The Effect of Positive Reinforcement on Hourly Compliance in Nasal Continuous Positive Airway Pressure Users with Obstructive Sleep Apnea 1- 3
EUGENE
c.
FLETCHER and RITA A. LUCKETT
Introduction
Since the first descriptions of nasally applied expiratory positive airway pressure (nCPAP), this mode of therapy has become the most popular and effectivenonsurgical method for treating obstructive sleep apnea (GSA) (1-4). In many cases, when properly applied using adequate pressure and with sufficient patient education and acceptance of therapy, nCPAP may eliminate apnea, reverse symptomatology, improve daytime blood gases (5), and ameliorate the cardiovascular sequelae of GSA (6). However, similar to the problem presented with chronic home oxygentherapy for diseases causing hypoxemia (7, 8), the therapeutic success of nCPAP relies heavily upon patient acceptance of the device since its therapeutic potential may only be realized while the device is being used. Previous reports describe compliance with nCPAP mainly in terms of the number of patients with GSA able to use it beyond their initial trial night (agree to use a machine at home) or those continuing use after some home therapy. From 15 to 300/0 of GSA patients tested with nCPAP in the sleeplaboratory either cannot tolerate it on the initial night or refuse to take it home as chronic therapy (9-14). At follow-up ranging from 2 to 24 months, 17 to 33% of those taking the machines home havediscontinued use or are using it only intermittently (9-14). With the exception of one study (15), most data on compliance are retrospective, limited to single point-in-time surveys collected from questionnaires that depend on subjective reports of compliance, and omit data on those not returning surveys. Furthermore, the methods of educating GSA patients and acclimating them to chronic home nCPAP use vary widely and active intervention to promote compliance is nonuniform or nonexistent. Since there may be a difference be936
SUMMARY Previous reports havedescribed compliance with nasal continuous positive airway pressure (nCPAP) for the treatment of obstructive sleep apnea (OSA) only in terms of the number of patients able to use it beyond their initial trial night or those continuing after some home use. Because of a possible difference between the level of compliance (mean number of hours of use per 24 h) needed for symptomatic relief of OSA versus cardiovascular Improvement, the level of hourly compliance in chronic nCPAP users may be important. The first part of this study prospectively examines compliance In a stable population of OSA patients already using nCPAP for 6 months to 2 yr. The second part is a prospective randomized, crossover study examining the effect of weekly (three times) then monthly (twice) positive reinforcement on hourly compliance of new nCPAPusers for 3 months versus no reinforcement for 3 months. Positive reinforcement consist~d of telephone discussions with the patients about the severity or complications of OSA, benefits of nCPAp,and suggestions about minimizing side effects. Using self-assessment scales, each patient reported the perceived level of improvement from the untreated to the treated condition and the prevalence and severity of side effects from the nCPAPtherapy. The level of compliance in stable, chronic nCPAP users with OSA was 6.1 ± 2.2 h/24 h (n 9). For the new nCPAP users during the nonreinforced period, the mean compliance was 6.0 ± 2.8 h/24 h: that during the reinforcement period was 6.0 ± 2.7 h/24 h (NS). There was no significant correlation between perceived improvement in OSA symptoms or between the perceived side effects of nCPAPversus hourly compliance. Positive reinforcement by telephone does not Improve hourly compliance with chronic use of nCPAP for OSA.
=
AM REV RESPIR DIS 1991; 143:936-941
tween the hours per night of nCPAP use needed for symptomatic improvement from GSA versus that needed to allow improvement in hemodynamic abnormalities, such as systemicand pulmonary hypertension, we first thought that it was important to prospectively establish by objective criteria the level of hourly use (compliance) in a stable population of chronic home nCPAP users. Second, we were interested in whether maintenance of communication, attention, and a show of interest in the patient in a cost-effective manner improved chronic compliance with nCPAP therapy. We attempted to demonstrate this using a prospectivecrossover method of positive reinforcement by telephone versus no positive reinforcement, following a standardized program of optimal education and acclimation to the device before home use. Wehypothesized that nightly use of the nasal CPAP machine once it is in the home may vary greatly between patients and that hourly compliance may be improved by close support and follow-up from the clinic or physician prescribing the therapy. Final-
ly,we sought to determine whether nightly compliance to therapy may be related to improvement in symptoms or the development of nCPAP side effects. Methods Overview The study was divided into two parts. First, describing compliance as the number of hours used per night, we prospectively examined compliance in a stable population of OSA
(Received in original form February 6, 1990 and in revised form December 11, 1990) 1 From the Department of Medicine, Pulmonary Disease Section, Veterans Administration Medical Center, Baylor College of Medicine, Houston, Texas. 2 Supported in part by an equipment grant from Respironics Inc., Pittsburgh, Pennsylvania, and by the General Research Service of the Department of Veterans Affairs. 3 Correspondence and requests for reprints should be addressed to Eugene C. Fletcher, M.D., Associate Professor of Medicine, Pulmonary Disease Section 111i, Veterans Administration Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030.
RATE OF DESATURATION DURING APNEA
patients actively using nCPAP for 6 months to 2 yr. Each subject was issued a new machine (Sleep Easy® II or III; Respironics Inc., Monroe, PA) with a Hobbs meter. This meter records the time that the machine is in operation (but not necessarily worn by the patient). Meter readings were recorded at the time of issue and monthly thereafter (by telephone report from patient) for 6 to 9 months. The subjects were told that the reason for switching machines was to update to a later model, and no active intervention to alter compliance occurred. The second part of the study was a prospective, randomized, crossover study examining the effect of positive reinforcement on nightly compliance of new nCPAP users. Following polysomnographic diagnosis of GSA, the subject slept a second night during which increasing levels of pressure were applied, determining the optimal nCPAP pressure to minimize apnea frequency. The patient was subsequently hospitalized one to three nights during which he or she slept with the nCPAP device at the predetermined optimal pressure and with the best fitting nasal mask to minimize leaks and mask discomfort. During this short period of hospitalization, each subject viewed a videotape produced by Respironics about GSA and use of the nCPAP machine. They were given instructions on the care and use of the machine by a nurse familiar with its operation. They wererequired to demonstrate proficiency with use and cleaning of the device and reading of the Hobb's meter before being discharged. In an alternating fashion new patients were assigned to one oftwo groups: telephone reinforcement about GSA and nCPAP use during the first 3 months (weekly for 3 wk then monthly for 2 months) or a control group called monthly but asked only for the hour reading on their machine's Hobb's meter and to report side effects of the therapy and perceived level of improvement from pretherapy symptomatology. If the subject reported side effects or problems with the equipment, suggestions weremade to correct these problems. Unscheduled calls from patients reporting side effects, problems with equipment, or to ask questions were accepted during all portions of the study. Suggestions by the study nurses for solving problems and answers to patient questions were offered throughout the study. During the course of the study patients brought their machines to the clinic one or more times to have specificproblems checked, mask or spacer size changed, and filters or valves changed. At that time meter readings were checked and back-calculation used to verify use times based upon meter readings reported over the telephone. This was done to detect intentional or unintentional misreadings of the meter. At the end of the first 3 months, reinforcement was used on the group who initially receivednone and stopped in those initially reinforced. After all subjects completed the reinforced and nonreinforced periods, they were followed an additional 3 months to confirm persistence of compliance.
