RICERCA ITALIANA     

Sleep-debt effects on procedural learning and on clinical and cognitive performance in resident physicians: the protective function of naps

Università degli Studi di Bologna

Abstract

The project will examine two of the most crucial worldwide problems of the new programs of postgraduate medical education: the estimate of the consequences of sleep debt due to working overnight on clinical activities and learning processes, and the evaluation of the efficacy of different nap strategies in counteracting the effects of sleep loss and fatigue in resident physicians. There are estimated to be about 5,000 resident physicians per year in Italy.
Decreases in vigilance during long periods of sustained wakefulness are a serious problem for activities in industrial and transport areas as well as in public health. Both field surveys and laboratory studies have shown that alterations of the sleep-wake cycle almost inevitably leads to an increase of “human errors” as well as to a deterioration of cognitive and professional performance and of learning processes.
Working overnight is a prominent feature of medical residency as part of a comprehensive 24-hour medical service to patients, and is included in educational programs of post-graduate training for young physicians (residents). This training also requires an involvement in on-call duties and urgencies.
Several studies have emphasized the effects of sleepiness on medical errors of commission and, to a lesser extent, omission. These errors are consequent to decrements in vigilance and in psychomotor abilities as well as in executive functions and cognitive flexibility (needed to correct errors and revise plans). Executive functions have shown to depend on the functional integrity of the prefrontal cortex (PFC), which is quite severely impaired after 24-h of waking. Moreover, there is laboratory evidence that one night of sleep deprivation impairs the consolidation of recently acquired perceptual, motor and spatial skills.
The indications so far obtained as to the consequences of sleep debt on various types of cognitive processes cannot be generalized, as shown by field studies using non-homogeneous procedures of sampling, testing of performance and outcomes of the learning process, evaluation of sleep debt and measurement of sleepiness and stress. Moreover, several aspects of the strategy suggested to counteract sleep debt still remain uncertain, such as the optimal number, duration and circadian location of one or more naps.
The present project attempts to gather pertinent and consistent evidence regarding both these issues. Indeed, the 5 Research Units have planned their study adopting the same experimental design, so that homogeneous and large samples (200 or 400 participants per Unit, for a total of 1200 participants) of resident physicians will be assessed at the same timetable on distinct tasks. The data obtained will allow evaluating a) the effectiveness of three nap strategies (namely, one nap located either during the working night or in the following afternoon and two naps both at night and in the afternoon) in counteracting the negative influence of sleep debt consequent to working overnight and b) the effects of sleep debt on: b1) the learning process of perceptual and motor skills (learnt before or after the working night) and psychomotor vigilance and reasoning abilities (Unit of Bologna), b2) cognitive flexibility and executive control (as measured by a task-switch test: Unit of L’Aquila), b3) the attention and inhibition of a dominant response (as a further part of the executive functions: Unit of Trieste), b4) the flexibility in adopting strategies for problem solving and in inhibiting mechanical responses (Unit of Naples), and b5) clinical reasoning skills (as assessed using simulated clinical tasks) and clinical performance (as assessed by the interns’ tutor: Unit of Pisa).
A large spectrum of high-level executive functions (Units of L’Aquila, Trieste and Naples) will be assessed, as well as of learning process (Bologna) and vigilance and performance during abstract and clinical reasoning (Units of Bologna and Pisa). Statistical analyses, with comparison of the scores obtained during the various tasks in different moments of the experimental routine (the days before and after the working night, the day after a recovery night and after a week), will allow us to establish a) which specific function and learning process is negatively influenced by sleep debt in a transitory or persistent manner, and how; b) how effective each of the three nap strategies is in counteracting the negative influence of sleep debt; c) whether the effects of sleep debt and/or of nap strategy vary as a function of gender and clinical experience (4th year residents as opposed to 2nd years); d) whether the prolonged partial deprivation of sleep (more likely in 4th year than in 2nd year residents) exerts an additional negative influence.
The expected findings may provide useful indications as to how to schedule the periods of night-work of resident physicians to reduce the negative influence of sleep debt on performance and learning. <<<

