A cross-sectional multilevel study on nurses’ experiences with health information systems: A key to understanding documentation hazards and technology-induced errors in different working environments

Abstract

Objective: This study aimed to investigate how nurses’ backgrounds, documentation skills, and experiences with documentation practices and information systems usage influence documentation hazards and to determine whether these hazards are linked to technology-induced errors (TIEs). Methods: An online survey was conducted to collect data from 3065 registered nurses working in Finnish hospitals and in acute, primary, and home care services regarding their experiences with electronic health records (EHRs) or client information systems (CIS). The data were analysed using linear and logistic multilevel models to identify patterns and correlations. Results: User interaction with EHR/CIS systems significantly influenced documentation hazards across different work environments. Perceived system-provided documentation support and documentation hazards were identified as contributors to TIEs. Conclusions: Improving system design and documentation support is a desirable goal, but it is not sufficient to mitigate documentation hazards and promote efficient practices. To achieve the best possible results, skilled users are needed to operate these systems.

Keywords

client information systems (CIS)documentation electronic health records (EHR)patient safety

Introduction

The challenges associated with health information technology can be understood through the concept of technology-induced errors (TIEs).¹ These are unintended adverse outcomes in healthcare that arise from the design, implementation, or use of digital systems.^1,2 The concept was first described in 2005 and has since been analysed from incident reports across multiple countries, e.g.,.^3–5 TIEs can be broadly classified into four categories: (1) clinical errors related to medication dispensing and recording, (2) procedural errors involving prescribing or legal issues such as documentation under the wrong patient, (3) human-related errors stemming from inadequate knowledge, skills, or usability challenges, and (4) technical issues associated with information transfer or system/software problems.^3,4,6,7 The frequency of such errors highlights their importance: while not all system issues result in direct patient harm, they create latent conditions that increase the likelihood of safety incidents.^1,8,9

A retrospective analysis of 2627 health information technology incident reports in England and Wales found that approximately 82% caused no harm, 13% resulted in low harm, 4% in moderate harm, 1% in severe harm, and fewer than 1% were fatal; notably, about 75% of these incidents were deemed preventable.¹⁰ The results highlight the need to focus on identifying aspects of technology use, people, and the environment that relate to technology-induced errors.

One critical source of technology-induced errors is hidden in documentation hazards.⁸ These refer to risks to patient safety that arise when clinical documentation is incomplete, inaccurate, misleading, poorly structured, or excessively detailed.^11–13 The Institute of Medicine’s report on Health IT and Patient Safety (2012) identified usability flaws and limited interoperability as key contributors to such hazards. Documentation hazards are risks to patient safety that result from unclear documentation practices and insufficient system support, particularly in electronic health records. In practice, documentation hazards may lead to delayed care, inappropriate treatments, or breakdowns in information continuity across care settings.¹⁴ While not all usability, interoperability, skills, or workflow problems in healthcare result directly in harm, many of these challenges represent latent conditions that increase the likelihood of technology-induced errors, thereby posing a risk to patient safety through documentation hazards.^8,13

The working environment, both physical and electronic, plays a central role in shaping how documentation hazards emerge. Poor user-friendliness, unreliable system functionalities, and fragmented workflows place additional strain on nurses’ documentation practices.^15–18 High patient volumes, time pressures, and limited knowledge of system functionalities increase the likelihood of errors in documentation practice.^17–20 These challenges are observed across professional groups, irrespective of variations in educational background, professional experience, and training, underscoring the pervasive nature of the problem.²¹

Education and in-house training improve nurses’ ability to use electronic records effectively, yet significant challenges remain, according to research.^22–25 Systematic and scoping reviews have highlighted documentation burdens associated with the use of electronic health records, including increased workload,²⁶ nurse-patient interactions,²⁷ interprofessional practice,²⁸ information overload,²⁹ interoperability challenges,³⁰ and fragmented workflows.³¹ While these studies provide important insights into the implementation of health information systems and understanding the risks for patient safety, less is known about how documentation hazards specifically relate to technology-induced errors, particularly from the perspective of nurses, who are the largest professional group using information systems in daily practice.

The Finnish health and welfare system provides a unique setting to explore this issue. Finland follows the Nordic welfare model, offering universal, tax-funded access to healthcare and social services.³² Since 2023, services have been organized by 21 Wellbeing Service Counties, supported by a national digital infrastructure (Kanta Services). Healthcare organizations primarily use electronic health records (EHRs), while social care relies on client information systems (CISs). Home-care and housing services sit at the intersection of these domains, making seamless and accurate documentation essential for continuity of care.³³ Registered nurses work in both healthcare and social care, across public and private sectors. Midwives and public health nurses receive a registered nurse qualification as well.^34,35

Despite nearly two decades of research into technology-induced errors, the relationship between documentation hazards and technology-induced errors has not been thoroughly examined from a nursing perspective. Nurses’ experiences are significant, as they provide direct insight into how daily documentation practices influence patient safety in everyday work.

