Detecting Patient Safety Errors by Characterizing Incidents Reported by Medical Imaging Staff

BRIEF RESEARCH REPORT

published: 18 March 2022

doi: 10.3389/fpubh.2022.846604

Frontiers in Public Health | www.frontiersin.org 1 March 2022 | Volume 10 | Article 846604

Edited by:

Sandra C. Buttigieg,

University of Malta, Malta

Reviewed by:

Francis Zarb,

University of Malta, Malta

Guilherme Couto,

University of Malta, Malta

*Correspondence:

Tarja Tarkiainen

tarja.tarkiainen@ppshp.ﬁ

Specialty section:

This article was submitted to

Radiation and Health,

a section of the journal

Frontiers in Public Health

Received: 31 December 2021

Accepted: 22 February 2022

Published: 18 March 2022

Citation:

Tarkiainen T, Sneck S, Haapea M,

Turpeinen M and Niinimäki J (2022)

Detecting Patient Safety Errors by

Characterizing Incidents Reported by

Medical Imaging Staff.

Front. Public Health 10:846604.

doi: 10.3389/fpubh.2022.846604

Detecting Patient Safety Errors by

Characterizing Incidents Reported by

Medical Imaging Staff

Tarja Tarkiainen

, Sami Sneck

, Marianne Haapea

, Miia Turpeinen

and

Jaakko Niinimäki

Research Unit of Medical Imaging, Physics and Technology, Oulu University Hospital, University of Oulu, Oulu, Finland,

Administrative Centre, Oulu University Hospital, Oulu, Finland,

Medical Research Centre, Oulu University Hospital,

University of Oulu, Oulu, Finland,

Administrative Centre, Research Unit of Biomedicine, Oulu University Hospital, University

of Oulu, Oulu, Finland

The objectives of the study were to characterize events related to patient safety reported

by medical imaging personnel in Finland in 2007–2017, the number and quality of

reported injuries, the risk assessment, and the planned improvement of operations.

The information was collected from a healthcare patient safety incident register system.

The data contained information on the nature of the patient safety errors, harms and

near-misses in medical imaging, the factors that lead to the events, the consequences

for the patient, the level of risks, and future measures. The number of patient safety

incident reports included in the study was 7,287. Of the incident reports, 75% concerned

injuries to patients and 25% were near-misses. The most common consequence of

adverse events and near-misses were minor harm (37.2%) related to contrast agent,

or no harm (27.9%) related to equipment malfunction. Supervisors estimated the risks

as low (47.7%) e.g., data management, insigniﬁcant (35%) e.g., verbal communication

or moderate (15.7%) e.g., the use of contrast agent. The most common suggestion for

learning from the incident was discussing it with the staff (58.1%), improving operations

(5.7%) and submitting it to a higher authority (5.4%). Improving patient safety requires

timely, accurate and clear reporting of various patient safety incidents. Based on

incident reports, supervisors can provide feedback to staff, develop plans to prevent

accidents, and monitor the impact of measures taken. Information on the development

of occupational safety should be disseminated to all healthcare professionals so that the

same mistakes are not repeated.

Keywords: incident reporting, medical imaging, errors, patient safety, risks

INTRODUCTION

The World Health Organization (WHO) has constructed global priorities for research to collect

information on patient safety incidents. The focus is on the underlying processes and organizational

factors that lead to unsafe care a nd adverse events (

1, 2). The goal is to help organizations

prevent a nd reduce the occurrence of serious incidents (

3). Howe ver, it is well-known that

although information on incidents is collected, some are always underreported, under-analyzed,

and under-utilized (4–7).

Tarkiainen et al. Analyzing Patient Safety Errors

Countries use diﬀerent methods to investigate and analyze

healthcare incidents. For example, the USA uses the root

case analysis method of the Joint Commission (8). At th e

beginning of 2000, the development of patient safety incident

reporting systems was initiated in 13 European countries:

Austria, Belgium, the Czech Republic, Denmark, Ireland, France,

the Netherlands, Norway, Scotland, Spain, Sweden, Switzerland,

and the United Kingdom. These countries collect information

on patient safety incidents on three diﬀerent levels: only sentinel

events (based on law), speciﬁc clinical domains, or healthcare

system-wide (including near-misses) (

9).