937
Positive reinforcement consisted of telephone discussions with the patients about the severity of the disease, complications of apnea, the benefits of nCPAP, and suggestions about minimizing side effects. The content of the following paragraph was communicated in its entirety in a discussion format that was intended to be informative while conveying some urgency and working all elements of the paragraph into the conversation. The purpose of this telephone discussion was to convey a sense of interest in the patient's welfare as it relates to treatment of his or her condition through use of the nCPAP machine.
from zero (none), 1 (occasional), and 2 (moderate or tolerable) to 3 (severeor intolerable) and summed. The nine side effects were nasal stuffiness, dry mouth, difficulty breathing, mask discomfort, headache, machine noise, dry eyes, frequent awakening, and travel restriction. The maximum possible side effect score for each patient was 27.
Statistical Methods Mean hourly use per night was calculated from the meter readings in the stable nCPAP users and in the new patients during periods of positive versus no reinforcement. Compliance in stable users was compared to new users by the t test for unpaired data, and between reinforced and nonreinforced periods in new users by the t test for paired data. Hourly compliance in new users between reinforced, nonreinforced, and the 6- to 9-month follow-up period was compared using one-way analysis of variance. Correlation between baseline apnea severity (apnea index, mean level of desaturation, and mean apnea duration), level of nCPAP pressure required, baseline symptomatology, perceived improvement, and perceived side effects was examined for correlation with hourly compliance using leastsquares regression. The same variables were analyzed by least-squares regression with stepwise model selection to examine their relative usefulnessas combined predictors of compliance (16).The null hypothesis was rejected at p < 0.05. The protocol was approved by the Human Studies Committee of this institution, and all subjects signed informed consent before participation.
Mr.lMrs. , you have a very severe form of sleep apnea. You stop breathing during the night for periods up to 45 s or a minute and you do this as frequently as 50 to 100times per hour. The lowering of oxygen in your bloodstream may have bad effects on your heart and brain. It may cause heart failure and fluid in the legs. In rare cases patients with sleep apnea, like yourself, may die in their sleep. It is very important for you to use your CPAP at all times that you are sleeping. When you do not use the CPAP, you have apneas even though you may not know it. The only time you are free of apneas is when you are using the machine. If you haveproblems with the machine or discomfort from the mask, we are here to try to help you so that you use the machine as much as possible. Call us if you need help with the machine or if you haveany questions. Remember, use the machine every time you lie down to go to sleep, even for afternoon or evening naps.
Symptomatology and Side Effects
Results
Before using nCPAP each new subject responded to 25 questions regarding the level of GSA symptomatology in various areas of daytime and nocturnal function (appendix 1). The patient rated himself or herself as zero (no symptoms) to 3 (maximal symptoms) on each of the 25 questions. The maximum score was divided by 25 to determine the level of baseline symptoms. Following the first week of nCPAP therapy and at each follow-up call, using self-assessment scales, each patient reported the perceived level of improvement from the untreated to the treated condition. Improvement scores ranged from zero (no improvement) to 3 (much improvement). The perceived level of nine side effects was scored
A total of 20 patients were enrolled in the study: 8 males and 1 female in the stable, chronic nCPAP user group and 11 males in the new user intervention group. A single male subject in the intervention group moved from the area shortly after taking his nCPAP machine home andno data wereavailable. Thus data included in the crossover portion of this study are based upon the remaining 10 subjects. The mean age of the patients in the stable chronic group was 60, ideal body weight 155070, apnea index 55 h:', mean arterial desaturation 74070, and
TABLE 1 AGE AND BASELINE APNEA CHARACTERISTICS IN STUDY SUBJECTS Stable. Chronic User Group,
n
= 9
(Mean ± SO) Age IBW,% Apnea index, apneas/h in bed Average low saturation, % Average duration, s
59.6 155 55 74 26
± ± ± ± ±
6.1 34 20 9 11
New User, With or Without Reinforcement Group, n =7 (Mean ± SO)
51.6 158 49 65 27
± ± ± ± ±
12.8 20 26 12 8
P NS NS NS NS NS
FLETCHER AND LUCKETT
938 TABLE 2 COMPLIANCE, PERCEIVED IMPROVEMENT, AND PERCEIVED EFFECTS IN NEW NCPAP USERS Reinforced Period
Nonreinforced Period
Mean
SD
Mean
SD
P
Hourly use, h/night
5.95
2.74
6.04
2.n
NS
Perceived improvement* (possible score 0-3)
2.49
0.65
2.42
0.63
NS
Perceived side effectst (possible score 0-27)
3.37
1.59
2.76
2.22
NS
• Perceived improvement evaluates the patient's perception of how much relief the n~PAP p~ovides from symptoms before its use. A score of zero implies no relief and a score of 3 ma~(Imal. rehef. t Perceived side effects evaluates the patient's perceived level of problems from nine Side effects of nCPAP listed in the text. Zero represents no side effects and a SUbject with the worst possible level of all nine side effects could score 27 points.
mean apnea duration 26 s (table 1). The intervention group did not vary from the chronic group in any of these variables. The level of compliance in the stable chronic nCPAP users was 6.1 ± 2.2 hi 24 h. For the new nCPAP users (n = 10) the mean compliance was 6.0 h/24 h for both the reinforced and nonreinforced periods (table 2 and figure Ij.During the first 3 months of the study there was no difference in compliance between those patients who receivedreinforcement (6.58 ± 1.8 h/24 h; n = 6) and those not reinforced (5.55 ± 3.9; n = 4). During the second 3 months of the study there was no difference in compliance between those receiving reinforcement (5.0 ± 3.9; n = 4) and those no longer receivingreinforcement (6.37 ± 2.1; n = 5). Compliance at 6 to 9 months (follow-up period) for the new nCPAP users was 6.55 ± 3.50 h/24 h. None of the nightly compliance figures varied significantly from the others. There was no difference in the level of perceived improvement or level of perceived side effects between reinforced and nonreinforcement periods. There was no significant correlation between the level of baseline symptoms, perceivedimprovement, perceivedside ef-
fects, apnea index, mean desaturation level, mean apnea duration, or nCPAP pressure required for optimal treatment of apneas versus the level of compliance (hourly usage) during either the reinforced or nonreinforced period (table 3). Least-squares regression with stepwise model selection did not show any related variables that together correlated with hourly compliance. The correlation between baseline symptoms and level of perceived improvement was 0.62 (p < 0.05). No patient failed to use the machine at least part-time, but four patients did not use it on a nightly basis. Patient 4 (figure 1), with the lowest hourly compliance (0.6 ± 1.1 h/24 h), used the machine only intermittently and stopped using it altogether during the 6- to 9month follow-up period. Patient 8 (2.4 ± 1.8 h/24 h) used the machine on week nights only. Patient 6 (3.9 ± 1.4 h/24 h) used the machine intermittently depending upon symptoms but lost 38 pounds by the end of the 6- to 9-month followup period and discontinued use of nCPAP. Patient number 3, who used the machine intermittently (5.3 ± 1.1 hi 24 h), expired from a probable coronary
12 'C
Q)
III
::::l Q.
