What is known about this topic?
- There is temporal variation in some clinical processes in HASU (see Chapter 3), and no measurable variation in ‘front-door’ processes and patient outcomes.
- Consistency in ‘front-door’ processes is partly achieved by extending nursing roles at night and at the weekend (see Chapter 4).
- Senior nursing is perceived to be an important component of achieving high standards of HASU care (see Chapter 4).
What this analysis adds
- There is variation in the nurse staffing by time of the week, particularly in terms of the seniority of key roles, rather than the number of nurses overall.
- Senior nursing roles are more differentiated than junior roles, which can be filled by bank nurses and other nursing staff.
- When senior nurses take on the nurse-in-charge role during the day and during the week, it has an impact on the delivery of clinical interventions.
Introduction
London HASUs reduced variations in the delivery of ‘front-door’ clinical interventions overall41 and at different times of the day and week (see Chapter 3).34 The London stroke service standards recognise the pivotal role played by stroke nurses in delivering high-quality care and patient outcomes, and, therefore, in reducing this variation. An A1 standard (i.e. a standard that must be met if a service is to be classified and funded as a HASU36,39) requires consistent nurse staffing levels at all times:
Provision of 24/7 nursing workforce to provide: 2.9 WTE [whole-time equivalent] nurses per bed; 80 : 20 trained to untrained skill mix.
However, this standard leaves a degree of flexibility in terms of how nurses are organised in a HASU. For instance, it does not specify nurse leadership roles, the proportion of senior nurses, or the roles and responsibilities of the nurse in charge.
Our analysis in Chapter 3 showed that London HASUs operate a 24/7 service with respect to nursing-related care, rapid access to brain scans and administration of thrombolysis to eligible patients, but not other aspects of care. It also showed that there was no variation among patients treated at London HASUs by day and time of admission across the week in terms of 3-day mortality and mRS score at hospital discharge. It has been suggested that staffing levels, in particular nurse-to-bed ratios, can influence variations in quality of care and outcomes by day and time of admission across the week.60 Recently, several studies have found that stroke units with higher ratios of nurses trained in dysphagia screening are associated with lower risk of death87 and lower risk of stroke-related pneumonia.88,89 The measures used in these studies were derived from the SSNAP Organisational Audit, which is based on a one-off snapshot of staffing, carried out once every 2 years, and does not account for day-to-day variation. We used a novel data set of London HASU nurse staffing based on detailed information from nursing rotas for every time of the day in all HASUs in London, applied to the SSNAP data, to examine the relationship between nurse staffing in the HASU and processes of care and outcomes. This was informed and supported by a qualitative analysis of temporal variation of the organisation of stroke care and nursing roles on the HASU. The results from this analysis indicate that nurse staff profile – seniority and separation of roles at the HASU – contributes to the quality of processes management within the ward and there are variations in nurse staffing by day and night and across the week. Subsequently, the aims of the analysis were to investigate (1) how nurse roles on the HASU are organised, (2) whether or not nurse staffing affected outcomes and quality of acute stroke care in London HASUs and (3) whether or not there is variation in those effects by day and time of admission across the week.
Methods
This was a mixed-methods analysis. Data collection and analysis are described in Chapter 2. In Chapter 4, we identified aspects of organisation that are important for care delivery and outcomes, especially the roles of nursing staff on the HASU, including the seniority and specialisation of nursing staff. For every patient who was admitted to a London HASU during the study period, we identified the characteristics of the nurse staffing on the HASU at the time they were admitted to the hospital and tested whether or not these factors were associated with processes and outcomes. The quantitative results were discussed in the light of the qualitative findings, and were then used to explain the variation of acute stroke care and outcomes.
Data and measures
Interview and observational data collection are explained in Chapter 2, Qualitative data collection, as are quantitative nurse staffing data collection and linkage. Each day was divided into 48 30-minute shifts. Every patient admitted within each of these shifts was assigned a corresponding nurse staffing measure during the same time period.
For this analysis, the following quality of care indicators were used, as nurses’ input was essential to meet the standards of care: brain scan within 1 hour, dysphagia screen within 4 hours, admission to a stroke unit within 4 hours, administration of thrombolysis to eligible patients and door-to-needle time. Outcomes were measured as whether or not the patient died within 3 days and disability using the mRS score (0–2 vs. 3–6; moderate, moderately severe, severe disability or death) at the end of the inpatient stay. We also analysed the effect of nurse staffing on LOS in HASU and total LOS in the hospital.
We matched the nurse staffing data set with the SSNAP data set based on the time of admission to the hospital, not the HASU, in line with the way in which clinicians describe the ‘clock start’ of other process measures in the SSNAP audit. As reflected in Chapter 5, many of the processes of care analysed are likely to be delivered in the ED on arrival at the hospital. Note that data were not collected on medical staffing, so we were unable to control for this in our analyses.