Aim and research questions

This study aims to investigate how nurses’ backgrounds, documentation skills, and experiences with documentation practices and health information system usage influence documentation hazards and to determine whether these hazards are linked to technology-induced errors.

The following research questions were formulated:

1. What perceived factors related to electronic health records (EHR) and client information systems (CIS) usage are associated with documentation hazards?

2. How are documentation hazards and nurses’ backgrounds and skills associated with technology-induced errors?

Data and methods

Setting and design

In Finland, electronic health records (EHRs) were adopted in healthcare in the early 1990s. In comparison, client information systems (CISs) were introduced in social care in the early 2000s, marking significant advancements in documentation practices. The EHR and CIS brands vary between service providers; for example, the five university hospitals use three different brands: OMNI360, Esko, and Apotti (Epic). Some 10 EHR brands/systems with various information systems for specialties are used in healthcare, and some 10 in social services.^34,35 Finnish legislation distinguishes between patient and client data. Therefore, both EHR and CIS have been introduced; however, in some cases, the same brand name is used for both healthcare and social services.³⁶ This cross-sectional survey study explored nurses’ experiences with EHR and CIS related to patient safety. A research plan for data collection was reviewed and approved by the National Institute for Health and Welfare (THL) Research Ethics Committee (decision THL/634/6.02.01/2023 §928).

Data collection

The data used in this cross-sectional multilevel study come from online repeated cross-sectional studies that monitor digital healthcare and social care.^34,35,37,38 In spring 2023, feedback from 17 pilot testers representing the target population led to adjustments to the questionnaire. The questionnaire, freely available in Finnish, Swedish, and English, was distributed by the Finnish Nurses Association and Tehy, a trade union representing qualified health and social care professionals. Inclusion criteria were the following: Registered Nurse, member of a trade union or professional association, age 18–65 years old (e.g. working age). This information is based on the membership registers. Exclusion criteria included those who did not use electronic health records (EHR) or client information systems (CIS). This information is based on the respondents’ responses. The survey closed for those who responded negatively to this very first question in the electronic survey about EHR and CIS use. Members were limited to individuals aged 18–65 who had completed a nursing degree, and 3065 registered nurses completed the survey. Of these, 95 respondents were excluded based on their answers, indicating that they did not use EHR or CIS. Based on the research questions, this study presents the results of variables focusing on perceived documentation hazards, such as documenting the same client/patient information more than once for the same care event, and technology-induced errors, including faulty system functions (Table 1).

Table 1.

Variables used in the analysis.

Measure	Variables and abbreviations used in analyses	Scales
Background	Healthcare professional	Registered nurse
		Public health nurse
		Other
	Highest degree that you have completed (=highest degree)	Nursing, a postsecondary degree
		Bachelor of health care (nursing)
		University of applied sciences master’s degree
		A university degree
	Graduation year	Continuous
Skills/experience	Skills needed to use EHR/CIS (=skills)	(1) Fair – excellent (4)
Skills/experience	Experienced as an EHR/CIS user (=EHR/CIS experience)	(1) Beginner – highly experienced (5)
Perceived documentation support	System support for documentation (=documentation support)	(1) Totally disagree – totally agree (5)
Working environment	Where do you work more specifically? (=work environment)	Inpatient ward
		Emergency clinic, intensive care or monitoring unit
		Outpatient clinic
		Patient/client appointment in public health care centre
		Inpatient ward in public health care centre
		Home nursing or hospital-at-home
		Service housing
		Other
Documentation practice	Number of daily logins (=logins)	(1) One – five or more (5)
	Brand name of EHR/CIS (=EHR/CIS brand name)	Effica Health Care
		Apotti (Epic)
		Esko
		Lifecare
		Mediatri
		OMNI360
		Pegasos
		Other
Documentation hazards^a	During the last week, have you documented the same client/patient information more than once for the same care event? (=documentation hazard)	No, documenting information once is enough
		Yes, I document information in two different EHR/CIS
		Yes, first on paper, then into the EHR/CIS
		Yes, to be sure
		Yes, for some other reason^b
Technology-induced errors (TIE)^a	Faulty system function has nearly caused a serious adverse event for a patient.	(1) Fully disagree – fully agree (5)
Technology-induced errors (TIE)^a	Faulty system function has caused a serious adverse event for a patient.	(1) Fully disagree – fully agree (5)

^aDependent variables.

