Staffing in acute hospital wards: part 2. Relationships betweengrade mix, staff stability and features of ward organizationalenvironment

ANN ADAMS P h D , R G N1 and SENGA BOND P h D , F R C N , R N

2

1Senior Research Fellow, Centre for Primary Health Care Studies, School of Health and Social Studies, University ofWarwick, Coventry and 2Professor of Nursing Research, Centre for Health Services Research, The University ofNewcastle upon Tyne, Newcastle upon Tyne, UK

Correspondence

A. Adams

Senior Research Fellow

Centre for Primary Health

Care Studies

School of Health and Social Studies

University of Warwick

Coventry CV4 7AL

UK

E-mail: a.e.adams@warwick.ac.uk

A D A M S A. & B O N D S. (2003) Journal of Nursing Management 11, 293–298

Staffing in acute hospital wards: part 2. Relationships between grade mix, staffstability and features of ward organizational environment

This paper explores relationships between grade mix, staff stability, care organization

and nursing practice. The data were collected in the mid-1990s from a nationallyrepresentative sample of 100 acute hospital wards and 825 nurses. Analyses provides

important insights for managers seeking to achieve the strategic aims set out in con-

secutive National Health Service (NHS) human resource management policies.

Hypotheses about ward clinical grade mix were not well supported. Where there

was rich grade mix, nurses reported better collaborative working with other dis-

ciplines and greater influence. However, it was expected that wards practising

�devolved� nursing would have a richer grade mix and that the latter would lead to

more innovative practice and nurses experiencing greater job satisfaction.

No evidence to support any of these hypotheses was found although the opposite

scenario – a link between poor grade mix, unprogressive practice and perceived

lower standards of care – was supported. Wards practising the �devolved� system

rely on adequate numbers of nurses rather than a rich grade mix, and do not

necessarily provide a more stable, retentive work environment for nurses.

By contrast, findings about staff stability were largely as expected. A strong link

between staff stability and standards of professional nursing practice was found,

indicating that staff stability is more important than a rich grade mix for achieving

innovative, research-based practice. However, staff instability undermined cohesion

with nurse colleagues, collaborative working with doctors, and nurses’ ability to

cope with the workload.

Overall, both the papers demonstrate that staffing resources and prevailing ethos

of care are more important predictors of care processes and job satisfaction than

organizational systems. They identify the detrimental effects on nurses and their

work of having few staff and a weak grade mix, and the importance of staff sta-

bility. Higher standards of nursing practice are achieved where stability is high,

independently of staffing characteristics.

Keywords: nurses, staff stability, grade mix, ward organizational practice, work envi-

ronment, acute hospital wards, standards of practice

Accepted for publication: 21 October 2002

Journal of Nursing Management, 2003, 11, 293–298

ª 2003 Blackwell Publishing Ltd 293

Introduction

Part 1 of this paper (Adams & Bond 2003) examined

the effects of number of nurses on care processes and

job satisfaction. Part 2 focuses on clinical grade mix and

staff stability.

While descriptive definitions of skill mix in nursing

are available (Gibbs et al. 1990, Nessling 1990,

Whynne 1995, Adams et al. 1998), there are no

standards for measuring it. Evaluating the effects of

different skill mixes on care processes and outcomes

therefore remains problematic. A proxy measure for

skill mix is clinical grade mix. Used by Carr-Hill et al.

(1992), a positive association between the ratio of more

senior nursing grades and quality care was found.

In the absence of skill mix measures, this study also

uses clinical nursing grade mix as a proxy measure of

skill. Associations are examined between clinical grade

mix, organizational practice and nurses� perceptions of

their working environment. This provides further evi-

dence about the importance of clinical grade mix in

nursing care provision, generated from survey data

collected from a nationally representative sample of

wards. This large survey contrasts with findings derived

from observational data from a small sample of wards

by Carr-Hill et al. (1992).