:
-
8
(J
c
.c
Cl
Z
4
o
•
•
Reinforced
NonReinforced
..... 4
6-9 Month Follow-up
Fig. 1. Hourly compliance for all 10 subjects completing the "positive reinforcernenr (left), "no reinforcemenf (middle), and 6- to 9-month follCMI-uP period (right) of nCPAP therapy. Individual subjects are denoted by the number to the right of the follCMI-up data. The four individuals with the lowest compliance were intermittent users. See text for explanation.
event (during daytime hours) near the end of the 6- to 9-month follow-up period. All other patients used the machine on a nightly basis, but some slept with the machine for less than the total number of hours in bed. Two patients reported occasional mask removal during sleep that was usually associated with arousal and replacement of the mask. The correlation coefficient between hours used calculated from weekly and monthly telephone meter readings by the patients (n = 19)and that taken directly from the Hobb's meter at the time of distribution and return was 0.990 (a = 0.27, b = 1.07; p < 0.0001). Discussion
Previous reports on compliance with home nCPAP in OSA patients have been mainly retrospective and subjective and involve no standardized program of introduction, education, or acclimation to the device. Furthermore, no studies have been interventional, trying to improve compliance by reinforcement. There are three obvious factors that could affect nCPAP compliance: (1) motivation, which is tied to severity of symptoms and satisfaction with the mode of therapy; (2) perceived complications or discomfort from use of the device; and (3) degree of education about and understanding of the pathophysiology of the OSA and the importance of treating it. The present study was designed to accumulate objective data on hourly compliance in a stable population of chronic nCPAP users. It was further designed to address compliance factors by examining the level of symptoms and perceived improvement by examining the degree of side effects (to include systematic treatment of discomfort and side effects) and by maximizing education, acclimation, and reinforcement of use. Our results reveal that hourly use in OSA patients who just acquired nCPAP did not improve over that of a stable population of users in whom no organized attempt at education, acclimation, or positive reinforcement occurred. Compliance did not correlate with the level of baseline symptoms, perceived improvement, level of side effects, parameters of apnea severity, or level of nCPAP pressure required to alleviate apnea. Only one study to date has prospectively collected objective data on chronic compliance with nCPAP. Krieger and Kurtz (15) evaluated 45 subjects every 3 months (range < 1 to 40 months). The mean rate of use was 5.14 ± 0.31 h/day,
RATE OF DESATURATION DURING APNEA
939 TABLE 3
CORRELATIONS BETWEEN SYMPTOMS, PERCEIVED IMPROVEMENT OR SIDE EFFECTS, AND THE LEVEL OF NIGHTLY COMPLIANCE Nonreinforced Period
Reinforced Period
Baseline symptoms Perceived improvement Perceived side effects Apnea index Required nCPAP pressure
R*
P
R
P
0.23 0.51 -0.21 -0.38 0.17
NS NS NS NS NS
0.34 0.53 -0.33 -0.35 0.00
NS NS NS NS NS
• Coefficient of correlation.
which is not very different from the chronic compliance rate of 6.1 h/night in the current "stable users" group. Regarding reasons for compliance versus noncompliance, the current study agrees with some findings of other studies. For example, Waldhorn and colleagues (14) found no correlation between severity of the apnea (based upon pretreatment apnea/hypopnea index), frequency of adverse reactions during therapy, or level of nCPAP pressure prescribed versus therapeutic noncompliance (in this case discontinuation of nCPAP after home placement). They did think there was a higher level of symptoms in the compliant versus noncompliant group, but symptoms were relieved in both groups to an equal degree. Our results may vary with the last point because no patient with only minimal symptoms or without daytime hypersomnolence was included in our study. The mean score on baseline symptomaticity was 1.7 of a maximal score of 3.0, whereas any score above 1.0 in our laboratory implies moderate symptomaticity. Other studies of nCPAP have reported initial noncompliance rates (refused to take the machine home) of 15 to 25070 and follow-up discontinuation after home use of 18to 30% (9-14). The present study does not examine either of these figures because 9 of 10 of the new users continued to use the machine (but with variable hourly compliance) at 9 months of follow-up and the stable, chronic user group were all currently using the device at the time of recruitment. One criticism of this study could be that the time period of follow-up may have been too short and that if followed 12months or longer, compliance would have decreased. Schweitzer and coworkers (10), having surveyed 96 patients (use range 2 to 19 months) who agreed to home nCPAP therapy, found that most patients who will stop therapy do so within the first 3 to 4 months. That compliance at 9
months was not significantly different from that at 3 and 6 months in our subjects agrees with the findings of Schweitzer. Indeed, the study by Krieger and Kurtz (16) seems to show an overall increase in hourly nCPAP compliance with time, which they attribute to "progressive adaptation to the constraints of CPAP therapy" or to improved nCPAP devices. Another criticism could be that recruitment into a study could induce a nonspecific motivational effect related solely to a change in patient condition: being selected to be in a special (research) group, the so-called Hawthorne effect. Since it is necessary to meet human study standards of informed consent, it is impossible to eliminate the nonspecific effect of selection to a special group. However, the Hawthorne effect cannot be maintained over time without a continued change in conditions. We believe that had the Hawthorne effect altered this study a decrease in compliance would have been noted in the 6- to 9-month follow-up, and this clearly was not the case. Previous studies have shown that tracheostomy improves symptoms as well as hemodynamic abnormalities, that is, pulmonary and systemic hypertension. In a study performed by Motta and colleagues six patients with symptomatic OSA received tracheostomies because of severesymptoms and the presence ofpulmonary and systemic hypertension (17). Within weeks after this procedure, symptoms disappeared, daytime Pa02 increased from 38 to 71 mm Hg, and substantial reductions were noted in the mean pulmonary artery and systemic pressures. Nasal CPAP has been shown to markedly improve the hypersomnolence and other central nervous system symptoms of OSA and to correct chronic respiratory failure (6). Unlike tracheostomy, however, nCPAP has never been shown to reverse the chronic hemody-
namic abnormalities of OSA. Both modalities are capable of eliminating apnea, which would theoretically eliminate the hypoxic stimulus to systemic and pulmonary hypertension. However, tracheostomy patients are almost by definition 100% compliant since the apparatus for eliminating the apneas is always in place. Many nCPAP patients, on the other hand, may find the apparatus uncomfortable, for whatever reason, and remove it during the night or not wear it every night. They may titrate their usage of the machine to eliminate symptoms of hypersomnolence but not use it enough to completely eliminate apneas. The outcome of this type of use could be improved symptoms but a continued hypoxic stimulus provoking hemodynamic abnormalities. For this reason nightly and hourly compliance may be an important question in nCPAP therapy. It appears from the results of the present study that the degree of symptomaticity, the level of side effects, or the degree of improvement does not individually predict successful compliance with nCPAP therapy. Likewise, pre-prescription education and acclimation to the machine does not raise the level of compliance above that achieved in a stable population who have accepted nCPAP therapy for a minimum of 6 months without a specific educational or acclimation program. A practical method of positive reinforcement (by telephone) does not appear to alter compliance. This does not suggest that some minimum levelof education, acclimation, reinforcement, and side effect solving is unnecessary to achieve a moderate level of compliance. It does suggest that no area deserves particular emphasis over another and that those patients who comply do so because of innate motivational factors that we have yet to identify. Future studies should combine objective evidence of nCPAP compliance with objective evidence of symptom improvement, such as the multiple sleep latency test, and blood gas and hemodynamic improvement. The results of this study have alerted us to the great variation in hourly use of nCPAP among patients who take a machine home and continue to report use of the machine during clinical follow-up. We do not monitor hourly use in all patients taking nCPAP machines home since we believe that resolution of symptoms remains an excellent indicator of machine use. However, we monitor nightly and hourly compliance in patients in whom cure of apnea is essential for treat-
FLETCHER AND WCKETT
940
ment of abnormal hemodynamics, such as patients with concomitant chronic obstructive pulmonary disease or daytime alveolar hypoventilation with prior episodes of right heart failure. We believe that reasonable compliance should be established in such cases and perhaps more aggressive therapy (tracheostomy) used in those whose hourly use is low because of discomfort or neglect and in whom cure of apnea may be essential for improvement of pulmonary hypertension and blood gases (18). One further application for such monitoring is in the OSA patient who continues to complain of somnolence while claiming to be using his or her machine. References 1. Sullivan CE, Berthon-lones M, Issa FG, Eves L. Reversal of obstructive sleep apnea by continuous positive airway pressure applied through the nares. Lancet 1981; 1:862-5. 2. Mahadevia AK, Onal E, Lopata M. Effects of expiratory positiveairway pressure on sleep-induced respiratory abnormalities in patients with hypersomnia-sleepapnea syndrome. Am RevRespir Dis 1983; 128:708-11.