Principal Investigator

Carlo Cipolli Università degli Studi di BOLOGNA

Research Objectives

The project will examine two of the most crucial worldwide problems of the new programs of postgraduate medical education, namely the estimate of the consequences of sleep debt due to working overnight on the clinical activities and learning processes, and the evaluation of the efficacy of different nap strategies in counteracting the effects of sleep loss and fatigue in resident physicians. Their number can be estimated in about 5,000 per year in Italy. The formal procedure of involvement of resident physicians in the clinical activities of all divisions of the hospitals included in the network of each postgraduate program has been recently approved and will be applied in the present academic year (see D.P.C. M. 6/7/2007). Consequently, this will lead to an increase of the risk of an heavy exposure to working night as a routine and, thus, of the interest in better designing the work-schedule within clinical division and in suggesting strategies to better counteract the negative effects of sleep debt.

The present project attempts at gathering pertinent and consistent evidence on both these issues by examining a large sample (more than 1200) of resident physicians. The 5 Research Units will apply the same experimental design in order to assess resident physicians at the same hours and with the same timing on distinct tasks. The main aim is to gather coherent indications about the deterioration, as a consequence of sleep debt induced by working overnight, of professional (as assessed on simulated clinical tasks) and cognitive performance (measured on a wide spectrum of high-level executive functions), as well as of the learning process of new perceptual and motor skills. A further aim will be to assess the efficacy of different napping strategies in counteracting the effects of sleep debt and fatigue on the above mentioned abilities.

In particular, the Unit of Bologna will ascertain: a) how much sleep deprivation due to night-work impairs psychomotor vigilance and reasoning ability compared with the condition of normal sleep (i.e., the performance of the control groups) and how effective is such countermeasure as one- or two-nap strategy; b) how much sleep deprivation impairs, in a transitory and persistent (perhaps permanent) manner the consolidation of procedural (perceptual and motor) skills acquired before and after sleep deprivation compared with the control condition.

The Unit of L’Aquila will ascertain: a) whether sleep loss during night shifts selectively affects the control processes involved in task switching and how effective is a one- or two-naps strategy is in counteracting the influence of sleep loss on such cognitive ability; and b) which specific component (endogenous or decay) of task switching is the most negatively affected by sleep deprivation and how it is sensitive to the combined influence of sleep deprivation and fatigue due to working during the night.

The Unit of Trieste will ascertain: (a) whether sleep deprivation impacts in a transitory or long-lasting way performance and familiarization to unknown tasks measuring the efficiency of attentional networks and executive functions; (b) how effective are distinct napping strategies against the negative effects of sleep loss taking into account not only the presence/absence of one or two naps but also their positioning throughout the day.

The Unit of Naples will establish, by detecting perseverations or abstract thinking deficits: a) whether and to what extent sleep debt impairs the flexibility of choice in “problem solving” processes (which are deeply involved in clinical duties, overall in emergency division) and the abilities to monitor on a moment-by-moment basis the execution of repetitive task and the decision of maintaining/changing strategy; b) how nap strategies are effective in reducing this negative influence of sleep debt.

The Unit of Pisa will establish: a) how sleep deprivation and clinical performance are related by using objective and subjective evaluations of fatigue and sleepiness and objective and subjective indices of clinical performance (by measuring diagnostic and procedural performance in simulated clinical tasks as well as evaluating the quality perception of the clinical practice with an ad-hoc questionnaire administered both to residents and supervisors); and b) how nap strategies are effective in reducing the negative influence of sleep debt on clinical reasoning, decision making and performance.

All the Units will apply the same experimental design on very homogeneously selected and balanced (for age and clinical practice, in term of years of residency program) samples of resident physicians. They will also adopt the same procedures to monitor sleep/wake patterns (using actigraphic recordings) during the day 1-3 of the experiment, to evaluate the subjective perception of sleepiness, anxiety and stress, and to describe the sleep habits of the participants (by means of a sleep diary). Therefore, it will be also established: a) whether there are gender-related and/or practice-related variations in the size of the sleep-debt effect in one or more measures of clinical and cognitive performance measured by the single Units; b) whether there are gender-related and/or practice-related variations in the effectiveness of the nap strategies to counteract the sleep debt in one or more measures of clinical and cognitive performance; c) whether there are additional positive (depending on the adaptive habit) or negative (depending on prolonged/cumulative partial sleep debt) effects consequent to periodical night shift (in particular in the 4-year residents); d) how much and for which clinical and cognitive abilities and learning processes the self-evaluations of sleepiness, fatigue, anxiety and stress are predictive of performance deterioration.