To examine trends across the week we used four time periods of time of admission to the hospital, similar to broad trends used in Chapter 3, Data and measures: Monday to Friday, 08.00 to 19.59; Monday to Friday, 20.00 to 07.59; Saturday and Sunday, 08.00 to 19.59; and Saturday and Sunday, 20.00 to 07.59. This is different from previous research in which the only comparison was between weekdays and weekends.60
Sequential mixed-methods analysis
Qualitative analysis
Interview and observational data analysis methods are summarised in Chapter 4, Analysis. Findings related to nursing seniority, highlighted in Chapter 4, Results, informed the quantitative analysis.
Quantitative analysis
We examined whether or not nurse staffing varied across the week. We tested for statistically significant variations across the four time periods using chi-squared tests.
To evaluate whether or not nurse staffing affected 24/7 access to high-quality acute stroke care, we tested the impact of nurse staffing on the variation in the quality of acute stroke care and outcomes by day and time of admission in London HASUs. We ran three sets of patient-level regression models following the results from qualitative observations. First, we ran exactly the same logistic model for the selected outcomes and quality-of-care measures used in the analysis reported in Chapter 3: Statistical analysis for only London HASUs, including the four time periods and the same baseline patient and case mix covariates. In addition, these models included the nurse staffing variables (band of the nurse in charge, whether or not there was a dedicated thrombolysis nurse who was different from the nurse in charge, total number of nurses on duty in the HASU) plus the number of beds in each HASU, and whether or not the patient was admitted to the HASU during the nursing shift handover period. Handover is the half-hour time period between the day and the night shifts (e.g. 07.30–20.00 and 19.00–08.00) to transfer responsibility and accountability for some or all aspects of patient care. Second, we re-ran these models for each of the four time periods separately. Third, based on input from the qualitative findings showing that in the delivery of ‘front-door’ services patients eligible for tissue plasminogen activator (tPA) are prioritised, we re-ran the models stratified by subgroups of patients eligible and not eligible for thrombolysis at the time of arrival to the hospital.
For each outcome and quality-of-care measure we tested whether or not the nurse staffing variables were significantly different from zero (individually significant in the case of whether or not there was a dedicated thrombolysis nurse who was different from the nurse in charge and the number of nurses on duty in the HASU, and jointly significant for the categories describing the band of the nurse in charge). For the categorical variables (band of the nurse in charge and whether or not there was a dedicated thrombolysis nurse who was different from the nurse in charge) we computed average predicted probabilities for each outcome, controlling for the covariates. For variables included as linear terms (number of nurses on duty in the HASU) we computed average marginal effects showing the increased probability that each outcome would be achieved if the number of nurses on duty increased by one unit.
Using the second and third sets of models, for every measure we also tested whether or not the coefficients on the nursing variables were significantly different between the four time periods and two subgroups, using likelihood ratio tests. This provides an indicator of whether or not nurse staffing affects variations in quality of care and outcomes by time of admission and by subgroups in London HASUs.
Results
Our results present a multifaceted analysis of how senior nurses contribute to HASU processes and patient outcomes. First, we outline the difference between junior and senior nurses in terms of tasks on the ward, and give contextual information about where they conducted these tasks.
Nursing roles on the hyperacute stroke unit
Junior nursing roles on the hyperacute stroke unit
In our data, the average daily number of patients arriving Monday to Friday was 2.2 in hours and 1.4 out of hours. Arriving Saturday to Sunday, the average daily number of patients was 2 in hours and 1.4 out of hours. The London stroke service standards specified a certain number of nurses per HASU bed and specified that these were to be provided on a consistent (24/7) basis. This created a sense of continuity and a consistency in the level of bedside care. The level of seniority was not specified. In some HASUs the nurses themselves were assigned to a specific room or number of rooms, making the link between beds and nurses concrete. These nurses were broadly known as ‘junior nurses’.
Junior nurses engage in activities on the HASU that mainly involve direct patient care with some administrative duties. A list of junior nurse roles in the HASU, as evidenced by our interview and observation data, is provided in Table 13.
| Role | Task |
|---|---|
| Administration | Answer telephone (when no ward clerk) Stock take |
| Patient assessment and care planning | Admission care plans Skin integrity Weight Nutrition Mood Continence Risk assessment, mobility, fluid balance, temperature, falls risk |
| Stroke nursing | Stroke assessment in the ED (if trained) Dysphagia screening (if trained) Administration of thrombolysis |
| Supervision | Monitoring for deterioration Monitoring confusion Blood pressure measurement Regular observations |
| Therapy activities | Practising movements (e.g. chair to standing) Practising activities of daily living |
| Traditional nursing | Washing patients and personal care, feeding Helping patients to sit out of bed Giving medications (if trained) Assisting with fluids and nutrition intake Turning patients Escorting to investigations if needed Escorting to the toilet |
Senior nursing roles interacting with the hyperacute stroke unit
There were other nursing roles, however, that are not tied to bed numbers, and these have specific functions (Figure 12).
