Sleep Histories Are Seldom Documented on a General Medical Service.

ABSTRACT

Background. Sleep disorders are common, but the frequency of sleep history documentation in hospitalized patients is unknown.

Methods. We reviewed 442 initial histories and physical examinations recorded by 122 house officers and 47 medical students in 208 consecutive general medicine ward patients.

Results. Any reference to sleep was recorded in only 18 patients (9%), including 12 of 141 (9%) with conditions associated with obstructive sleep apnea. Sleep histories were recorded more often in women (13% vs 4%) and less often than histories of cigarette smoking or alcohol use. Medical students recorded such histories more often than did house officers. Patients with sleep histories more often had pulse oximetry (78% vs 37%), pulmonary function testing (11% vs 1%), arterial blood gas analysis (67% vs 30%), or electrocardiograms (78% vs 49%).

Conclusions. Sleep histories are documented infrequently in hospitalized patients. Patients with a recorded sleep history more often have tests that suggest increased concerns about cardiorespiratory risk and/or a different process of care.

SLEEP DISORDERS affect approximately 12% of the population [1] and have been associated with reduced quality of life, [2] Motor vehicle accidents, [3,4] and other catastrophes. [5] Sleep-disordered breathing, such as obstructive sleep apnea (OSA), has been estimated to affect approximately 2% to 4% of adults [6-8] and has been associated with hypertension, [9-10] cardiovascular disorders, [11-14] stroke, [14] and mortality. [15] Because approximately 80% to 90% of clinically significant OSA cases are not recognized, [16] the recognition and treatment of sleep disorders should be a priority of health professionals. Sleep histories might help physicians and other health professionals to identify potentially treatable conditions, but the frequency with which the sleep history essential to diagnosis is done in an acute care setting is unknown. In ambulatory settings, sleep histories are seldom documented and are addressed considerably less often than histories pertaining to alcohol use and cigarette smoking. [17]

We evaluated the performance of house officers and medical students in addressing the following questions: How often is any sleep history documented in the medical records of patients hospitalized on an acute general medicine ward service? How does the frequency of documented sleep histories compare with that of other health promotion histories (cigarette smoking and alcohol use)? Does the frequency of documented sleep histories relate to the level of health professional education or discernible differences in patient demographic characteristics or approaches to patient evaluation?

METHODS

We reviewed the medical records of all hospitalized patients admitted to internal medicine ward services at Wake Forest University Baptist Medical Center and assessed the performance of house officers and medical students. This 3-year accredited internal medicine residency program and 4-year accredited medical school is affiliated with Wake Forest University School of Medicine and North Carolina Baptist Hospital. The house officers were in their first through third years of training (postgraduate year [PGYJ] 1, PGY 2, and PGY 3) and students were in their third and fourth years of medical school (MSIII, MSIV).

Medical records of all patients hospitalized on four general internal medicine ward services were reviewed on 5 days: October 4, 1997, December 14, 1997, February 18, 1998, September 20, 1998, and October 7, 1998. Each patient was seen by one to three house officers and/or medical students. A total of 441 initial medical histories and physical examinations were recorded in 210 consecutive patients by 2 fellow physicians, 122 resident physicians, and 47 medical students. A total of 435 medical records were included and 6 were excluded. Of the excluded records, 4 represented "History and Physicals" that were included in a previous review date and 2 were done by fellows from subspecialty services. The initial comprehensive admission evaluations are approximately 1-hour assessments, as opposed to brief daily progress notes.

Data collected included patient demographic characteristics, initial complaints, medical history, medications, allergies, initial vital signs, social and family histories, and a review of symptoms. Diagnostic tests (including chest radiographs, polysomnograms, electrocardiograms, thyroid function testing, and pulse oximetry [continuous or spot measurements]) and specific management changes (including patient instruction about weight loss, smoking cessation, or referrals to pulmonary, otolaryngology, psychiatry, or neurology consultants) were recorded. Whether a sleep history was documented in any part of the record and the frequency with which sleep disorders were listed in the patient's problem list were recorded. A sleep history was counted if the house officers or medical students noted the presence or absence of insomnia, snoring, not breathing (witnessed apnea), nocturnal choking, nonrestorative sleep, excessive daytime hypersomnolence, and sleep symptoms interfering with daily function, as well as any other reference to sleep.

Patients were considered to have conditions associated with OSA if they had a history of stroke or myocardial infarction (MI), a documented history of obesity or characterization by the medical students or house officers as obese, and/or the presence of hypertension. Patients were considered hypertensive if there was any documentation in the record of a history of hypertension and/or they were being treated for a blood pressure of [greater than]140 mm Hg systolic or [greater than]90 mm Hg diastolic at hospital admission. Patients were considered obese if a history of obesity was reported or the examiner characterized the patient as obese.

