May - August 2004: 
Volume 17, Issue 2

Click on the image to download the Issue in PDF format.


Compliance with CPAP treatment of patients with Sleep Apnea Syndrome.
The objective of this study was to evaluate compliance with the use of the CPAP machine in a sample of Greek patients with SAHS, as well as to investigate which factors influence compliance in these patients. We studied 138 consecutive patients (114 men, 24 women, mean age±SD: 52±10 years) who were scheduled for a follow up visit in the last 3 months of the year 1999. The single one eligibility criterion was the presence of an integral time clock in the CPAP machine. The patients had the machine in their possession for 20 to 2,315 days (mean value: 379 days). The total time of CPAP use was 57 to 14,388 hours (mean value: 1733 hours), while daily CPAP use varied from 0.429 to 10.08 hours (mean: 5.54 h/day). Analysis of study data showed that 77.5% of the patients used the CPAP machine for more than 4 hours daily; adjusted compliance was 77%. Apnea/hypopnea index (AHI) was the only parameter that correlated with daily use (p<0.01) and adjusted compliance (p<0.03). An inverse significant correlation between adjusted compliance and duration of CPAP treatment was found. Women used the CPAP machine for a shorter daily time, showed lower adjusted compliance [64±27 vs. 80±28 (p<0.02)] and reported more problems with their partners (p=0.0014) compared to men. Conclusively, the studied sample of Greek SAHS patients shows a generally satisfactory compliance with CPAP therapy, which is similar to that reported in the literature. However, women adduce a variety of reasons to justify their low CPAP use, even though they have accepted this treatment. Pneumon 2004, 17(2):167-176.
Full text


Since the publication of the first relevant study in Lancet in 19811, continuous positive airway pressure (CPAP) therapy has been the standard treatment for sleep apnea/hypopnea syndrome (SAHS); this therapy is also often recommended for the management of increased upper airway resistance syndromes2 or hypoventilation syndromes during sleep. In recent years, the indications for CPAP use have expanded so as to include heart failure and Cheyne-Stokes respiration3.

In the case of SAHS, CPAP constitutes a "technical" solution to a mechanical problem. It's a highly effective therapy with minimal objective complications. However, therapeutic efficacy persists only for as long as treatment is being daily used; hence, the majority of patients need lifelong treatment. This is essentially the single drawback of CPAP therapy, since it requires patients to fully comply with the instructions they were given in sleep clinic. It is generally recognized that patients with chronic conditions show poor compliance with prescribed treatment. As regards chronic pulmonary diseases in particular, patient compliance with chronic oxygen therapy has been shown to range from 50 to 65%4; compliance with nebulizer therapy ranges from 50-57% and compliance with inhaler therapy from 48-67%6.

Despite high cost of CPAP therapy, net benefit for the patient, the society and health economics has been shown to be a multiple of cost7. Hence, it is extremely important to monitor compliance with CPAP therapy, which is also an indicator of the effectiveness of the activity of a sleep clinic.

In this study, we examined patient compliance with prescribed CPAP therapy in our sleep clinic, as well as factors that may influence treatment utilization. To our knowledge, this is the first study to report this kind of data from a Greek population.


One hundred thirty-eight consecutive patients receiving CPAP treatment, who visited our Sleep Clinic for their regular follow-up in the last three months of the year 1999, were enrolled in the study. An indispensable requirement was the presence of an integral time clock in the CPAP machine. One hundred thirty-six patients had sleep apnea syndrome and two patients suffered from increased upper airway resistance syndrome. Diagnosis had been previously made (38 days to 76 months earlier) with polysomnographic studies. All patients used the standard CPAP method, except from three patients who used auto-CPAP and another three patients who had switched to BiPAP due to intolerance of the high expiratory pressures of standard CPAP. None of the patients needed supplemental oxygen therapy. In accordance with the standard operating procedures of our Clinic, during pressure titration as well as in the following days efforts to solve emerging problems and train the patients were made. It is worthy of note that all scheduled patients were enrolled in the study.