^bE.g., to ensure data is saved when the information system crashes and does not save the data; to provide information flow when patient care transitions to another place, for example, to an operation theatre, where specific auxiliary systems measure vitals that require separate documentation.

Statistical analysis

First, the study examined factors related to nursing documentation in a single location or several locations in EHR/CIS (documentation hazards) (Table 1). Second, it revealed factors associated with perceived TIEs besides nursing documentation. The factors used in the modelling include the respondents’ socio-demographic factors and working environments, as well as their perceived assessment of EHR/CIS that they use in their work with patients or clients.

While working environments do not constitute uniform units, their structural characteristics create a natural hierarchical arrangement in the data. This is particularly evident in the Finnish context, where the healthcare and social care sectors are strictly regulated, resulting in nurses being embedded within distinct institutional settings. The shared characteristics tend to induce intra-group similarity among nurses, violating the assumption of independent observations in conventional linear regression models and resulting in underestimated standard errors. To account for the hierarchical structure of the data and the clustering of respondents within regulated healthcare and social care environments,³⁷ we employed multilevel modelling techniques.³⁹ Specifically, we used fixed-effects logistic regression models to analyse five binary documentation hazard variables (e.g., duplicate documentation for the same care event), and linear multilevel regression models for two continuous technology-induced error (TIE) variables.

Maximum likelihood estimation (MLE) was used as the estimation method. Random intercepts were extracted to assess variation between work environments after controlling for individual-level covariates. The models included socio-demographic variables, work environment characteristics, and respondents’ subjective assessment of the electronic health record (EHR) and client information systems (CIS) used in their daily work. Cases with missing data on key variables were excluded from the analysis using listwise deletion.

All statistical analyses were conducted using Stata/SE for Windows 18.0. Descriptive statistics are presented in Appendix A (Tables 4–6).

Results

In total, 2970 respondents were included in the analysis. Most respondents (82%) were nurses, and most of them (85%) worked in healthcare, with 13% working in social care. Their mean age was 47 years (std. 10.6); the oldest was born in 1958, and the youngest in 2000, and 92% of the respondents were female.

First, the factors associated with perceived technology-induced errors (near misses and adverse events) were examined. The results for both perceived TIEs were similar, so only the technology-induced adverse events are reported here (Table 2). The results for near misses are presented in Appendix B (Table 7). Nurses’ skills with EHR/CIS, graduation year, or professional backgrounds were not associated with TIEs. In general, regardless of daily logins or perceived support for documentation, Apotti users reported more perceived TIEs (β ≈ 0.50, p < 0.001). The number of daily logins increased only slightly, whereas perceived TIEs (β ≈ 0.08, p < 0.01) increased, and perceived documentation support decreased (β ≈ −0.19, p < 0.05). Documentation hazards contributed to perceived TIEs. While documenting in one place decreased perceived TIEs (β = −0.30, p < 0.001), documenting in two locations “to be sure” (β = 0.46, p < 0.001), on paper first (β = 0.33, p < 0.001), and for other reasons (β = 0.48, p < 0.001) increased perceived TIEs.

Table 2.

Random intercept logistic regression models for factors associated with documentation hazards.