FIGURE 12 Senior nursing roles and working sites. (Figure not reproduced here; see the original article.)
Senior nurses had very different roles (Table 14) from junior nurses (see Table 13), although we did find that some HASUs had trained their junior nurses to fulfil specific tasks, such as thrombolysis and dysphagia screening. Senior nurses were always present in the daytime during the week, but were not always present during the nights or at weekends. Someone in the ward will always fulfil the role of nurse in charge and thrombolysis nurse, but at night and at weekends this may be a lower band nurse or the same person may complete both roles. However, the actual tasks performed by these roles are different in weekday hours (Tables 14 and 15). It is worth noting that when a senior nurse is actively involved in patients’ care, their ward management and oversight activities are reduced.
| Role | Task |
|---|---|
| Communication roles | Update/liaise with families Act as a link between doctors, nurses and therapists Update patient boards or lists |
| Patient care roles | Replace missing nurses to provide direct patient care Take over the roles that other nurses are unable to do (e.g. if staff have not completed training) Dysphagia screening Attend thrombolysis calls and assess patients in the ED with doctor Assess patients to determine their needs |
| Support roles | Oversee and mentor junior nurses Training junior nurses Supporting doctors Supporting new staff |
| Co-ordination and oversight | Escalate urgent patient issues Bed management Complete rotas and ensure that staffing levels are adequate Co-ordinate the transient ischaemic attack clinic Co-ordinate ward staff to ensure that tasks are completed (e.g. paperwork, repatriation) Co-ordinate the discharge or transfer of patients, including transport, medications, etc. Oversee the stroke pathway |
| Role | Task |
|---|---|
| Education | Provide education to patients, such as risk factors, lifestyle, driving, medication Teaching at university Provide advice to patients and families after discharge |
| Patient assessment | Assess, admit and manage non-urgent referrals |
| Service management | Attend MDT meetings Managerial role Nurse-led ward rounds to review care plans Ward rounds with doctors Make sure doctors’ orders are carried out Attend strategy meetings Manage investigation of complaints, clinical incidents and near misses Introduce and embed trust policy and care initiatives Manage ward budget Responsible for trust-based ward quality metrics Responsible for upkeep of ward environment |
How senior is the nurse in charge of the ward at different times of day and days of the week?
Across all HASUs, for the majority of patients admitted to a hospital, the nurse in charge was a band 6 nurse (for 68–90% of patients in each time period across the week) (Figure 13). The nurse in charge was band 5 for 7–29% of patients across the week, with higher proportions of band 5 nurses in charge during the night. The nurse in charge was band 7 for 0–13% of patients across the week, mainly during the day on weekdays. The nurse in charge was rarely a band 8 nurse or a bank or agency nurse.
FIGURE 13 Band of nurse in charge, by day and time of admission. Variation across the week (p < 0.001). (Figure not reproduced here; see the original article.)
For 61% of patients there was a dedicated thrombolysis nurse who was separate from the nurse in charge on arrival (Figure 14), varying from 36% to 80% across the week. For patients admitted at night, there was a greater likelihood that the thrombolysis nurse and nurse in charge were roles held by one person.
FIGURE 14 Whether or not there were separate nurses managing thrombolysis and the ward, by day and time of admission. Variation across the week (p < 0.001). (Figure not reproduced here; see the original article.)
The mean number of HASU nurses varied between 8 and 11 across the week, with higher numbers working during the day (Figure 15). The most common band of nurse in each time period was band 5, with, on average, three to five band 5 nurses working across the week, followed by health-care assistants (two or three nurses working across the week). The next most common were band 6 nurses (one or two nurses) and bank or agency nurses (one or two nurses).
FIGURE 15 Number of nurses on duty involved in patient care by band, by day and time of admission. Variation in all but band 8 across the week (p < 0.001). HCA, health-care assistant. (Figure not reproduced here; see the original article.)
Although we have granular data describing nurse staffing for every half-hour of the day over the study period, nurse staffing varies largely by shift. On average, the number of band 7 and 8 nurses was small (< 1), seen mainly during the day on weekdays. Band 7 and 8 ward managers in most cases are not included in rotas, as they do not provide patient care. There was greatest variation across the week in band 5 nurses, health-care assistants and bank and agency nurses, with higher numbers of band 5 nurses and health-care assistants during the day, and higher numbers of bank and agency nurses at night.
How does nursing seniority impact on clinical processes and patient outcomes?
Results we report are for patients admitted during all time periods combined and in each of four time periods separately. In Report Supplementary Material 2, Table S4, we report analogous results but only for patients eligible for thrombolysis on admission to the hospital (according to the Royal College of Physicians’ definition of eligibility). In Report Supplementary Material 2, Table S5, we report analogous results for patients not eligible for thrombolysis.