A patient compliance questionnaire was developed based on the literature8 with the addition of questions relevant to our study objectives. This questionnaire includes items that investigate CPAP associated problems and symptoms. Fourteen items relate to the upper airways and the equipment applied to the airways, 10 items relate to the patient experience with CPAP and patient attitude towards CPAP therapy and its effect on the quality of sleep, 5 items deal with the effect of CPAP therapy on the relationship with the spouse and 5 items examine the duration of sleep and the extent of CPAP utilization (hours or days/month). Questionnaires were completed with YES or NO answers at every follow-up visit in the Sleep Clinic.

The actual daily use of the CPAP machine was calculated by dividing the number of hours of use recorded on the time clock by the number of days elapsed since the acquisition of the machine (equation 1). The accuracy of patient reported CPAP use was verified by subtracting the product of the number of days of a month by the number of hours of use (as reported) from the product of the actual hours of daily use by 30 (equation 2). Adjusted compliance rate (AC%), i.e. hours of sleep during which CPAP was used as a percentage of the total hours of sleep in a 24 hour period for the days of a month on which CPAP was used (as reported by the patients), was also calculated (equation 3).

Equations used:

1. Hours of daily use (HDU) = hours of use recorded on the clock/days of CPAP machine possession.

2. Report accuracy (RA) = HDU * 30 - [no. of days (/month) * hours of sleep].

3. Adjusted compliance (AC) rate = [HDU 30/days (/month) * hours of sleep] * 100.

The statistical package SPSS was used for the statistical analysis.



Table 1. Anthropometric characteristics of study patients and data from multiparametric studies

                                              Males Ν=114                Females Ν=24                    P

Age (years)                            50.31 ± 10.18                 58.96 ± 8.86                  <0.000

ΒΜΙ (kg/m2)                             34.68±5.45                    41.46±6.44                   <0.000

Marital status (M)                      105 (92%)                       21(87%)                      <0.01

Education (years)                     10.7±3.77                      7.75±3.49                    <0.003

Smoking                                   44 (38.6%)                       6 (25%)                         NS

Alcohol consumption                      27                                   0                           <0.000

Apnea/Hypopnea Index          62.36±25.29                  53.59±27.32                     NS

Apnea Index                           37.61±25.82                   22.3±19.09                   <0.003

Mean SaO2%                            90.0±4.41                   89,4,92±4.29                    NS

Minimal SaO2%                        62.9±14.04                    61.5±16.57                      NS

Desaturation index                    51.1±25.5                     52.6±26.39                      NS

ΒΜΙ = body mass index, Alcohol consumption = at least one unit (9 gr) daily, apnea/hypopnea index = number of apnea/hypopnea episodes per hour of sleep, desaturation index = number of events during which SaO2 decreased by at least 4% from baseline, per hour of sleep, M= married.


Significant differences in mean age, body mass index (BMI), level of education, alcohol consumption and apnea index between males and females were found (Table 1). Compliance data were analyzed both in total and by gender.

The time of CPAP machine possession varied from 20 to 2,315 days (mean±SD 562±505 days; median 379 days) (Table 2). The total hours of use ranged from 57 to 14,388 (mean±SD 2,911±2771; median 1733 hours), whereas hours of daily use (HDU) ranged from 0.429 to 10.08 (mean±SD 5.37±2.00; median 5.54 hours daily). According to patient reported data, 3 (2%) patients used CPAP for up to 10 days/month, 6 (4%) for 11-20 days/month, 9 (7%) for 21-25 days/month and 120 (87%) for more than 25 days/month. The calculations for HDU (irrespective of the number of days of CPAP use) indicated that 8 (5.8%) patients used the CPAP machine for 0-2 hours/day, 23 (16.6%) patients for 2.01-4 hours/day, 49 (35.5%) patients for 4.01-6 hours/day and 58 (42.0%) patients for more than 6.01 hours/day.

According to their reports, female patients used CPAP for more hours daily and more days per month compared to male patients; however, their actual time of CPAP use (HDU) was shorter than that of males. Still, this difference was not significant (Table 2). Interestingly though, males showed a significantly higher adjusted compliance rate compared to females (p<0.02).



