The Carter review into efficiency in the NHS recommended a new staffing metric be adopted to measure nursing resources, but workforce experts are not convinced
The Carter review of efficiency in the NHS published this year recommended that a new staffing metrics of care hours per patient day be adopted to measure and compare how hospitals are using their nursing resources. Some workforce experts have warned that this process fails to recognise the complexity of care and could lead to unsafe staffing levels. This article aims to clarify what is meant by care hours per patient day and factors that need to be considered before they are used to determine staff numbers.
Citation: Hunt J (2016) How should nursing productivity be measured? Nursing Times; 112, 39/40: 18-20.
Author: Jennifer Hunt is visiting professor at Anglia Ruskin University.
- This article has been double-blind peer reviewed
- Scroll down to read the article or download a print-friendly PDF here
Nurses make up the largest single group of healthcare workers in acute hospitals and absorb a high proportion of the total budget. As a result, it is no surprise that the Carter review (2016) of NHS productivity and efficiency in non-specialist acute hospitals looked at the use of nursing resources.
Lord Carter’s (2016) review recommends that care hours per patient day (CHPPD) should become the principal measure of how hospitals use nurses and healthcare assistants from April this year. However, many nurses have found this new metric confusing, with terms such as CHPPD used in the review not being fully understood. It is important that nurses understand these terms and their implications so that they can use their expert knowledge and skills to influence decisions on how nurse numbers and grades are determined.
Analysis of registered nurse (RN) and healthcare support worker (HCSW) staffing data in the hospitals participating in the Carter review shows variation, which is not surprising. All studies on nursing workforce in hospitals show differences in total numbers, skill and grade mix of staff. This can be due to a number of reasons, such as the presence or absence of an intensive care unit or coronary care unit in hospitals which have 1:1 RN staffing. However, Carter goes further by stating explicitly that this variation is ‘unwarranted’.
He suggests that to ensure optimum use of staff resources, such as nursing, benchmarks and indicators should be standardised so the same metrics are produced across all hospitals. For nursing, Carter proposes using CHPPD as the preferred option to replace other commonly used measures, such as wholetime equivalents (WTEs) or nurse-patient ratios (NPRs). He suggests its use would enable more accurate comparisons, ensure productivity and efficiency are evaluated more easily and show how many staff are required and how many are available. In this article, CHPPD are explained and key issues involved in using the metric are outlined.
What are CHPPD?
Counting staff hours
CHPPD are calculated by adding the hours of RNs and HCSWs providing care during 24 hours and dividing the total by the total number of patients at the midnight census. Examples are outlined in Box 1.
To ensure consistency, clarification is needed as to what constitutes ‘worked hours’, in particular whether they mean:
- Official shift hours excluding meal breaks;
- Actual worked hours, including additional unpaid hours if staff are not able to take their meal breaks and do not leave on time;
- Only direct care hours.
There is also the question of whether ‘worked hours’ include all, none or a specified percentage of RN hours of staff such as ward managers, clinical specialists, student mentors and new graduates.
Counting patient hours
For patient hours, Carter proposes that a single time point be used to count the number of occupied beds, namely the midnight census. This is the most common approach used to record patient numbers and occupancy because it is easy to record and has high reliability. However, it may underestimate patient numbers and care hour requirements because it does not capture ‘churn’ – admissions, transfers, discharges, deaths and patients occupying a bed for less than 24 hours (Fieldston et al, 2012; Simon et al, 2011; Beswick et al 2010).
In the example in Box 1, the Wednesday/Thursday midnight census shows there were six patients in a six-bed room, but only five patients 24 hours later when the midnight census was recorded for Thursday/Friday. Yet on Thursday, nine different patients actually occupied beds, all of whom received care.
In addition, the midnight census does not take into account differences in the amount of care (acuity and dependency) each patient needs, nor the fact that a shorter length of stay increases the total care hours required per patient episode as shown in Figure 1 (Lawless, 2014), attached.
Metrics such as hours per patient day (HPPD) and nursing hours per patient day (NHPPD), referred to by Carter as CHPPD, have been used for decades in the US to examine nursing productivity both within and between hospitals. They are also used to determine staffing levels based on national or regional benchmarks and establish budgets for nursing departments. NHPPD usually refer to qualified nurses in the US, so may include only RNs with degrees or both RNs and those with older diplomas and/or licensed practical nurses. HPPD might or might not include all care staff. NHPPD are also used in several states in Australia and are usually used to describe qualified nurses only. This is why it is important to understand which staff are included before accepting staffing data.