	M1	M2	M3	M4	M5
Health care professional (ref. Registered nurse)
Public health nurse	0.827	1.175	1.361	0.87	1.670*
Public health nurse	(0.61–1.12)	(0.72–1.91)	(0.93–2.00)	(0.48–1.57)	(1.05–2.65)
Other	0.420***	2.628***	1.619	2.385**	4.207***
Other	(0.29–0.62)	(1.54–4.48)	(0.95–2.76)	(1.33–4.27)	(2.58–6.85)
Brand name of EHR/CIS (ref. Lifecare)
Effica healthcare	1.347	1.394	0.6	0.627	0.68
Effica healthcare	(0.74–2.46)	(0.61–3.20)	(0.24–1.47)	(0.18–2.19)	(0.25–1.85)
Apotti (Epic)	0.932	0.305***	1.117	2.000**	1.081
Apotti (Epic)	(0.72–1.21)	(0.17–0.53)	(0.79–1.57)	(1.32–3.03)	(0.74–1.58)
Esko	0.517**	2.670***	1.444	2.903**	1.821
Esko	(0.34–0.79)	(1.51–4.73)	(0.79–2.63)	(1.47–5.72)	(0.97–3.40)
Mediatri	1.25	0.374*	0.956	1.219	0.340*
Mediatri	(0.83–1.89)	(0.17–0.84)	(0.56–1.64)	(0.61–2.42)	(0.14–0.82)
OMNI360	1.137	1.04	0.93	0.681	0.826
OMNI360	(0.83–1.57)	(0.65–1.67)	(0.61–1.43)	(0.37–1.27)	(0.50–1.36)
Pegasos	1.039	0.84	0.917	1.386	0.789
Pegasos	(0.72–1.50)	(0.48–1.48)	(0.56–1.51)	(0.72–2.67)	(0.42–1.48)
Other	0.766	1.726	1.06	1.391	0.851
Other	(0.47–1.24)	(0.94–3.16)	(0.54–2.09)	(0.60–3.24)	(0.38–1.88)
Number of logins	0.797***	1.571***	1.152**	1.197**	1.173**
Number of logins	(0.74–0.86)	(1.40–1.76)	(1.04–1.28)	(1.05–1.36)	(1.04–1.32)
Documentation support	1.218	0.746	1.109	0.822	0.635*
Documentation support	(0.96–1.55)	(0.50–1.12)	(0.80–1.53)	(0.57–1.19)	(0.44–0.92)
Skills needed to use EHR/CIS	0.992	1.169	1.37	1.053	0.769
Skills needed to use EHR/CIS	(0.74–1.33)	(0.72–1.91)	(0.93–2.02)	(0.69–1.60)	(0.49–1.19)
Skills * documentation support	1.051	0.969	0.862*	0.901	1.042
Skills * documentation support	(0.96–1.15)	(0.84–1.12)	(0.77–0.97)	(0.78–1.03)	(0.91–1.19)
Var (work environments)	1.088	1.27	1.011	1.487	1.191
Var (work environments)	(0.98–1.21)	(0.93–1.73)	(0.94–1.08)	(0.94–2.36)	(0.95–1.50)
Number of observations	2045	1382	1466	1353	1385

Odds ratios; 95% confidence intervals in brackets below.

Note. The dependent variables were derived from the multiple-choice question, “During the last week, have you documented the same client/patient information more than once for the same care event?” and are as follows: M1 = Documenting once is enough; M2 = Two different systems; M3 = First on paper, then into the system; M4 = Yes, to be sure; M5 = Other reason. EHR/CIS (electronic health records/client information systems) with brand names, the most common, Lifecare, was put in the reference group.

Next, the factors contributing to documentation hazards were examined (Table 3). Regardless of the system used or other factors, “others” documented significantly more in multiple locations, especially in two systems (OR = 4.52, p < 0.001) and for other reasons (OR = 4.68, p < 0.001). The most notable results were found between different brand names. Compared to the most used system (Lifecare), Apotti (Epic) and Esko were perceived as inferior. Regardless of how many systems nurses log in to daily, both systems increased the “to be sure” double documenting (OR = 2.36 & 3.39, p < 0.01). Expectedly, the number of daily logins decreased, with documentation concentrated in only one location (OR = 0.78, p < 0.001), while increasing double documentation, especially in two locations (OR = 1.65, p < 0.001). Nurses perceived that documentation support or user proficiency was not associated with documentation hazards on their own. However, the interaction between these variables contributed to documenting in one location (OR = 1.12, p < 0.05), in two locations (OR = 0.82, p < 0.05), and “to be sure” (OR = 0.84, p < 0.05).

Table 3.

The results multilevel logistic regression for factors associated with documentation hazards among nurses.