Many countries have their own national and local error

reporting systems aiming to identify gaps in the system. If

events are only known to individuals, countermeasures cannot be

developed. The data, especially on medication-related incidents,

are collected from the Medication Error Reporting System

(MERS), which is used in the United Kingdom and the

United States (4, 9, 10). However, the role of medical imaging

in these incidents is not so well-known (2). Australia and New

Zealand use a database called the Radiology Events Re gister (The

RaER). Reporting incidents on the RaER is web-based, voluntary,

conﬁdential, and anonymous. The system collects information

via narrative texts about events, their outcomes and contributing

factors, and how to prevent or reduce them. The RaER does not

replace state and hospital-based incident reporting, but it has

been de c lared a quality assurance resource (2, 11).

In 2005 in Finland, the Ministry of Social Aﬀairs and Health

launched national-level coordination and strategic guidance for

patient safety measures. A steering group focused on three areas

of patient safety: education and culture, tools, and reporting

(

9). The Finnish healthcare system is based on municipal

primary care (about 190 units), specialized care (20 hospital

districts), occupational health services, and private services. It

is decentralized and has multiple funding resources (e.g., the

state, municipalities, households, voluntary private, and state

national health insurance organizations). Hospital districts guide

and control the de velopment of imaging services, and other

special services provided by municipal healthcare research,

development, and training activities of coordinated municipal

health information systems (9, 12, 13).

Finland utilizes an electronic Reporting System for Safety

Incidents (RSSI) in Health Care Organizations (HaiPro, Awanic

Ltd 2015c). The system was developed in 2006 and launched

in 2007. The aim of the RSSI is to increase the documentation

of incidents and to utilize this dat a to improve the quality and

safety of care (

14). Staﬀ report medical and nursing treatment

incidents, including near-misses. A further aim of the system

is to report, analyze and learn of incidents, near-misses and

patient safety risks, and to improve patient care processes. The

RSSI is designed for internal use in h ealthcare units (university

and local hospitals, social units, and health centers) and it is

accessed online via the organization’s intranet. Use of the system

is voluntary, anonymity, and conﬁdentiality are guaranteed, and

Abbreviations: FSPS, The Finnish Society for Patient Safety; MERS, Medical Error

Reporting System; RaER , t he Radiology Events Register; RSSI, Reporting Systems

for Safety Incidents.

no one is blamed. It provides information on the data reported,

how local organizations have learnt from their incidents, and how

the process of patient safety has been improved. The Finnish

Society for Patient Safety (FSPS) manages the research permit

practices related to the patient incident reports (

9, 15, 16).

Finnish healthcare professionals can report diﬀerent kinds

of safety incidents, including medical imaging errors. The dat a

are collected using web-based forms that contain predeﬁned

selection lists and spaces for free text (14, 16, 17). At present,

RSSI d ata are mainly collected on the local and organizational

level and are not analyzed or utilized on the regional or national

level (9, 14–16).

The purpose of this study was to characterize the medical

imaging reports of incidents on the national level. To our

knowledge, this has not previously been studied in Finland. Our

aim was to obtain information on near misses and adverse events

of medical imaging. In addition, we wanted information on how

the recurrence of incidents were prevented.

MATERIALS AND METHODS

Data Collection

This study was an applied research that aims to highlight the

practical problems in patient safety and ﬁnd solutions to them.

Awanic Ltd collected the dat a from a web-based national error-

reporting dat abase and sent it to the researcher (ﬁrst writer). The

data comprised safety incidents (N = 7,409) related to medical

imaging that posed risks to patient safety in 2007–2017. The

study involved 125 radiology departments from 18 regional and

university hospital or other health service districts (Awanic L td).

The FSPS granted permission to use the medical imaging

error data in the present study. In Finland, a registry study does

not require ethical permission. The study followed t he basic

principles of research ethics (e.g., anonymity, conﬁdentiality).

The RSSI database is not linked to other databases, such as

patient records. Disclosure of the data is subject to permission.

As participation was voluntary, the medical units had t h e option

of not participating in the study.

Data on Medical Imaging Incidents

Incident reports (N = 7,409) were collected from January 2007

to December 2017, and 7,287 reports were included. Reports

on nuclear medicine, radiotherapy, internal tests, duplicates, and

reports not associated with imaging (n = 122) were excluded. All

medical imaging-related notiﬁcations were included, even if the

informer was not imaging staﬀ (e.g., nurse, physician, secreta ry,

or other). As some of the informers had not ﬁlled in every section

of the reports, some information was missing. However, forms

were included if the most rele vant sections had been completed.

