Article Text
Abstract
Introduction Local and national awareness of the need to improve the recognition and response to the clinical deterioration of hospital inpatients is high. The authors designed and implemented a programme to improve recognition of deteriorating patients in their hospital; a new observation chart for vital signs was one of the major elements. The aim of the study is to evaluate the impact of the new chart and associated education programme on the completeness of vital-sign recording in ward areas.
Methods The setting is a university-affiliated teaching hospital in Sydney, Australia. Three study periods, each lasting 14 days (preintervention, 2 weeks postintervention, 3 months postintervention), were carried out in three wards. The new observation chart was supported by an education programme. The primary outcome measures were the ascertainment rates of individual vital signs as a proportion of total observation sets.
Results Documentation of respiratory rate increased from 47.8% to 97.8% (p<0.001) and was sustained at 3 months postintervention (98.5%). Collection of a full set of vital signs also improved by a similar magnitude. Basic neurological observation for all patients was introduced in the new chart; the uptake of this was very good (93.1%). Ascertainment rates of blood pressure and oxygen saturation also increased by small but significant amounts from good baseline rates of 97% or higher.
Conclusion The introduction of a new observation chart, and education regarding its use and importance, was associated with a major improvement in the recording of respiratory rate and other vital signs.
- Vital signs (MeSH)
- monitoring
- physiological (MeSH)
- patient care management (MeSH)
- quality of healthcare (MeSH)
- healthcare quality improvement
- patient safety
Statistics from Altmetric.com
- Vital signs (MeSH)
- monitoring
- physiological (MeSH)
- patient care management (MeSH)
- quality of healthcare (MeSH)
- healthcare quality improvement
- patient safety
Introduction
Improving the recognition and response to the deteriorating inpatient has become an important focus for the improvement of safety and quality in healthcare.1–5 This is appropriate, and based around the association between abnormal vital signs and the occurrence of subsequent major adverse events such as cardiac arrest or unscheduled admission to the intensive care unit (ICU).6–9 Abnormalities in respiratory rate (RR) are strongly associated with serious deterioration, but this vital sign is one of the most frequent omissions in general ward areas.10–12
Detection of deterioration is based on objective physiological criteria (afferent limb) and the implementation of medical emergency teams (efferent limb)13 to respond to these patients. We designed and implemented an integrated programme to improve our recognition and response to clinical deterioration in ward areas. This programme was customised to our hospital and encompassed changes in hospital policy, education, models of care and documentation.
The aim of this study is to describe the development and implementation of a new observation chart and education programme, and to evaluate its impact on the completeness of vital-sign recording in ward areas.
Methods
Setting
This was a prospective, before-and-after-intervention study conducted between May 2009 and August 2009 in an Australian tertiary-referral, university-affiliated teaching hospital in Sydney. The project was approved by the institutional human research ethics committee, and the need for individual patient consent was waived. The project was devised and implemented as part of a hospital-wide initiative called the Royal Prince Alfred Hospital (RPAH) Clinical Emergency Response System (CERS).14
The CERS comprises a steering committee representing executive, patient safety and clinical personnel. The clinical representatives include medical and nursing staff, senior and junior staff, and intensive care and ward staff. The committee designed and implemented a new observation chart, education programme and data collection in support of their goals; it supervises the work of one nursing staff member appointed to coordinate the system. Ward and switchboard practices regarding the triggering mechanism were reformed, and a modified medical emergency team (MET) was introduced. This efferent limb has two graded levels—the first level is a senior house officer or specialist registrar usually responsible for that patient's care; the second level is a specialist registrar and nurse from the ICU. We also have a cardiac arrest team response. Escalation is determined according to the level of clinical concern and the results of the primary response (see figure 1)
Intervention
Design of new observation chart
Senior clinical nurses in pilot ward areas, working with the steering committee, identified that the current observation chart was outdated. The old observation chart plotted vital signs over two pages, and respiratory rate was at the bottom of the first page. The redesign of the observation chart was undertaken in collaboration with the hospital's forms committee and medical-records department. The key features of the new observation chart are: (1) integration of the physiological triggers for the escalation plan, (2) graphing of all vital signs on one side of a standard A4 page, (3) elevation of respiratory rate to the top of the chart, (4) provision for individual modification of the physiological triggers for activation of the escalation plan and (5) the addition of basic neurological monitoring as a routine observation in all patients: level of consciousness was documented, measuring responsiveness with the Alert, Verbal, Pain or Unresponsive (AVPU) method.15
Implementation of education programme and new observation chart
An extensive education programme was developed prior to the implementation of the new observation chart. The education programme consisted of an orientation to the new chart and CERS, and basic vital-signs assessment skills using manual techniques. It was a requirement for all nursing staff to attend the education programme prior to the implementation of the new observation chart and CERS.
Prospective audit of vital-sign ascertainment
Three ward areas were chosen to measure and analyse the impact of the new observation chart and education programme. Ward A is a mixed medical/surgical ward, Ward B surgical and Ward C medical. Three study periods were chosen: preintervention, 2 weeks postintervention and 3 months postintervention. The preintervention study period occurred during the final month before intervention. Each study period consisted of all discharges during a consecutive 14-day period. All observation charts were photocopied for analysis later. We ensured complete capture of all patients discharged during the study period by cross-referencing the hospital information system.