For each patient, a record was generated that included the number of evaluations, and whether a history had been performed, as well as the patient's demographics. Chi-square and Wilcoxon rank-sum tests were used to assess the univariate associations between a patient's characteristics and the performance of a sleep history. [18] Logistic regression was used to assess the joint effects of the patient characteristics. [19] A general estimating equation analysis was used to assess the level of education of the person taking a sleep history. This approach allowed consideration of multiple observations per patient. Differences were regarded as statistically significant at p [less than] .05.

RESULTS

Demographic characteristics of patients are summarized in Table i. Ages ranged from 18 to 103 years, with a median of 60 years. Half of the patients were hypertensive, and 11% were obese; 68% of patients were hypertensive, obese, had had cerebrovascular accidents and MIs, and were regarded as having conditions associated with OSA. [9-14]

Of the house officers, 63% were men; 54% were PGY 1, 30% PGY 2, and 16% PGY 3. In general, upper-level house officers (PGY 2 and 3) had received previous sleep instruction (about 2.5 hours) during clinical conferences, while the extent of interns' previous sleep instruction was unknown. Twenty percent of the house staff had attended Wake Forest University School of Medicine, and the precise extent of their participation in sleep instruction provided was not quantified.

Of the medical students, 51% were men; 57% were in their third year, and 43% were in their fourth year of medical school. All had received previous multidisciplinary instruction regarding sleep problems during their second preclinical year, and most had shown familiarity with the essentials of a sleep history on previous written examination and/or an objective structured clinical examination.

Sleep histories were documented by medical students and/or house officers in 9% of medical records of 208 patients and in only 7% of the 435 histories documented overall (Fig 1). Sleep histories were recorded more often in women than in men (13% vs 4%, P = .027). Otherwise, patients with and without documented sleep histories had similar demographic characteristics and frequencies of conditions associated with sleep disorders, including OSA (Table 1). Sleep histories were more often documented in patients with depression, previous MI., stroke, syncope, hypertension, or obesity but these differences were not statistically significant. Only 9% of the 141 patients who were considered to have a condition associated with OSA had a sleep history recorded (Fig 2). In the 18 patients for whom a sleep history was documented, a total of 19 notations were made related to sleep, including 4 complaints of hypersomnolence and 15 complaints of difficulty sleeping. Similarly low frequencies of sleep histories were observed during each of the five periods of review.

In these patient encounters, medical students documented sleep histories more often than house officers (15% vs 6%, P= .01). Male house officers and medical students more often documented sleep histories than female house officers and medical students (10% vs 3%, P= .037).

Histories related to cigarette smoking and alcohol use were documented for 81% and 72% of patients, respectively, significantly more often than sleep histories (both, P [less than] .001). Trainee level of education was related to the likelihood of recording smoking and alcohol histories, with medical students more often documenting these behaviors. Of the 18 patients with a documented sleep history, 50% were associated with a smoking and/or alcohol history.

The relationship between documented sleep histories and diagnostic testing is summarized in Table 2. House officers ordered pulse oximetry (14/18 [78%] vs 71/190 [37%]; P[less than].001), arterial blood gas measurements (12/18 [67%] vs 56/190 [30%]; P[less than].001), pulmonary function tests (2/18 [11%] vs 2/190 [1%]; P = .038), and electrocardiograms (14/18 [78%] vs 93/190 [49%]; P = .019) more frequently in patients with a sleep history than in patients without recorded sleep histories. Other diagnostic tests potentially related to OSA or cardiorespiratory dysfunction (chest radiograph, thyroid function test, Holter monitor, and echocardiogram) were done with similar frequencies.

Retrospective review of Wake Forest University Baptist Medical Center sleep laboratory records showed that four polysomnograms (PSGs) were ordered in the 208 patients; 2 of these patients had a recorded sleep history at admission, and the other two did not. In three of the four PSGs ordered, the study was recommended by consultants; two by pulmonologists and one by a neurologist. The fourth PSG occurred after house staff (ie, nurse and intern) had witnessed sleep apnea. In all 4 patients a condition associated with OSA was documented, and 3 received a diagnosis and were ultimately treated. Sleeping pills were ordered for 53 patients, including 3 of 18 (16.7%) with and 50 of 190 (26.3%) without recorded sleep histories (P[greater than].05).

