Marga Hoogendoorn

99 influence on how satisfied nurses are about the workload. Comparing the mean NASA- TLX in our study (NASA-TLX 43.3) with the results of other studies, the overall perceived workload with the NASA-TLX was relatively low. A study of Hoonakker et al (2011) in 757 ICU nurses in 7 hospitals and 17 different ICUs showed a mean overall workload of 70.4 27 . Those ICUs included however also workload of nurses on a burn-unit, pediatric or neonatal unit. But also other studies showed NASA-TLX scores between 59.95 and 70.2 4,5 . A possible explanation for our lower NASA-TLX score is the high number of postoperative patients in our study group (32%); the workload of a planned postoperative ICU patient is relatively low compared to unplanned surgical or medical patients 11 . The low nursing workload and the negative impact of this workload on nurses is also confirmed in different remarks we found in the free text comments. Nine out of the thirteen nurses left a comment about a quiet shift stating: ‘very quiet’ or ‘too quiet’, ‘little or no challenge’, ‘workload too low to be pleasant’, ‘sometimes it feels more pleasant to have a busier shift’. Only one nurse left a comment about a busy (restless) shift with too few nurses for the work to be done. The results of the satisfaction about workload of the nine nurses with comments due to a quiet shift showed a wide range in the satisfaction rate (2 till 10). this shows that there is dissatisfaction with the workload at both a very high and very low workload. These qualitative results seem to support that there is an optimal point in the nursing workload. This optimal point is important because of the impact of nursing workload on job satisfaction, burn-out or intention to leave 12,13 . Planning the nursing staff should not be based on the number of patients per nurse, but on both the objective and perceived nursing workload. Strength and limitations To the best of our knowledge this study is the first one that assessed the relationship between workload satisfaction and both the objective as well as the perceived nursing workload. Many studies described nursing workload or job satisfaction, but none of those studies analyzed the association between these concepts. Therefore, this study contributes to a better understanding of nursing workload and how to use the concept of workload as a human resource tool. A strength of this study is the completeness of data. During this study both the Nursing Activities Score and the NASA-TLX was filled in by almost all the nurses. We included data in a period of 13 months and therefore the data are representative for workload all over the year. Because we asked the nurse to fill in the questionnaire at the end of the shift and after the handover the study itself did not affect the measured nursing workload. Some limitations of our study need to be taken into consideration. Whereas workload satisfaction is a very complex concept we used a simple one-dimensional question for