Table 2. CPAP use data

                                               Total Ν=138             Males Ν=114           Females Ν=24               p

Days of CPAP possession         562±505                   530±485                   708±581                  NS

Hours of sleep/24 hours*         7.32±1,21                 7.24±1.10                 7.73±1.62                 NS

Hours of use/24 hours*            6.18±1,45                 6.69±1.36                 7.21±1.82                 NS

Days of use/month*                 28.67±3,91               28.47±4.19               29.58±2.04                NS

Time clock reading                 2,911± 2,771             2,799±2,756             3,473±2,842               NS

Hours of daily use (HDU)         5.37±2.00                 5.44±1.92                 5.02±2.37                 NS

Adjusted compliance (%)             77±28                       80±28                       64±27                  <0.02

CPAP pressure                        10.85±2.49               10.94±2.51               10.29±2.43                NS

*As reported by the patients.

Evaluation of report accuracy by subtracting the actual hours of daily use from patient reported number of hours of use showed that 43 patients gave accurate re ports, 14 patients reported fewer hours of use and 81 patients overestimated their use of CPAP therapy. Hence, the odds ratio (OR) of a patient reporting more than the actual hours of use is 1.42.

Linear regression analysis was used to determine whether age, BMI, laboratory parameters that define the syndrome and its severity, as well as the duration of therapy are associated with HDU and adjusted compliance. Only the apnea/hypopnea index showed a significant positive association with HDU (p<0.01) and AC (p<0.03). Duration of CPAP treatment had no effect on HDU (r=-.0134, p=0.128), but was inversely related to AC (r=-0.215, p=0.015). Duration of treatment, HDU and AC by gender are presented in Table 3. There was no difference between men and women in the duration of treatment, as assessed by chi-square test (Table 3).



Table 3. Hours of daily CPAP use (HDU) and adjusted compliance (AC) with treatment in relation to the duration of CPAP treatment (n: no. of patients)

                                                         Males                                                  Females

Duration of treatment (years)               n              HDU               AC(%)             n              HDU           AC (%)

Less than 1 year                 58        5.6±2.0       83±28                      9        5.2±2.6        73±39

1-2 years                             24        5.2±1.9       74±25                      5        6.3±2.1        74±11

2-3 years                             13        6.2±1.2       87±16                      4        3.5±2.6        47±26

3-4 years                             14        4.7±2.3       73±34                      3        6.0±1.2        64±10

4-5 years                               3        4.4±1.0       58±16                      1            2,6               37

5-6 years                               1            7.8             100                        2        3.6±1.1        57±24

6-7 years                               1            4.3              68                          -


We investigated patient reported problems that were associated with the use of mask and with air pressure. A commonly reported problem was redness of the face, which resolved during the day and pain on the nose as a result of pressure of the mask on this area. Ulceration on the root of the nose developed at the beginning of CPAP treatment in 25% of the patients, whereas healing with scarring occurred in 6%. Women complained more often about dryness of the nose or nasal catarrh compared to men. Furthermore, more women than men reported a feeling of breathlessness when the mask was on, eye irritation and headache during the next day. None of these complaints had any effect on HDU or AC.

Complaints about noise produced by the CPAP machine and difficulty getting back to sleep once woken up were more common among women than among men (p=0.03) (Table 5). Patients who said that they resented resuming CPAP in case they happened to get up during the night had significantly fewer HDU (p=0.00018) and lower AC% (p=0.0412) compared to those who would resume CPAP.



Table 4. Common patient reported problems associated with the use of mask and symptoms during sleep with the use of CPAP or in the next morning

Problem-Symptom                                                    Total                    Males            Females           p

                                                                             YES        NO          YES (%)          YES (%)

The mask causes a feeling of breathlessness    6    (4%)    132               3                     12              0.03

The mask has caused sores on my nose          35  (25%)    103              25                    21               NS

Scars are permanent                                          7    (6%)    131               4                      8                NS

The mask irritates and leaves signs on my face 52 (38%)     86               36                    46               NS

The mask causes pain on the root of my nose 32  (23%)    106              20                    37               NS

My nose bleeds often                                        9    (7%)    129               7                      4                NS

My nose is very dry                                          31  (22%)    107              19                    37              0.05

My nose is congested                                      20  (14%)    118              12                    25               NS