3. Rapoport DM, Garay SM, Goldring RM. Nasal CPAP in obstructive sleep apnea: mechanisms of action. Bull Eur Physiopathol Respir 1983; 19:616-20. 4. Remmers lE, Sterling lA, Thorarinsson B, Kuna S. Nasal airway positive pressure in patients with occlusive sleep apnea. Am Rev Respir Dis 1984; 130:1152-5. 5. Berthon-lones M, Sullivan CEo Time course of change in ventilatory response to CO 2 with longterm CPAP therapy for obstructive sleep apnea. Am Rev Respir Dis 1987; 135:144-7. 6. Sullivan CE, Berthon-lones M, Issa FG. Remission of severeobesity-hypoventilationsyndrome after short-term treatment during sleep with nasal continuous positiveairwaypressure. Am Rev Respir Dis 1982; 128:177-81. 7. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease. Ann Intern Med 1980; 93:391-8. 8. Report of the Medical ResearchCouncil Working Party. Long-term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981; 1:681-5. 9. Sanders MH, Gruendl CA, Rogers RM. Patient compliance with nasal CPAP therapy for sleep apnea. Chest 1986; 90:330-3. to. SchweitzerPK, Chambers GW, Birkenmeir N, Walsh lK. Nasal continuous positive airway pressure (CPAP) compliance at six, twelve and eighteen months. Sleep Res 1987; 16:186.
11. Baker lP, Rose V, Ware C. Obstructive sleep apnea: therapeutic compliance. Trans Am Clin Climatol Assoc 1987; 99:224-30. 12. Kribbs NB, Kline LR, Pack AI, Dinges DF. Intermittent nasal CPAP therapy and sleepdisordered respiration. Am Rev Respir Dis 1989; 140:A115. 13. Nino-Murcia G, McCann CC, Bliwise DL, Guilleminault C, Dement WC. Compliance and side effects in sleep apnea patients treated with nasal continuous positive airway pressure. West 1 Med 1989; 150:165-9. 14. Waldhorn RE, Merrick TW, Nguyen MC, O'Donnell AE, Sodaro 1, Potolicchio Sl. Longterm compliance with continuous positive airway pressure therapy of obstructive sleep apnea. Chest 1990; 97:33-8. 15. Krieger 1, Kurtz D. Objective measurement of compliance with nasal CPAP treatment for obstructive sleep apnea syndrome. Eur Respir 1 1988; 1:436-8. 16. SAS/STAT users guide. Edit. 6.03. Cary, NC: SAS Institute Inc., 1988; 818. 17. Motta 1, Guilleminault C, Schroeder rs, Dement WC. Tracheostomy and hemodynamic changes in sleep induced apnea. Ann Intern Med 1978; 89:454-8. 18. Fletcher EC, Schaaf rw, Miller 1, Fletcher 1. Long term cardiopulmonary sequelae in patients with sleep apnea and chronic lung disease. Am Rev Respir Dis 1987; 135:525-33.
941
RATE OF DESATURATION DURING APNEA
APPENDIX 1. QUESTIONNAIRE FOR SLEEP DISTURBANCES The purpose of this questionnaire is to determine if daytime sleepiness may be related to breathing disturbances during sleep. Please answer questions to the best of your ability. Ask the technician if you need help.
o
1
3
2
1. Do you fall asleep while watching TV or at the movies? Never Very rarely Occasionally
Frequently
Not applicable
_
2.
Do you fall asleep reading newspapers, books or magazines? Never Very rarely Occasionally Frequently
Not applicable
_
3.
Do you fall asleep while talking with someone in person or over the telephone? Never Very rarely Occasionally Frequently _
4.
Do you get extremely sleepy while driving or fall asleep at stoplights? Never Very rarely Occasionally Frequently
Not applicable
_
5.
Have you had an automobile accident or driven off the road because of sleepiness? Never One time Two times More than two times
Not applicable
_
6.
Do you now, or have you fallen asleep while at work? Never Very rarely Occasionally
Not applicable
_
Frequently
7.
Do you fall asleep at public gatherings (movies, sports events, plays, etc.)? Never Very rarely Occasionally Frequently
8.
Do you get sleepy whenever you are inactive or bored? Never Very rarely Occasionally
9.
Frequently
Not applicable _._ _ _
Do you ever fall asleep in situations where you particularly need to stay awake? Never Very rarely Occasionally Frequently _
10. Do you ever feel confused when you awaken from sleep? Never Very rarely Occasionally
Frequently
_
11. Do you ever wake up in the morning with headaches? Never Very rarely Occasionally
Frequently
_
12. Do you ever "wet the bed" during the night? Never Very rarely Occasionally
Frequently
_
13. Do you ever sleepwalk? Never Very rarely
Frequently
_
Occasionally
14. Have you performed simple tasks during the day that you later don't remember doing? Never Very rarely Occasionally Frequently _ 15. Do you have vivid dreams as you are going to sleep or waking up? Never Very rarely Occasionally Frequently
_
16. Do you snore or has someone told you that you snore loudly? Never Very rarely Occasionally Frequently
Don't know
17. Has your bed partner told you that you stop breathing during sleep? Never Very rarely Occasionally Frequently
Not applicable
_
18. Has your bed partner awakened you for fear that you were not breathing? Never Very rarely Occasionally Frequently
Not applicable
_
19. Is your sleep restless with frequent turning or wild movements? Never Very rarely Occasionally Frequently
Not applicable
_ _
_
20.
Have you ever struck your bed partner or injured yourself during sleep? Never Very rarely Occasionally Frequently
Not applicable
21.
Does you bed partner sleep in another room because you disturb their sleep? Never Very rarely Occasionally Always _
Not applicable
_
22.
Do you find that your mind is not working as quickly or that you are more forgetful? Never Very rarely Occasionally Frequently _
Not applicable
_
23.
Have other people told you that recently you are more irritable or short-tempered? Never Very rarely Occasionally Frequently _
Not applicable
_
24.
Do you have trouble "having sex" recently? Never Very rarely Occasionally
Frequently
_
No interest
25.