	M1	M2	M3	M4	M5
Health care professional (ref. Registered nurse)
Public health nurse	0.744	1.346	1.680*	1.014	1.811*
Public health nurse	(0.53–1.04)	(0.78–2.33)	(1.11–2.53)	(0.52–1.98)	(1.08–3.02)
Other	0.358***	4.516***	2.053*	3.081**	4.678***
Other	(0.23–0.57)	(2.41–8.46)	(1.11–3.81)	(1.56–6.10)	(2.62–8.37)
EHR/CIS brand name (ref. Lifecare)
Effica healthcare	1.56	1.649	0.404	0.655	0.529
Effica healthcare	(0.76–3.19)	(0.66–4.15)	(0.12–1.37)	(0.15–2.94)	(0.15–1.83)
Apotti (Epic)	0.852	0.399**	1.126	2.360***	1.173
Apotti (Epic)	(0.63–1.15)	(0.21–0.77)	(0.76–1.67)	(1.45–3.83)	(0.76–1.81)
Esko	0.408***	3.780***	1.737	3.389**	2.217*
Esko	(0.25–0.66)	(1.95–7.34)	(0.87–3.45)	(1.52–7.55)	(1.12–4.38)
Mediatri	1.186	0.575	0.789	1.199	0.459
Mediatri	(0.75–1.88)	(0.25–1.34)	(0.42–1.50)	(0.53–2.69)	(0.19–1.13)
OMNI360	1.046	1.182	1.004	0.805	0.791
OMNI360	(0.73–1.51)	(0.68–2.05)	(0.61–1.64)	(0.39–1.67)	(0.44–1.42)
Pegasos	1.107	0.877	0.947	1.101	0.669
Pegasos	(0.73–1.68)	(0.46–1.68)	(0.55–1.64)	(0.49–2.46)	(0.32–1.39)
Other	0.541*	3.504***	0.963	2.181	1.102
Other	(0.30–0.98)	(1.70–7.23)	(0.37–2.50)	(0.80–5.93)	(0.42–2.89)
Logins	0.780***	1.652***	1.175**	1.189*	1.241**
Logins	(0.71–0.85)	(1.44–1.89)	(1.05–1.32)	(1.02–1.39)	(1.09–1.41)
Documentation support	1.035	1.129	0.993	0.967	0.85
Documentation support	(0.81–1.32)	(0.72–1.77)	(0.71–1.39)	(0.65–1.43)	(0.58–1.24)
Skills	0.855	1.826	1.125	1.271	0.979
Skills	(0.61–1.20)	(0.99–3.37)	(0.71–1.78)	(0.78–2.08)	(0.58–1.66)
Skills * documentation support	1.121*	0.822*	0.896	0.838*	0.943
Skills * documentation support	(1.01–1.24)	(0.69–0.99)	(0.78–1.03)	(0.71–0.99)	(0.80–1.11)
Number of observations	1593	1086	1146	1054	1092

Odds ratios; 95% confidence intervals in brackets below; *p < 0.05 **; p < 0.01 ***; p < 0.001.

The interaction effects are depicted in Figure 1. In general, the figure suggests that documenting in one location results from a skilled nurse using a well-functioning system. Pane 1A shows that if the system is perceived to provide poor documentation support, the difference between nurses with fair and excellent skills is negligible. With good documentation support, nurses with excellent skills are over 20% more likely to document in one location than their colleagues with only fair skills. Pane 1B depicts this same interaction in relation to documenting in two locations. Here, the nurse’s improving abilities, combined with perceived poor documentation support, increase the probability of documenting in two locations. Lastly, pane 1C shows this effect in relation to double documenting “to be sure.” Nurses with excellent skills are approximately six times more likely to document in two locations “to be sure” when comparing systems that provide good and poor support for documentation.

Figure 1.

Interaction between system support and user proficiency supports single-location documentation.

Lastly, random intercepts from the above models were extracted to depict differences in documentation hazards across working environments, after accounting for other factors. According to the results (Figure 2), documenting in one location was the most common in inpatient/client appointments in public health services and the least typical during in-home nursing or hospital-at-home care. In inpatient wards of public health centres and hospitals, the predominant reason for documenting the same information in different locations was the “yes, to be sure” documentation hazard type. This remained the most common explanation even after controlling for working environment–related factors (random intercepts from Table 3).

Figure 2.

Documentation hazard types in working environments after related factors are controlled (random intercepts from Table 3).

Discussion

This study investigated the factors associated with documentation hazards and explored how these hazards, combined with nurses’ backgrounds, relate to perceived technology-induced errors. It is part of a broader research series examining the digital transformation of Finland’s healthcare and social care system. Specifically, the study focused on the use of electronic health records (EHR) in healthcare and client information systems (CIS) in social care.^34,35,37

Documentation hazards were measured by assessing how often users recorded the same information in multiple ways using EHR/CIS. While duplication through actions like copy-pasting¹² was not specifically studied, it was recognized that data was recorded in numerous locations within the same EHR. Documenting in one location was most common during inpatient or client appointments in public health services, and least common in in-home nursing or hospital-at-home settings. Conversely, documenting in multiple locations (“yes, to be sure”) was most frequent in hospital inpatient wards and public health centre wards. This practice appears to reflect both human and environmental factors, including faulty system design, poor interactions between system components,^6,21,40 and limited users’ knowledge of EHR functions.²⁴

The Finnish Patient Safety Strategy emphasizes the importance of a systematic and comprehensive orientation to new roles, environments, workplace devices, and information systems as prerequisites for safe professional practice.²⁶ Documenting in a single location minimizes the risk of adverse events, whereas documenting in multiple locations increases such risks. Even without immediate safety incidents, such hazards can delay care processes and increase professional workload.^3,20,25