These items of evidence will provide reliable indications about the more frequent types of performance deterioration for male and female residents, will help the choice of the most appropriate napping strategy and will give hints about the effectiveness of adaptive habits to sleep restriction. These indications may be useful in scheduling the periods of night-work of resident physicians in order to reduce the negative influence of sleep debt on performance and learning.

Taken as a whole, the present project has several innovative characteristics, aiming to provide for the first time convergent indications about the actual effects of working the night shift on several practical and cognitive abilities of a very large sample of resident physicians engaged in the most advanced program of clinical education, as well as about the effectiveness of different nap strategies in counteracting such negative effects. A further strength of the project comes from the complementary competences of the researchers of the five Units, that will allow to explore all the most important abilities supposedly affected by sleep loss. <<<

First Results

The project will lead to collect a great amount of data, a large part of which of potential interest to deepen our knowledge of the temporary and delayed (and perhaps permanent) effects of sleep deprivation on learning of new skills by resident physicians (reducing, in this way, the scope of the postgraduate programs). Moreover, the data will allow to establish for the first time the effectiveness of three nap strategies in counteracting the negative influence of sleep deprivation and, thus, providing important indications as to the basic questions of the protective function of naps, their optimal duration and circadian location.

Each Unit will provide potentially important findings, and more information will come from the integration of the findings of the 5 Units. In general terms, the analysis of the subjective and objective data of sleepiness, fatigue and stress collected by all Units of the project may provide further insight into the relationship between sleep deprivation (in controlled situations) and clinical performance. In particular, they will be useful to estimate the prediction power of subjective measures of sleepiness and of deprivation- and fatigue-related stress with respect to the deterioration of cognitive and clinical performance and of learning processes.
Given that the same experimental design is shared by all the research units and the homogeneity of the samples, it seems legitimate to argue from the data collected by distinct Units a coherent picture of the effects of sleep deprivation on clinical and cognitive performance and learning processes and of the antagonistic power of nap strategies.
In this way some reliable and generalized indications as to the optimal work schedules for low- and high-skilled residents may be provided to the Agencies responsible of planning and monitoring the formative system of Italian residencies.

The main results expected from the single Units are the following.

Unit of Bologna
In keeping with the general hypothesis that sleep has an important role both in maintaining effectiveness in the performance of psychomotor vigilance and reasoning abilities, and in consolidating previously acquired (consolidation phase) and newly acquired procedural skills (acquisition phase), four main results are expected. Sleep deprivation due to night-work should impair:
1) psychomotor vigilance and reasoning abilities compared with the control condition (i.e., the performance of the control group). The impairment should be higher for sleep-deprived no-nap subgroups, compared with the two subgroups taking one nap and, above all, compared to the two subgroups taking two naps. The performance impairment should be transitory, performance scores being expected to reach baseline value after one recovery night and especially after 7 recovery nights;
2) the consolidation of procedural skills acquired BEFORE sleep deprivation compared with the control condition. The impairment should be lower in patients taking a night-time nap than those taking an afternoon nap and even lower in patients taking both night-time and daytime naps. The impairment should be permanent, the scores of Texture Discrimination Task (for perceptual skills) and Finger Tapping Task being expected to remain worse than those of participants in the control condition (i.e., without sleep deprivation) after one as well as after 7 recovery nights;
3) the consolidation of procedural skills acquired AFTER sleep deprivation compared with the control condition. The impairment should be lower for patients taking a night-time nap than those taking an afternoon nap and even lower in patients taking both night-time and daytime naps. The impairment should be permanent, the scores of Texture Discrimination Task (for perceptual skills) and Finger Tapping Task being expected to remain worse than those of participants in the control condition (i.e., without sleep deprivation) after one as well as after 7 recovery nights;
4) the learning process (and perhaps performance requiring psychomotor vigilance and reasoning ability) more severely in the participants (above all of the 4th year) with than in those without prolonged partial deprivation of sleep (as assessed by sleep diary) and in the participants without than in those with adaptive sleep habits (again, as assessed at sleep interview and sleep diaries). Both performance deterioration and learning impairment should be less severe after fairly long (about 30 min) than short naps (10 or less min), as retrospectively assessed by actigraphic data.