Nurse staffing had a statistically significant impact on having a brain scan within 1 hour, dysphagia screening within 4 hours and the probability of being discharged with at least moderate disability (Table 16). It did not have a significant impact on administration of thrombolysis, admission to the HASU within 4 hours, inpatient mortality within 3 days or LOS on HASU and in the hospital.
| Components of LOS/costs | All | 08.00–19.59 | 20.00–07.59 | ||
|---|---|---|---|---|---|
| Weekday | Weekend | Weekday | Weekend | ||
| Brain scan within 1 hour | |||||
| Band 8 | 0.80 (0.67–0.92) | 0.82 (0.70–0.94) | |||
| Band 7 | 0.63 (0.58–0.67) | 0.61 (0.56–0.66) | 0.59 (0.46–0.73) | 0.70 (0.42–0.97) | |
| Band 6 | 0.61 (0.60–0.63) | 0.59 (0.58–0.61) | 0.64 (0.61–0.67) | 0.64 (0.61–0.66) | 0.65 (0.61–0.70) |
| Band 5 | 0.56 (0.53–0.60) | 0.55 (0.49–0.60) | 0.55 (0.48–0.62) | 0.55 (0.48–0.62) | 0.61 (0.50–0.71) |
| Band/agency | 0.62 (0.54–0.70) | 0.61 (0.47–0.74) | 0.63 (0.47–0.79) | 0.66 (0.51–0.80) | 0.78 (0.51–1.04) |
| p-value | 0.01 | 0.02 | 0.15 | 0.11 | 0.49 |
| Separate nurses | 0.59 (0.58–0.61) | 0.58 (0.56–0.60) | 0.60 (0.57–0.64) | 0.62 (0.58–0.66) | 0.63 (0.56–0.69) |
| One nurse | 0.64 (0.62–0.66) | 0.64 (0.60–0.67) | 0.67 (0.62–0.72) | 0.63 (0.59–0.67) | 0.66 (0.61–0.72) |
| p-value | 0.00 | 0.01 | 0.04 | 0.63 | 0.46 |
| RN, n | 0.01 (0.13) | 0.02 (0.00) | 0.02 (0.02) | –0.03 (0.00) | –0.01 (0.45) |
| HCA, n | 0.00 (0.29) | 0.00 (0.82) | –0.01 (0.30) | 0.03 (0.01) | 0.02 (0.40) |
| Dysphagia screening within 4 hours | |||||
| Band 8 | |||||
| Band 7 | 0.76 (0.72–0.80) | 0.72 (0.67–0.77) | 0.79 (0.67–0.92) | 0.85 (0.58–1.12) | |
| Band 6 | 0.76 (0.74–0.77) | 0.74 (0.72–0.75) | 0.78 (0.76–0.81) | 0.78 (0.75–0.80) | 0.79 (0.75–0.82) |
| Band 5 | 0.70 (0.67–0.73) | 0.71 (0.66–0.76) | 0.65 (0.59–0.72) | 0.70 (0.64–0.77) | 0.77 (0.67–0.86) |
| Band/agency | 0.74 (0.67–0.81) | 0.72 (0.60–0.84) | 0.77 (0.63–0.92) | 0.82 (0.71–0.92) | 0.69 (0.40–0.98) |
| p-value | 0.01 | 0.61 | < 0.01 | 0.13 | 0.74 |
| Separate nurses | 0.79 (0.78–0.80) | 0.77 (0.76–0.79) | 0.80 (0.78–0.83) | 0.81 (0.77–0.84) | 0.86 (0.82–0.90) |
| One nurse | 0.66 (0.64–0.68) | 0.60 (0.57–0.64) | 0.65 (0.60–0.70) | 0.73 (0.69–0.77) | 0.71 (0.65–0.77) |
| p-value | 0.00 | 0.00 | 0.00 | 0.01 | 0.00 |
| RN, n | 0.00 (0.72) | 0.02 (0.00) | 0.00 (0.97) | –0.03 (0.00) | –0.03 (0.02) |
| HCA, n | 0.05 (0.00) | 0.05 (0.00) | 0.04 (0.00) | 0.08 (0.00) | 0.03 (0.18) |
| Admission to the HASU within 4 hours | |||||
| Band 8 | 0.64 (0.49–0.79) | 0.66 (0.51–0.81) | |||
| Band 7 | 0.63 (0.59–0.68) | 0.61 (0.56–0.66) | 0.61 (0.47–0.75) | 0.60 (0.28–0.91) | |
| Band 6 | 0.65 (0.63–0.66) | 0.62 (0.60–0.64) | 0.68 (0.65–0.71) | 0.68 (0.66–0.71) | 0.70 (0.66–0.74) |
| Band 5 | 0.61 (0.57–0.64) | 0.60 (0.55–0.66) | 0.51 (0.44–0.59) | 0.62 (0.56–0.69) | 0.75 (0.66–0.84) |
| Band/agency | 0.67 (0.60–0.75) | 0.67 (0.54–0.79) | 0.61 (0.45–0.77) | 0.78 (0.66–0.90) | 0.56 (0.26–0.86) |
| p-value | 0.22 | 0.87 | 0.00 | 0.14 | 0.38 |
| Separate nurses | 0.65 (0.63–0.66) | 0.62 (0.60–0.63) | 0.67 (0.63–0.70) | 0.69 (0.65–0.73) | 0.73 (0.67–0.79) |
