On 17 January, the government announced the publication of performance league tables on hospital consultants so patients can judge which hospitals are the safest for treatment. Our research suggests that patients looking for the best chance of a cure would benefit just as much from league tables on the performance of HR directors, showing which HR practices they have implemented and whether they are members of the hospital board.

In a study of 61 hospitals in England, we found strong associations between HR practices and patient mortality. As reported in People Management, (11 Oct 2001 and 8 Nov 2001), the extent and sophistication of appraisal systems in hospitals was closely related to lower mortality rates, but there were also links with the quality and sophistication of training and the number of staff trained to work in teams.

Why is this? Research into ‘progressive’ HRM practices suggests that these measures enhance productivity and profitability by improving the knowledge, skill, motivation and performance of employees (Huselid, 1995; Patterson et al., 1997). Research has shown that specific HRM practices, such as selection and training, are associated with improved job performance. Studies have also found that progressive HRM practices can enhance behaviour. People are more cooperative and helpful with colleagues and are better teamworkers.

But little of this research has been conducted in hospital settings and we don’t know whether HRM practices here are related to performance in the same way.

A US study examined the relationship between the organisation of nursing care and mortality rates (Aiken, Smith, and Lake, 1994). Hospitals that were able to attract and retain good nurses and provide good nursing care (termed ‘magnet’ hospitals) were compared with other hospitals. It found that mortality rates were around 5 per cent lower in the magnet hospitals than the others. But this isn't conclusive. It could just be that nurses were attracted to work in places that had a good reputation and were more likely to stay.

Our aim was to show not just whether there is a link between HR practices and quality of care, but which practices affect these outcomes. Chief executives and HR directors from 61 hospitals completed a questionnaire detailing HR strategy, policies and procedures. The survey gathered information on four areas: hospital characteristics; hospital HRM strategy; employee involvement strategy and practices; and HRM practices and procedures. Questions on HRM practices and procedures were asked separately in each of the main occupational groups – doctors, nurses and midwives, professions allied to medicine (eg physiotherapists), ancillary staff, professional and technical staff, administration and clerical staff, and managers. HR departments had between two and 60 staff (an average of 21) and seven, on average, had professional qualifications.

Those we surveyed and interviewed were also asked about the size of the hospital training budget: the extent to which each occupational group had access to a formal written statement about training policy and entitlements (48 out of 61 said this was available to all staff groups; seven said it was available to none; and 13 to some), the percentage of staff in each occupational group receiving at least three days of formal off-the-job training in the previous year (92 per cent of doctors and 26 per cent of ancillary staff) and the frequency of training needs assessments for each of the main occupational groups (90 per cent were assessed annually).

The HR directors gave information about the percentage of staff in the hospital working in teams. Twelve hospitals claimed that all their staff worked in teams, 23 said between 80 and 99 per cent and 24 trusts failed to provide any information. However, training in teamworking was less common. Most hospitals reported that less than a third of their staff had been trained in teamworking.

They were also asked what percentage of staff in each occupational group had received an appraisal in the last year; the frequency of appraisals; the number of staff conducting appraisals in each group who were trained to do so; and what methods were used to evaluate the appraisal system and process (eg appraisers and appraisees completing evaluation forms, monitoring by the HR department). There was considerable variation, with some staff in some hospitals having no appraisals (usually ancillary staff, doctors and administrative staff), while other hospitals reported appraising all staff. There was most variation in the percentages of staff receiving training in conducting appraisals, with many hospitals failing to provide it.

Overall, appraisal was carried out annually in 85 per cent of cases. HR departments checked that appraisals were carried out in 43 of the 71 hospitals from which we gathered data (some of these hospitals were not used in the analysis because of incomplete data).

Measuring mortality was more difficult than imagined because of variations between regions in health and socioeconomic status. We were able to gather data on deaths following emergency surgery, non-emergency surgery, admission for hip fractures, admission for heart attacks, re-admission rates and a mortality index.