DISCUSSION

The sleep history is key to the recognition and diagnosis of potentially treatable sleep disorders. [20,21] The presence of daytime hypersomnolence that interferes with activity, apnea, or witnessed choking [22,23] and obesity [24] are sensitive indicators of sleep disturbance, but observations suggest that the sleep history is sought infrequently. [21] In a hypothetical insomniac elderly patient vignette, only 40% of physicians elicited a sleep history. [25] In a study involving standardized simulated patients, medical interns infrequentiy (13%) and community-based physicians never elicited a sleep history. [26] Although there is a general impression that sleep problems are underdiagnosed [16] and seldom documented, [17] the frequency of recorded sleep histories in hospitalized general medical patients has not been reported previously.

Our findings of a documented sleep history in only 9% of hospital charts is consistent with previous reports and indicates a general underprioritization of sleep problems. The liberal criteria we used to define sleep history, and the finding that approximately two thirds of these patients had conditions associated with OSA makes the low emphasis on sleep all the more remarkable. Interestingly, sleep histories were documented more often in women. This observation is consistent with previous reports of a higher frequency of sleep complaints among older women, and the increased frequency of insomnia in this group. [27,28] Otherwise, the likelihood of a recorded sleep history did not appear to be related to a variety of patient characteristics.

Medical students and house officers differed in documentation of sleep histories. Students were more likely to record histories relating to sleep, smoking status, or alcohol use. While this finding might reflect previously described tendencies of medical students to record more comprehensive histories than house officers, the over twofold higher rate of sleep histories may also reflect differences in previous instruction of these medical students about sleep problems and their clinical importance.

On the other hand, the low frequency of recorded sleep histories by medical students is of concern, since they had not only received previous preclinical sleep instruction, but also had demonstrated their knowledge of the sleep history on a written examination and performance of sleep histories during interviews with simulated patients. As part of their residency curriculum, many of the upper level house officers had also received sleep instruction (though not to the degree that medical students had received). We noted no significant variation in sleep histories with the level of house officers' training or whether they attended Wake Forest University School of Medicine.

Despite the extent of specific sleep instruction, documentation of sleep histories remained low overall. This finding suggests that despite previous educational interventions (with short-term success), sleep histories remain underprioritized. It is unclear whether the lack of documentation of sleep histories reflects house officers' priorities, their time constraints, possible suboptimal attiudes toward sleep disorders, and/or their questions about the level of evidence supporting an intervention and whether it has a meaningful impact on patient outcomes with sleep disorders. Further definition of the factors influencing (and barriers to) sleep histories by health professionals is necessary.

Also noteworthy, for both medical students and house officers, sleep histories were documented less often than other socially important histories such as cigarette smoking and alcohol use (P [less than] .001). It may be suggested that a comparison between the sleep histories and cigarette smoking and alcohol use may be inappropriate since they may appear in traditionally different areas of the recorded "History and Physical" ("Review of Systems" and "Social History," respectively). In our review of these charts, there appeared to be no consistent area in which clinicians documented sleep problems, and we credited sleep histories whenever they occurred.

Of these 208 patients, only 18 had a sleep history recorded, and 3 patients (1.4%) were ultimately shown to have both associated conditions and PSG confirmation of OSA. There was no discernible relationship between sleep history documentation and patient treatment. Overall, however, the process of care appeared to differ, since pulse oximetry, arterial blood gas analysis, pulmonary function tests, and electrocardiograms were more likely to be ordered for patients who had any sleep history recorded (by medical students and house officers). It is not certain whether these enhanced clinical concerns about cardiorespiratory disease in patients with sleep histories are a reflection of recent documentation of the morbidity of sleep disorders, a more comprehensive approach (independent of sleep concerns) on the part of the caregiver, or some other factor. Accounting for these findings requires clarification.

Important limitations of this report must be acknowledged. Because it is well-established that chart audits represent imprecise measures that may underestimate health care provider behaviors, [29,30] it is likely that sleep was addressed more often than we appreciated. Nevertheless, the exceedingly low proportion of documented histories in a high-risk population despite our use of a very liberal definition of sleep history is noteworthy. Because a slightly higher proportion of patients with sleep complaints had depression, and since depressed patients commonly have sleep complaints related to their illness, it remains possible that the frequency of recorded sleep histories among these patients overestimates the concern about primary sleep disorders. Importantly, this review did not differentiate whether the patient volunteered the sleep complaint, or whether it was elicited by the caregiver.

The generalizability of our observations to other institutions is another concern. Because sleep issues are particularly emphasized at our center and are not prioritized in most curricula, we are concerned that the scope of this problem is even more severe elsewhere than at our institution. That sleep histories were documented less often than histories of cigarette smoking and alcohol use is not surprising, and it indicates that sleep is not as highly prioritized as these other areas. This observation in our hospitalized patients is similar to that in ambulatory clinic patients [9] and identifies important lost health promotion opportunities. If treatable sleep problems are to be appropriately prioritized, then barriers to taking sleep histories must be defined and addressed.

