider term health information systems. clinical applications including the clinical patient record (CPR, only 13 % of hospitals have operational CPRs in 2002 [5]). Reasons for this still disappointing picture are discussed in the literature and comprise among others (i) low ICT budget in hospitals [6], (ii) little awareness of the benefits of ICT [7], (iii) lack of standards to overcome islands [4, 8], (iv) technical problems such as fragile systems [9] and (v) too little involvement of users [3]. According to the HIMSS survey of 2002 [5] the three top ranking barriers in the United States were (1) lack of financial support, (2) vendors inability to effectively deliver and (3) achieving end user acceptance. In Germany, the most recent KIS study [10] reveals a similar picture concerning barrier number one. The three major impediments according to the German study were (1) budget too small, (2) missing/insufficient strategy, (3) no evidence for return of investment and software problems. Nursing systems that support patient care and management go back to the eighties (in Anglo-American countries) and the nineties (in German speaking countries) [11] but its implementation in hospitals seems to come with a greater delay than clinical patient record systems. Neither the HIMSS nor the KIS study mentions applications for nursing distinctly. This is surprising as nurses form the largest professional group in hospitals [12] and are well recognized for their role in information gathering and communicating. Recent definitions of Nursing Informatics therefore make the role of nurses as information integrators explicit [13]. Moreover, applications for nurses are particularly interesting because nursing embraces a whole variety of different tasks covering clinical (e.g. nursing process) as well as managerial issues (e.g. human resource management, logistics ) [14]. As nursing is moving into the Information Age it is of interest to healthcare providers and vendors alike which nursing applications are in use to date and what are the prospects and priorities for the future. Since nurses typically do not decide alone about new investments into nursing system we wanted to know whether there is a difference between the nurses and the hospital managers point of view. Understanding nursing applications as part of the overall hospital or health information system we further wanted to put the results into the context of other solutions implemented and of organizational and financial characteristics of the hospital. Materials and Methods We therefore conducted a nation wide survey including all acute hospitals in Germany (n = 2182 according to 376

2 Two questionnaires one specific for nurses, the other specific for hospital managers were developed. Both of them contained overlapping questions about basic hospital characteristics (number of beds, parent organisation, ICT budget, number of wards with at least one PC workstation), HIS-modules implemented at the site and clinical modules planned. In case the answers given by the managment and nursing did not correspond the answer was coded as missing. Also, both questionnaires included statements about the priority of nursing applications as compared to a selection of HIS modules for management and medicine/nursing (table 1). Table 1: Priority of modules - example from questionnaire Introduction of ROSTERING software is more important than... care planning care documentation ordering system + result reporting 3D visualisation of medical images electronic archive meal ordering management information system ing internal laboratory theatre nursing school management material management time data entry Priority of a system was defined by the percentage of numbers of yes and equal answers of all answers possible yielding a value range from 0 (low priority) to 100 (high priority). In addition the nursing questionnaire contained detailed questions about the nursing modules implemented and about requirements for nursing modules. Altogether the management questionnaire consisted of 15 items, the nursing questionnaire of 30. Each hospital of the population of German hospitals received the two questionnaires, one addressed to the general manager of the hospital and the other to the nursing manager/nursing director. Mailing took place in March 2002 with the hospitals requested to answer back until August No reminders were sent. Results yes no equal do not know Sample Out of the 2182 letters mailed 13 were returned because the hospital had closed down. From the remaining 2169 hospitals 600 answered which makes an overall response rate of 27.6 %. Among those were 95 hospitals for which both the general manager and the nursing manager/director replied. Split into the two professions we achieved a response rate for the nursing managers/directors of 20.7 % and for the general managers of 11.4 % yielding a total n = 695 returned questionnaires. The goodness of fit between our sample and the population was estimated based on the variables location (federal state where hospital is located), parent organisation and size (number of beds). To minimize the β-error, α was set to 0,25. All chi-square values were significantly different from zero (table 2) meaning that the sample was not representative for the entire population with respect to the three variables. Table 2: Goodness of fit between sample and population χ 2 χ 2 df α;df location parent organisation size Precision of the results based on the sample size of 600 hospitals and a proportion of 50 % is within the limits of 46 % and 54 % (confidence interval at α = 0,05 two sided). For a proportion of 80 % the confidence interval stretches from 77 % to 83 %. ICT and HIS modules Based on the answers given by the hospital management and nursing management the following distribution of administrative HIS modules in German hospitals resulted (table 3). Accounting was the most frequent module implemented, followed by other traditional management applications and DRG computing. Archive management belonging to the administrative functions in a hospital ranked lowest. Table 3: Administrative HIS modules implemented (n = 597) percentage accounting 84 % asset analysis and reporting 70 % human resource management 68 % material management + pharmacy 68 % controlling 68 % DRG 59 % electronic archive 15 % Among the clinical modules the laboratory system was the one most of the hospitals had in use. Electronic clinical patient records were implemented in 19 % of the hospitals. The complete list of clinical modules is shown in table 4. When asked about their ICT budget only 57.4 % answered the question. The distribution of budgets peaked at between 50 and 150 thousand Euro (modal value and median) for new investments in 2002 resulting in 331 Euro investment budget per bed (median) over all hospitals with a hugh range from 22 Euro budget/bed to over 5995 Euro budget/bed. Status quo in nursing applications Table 5 shows the answers of the nursing managers/directors to the question what nursing related modules they had implemented. Relevance to nursing was defined by degree of usage by nurses. (In Germany order entry systems are used by nurses mainly.) With 51 % rostering was the most frequent module in use whereas care planning achieving a mere 6 % -was the least frequent. Difference between hospital and nursing managers From the set of data on future implementations ( current year, within next two years, planned without time frame ) only data on systems planned for the current year were used to estimate future implementations because they are the most reliable 377

3 to predict what will come. Plans and priorities (for calculation see Material and Methods) of nursing managers/directors and hospitals managers from the same hospitals (n = 95) were compared. Table 4: Clinical HIS modules implemented (if not specified n = 597) Nursing in the Information Age The three top ranking nursing applications for which half of all hospitals employs information technology cover human resource management (rostering) and logistics (ordering system and patient management), all of them classical nursing management tasks. Systems that support the nursing process such as care planning and documentation are much less frequently used. A low popularity of the nursing process, poor documentation - also on paper - and terminology problems may account for this finding. However, it may not be a nursing problem alone because also clinical patient record systems are not that frequent (operational only in each fifth hospital). This shows a general reluctance of hospitals to embark on clinical documentation. The number of nursing applications installed was independent of the hospital size and the budget per bed and there was no difference between public, church run and private hospitals. Obviously there must be other factors influencing and accompanying the fact whether a hospital makes use of ICT in nursing or not. One of those factors might be readiness for innovation. As care planning and documentation are no mainstream computer applications in hospitals it certainly needs innovative people willing to take a risk and to decide in favour of such modules. The mean priority given by nursing managers to four major nursing modules was significantly higher than that of the hospital managers, still the absolute values given by the hospital managers to nursing modules were quite high and the priority for the ordering system was very similar to the one of the nursing managers. Both professional groups agreed in the rank order of the nursing modules: the ordering system received the highest priorities, next came care documentation and rostering and finally care planning. As to the plans for the current year there was no difference between the two professional groups. Despite a differpercentage laboratory 69 % surgery (documentation) 68 % anesthesia (documentation) 56 % outpatient management (n = 581) 43 % clinical workplace 40 % radiology information system (n = 581) 30 % clinical medical record 19 % PACS (n = 585) 12 % bed management 9 % Table 5: Nursing modules implemented (n = 448) percentage rostering 51 % ordering system (examinations, drugs, material) 48 % patient management (ADT) used on wards 45 % meal ordering 23 % management of nursing school run by hospital 20 % time data entry in conjunction with rostering 14 % care documentation 7 % care planning 6 % A two way ANOVA for dependent samples resulted in significant (α = 0.05) differences of priority between nursing and hospital management and between the modules (table 6). Pairwise comparisons showed that the priority of ordering systems was higher than that of any of the other modules and that the priority of care documentation was significantly higher than that of care planning. Table 6: Priorities for four major nursing modules nursing mgnt. hospital mgnt. mean SD mean SD rostering ordering system care documentation care planning Table 7 shows the plan for the current year for installing the four major nursing modules. Data were split by the two professional groups. None of the four χ 2- tests (based on 2x2 cross tables each with df = 1) resulted in a significant difference between the nursing and the hospital managers. Context factors By context factors we mean size, budget per bed and parent organisation. The Pearson correlation between number of nursing modules implemented and (i) size (r = 0.16, p = 0.001) and (ii) budget per bed (r = -0.03, n.s.) resulted in a very low and no correlation at all. There was also no significant difference in the number of nursing modules in use between different types of parent organisations (public, church run and private) as calculated by the Kruskal-Wallis-Test (n = 448, df = 2). Table 7: Plans for the current year (percentage yes of professional group) nursing hospital n mgnt. mgnt. rostering 41 % 28 % 119 ordering system 38 % 31 % 102 care documentation 8 % 18 % 157 care planning 7 % 11 % 144 Discussion Sample This survey reflects the current situation of ICT in hospitals including nursing in Germany whereby Germany may be regarded as a good example of highly industrialised countries. Achieving a response rate of 27.5 % without sending reminders is a good result and forms a decent basis for data interpretation. In terms of representativeness this sample differs significantly from the German population in size of the hospitals, location and parent organisation with a bias towards large hospitals and public hospitals, in large states (e.g. North Rhine Westphalia). Though not fully representative the sample includes hospitals in any of the categories of the variables and therefore allows general statements on the current situation. The sample size is large enough to estimate the population proportions with sufficient precision. 378