I have nasal catarrh the next day                      17  (12%)    121              10                    25              0,03

I use a heated humidifier                                  17  (12%)    121              12                    12               NS

My eyes are irritated                                         23  (17%)    115              13                    33              0.01

My ears hurt                                                       6    (4%)    132               2                      8                NS

I have chest pain                                                5    (4%)    133               4                      0                NS

It brings me headache                                        6    (4%)    132               3                     12              0.03

It causes aerophagia                                          5    (4%)    133               3                      8                NS



Table 5. Problems related to patient sleep

                                                                                              Total              Males     Females         p

                                                                                      YES          NO      YES (%)   YES (%)

When I think that I have to use CPAP I get upset       29 (21%)     109           21             21             NS

The noise of the machine disturbs me                        22 (16%)     116           13             29            0.05

I have difficulty starting to sleep when I use CPAP     12  (8%)     126             9               8             NS

If I happen to wake up during the night, I have difficulty 25          (18%)     113             15             33            0.03

getting back to sleep

If I get up during the night, I’ m not willing                  33 (24%)     105           24             25             NS

to put the mask on again

I think I sleep worse when I use CPAP                         2  (1%)     136             0,9            4             NS

Symptom improvement palliates the inconvenience  136 (99%)         2           98           100             NS


An attempt to investigate the impact of CPAP treatment on the relationship of married patients (n=126) with their spouses was also made. The approach we employed and the results we obtained are shown in Table 6. Forty three per cent of women reported that the spouse slept worse when they used CPAP and 33% that the spouse was more disturbed by CPAP than by their snoring. Differences between female and male patient reports were significant (Table 6). Interestingly, patients who reported that their spouse reminded them to use CPAP had significantly fewer HDU (p=0.0049) and lower AC% (p=0.0040) compared to patients who were not reminded by their spouses. One-way analysis of variance (ANOVA) showed that marital status had no effect on HDU or AC%. Patients who indicated that CPAP had an unfavorable effect on their sex life and on the spouse's sleep were significantly older than patients who did not suffer such adverse CPAP effects (Table 7).



Table 6. Effect of CPAP treatment on the relationship of the patient with his/her partner

                                                                                                   Total              Males     Females      p

                                                                                                  N=126            N=105       N=21

                                                                                            YES          NO      YES (%)   YES (%)

CPAP has a negative effect on intimacy with my partner 27 (21%)       99          19             33          NS

CPAP has an unfavorable effect on my sex life               14 (11%)     112          10             19          NS

My partner complaints about worse sleep                        29 (13%)       97          19             43        0.012

My partner finds CPAP treatment more disturbing than snoring 16       (12%)    110               9          33          0.0014

My partner’s attitude discourages me from using CPAP    8   (7%)     118            6             10          NS

My partner reminds me that I have to use CPAP             76 (60%)       50          61             57          NS

CPAP pressure is too high and causes discomfort          18 (14%)     108          15             10          NS



Table 7. Mean age (±SD) of patients who gave a positive or negative response to the items concerning the effect of CPAP on their relationship with their partner

                                                                                                     Age (years)

                                                                                                YES                  NO                    p        

CPAP has a negative effect on intimacy with my partner        53±11              51±10                 NS

CPAP has an unfavorable effect on my sex life                      58±2                51±1               0.014

My partner complaints about worse sleep                               56±2                50±1               0.008

My partner finds CPAP treatment more disturbing than snoring 57±3             51±1               0.033



The first problem that emerges when CPAP is recommended as the treatment of choice to a SAHS patient is non-acceptance of the CPAP machine. The patient may reject CPAP when he/she uses the mask for the first time the night before CPAP titration or after the first night of CPAP use. Several studies have assessed non-acceptance and report rejection rates from 4.5%9 to 26.8%10. This parameter was not assessed in the present study, since our patients already possessed their own CPAP machine and had used it for 20 days to 77.2 months.