From the Department of Internal Medicine, Division of Pulmonary and Critical Care, and the Departments of Public Health Sciences and Psychiatry and Behavioral Medicine, Wake Forest University Baptist Medical Center, Winston-Salem, NC.

Funded by the National Heart, Lung, and Blood Institute, Sleep Academic Award, Grant No. 1 K07 H005647-O1.

Reprint requests to Andrew M. Namen, MD, Wake Forest University School of Medicine, Section on Pulmonary/Critical Care Medicine, Medical Center Blvd, Winston-Salem, NC 27157.

References

(1.) Klink M, Quan SF: Prevalence of reported sleep disturbances in a general adult population and their relationship to obstructive airways diseases. Chest 1987; 91:540-546

(2.) Gottlieb DJ, Whitney CW, Bonekat WH, et al: Relation of sleepiness to respiratory disturbance index: the Sleep Heart Health Study. Am J Respir Crit Care Med 1999; 159:502-507

(3.) Findley EC, Unverzgat ME, Suratt PM: Automobile accidents involving patients with obstructive sleep apnea. Am Rev Res Respir Dis 1988; 138:337-340

(4.) Teran-Santos J, Jimenez-Gomez A, Cordero-Guevara J, et al: The association between sleep apnea and the risk of traffic accidents. N Engl J Med 1999; 340:847-851

(5.) National Commission on Sleep Disorders Research: Wake Up America: A National Sleep Alert. Report of the National Commission on Sleep Disorders Research. Washington, DC, National Commission on Sleep Disorders, 1993, pp 1-4

(6.) Young T, Palta M, Dempsey J, et al: The occurrence of sleep disorder breathing among middle aged adults. N Engl J Med 1993; 328:1230-1235

(7.) Krieger J, Maglasiu N, Sforza E, et al: Breathing during sleep in normal middle-aged subjects. Sleep 1990; 13:143-154

(8.) Ancoli-Israel 5, Kripke DF, Klauber MR, et al: sleep-disordered breathing in community-dwelling elderly. Sleep 1991;14:486-495

(9.) Nieto FJ, Young TB, Lind BK, et al: Association of sleep-disordered breathing, sleep apnea, and hypertension in a community-based study. JAMA 2000; 283:1829-1836

(10.) Peppard PE, Young T, Palta M, et al: Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342:1378-1384

(11.) Parish JM, Shepard JW Jr: Cardiovascular effects of sleep disorders. Chest 1990; 97:1220-1226

(12.) Bradley TD, Rutherford R, Grossman RF: Role of daytime hypoxemia in the pathogenesis of right heart failure in obstructive sleep apnea syndrome. Am Rev RespirDis 1985; 131:835-839

(13.) Fletcher EC, Shah A, Qian W: "Near miss" death in obstructive sleep apnea: a critical care syndrome. Crit Care Med 1991; 19:1158-1164

(14.) Partinen M, Palomaki H: Snoring and cerebral infarction. Lancet 1985; 2:1325-1326

(15.) He J, Kryger MH, Zorick FJ: Mortality and apnea index in obstructive sleep apnea. Chest 1988; 94:9-14

(16.) Young T, Evans L, Finn L, et al: Estimation of the clinically diagnosed proportion of sleep apnea syndrome in middleaged men and women. Sleep 1997; 20:705-706

(17.) Namen AM, Wymer A, Case D, et al: Performance of sleep histories in an ambulatory medicine clinic: impact of simple chart reminders. Chest 1999; 116:1558-1563

(18.) Bland M: An Introduction to Medical Statistics. Oxford, England, Oxford University Press, 1987

(19.) Hosmer Dw, Lemeshow 5: Applied Logistic Regression. New York, Wiley-Interscience, 1994

(20.) Netzer NC, Stoohs RA, Netzer CM, et al: Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med 1999; 131:485-491

(21.) Millman RP: Do you ever take a sleep history? Ann Intern Med 1999; 131:535-536

(22.) Viner 5, Szalai JP, Hoffstein V: Are history and physical examination a good screening test for sleep apnea? Ann Intern Med 1991; 115:356-359

(23.) Roehrs T, Zorick F, Wittig R: Predictors of objective level of daytime sleepiness in patients with sleep-related breathing disorders. Chest 1989; 95:1202-1206

(24.) Crocker B, Olson L, Saunders N, et al: Estimation of the probability of disturbed breathing during sleep before a sleep study. Am Rev Respir Dis 1990; 142:14-18