Staff stability

Improving recruitment and retention of nurses is cur-

rently one of the biggest challenges to achieving the

current NHS plan (DoH 2000). Sickness absence and

turnover are both common measures of staff stability,

and both have been linked to job satisfaction, perceived

work stress and number of staff. Low job satisfaction

and high stress are associated with both increased

sickness absence and turnover (Cavanagh & Coffin

1992). Greater numbers of staff are also associated

with increased sickness absence, while a rich grade

mix and increased responsibility have been associated

with reduced sickness absence (Clark 1975, Hoare

1985).

Interviews with nurses (Adams & Bond 1995) con-

firmed the importance they attach to staff stability in

relation to the social aspects of work organization:

creating cohesion amongst peers and developing effec-

tive working relationships. They also associated these

social aspects of ward life with the standards of practice

they were able to achieve, believing that greater stability

enhances quality of patient care. In the same study,

nurses reported that common staff substitution prac-

tices, i.e. using bank or agency staff, or nurses borrowed

from other wards, were of limited value. These staff are

considered �outsiders�, unfamiliar with culture and

practices, ignorant of patients� needs and requiring

substantial support. They add significant pressures on

nurses already coping with the stressful effects of staff

shortages.

Staff substitution is long standing and widespread

throughout the NHS. Procter (1995, unpublished data)

reported that bank and agency nurses were even used in

nursing development units (NDUs) to enable permanent

staff to be released for professional development activ-

ities. With increased emphasis on continuing profes-

sional development and life-long learning (UKCC 1996,

DoH 1999), it seems likely that staff substitution

practices will remain a permanent feature unless staff

release time is built into ward staffing establishments,

and that such establishments can be obtained.

While the negative effects of nursing staff instability

are well documented at the national (e.g. Buchan et al.

1998) and hospital levels (McClure et al. 1982, Aiken

et al. 1998), there is little empirical evidence available

about its effects at ward level. The above discussion

highlights the importance of examining the effects of

staff instability on a wider range of care processes and

nurses� job satisfaction, as well as identifying factors

that create or exacerbate it. Whether different forms of

instability, e.g. those associated with sickness/turnover,

absence and staff substitution practices have different

effects, remain important for health service managers.

Aims

This paper tests two sets of hypotheses. First, we

hypothesize a direct association between clinical grade

mix and ward organizational structures, standards of

practice and attitudes to care provision. We predict that

where grade mix is poor (i.e. where there is a low

proportion of more senior nursing grades within a

ward), nursing practice will be hierarchical, task ori-

entated and unprogressive (subsequently referred to as

�hierarchical practice�). However, where it is rich (i.e.

where there is a higher percentage of more senior nur-

sing grades within a ward), nursing will be character-

ized by devolved organizational structures and more

innovative, research-based practice (subsequently

referred to as �professional practice�). We further predict

that where grade mix is rich, nurses will have better

working relationships with other disciplines, exert more

influence over ward events and care provision and

experience greater job satisfaction.

Secondly, we hypothesize a link between staff sta-

bility and nurses� perceived standards of professional

A. Adams and S. Bond

294 ª 2003 Blackwell Publishing Ltd, Journal of Nursing Management, 11, 293–298

nursing practice. We predict that where instability is

high, ward nursing practice will be �hierarchical�, but

that where instability is low, practice is more likely to

be innovative and research-based. We further predict

that higher level instability will undermine the quality

of nurses� working relationships, their felt ability to

cope with the ward workload and their job satisfaction.

The study methods are described in Part 1 of this paper.

Data analysis

Clinical grade mix and sickness absence

In order to test our hypotheses, measures of clinical

grade mix and staff stability were created. Two meas-

ures of grade mix were created: based on the percent-

ages of nurses of clinical grades C and above and

clinical grades E and above. A ratio was calculated

between the number of days lost because of sickness

absence and the number of nurses within each ward.

In the first stage of subsequent data analysis descrip-

tive statistics characterized grade mix and staff stability.

The second stage comprised analysis of variances using

wards classified as the three different types of organ-

izational system described in Part 1 as an independent

variable. Finally, associations between ward grade mix

and staff stability were examined by correlating these

measures with WOFS subscales, using Pearson’s corre-

lation coefficient.