Previous research has shown that poor usability and functionality are linked to safety concerns and workflow dysfunctions.^5,9,15,18 Documenting in two locations was most common when the work was performed in the client’s home, which may indicate poor interoperability between the information system and mobile applications. Furthermore, employees in the “Other” environment, e.g, laboratory, document significantly more frequently in two or more locations, reflecting differences between working environments. The working environment was closely tied to documentation hazards and strongly associated with incidents where faulty system functionality contributed to perceived technology-induced errors. Information overload may play a role, as the frequency of safety incidents has been shown to vary across hospital wards.¹⁷ Daily login frequency was also related to documentation hazards. More logins reduced the likelihood of documenting once and increased documentation in two locations, often across different systems. This behaviour may reflect workflow adjustments following the implementation of a new system.^13,26,27

An interaction effect was observed between perceived system support for documentation and user proficiency. While better system support reduces documentation hazards, the reduction is more pronounced when user proficiency simultaneously increases. However, when users perceive the system as unsupportive, higher proficiency is associated with a greater likelihood of using paper-first recording, suggesting that proficient users may revert to manual methods to ensure accuracy or reliability.

Several studies have emphasized the importance of education, as users may struggle with EHRs due to insufficient introduction and training.^16,22–24 Conversely, poor system support and skilful users are associated with increased documentation across multiple locations. This may indicate that experienced users, aware of the system’s shortcomings, adopt additional measures to ensure completeness. Nonetheless, documentation errors can occur regardless of qualifications, training, work experience, or working hours.¹⁵ Enhancing system support for documentation could reduce the need for workarounds, such as paper-first or double-recording, particularly among proficient users.²¹

The concept of technology-induced error (TIE) was first introduced in 2005.¹ Since then, the phenomenon has been extensively studied, particularly focusing on faulty system design, implementation, customization of EHR, and poor interactions between system components.^4,6 In this study, TIE was assessed using two statements:

“Faulty system function has nearly caused a serious adverse event for a patient.”

“Faulty system function has caused a serious adverse event.”

Factors associated with perceived TIEs were assessed using the most widely used system, Lifecare, as a reference. Among the examined systems, only one stood out negatively: Apotti (Epic). Our findings indicate that Apotti (Epic) is the only system associated with increased perceived TIEs. Despite several years of implementation, user feedback on Apotti (Epic) remains critical, although some improvements have been reported.^20,21,23,40 Importantly, the Finnish adaptation of the Epic system complies with national data structures and legal requirements,^32,36 yet user-reported safety concerns persist.

The hospital working environment plays a significant role in nurses’ evaluations of EHR usability and whether these systems meet their intended goals of improving care quality and safety.^27,29 This influence was evident in this study, regardless of the level of EHR adoption. While the number of daily logins slightly increased perceived TIEs, the system’s documentation support emerged as a crucial factor in reducing the risk of adverse events. Poor usability contributes to significant safety concerns and workflow inefficiencies.^5,12

Perceived TIEs were less common when documentation was available in a single location. In contrast, paper-first-based recording and documenting across multiple locations increased the likelihood of reported TIEs. These findings highlight the importance of robust system support in minimizing risks and improving workflow efficiency.

Strengths and limitations

Examining the factors associated with documentation hazards and how these hazards and nurses’ backgrounds relate to perceived technology-induced errors revealed the link between documentation practices and the interaction between system operations. The results cannot be generalized to countries with different EHR/CIS systems and legislation, or to countries where electronic documentation is not as widely used as in Finland.

The survey focused on self-reported data; for instance, it is uncertain what kind of knowledge and skills the participants possess or how they actually interact with the systems. Although we did not specifically ask what type of information they document in two places, “yes, to be sure,” or the use of paper for documentation first, besides electronic documentation, these customs are challenges for both system developers and introduction. Neither do we have any concrete information about the working environment, per se; we only know it as an administrative unit. Nurses work in both healthcare and social care, where both environments offer inpatient or home care services and long-term care.^34,35 The documentation may encounter difficulties related to whether a mobile version of the system is used. Sometimes, documentation is done at the office after home visits. Technology-induced errors occurred across various environments using EHRs and CISs; some systems appeared more prone to these errors.

While the study respondents represented the target group, as evidenced by similar findings in surveys conducted in 2017 and 2020,^20,23 the timing of data collection during the spring of 2023 may pose a limitation. This period coincided with the launch of the Wellbeing Service Counties, which could have influenced participation and responses.³³ As the context and practices of healthcare and social services evolve, new organizations frequently face challenges during their early stages. Thus, the context in which the study was conducted may have influenced the results. From one perspective, new organizations have new contact details, including new employee email addresses, which may contribute to difficulty reaching respondents.