The information on the web-based forms and in the free-

text descriptions that was related to medical imaging incidents

were analyzed. Figure 1 presents the patient incident reporting

process. The following information was reported: date, time,

place (e.g., medical imaging) of the incident, and the unit of

the person reporting. It was also possible to report the nature

of the incident (adverse event, near miss) and the incident

type, and to more speciﬁcally describe the event and how it

Frontiers in Public Health | www.frontiersin.org 2 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

FIGURE 1 | The sequence of the patient incident reporting process.

occurred, t he consequences for the patient and the unit, the

conditions at the time of the incident, and other contributing

factors. The staﬀ also classiﬁed the causes of the inci dents into

eight main categories; drug or contrast agent, information ﬂow or

information management, other treatment or follow-up, imaging

(not further speciﬁed), device or its use, accident/injury/damage,

an invasive procedure, or diagnosis related.

The supervisor, usually the radiographer in charge or senior

radiologist reviews the incident report and assigns a level of risk

according to pre-deﬁned categories: insigniﬁcant, low, moderate,

high, and extreme. The supervisor also includes the contributing

factors of the incident and des cribes measures to be taken to

prevent the event reoccurring: (1) no measures, (2) informing of

the event, (3) submitting it to a higher aut h ority for a decision, or

(4) designing improvement measures (14, 15).

The data were processed statistically using Microsoft 365

Excel (2016) and IBM SPSS statistics 25 (SPSS Inc., Chicago, IL,

USA) for Windows. The results were reported using frequencies,

percentages and cross-t abulation. The qualitative data will be

analyzed in more detail in a subsequent article.

RESULTS

Identiﬁcation of Adverse Procedural

Incidents Associated With Medical Imaging

This study included 7,287 adverse incidents that aﬀected the

safety of patients undergoing radiological procedures. The

number of adverse incidents reported by staﬀ rose (from 2007)

by 10 per annum to 1,438 per annum over 10 years, making

an average annual increase of 21.2%. Of the incidents reported,

75% concerned injuries to patients and 25% were near misses.

Most of the patient injuries (28.9%) and near-misses (35%)

were associated with the wrong patient, study or procedure.

The other reasons for near miss reporting were for instance:

data ﬂow and data management (30%), malfunction of device

(10%), error of radiology report or diagnosis (6%), and referral

error (3%). The other reasons concerning patient injuries were:

related to medication or contrast agent (17%), data ﬂow and

data management (15%), malfunction of device (14%), error of

radiology report or diagnosis (10%), and referral errors (2%).

The most common consequences of the near-misses and

adverse events were minor harm (37.2%) or no harm (27.9%).

The most serious eﬀects on patients were moderate harm, at 7 .7%

(563 cases), and serious harm, at 0.8% (61 cases). In 10.5% of the

events, the consequences were unknown (e.g., a patient moved

to another ward) and 15.8% were not assessed at all (missing).

During 2007–2011, t he consequences of events were assessed

only occasionally, but from 201 2, more regularly. The most

common injuries to the patients were unnecessary or excessive

radiation (wrong patient, side, or site), allergic re action, burn,

infection, falling, fainting , pain, aches, extra dose of contrast, and

contrast media to tissue.

After the staﬀ has assessed th e consequences, supervisors

assess the risk posed by the incident. The risk analysis matrix

allows a supervisor to select the risk category. A supervisor selects

the likelihood of an event: rare, unlikely, possible, likely, and

almost certain; and its typical consequences: very minor, minor,

moderate, major, or severe. These are divided into diﬀerent risk

types: insigniﬁcant (e.g., data ﬂow or data management) low

(e.g., equipment malfunction), moderate (use of contrast agent or

accident), high (unexpe ct ed reaction of the patient), and extreme

risk (serious injury to the patient). A supervisor may assess

the incident as rare and its consequences very minor, in which

case its risk type is insigniﬁcant or the probability of incident

almost certain with severe consequences in which case the risk

is extreme.

Supervisors assessed 69.4% (5,011 in total) of the patient risks,

although 30.6% (about 200 per year) were not evaluated. In

2010–2017, 1.3% (66) of all risks were assessed as high and 0.2%

(10) as extreme. Risks were generally estimated as low (47.7%),

insigniﬁcant (35%), or moderate (15.7%). Figure 2 presents the

annual distribution of the reported incidents by risk leve l and

the number of reports during 2010–2017. In 2007–2009, th e risk

assessment was limited (14 reports) and are thus not taken into

account in this ﬁgure.