Do your ankles swell? Never Very rarely
Frequently
_
Not applicable
Occasionally
No opportunity _
_
EUGENE
c.
FLETCHER and RITA A. LUCKETT
Introduction
Since the first descriptions of nasally applied expiratory positive airway pressure (nCPAP), this mode of therapy has become the most popular and effectivenonsurgical method for treating obstructive sleep apnea (GSA) (1-4). In many cases, when properly applied using adequate pressure and with sufficient patient education and acceptance of therapy, nCPAP may eliminate apnea, reverse symptomatology, improve daytime blood gases (5), and ameliorate the cardiovascular sequelae of GSA (6). However, similar to the problem presented with chronic home oxygentherapy for diseases causing hypoxemia (7, 8), the therapeutic success of nCPAP relies heavily upon patient acceptance of the device since its therapeutic potential may only be realized while the device is being used. Previous reports describe compliance with nCPAP mainly in terms of the number of patients with GSA able to use it beyond their initial trial night (agree to use a machine at home) or those continuing use after some home therapy. From 15 to 300/0 of GSA patients tested with nCPAP in the sleeplaboratory either cannot tolerate it on the initial night or refuse to take it home as chronic therapy (9-14). At follow-up ranging from 2 to 24 months, 17 to 33% of those taking the machines home havediscontinued use or are using it only intermittently (9-14). With the exception of one study (15), most data on compliance are retrospective, limited to single point-in-time surveys collected from questionnaires that depend on subjective reports of compliance, and omit data on those not returning surveys. Furthermore, the methods of educating GSA patients and acclimating them to chronic home nCPAP use vary widely and active intervention to promote compliance is nonuniform or nonexistent. Since there may be a difference be936
SUMMARY Previous reports havedescribed compliance with nasal continuous positive airway pressure (nCPAP) for the treatment of obstructive sleep apnea (OSA) only in terms of the number of patients able to use it beyond their initial trial night or those continuing after some home use. Because of a possible difference between the level of compliance (mean number of hours of use per 24 h) needed for symptomatic relief of OSA versus cardiovascular Improvement, the level of hourly compliance in chronic nCPAP users may be important. The first part of this study prospectively examines compliance In a stable population of OSA patients already using nCPAP for 6 months to 2 yr. The second part is a prospective randomized, crossover study examining the effect of weekly (three times) then monthly (twice) positive reinforcement on hourly compliance of new nCPAPusers for 3 months versus no reinforcement for 3 months. Positive reinforcement consist~d of telephone discussions with the patients about the severity or complications of OSA, benefits of nCPAp,and suggestions about minimizing side effects. Using self-assessment scales, each patient reported the perceived level of improvement from the untreated to the treated condition and the prevalence and severity of side effects from the nCPAPtherapy. The level of compliance in stable, chronic nCPAP users with OSA was 6.1 ± 2.2 h/24 h (n 9). For the new nCPAP users during the nonreinforced period, the mean compliance was 6.0 ± 2.8 h/24 h: that during the reinforcement period was 6.0 ± 2.7 h/24 h (NS). There was no significant correlation between perceived improvement in OSA symptoms or between the perceived side effects of nCPAPversus hourly compliance. Positive reinforcement by telephone does not Improve hourly compliance with chronic use of nCPAP for OSA.
=
AM REV RESPIR DIS 1991; 143:936-941
tween the hours per night of nCPAP use needed for symptomatic improvement from GSA versus that needed to allow improvement in hemodynamic abnormalities, such as systemicand pulmonary hypertension, we first thought that it was important to prospectively establish by objective criteria the level of hourly use (compliance) in a stable population of chronic home nCPAP users. Second, we were interested in whether maintenance of communication, attention, and a show of interest in the patient in a cost-effective manner improved chronic compliance with nCPAP therapy. We attempted to demonstrate this using a prospectivecrossover method of positive reinforcement by telephone versus no positive reinforcement, following a standardized program of optimal education and acclimation to the device before home use. Wehypothesized that nightly use of the nasal CPAP machine once it is in the home may vary greatly between patients and that hourly compliance may be improved by close support and follow-up from the clinic or physician prescribing the therapy. Final-
ly,we sought to determine whether nightly compliance to therapy may be related to improvement in symptoms or the development of nCPAP side effects. Methods Overview The study was divided into two parts. First, describing compliance as the number of hours used per night, we prospectively examined compliance in a stable population of OSA
(Received in original form February 6, 1990 and in revised form December 11, 1990) 1 From the Department of Medicine, Pulmonary Disease Section, Veterans Administration Medical Center, Baylor College of Medicine, Houston, Texas. 2 Supported in part by an equipment grant from Respironics Inc., Pittsburgh, Pennsylvania, and by the General Research Service of the Department of Veterans Affairs. 3 Correspondence and requests for reprints should be addressed to Eugene C. Fletcher, M.D., Associate Professor of Medicine, Pulmonary Disease Section 111i, Veterans Administration Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030.
RATE OF DESATURATION DURING APNEA
patients actively using nCPAP for 6 months to 2 yr. Each subject was issued a new machine (Sleep Easy® II or III; Respironics Inc., Monroe, PA) with a Hobbs meter. This meter records the time that the machine is in operation (but not necessarily worn by the patient). Meter readings were recorded at the time of issue and monthly thereafter (by telephone report from patient) for 6 to 9 months. The subjects were told that the reason for switching machines was to update to a later model, and no active intervention to alter compliance occurred. The second part of the study was a prospective, randomized, crossover study examining the effect of positive reinforcement on nightly compliance of new nCPAP users. Following polysomnographic diagnosis of GSA, the subject slept a second night during which increasing levels of pressure were applied, determining the optimal nCPAP pressure to minimize apnea frequency. The patient was subsequently hospitalized one to three nights during which he or she slept with the nCPAP device at the predetermined optimal pressure and with the best fitting nasal mask to minimize leaks and mask discomfort. During this short period of hospitalization, each subject viewed a videotape produced by Respironics about GSA and use of the nCPAP machine. They were given instructions on the care and use of the machine by a nurse familiar with its operation. They wererequired to demonstrate proficiency with use and cleaning of the device and reading of the Hobb's meter before being discharged. In an alternating fashion new patients were assigned to one oftwo groups: telephone reinforcement about GSA and nCPAP use during the first 3 months (weekly for 3 wk then monthly for 2 months) or a control group called monthly but asked only for the hour reading on their machine's Hobb's meter and to report side effects of the therapy and perceived level of improvement from pretherapy symptomatology. If the subject reported side effects or problems with the equipment, suggestions weremade to correct these problems. Unscheduled calls from patients reporting side effects, problems with equipment, or to ask questions were accepted during all portions of the study. Suggestions by the study nurses for solving problems and answers to patient questions were offered throughout the study. During the course of the study patients brought their machines to the clinic one or more times to have specificproblems checked, mask or spacer size changed, and filters or valves changed. At that time meter readings were checked and back-calculation used to verify use times based upon meter readings reported over the telephone. This was done to detect intentional or unintentional misreadings of the meter. At the end of the first 3 months, reinforcement was used on the group who initially receivednone and stopped in those initially reinforced. After all subjects completed the reinforced and nonreinforced periods, they were followed an additional 3 months to confirm persistence of compliance.