The survey used for data collection had some limitations in addressing the research topic. The variables related to documentation hazards and TIEs were limited to two statements based on nurses’ experiences.³⁷ In contrast, earlier studies have utilized patient safety incident reports to classify types of TIEs.^3–5 While such reports are valuable for identifying potential safety hazards, they do not provide insights into the magnitude of these hazards.^8,14 Despite these limitations, the survey gathered data from nurses working in diverse environments and offered valuable insights into real-world documentation hazards.

Conclusions

The interaction between system support and user proficiency shapes documentation practices. Without adequate system support, even skilled users may resort to practices such as paper-first recording, which reflects inefficiencies or a lack of trust in the system. Improving system design and documentation support is a desirable goal, but it is not sufficient to mitigate documentation hazards and promote efficient practices. To achieve the best possible results, skilled users are needed to operate these systems. Introduction and continuous training should focus on building proficiency and enhancing user confidence in effectively using the system’s features across various work environments.

Footnotes

Acknowledgments

The authors wish to thank the Finnish Nurses Association and the National Association of Health and Welfare Professionals (TEHY) for distributing the survey to their members, as well as all registered nurses who responded.

ORCID iD

Kaija Saranto

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Appendix

References

Kushniruk

Triola

Borycki

, et al. Technology-induced error and usability: the relationship between usability problems and prescription errors when using a handheld application. Int J Med Inf 2005; 74(7-8): 519–526.

Sittig

Singh

. Defining health information technology-related errors: new developments since to err is human. Arch Intern Med 2011; 171(14): 1281–1284.

Magrabi

Liaw

Arachi

, et al. Identifying patient safety problems associated with information technology in general practice: an analysis of incident reports. BMJ Qual Saf 2016; 25(11): 870–880.

Westbrook

Reckmann

, et al. Effects of two commercial electronic prescribing systems on prescribing error rates in hospital inpatients: a before and after study. PLoS Med 2012; 9(1): e1001164.

Palojoki

Saranto

Reponen

, et al. Classification of electronic health record-related patient safety incidents: development and validation study. JMIR Med Inform 2021; 9(8): e30470.

Borycki

Kushniruk

Bellwood

, et al. Technology-induced errors: the current use of frameworks and models from the biomedical and life sciences literature. Methods Inf Med 2012; 51(2): 95–103.

Saranto

Kuusisto

Vehko

. A review of technology-induced errors. Stud Health Technol Inf 2025; 323: 463–467.

Sittig

Wright

Coiera

, et al. Current challenges in health information technology-related patient safety. Health Inf J 2020; 26(1): 181–189.

Lindén-Lahti

Kivivuori

Lehtonen

, et al. Implementing a new electronic health record system in a university hospital: the effect on reported medication errors. Healthcare 2022; 10(6): 1020.

10.

Martin

Ghafur

Cingolani

, et al. The effects and preventability of 2,627 patient safety incidents related to health information technology failures: a retrospective analysis of 10 years of incident reporting in England and Wales. Lancet Digit Health 2019; 1(3): e127–e135.

11.

McCarthy

Fitzgerald

O’Shea

, et al. Electronic nursing documentation interventions to promote or improve patient safety and quality care: a systematic review. J Nurs Manag 2019; 27(3): 491–501.

12.

Davy

Borycki

. Copy and paste in the electronic medical record: a scoping review. Knowl Manag E-Learn 2021; 13(4): 522–535.

13.

Tubaishat

. The effect of electronic health records on patient safety: a qualitative exploratory study. Inf Health Soc Care 2019; 44(1): 79–91.

14.

Committee on Patient Safety and Health Information Technology, Institute of Medicine . Health IT and patient safety: building safer systems for better care. National Academies Press, 2011.

15.

Kutney-Lee

Sloane

Bowles

, et al. Electronic health record adoption and nurse reports of usability and quality of care: the role of work environment. Appl Clin Inf 2019; 10(1): 129–139.

16.

Jachan

Müller-Werdan

Lahmann

. Patient safety: factors for and perceived consequences of nursing errors by nursing staff in home care services. Nurs Open 2021; 8(2): 755–765.

17.

Nijor

Rallis

Lad

, et al. Patient safety issues from information overload in electronic medical records. J Patient Saf 2022; 18(6): e999–e1003.

18.

Davis

Shepheard

. Clinical documentation integrity: its role in health data integrity, patient safety and quality outcomes and its impact on clinical coding and health information management. Health Inf Manag 2024; 53(2): 53–60.