The highest number of patient harm-related reports

concerned computer tomography CT (27%); the second,

problems with computers and telecommunicat ions (25.6%);

and the third, plain radiography (16.9%). Figure 3 shows

how the consequences for patients were distributed among

diﬀerent modalities, communications, and other causes. Most

Frontiers in Public Health | www.frontiersin.org 3 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

FIGURE 2 | Annual reports and risk assessments in medical imaging in Finland 2010–2017. *In 2007–2008, the risks were not assessed, and in 2009 only 14 were

assessed, so these years are not included in the ﬁgure. **High and extreme risks are combined (n = 17).

of the severe and moderate harm cases occurred in ﬂuoroscopy

(2.8%) and mammography (2.1%). Approximately 73% of CT

scan-related injuries were no or minor harm, and 15% were

moderate or severe.

Incident Reporting by Staff

Most incidents were reported by radiographers (62.1%), other

nursing staﬀ (e.g., practical nurse) (21%), other s taﬀ (e.g.,

oﬃce worker) (7.9%), and radiologists or other physicians

(7.4%). Event types were evaluated in 7,083 reports (97.2%).

Staﬀ assessed near-misses and adverse events using the RSSI

form’s predetermined categories. About 26% of the patient

consequences were not assessed at all or were reported as

unknown. This is probably due to th e fact that the informer is

not working on radiology unit or the patient moves to another

ward so the late eﬀects of the event remain unknown. The most

frequently reported type was associated with imaging (40%),

but this was not further speciﬁed. A total of 20 .1% events were

related to information ﬂow or information management, 13.9%

concerned devices, and 13.4% were related to contrast media.

Staﬀ reported fewer events relat ed to patient care or follow-up

(4.6%), accidents or injuries (4%), diagnostic events (2.3%), and

invasive procedures (1.8%).

Staﬀ did not assess the extent of t he risk. Supervisors estimated

risks on the basis of the narrative content of the report and the

risk matrix. Assessments by nursing supervisors showed that the

patient was most likely to be at a low (40.7–50.3%) or insigniﬁcant

risk (27.6–36.3%). The share of moderate risk was from 14.3 to

26.3%, high risk under 2% (0.9–1.9%), and extreme risk under

1% (0–0.1%). Medical super v isors estimated that the patient was

at a high risk in 3.5% and at an extreme risk in 1.6% of the cases.

Supervisors’ Proposals for Measures

A measure to prevent incidents was proposed by 89.4% of the

supervisors. Generally (58.1%) supervisors suggested discussing

in staﬀ meetings. In under 6% (5.7%) of incidents improvement

in working methods were recommended and in 5.4% of cases

reports were delegated to a higher authority. Figure 4 shows the

distribution of these measures in relation to the patient’s incident

assessment. In all categories, the most common recommendation

by the supervisors was to inform the parties that were involved

(patient, ward, attending physician) and to discuss about the

incident with staﬀ (52.6%). In the severe harm category, 22.8%

were submitted to a higher aut h ority and 19.3% to plan an

improvement measure. The most general solution with minor

harm category was “no action” (30%) and for preventing

a recurrence of the event recommended, “informing of and

Frontiers in Public Health | www.frontiersin.org 4 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

FIGURE 3 | Patient injuries assessed by supervisors in different modalities and other contexts in Finland 2010–2017. *Missing or unknown 26.3% (n = 1,916).

discussing” (67%). When the patient injury was assessed as minor

(e.g., extravasation during intravenous injection), it was classiﬁed

as “no action” category.

DISCUSSION

In this study, we a nalyzed what kind of medical imaging errors

were reported in Finland, how these incidents and consequences

aﬀected risks, and what measures supervisors took to improve

patient safety. The strength of the study is that th e data were

large and comprehensive, covering 11 years. The average annual

number of radiology examinations in Finland is 2.5 million,

which is 1,084 examinations per thousand inhabitants. The

percentage of reported patient adverse events i s low, only 0.06%

of radiology examinations per year. It is likely that only a part of

the incidents were reported also in this study, as previous studies

have found t hat some cases always remain unreported (

2, 18).