Unit of L’Aquila
The first three experiments carried out by the present Unit have the main goal of increasing our knowledge about specific frontal lobe functions across the 24 hours and in situations of decreased vigilance. More specifically:
a) The first two pilot experiments are expected to define the sensitivity of the task switching paradigm to the influence of sleep debt in a laboratory setting. This step is particularly important for the subsequent choice of a task switching paradigm which is highly sensitive to one night of total sleep deprivation. Then the most appropriate task configuration will be used in the “on the field” study.
If sleep deprivation exerts its negative effects on the endogenous component of task set reconfiguration, then in the first pilot experiment an increase of switch costs after the sleep deprivation session is to be expected. Otherwise, if the sleep debt negatively impacts the decay component of the task switching performance, then the switch costs will increase after the deprivation session of the second pilot experiment. We cannot exclude that both the task switching components are sensitive to the experimental manipulation. Moreover, the comparison of the effects of the pilot experiments (carried out in a typical laboratory controlled inactive condition) with those of the “on the field” study, will allow us to disentangle the simple effects of sleep loss per se from the cumulative effects of sleep debt and operational fatigue due to working overnight.
b) The “time of day” experiment will establish whether the executive functions involved in task switching show some circadian and /or ultradian covariation with subjective and objective sleepiness. Whether executive processes are affected by circadian and/or ultradian variations in physiological functioning, then switch costs will vary along the daytime showing, for example, a dip during the postprandial hours.
Finally, the main study “on the field” will allow to evaluating the sensitivity of task switching performance in showing the interaction effects of sleep debt and fatigue in real life medical conditions. Thus, this experiment has a main applicative interest, allowing to extend the basic knowledge about the fontal lobes executive functions to a real work setting, and specifically to the medical decision making in situations of sleep loss and fatigue.

Unit of Trieste
The findings of the study are expected to asscertain:
(a) how much sleep deprivation impacts on performance and familiarization to unknown tasks which measures the efficiency of attentional networks and executive functions;
(b) the possible delayed and long-lasting effects (after both one recovery night and a week characterized by a normal sleep-wake rhythm);
(c) the (different) efficacy of napping strategies against the negative effects of sleep loss taking into account not only the presence/ absence of one or two naps but also their circadian location;
(d) the effect of gender and prolonged exposure to sleep deprivation.
Moreover, the results of the study should allow establishing whether sleep loss entails a generalized impairment of attentional processes (as suggested by experimental studies using almost exclusively vigilance tests) or a selective impairment of specific components.
Possible operative indications of the expected results could regard the reconfiguration of the work-schedule within clinical divisions and the formulation of guidelines and recommendations as to the strategies to better counteract the negative effects of sleep loss in residency.

Unit of Naples
According to a fairly consistent literature about the effects of sleep loss on metabolic and neurophysiological level in the Central Nervous System, we expect that superior executive functions in our sample will be deeply impaired during long work shifts. Therefore, changes in performance at more complex tasks should result greater than those concerning simpler neurocognitive tasks, as these do not imply monitoring and adaptation strategies to relevant variations in the environmental requests.
It is also expected that one or two naps of appropriate duration (between 20 and 45 minutes), just as those allowed to some subgroups of the sample assessed, can partly counteract the sleep-debt reduction of efficiency of central executive functions, both for the acquisition of new skills after sleep deprivation and for the maintenance, despite fatigue and sleep loss consequences, of what has been previously consolidated.
Possible applicative fall-out of this study is a data-driven rearrangement of timetables and schedules of shift working amongst junior residents, who will gradually be more and more in charge of delicate tasks and required to take great responsibilities while on night-shift and on-call duty.