937
Positive reinforcement consisted of telephone discussions with the patients about the severity of the disease, complications of apnea, the benefits of nCPAP, and suggestions about minimizing side effects. The content of the following paragraph was communicated in its entirety in a discussion format that was intended to be informative while conveying some urgency and working all elements of the paragraph into the conversation. The purpose of this telephone discussion was to convey a sense of interest in the patient's welfare as it relates to treatment of his or her condition through use of the nCPAP machine.
from zero (none), 1 (occasional), and 2 (moderate or tolerable) to 3 (severeor intolerable) and summed. The nine side effects were nasal stuffiness, dry mouth, difficulty breathing, mask discomfort, headache, machine noise, dry eyes, frequent awakening, and travel restriction. The maximum possible side effect score for each patient was 27.
Statistical Methods Mean hourly use per night was calculated from the meter readings in the stable nCPAP users and in the new patients during periods of positive versus no reinforcement. Compliance in stable users was compared to new users by the t test for unpaired data, and between reinforced and nonreinforced periods in new users by the t test for paired data. Hourly compliance in new users between reinforced, nonreinforced, and the 6- to 9-month follow-up period was compared using one-way analysis of variance. Correlation between baseline apnea severity (apnea index, mean level of desaturation, and mean apnea duration), level of nCPAP pressure required, baseline symptomatology, perceived improvement, and perceived side effects was examined for correlation with hourly compliance using leastsquares regression. The same variables were analyzed by least-squares regression with stepwise model selection to examine their relative usefulnessas combined predictors of compliance (16).The null hypothesis was rejected at p < 0.05. The protocol was approved by the Human Studies Committee of this institution, and all subjects signed informed consent before participation.
Mr.lMrs. , you have a very severe form of sleep apnea. You stop breathing during the night for periods up to 45 s or a minute and you do this as frequently as 50 to 100times per hour. The lowering of oxygen in your bloodstream may have bad effects on your heart and brain. It may cause heart failure and fluid in the legs. In rare cases patients with sleep apnea, like yourself, may die in their sleep. It is very important for you to use your CPAP at all times that you are sleeping. When you do not use the CPAP, you have apneas even though you may not know it. The only time you are free of apneas is when you are using the machine. If you haveproblems with the machine or discomfort from the mask, we are here to try to help you so that you use the machine as much as possible. Call us if you need help with the machine or if you haveany questions. Remember, use the machine every time you lie down to go to sleep, even for afternoon or evening naps.
Symptomatology and Side Effects
Results
Before using nCPAP each new subject responded to 25 questions regarding the level of GSA symptomatology in various areas of daytime and nocturnal function (appendix 1). The patient rated himself or herself as zero (no symptoms) to 3 (maximal symptoms) on each of the 25 questions. The maximum score was divided by 25 to determine the level of baseline symptoms. Following the first week of nCPAP therapy and at each follow-up call, using self-assessment scales, each patient reported the perceived level of improvement from the untreated to the treated condition. Improvement scores ranged from zero (no improvement) to 3 (much improvement). The perceived level of nine side effects was scored
A total of 20 patients were enrolled in the study: 8 males and 1 female in the stable, chronic nCPAP user group and 11 males in the new user intervention group. A single male subject in the intervention group moved from the area shortly after taking his nCPAP machine home andno data wereavailable. Thus data included in the crossover portion of this study are based upon the remaining 10 subjects. The mean age of the patients in the stable chronic group was 60, ideal body weight 155070, apnea index 55 h:', mean arterial desaturation 74070, and
TABLE 1 AGE AND BASELINE APNEA CHARACTERISTICS IN STUDY SUBJECTS Stable. Chronic User Group,
n
= 9
(Mean ± SO) Age IBW,% Apnea index, apneas/h in bed Average low saturation, % Average duration, s
59.6 155 55 74 26
± ± ± ± ±
6.1 34 20 9 11
New User, With or Without Reinforcement Group, n =7 (Mean ± SO)
51.6 158 49 65 27
± ± ± ± ±
12.8 20 26 12 8
P NS NS NS NS NS
FLETCHER AND LUCKETT
938 TABLE 2 COMPLIANCE, PERCEIVED IMPROVEMENT, AND PERCEIVED EFFECTS IN NEW NCPAP USERS Reinforced Period
Nonreinforced Period
Mean
SD
Mean
SD
P
Hourly use, h/night
5.95
2.74
6.04
2.n
NS
Perceived improvement* (possible score 0-3)
2.49
0.65
2.42
0.63
NS
Perceived side effectst (possible score 0-27)
3.37
1.59
2.76
2.22
NS
• Perceived improvement evaluates the patient's perception of how much relief the n~PAP p~ovides from symptoms before its use. A score of zero implies no relief and a score of 3 ma~(Imal. rehef. t Perceived side effects evaluates the patient's perceived level of problems from nine Side effects of nCPAP listed in the text. Zero represents no side effects and a SUbject with the worst possible level of all nine side effects could score 27 points.
mean apnea duration 26 s (table 1). The intervention group did not vary from the chronic group in any of these variables. The level of compliance in the stable chronic nCPAP users was 6.1 ± 2.2 hi 24 h. For the new nCPAP users (n = 10) the mean compliance was 6.0 h/24 h for both the reinforced and nonreinforced periods (table 2 and figure Ij.During the first 3 months of the study there was no difference in compliance between those patients who receivedreinforcement (6.58 ± 1.8 h/24 h; n = 6) and those not reinforced (5.55 ± 3.9; n = 4). During the second 3 months of the study there was no difference in compliance between those receiving reinforcement (5.0 ± 3.9; n = 4) and those no longer receivingreinforcement (6.37 ± 2.1; n = 5). Compliance at 6 to 9 months (follow-up period) for the new nCPAP users was 6.55 ± 3.50 h/24 h. None of the nightly compliance figures varied significantly from the others. There was no difference in the level of perceived improvement or level of perceived side effects between reinforced and nonreinforcement periods. There was no significant correlation between the level of baseline symptoms, perceivedimprovement, perceivedside ef-
fects, apnea index, mean desaturation level, mean apnea duration, or nCPAP pressure required for optimal treatment of apneas versus the level of compliance (hourly usage) during either the reinforced or nonreinforced period (table 3). Least-squares regression with stepwise model selection did not show any related variables that together correlated with hourly compliance. The correlation between baseline symptoms and level of perceived improvement was 0.62 (p < 0.05). No patient failed to use the machine at least part-time, but four patients did not use it on a nightly basis. Patient 4 (figure 1), with the lowest hourly compliance (0.6 ± 1.1 h/24 h), used the machine only intermittently and stopped using it altogether during the 6- to 9month follow-up period. Patient 8 (2.4 ± 1.8 h/24 h) used the machine on week nights only. Patient 6 (3.9 ± 1.4 h/24 h) used the machine intermittently depending upon symptoms but lost 38 pounds by the end of the 6- to 9-month followup period and discontinued use of nCPAP. Patient number 3, who used the machine intermittently (5.3 ± 1.1 hi 24 h), expired from a probable coronary
12 'C
Q)
III
::::l Q.