In about a quarter of the cases, the injuries to the patient

had not been assessed. This result support those of international

studies that have identiﬁed diﬃculties in reporting beca use the

impact on the patient is unknown or the informer is from

another unit, in which case the information obtained from the

event is inaccurate (19–23). In the initial years, 2007–2011, forms

were only ﬁlled in occasionally, but this clearly increased in

later years. The probable reason for this is the greater national

attention paid to patient safety in Finland due to the Patient

Safety Implementation Plan (341/2011). This has increased the

identiﬁcation and reporting of incidents, and more hospitals

joining the RSSI (

24). Our research shows that when staﬀ

were encouraged to ﬁll in adverse incident forms, the number

of patient safety incident reports increased by 20% annually.

Therefore, we cannot directly conclude that the number of

complications increased every year.

In 2015, 624 radiologists and 3,262 radiographers were

working in Finland. In this study, most of the reporters were

radiographers and other nursing staﬀ. O ther physicians and

radiologists reported only 7.5%. In other Finnish studies the

number of incident reports made by physicians have been 2.5–

10% (

9, 17). Our result is comparable to those of previous

international studies that have found that physicians make fewer

incident reports (25, 26) and most patient incident reports are

made by nursing stuﬀ (19, 30). The physicians feel more obliged

to deal with serious cases than reporting near misses (19, 20, 30)

whereas nurses are more diligent and follow t he instructions (26).

Supervisors are provided with guidelines on how to handle

notiﬁcations. These guidelines oﬀer advice on selecting from the

diﬀerent options the correct type of event, the consequences for

the patient, the unit, the correct risk category, and suggested

development measures. The event should be evaluated according

to general practice, not on the basis of the individual patient.

This means that the risk assessment must take into account the

wider context of the individual case, which can be challenging.

Therefore, supervisors may assess risks as insigniﬁcant or minor,

even if the patient was harmed. Radiologists reported more

serious harm (2.2%) than radiographers in charge (0.7%) did,

and their estimated risk levels were higher than those of the

other staﬀ. This result is consistent with that of previous studies

(

23, 25, 26) that have found that physicians usually report the

Frontiers in Public Health | www.frontiersin.org 5 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

FIGURE 4 | Reported patient harms with proposed measures in 2007–2017 in Finland.

most serious patient injuries (se vere injury or infection, death).

In Finland this is probably due to the fact that the most severe

cases are referred to a radiology supervisor.

During the initial years, both physicians’ and radiographers’

information technology (IT)-related notiﬁcations were relatively

high in number. The number decreased each year as IT problems

were solved. There are usually problems when starting a new

operation, and knowing this risk, having a contingency plan to

investigate operational faults, and striving to remedy them as

soon as possible are essential.

On average, 10.5% (5.6–13.6%) of patient injuries related

to modalities were unclassiﬁed be cause the person ﬁling the

report did not know what had occurred to the patient after the

examination. Our previous study of patient injury reports due

to imaging (

27) provides more information on these incidents.

Usually for example, in interventions, patient injury is noticed

immediately, but in other modalities, the eﬀect of the event may

become apparent later.

In CT, 24.3% of incidents were related to contrast agents, due

to leakage of the contrast media into the surrounding soft tissues

instead of the normal intravascular compartment, failed injection

(the cannula came oﬀ or contrast agent was injected into tissues

or the contrast agent syringe did not work), or patients’ allergic

reactions. Contrast media extravasation (CMEV) is a well-known

complication in CT scanning. Most extravasations cause minimal

swelling or erythema, but skin necrosis may also occur (28).

Therefore, eﬀorts should be made to reduce the migration of

the contrast medium into t h e tissue: particularly in CT scanning,

where th e injection rate and volume of contrast medium is high.

The modalities most involved in patient incidents were CT,

radiography, and MRI. Our results are consistent with those of

other studies (

2, 27, 29), with t h e exception that in this study, MRI

incidents were more common than ultrasound incidents. This is

probably because in Finland, more contrast-enhanced MRI scans

are performed than contrast-enhanced ultrasounds. The higher

rates of moderate harms in ﬂuoroscopy and mammography were

related to information ﬂow, incomplete or unclear information

and equipment problems. Severe harms, such as serious infection

or accident, were related to invasive procedures.