We took great care not to bias interpretation of data and used different researchers to collect HR data from those collecting the mortality data and those analysing the results as we did not want the analysis to be influenced by knowledge of the hospitals. Even so, there is the danger that the results could be due to other factors such as geographical or regional variations in mortality. Hospitals in poor areas might have high mortality rates and might attract less able HR practitioners, the high flyers preferring to go to the top teaching hospitals. We therefore subtracted any effects caused by the size and wealth of the hospital measured by hospital income and local health needs since both are likely to vary with mortality. To obtain a measure of local health needs we used the government’s high level indicator statistics, which measures the ratio of deaths to expected number of deaths in a local authority area.

The final analyses showed a strong relationship between HRM practices overall and patient mortality. This was exciting but needed deeper investigation. With the link between HR practices and mortality established, we set out to determine which individual HR practices, if any, affected patient mortality.

We found that:
• Appraisal has the strongest relationship with patient mortality.
• The extent of teamworking in hospitals is also strongly related to patient mortality.
• Sophistication of training policies is linked to lower patient mortality.

However, our findings received a setback. A senior medic pointed out that the results could be due to differences in the numbers of doctors per 100 patients, since this factor has been shown to relate to patient mortality (Jarman et al., 1999). We gathered this data from hospital statistics and indeed found a relationship – more doctors in a hospital meant fewer deaths. But even after taking account of this, the links between the HR variables and patient mortality remained robust.

We know that in private sector organisations there are links between HR practices and organisational productivity (as our own research with the CIPD demonstrated: Patterson et al., 1997). Training, appraisal and teamworking were powerful factors that predicted productivity in the private sector. But this is the first study of hospitals that has established similar relationships with performance. Our previous research has shown that teamworking is a good predictor of quality and levels of innovation in patient care in NHS primary health and community mental health teams (see Borrill, et al., 2000 and www.research.abs.aston.ac.uk/achsor/achsor.html).

But we have encountered considerable scepticism from doctors who seek demonstrations of how HRM practices improve individual surgeons’ performance in ways that reduce mortality rates. Our answer is simple, though it may seem strange to those who deal with individuals rather than organisations. If you have HR practices that focus on effort and skill; develop people’s skills; encourage co-operation, collaboration, innovation and synergy in teams for most, if not all, employees, the whole system functions and performs better.

The effects show across the board, even in measures of performance as fundamental as patient deaths. If the receptionists, porters, ancillary staff, secretaries, nurses, managers and, yes, the doctors are working effectively, the system as a whole will function effectively. Our findings are based on HR practices applied across all staff combined rather than individual groups, suggesting that managing all staff well, not just medical and nursing staff, gets good performance.

Another striking finding is that where HR directors are members of the board of the hospital, the association between HR practices and patient mortality are even stronger. Just over half of those who replied to our survey said the person responsible for HR was a voting member of the board. This mirrors research in the private sector that shows that having political influence at the strategic apex of organisations is important to the potency of the HR function (see news, this issue).

We need to extend this research by including more hospitals. Without further research we cannot know whether the relationships described here are causal and, if so, in what direction. It could be that managers in hospitals that achieve low levels of patient mortality relax their focus on patient outcomes and give more attention to the management of employees. It may also be that hospitals that manage patient care well also happen to manage other aspects of organisational functioning well, including the management of employees, but there is no causal relationship between the two.

We can’t say for certain that emphasis on improving aspects of HR in hospitals will improve healthcare. However, the idea that good HR practices predict positive organisational performance has been demonstrated in a wide variety of studies in private and public sector organisations. Our detailed case studies (each of which involved extensive interviews and surveys of all staff groups) in 10 of the hospitals included in the sample suggested that staff themselves perceived two HR initiatives in particular led to improved patient care: the introduction of teamworking and sophisticated appraisal systems. (see case study, page 32).

The resource implications are considerable. The magnitude of returns for investments in HR practice, can be substantial. For example, Huselid demonstrated that a significant change (of one standard deviation) in each HRM practice associated with good organisational performance (eg appraisal) could produce increased sales of £18,000 per employee.

On this logic, if we take the strongest association – that between deaths following admissions for hip fractures and appraisal – for hospitals of equal size and local population health needs, a significant improvement in the appraisal system would lead to the equivalent of 1,090 fewer deaths per 100,000 admissions. This is equal to more than 1 per cent of all admissions. In other words it could save 12.3 per cent of hospital deaths.