4 ence in priorities nursing management and hospital management agreed on the plans for the current year. Care planning received both low priority and low frequency values for plans which coincides with the number of care planning systems already implemented. It seems as if neither nursing management nor hospital management esteem the utility of care planning systems very high. The overall picture of status quo and future of nursing applications is quite consistent: nursing uses ICT mainly for management tasks and will do so in the future following the priorities and future plans. 20 years ago... A comparison between our data and data from a survey on the use of ICT in German hospitals conducted 20 years ago [15] demonstrates the change achieved since these days. Out of the 902 hospitals 55.2 % used information technology for accounting, 19.6 % for outpatient management, 12.4 % for human resource management, 6.3 % for the laboratory, 0.3 % for medical documentation and 0.2 % for the discharge summary (clinical patient record). Looking at the numbers the trend is obvious: a major leap forward has taken place since the early eighties. Both managerial and clinical systems enhanced their penetration in German hospitals dramatically. In those days clinical systems were almost non existant and half of the hospitals still did not have their own accounting software. Interestingly, accounting and laboratory were and are still the top ranking systems within the categories managerial and clinical systems. But whereas accounting managed an increase only from 55 % to 84 %, laboratory jumped from 6% to 69%. If these systems stand pars-pro-toto for their category this means that clinical systems gained over proportionally compared to administrative systems during the last years. But the speed varies considerably also within the clinical category. Subsuming medical documentation and discharge summary under clinical patient record system there was an increase only from 0.3 %/0.2 % to 19 %. The 1982 study mentions nursing only in the context of intensive care with 2.9 % of the hospitals employing ICT for its use. As there were only extremely few systems for nursing til the late eighties [16] in Germany it is understandable the study did not include them. Conclusion The comparison of the data from our study and the 1982 study highlights the fact that an enormous increase in ICT use took place but 20 years seem to be too small a time frame to achieve the goals set 30 years ago. Care planning and documentation stands to date at a point where laboratory systems used to be 20 years ago. However, priorities for care planning are low and there are only few hospitals planning to implement care planning and documentation in the near future. Therefore, it is hard to predict how many of these systems will be in place in the next 5 to 10 years as we have only little insight into their true dynamics. The progress made with care planning and documentation will have to be monitored closely and follow-up studies will have to come to capture the trend. For the near future it seems certain that applications for nursing management will still outnumber clinical applications in particular those for the nursing process. If this is to change a new culture in planning and documenting clinical work has to grow supported by systems that immediately deliver what the clinicians need to know. Acknowledgments This work was kindly supported in part by the State of Lower Saxony Germany within the grant e-nurse. References [1] Bemmel JH, Musen MA. Handbook of Medical Informatics. Heidelberg: Springer Verlag 1997 [2] Ball MJ, O Desky RI, Douglas JV. Status and Progress of Hospital Information Systems. Int J Biomed Comput 1991: 29: [3] Ball MJ. Hospital information systems: perspectives on pro blems and prospects, 1979 and International Journal of Medical Informatics 2003: 69: [4] Kuhn KA, Guise DA. From Hospital Information Systems to Health Information System. Methods Inf Med 2001: 40: [5] 13th Annual HIMSS Leadership Survey sponsored by Superior Consultant Company. 2002, as accessed on August 25 th, 2003 [6] Iakovidis I. Towards personal health record: current situation, obstacles and trends in implementation of electronic healthcare record in Europe. International Journal of Medical Informatics 1998: 52: [7] Stead WW, Lorenzi NM. Health informatics: linking investment to value. J Am Med Inform Assoc 1999: 6: [8] McDonald CJ. The Barriers to Electronic Medical Record Systems and How to Overcome Them. J Am Med Inform Assoc. 1997: 4: [9] Haux R, Knaup P, Bauer AW, Herzog W, Reinhardt E, Überla K, van Eimeren W, Wahlster W. Information Processing in Healthcare at the Start of the third Millenium:Potential and Limitations. Methods Inf Med 2001: 40: [10]KIS_Umfrage_2002 of the HIS-working group of the German Socienty of Medical Informatics (GMDS) as accessed on Aug 28 th, 2003, [11]Hannah KJ, Ball MJ, Edwards M, Hübner U. Pflegeinformatik. Heidelberg: Springer Verlag 2002 [12]Gesundheitswesen - Grunddaten der Krankenhäuser und Vorsorge- oder Rehabilitationseinrichtungen 2001, Fachserie 12 / Reihe 6.1. Wiesbaden: Statistisches Bundesamt 2003 [13]Staggers N, Bagley Thompson C. The Evolution of Definitions for Nursing Informatics: A Critical Analysis and Revised Definition. J Am Med Inform Assoc 2002: 9:

5 [14]Hübner U. Pflegeinformatik. In: von Eiff W Ed. Der Krankenhausmanager. Heidelberg: Springer Verlang 2000, Band 1, Teil 5, Kapitel 10, pp 1-35 [15]Lordieck W, Reichertz PL. Die EDV in den Krankenhäusern der Bundesrepublik Deutschland. Heidelberg: Springer Verlag 1983 [16]Bessai B. Software für die Krankenpflege. Deutsche Krankenpflege-Zeitschrift 1991: 3: Address for correspondence Prof. Dr. Ursula Hübner University of Applied Sciences Osnabrück - Faculty of Business Management and Social Sciences Healthcare Informatics Unit, P.O. Box 1940 D Osnabrück, Germany u.huebner@fh-osnabrueck.de, 380