| One nurse | 0.63 (0.61–0.65) | 0.62 (0.59–0.65) | 0.61 (0.56–0.66) | 0.66 (0.63–0.70) | 0.68 (0.63–0.74) |
| p-value | 0.19 | 0.94 | 0.06 | 0.40 | 0.29 |
| RN, n | 0.02 (0.00) | 0.03 (0.00) | 0.02 (0.06) | 0.00 (0.89) | 0.02 (0.16) |
| HCA, n | 0.02 (0.00) | 0.03 (0.00) | 0.01 (0.37) | 0.03 (0.01) | –0.01 (0.52) |
| Thrombolysis rate | |||||
| Band 7 | 0.85 (0.75–0.94) | 0.87 (0.77–0.96) | |||
| Band 6 | 0.89 (0.87–0.91) | 0.89 (0.86–0.91) | 0.88 (0.84–0.91) | ||
| Band 5 | 0.84 (0.78–0.90) | 0.81 (0.70–0.92) | 0.81 (0.66–0.96) | ||
| Band/agency | 0.82 (0.66–0.99) | 0.87 (0.65–1.09) | 0.41 (0.06–0.76) | ||
| p-value | 0.37 | 0.52 | 0.05 | ||
| Separate nurses | 0.88 (0.86–0.90) | 0.87 (0.84–0.90) | 0.88 (0.82–0.94) | ||
| One nurse | 0.86 (0.82–0.90) | 0.90 (0.85–0.95) | 0.84 (0.76–0.92) | ||
| p-value | 0.35 | 0.36 | 0.50 | ||
| RN, n | 0.00 (0.71) | 0.00 (0.67) | 0.02 (0.35) | ||
| HCA, n | 0.01 (0.12) | 0.00 (0.61) | 0.02 (0.50) | ||
| Door-to-needle time < 1 hour | |||||
| Band 8 | 0.84 (0.55–1.13) | 0.86 (0.60–1.12) | |||
| Band 7 | 0.76 (0.65–0.87) | 0.72 (0.60–0.85) | 0.73 (0.35–1.10) | ||
| Band 6 | 0.85 (0.83–0.87) | 0.85 (0.82–0.88) | 0.90 (0.86–0.94) | 0.80 (0.75–0.86) | |
| Band 5 | 0.81 (0.74–0.88) | 0.85 (0.76–0.94) | 0.66 (0.45–0.87) | 0.81 (0.66–0.95) | |
| Band/agency | 0.80 (0.63–0.96) | 0.82 (0.58–1.05) | 0.95 (0.80–1.09) | 0.76 (0.43–1.08) | |
| p-value | 0.37 | 0.31 | 0.07 | 0.96 | |
| Separate nurses | 0.84 (0.81–0.86) | 0.83 (0.80–0.86) | 0.91 (0.87–0.95) | 0.78 (0.70–0.86) | |
| One nurse | 0.85 (0.81–0.89) | 0.87 (0.81–0.92) | 0.75 (0.62–0.87) | 0.82 (0.74–0.90) | |
| p-value | 0.58 | 0.28 | 0.01 | 0.56 | |
| RN, n | 0.01 (0.27) | 0.02 (0.02) | –0.01 (0.40) | –0.02 (0.18) | |
| HCA, n | 0.01 (0.20) | 0.01 (0.45) | 0.02 (0.22) | 0.01 (0.78) | |
| mRS score of 3–6 at HASU discharge | |||||
| Band 8 | 0.54 (0.40–0.68) | 0.58 (0.44–0.72) | |||
| Band 7 | 0.56 (0.52–0.61) | 0.55 (0.51–0.60) | 0.62 (0.51–0.74) | 0.27 (0.02–0.53) | |
| Band 6 | 0.60 (0.59–0.61) | 0.59 (0.58–0.61) | 0.61 (0.58–0.64) | 0.60 (0.58–0.63) | 0.58 (0.54–0.62) |
| Band 5 | 0.57 (0.54–0.60) | 0.60 (0.56–0.65) | 0.64 (0.58–0.70) | 0.51 (0.46–0.57) | 0.42 (0.33–0.51) |
| Band/agency | 0.65 (0.59–0.72) | 0.68 (0.57–0.80) | 0.70 (0.58–0.82) | 0.60 (0.46–0.73) | 0.66 (0.39–0.92) |
| p-value | 0.05 | 0.21 | 0.52 | 0.02 | 0.01 |
| Separate nurses | 0.58 (0.57–0.60) | 0.58 (0.56–0.59) | 0.61 (0.59–0.64) | 0.55 (0.52–0.59) | 0.57 (0.51–0.64) |
| One nurse | 0.62 (0.60–0.64) | 0.63 (0.60–0.66) | 0.63 (0.59–0.67) | 0.61 (0.58–0.65) | 0.54 (0.49–0.59) |
| p-value | 0.00 | 0.00 | 0.55 | 0.05 | 0.46 |
| RN, n | 0.00 (0.92) | 0.01 (0.21) | –0.01 (0.08) | 0.01 (0.30) | –0.03 (0.07) |
| HCA, n | 0.03 (0.00) | 0.02 (0.00) | 0.05 (0.00) | 0.05 (0.00) | 0.00 (0.84) |
| In-hospital mortality at 3 days | |||||
| Band 7 | 0.03 (0.01–0.04) | 0.03 (0.02–0.05) | |||
| Band 6 | 0.03 (0.02–0.03) | 0.03 (0.02–0.03) | 0.04 (0.03–0.06) | 0.03 (0.02–0.04) | |
| Band 5 | 0.02 (0.01–0.03) | 0.04 (0.02–0.07) | 0.01 (0.00–0.02) | 0.01 (0.00–0.03) | |
| Band/agency | 0.02 (0.00–0.05) | 0.03 (–0.02–0.08) | 0.07 (–0.04–0.18) | ||