(25.) Everitt DE, Avorn J, Baker MW: Clinical decision-making in the evaluation and treatment of insomnia. Am J Med 1990; 89:357-362

(26.) Haponik EF, Frye AW, Richards B, et al: The sleep history is neglected diagnostic information: challenges for primary care physicians.] Gen Intern Med 1996; 11:759-761

(27.) Maggi 5, Langlois JA, Minicuci N, et al: Sleep complaints in community-dwelling older persons: prevalence, associated factors, and reported causes.] Am Geriatr Soc 1998; 46:161168

(28.) Roberts RE, Shelma SJ, Kaplan GA: Prospective data on sleep complaints and associated risk factors in an older cohort. Psychosom Med 1999; 61:188-196

(29.) Leaf DA, Neighbor WE, Schaad D, et al: A comparison of self-report and chart audit in studying resident physician assessment of cardiac risk factors. J Gen Intern Med 1995; 10: 194-198

(30.) Gerbert B, Hargreaves WA: Measuring physician behavior. Med Care 1986; 24:838-847

TABLE 1

Documented Demographic and Comorbid Characteristics of Patients on a
General Medicine Ward


                                     Sleep   No Sleep
Patient                  Overall   History    History
Characteristics        (n = 208)  (n = 18)  (n = 190)  P Value

Median age (yr)             60       63        59       .105
Age range (yr)           18-103     40-87     18-103
Race (W:NW)              65%:35%   61%:39%   66%:34%    .690
Sex (M:F)                47%:53%   22%:78%   49%:51%    .027
Hypertension           117 (56%)  11 (61%)  106 (55%)   .475
Obesity                 23 (11%)   4 (22%)   19 (10%)   .121
Previous stroke         28 (13%)   4 (22%)   24 (13%)   .275
Previous MI             49 (24%)   6 (33%)   43 (23%)   .382
Conditions associated  140 (67%)  12 (67%)  128 (67%)   .952
 with OSA [*]
Syncope                 11 (5%)    2 (11%)    9 (5%)    .244
Angina                  32 (15%)   3 (17%)   29 (15%)   .744
Depression              27 (13%)   4 (22%)   23 (12%)   .263



W:NW = White-nonwhite ratio; M:F = male-female ratio; MI = myocardial
infarction; OSA = obstructive sleep apnea.

(*)Conditions associated with OSA include history of or current
treatment for hypertension with blood pressure [greater than] 140
mm Hg systolic or [greater than] 90 mm Hg diastolic, history of/or
physician characterizing the patient as obese, and/or history of stroke
or MI.
TABLE 2

Frequency With Which Diagnostic Tests Were Ordered by House Officers and
Medical Students


                              No. (%)         No. (%)
                              Overall   With Sleep History
Test                         (n = 208)       (n = 18)

Pulse oximetry                85 (41)        14 (78)
Arterial blood gas analysis   68 (33)        12 (67)
Chest radiograph             137 (66)        12 (67)
Pulmonary function test        4 (2)          2 (11)
Thyroid function test         18 (9)          2 (11)
Electrocardiogram            107 (51)        14 (78)
Holter monitor                20 (10)         2 (11)
Echocardiogram                42 (20)         3 (17)



                                    No. (%)
                             Without Sleep History
Test                               (n = 190)        P Value

Pulse oximetry                      71 (37)          .001
Arterial blood gas analysis         56 (29)          .001
Chest radiograph                   125 (66)          .940
Pulmonary function test              2 (1)           .038
Thyroid function test               16 (8)           .659
Electrocardiogram                   93 (49)          .019
Holter monitor                      18 (9)           .686
Echocardiogram                      39 (21)           1.0
FIGURE 1

Level of education related to documented health promotion histories.


                     Total     Student   Resident
                   Histories  Histories  Histories
                     n=435      n=67       n=368

Smoking Histories     69%      88% [*]    66% [*]
Alcohol Histories     58%      77% [+]    54% [+]
Sleep Histories       7%       15% [++]    6% [++]



(*)P = 0.002

(+)P [less than] 0.001

(++)P = 0.01

Note: Table made from bar graph
FIGURE 2

Frequency with which sleep histories were taken in patients with
conditions associated with obstructive sleep apnea. (HO = House officer)


                  % with a condition
                  associated with OSA  Sleep Hx

Medical Students          72%          [*] 15%
HO's 1                    65%          [*] 5%
HO's 2                    70%          [*] 5%
HO's 3                    68%          [*] 8%



(*)P[less than]0.001

Note: Table made from bar graph