Published ahead of print on June 1, 2004, doi:10.1164/rccm.200310-1421OC
© 2004 American Thoracic Society doi: 10.1164/rccm.200310-1421OC
Differences in Patient and Bed Partner–assessed Quality of Life in Sleep-disordered BreathingDepartment of Epidemiology, Johns Hopkins Bloomberg School of Public Health; and Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland Correspondence and requests for reprints should be addressed to Naresh M. Punjabi, M.D., Ph.D., Division of Pulmonary and Critical Care Medicine, Johns Hopkins Asthma and Allergy Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail: npunjabi{at}jhmi.edu
Sleep-disordered breathing (SDB) is associated with daytime sleepiness and impaired quality of life. Clinical experience suggests that there is a discrepancy between the information provided by the patient and the bed partner. Although this discrepancy is widely recognized, it is not known whether there are differences in quality of life ratings as reported by the patient and by the bed partner on the patient's behalf. Using the Short-Form 36 to assess quality of life in 122 patients with SDB (apnea–hypopnea index  5 events/hour), this study found that patients with SDB generally rate their quality of life higher than their respective bed partners. Systematic differences existed between the two raters in the following Short-Form 36 domains: physical functioning, general health, and vitality. Moreover, male patients reported a higher functional status compared with female patients relative to their respective bed partners. In contrast, no differences were noted between self and bed partner quality of life in normal subjects (n = 15) without SDB (apnea–hypopnea index < 5 events/hour) recruited from the general community. This study suggests that systematic differences exist between patient- and bed partner–assessed quality of life in SDB. Bed partner ratings provide supplemental information on quality of life impairment in SDB.
Key Words: bed partner • proxy ratings • quality of life • Short-Form 36 • sleep-disordered breathing Sleep-disordered breathing (SDB) is a chronic condition that is characterized by frequent episodes of partial or complete upper airway collapse during sleep. The resulting decrease or cessation of airflow is often associated with recurrent drops in the oxyhemoglobin saturation and arousals from sleep. Common symptoms include loud snoring, restless sleep, and excessive daytime sleepiness. Epidemiologic data from the Wisconsin Sleep Cohort study indicate that at least 4% of men and 2% of women in the general population have SDB associated with daytime sleepiness (1). Without treatment, patients with SDB often report a general impairment in their daily functioning and demonstrate decrements in attention, vigilance, and memory (2–9). Moreover, there is strong evidence indicating that SDB is independently associated with incident hypertension (10) and prevalent cardiovascular disease (11–13). It is now recognized that in addition to the cardiovascular morbidity and neurobehavioral consequences, individuals with SDB also manifest lower health-related quality of life compared with those without SDB (14–21). Previous studies on quality of life in SDB have employed a number of self-administered generic instruments, including the Medical Outcomes Study Short-Form 36 (SF-36) (22), Sickness Impact Profile (23), and the Nottingham Health Profile (24). Two disease-specific instruments for health-related quality of life in SDB are also now available and are increasingly being used in ongoing epidemiologic studies related to SDB (25, 26). A major assumption in studies on quality of life is that self-reported information reflects the full extent of the disease-related impairment in general health status. Clinical experience suggests that information derived from the bed partner or a proxy can be invaluable in the initial evaluation and subsequent follow-up of patients with SDB. Although bed partner– or proxy-derived histories can have a significant role in clinical practice, there are no available studies that have examined whether patients and proxies provide congruent information on the patient's general health status. Numerous studies in other chronic disease states (27–35) illustrate that proxies can provide supplemental information on disease-related symptoms and the patient's functional status. Thus, the primary objective of this study was to determine whether systematic differences existed in the assessment of quality of life provided by the patient and by a bed partner on the patient's behalf. In addition to characterizing the difference between the patient and the bed partner–reported quality of life in SDB, the current investigation sought to determine whether any discernable difference was associated with factors including patient's age, sex, obesity, or severity of underlying disease, as assessed by the apnea–hypopnea index (AHI). Finally, for comparative analyses, differences in self- and bed partner–reported quality of life were also examined in normal subjects recruited from the general community.