Unit of Pisa
The data collected by the Unit are expected:
1) to confirm in a controlled experimental setting (albeit naturalistic as a nocturnal duty in hospital is), the relationship between sleep deprivation and clinical performance (including a comparison between subjective experience of clinical performances and its objective evaluation);
2) to assess the combined effect of the level (low/high) of professional skills and night-duty experience on clinical performance. This kind of indications may be useful both in scheduling nocturnal duties inside the residency and in organizing the whole formative track;
3) to evaluate the effects of naps as a recovery strategy for clinical skills and the underlying reasoning and decisional abilities. Also this kind of information may be useful in scheduling night-shift not only for residents but also for specialists.
Additionally, the symmetrical trends of data on medical errors and learning and memory skills, reasoning abilities and executive functions may provide useful suggestions as to improve the formative track of the Italian residencies in reducing errors and improving learning and, thus, overcoming the frequently observed difficulty of learning from errors when sleep deprived.

Finally, as all the Units will share a common and homogeneous population for their study and will measure common variables (as stated above) it will be possible to explore simple relationship among data collected by the different groups. In detail it will be evaluated if clinical reasoning has the same pattern of efficiency that problem solving has or if it tends to be impaired after sleep deprivation and restored after recovery, being perhaps modulated by the level of clinical experience. Moreover as one of the most important aim of residency is “teaching” to junior physicians it will be also useful to assess the pattern of attenional and learning skills and their chronological relation with clinical practice skills. More generally we will be able to ascertain if and how the different cognitive functions took in account in the project (all claimed to be essential during the training and learning period of the residency) are affected by shift-work, fatigue and experience and to identify specific patterns of impairment as well as specific chronological correlation with clinical performances. <<<

