Background: With more liberal visiting hours in paediatric intensive care practice, parents’ presence at the bedside has increased. Parents may thus become involved in critical incidents as contributors or detectors of critical incidents or they may be affected by critical incidents.
Methods: Voluntary, anonymous, non-punitive critical incident reporting system. Parents’ involvement in critical incidents has been evaluated retrospectively (January 2002 to August 2007). The reports were analysed regarding involvement of parents, age of child, unit (paediatric intensive care or intermediate neonatal nursery), critical incident severity, critical incident category, actual or potential harm to patient and/or parent (minor, moderate, major), delay between the critical incident and its detection, and implemented system changes.
Results: Overall, 2494 critical incidents have been reported. There were 101 critical incidents with parental involvement: parents as contributors to critical incident (18; 0.7%), parents discovering a critical incident (11; 0.4%), parents affected by critical incident (72; 2.9%). The most vulnerable categories regarding contribution and detection were drugs, line/drain disconnection, trauma and hygiene. Ten critical incidents precipitated by parents were of moderate severity and seven of potential major severity (six line/drain disconnections). The majority of the events (six) detected by parents were of potential moderate severity and four were of major severity.
Conclusion: Because of their presence at the bedside, parents in the paediatric intensive care unit are inevitably involved in safety issues. It is not the parents’ duty to guarantee the safety for their children, but parents should be encouraged to report anything that worries them. Only an established safety culture allows parents to articulate their concerns.
Statistics from Altmetric.com
Concurrent with more liberal visiting hours practice in paediatric intensive care, parents’ presence at the bedside has increased. This may have important emotional benefits for the critically ill child.1 When present in the paediatric intensive care unit (PICU), parents are interacting with their child, observing health professionals in performing their duties at the bedside and interacting with them, and looking at monitors, other equipment (ventilator, infusion pumps, drainage systems, intravenous lines, etc) and charts. Furthermore, parents are analysing intensive care procedures and sometimes developing actions to protect their babies.2
The presence of parents, their interest and also participation in the care of their child may lead to their involvement in critical incidents in a threefold way: (1) parents may be contributors to critical incident; (2) parents may detect critical incident; (3) parents may be directly affected by critical incident. There is scarce information in the literature on this topic.3 4 We, therefore, analysed our critical incident database regarding the parents’ involvement.
The study was performed in a multidisciplinary, 19-bed, neonatal–paediatric intensive care unit (PICU) and a 16-bed neonatal intermediate care nursery (IMC) of a university teaching children’s hospital. In 2002, a hospital-wide voluntary, anonymous, non-punitive critical incident reporting system was implemented.5 Before starting the monitoring, several tutorial sessions and group discussions were held in order to familiarise nursing and medical staff with the system approach philosophy. The incident reporting system had to be non-threatening to staff, encourage team involvement and focus on deficiencies in the system rather than the individual. Critical incidents comprise harmful and potentially harmful events. Our reporting form consisted of seven sections: (1) age of the patient involved; (2) narrative section about the critical incident and its circumstances; (3) possible contributory factors; (4) actual harm to the patient; (5) was patient harm prevented by a check?; (6) proposals of measures to prevent any such critical incident in the future; (7) date and hour of day of the critical incident and its detection. In 2004, an electronic reporting form was introduced. Every 4 months, the reports were analysed by the quality assurance group and discussed within the PICU team. The quality assurance group consisted of one consultant, five senior nurses (of whom three have managerial functions), one person who was responsible for the PICU equipment, one pharmacist and one resident. The quality assurance group developed organisational measures in order to prevent the recurrence of the reported incidents.
The present study is a retrospective survey on all critical incident reports with parental involvement, from 1 January 2002 to 31 August 2007. Parental involvement was classified as (1) parents contributing to a critical incident; (2) parents detecting a critical incident; (3) parents being affected by a critical incident. We analysed only reports with the explicit statement that parents were involved. Reports were completed exclusively by nurses and doctors. Most parents were not aware of our critical incident system, as they were not routinely informed about it. So, there was never a request of a parent to submit a critical incident. However, patients and parents were encouraged to write their concerns and wishes on a questionnaire. These questionnaires were handled by the patient counselling officer of the hospital.