Study Sample and Covariate Data Patients with SDB were prospectively recruited from a consecutive series of new patients referred for polysomnographic evaluation. Eligibility criteria included age of 18 years or more and presence of a bed partner. Patients with a diagnosis of a sleep disorder other than SDB or on treatment with positive pressure therapy were ineligible. SDB was defined as an AHI of five events or more per hour during sleep. A total of 776 patients had polysomnography over the study period and met the definition of SDB. Of the 776 patients with SDB, 187 patients (24.1%) reported not having a bed partner, 456 patients (58.8%) reported having a bed partner, and 133 patients (17.1%) were nonresponders. From the 456 patients with a bed partner, 122 patient and bed partner pairs volunteered for the study with no discernable bias between participants and nonparticipants. For comparison, a sample of 15 subjects without SDB (AHI < 5 events/hour) or other medical conditions was also recruited from the local community. Approval for the study was acquired from the institutional review board, and informed consent was obtained from each participant and the bed partner. Each participant completed a questionnaire that included information on age, sex, race, marital status, educational level, and Epworth Sleepiness Scale. Quality of life was assessed with the SF-36, a generic quality of life instrument that yields eight domain scores and two summary measures (36). Information from the patients with SDB was collected before the start of therapy. Bed partners were asked to complete a questionnaire, which included information on age, sex, race, and their assessment of the participant's quality of life using the SF-36. Bed partners also assessed their own quality of life and daytime sleepiness using the SF-36 and the Epworth Sleepiness Scale (37), respectively. To ascertain the bed partner's perspective on the participant's quality of life, the SF-36 was structured such that the bed partner was reporting for the participant (see the online supplement). The polysomnogram consisted of continuous recordings of C3-A2 and C3-O1 electroencephalographic leads, right and left electrooculographic leads, and submental and bilateral anterior tibialis electromyograms. Respiration was monitored with thermocouples and thoracic and abdominal strain gauges, and oxygen saturation was obtained with a pulse oximeter. Sleep-stage scoring was performed on 30-second epochs according to standard criteria (38). Apnea was defined as a complete cessation of airflow for at least 10 seconds. Hypopnea was defined as any reduction in the airflow associated with an electroencephalographic arousal or a 4% drop in the oxygen saturation. The AHI was calculated as the total number of apneas or hypopneas per hour of total sleep time.
Statistical Analysis
The study sample consisted of 122 patients with 88 male (72.1%) and 34 female patients (27.9%). The characteristics of the patient sample are shown in Table 1. Polysomnography revealed that the patients had, on average, evidence of moderate to severe SDB with a mean AHI of 43.2 events/hour (SD, 25.6). Measurements of daytime sleepiness with the Epworth Sleepiness Scale revealed an average score of 12.9 (SD, 5.2). Bed partners of patients with SDB had a mean age of 49.2 years (SD, 10.4) and an average BMI of 27.6 kg/m2 (SD, 6.1). As a group, bed partners of patients with SDB did not report subjective daytime sleepiness on the Epworth Sleepiness Scale and had a mean score of 6.8 (SD, 4.1). In the sample of normal subjects without SDB (n = 15), the average age was 43.1 years (SD, 4.2), and the mean BMI was 23.8 kg/m2 (SD, 3.7). Table E1 (online supplement) provides a summary of other demographic variables for the normal subjects. There were no same-sex bed partners in either the patient or the normal samples.
Figure 1 shows the average SF-36 domain scores for patients with SDB and for male and female patients separately. Using SF-36 data from the general population as a reference, patients with SDB reported a lower functional status on all SF-36 domains and summary component scores. Moreover, female patients consistently reported more impairment than male patients on each of the SF-36 domains (Figure 1). Women had a lower AHI compared with men (median AHI, 26.4 vs. 43.0 events/hour, p < 0.016) despite a higher BMI (median BMI, 35.9 vs. 31.1 kg/m2, p < 0.005). In contrast to patients with SDB, self-reported SF-36 scores in normal subjects were at or higher than the age- and sex-adjusted values from the general population (see Figure 1 in the online supplement).