| p-value | 0.92 | 0.61 | 0.02 | 0.23 | |
| Separate nurses | 0.03 (0.02–0.03) | 0.03 (0.02–0.03) | 0.05 (0.03–0.06) | 0.03 (0.02–0.04) | |
| One nurse | 0.03 (0.02–0.03) | 0.04 (0.02–0.05) | 0.03 (0.01–0.04) | 0.03 (0.01–0.04) | |
| p-value | 1.00 | 0.35 | 0.10 | 0.97 | |
| RN, n | 0.00 (0.89) | 0.00 (0.87) | 0.00 (0.94) | 0.00 (0.57) | |
| HCA, n | 0.00 (0.24) | 0.00 (0.10) | 0.00 (0.98) | 0.00 (0.64) | |
| HASU LOS | |||||
| Band 8 | 3.19 (2.34–4.04) | 3.17 (2.30–4.04) | 1.92 (–1.26–5.10) | ||
| Band 7 | 2.93 (2.70–3.17) | 2.79 (2.55–3.04) | 3.44 (2.79–2.79) | 4.03 (1.41–6.65) | |
| Band 6 | 3.13 (3.06–3.20) | 3.13 (3.04–3.22) | 3.35 (3.20–3.20) | 3.00 (2.85–3.16) | 3.17 (2.93–3.41) |
| Band 5 | 3.34 (3.15–3.53) | 3.21 (2.92–3.50) | 3.79 (3.38–3.38) | 3.27 (2.87–3.67) | 3.39 (2.81–3.98) |
| Band/agency | 3.11 (2.67–3.56) | 2.94 (2.26–3.63) | 3.74 (2.80–2.80) | 2.72 (1.92–3.51) | 5.07 (2.64–7.49) |
| p-value | 0.11 | 0.16 | 0.22 | 0.53 | 0.14 |
| Separate nurses | 3.18 (3.10–3.26) | 3.16 (3.07–3.26) | 3.46 (3.29–3.62) | 2.95 (2.73–3.18) | 3.30 (2.94–3.66) |
| One nurse | 3.07 (2.96–3.18) | 2.91 (2.75–3.08) | 3.38 (3.14–3.62) | 3.13 (2.91–3.36) | 3.18 (2.87–3.49) |
| p-value | 0.13 | 0.02 | 0.63 | 0.31 | 0.64 |
| RN, n | –0.07 (0.00) | –0.06 (0.06) | –0.18 (0.00) | 0.02 (0.75) | –0.11 (0.15) |
| HCA, n | –0.08 (0.00) | –0.09 (0.00) | 0.04 (0.50) | –0.09 (0.24) | –0.20 (0.12) |
| Hospital LOS | |||||
| Band 8 | 7.84 (4.33–11.34) | 8.34 (4.45–12.23) | 12.36 (–16.79–41.50) | 3.09 (–4.98–11.16) | |
| Band 7 | 10.20 (8.76–11.64) | 9.30 (7.86–10.74) | 14.52 (9.24–19.80) | 7.84 (0.68–15.01) | |
| Band 6 | 11.13 (10.58–11.68) | 10.76 (10.03–11.49) | 13.72 (12.04–15.40) | 10.67 (9.55–11.79) | 12.84 (10.27–15.41) |
| Band 5 | 10.90 (9.81–12.00) | 11.68 (9.86–13.51) | 13.61 (10.71–16.52) | 9.40 (7.49–11.31) | 10.82 (7.13–14.51) |
| Band/agency | 11.20 (8.53–13.86) | 11.44 (7.03–15.85) | 16.96 (9.16–24.76) | 9.24 (5.02–13.45) | 15.37 (1.91–28.83) |
| p-value | 0.45 | 0.18 | 0.92 | 0.60 | 0.54 |
| Separate nurses | 11.04 (10.47–11.61) | 10.76 (10.03–1.49) | 14.20 (12.42–15.97) | 9.60 (8.27–10.92) | 14.27 (10.77–17.77) |
| One nurse | 11.03 (10.26–11.80) | 10.44 (9.31–11.57) | 12.89 (10.80–14.97) | 11.20 (9.72–12.67) | 11.14 (8.58–13.71) |
| p-value | 0.98 | 0.60 | 0.24 | 0.10 | |
| RN, n | –0.25 (0.05) | –0.10 (0.23) | –0.68 (0.05) | –0.03 (0.91) | –0.89 (0.12) |
| HCA, n | –0.10 (0.47) | –0.18 (0.16) | 0.69 (0.11) | –0.15 (0.72) | –1.59 (0.08) |
Impact of the seniority of the nurse in charge on clinical processes
After controlling for time of admission across the week, number of staff, and separation of ward management and thrombolysis roles, a higher band of the nurse in charge would result in a higher probability that a patient would undergo a brain scan within 1 hour (a probability of 0.56 if the nurse-in-charge was band 5, rising to a probability of 0.80 if the nurse in charge was band 8). Dysphagia screening within 4 hours was also affected by the band of the nurse in charge (see Table 16), with a probability of 0.70 if the nurse in charge was band 5 and rising to a probability of 0.76 if the nurse-in-charge was band 7.