Findings

Clinical grade mix and ward nursing practice

The percentage of nurses of clinical grades C and above

(i.e. holding a UKCC registered qualification) ranged

from 56 to 100% in the study sample of wards (mean

83%). The percentage of nurses of clinical grades E and

above ranged from 24 to 88% (mean 46%). There was

considerable variation between ward nursing estab-

lishments.

Analysis of variance demonstrated a significant

association between clinical grade mix (percentage of

grades C and above) and the three systems of nursing

organization (P < 0.04). However, findings were not as

expected. Wards practising the �centralized� system had

the richest grade mix, rather than wards practising the

�devolved� system. Wards practising �two tier�nursing had

the lowest percentage of professionally qualified nurses.

Analysis of WOFS subscales demonstrated a signifi-

cant association between clinical grade mix and �hier-

archical� practice. A low negative correlation

(r ¼ )0.17, P < 0.0001) was obtained between �hier-

archical practice� and the percentage of professionally

qualified nurses in the ward establishment; and a

stronger, negative correlation (r ¼ )0.21, P < 0.0001)

between �hierarchical practice� and the percentage of

ward nursing staff of clinical grades E and above. These

results demonstrate that wards with a poor grade mix

are more likely to be characterized by a hierarchical,

task-orientated, unprogressive approach to care provi-

sion. However, contrary to expectations, a rich grade

mix was not associated with �professional practice� in

our sample of wards, highlighting that there is no ne-

cessary link between having a more senior nursing staff

and innovative practice.

Despite this, some grade mix results were as expected.

Grade mix was found to be significantly and positively

associated with nurses� scores on the three WOFS

influence subscales and with their perceptions of

multidisciplinary collaboration. In each case, the rela-

tionship was stronger when using the higher level

measure of ward clinical grade mix (i.e. the percentage

of nurses of grades E and above). These results

emphasize that where grade mix is richer, nurses within

a ward perceive themselves to be more empowered and

more able to work collaboratively with other disci-

plines. This reflects findings from our earlier work

(Adams & Bond 1995), where perceptions of influence

and collaborative working were positively linked to

individual nurses� clinical grades. The higher the grade,

the more influence or the more collaboratively the nurse

perceived her/himself to work. The results also suggest

that where there is a rich grade mix, ward nurses may

perceive themselves to have more aggregate influence,

regardless of individuals� clinical grades.

Staff stability and ward nursing practice

(i) Use of bank and agency nurses

Most wards (72%) used bank or agency nurses once or

more every week. Forty-three per cent of wards used

them once or twice per week, while 38% of wards used

them three or more times per week. These figures are

indicative of the shortfalls in nursing staff establish-

ments in the mid-1990s and of the amount of NHS

revenue spent on expensive temporary replacement

staff, although subsequent spending on recruiting nurses

from abroad is likely to account for an even larger

proportion of NHS staffing monies. Analysis of vari-

ance demonstrated a strong relationship between the

use of temporary staff and systems of ward nursing

organization (P < 0.0001). However, contrary to

Staffing in acute hospital wards: part 2

ª 2003 Blackwell Publishing Ltd, Journal of Nursing Management, 11, 293–298 295

expectations, wards practising the �devolved� system

had the highest mean usage rates, while wards practis-

ing the �centralized� system used them least often.

Interviews with nurses told us that the use of bank

and agency nurses was associated with the degree of

cohesion nurses experienced with their peers. This was

demonstrated by a statistically low negative correla-

tion (r ¼ )0.13, P < 0.003) between the numbers of

temporary staff used in wards on both day and night

duties and nurses� scores on the WOFS �cohesion�subscale. Level of use of bank and agency nurses was

also significantly associated with the �staff organiza-

tion� subscale measuring nurses� perceptions of their

ability to cope with the ward workload (r ¼ )0.16,

P < 0.0001). This indicates that where bank and

agency nurse usage is higher, nurses are more likely to

perceive themselves unable to cope; whereas when

temporary staff usage is low, nurses are more likely to

perceive ward staffing levels as sufficient to cope with

the workload. To a lesser degree, usage of bank and

agency nurses was also associated with the amount of

influence nurses perceived themselves to have over the

�timing of ward and patient events� (r ¼ )0.10,

P < 0.01). The presence of more bank and agency staff

is therefore likely to undermine nurses� sense of control

over ward events.