:
-
8
(J
c
.c
Cl
Z
4
o
•
•
Reinforced
NonReinforced
..... 4
6-9 Month Follow-up
Fig. 1. Hourly compliance for all 10 subjects completing the "positive reinforcernenr (left), "no reinforcemenf (middle), and 6- to 9-month follCMI-uP period (right) of nCPAP therapy. Individual subjects are denoted by the number to the right of the follCMI-up data. The four individuals with the lowest compliance were intermittent users. See text for explanation.
event (during daytime hours) near the end of the 6- to 9-month follow-up period. All other patients used the machine on a nightly basis, but some slept with the machine for less than the total number of hours in bed. Two patients reported occasional mask removal during sleep that was usually associated with arousal and replacement of the mask. The correlation coefficient between hours used calculated from weekly and monthly telephone meter readings by the patients (n = 19)and that taken directly from the Hobb's meter at the time of distribution and return was 0.990 (a = 0.27, b = 1.07; p < 0.0001). Discussion
Previous reports on compliance with home nCPAP in OSA patients have been mainly retrospective and subjective and involve no standardized program of introduction, education, or acclimation to the device. Furthermore, no studies have been interventional, trying to improve compliance by reinforcement. There are three obvious factors that could affect nCPAP compliance: (1) motivation, which is tied to severity of symptoms and satisfaction with the mode of therapy; (2) perceived complications or discomfort from use of the device; and (3) degree of education about and understanding of the pathophysiology of the OSA and the importance of treating it. The present study was designed to accumulate objective data on hourly compliance in a stable population of chronic nCPAP users. It was further designed to address compliance factors by examining the level of symptoms and perceived improvement by examining the degree of side effects (to include systematic treatment of discomfort and side effects) and by maximizing education, acclimation, and reinforcement of use. Our results reveal that hourly use in OSA patients who just acquired nCPAP did not improve over that of a stable population of users in whom no organized attempt at education, acclimation, or positive reinforcement occurred. Compliance did not correlate with the level of baseline symptoms, perceived improvement, level of side effects, parameters of apnea severity, or level of nCPAP pressure required to alleviate apnea. Only one study to date has prospectively collected objective data on chronic compliance with nCPAP. Krieger and Kurtz (15) evaluated 45 subjects every 3 months (range < 1 to 40 months). The mean rate of use was 5.14 ± 0.31 h/day,
RATE OF DESATURATION DURING APNEA
939 TABLE 3
CORRELATIONS BETWEEN SYMPTOMS, PERCEIVED IMPROVEMENT OR SIDE EFFECTS, AND THE LEVEL OF NIGHTLY COMPLIANCE Nonreinforced Period
Reinforced Period
Baseline symptoms Perceived improvement Perceived side effects Apnea index Required nCPAP pressure
R*
P
R
P
0.23 0.51 -0.21 -0.38 0.17
NS NS NS NS NS
0.34 0.53 -0.33 -0.35 0.00
NS NS NS NS NS
• Coefficient of correlation.
which is not very different from the chronic compliance rate of 6.1 h/night in the current "stable users" group. Regarding reasons for compliance versus noncompliance, the current study agrees with some findings of other studies. For example, Waldhorn and colleagues (14) found no correlation between severity of the apnea (based upon pretreatment apnea/hypopnea index), frequency of adverse reactions during therapy, or level of nCPAP pressure prescribed versus therapeutic noncompliance (in this case discontinuation of nCPAP after home placement). They did think there was a higher level of symptoms in the compliant versus noncompliant group, but symptoms were relieved in both groups to an equal degree. Our results may vary with the last point because no patient with only minimal symptoms or without daytime hypersomnolence was included in our study. The mean score on baseline symptomaticity was 1.7 of a maximal score of 3.0, whereas any score above 1.0 in our laboratory implies moderate symptomaticity. Other studies of nCPAP have reported initial noncompliance rates (refused to take the machine home) of 15 to 25070 and follow-up discontinuation after home use of 18to 30% (9-14). The present study does not examine either of these figures because 9 of 10 of the new users continued to use the machine (but with variable hourly compliance) at 9 months of follow-up and the stable, chronic user group were all currently using the device at the time of recruitment. One criticism of this study could be that the time period of follow-up may have been too short and that if followed 12months or longer, compliance would have decreased. Schweitzer and coworkers (10), having surveyed 96 patients (use range 2 to 19 months) who agreed to home nCPAP therapy, found that most patients who will stop therapy do so within the first 3 to 4 months. That compliance at 9
months was not significantly different from that at 3 and 6 months in our subjects agrees with the findings of Schweitzer. Indeed, the study by Krieger and Kurtz (16) seems to show an overall increase in hourly nCPAP compliance with time, which they attribute to "progressive adaptation to the constraints of CPAP therapy" or to improved nCPAP devices. Another criticism could be that recruitment into a study could induce a nonspecific motivational effect related solely to a change in patient condition: being selected to be in a special (research) group, the so-called Hawthorne effect. Since it is necessary to meet human study standards of informed consent, it is impossible to eliminate the nonspecific effect of selection to a special group. However, the Hawthorne effect cannot be maintained over time without a continued change in conditions. We believe that had the Hawthorne effect altered this study a decrease in compliance would have been noted in the 6- to 9-month follow-up, and this clearly was not the case. Previous studies have shown that tracheostomy improves symptoms as well as hemodynamic abnormalities, that is, pulmonary and systemic hypertension. In a study performed by Motta and colleagues six patients with symptomatic OSA received tracheostomies because of severesymptoms and the presence ofpulmonary and systemic hypertension (17). Within weeks after this procedure, symptoms disappeared, daytime Pa02 increased from 38 to 71 mm Hg, and substantial reductions were noted in the mean pulmonary artery and systemic pressures. Nasal CPAP has been shown to markedly improve the hypersomnolence and other central nervous system symptoms of OSA and to correct chronic respiratory failure (6). Unlike tracheostomy, however, nCPAP has never been shown to reverse the chronic hemody-
namic abnormalities of OSA. Both modalities are capable of eliminating apnea, which would theoretically eliminate the hypoxic stimulus to systemic and pulmonary hypertension. However, tracheostomy patients are almost by definition 100% compliant since the apparatus for eliminating the apneas is always in place. Many nCPAP patients, on the other hand, may find the apparatus uncomfortable, for whatever reason, and remove it during the night or not wear it every night. They may titrate their usage of the machine to eliminate symptoms of hypersomnolence but not use it enough to completely eliminate apneas. The outcome of this type of use could be improved symptoms but a continued hypoxic stimulus provoking hemodynamic abnormalities. For this reason nightly and hourly compliance may be an important question in nCPAP therapy. It appears from the results of the present study that the degree of symptomaticity, the level of side effects, or the degree of improvement does not individually predict successful compliance with nCPAP therapy. Likewise, pre-prescription education and acclimation to the machine does not raise the level of compliance above that achieved in a stable population who have accepted nCPAP therapy for a minimum of 6 months without a specific educational or acclimation program. A practical method of positive reinforcement (by telephone) does not appear to alter compliance. This does not suggest that some minimum levelof education, acclimation, reinforcement, and side effect solving is unnecessary to achieve a moderate level of compliance. It does suggest that no area deserves particular emphasis over another and that those patients who comply do so because of innate motivational factors that we have yet to identify. Future studies should combine objective evidence of nCPAP compliance with objective evidence of symptom improvement, such as the multiple sleep latency test, and blood gas and hemodynamic improvement. The results of this study have alerted us to the great variation in hourly use of nCPAP among patients who take a machine home and continue to report use of the machine during clinical follow-up. We do not monitor hourly use in all patients taking nCPAP machines home since we believe that resolution of symptoms remains an excellent indicator of machine use. However, we monitor nightly and hourly compliance in patients in whom cure of apnea is essential for treat-
FLETCHER AND WCKETT
940
ment of abnormal hemodynamics, such as patients with concomitant chronic obstructive pulmonary disease or daytime alveolar hypoventilation with prior episodes of right heart failure. We believe that reasonable compliance should be established in such cases and perhaps more aggressive therapy (tracheostomy) used in those whose hourly use is low because of discomfort or neglect and in whom cure of apnea may be essential for improvement of pulmonary hypertension and blood gases (18). One further application for such monitoring is in the OSA patient who continues to complain of somnolence while claiming to be using his or her machine. References 1. Sullivan CE, Berthon-lones M, Issa FG, Eves L. Reversal of obstructive sleep apnea by continuous positive airway pressure applied through the nares. Lancet 1981; 1:862-5. 2. Mahadevia AK, Onal E, Lopata M. Effects of expiratory positiveairway pressure on sleep-induced respiratory abnormalities in patients with hypersomnia-sleepapnea syndrome. Am RevRespir Dis 1983; 128:708-11.