There was a significant variation in the effect of nurse staffing measures during the four time periods across the week; this reflects the importance of the contribution of senior nurses to co-ordination within (improving the likelihood of dysphagia screening on time) and outside (improving the likelihood of brain scanning on time) the HASU. The effect of nurse-in-charge seniority was significant in hours, Monday to Friday, for brain scanning; in hours, Saturday to Sunday, for dysphagia assessment; and out of hours, Monday to Sunday, for the mRS measure. The probability of admission to the HASU within 4 hours did not vary significantly with nurse staffing, except for a higher probability of patients being admitted in hours over the weekend if a more senior nurse was in charge of the ward.
Impact of separating the nurse in charge and thrombolysis nurse roles on clinical processes
The availability of a dedicated thrombolysis nurse who was not the nurse in charge was associated with a higher probability of completing the dysphagia assessment within 4 hours by 13 percentage points (from 66% to 79%). The probability of completing the brain scan within 1 hour was higher when the nurse in charge was also the thrombolysis nurse. This result was particularly significant for patients not eligible for thrombolysis, admitted in hours, Monday to Friday.
The effect of having a separate tPA nurse and nurse in charge was significant in hours, Monday to Sunday, for brain scanning, in hours and out of hours, Monday to Sunday, for dysphagia assessment, and in hours, Monday to Friday, for the mRS measure. Better outcomes for dysphagia assessment and disability measures were associated with having separate nurses for these roles, whereas for brain scanning better outcomes were achieved when there was one nurse managing the ward and responsible for tPA. This result is reflecting the fact that a more senior nurse on duty is usually taking the role of the nurse in charge, and when they are busy with a new patient (e.g. arranging the brain scan) there is no one on the ward to provide the leadership that supports rapid dysphagia screening.
For example, we observed a new patient arriving in the evening at site H8. The nurse in charge was asked to supervise a dysphagia screen by one of the junior nurses who was being trained, but the tPA nurse was still down in the ED. She had the time to contribute to training and make sure the patient was screened properly. By contrast, in an in-hours observation in H1, the nurse in charge was a band 5 nurse who had stepped in for the day and was also the tPA nurse. She was struggling to cover all the patients as there were some sickness absences and was struggling, in particular, with some ‘confused’ patients [in-hours observation 1 (IH01) H1]. It is notable that these effects were observed for patients not eligible for tPA, mostly.
Impact of nursing on health outcomes
For the health outcomes measures, the availability of a dedicated thrombolysis nurse, different from the nurse in charge, and a more senior nurse in charge were associated with a lower probability of being discharged with at least moderate disability (mRS score of ≥ 3). Inpatient mortality was not significantly related to the nurse staffing, in hours or out of hours.
The number of nurses had a statistically significant impact on some processes and outcomes, but the size of the effect was very small (e.g. ± 5 percentage points when one extra nurse was working), suggesting a limited effect of changes in the number of nurses, once controlling for the roles and seniority. We noticed that an increase in the number of regular nurses was associated with a lower HASU LOS in hours, Saturday to Sunday: having one additional nurse decreases the LOS by 0.18 days (p < 0.01). On the other hand, one result from Chapter 3 suggested that patients admitted within the same time window were staying 0.4 days longer on the HASU. Combined, these findings indicate that employing two more regular nurses might bring down the LOS on HASU, but further analysis on cost and effects needs to be carried out to check whether or not employing extra staff compensates for the reduction of cost due to shorter LOS.