Timescale

24 months

National and international background

It has been estimated that about one sixth of the workforce in western countries operates outside standard daytime hours, with a large proportion of night-time work hours (Boivin et al., 2007). The social and economic effects of working within atypical schedules and with sleep curtailment have been long outlined, as leading to increased risk of accidents, impaired performance and negative effects on health, well-being and life satisfaction (Rosekind, 2005). Shift work, in particular, has also shown to be associated with an increased perception of mental and physical fatigue and, not rarely, with sleep disorders, such as excessive sleepiness and/or sleep disruption (Am Acad Sleep Medicine, 2005, 2nd Ed, 2005).
In particular, decreases in vigilance during sustained wakefulness are a serious problem for applied areas such as public health and medical care. Working overnight is a prominent feature of medical residency as part of a comprehensive 24-hour medical service to patients and is a prominent feature of educational programs of post-graduate training for young doctors (interns and residents). This training also consists of a progressive involvement in on-call duties and urgencies in the teaching hospitals. In these clinical activities time pressure is usually high, also because fatigue is associated with reduced alertness as a consequence of sleep debt and with a high level of stress in taking decisions regarding newly acquired or unfamiliar diagnostic and therapeutic manoeuvres.
The risk of errors (both of commission and omission) and mistakes as a consequence of sleep debt in resident physicians is well known, having been firstly pointed out by Friedman and colleagues (1971), who reported that interns made almost twice as many errors reading electrocardiograms after an extended (24 hours or more) work shift than after a night of sleep. Recent field surveys have confirmed that interns and residents working extended work shifts and on-call periods have significantly more attentional failures, induce more preventable adverse events, and make significantly more serious errors (Grantcharov et al., 2001; Eastridge et al., 2003; Lockley et al., 2004; Landrigan et al., 2004; Barger et al., 2006).
The deleterious effects of sleep deprivation on procedural and diagnostic tasks are reliably attributable to sleep debt consequent to alterations of the circadian rhythms in wake/rest cycle (Philibert, 2005). Indeed, the frequency of errors made by residents decreases after a period of recovery of sleep debt (Lockley et al., 2004; Landrigan et al., 2004). The attribution of these negative effects to the acute sleep deprivation inherent to on-call and night-shift activities of residents, who are engaged in often unexpected and complex clinical tasks, is supported by the findings of studies. There is consistent evidence that even one night of sleep deprivation does indeed lead to significant deterioration in complex skills, such as anticipating consequences, keeping track of events, being innovative, and avoiding distractions and irrelevant stimuli. Unlike rule-based, convergent and logical skills, the above-listed skills depend heavily on the functional integrity of the prefrontal cortex (PFC), the cortical region that appears to be especially vulnerable to the effects of prolonged wakefulness (Ferrara et al., 2002; Finelli et al., 2001; Maquet, 2000). Several recent studies have indicated that the impaired functioning of PFC is associated with a reduced effectiveness in high-level and complex skills, and in particular, the so-called executive functions. These studies have ruled out the view, which was long prevailing in the research area of the effects of sleep deprivation, that high-level complex skills are relatively unaffected by SD because of the interest they generate and because of the implicit encouragement for participants to apply compensatory effort to overcome their sleepiness. Indeed, recent laboratory studies have shown that personal motivation to performance is not sufficient to secure, after sleep loss, an elevated standard of efficiency in a wide variety of high level cognitive abilities (Jones &amp; Harrison, 2001). These abilities largely rest upon the functional integrity of the prefrontal cortex, and it has been shown that its metabolic activity can be seriously reduced while the subject is performing a wide variety of tasks, after being awake for 24 hrs. (Muzur et al., 2002).
The consistent indications of these studies prompt us to focus both on the mechanisms through which sleep deprivation interferes with cognitive processing, and the high-level executive functions (or processes) which are actually impaired by sleep debt. To this aim, a comprehensive theoretical framework within which cognitive performance decrements can be interpreted is required. Although it is usually maintained that performance decrements after sleep loss are mainly provoked by generic attentional deficits (Lamond et al., 2007; Quigley et al., 2000), attention is a multidimensional cognitive ability (Demeyere &amp; Humphreys, 2007; Fan et al., 2005; Mirsky et al., 1991). Hence, it is not enough to establish a generic relationship between sleep loss and attention; it is necessary to specify whether, following sleep deprivation/reduction, there is a general impairment of attention processes or a selective one (Fimm et al., 2006; Doran et al., 2001; Jennings et al., 2003; Vgontzas et al., 2004). Accordingly, the so-called “executive functions” should not be considered a unitary generic ability, but should be differentiated. The term “executive functions”, used correctly, should refer to those functions which operate when routine actions are inadequate for executing a task (Posner &amp; Di Girolamo, 1998). Complex executive functions, including executive attention, working memory and divergent thinking, involve the abilities to plan and coordinate wilful action in the face of alternatives, to monitor and update actions as necessary, and to suppress distracting material by focusing attention on relevant information (a process called “inhibition”) (Sagaspe et al.,2006). Thus, it is apparent that, amongst all the categories of professional workers, “on call doctors” have a special need to preserve these abilities. Beyond vigilance and the simplest operative skills, complex cognitive functions allow them to cope with responsibilities as well as with difficult, and frequently unfamiliar, tasks, through “ decision making” and “problem solving” processes. This seems to be the case especially for those junior doctors who are less experienced in their daily working routines and who run a high risk of being affected by reduced efficiency of executive functions. The few experimental indications so far obtained are consistent in showing a negative impact of sleep debt on complex neurocognitive functions (Gottselig et al., 2006). Also, neurophysiological and functional metabolism studies have suggested a particular sensitivity to sleep loss of the prefrontal cortex, i.e. of those neural circuits which regulate superior cognitive functions (Durmer e Dinges, 2005).