19.

Feely

Edbrooke

Bower

, et al. Allied health professionals’ experiences and lessons learned in response to a big bang electronic medical record implementation: a prospective observational study. Int J Med Inf 2023; 176: 105094.

20.

Vehko

Hyppönen

Puttonen

, et al. Experienced time pressure and stress: electronic health records usability and information technology competence play a role. BMC Med Inf Decis Making 2019; 19: 160.

21.

Hertzum

Ellingsen

Cajander

. Implementing large-scale electronic health records: experiences from implementations of epic in Denmark and Finland. Int J Med Inf 2022; 167: 104868.

22.

Hübner

Shaw

Thye

, et al. Technology informatics guiding education reform – TIGER. Methods Inf Med 2018; 57(Suppl 1): e30–e42.

23.

Kinnunen

Kuusisto

Koponen

, et al. Nurses’ informatics competency assessment of health information system usage: a cross-sectional survey. Comput Inform Nurs 2023; 41(11): 869–876.

24.

Lee

Caton

Knight

. Evaluating digital competencies for allied health professionals in the United Kingdom. Digit Health 2023; 9: 20552076231176658.

25.

Nordic Council of Ministers . A nordic framework for patient safety knowledge and skills: competence areas for safe healthcare. Nordic Council of Ministers, 2024.

26.

Sloss

Abdul

Aboagyewah

, et al. Toward alleviating clinician documentation burden: a scoping review of burden reduction efforts. Appl Clin Inf 2024; 15(3): 446–455.

27.

Forde-Johnston

Butcher

Aveyard

. An integrative review exploring the impact of electronic health records (EHR) on the quality of nurse-patient interactions and communication. J Adv Nurs 2023; 79(1): 48–67.

28.

Robertson

Rosbergen

ICM

Burton-Jones

, et al. The effect of the electronic health record on interprofessional practice: a systematic review. Appl Clin Inf 2022; 13(3): 541–559.

29.

Clarke

Ashrafian

, et al. The impact of electronic health record interoperability on safety and quality of care in high-income countries: systematic review. J Med Internet Res 2022; 24(9): e38144.

30.

Chishtie

Sapiro

Wiebe

, et al. Use of epic electronic health record system for health care research: scoping review. J Med Internet Res 2023; 25: e51003.

31.

Moy

Schwartz

Chen

, et al. Measurement of clinical documentation burden among physicians and nurses using electronic health records: a scoping review. J Am Med Inf Assoc 2021; 28(5): 998–1008.

32.

Keskimäki

Tynkkynen

Reissell

, et al. Finland: health system review. Health Syst Transit 2019; 21(2): 1–166.

33.

Karanikolos

Tynkkynen

Keskimäki

. Finland: health system summary. European Observatory on Health Systems and Policies, 2024. [cited 2024 Sep 16]. Available from: https://www.euro.who.int/en/about-us/partners/observatory/publications/health-system-reviews-hits/full-list-of-country-hits/finland-hit-2019

34.

Reponen

Keränen

Ruotanen

, et al. Availability and use of e-health in Finland. In: Vehko

(ed). E-health and e-welfare of Finland: checkpoint 2022. Finnish Institute for Health and Welfare (THL), 2022, pp. 61–95.

35.

Salovaara

Silén

Kyytsönen

, et al. Availability and use of e-welfare in Finland. In: Vehko

(ed). E-health and e-welfare of Finland: checkpoint 2022. Helsinki: Finnish Institute for Health and Welfare (THL), 2022, pp. 96–114.

36.

Finland . Act on processing client data in social and health care 703/2023. Finlex Data Bank, 2023. [cited 2024 Sep 10]. Available from: https://www.finlex.fi/fi/laki/smur/2023/20230703

37.

Hyppönen

Lääveri

Hahtela

, et al. Smart systems for capable users? Nurses’ experiences on patient information systems 2017. Finn J eHealth eWelfare 2018; 10(1): 30–59.

38.

Finnish National Board on Research Integrity (TENK) . The ethical principles of research with human participants and ethical review in the human sciences in Finland. TENK, 2019. [cited 2024 Nov 20]. Available from: https://tenk.fi/sites/default/files/2021-01/Ethical_review_in_human_sciences_2020.pdf

39.

Snijders

TAB

Bosker

. Multilevel analysis: an introduction to basic and advanced multilevel modelling. 2nd ed. Sage, 2012.

40.

Hertzum

Ellingsen

. The implementation of an electronic health record: comparing preparations for epic in Norway with experiences from the UK and Denmark. Int J Med Inf 2019; 129: 312–317.