Discussion
There is variation in nurse staffing by time of the week, particularly in terms of the seniority of key roles, rather than the number of nurses overall. This is an important finding, as senior nursing roles are more differentiated than junior roles, in which bank nurses and other nursing staff can more easily substitute for missing personnel.
Qualitative analysis indicates that senior nurses tended to be more able to manage processes within the ward. Many HASUs timetable nurses at higher bands (e.g. band 7) to take on senior nursing tasks, such as the nurse-in-charge role, during the day in the week. This has an impact on the delivery of clinical interventions: nurse staffing had a direct effect on quality-of-care measures soon after arrival at the hospital (brain scan within 1 hour and dysphagia screen within 4 hours). In these cases, quality of care was better if the nurse in charge was a higher band and/or there were more nurses on duty. We explain why nurses with more experience at a higher banding would be more able to fulfil the many tasks expected of them within the time frame expected by the stroke standards. Although some HASUs provide training in particular tasks for more junior nurses, this does not enable them to juggle the many demands of the senior role that are so different from junior tasks.
Meeting the targets set by a financially incentivised set of standards has positive consequences for patient care. Variations in nurse staffing numbers across London HASUs have very limited effects on the likelihood of achievement of HASU standards. This may be because minimum nursing levels in London HASUs, which are stipulated in performance standards, are sufficient to address variations in quality of care and, therefore, variations in nurse staffing over and above this minimum have little additional impact.
We observed limited variation in the number of nurses but greater average seniority of nurses working regular working-week hours (Monday–Friday, 08.00–19.59) than out of hours (Monday–Friday, 20.00–07.59) and at weekends, yet the actual number of patients arriving is not that different. It is partially attributed to the fact that the regular work shifts of senior nurses (Monday–Friday, 08.00–16.00 or 09.00–17.00) do not completely overlap with regular ward working shifts (Monday–Friday, 08.00–19.59).
In the quantitative analysis we did not find variation by nurse staffing in the performance of care processes related to thrombolysis. This finding is in line with our qualitative observations suggesting a certain hierarchy favouring tPA patients, despite London stroke service standards that specify that most patients should be undergoing rapid ‘front-door’ services.
On the other hand, the achievement of key processes for patients not eligible for thrombolysis might differ if there is not enough senior nurse leadership, and separate responsibilities of the nurse in charge and thrombolysis nurse. The greater leadership capacity of senior nurses increased the likelihood of adherence to the system: senior nurses know that the London system is meant to benefit all stroke patients, whereas less senior staff might bend in the face of staff who want to prioritise only the tPA subsample.
Our findings suggest that one nurse managing thrombolysis and the ward is sufficient for rapid access to brain scanning, but separate nurses for these roles would be more efficient for dysphasia screening within 4 hours. We noted, from observations in our qualitative study, that brain scans were more likely to be undertaken soon after arrival at the ED, whereas dysphagia screening was more likely to occur on the HASU. We also observed that the tPA nurse was more likely to spend time at the ED (i.e. seeing patients soon after they arrived at the hospital). If the tPA nurse was also the nurse in charge, then this suggests that the nurse in the ED would be relatively senior, which might explain why brain scans within 1 hour were more likely to be achieved. Conversely, if the nurse in charge is not the tPA nurse, then the nurse in charge is more likely to be located on the HASU, which is where dysphagia assessments typically occur. The senior nurse being located on the HASU might therefore explain why dysphagia screening is more likely to be completed within 4 hours if the nurse in charge and tPA nurse roles are separate.
Improved quality-of-care measures and their contribution towards health outcomes might be expected by ensuring better senior nurse coverage out of hours and during the weekend, especially for patients not eligible for thrombolysis. There are advantages and disadvantages when the nurse in charge is the thrombolysis nurse.
Limitations
There are several weaknesses in our analysis. First, nurse staffing shifts were not the same in all eight HASUs. We chose 12-hour time periods, starting at 08.00, to be consistent with the in-hours and out-of-hours definition adopted in Chapter 3. However, in several HASUs, the first shift started at 07.00 or 07.30, and in one HASU there were three daily shifts instead of two. This was taken into account in the handover control variable. Second, mRS score at discharge from a HASU is a complex measure that accounts for many aspects of patient care and, as uncovered in our qualitative observations, is also affected by the organisation of repatriation processes, availability of local services and HASU demands, which might not always reflect an outcome of quality of care. Third, our data set did not include medical staffing and, therefore, we were unable to control for this in our analyses, which could confound the results. Future research exploring the interaction between medical and nursing staffing in this context would be beneficial. Fourth, similar to what is discussed in Chapter 3, Discussion, there was limited follow-up of patients, including mortality and disability indicators not available in the SSNAP.
Summary
- There was variation in nurse staffing by time of the week, particularly in terms of the seniority of key roles, rather than the number of nurses overall.
- Senior nursing roles were more differentiated than junior roles, which can be substituted by bank nurses and other nursing staff.
- When senior nurses took on the nurse-in-charge role during the day in the week, it had an impact on the delivery of clinical interventions.