Barriers in the implementation of health information systems: a scoping review.

• Author: Stamatian, Florin

1. Background

   1.1. Why is health information relevant?

   Health-related information is vital for the field of public health, since it is the foundation for policies, campaigns and programs aimed at maintaining, promoting and improving health (Bryant, 2002, pp. 89-98). Even though information comes through various channels, vital statistics and cause-of-death statistics, to name two of the most rich data sources, serve as solid tools for drafting and implementing public health policies. Country health information systems include a range of population-based and health facility-based data sources (WHO, 2013). However, there are significant disparities between countries, with little improvement in the amount and quality of data especially in low-income countries (Mahapatra et al., 2007, pp. 1653-1663). Up to 2007, only 125 of 193 countries have provided to the World Health Organization data regarding the causes of death at least once after the year of 1950. Moreover, data coming from different backgrounds generally differ with regards to accuracy, relevance, comparability, timeliness and accessibility (Detmer, 2003). To overcome this barrier, various systems have been implemented to support the collection of comparable data from different countries. The European Community Health Indicators Monitoring, for instance, is a European-wide initiative, launched by the European Commission, to collect health indicators from the countries across the continent. In time, its initiators have produced a list of 88 indicators, grouped into five major categories (demographic and socio-economic factors, health status, determinants of health, health interventions: health services, health interventions: health promotion), which provide a realistic picture of the health system they are coming from (European Community Health Indicators Monitoring, 2012).

   The financial difficulties which were brought by the generalized financial crisis led to serious effects on the budgets allocated to health worldwide. Although these changes represent both a problem and a challenge, because they can act as important catalysts for structural reforms, they inevitably produced a reduction in the investments in health information systems (Quaglio et al., 2013).

   1.2. How did the technology progress impact the collection and use of health information?

   The debate over the utility of health information systems in clinical settings is a long one. Some have argued that the arguments for their implementation are prevailing over the ones against the implementation of these systems (Maffei, 2006), whereas others have highlighted the increased costs associated with putting health information systems in place (Schmitt and Wofford, 2002; Mildon and Cohen, 2002).

   Electronic Health Records are moving from an institutional to an inter-institutional level, which raises concerns about the security and privacy of the shared data (Haas et al., 2011). In order to prevent any threat to become reality, the institutions handling EHRs must pay attention to three security goals, namely confidentiality, integrity and availability (Haas et al., 2011).

   Clinical decision support systems are able to improve healthcare quality and safety by enabling health professionals to make better clinical decisions (Johnston et al., 1994; Kuperman et al., 2007; Wright et al., 2009). Recently, their uptake has seen an increase, although it was recognized that more efforts need to be invested towards their evaluation and research, to ensure patients safety (Ammenwerth et al., 2013, pp. 1-3).

   1.3. How differently are health information systems implemented?

   The degree of health information systems implementation is variable, with countries such as Denmark (where 100% of the hospitals and pharmacies are currently using health information systems) leading the ranking (Deutsch, Duftschmid and Dorda, 2010). Other countries, such as the United States of America, the United Kingdom and Australia have also supported the funding of Electronic Health Records, Computerized Provider Order Entry and other consumer health information systems (Waterson, Hoonakker and Carayon, 2013; Stroetmann, 2012).

   1.4. Health information systems and safety issues

   Despite the benefits that the implementation of health information systems is bringing, attention needs to be paid to the safety of the systems (Menachemi and Collum, 2011). Efforts need to concern the standardization and oversight for the design, implementation and use of health information systems, in order to maintain and promote patient safety (Magrabi et al., 2013). Although health information systems are meant to increase patient safety, by supporting the clinical decision process, concerns have been raised that they can, paradoxically, represent a threat to patient safety (Ash, Berg and Coiera, 2004; Magrabi, 2010). These unwanted events, called 'e-iatrogenesis' (Weiner et al., 2007), are in fact part of a bigger category of unintended consequences related to computerized provider order entry (Campbell et al., 2006). It was argued that the identification and understanding of these nine types of consequences (see Table 1) will lead to better managing the computerized provider order entries

   (Campbell et al., 2006). In a systematic review focused on the use of health information systems in primary care, authors have concluded that the concerns about privacy, patient safety and provider-patient relations can be prevented or solved by strong leadership, effective project management and standard-setting (Ludwick and Doucette, 2008). Based on previous experience, some other critical areas encountered were change management, project management, implementation strategy and basic legal requirements (Deutsch, Duftschmid and Dorda, 2010).

   1.5. Purpose of the study

   Building on the available literature published so far, our literature review aims at providing an overview of the publications in which barriers encountered in the process of implementing the health information systems were identified. By referring to the implementation process we include the setting up and the running phases and the whole review will be based on this view. We also aim to highlight the most important lessons which were learnt after experiencing the development and implementation of health information systems at various levels in the healthcare system.

2. Methodology

   Our literature review focused on the scientific literature dedicated to health information systems and their implementation in Europe and the United States of America (USA). Furthermore, we focused on each European Union (EU) member state and each Non-EU country, as well as each of the fifty states of the United States of America with the purpose of comparing how the HIS implementation processes were undertaken. There is information regarding the implementation of health information systems in both Europe and the United States of America separately, but due to the fact that we have not found any review focused on the comparison between the two settings, we chose to address this issue.

   We performed the search by using five databases: PubMed, ScienceDirect, SCOPUS, Google Scholar and the Cochrane Library. Although initially we used all these five databases, only PubMed, Science Direct and Scopus retrieved articles related to our research question, therefore, we excluded Google Scholar and the Cochrane Library from our search databases.

   We set our inclusion criteria as being (i) articles in English, published in peer-reviewed journals exclusively, (ii) regardless of their publication date. In the above mentioned databases, we searched for the exact phrase 'implementation health information system' along with the name of each European country/USA state (accounting for each one of the 100 territories).

   Following the initial search, results were imported into a reference manager (Mendeley Desktop 1.10.1) and classified by the setting reporting data from (USA state/European country) and by the search engine.

   The tables below show the number of results obtained on each database, for Europe (Table 2) and for the USA (Table 3).

   Relevance was then tested by scanning titles and abstracts of each paper, as part of our second methodological step. During this second step, we traced a trend in emphasizing the barriers and challenges in implementing a health information system. Therefore, we limited our analysis only to this topic, since the literature we reviewed lacked articles focused on the differences in the HIS implementation process between the US, EU and Non-EU countries, fact that lead us to the third and final step. This involved reading all the articles that were chosen during the previous phases. One criterion for this phase was the availability of the full-text version of the papers. If access to full text was granted, the word 'barrier' and its synonyms ('obstacle', 'issue', 'problem', 'challenge', 'risk') were searched in the manuscripts. Where the search could not be performed, the document was scanned and if the terms were not found, the paper was considered irrelevant. Then, in order to aggregate the findings, a data extraction tool was created, containing the name of the state/country that the article reports data on, the article citation and the challenges identified.

3. Results

   Following the literature...