Use of bank and agency nurses is significantly asso-

ciated with nurses� perceptions of the standard of

practice achieved within wards, measured by the WOFS

�professional practice� subscale (r ¼ )0.13, P < 0.004).

Nurses perceive standards to be lower when more bank

and agency staff are used. Conversely, a stable ward

establishment is associated with higher perceived

standards of practice. This finding suggests that main-

tenance of staff stability, over and above staff number

and competency issues, is conducive to creating an

environment with innovative nursing practice.

(ii) Sickness absence

Working days lost because of sickness absence in the

12 weeks prior to data collection ranged from 0 to 185

in the sample of wards. The ratio of days lost in sickness

absence to the size of the ward nursing establishment,

demonstrated a range of between 0 and 5.1 days lost

per full time equivalent (FTE) nurse in the same period,

with a mean value of 2 days lost per nurse. It is

recognized that long-term sickness will inflate a ward’s

sickness absence/nurse ratio, and this was evident in

41% of wards.

Contrary to expectations, sickness absence was not

related to either numbers of staff or the clinical grade

mix within wards. Neither was it associated with a

ward’s system of nursing organization. However, sta-

tistically significant associations were found between

sickness absence and the two WOFS subscales meas-

uring standards of ward nursing practice. This suggests

ward attitudes to care provision, and not ward organ-

izational systems or structures, are associated with

nurses� sickness absence rates. The correlations were not

large. Between sickness absence and �hierarchical prac-

tice� (r ¼ 0.19, P < 0.0001), indicating that nurses are

more likely to be �off sick� in wards characterized by this

ethos of care. Between sickness absence and �profes-

sional practice� (r ¼ )0.09, P < 0.05), suggesting that

perceived standards of practice are undermined by

higher sickness absence rates.

Sickness absence was also found to undermine nurses�perceptions of their ability to cope with the ward

workload ()0.09, P < 0.05), and collaborative working

with medical colleagues ()0.11, P < 0.01). Together

results highlight the importance of maintaining a stable

ward staff.

Discussion

Both parts of this paper examined relationships

between ward staffing resources, systems of nursing

organization and nurses� perceptions of their working

environment. The hypothesized relationships concern-

ing ward clinical grade mix were not well supported.

Having a richer grade mix was found to be significantly

and positively related to nurses� perceptions of colla-

borative working with other disciplines and the amount

of influence they perceived themselves to have. How-

ever, it was also expected that there would be a richer

grade mix in wards practising �devolved� nursing and

that wards with richer grade mixes would be charac-

terized by more innovative practice and nurses experi-

encing greater job satisfaction. No evidence to support

any of these hypotheses was found. However, there was

support for a link between a poor grade mix and

�hierarchical practice�.By contrast, the findings related to staff stability were

largely as expected. Evidence of a strong link between

staff stability and standards of professional nursing

practice was found. As hypothesized, where instability

was high, standards were perceived to be low: where

instability was low, standards were perceived to be

higher. Staff instability was also shown to undermine

nurses� working relationships with their peers and high

sickness absence in particular was associated with

poorer working relationships with medical staff. While

no direct link between staff instability and job satis-

A. Adams and S. Bond

296 ª 2003 Blackwell Publishing Ltd, Journal of Nursing Management, 11, 293–298

faction was found, there was an association between

staff instability and nurses� stress, evident in their low-

ered perceptions of their ability to cope with the ward

workload.

Findings related to ward organizational systems and

staffing resources were frequently found to be incon-

sistent with common beliefs about nursing organiza-

tion. We began to develop this theme in the first part of

this paper and there is further evidence here. In partic-

ular, expected findings related to wards practising the

�devolved� system were not supported. Evidence of the

need for a richer clinical grade mix rather than an

adequate number of nurses to operate this system was

not found; nor was there any evidence that wards

practising this system provided a more stable, retentive

work environment for nurses. However, findings may

reflect a greater concern among nurses within devolved

systems to use bank and agency staff to release one

another for staff development activities (Procter 1995,

unpublished data).