This study has some limitations. First, the classiﬁcation of

adverse events and near-misses was based on the incident

description, which was occasionally limited and not all sections

of the form had always been completed. Therefore, the “not

known” and “missing” category made up 20–30% of incidents.

Second, as only a small proportion of incidents are reported,

and even less near-misses (

4–7), our data most likely represents

only a part of th e actual problems in incident reporting. Third,

because the RSSI form does not contain own category for imaging

staﬀ, the exact number of incidents cannot be measured. Fourth,

the results can only be generalized to a certain extent, as they

Frontiers in Public Health | www.frontiersin.org 6 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

were collected in one country. And ﬁfth unfortunately, the

supervisors’ development proposals for resolving the incident

that had occurred were in written form, and due to the large

amount of data, could not be discussed in more detail in

this artic le.

In our study, the number of reported near-misses was lower

(26.3%) than the number of incidents (72.6%). This ma y be

because incidents are easier to identify and staﬀ consider it

more important to report events that have consequences for the

patient. It is important to note that when we invest in patient

safety by collecting data on near-misses, we can also use them

in eﬀe ct on damage prevention (30, 31).

Supervisors planned some sort of measure to prevent t he

recurrence of incidents in almost 90% of c ases. The most

common solution, at over 50%, was discussion with the staﬀ.

Conversation is a quick and easily organized solution, but in our

opinion not necessarily the most eﬀective. It is good practice

and can solve the problem, but if the incident is repeated, it

would be better to investigate adverse event’s root causes. It is

important to notice that even when the patient was not harmed

some supervisors planned a development of operational safety. It

is good that all cases are taken seriously.

The more severe the damage, the hig her the proportion of

improvement measures and notiﬁcations at a higher level. It

is common that when the harm is assessed as severe (serious

injury or death of the patient, resource shortage or IT problems)

the proportion of improvement measures and notiﬁcations are

submitted to a higher authority (

30). This is probably due to the

desire to receive support and at th e same time inform the senior

management of the most serious events. Senior management

should also answer questions about resourcing and communicate

with equipment and other suppliers.

CONCLUSION

Improving patient safety requires timely, accurate and clear

reporting of various patient safety deviations. Therefore, it is

important that national data on patient safety incidents are

collected and analyzed. Based on the incident reports, supervisors

should provide feedback to staﬀ, develop a plan to prevent

accidents, and monitor the impact of the measures taken.

Information on the development of operational safety should be

disseminated widely to all healthcare professionals so that others

can avoid making the same mistakes and improve their own

patient safety by le a rning from others.

IMPLICATIONS FOR PRACTICE

• To prevent incidents, near misses should be reported

more frequently.

• Information on measures tak en to improve operations needs

to be more widely disseminated.

DATA AVAILABILITY STATEMENT

The datasets presented in this article are not readily available

because, the data that support the ﬁndings of this study

are available from the Finnish Society for Patient Safety but

restrictions apply to the availability of these data, which were

used under license for the current study, and so are not publicly

available. Data are however available from the authors upon

reasonable request and with permission of the Finnish Society for

Patient Safety. Requests to access the datasets should be directed

to https://tarja.tarkiainen@ppshp.ﬁ.

ETHICS STATEMENT

Ethical review and approval was not required for the

study on human participants in accordance with the

local legislation and institutional requirements. Written

informed consent for participation was not required for this

study i n accordance with the national legislation and the

institutional requirements.

AUTHOR CONTRIBUTIONS

TT analyzed and interpreted the data together with JN. MT

and SS veriﬁed and assessed the accuracy of the text. MH as

a professional bio statistician veriﬁed the table and ﬁgures. All

authors read and approved the ﬁnal manuscript.

REFERENCES

1. World Health Organization. Global Priorities for Research in Patient Safety.

(2008). Available online at: https://www.who.int/patientsafety/research/

priorities/global_priorities_patient_safety_research.pdf (accessed December

27, 2021).