The reports were analysed regarding age of child, unit (PICU or IMC), critical incident severity, critical incident category, actual harm to patient and/or parents, delay between the critical incident and its detection, implementation of system changes. The critical incidents were categorised into the following sections: drugs, communication, documentation, hygiene, monitoring, trauma, lines and drains, general patient management, respiration, equipment and miscellaneous. Actual and potential critical incident severity was graded (agreement of the members of the quality assurance group)3: minor (no interventions required), moderate (requiring routine therapy available outside the PICU), major (need for therapeutic interventions specific to the PICU or death). The critical incidents with parental involvement were compared with the rest of the critical incidents without parental involvement.
Until 31 December 2003, parents were not allowed to visit their child during the morning round (07:30 to 11:00), and it was recommended to them not to stay in the PICU during night (22:00 to 07:00). From 1 January 2004 onwards, visiting hours have been liberalised and parents are encouraged to stay at the bedside whenever they choose to.
Averages were given as median (range). Differences between groups were analysed using the Mann–Whitney U test. A p value of <0.05 was considered significant.
In the studied period (5 years and 8 months), a total of 2494 critical incidents (40/1000 patient days) have been reported, 1928 critical incidents in the PICU and 566 critical incidents in the IMC. There were 101 critical incidents in which a parent was involved. In 18 reports, a parent contributed to the occurrence of the error (0.7%); in 11 reports, a parent discovered an error (0.4%); and in 72 reports, a parent was the affected individual (2.9%). Table 1 summarises the categories of these critical incidents, the actual/potential harm to patients/parents, the delay between critical incident and its detection, and the age of the involved children. The median delay between occurrence of critical incidents and their detection was longest for the subgroup of parents detecting an error (10 h) and shortest for the subgroup of parents contributing to error (0.25 h). Comparisons between groups for delay (between incident and detection) and patient age were not significant. All of the critical incidents detected by parents occurred only after full liberalisation of visiting hours. Different system changes as a result of the reported incidents were implemented in four instances in the “contribution group”, in four instances in the “detection group” and in nine instances in the “affected group”. Table 2 gives examples of reported events.
In the “contribution” and “detection” groups, drug incidents, line/drain disconnections, trauma and hygiene accounted for most reports (38%, 28%, 10% and 10%, respectively) (table 1). In the remaining critical incidents (>2000), drugs were also the most important category (33%), but line/drain disconnections were quite rare (2.7%), as were trauma (0.2%) and hygiene (3%) (table 1). The eight children affected by disconnections were young (median age 3.5 weeks, range 4 days to 1.5 years).
In our sample, parents contributed to 0.7% of all reported critical incidents; they were involved in the detection of critical incidents in 0.4%; and they were affected in 2.9% of all critical incidents. The respective numbers in the literature are 0.5% and 0.6% for contribution, 1.9% and 0.7% for detection3 4 and 0.7% for adverse affection.4 The relatively high percentage of 1.9% of critical incidents detected by parents in our previous study may have been related to the prospective nature of that survey3: the reporting form of our previous study had a section on the function of the person precipitating or detecting a critical incident. On the other hand, the methodology was consistent with the present study, in particular, it was also the nurses and doctors filling in the forms who judged the parental involvement. We identified a higher percentage of critical incidents on affected parents in the present survey compared with the study of Suresh et al.4 The latter study, conducted in neonatal intensive care units of the Vermont Oxford Network, used the same methodology regarding assessment of parental involvement.4 Our higher percentage of adverse parental affection may be related to the high number of reported breast milk confusions (25): each of them resulted in subsequent viral testing in the “donor”-mother. Within the scope of a quality improvement project, we encouraged the nursing staff to report breast milk confusions comprehensively. Actions have been taken for improvement (clear labelling of each bottle with content and name of patient), and subsequently, the reports of this type of incident have almost entirely vanished.
