Those involved in the improvement science community have much to celebrate as more resources become available to improve patient safety. Professor Kathleen Steven's letter in this issue reminds us there is still much to do and that progress depends on building the scientific basis for improved patient safety and health outcomes. Professor Stevens' letter builds on Professor Pearson's promotion of the Improvement Science Research Network (ISRN) in a recent editorial.1 For Professor Stevens the further development of improvement science requires better opportunities to collect data from broader populations and to test findings more broadly. Clearly, Professor Stevens intends the ISRN to lead by enabling access to research instruments and bibliographies organised around key concepts in improvement science and by providing media supports - newsletters, online interaction and webcasts.

The ISRN structure has similarities with the Joanna Briggs Institute (JBI), which has brought to the clinical and research community resources that are now widely used and appreciated. With access to online databases, the Cochrane Library, JBI resources, the Physicians' Information and Education Resource (PIER), BMJ Clinical Evidence, emedicine, UpToDate, quantitative and qualitative research methods of increasing sophistication, and now the ISRN, as well as other resources, the improvement science community has a wealth of supports to call on in the quest for patient safety and better healthcare outcomes.

Yet, the achievement of improved patient care will continue to elude us until the ethics of healthcare are taken more seriously, by which I mean that there is a challenge in healthcare delivery that cannot be resolved by empirical methods alone, even those that rely on inductive reasoning. The lack of compassion in the care of the elderly in National Health Service (NHS) hospitals requires far more critical reflection on the realities of everyday practice than is possible based on information science alone. Whereas it is possible to benchmark performance across healthcare systems using quantitative and qualitative indicators, and to show improved facility or system performance against baseline measures, what appears to be an erosion of compassion cannot be solved by more and better data, higher-level aggregations of data or collaboration within the quality improvement industry.

According to a popular UK newspaper, the Daily Mail, millions of elderly patients suffer pressure ulcers, hip fractures following falls or malnutrition during NHS hospital stays and most of these cases could be avoided.2 The reporter goes on to mention that there have been complaints for years of nurses not bothering to help elderly people eat their meals; of food placed out of the reach of patients and thrown away uneaten; of failure to identify 70% of patients with malnutrition, of 200 000 falls a year on NHS property; and of a 10% incidence rate of pressure ulcers. In response to these figures, NHS managers have reminded nurses of what they should be doing:

Nurses and midwives have a responsibility to make sure people under their care are appropriately nourished and hydrated - they need to have food and drink. This is a basic human requirement and fundamental to the care and recovery of patients.3

An earlier report released in November 2009 highlighted poor end-of-life care in the NHS.4 The report of the National Confidential Enquiry into Patient Outcome and Death draws attention to widespread poor communication between NHS doctors and among NHS doctors, patients and relatives. The final days of many patients are made worse by poor care as a result. Appropriate end-of-life care is prevented by doctors failing to judge on admission that a patient is at the end of life. Therefore, patients are deprived of the possibility of dying at home with their loved ones around them, of effective pain relief and of treatments that might help them. Of the 3000 patients included in the review, more than half were not seen by a senior consultant early enough and suffered less than adequate care as a result; and more than one-third did not get a high enough standard of care. In one quarter of the patients the delay in seeing a consultant was clinically important and access to computed tomography and magnetic resonance imaging scanning was a substantial problem. In one-fifth of hospitals 'very junior trainee doctors': signed "do not attempt resuscitation" orders. The poor standard of care was not confined to the elderly. The reviewers found evidence of hospitals having major challenges in delivering an acceptable standard of care when dealing with very sick children. Nevertheless, the elderly bear most of the burden of substandard end-of-life care.

Without data, the scale of these problems would be unknown and without research understanding the reasons for poor care of the elderly and other vulnerable patients would be anecdotal. Hence, improvement science is needed to identify and address healthcare practices in need of urgent reform. Benchmarking against system-wide performance can be and is an impetus for change, but it is not a complete answer to substandard care. Neither are recommendations on policy development, healthcare restructuring or more monitoring systems sufficient in themselves.

Strategies are required that address discrimination against the elderly in healthcare funding, problems in health workforce composition and deployment, inadequate staffing patterns, dysfunctional organisational cultures, the continuing gulf between policy statements and the institutionalised neglect that now passes for care of the elderly. There seems to be something morally lacking in a system that neglects patients almost to the point of abuse, which is not corrected despite surveillance by governmental watchdogs, patient advocacy organisations, improved data capture, sophisticated data analysis and policy initiatives in response to successive damning reports. Such strategies increase accountability in that healthcare organisations have access to data systems that encourage self-criticism and use of information to improve performance. However, such valuable and systematic approaches to improvement are unfortunately not enough to safeguard vulnerable patients.

Hospital executives in the USA have recognised that more is required and the emphasis is shifting from organisational surveillance to self-surveillance by the healthcare professional. One such initiative was taken at Northfield Hospital in Minnesota where an I promise campaign was introduced on the assumption that by making formal pledges, health professionals would be more likely to deliver the care they know they should be providing. After all, we keep our word, don't we?5

This is not to say that there are no longer compassionate nurses and doctors, nor to deny that there are nurses and doctors whose consciences trouble them because they cannot provide the care they know is necessary. Yet it seems an inescapable conclusion that we can no longer take for granted the idea that health professionals are motivated by compassion as much as by power, status and the need to provide for their families.

Rather my intention is to draw attention to what seems to be a crisis of compassion in the NHS. If there is such a crisis it is not something that can be easily measured and benchmarked; nor something that can be shown on a percentile diagram; still less something that can be corrected by making promises staff do not have the resources to keep. Perhaps these strategies will improve the care of the elderly in the NHS - I hope they do. If they do not, compassion, a value long associated with the health professions, will continue to be observed more in the wording of mission and value statements than in the care of the elderly, and not only in the NHS.

References