The preintervention period was the baseline for our previous institutional practice using the old observation chart. The 2-week postintervention period occurred after the launch of the new observation chart, and nursing and medical education regarding the importance of abnormal vital signs and the new CERS. All education was complete within 1 week of the launch of the observation chart, and there was no more formal education, regarding the observation chart, after this period. The CERS coordinator was in frequent informal contact with individual staff after the intervention. The 3-month postintervention study period was carried out to evaluate whether any changes had been sustained. Nursing managers on each of the wards were aware of the study taking place, but were not informed of the specific audit dates. Patients were excluded from the 2-week and 3-month audit if they had been transferred into a study ward, during the study period, from another ward of the hospital that was not yet using the new observation chart, as part of their stay would have utilised the old chart.
Four primary vital signs were studied: RR, blood pressure (BP), oxygen saturation (SpO2) and heart rate (HR). An observation set was considered to be any documentation of one or more of these vital signs at any time point on the chart. A complete observation set was defined as the simultaneous documentation of all four primary vital signs. Temperature is collected as a vital sign, but was not analysed, since it is not an activation criterion for the CERS. Data were entered into a spreadsheet (Microsoft Excel for Mac).
Outcomes and statistical methods
The primary outcome measure for each vital sign was the proportion of observation sets in the entire study period, which included documented collection of that vital sign (RR, BP, SpO2, HR, complete set 4 of 4). We also measured the same outcome for AVPU in the postintervention periods.
The levels of activity in each ward area were summarised using the number of discharged patients, percentage of medical (non-surgical) patients, length of stay, and median and mean number of observation sets per patient. The proportion outcome measures were compared using a χ2 test (three groups for RR, BP, SpO2, HR, complete set; two groups for AVPU). Results were considered significant if p<0.05. SPSS v14.0 software was used for the χ2 test.
Results
The previous observation chart is shown in figure 2A,B. The new observation chart is shown in figure 3. The characteristics of the three ward areas studied are summarised in table 1. The level of activity, measured by the number of patient discharges during the study period, was greater in the period 2 weeks postintervention owing to seasonal factors. The total number of observations in the preintervention, 2 weeks postintervention and 3-month postintervention study periods (2557, 2435, 2250 observations respectively) remained similar. The number of observation sets per patient was skewed in all study periods, consistent with small numbers of patients staying for longer periods.
The impact of the new observation chart and education programme on completeness of vital-sign ascertainment is displayed in table 2. The rate of RR ascertainment and the completeness of observation sets was approximately 48% prior to the intervention. Both rates doubled and became nearly complete after the intervention, and this change was sustained at 3 months. The rate of BP and oxygen saturation ascertainment was high prior to intervention, but this also increased by a small and significant amount in the period after intervention. Heart-rate measurement was almost 100% prior to the intervention, and this did not decrease after the intervention. The introduction of AVPU scoring was well supported, with a sustained compliance rate of around 92–93% after intervention.
Discussion
Summary of key findings
The ascertainment rate of RR increased from 47.8% to 97.8% between the preintervention and 2 weeks postintervention periods (p<0.001); this change was sustained at 3 months (98.5%). We also observed similar and highly significant increases in the rate of collection of a complete observation set. The uptake of AVPU after intervention was very good (93.1% at 2 weeks postintervention), and this also remained steady 3 months postintervention (92.5%, p=0.45). Small but significant improvements in the ascertainment rates of BP and SpO2 occurred which were already excellent at baseline (97% or higher).
Context
This report evaluates an important element of the afferent limb in our institution's modified medical emergency team. The positive impact of a well-designed observation chart has been reported16 17 but has received relatively little scientific attention. The key physical features of our chart that we believe are most important are: (1) elevation of RR to the front and top of the chart; (2) visual display of all four primary vital signs plus AVPU on a single side of standard A4-sized page; and (3) the use of colour coding and banding to highlight abnormal values. It is probable that a chart that is easy to use will be associated with improved afferent limb function, and delays in triggering are associated with increased mortality.18 19
Interpretation
The positive effect of the improved visual amenity of the chart appears to be strong, since it was associated with increased vital-sign ascertainment despite more vital signs being observed after introduction, and higher numbers of admissions in the postintervention period. Persistence of the positive changes at 3 months later also encourages us in this view. Other factors likely to be involved in the success that we experienced are the support of the hospital executive and patient safety staff, and ongoing informal detailing of ward staff by the CERS coordinator. This included debriefing staff and the promotion of examples of good observation practice. We have not analysed the statistical significance of the number of observation sets appearing to decrease slightly during the course of the study periods. The numbers appear similar, and any univariate association would be difficult to interpret without adjusting for unmeasured covariates such as acuity.
Limitations
We did not quantitate the burden of the education programme in terms of time and human resources. We were not able to randomly introduce the education package and observation chart separately, or in a crossover manner, which would objectively reveal their relative contribution. A multifaceted approach which includes education is important for success,17 20 but we believe that the predominant driver of the changes that we observed is the observation chart itself. We have not audited the clinical appropriateness of the frequency with which observations were collected in ward areas, or their accuracy and precision.
Conclusion
We implemented a new programme to improve the safety and quality of care provided to patients who are deteriorating in ward areas. A new observation chart was a simple and effective part of this programme and was associated with major and sustained improvements in RR ascertainment. The documentation of vital signs has improved across the institution as a result.
References
Footnotes
Competing interests None.
Ethics approval Ethics approval was provided by the Sydney South West Area Health Service Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.