As our critical incident monitoring is voluntary, the identified numbers of parental involvement do not give the real picture. Furthermore, as only reports were analysed in which it was explicitly stated that parents had been involved, our numbers depend on the judgement of the reporting doctors and nurses. Whereas the categories “detection” and “contribution” seem more objectively determined, the category “parents affected” needs to be interpreted with caution. Indeed, parents are more or less emotionally affected by most of the critical incidents, especially critical incidents leading to patient harm. Compared with other units,4 6 7 the total number of reported critical incidents was relatively high, indicating a good safety culture which may enable picking up a substantial percentage of occurring critical incidents.7 Whatever the exact number of parental involvement in our PICU, this survey shows that parents are involved in critical incidents.
Interestingly, in the group of critical incidents that have been detected by parents, the delay between occurrence and detection was longest (although statistically not significant) and all these critical incidents occurred only after introduction of visiting hours around the clock. This observation indicates that parents may only detect critical incidents when they spend longer hours at the bedside and that they may need more time to catch up a critical incident compared with healthcare personnel. On the other hand, the critical incidents precipitated by parents were detected by the nursing and medical staff after a short delay (0.25 h), indicating good supervision.
Regarding parents’ contribution to and detection of critical incidents, drugs, line/drain disconnections, trauma and hygiene were the most important categories. Line/drain disconnections accounted for 28% of the critical incidents compared with only 2.7% in the incidents without parental involvement. It concerned mainly young infants nursed by a parent on his/her lap. The quality assurance group has taken the mention of parent involvement in critical incident reports as a subtle indicator of relevant problems and respective measures have been implemented (eg, presence of pharmacists in patient rounds, guidelines regarding the handling of infants with line and drains, parents’ instruction in correct hygiene behavior).
It has been recognised that more liberal visiting hours, which allow parents to be at the bedside for prolonged periods, have important emotional benefits for the critically ill child.1 The present study shows that the presence of the parents may also impact on safety issues, in a good and bad way. We showed that most of the critical incidents precipitated by parents were of actual moderate severity and some of them had the potential for major severity. On the other hand, the majority of the events discovered by parents were of potential moderate severity. In a neonatal intensive care unit, it has been shown that mothers commonly developed actions to protect their premature babies.2 This may be also true for the paediatric intensive care unit. Parents may thus interfere with the care of their child and thus evoke critical incidents as well as prevent such incidents. However, we believe that healthcare professionals have to be watchful for increasing independent parental involvement in the medical care of their child. Parents should be encouraged to report anything that worries them.8 There should be a safety culture in the unit that allows parents to articulate their concerns (eg, if a doctor does not wash his/her hands before touching a baby).9 Parents may be asked questions by the hospital staff such as “Are there aspects of your child’s care that you find concerning?” or “What do you worry about when you leave your child?”9 A prerequisite for meaningful answers by parents is their comprehensive information on the treatment plans of their child. In adult patients, it has been shown that patients who were well informed were less likely to accept ineffective or risky procedures.8 Once a critical incident has been precipitated or detected by parents, it should be analysed very carefully in order to implement system changes with the potential to eliminate these incident in the future.
The incidents on adversely affected parents are an indication of the parents’ role as care givers of their child and their close involvement in treatment plans. Most of these reports relate to communication failures. This underscores the importance of timely and comprehensive information of parents in order to avoid additional stress. Full and open disclosure applies especially to adverse events and medical errors.10
Because of their presence at the bedside, parents in the PICU are inevitably involved in safety issues. They may precipitate or detect critical incidents and they may be affected by critical incidents. The most vulnerable categories regarding contribution and detection were drugs, line/drain disconnection, trauma and hygiene. It is not the parents’ duty to guarantee the safety for their children, but parents should be encouraged to report anything that worries them. Only an established safety culture allows parents to articulate their concerns.
Funding Competing interests: None.
Competing interests None.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.