In patients with SDB, statistically significant differences were observed between self- and bed partner assessments of several SF-36 domains. Bed partner–reported quality of life scores (Figure 2) were consistently lower compared with those provided by the patient for physical functioning, general health, vitality, and physical component summary score resulting in a positive and significant difference (p < 0.001) for these domains that was independent of age, sex, BMI, and AHI. The associations between the observed difference between the two quality of life assessments and factors including patient's sex, age, and the AHI were subsequently examined. Men and women demonstrated a similar tendency to underestimate their functional status compared with their bed partners on the physical functioning and general health domains (Figure 3). However, a statistically significant (p < 0.01) and negative interaction between sex and rater was noted for several SF-36 domains with the opposing differences in men compared with women. Male patients generally reported a higher vitality score compared with their bed partners, whereas female patients reported a lower vitality score compared with their bed partners (Figure 3). A negative interaction between sex and rater was also observed for social functioning and the mental component summary scores (Figure 3). Patient's age and physiologic measures of disease severity, including the degree of nocturnal desaturation and AHI, were not associated with the difference in SF-36 scores between the two raters.
Figure 4 displays the bed partner's self-reported quality of life. No significant differences were noted for any of the SF-36 domains between the bed partner sample and the expected normative values. Of the bed partner variables, age, BMI, and the Epworth score were not associated with the difference between the patient- and the bed partner–assessed SF-36 values. The effect of bed partner's own quality of life on the difference between the two raters was then examined. Using a cut-point of the median score in the bed partner's physical and mental component scores to define high and low quality of life subgroups, no systematic associations between the bed partner's own quality of life, as defined by either one of the two summary scores, were noted with the difference in assessments provided by the patient and the bed partner.
The primary objective of this study was to compare quality of life assessments provided by patients with SDB with the assessments provided by bed partners on the patient's behalf. This study has several key findings. First, self-reported quality of life scores were generally lower in patients with SDB compared with age- and sex-adjusted normative values. Second, systematic differences exist between quality of life as assessed by the patient and the bed partner. Bed partner–reported quality of life scores were lower compared with those reported by the male patient. In contrast, bed partner–reported quality of life scores were generally higher compared with those reported by the female patient. Third, except for the patient's sex, other characteristics such as age, BMI, AHI, and bed partner's own quality of life were not associated with the difference between patient and bed partner scores on the SF-36. Finally, no significant differences were noted in self- and bed partner–reported quality of life in normal subjects without SDB recruited from the general community. The finding that there are systematic differences between self- and bed partner–reported quality of life in SDB was not unexpected. Patients are often unaware of the nocturnal signs of SDB, such as witnessed apneas and restless sleep, and tend to underreport the symptoms of daytime sleepiness. This study indicates that in addition to the underreporting of symptoms and signs associated with SDB, patients generally score higher on quality of life assessments compared with information provided by their bed partners. The discrepancy between self and bed partner reports may, in part, be related to patient's lack of recognition or denial of the potential impact that SDB has on general health. Because SDB is often insidious, patients may unknowingly adjust their expectations of general health during the preclinical phase. Alternatively, the observed differences between patient- and bed partner–reported quality of life might result if bed partners exaggerate the disease-related impairment from SDB. Nocturnal symptoms of SDB (i.e., loud snoring) often interfere with the bed partner's ability to sleep (39) that may, in turn, cause a reduction in their functional status and bias their reports. However, in this study, no significant associations were noted between the bed partner's own quality of life and the difference in ratings provided by the patient and the bed partner. In fact, bed partner reports of their own quality of life were, on average, not significantly different than age- and sex-adjusted expected values. The fact that SDB does not have a significant impact on the bed partner's quality of life, as assessed by the SF-36, has also been noted in a recent study by Parish and Lyng (40). Surprisingly, despite the absence of any significant differences from normative data, these investigators observed significant improvements in the bed partner's quality of life 6 weeks after the patients had initiated positive pressure therapy. The results of this study demonstrate that the difference in perceived quality of life is influenced by sex, as reflected in male patients reporting a higher functional status compared with female patients relative to their respective bed partners. The effect of sex could be related to the fact that female bed partners in the study sample may have felt burdened by the patient's impairment and therefore rated the patient's