The combined findings of the two parts of this paper

suggest that the relationship between staffing resources,

care processes and job satisfaction is of greater signifi-

cance than that between systems of nursing organiza-

tion, and care processes and job satisfaction. Systems of

nursing organization appear to be an intervening vari-

able in the original relationship. However, these results

may reflect limitations in the data as the number of

wards practising either the �devolved� and �centralized�systems was small, and no contextual information

about reasons for staffing data was available.

By contrast, ward culture or attitude to care (meas-

ured by the �hierarchical practice� and �professional

practice� subscales) was significantly related to both

ward staffing resources and stability, in line with the-

oretical predictions. This provides further evidence that

it is the ward ethos or philosophy of care and not the

organizational system or structure which has more

impact on care processes and nurses� job satisfaction

(Adams & Hardey 1992). Where low numbers and

poor grade mix were present, nurses perceived their

wards to be characterized by practice devoid of

innovation and where they perceived themselves to be

both disempowered and their work devalued by

management and medical colleagues. This negative

ward ethos was also found to be significantly associated

with staff instability: with high sickness absence and

high usage of bank and agency nurses.

Carr-Hill et al. (1992) found evidence of a link

between clinical grade mix and the quality of care

provided for patients, where a �rich� grade mix was

associated with higher quality care, and vice versa.

However, in our findings, a more professional approach

to practice was not found to be associated with

available staffing resources, but rather with staff sta-

bility. This was an unexpected finding. Substituting

bank and agency nurses for absent members of perma-

nent ward staff and high sickness absence rates created

working conditions which undermine nurses� per-

ceptions of professional practice, i.e. practice that is

research based, characterized by innovation, staff dev-

elopment and where patients and families are involved

in care decisions. These findings suggest a link between

innovative practice and established working relation-

ships, which are upset by staff instability.

Nurses� personal reactions to their work were affected

by both ward staff characteristics and staff instability.

The WOFS subscale �staff organization� relates to

nurses� perceptions of their ability to cope with the

workload within existing staffing and modes of organ-

izational. As such it can be used as a proxy measure of

work stress. Where there were more staff, nurses per-

ceived themselves more able to cope with the ward

workload. Where staffing was unstable because of high

sickness absence, high turnover and high usage of bank

and agency nurses, nurses perceived themselves less able

to cope and under more stress. Similarly, nurses� job

satisfaction was enhanced where number of staff was

higher.

Conclusions

This paper provides a systematic analysis of relation-

ships between ward staffing resources, systems of ward

nursing organization and nurses� perceptions of their

work environment. The findings highlight important

issues for health service managers concerned with

building effective ward nursing establishments.

Four key findings have emerged. First, staffing

resources were found to have a bigger impact on care

processes and nurses� job satisfaction than systems of

ward nursing organization. Many postulated theoretical

relationships between organizational systems, care

processes and job satisfaction, and between systems

and staffing resources were not supported by our

analysis.

Secondly, ward ethos or philosophy of care was more

consistently related to staffing resources and nurses�perceptions of their work environment than were sys-

tems of nursing organization. These two findings high-

light that system of ward nursing organization is an

intervening variable in the relationship between staffing

resources and nurses� perceptions of their work envi-

ronment and the standard of practice they achieve at

Staffing in acute hospital wards: part 2

ª 2003 Blackwell Publishing Ltd, Journal of Nursing Management, 11, 293–298 297

ward level, but that it is not the major predictor of the

latter. This needs to be taken account of in future

research, and emphasizes the need to base staffing

policy decisions on empirical evidence, that is open to

challenge, rather than conjecture.

Thirdly, the analysis has highlighted the detrimental

effects of having few staff and a poor clinical grade mix

within a ward. These were conditions associated with

hierarchical organizational structures and attitudes to

nursing, which militate against achieving high standards

of professional practice. These structures and attitudes

are in turn associated with high levels of sickness

absence and high staff turnover rates.

Finally, the findings provide strong support, over and

above any economic argument, for making staff sta-

bility a priority in HRM. Staff stability, and not num-

bers or grade mix was found to be positively associated

with the level of professional practice achieved within a

ward and with the degree of cohesion amongst nurses

and improved ability to cope with the ward workload.

