Disentangling The Web

Information is power, and the popular notion is that we are entering the Information era (an epoch following the Nuclear era). But, it has always been so, from the traditional operations of the spymaster to current manipulations of the spinmeister: those with the real facts have the advantage over those without. Information is also a very saleable commodity, no less so in healthcare than in the rest of the world. A byproduct of this is that scientific information is starting to be regarded as a proprietary asset belonging to individuals rather than being in the public domain. Unfortunate examples have been the instances of inopportune suppression of negative data by some owners in the insurance and pharmaceutical industries.^1 –3 This situation is particularly regrettable because much of the research takes place in institutions either directly or indirectly funded with taxpayer dollars.

Healthcare is a very big item with enormous financial implications to those who manufacture automobiles, purchase health insurance for employees, or actually deliver healthcare itself. The United States does not have a unified healthcare system, but a hodgepodge of systems that leave about one sixth of the population uncovered. Most of the rest of us are in a state of anxiety because of the uncertainty of possible exposure to financial ruin in the event of serious illness. The safety net of Medicaid is being threatened by the limitation of benefits, as in the new Florida Plan, which gives patients a meaningless power to make limited choices, which are unlikely to be fully understood, in a system designed to limit expenditure. This is not limited to Medicaid. Most of us have heard patients and their physicians bemoaning the time and effort spent in convincing the managers of health insurance plans of the necessity of some or other treatment to maintain a sick person’s life or return to well-being. Getting representative data on patient care in such a disorderly structure is very laborious, as there is no central authority to create a databank that would contain all the logistical and outcome data for the country. It is very frustrating to try to find out how services are being delivered, what is available and what is needed, and who is available to provide what is needed. Good data are therefore very precious and very necessary to make plans that will really maximize the appropriate use of resources for the benefit of the community.

One of the benefits of a consolidated leadership in a comprehensive nationwide scheme for delivering health services is the ability to both collect meaningful data and use it to improve the outcomes of and benefit to those in need of healthcare. It also gives the opportunity for the costs of delivering services to be appropriately managed to get the best value, even though the realists among us know that this often does not happen. A good example of meaningful data is the report published by the Royal College of Obstetricians and Gynaecologists, the triennial Report on Confidential Enquiries into Maternal Deaths in the United Kingdom⁴ (originally England and Wales), which started with the years 1952 to 1954, when the average Englishwoman had a 1 in 1500 risk of dying directly as a result of pregnancy or childbirth. In 2002, the risk had decreased to 1 in 19 020—a 12-fold improvement. Early on, preventable sources of death were identified (eg, poor anesthetic management) and corrected. Now thromboembolic complications are a major source of concern, and they too are being addressed. These reports are the most powerful type of information because the data are nominal. There is no need to resort to statistical probability because the entire population is in the databank. The information gives an unarguable guide as to what was needed to improve obstetrical outcomes.

When it comes to finding out about the organization of any aspect of healthcare, the predominant investigative tool is the questionnaire. It is therefore not surprising that most of those in charge of anything considered significant are bombarded with survey sheets from a large variety of investigators. The forms that get filled out first and fullest are those from internal and external administrative bodies that control institutional survival or continued employment. The redundancy is depressing. Too often a departmental chair lets out an audible groan as 5 different authorities on 5 differently formatted forms request the same information. It appears that the fragmented system we live in has multiple unlinked databases, demanding the same information and all being kept secret from one another. Because of this Byzantine arrangement, the questionnaires have become a disheartening imposition, and people have lost the inclination to fill them out unless they are paid to do so or those surveyed have very direct interest in the results. In the clinical arena, the latter is very evident, with some of the European surveys on sepsis in intensive care units (ICUs),⁵ ethics,^6,7 and pain relief⁸ enjoying a high percentage of returns. This is so different from the United States, where investigators have to struggle to get a reasonable return from a data-gathering request. There is a perpetual worry as to whether or not the information is representative or generalizable to a larger population. The techniques of comparing studies by some form of meta-analytical approach then become key to getting more value from the data. With meta-analysis, the sum becomes greater than the parts. Adjustments can be tailored to strengthen a set of data that could be unrepresentative when taken in its original form. This realistic approach makes for a greater validity in the findings reported by Kirchhoff and Dahl⁹ from a national survey of critical care facilities and units, developed by the American Association of Critical-Care Nurses, published in this issue of the journal.

The designers of the survey are to be particularly commended for their skill in making the data as representative as possible, so that much of it can be generalized to a wider range of facilities providing critical care services. They have provided a report card on much of what we have and what we do, which should give us all a very worthwhile insight for future planning and needs. There are certain questions that beg to be asked and should provide scope for further enquiry.

Whenever the question arises about the number of ICU beds needed (the United States has most ICU beds per capita of any country), there always seems to be an outcry for more because of a lack of availability. The survey data show that 45% of units are seeing an increasing problem with discharging patients who no longer require unit services and are in need of long-term care or some form of progressive care, and at the same time progressive care units are taking on a greater load. Although most units surveyed had palliative care services, only a quarter had formal end-of-life protocols for care of dying patients, but comfort care measures were certainly in place. Here is an area that might be usefully investigated with a view to lightening the workload in ICUs. Do palliative care services need to be used earlier and on a greater number of patients? It is also obvious that if discharging patients from ICUs is difficult, administrative policies to maximize bed utilization may be unsound if patients are kept at a more rigorous and expensive level of care than they need.

Staffing is another key issue. Some of the already available data on the level of staffing^5,10,11 indicate that the numbers of staff required for optimal outcomes for patients are essentially unreachable in our current environment of resource allocation and style of critical care management. It is therefore cheering to find that the majority of respondents to the survey felt that they managed to have adequate staffing most of the time, although some manipulation that might reduce the caliber of the nurses had to be done when shortages arose. By and large, staffing is planned on the basis of patient load. It also seems to takes a long time to fill a vacancy—2 months for an ICU nurse—which is undoubtedly outfall from the current overall shortage. Orientation programs may take another 2 months or longer. The data show that only 42% of units have clinical nurse specialists (CNSs). With this information, it would be interesting to find out if increasing the involvement of CNSs in the ICU could bring nurses fully on-line sooner and increase professional satisfaction to cut down on turnover.

Hourly payment statistics are difficult to evaluate, as are the increments for 10 years of experience or higher qualification. We are expecting increasingly higher levels of academic preparation for nurses, which obviously have to be recognized in the paycheck. This is an area that has to be tracked, as market forces outside the purview of organized nursing may ultimately dictate what salaries will be. So far, it appears that about two thirds of the facilities surveyed did not have a collective bargaining unit representing the nursing staff, and 10% were fully unionized with a closed-shop policy in place. The effects of these data must be correlated with salaries and benefits. (Educational benefits are reasonably widespread.)

One of the findings of the survey we find disappointing is that the visitation policies of 44% of intensive care units have remained restrictive, despite dozens of studies in the past 2 decades documenting the benefits of policies supporting open visiting. Nonetheless, the amount of open visiting is considerable, particularly in the larger institutions and progressive care areas.

In this climate of uncertainty, it is particularly laudable when organizations such as the American Association of Critical-Care Nurses invest resources to acquire and put healthcare information where it properly belongs—in the open. (A full report of the survey is available from the American Association of Critical-Care Nurses.)