quality of life lower compared with their male bed partners. Alternatively, men may not fully appreciate the impairment caused by SDB on their affected bed partners. Thus, it remains to be determined whether men underestimate and/or women overestimate the decrements in their bed partner's health status from SDB. It is also possible that the observed discrepancies in quality of life ratings are related to the differences in the clinical manifestations of SDB in men and women (41). Finally, differences in thresholds for seeking medical attention between men and women could also explain the sex differences in perceived quality of life. Referrals to sleep laboratories for women are significantly less than for men (42) possibly because of differences in symptom profiles and/or failure of the healthcare professionals in responding to the subtle symptoms of SDB in women (41, 43). Although differences in quality of life as assessed by the patient and the bed partner have not been previously examined in SDB, the results of this study are consistent with several reports in other chronic diseases that have used proxy assessments of quality of life. In a study of patients with stroke, Segal and Schall (44) showed a poor correlation between patient- and proxy-reported health status. Similarly, in a study of people who are older and disabled, Pierre and colleagues (45) showed poor to moderate agreement between patient and proxy-reported quality of life. Studies in other chronic illnesses (46) including cancer (32, 34, 35) have confirmed that patients often underestimate the impairment in their quality of life compared with their proxies. Because many chronic illnesses are associated with major decrements in cognitive and neuropsychological function, the proxy serves as an invaluable source of patient-related information that would be otherwise unavailable. Despite the fact that SDB is associated with relatively mild neurocognitive impairment (2–8), this study provides support for the notion that proxies (i.e., bed partners) of patients with SDB may provide an additional and an alternative source of information with regard to the functional impact of SDB. Although the clinical value of patient- and bed partner–derived histories was not examined in this study, the disparity in health status as reported by the patient and the bed partner provides indirect evidence on questions of potential clinical significance. Of the eight SF-36 domains, significant differences between the patient and bed partner were observed for vitality, general health, and physical functioning. Thus, responses to questions regarding patient's level of fatigue, energy level, and limitations in performing physical activities should also be obtained from a bed partner or proxy when available (see the online supplement for specific questions). Several limitations should be considered in the interpretation of this study. First, the patient sample used in this study was recruited from a clinical population. The use of a clinical sample can bias the difference in quality of life assessments between the patient and the bed partner away from the null as male patients with SDB are frequently referred for clinical evaluation because of concerns of their spouse or bed partner. Second, patient and bed partner pairs included in this study represent a convenient sample and thus may not provide generalizable estimates. However, comparisons between patients that were and were not enrolled revealed no significant differences in age, sex, BMI, or physiologic measures of disease severity. Third, this study required that bed partners complete the SF-36 on the behalf of the patient. The SF-36 was designed and validated as a self- and not as a proxy-administered instrument. Previous studies in other chronic illnesses have used a similar paradigm by employing self-administered instruments to assess proxy ratings. Moreover, data from normal individuals in this study indicate that self and bed partner reports on general health are not significantly disparate. In conclusion, the results of this study reveal systematic differences in the perception of quality of life as assessed by the patient and a bed partner. Clinicians frequently need to consider quality of life when making decisions about the need for treatment of SDB. Depending exclusively on patient report of quality of life may be insufficient. Additional ratings from a bed partner or other family members that observe the patient on a daily basis could provide useful information on the impairment associated with SDB. Finally, future clinical and epidemiologic studies that examine quality of life in SDB should consider inclusion of bed partner–assessed quality of life as a supplemental measure for assessing the impact of SDB and the degree of improvement with therapy.
The authors thank Terri E Weaver, Ph.D., F.A.A.N., R.N., C.S. (University of Pennsylvania School of Nursing), for her assistance in the review and interpretation of the data presented in this article.
Supported by the Agency for Healthcare Research and Quality (HS10786) and the National Institutes of Health (HL04065 and HL75078). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: J.G.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this article; D.E.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this article; P.L.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this article; N.M.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this article. Received in original form October 18, 2003; accepted in final form May 28, 2004
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
Point estimates and 95% confidence intervals for the difference between the patient and the bed partner are shown. Values are adjusted for the patient's age, sex, body mass index (BMI), and apnea–hypopnea index (AHI).