3. Rapoport DM, Garay SM, Goldring RM. Nasal CPAP in obstructive sleep apnea: mechanisms of action. Bull Eur Physiopathol Respir 1983; 19:616-20. 4. Remmers lE, Sterling lA, Thorarinsson B, Kuna S. Nasal airway positive pressure in patients with occlusive sleep apnea. Am Rev Respir Dis 1984; 130:1152-5. 5. Berthon-lones M, Sullivan CEo Time course of change in ventilatory response to CO 2 with longterm CPAP therapy for obstructive sleep apnea. Am Rev Respir Dis 1987; 135:144-7. 6. Sullivan CE, Berthon-lones M, Issa FG. Remission of severeobesity-hypoventilationsyndrome after short-term treatment during sleep with nasal continuous positiveairwaypressure. Am Rev Respir Dis 1982; 128:177-81. 7. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease. Ann Intern Med 1980; 93:391-8. 8. Report of the Medical ResearchCouncil Working Party. Long-term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981; 1:681-5. 9. Sanders MH, Gruendl CA, Rogers RM. Patient compliance with nasal CPAP therapy for sleep apnea. Chest 1986; 90:330-3. to. SchweitzerPK, Chambers GW, Birkenmeir N, Walsh lK. Nasal continuous positive airway pressure (CPAP) compliance at six, twelve and eighteen months. Sleep Res 1987; 16:186.
11. Baker lP, Rose V, Ware C. Obstructive sleep apnea: therapeutic compliance. Trans Am Clin Climatol Assoc 1987; 99:224-30. 12. Kribbs NB, Kline LR, Pack AI, Dinges DF. Intermittent nasal CPAP therapy and sleepdisordered respiration. Am Rev Respir Dis 1989; 140:A115. 13. Nino-Murcia G, McCann CC, Bliwise DL, Guilleminault C, Dement WC. Compliance and side effects in sleep apnea patients treated with nasal continuous positive airway pressure. West 1 Med 1989; 150:165-9. 14. Waldhorn RE, Merrick TW, Nguyen MC, O'Donnell AE, Sodaro 1, Potolicchio Sl. Longterm compliance with continuous positive airway pressure therapy of obstructive sleep apnea. Chest 1990; 97:33-8. 15. Krieger 1, Kurtz D. Objective measurement of compliance with nasal CPAP treatment for obstructive sleep apnea syndrome. Eur Respir 1 1988; 1:436-8. 16. SAS/STAT users guide. Edit. 6.03. Cary, NC: SAS Institute Inc., 1988; 818. 17. Motta 1, Guilleminault C, Schroeder rs, Dement WC. Tracheostomy and hemodynamic changes in sleep induced apnea. Ann Intern Med 1978; 89:454-8. 18. Fletcher EC, Schaaf rw, Miller 1, Fletcher 1. Long term cardiopulmonary sequelae in patients with sleep apnea and chronic lung disease. Am Rev Respir Dis 1987; 135:525-33.
941
RATE OF DESATURATION DURING APNEA
APPENDIX 1. QUESTIONNAIRE FOR SLEEP DISTURBANCES The purpose of this questionnaire is to determine if daytime sleepiness may be related to breathing disturbances during sleep. Please answer questions to the best of your ability. Ask the technician if you need help.
o
1
3
2
1. Do you fall asleep while watching TV or at the movies? Never Very rarely Occasionally
Frequently
Not applicable
_
2.
Do you fall asleep reading newspapers, books or magazines? Never Very rarely Occasionally Frequently
Not applicable
_
3.
Do you fall asleep while talking with someone in person or over the telephone? Never Very rarely Occasionally Frequently _
4.
Do you get extremely sleepy while driving or fall asleep at stoplights? Never Very rarely Occasionally Frequently
Not applicable
_
5.
Have you had an automobile accident or driven off the road because of sleepiness? Never One time Two times More than two times
Not applicable
_
6.
Do you now, or have you fallen asleep while at work? Never Very rarely Occasionally
Not applicable
_
Frequently
7.
Do you fall asleep at public gatherings (movies, sports events, plays, etc.)? Never Very rarely Occasionally Frequently
8.
Do you get sleepy whenever you are inactive or bored? Never Very rarely Occasionally
9.
Frequently
Not applicable _._ _ _
Do you ever fall asleep in situations where you particularly need to stay awake? Never Very rarely Occasionally Frequently _
10. Do you ever feel confused when you awaken from sleep? Never Very rarely Occasionally
Frequently
_
11. Do you ever wake up in the morning with headaches? Never Very rarely Occasionally
Frequently
_
12. Do you ever "wet the bed" during the night? Never Very rarely Occasionally
Frequently
_
13. Do you ever sleepwalk? Never Very rarely
Frequently
_
Occasionally
14. Have you performed simple tasks during the day that you later don't remember doing? Never Very rarely Occasionally Frequently _ 15. Do you have vivid dreams as you are going to sleep or waking up? Never Very rarely Occasionally Frequently
_
16. Do you snore or has someone told you that you snore loudly? Never Very rarely Occasionally Frequently
Don't know
17. Has your bed partner told you that you stop breathing during sleep? Never Very rarely Occasionally Frequently
Not applicable
_
18. Has your bed partner awakened you for fear that you were not breathing? Never Very rarely Occasionally Frequently
Not applicable
_
19. Is your sleep restless with frequent turning or wild movements? Never Very rarely Occasionally Frequently
Not applicable
_ _
_
20.
Have you ever struck your bed partner or injured yourself during sleep? Never Very rarely Occasionally Frequently
Not applicable
21.
Does you bed partner sleep in another room because you disturb their sleep? Never Very rarely Occasionally Always _
Not applicable
_
22.
Do you find that your mind is not working as quickly or that you are more forgetful? Never Very rarely Occasionally Frequently _
Not applicable
_
23.
Have other people told you that recently you are more irritable or short-tempered? Never Very rarely Occasionally Frequently _
Not applicable
_
24.
Do you have trouble "having sex" recently? Never Very rarely Occasionally
Frequently
_
No interest
25.
Do your ankles swell? Never Very rarely
Frequently
_
Not applicable
Occasionally
No opportunity _
_