A wide spectrum of the cognitive processes included in the domain of the high-level executive function is at the basis of the present project. This project attempts not only to assess the influence of sleep deprivation on the functioning of several distinct high-level executive functions (Units of L’Aquila, Napoli and Trieste) and on clinical performance (Unit of Pisa) and reasoning and psychomotor skills (Unit of Bologna), as well as on the learning of new skills acquired before or after sleep deprivation (again, Unit of Bologna), but also on the effectiveness of 3 nap (i.e., short period of sleep) strategies as a countermeasure of the negative influence of sleep deprivation. These two main aims are combined in the present project because sleep debt consequent to working overnight has so consistently shown to be detrimental for many clinical activities that not only the period of on-call duty has been reduced by law in USA (2001-3), but also scientific societies of sleep research have recommended the use of nap strategies to reduce the detrimental effects of sleep debt in resident physicians (Arora et al., 2006; Baldwin et al., 2003, 2004).
However, while several field studies have proved fairly consistently that naps are effective in relieving fatigue, sleepiness and the decrease in residents’ performance (Howard et al., 2002), it remains uncertain whether naps can prevent or repair another parallel negative effect of sleep loss, namely that of an impairment in the learning process of new skills and clinical manoeuvres. This additional risk of sleep debt is not negligible, given that laboratory studies have provided consistent evidence that sleep loss impairs the learning process of perceptual, motor and spatial skills when the acquisition phase is followed by a 24-h or longer period of sleep deprivation (Stickgold et al., 2000; Walker et al., 2002; Fisher et al., 2002; Ferrara et al., 2006). Moreover, this impairment is not contingent (i.e., lasting for the duration of the day after sleep deprivation), but persists (at least) over some days and may be permanent. Finally, given that the acquisition of new skills is less efficient after a period of sleep deprivation, the learning process of new skills is also influenced in a negative way by sleep deprivation, not only in a contingent manner (i.e., after some hours or one day), but also plausibly in a persistent (for example, after a week from acquisition) or permanent manner (as suggested by data on perceptual learning: Stickgold et al., 2000; Walker et al., 2003).

There is increasing agreement that systematic investigations regarding the possibly different effect of sleep loss on refined skills involved in clinical activities, and the actual protective function of nap strategies, are urgently required. On one hand, in clinical decision making, so-called task switching, i.e. the quick and flexible adaptation of behavior to the changing characteristics and requests of the environment, is crucial. Neuroimaging studies demonstrated that task switching performance recruits various PFC regions. However, the effects of sleep deprivation and of circadian factors on this executive function have never been investigated, notwithstanding the fact that the task switching paradigm has been widely used to investigate the executive control of cognition.
On the other hand, several indications provided by the studies so far carried out cannot be generalized, because of methodological limitations. In particular, the effects of sleep deprivation on clinical and cognitive performance have not been assessed with respect to a) gender (although it is known that in shift-workers the altered schedule of sleep/wake is more detrimental for women’s performance rather than men’s, and furthermore women make up a large proportion of residents); b) the adoption of an adaptive strategy in residents with a longer experience of working overnight; c) the possible concomitant presence of a long partial sleep deprivation. After a week of sleep restriction (less than 5 hours of sleep per day) several components of clinical performance (such as executive function, reasoning abilities and psychomotor vigilance: Dinges et al, 1997) have shown to be impaired in a more or less substantial manner. Finally, it has not as yet been considered that clinical medicine is fraught with ill-defined problems that clinicians solve in a variety of ways, depending on their specific knowledge in an area and their experience (Grant et al., 1988), moving through 3 kinds of mental representations, from basic mechanisms of disease, to illness scripts, to a bank of cases derived from experience (Schmidt et al.,1990). How much the differences in clinical experience and, arguably, in clinical reasoning strategies are capable of reducing the effects of sleep deprivation on medical performance (and, thus, on the quantity and type of errors) has not been addressed to date, even though it may have a huge importance in the organization of the postgraduate program of resident physicians.

For these two types of reasons systematic studies where all the above factors are taken into account appear to be urgent, also in order to establish the effectiveness of countermeasures (such as naps or drugs) to reduce the acute and possibly chronic negative effects of sleepiness during residency training. While several studies have reported that eliminating extended work shifts can reduce serious errors (Lockley et al., 2004; Landrigan et al., 2004), little is known regarding the effects of napping on clinical performance. Indeed, while a number of experimental studies have focused on the potential beneficial effects that a short nap could exercise on a subject’s performance (Monk et al., 2001; Tietzel &amp; Lack, 2001), the results reported refer almost invariably to changes in sleepiness level. And even when specific cognitive abilities are considered (Tietzel &amp; Lack, 2002; Gillberg et al., 1996), most of the time researchers have chosen to use simple monotonous tasks which at best are able to detect coarse variations in the subject’s vigilance/alertness levels, but cannot be sensitive to possible positive effects on high level cognitive processes. In addition to the small number of studies in the area of the relationship between napping and clinical performance, the available findings regarding the effectiveness of napping on clinical performance are often discrepant (Driskell et al., 2005). This is a non-secondary consequence of the fact that research on the above topic has been developed in laboratory settings rather than in natural care environments. <<<