2. Jones DN, Hannaford N , Thomas MJ, Mandel CJ, Grimm J, Schultz

TJ, et al. Where failures occur in the imaging care cycle: lessons

from the radiology events register. J Am Coll Radiol. (2010) 7:593–

602. doi: 10.1016/j.jacr.2010.03.013

3. Kruskal JB, Siewert B, Anderson SW, Eisenberg RL, Sosna J. Managing

an acute adverse event in a radiology department. RadioGraphics. (2008)

5:1237–50. doi: 10.1148/rg.285085064

4. Noble DJ, Pronovost PJ. Underreporting of patient safety incidents reduces

health care’s ability to quantify and accurately measure harm reduction. J

Patient Saf. (2010) 4:247–50. doi: 10.1097/PT S.0 b0 13 e3 18 1f d169 7

5. Rafter N, Hickey A, Condell S, Conroy R, O’Connor P, Vaughan D, et al.

Adverse events in healthcare: learning from mistakes. QJM. (2015) 4:273–

7. doi: 10.1093/qjmed/hcu145

6. Murﬀ HJ, Patel VL, Hripcsak G, Bates DW. Detecting adverse events for

patient safety research: a review of current methodologies. J Biomed Inform.

(2003) 36:131–43. doi: 10.1016/j.jbi.2003.08.003

7. Classen DC, Resar R, Griﬃn F, Federico F, Frankel T, Kimmel N, et

al. ‘Global trigger tool’ shows that adverse events in hospitals may be

ten times greater than previously measured. Health Aﬀ. (2011) 4:581–

9. doi: 10.1377/hlthaﬀ.2011.0190

8. Parker J. Root Cau se Analysis in Health Care Tools and Techniques. The Joint

Commission Resources (2017).

9. Doupi P. Nation al Reporting Systems for Patient Safety Incidents: A Review

of the Situation in Europe. Jyväskylä: National Institute for Health and

Welfare (2009). Available online at: https://www.julkari.ﬁ/handle/10024/

80105 (accessed March 4, 2022).

Frontiers in Public Health | www.frontiersin.org 7 March 2022 | Vo lume 10 | Article 846604

Tarkiainen et al. Analyzing Patient Safety Errors

10. Kohn LT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer

Health System. Washington, DC: Institute of Medicine (US) Committee

on Quality of He alth Care in America Washington; National Academies

Press (2000).

11. Hannaford N, Mandel C, Crock C, Buckley K, Magrabi F, Ong M, et

al. Learning from incident reports in the Australian medical imaging

setting: handover and communication errors. Br J Radiol. (2013)

86:1022. doi: 10.1259/bjr.20120336

12. Tynkkynen LK, Chydenius M, Saloranta A, Keskimäki I. Expanding choice

of primary care in Finland: much debate but little change so far. Healthpol.

(2016) 3:277–234. doi: 10.1016/j.healthpol.2016.01.015

13. Ministry of Social Aﬀairs and Help. Social Welfare and Health Care System

in Finland. (2021). In English Areas of expertise. Available online at: https://

stm.ﬁ/en/social-and-health-services/responsible-agencies (accessed March 4,

2022).

14. Knuuttila J, Ruuhilehto K, Wallenius J. Terveydenhuollon vaaratapahtumien

raportointi. In English; Health Incident Reporting National Agency for

Medicines. Lääkelaitoksen julkaisusarja. Helsinki (2007). Available

online at: https://www.valvira.ﬁ/documents/14444/50159/LH-2007-1_

vaaratapahtumien_raportointi.pdf (accessed March 4, 2022).

15. Awanic Ltd. Reporting System for Safety Incidents in Health Care

Organizations. Available online at: https://awanic.ﬁ/haipro/eng/ (accessed

December 27, 2021).

16. Holmström AR, Airaksinen M, Weiss M, Wuliji T, Chan XH,

Laaksonen R. National and local medication error reporting

systems: a survey of practices in 16 countries. J Patient Saf. (2012)

4:165–76. doi: 10.1097/PTS.0b013e3182676cf3

17. Härkänen M, Turunen H, Saano S, Vehviläinen-Julkunen K. Detecting

medication errors: analysis based on a hospital’s incident reports. Int J Nurs

Pract. (2015) 2:141–6. doi: 10.1111/ijn.12227

18. Zygmont ME, Itri JN, Rosenkrantz AB, Duong PAT, Gettle LM, Mendiratta-

Lala M, et al. Radiology research in quality and safety: current trends

and future needs. Acad Radiol. (2017) 3:263–72. doi: 10.1016/j.acra.2016.

