In March of 2002, Secretary of Health and Human Services (HHS) Tommy Thompson transmitted to Congress the results of his Department’s four-year investigation into the staffing needs of nursing homes, entitled “Appropriateness of Nurse Staffing Ratios in Nursing Homes.” This landmark study, conducted by some of the nation’s top nursing home experts, documented that most U.S. nursing homes do not have enough staff even to keep nursing home residents physically safe, let alone to provide good care.1 The report’s analysis supported national minimum staffing standards guaranteeing a sliding minimum of between 2.9 and 3.2 hours of nursing assistant care per 24-hour day, varying according to nursing homes’ particular mix of residents with different kinds of needs.

Since the current national average for all nursing homes is just 2.3 hours of nursing assistant care per day, adopting these recommendations would require increasing nursing assistant staffing by more than twenty percent nationally. And, since Medicare and Medicaid pay for most nursing home care, substantial increases in federal funding for such care would be required. In the current political climate, none of this is likely to happen. Indeed, Secretary Thompson’s letter to Congress accompanying the report actually dismissed its findings and recommendations and called instead for further study.

Given the current politics of nursing home reform, it is important to contribute to the debate by investigating the question of nursing home staffing from many vantage points. Participant observation research I conducted in 2004 suggests that, far from being unrealistically high, the standards proposed by the authors of the HHS report are probably still too low. To understand why, it is necessary, first of all, to understand how the HHS authors arrived at their “optimal” staffing thresholds of 2.9 to 3.2 hours of nursing assistant care per day.

Simulations and Observations
The investigators began with existing time-study data on average completion times for individual care tasks depending on the residents’ varying “acuity” or need. Then they constructed computer simulation models predicting the staffing requirements for nursing facilities with different mixes of residents. These models took into account the distribution of different kinds of nursing facilities, the varying needs of residents, and the proportions of different kinds of residents in the nursing home population.2 The HHS simulation models were also based on a number of conservative assumptions: for example, that aides face few unscheduled care tasks, that they are always able to be “on-task,” and that time spent moving from one resident to another is negligible. Thus, there are already some good reasons to think that if the HHS analysis erred, then it did so in the direction of underestimating staffing requirements for good care.

My participant observation research on nursing home work adds to the list of reasons for viewing the HHS “optimal” staffing thresholds as probably too low. For five months in 2004, I worked part-time as a nursing assistant on the day shift in a well-staffed,3 religiously affiliated, non-profit nursing home. I discovered two things relevant to the conclusions of the HHS study.4

One finding was that the task-completion times the HHS research team used as input data matched my own observations and experiences well. For example, it routinely took experienced aides at my facility about 15 minutes to perform the tasks known as “a.m. care” for the residents who needed the most assistance. This matches well with the figure of 14 minutes that the HHS researchers used as their input data. All the other tasks also matched up well; for example, experienced aides were able to give most residents their biweekly showers in about 15 minutes, the same figure used by HHS.

My second discovery, however, reveals why these input figures are still too low: a.m. care cannot be completed in 15 minutes without jettisoning official procedures. In other words, the average task times used by HHS are realistic but reflect nursing assistants’ informal, timesaving work methods. These methods contravene official care rules designed to protect the safety of both workers and residents and to prevent the spread of infectious disease. The aides I worked with deviated from official procedures, not because they were indifferent to safety or infection control, but because there was no other way to complete their assigned work on time.

Informal Practices
What kinds of informal practices were aides forced to adopt in order to complete their care tasks within time frames comparable to the ones that provide the basis for the HHS report’s staffing thresholds? The list is lengthy. Aides never used the required “gait belts” for assisting patients from bed to chair or the reverse. They ignored rules requiring two aides to perform transfers with mechanical lifts; sometimes workers did not use the lifts at all but simply picked residents up and carried them. Although aides were officially required to perform rounds to check on the condition of all patients at the beginning of a shift, this was unheard of in practice. No one ever followed official procedures for washing residents using two washbasins (one for soapy water, one with clean water) – aides saved precious minutes by not using washbasins at all.

Perineal care in practice never followed detailed rules for folding and refolding washcloths so that each of nine specific wipes would use a clean surface; instead it often meant a single swipe between the resident’s legs with a soapy cloth, before the resident was turned and washed from behind. Rules against leaving residents on the toilet alone were also routinely broken; one could rarely afford to stand there waiting and watching while a resident sat on the toilet. The only way to get done in time was to get the resident up on the toilet, go on to the next resident, and come back five or ten minutes later to finish the job. Residents who could not be counted on to remain seated were simply changed in bed, regardless of nursing orders requiring that they be assisted to the toilet.

The effects on the quality of care (and of work) of these informal solutions to the understaffing problem are obvious. When aides set aside rules for lifting and transferring patients, they increase the risk of falls and injuries to residents (and to themselves). When hurried aides wash their hands less frequently or less thoroughly than they should, they increase the risk that they might transmit infectious diseases – to which the elderly and sick are especially vulnerable – from patient to patient. When they do not perform rounds, they increase the chance that a resident’s suffering will go unnoticed until later in the shift. And of course, the very effort to complete care tasks in extremely compressed periods of time compromises the emotional aspects of care.

Impossible Demands
Sociologists of work have known for decades that workers often adopt informal practices in order to “get the work out” or to make it easier – and that workers in public service jobs are particularly likely to resort to such practices because of chronic under-funding in public services. Despite this knowledge, the debate over nursing home staffing is not particularly well informed by detailed knowledge about what nursing assistants actually do and the conditions under which they do it.

It is impossible to arrive at realistic staffing estimates unless they are built on a solid understanding of the nursing home labor process and its requirements. Without such an understanding, the “realistic” task times used by government agencies only serve to conceal the harm that the prevailing notion of “good” staffing does to nursing home residents and workers alike. Moreover, without such an understanding, the informal practices my research documents might appear to be examples of individual deviance rather than what they are: inevitable responses of the hard-pressed to impossible work demands.

At present, no one knows what levels of nursing assistant staffing are required as a necessary condition for good nursing home care. In a way, Tommy Thompson was right: we do need more research – not to find out whether national nursing home staffing standards are “appropriate,” but rather to learn how much higher staffing should be to enable aides to follow existing care rules. The HHS modeling approach is sensible, but there is an urgent need for new, ethnographically informed time-study research to provide more realistic input data.

NOTES

1. US Department of Health and Human Services, “Appropriateness of Minimum Nurse Staffing Ratios in Nursing Homes, Phase II,” Washington, D.C., Centers for Medicare and Medicaid Services, 2001.

2. The HHS modeling approach is very similar to the time and motion studies pioneered by advocates of scientific management. The logic behind the simulations is as follows: If we know (a) what care tasks need to be performed; (b) how long, on average, it takes to perform these care tasks for different kinds of residents; and (c) the proportion of different kinds of residents in the nursing home population, it is possible to compute the number of nursing assistants needed to care for a given population.

3. The nursing home I studied provided an average of 2.44 hours of nursing assistant care per day for each resident, well above the industry average. This translated into a day-shift ratio of one nursing assistant for every 7.4 residents.

4. A more detailed account of these findings can be found in Steven H. Lopez. 2006. “Culture-Change Management in Long Term Care: A Shop Floor View,” Politics and Society Vol. 34, No. 1, forthcoming.

Steven Henry Lopez is assistant professor of sociology at The Ohio State University.

Labor and Employment Relations Association
Tel: 217-333-0072 | Fax: 217-265-5130 | LERAoffice@uiuc.edu | www.lera.uiuc.edu