Interpreting CHPPD data and deciding how best the metric can be used is complicated. In the US, Kirby (2015) has argued that there are important reasons why hospitals should not be using HPPD as their key metric either to compare hospitals or to evaluate nurse staffing, including:
- The results are not adjusted to account for factors that might require more nursing care hours, such as age of patients, severity of patient illness or differences in the amount of nursing care required for patients with the same diagnosis (Jenkins and Welton, 2014);
- The measure does not account for differences in frequency of admissions and discharges, or other factors that might affect nurse staffing needs;
- Varying physical layouts and the extent of other support services make different demands on nurses’ time;
- Variations in nurses’ experience, skill sets and competency vary widely between hospitals and influence the time spent in delivering care and how nurses respond to patient care needs.
Lord Willis has echoed these concerns stating that the Carter review “fails to recognise the complexity of care provided by nurses and could lead to unsafe staffing levels” (Lintern, 2016). At the very least, one needs to know both the budgeted CHPPD and the actual hours available, be that at hospital or ward level, and how the numbers and grades are determined. Meaningful comparisons can be made only between units matched on key criteria, such as staff grade and experience, patient diagnosis, treatment needs, acuity and dependency, support staff – just as would be done in any clinical trial.
A focus on CHPPD needs to link to quality outcomes (patient, nurse and hospital). Without these, just counting and comparing CHPPD provides the price but not necessarily the true cost. We should be wary of focusing on productivity and efficiency without also factoring in effectiveness. We now have considerable evidence to show that RN numbers and RN/unqualified skill mix ratios are related to patient outcomes (Needleman, 2011; Kane et al, 2007; Rafferty et al, 2007) and, as the number of patients looked after by each nurse increases, so does the number of adverse outcomes such as mortality and morbidity (Aiken et al 2014; Ball et al, 2013; Rafferty et al, 2007). Similarly, there is evidence to show that better patient outcomes are achieved with graduate RNs (Aiken et al, 2014).
CHPPD can be useful but can also be misunderstood and misused. Unless there are very precise definitions for all the key terms, we will be comparing oranges and apples. As nurses, we need to be sure that we know how the metric is defined and how it is used. We must ensure that the variations, which we know exist in the amount of nursing care required even for patients with the same diagnosis, are measured. We need to reflect on the research that has already been done and take the opportunity presented by developments in the capture of real-time information to ensure that the contribution of nurses is fully recorded and understood. Of course, we need to take into account productivity and efficiency, but it is effectiveness that matters most to patients and their families and that should be our main focus in determining nurse staffing numbers and skill mix.
- Lord Carter’s (2016) review of NHS productivity recommended that care hours per patient day (CHPPD) should be used to measure nurse and healthcare assistant resource
- This method replaces other commonly used measures such as whole time equivalents or nurse-patient ratios
- CHPPD are calculated by adding the hours of RNs and HCSWs providing care during 24 hours and dividing by the total number of patients at midnight
- Using the midnight census does not consider the ‘churn’ of patients during the day
- The approach does not consider the complexity of care or experience of staff
Aiken L et al (2014) Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study. The Lancet; 383: 9931, 1824-1830.
Ball JE et al (2013) ‘Care left undone’ during nursing shifts: associations with workload and perceived quality of care, BMJ Quality and Safety; doi:10.1136/bmjqs- 2012-001767.
Beswick S et al (2010) Comparison of nurse workload approaches. Journal of Nursing Management; 18: 5, 592-598.
Fieldston E et al (2012) Traditional measures of hospital utilization may not accurately reflect dynamic patient demand: findings from a children’s hospital. Hospital Pediatrics; 2: 1, 10-18.
Jenkins P and Welton J (2014) Measuring direct nursing cost per patient in the acute care setting. Journal of Nursing Administration; 44: 5, 257-262.
Kane RL et al (2007) The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Medical Care; 45: 12, 1195-1204.
Kirby K (2015) Hours per patient day: not the problem, nor the solution. Nursing Economics; 33: 1, 64-66.
Lawless J (2014) Nursing works: patients matter and nurses make the difference. Presentation to Safe Staffing Alliance. (unpublished).
Lintern S (2016) Carter Review’s care hours metric ‘too blunt a tool’, warns peer. Nursing Times.
Merrifield N (2016) Nursing workforce experts question new staffing measure. Nursing Times.
Needleman J et al (2011) Nurse staffing and inpatient hospital mortality. New England Journal of Medicine; 364: 1037-1045.
Rafferty AM et al (2007) Outcomes of variation in hospital nurse staffing in English hospitals: cross-sectional analysis of survey data and discharge records. International Journal of Nursing Studies; 44: 2175-2182.
Simon M et al (2011) Midnight census revisited: reliability of patient day measurements in US hospital units. International Journal of Nursing Studies; 48: 1, 56-61.