07.021

19. Howell AM, Burns EM, Bouras G, Donaldson L J, Athanasiou T,

Darzi A. Can patient safety incident reports be used to compare

hospital safety? Results from a quantative analysis of the English

National Reporting and Learning System Data. PLoS ONE. (2015)

12:e0144107. doi: 10.1371/journal.pone.0144107

20. Pfeiﬀer Y, Manser T, Wehner T. Conceptualising barriers to incident

reporting: a psychological framework. Qual Saf Health Care. (2010)

6:e60. doi: 10.1136/qshc.2008.030445

21. Vincent CA. Analysis of clinical incidents: a window on the

system not a search for root. Qual Saf Health Care. (2004)

4:242–3. doi: 10.1136/qshc.2004.010454

22. Wakeﬁeld DS, Wakeﬁeld BJ, Uden-Holman T, Borders T, Blegen M, Vaughn

T. Understanding why medication administration errors may not be reported.

Am J Med Qual. (1999) 2:81–8. doi: 10.1177/106286069901400203

23. Evans SM, Berry JG, Smith BJ, Esterman A, Selim P, O’Shaughnessy J,

et al. Attitudes and barriers to incident reporting: a collaborative hospit al

study. Qual Saf Health Care. (2006) 1:3 9– 43 . doi: 10.1136/qshc.2004.

012559

24. Finlex 2011/Asetus 341. Sosiaali- ja terveysministeriön

asetuslaadunhallinnasta ja potilasturvallisuuden täytäntöönpanosta

laadittavasta suunnitelmasta. Statute of the Ministry of Social Aﬀairs and

Health. A plan for quality management and implementation of patient safety

(in English). Available online at: https://www.ﬁnlex.ﬁ/ﬁ/laki/alkup/2011/

20110341 (accessed March 4, 2022).

25. Lawton R, Parker D. Barriers to incident reporting in a healthcare system.

Qual Saf Health Care. (2002) 1:15–8. doi: 10.1136/qhc.11.1.15

26. Kingston MJ, Evans SM, Smith BJ, Berry JG. Attitudes of doctors and nurses

towards incident reporting: a qualitative analysis. Med J Aust. (2004) 1:36–

9. doi: 10.5694/j.1326-5377.2004.tb06158.x

27. Tarkiainen T, Turpeinen M, Haapea M, Liukkonen E, Niinimäki J.

Investigating errors in medical imaging: medical malpractice cases in Finland.

Insights Imaging. (2021) 12:86. doi: 10.1186/s13244-021-01011-8

28. Belzunegui T, Louis CJ, Torrededia L, Oteiza J. Extravasation of radiographic

contrast material and compartment syndrome in the hand: a case report.

Scand J Trauma Resusc Emerg Med. (2011) 19:42. doi: 10.1186/1757-7241-19-9

29. Mansouri M, Aran S, Shaqdan KW, Abujudeh HH. How often are patients

harmed when they visit the computed tomography suite? A multi-year

experience, in incident reporting, in a large academic medical center. Eur

Radiol. (2016) 7:2064–72. doi: 10.1007/s003 30 -0 1 5- 40 6 1- 0

30. Rauhala A, Kinnunen M, Kuosmanen A, Liukka M, Olin K, Sahlström M,

et al. Mitä vapaaehtoiset vaaratapahtumailmoitukset kertovat? What does

voluntary reporting system of patient safety incidents tell? (in English).

Suomen Lääkärilehti. Finnish Medical J. (in English). (2018). Available online

at: https://www.laakarilehti.ﬁ/pdf/2018/SLL462018-2716.pdf (accessed March

4, 2022).

31. Macrae C. The problem with incident reporting. BMJ Qual Saf. (2015)

2:71–5. doi: 10.1136/bmjqs-2015-004732

Conﬂict of Interest: The authors declare that the research was conducted in the

absence of any commercial or ﬁnancial relationships that could be construed as a

potential conﬂict of interest.

Publisher’s Note: All claims expressed in this article are solely those of the authors

and do not necessarily represent those of their aﬃliated organizations, or those of

the publisher, the editors and the reviewers. Any product that may be evaluated in

this article, or claim that may be made by its manufacturer, is not guaranteed or

endorsed by the publisher.

open-access article distributed under the terms of the Creative Commons Attribution

License (CC BY). The use, distribution or reproduction in other forums is permitted,

provided the original author(s) and the copyright owner(s) are credited and that the

original publication in this journal is cited, in accordance with accepted academic

practice. No use, distribution or reproduction is permitted which does not comply

with these terms.

Frontiers in Public Health | www.frontiersin.org 8 March 2022 | Vo lume 10 | Article 846604