Even with a weaker relationship – that between team working and deaths following emergency surgery – 25 per cent more staff working in teams could produce 275 fewer deaths per 100,000 or 7.1 per cent of the total.

While this does not invalidate the need to employ more – and better – doctors, the focus on medical staff alone may be missing a crucial point. A hospital is a work community and it is effective management of all staff within that community that makes the difference to health.

Happiness is...teamwork
Now that community health service staff at Calderdale and Huddersfield NHS Trust work evenings and weekends, patients can be discharged from hospital more quickly. This is thanks to multidisciplinary team-working that has handed responsibility for service planning to those involved in its delivery – district nurses, health visitors, practice managers, administrative staff, practice nurses and GPs.

“Now people close to the action are empowered to plan joint care initiatives, discharges from hospital can be accommodated far more comprehensively than before. Networks have developed between discharge, planning teams and primary care teams so resources are allocated to meet needs,” explains Lynda Hanson, currently director of operations and facilities at Calderdale and Huddersfield NHS Trust, who pioneered the project.

The trust set up 23 self-managing teams, each under trained team coordinators. The teams receive on-going HR, clinical and budgetary support from three senior managers. Weekly support meetings promote inter-team communication and collaboration.

“We get far more ownership of issues,” says Hanson. “People are happier in their jobs because they are more involved.”

Evaluation of the teamworking initiative revealed various benefits
such as:
• A climate of cooperation has facilitated better inter-departmental and multi-agency communication and practices. Communication with agencies such as social services has improved.
• Patients are receiving more integrated care, and quicker discharge times, so hospital beds are available more quickly.
• There is more community care for children with problems such as diabetes and cystic fibrosis.
• Teams can target their budgetary resources at local health needs and set up specialist clinics to meet these needs.
• Specialised training requirements have been identified to meet local needs. Many team members have developed new skills in areas such as ‘nurse prescribing’.
• Better quality of working life. Individuals reported that they felt involved in decision-making processes, they benefited from the professional and emotional support they gained as part of a team, and they experienced greater job satisfaction.
• Reduced sickness absence rates.

“This research will be of enormous value in shaping HR strategy both nationally and locally. It doesn't change the direction of travel we are heading in but it certainly does reinforce some key issues, such as the importance of appraisals – gaining acceptance of the value of appraisals as something which is of direct benefit to patients as well as staff. And most especially the importance of teamworking and the beneficial impact it has on patient services and clinical performance.

“There will be some scepticism among clinical colleagues about the research until it has had wider airing or been replicated elsewhere. There are so many variables that ‘impact on mortality rates’ critics will say: ‘Are the statistics robust enough to really demonstrate it?’ If you are looking at mortality rates then you are looking at a much more complex set of issues than bottom line profitability.”
Mike Griffin,
HR Director, King's College Hospital NHS Trust

“There is very little research on the NHS linking HR policy and practice to clinical outcomes, so this is significant.
“What Michael West says – that good HR practice involving all or most employees makes the whole system work better – is also crucial. The Commission for Health Improvement would agree. It is not just a question of training or appraising doctors and nurses, you must include everyone.

“Emerging trends from CHI reviews [of hospital practices] show that HR policy is one of the crucial things at the centre of what makes the difference between a good and a bad organisation in terms of patient care.
“To be more specific, there is a lot of stuff coming out on leadership – that is leadership at all levels, at the top of the organisation and of clinical and non-clinical teams. Another is teamworking – the negative impact a dysfunctional team can have on the quality of patient care.

“And when we describe a poorly performing organisation, invariably the application of the appraisal process tends to be poor.

“At the moment a lot of prominence is given to waiting times and financial indicators and there is no mention of HR. We are looking to balance that out: my challenge to the NHS would be to develop an HR strategy that is patient-centred.
Harry Hayer,
HR director, Commission for Health Improvement

Further information
Michael West will be speaking about ‘Promoting creativity in teams’ on the first day of the CIPD’s HRD 2002 conference, 16-18 April at Olympia, London.
For more details about the conference and exhibition, call the CIPD on 020 8263 3434
or visit www.cipd.co.uk/HRD