These findings re-emphasize the importance and

urgency of implementing solutions to NHS nurse

recruitment and retention problems (NHSE 2000).

Employing an adequate number of nurses and a suffi-

ciently rich grade mix have been shown to be associated

with a range of care processes and job satisfaction.

Manipulating systems of work organization in them-

selves will not bring about these benefits in the absence

of sufficient staff of the right calibre. Employing bank

and agency staff is not an effective solution. Their

long-term use can set up a negative cycle of events,

precluding the development of cohesive working

relationships, lowering morale and fuelling nurse turn-

over. Commitment is required to developing stable

teams comprising adequate staff of the right grade mix,

in line with the strategic aims of Working Together

(DoH 1998). Such a strategy is likely to benefit patients,

nurses and the NHS Trusts alike.

Acknowledgements

We would like to thank all the nurses and managers who sogenerously gave their time to assisting us in this study, andresearch staff at the Universities of Surrey and Newcastle whoassisted with data collection. The study was supported by agrant from the Department of Health. The views expressed arethose of the authors and not the funding body.

References

Adams A. & Bond S. (1995) Nursing Organisation in Acute Care:

The Development of Scales. Report no. 71. Centre for Health

Services Research, University of Newcastle upon Tyne, New-

castle upon Tyne.

Adams A. & Bond S. (2003) Staffing in acute hospital wards:

part 1. The relationship between nurses and ward organiza-

tional environment, Journal of Nursing Management 11 (5),

287–292.

Adams A. & Hardey M. (1992) Primary panacea? Nursing Times

88 (26), 34–35.

Adams A., Hugsden E., Chase J., Arber S. & Bond S. (1998) Skill

Mix Decisions in the NHS. Report submitted to the Depart-

ment of Health.

Aiken L.H., Sloane D.M. & Sochalski J. (1998) Hospital orga-

nisation and outcomes. Quality in Health Care 7, 222–226.

Buchan J., Seccombe I. & Smith G. (1998) Nurses� Work: An

Analysis of the UK Labour Market. Ashgate Press,

Aldershot.

Carr-Hill R., Dixon P., Griffiths M., Higgins M., McCaughan D.

& Wright K. (1992) Skill Mix and the Effectiveness of Nursing

Care. Centre for Health Economics, University of York,

Yorkshire.

Cavanagh S.J. & Coffin D.A. (1992) Staff turnover among hos-

pital nurses. Journal of Advanced Nursing 17, 1369–1376.

Clark M. (1975) Time Out: A Study of Absenteeism. Royal

College of Nursing, London.

Department of Health (1998) Working Together: Securing a

Quality Workforce for the NHS. DoH, London.

Department of Health (1999) Continuing Professional Develop-

ment: Quality in the New NHS. DoH, London.

Department of Health (2000) The NHS Plan: A Plan for Invest-

ment, a Plan for Reform. DoH, London.

Gibbs I., McCaughan D. & Griffiths M. (1990) Skill Mix in

Nursing: A Selective Review of the Literature. Centre for

Health Economics, University of York, Yorkshire.

Hoare C. (1985) Off Sick: An Investigation into the Absence

Behaviour of Nurses in Training. Unpublished MSc Thesis,

University of Surrey, Surrey.

McClure M., Poulin M., Sovie M. & Wandelt M. (1982) Magnet

Hospitals: Attraction and Retention of Professional Nurses.

American Nurses� Association, Kansas City.

Nessling R. (1990) Skill Mix: A Practical Approach for Health

Care Professionals. Manpower Monograph no. 2. DoH,

London.

NHSE (2000) Recruiting and Retaining Nurses, Midwives and

Health Visitors in the NHS. Department of Health, London.

UKCC (1996) The Future of Professional Practice: The Council’s

Standards for Education and Practice Following Registration

(PREP) – Transitional Arrangements for the Specialist Practi-

tioner Title/Qualification. UKCC, London.

Whynne T. (1995) Skill mix in nursing: efficiency and quality?

Journal of Nursing Management 3, 189–191.

A. Adams and S. Bond

298 ª 2003 Blackwell Publishing Ltd, Journal of Nursing Management, 11, 293–298