There is no unanimity on the definition of compliance and the best approach to its evaluation in the literature. The standard compliance measure is the mean machine run time (hours) in 24 hours11. Some studies examine compliance with the prescribed effective pressure12,13. Such studies are possible only if special CPAP machine models that allow recordings of time of use and mask pressure are used. In our study, the majority of the patients used standard CPAP machines from various manufacturers that had only an integral time clock, the presence of which was known to the patients. The prescribed effective pressure has been shown to be delivered for 89±3%14 or, according to other reports, for 94-98% of the hours of use recorded by the standard time clock15. According to their reports, our patients used the CPAP machine for 28.7±3.9 days per month (mean+ SD), whereas hours of daily use (HDU) in the total of patients were 5.4±2.0 (median 5.54). Our results are in agreement with available literature data. In the larger prospective study that included 1103 patients who accepted CPAP therapy, some of whom were followed up for 10 years, the median CPAP use was 5.7 hours/day9. In other studies, the mean daily use ranged for 4.7 to 5.7 days14,16. The higher mean daily use (6.6±2.2 hours/day) has been reported in a retrospective study conducted in France which included 3,225 patients, who constituted just 60% of the total of patients that were administered the study questionnaire.11 It is therefore likely that the majority of the participants in this survey were patients who used CPAP regularly.

It is recognized that SAHS has a negative impact on vital biologic functions with adverse effects on the cognitive, mental and physical health status of the patients. The association of the syndrome with high cardiovascular risk has already been established17. However, the minimum daily use of CPAP that improves symptoms and reduces cardiovascular risk has not been defined yet. Several studies investigate the relationship between compliance and symptoms improvement. Improvement in subjunctive sleepiness, cognitive function, mood and quality of life parameters (SF36 scale) in patients with mild SAHS and mean daily CPAP use of 2.8±2.1 hours was reported by Engleman et al18; Kingshott et al reported improved subjunctive sleepiness with 4.8±2.4 HDU19, whereas Bennett et al found no relationship between compliance (median 5, range 1.2-7.2 hours/day) and improvement in SF-36 scale20.

The minimum mean daily CPAP use that is defined as "satisfactory" varies greatly in different studies. There are centers that use the values of <1 hour15 or £2 hours of daily use10 to ask for the return of the CPAP machine. The most widely accepted threshold of "satisfactory" use is 4 hours.13 In the literature, however, there are studies that use a "satisfactory treatment" threshold of 2 hours/day10; others indicate a minimum effective use of 4 hours/day for 5 days/week (i.e. mean daily use 2.8 hours)20, or ³4 hours/day for ³5 days/week15, or ³5 hours/day22. Hence, the percentage of patients who comply with CPAP treatment, which is also a measure of the effectiveness of the activity of a sleep clinic, cannot be compared between centers. 77.5% of our patients used the CPAP machine for >4 hours/day, 86.9% for >3 hours/day and 94.2% for >2 hours/day. In the study by Popescu et al, the respective percentages of 128 patients at 1 year of CPAP use were 70.4%, 78.4% and 82.4%10.

On the other hand, adjusted compliance (AC%), i.e. the percentage of hours of sleep during which the patient uses the CPAP machine, allows more useful inferences. AC provides an individualized aspect of the problem and adjusts compliance to patient habits and needs. The authors that introduced the term reported a mean daily AC of 66% in their patients. In our study, the mean daily AC was 77±28%. Interestingly, we found a significant difference between men (80±28%) and women (64±27%) in adjusted compliance, although the duration of treatment was similar in these two groups. Duration of treatment was not found to have a significant effect on HDU, probably due to the small percentage of patients who used CPAP for more than three years; however, it did have a significant effect on AC.

Many investigators have tried to determine the factors that predict long-term use of CPAP. In a large study by McArdle et al9, factors that predicted long-term CPAP use included snoring, day-time sleepiness and AHI>30/hour, particularly in the presence of a high Epworth score (³10). The same study indicated that CPAP use of ³2 hours/day at 3 months reliably predicts long-term CPAP use. The risk of stopping CPAP therapy for patients with daily CPAP use of less than 2 hours is 12.7.

Studies of the factors that influence daily CPAP use have often produced conflicting results. Such factors that have been examined in various studies include subjunctive and objective sleepiness14, pre-treatment perception of general health status (SF-36 scale)20, respiratory function parameters15,16, and polysomnography parameters13-16. Apparently, general satisfaction from CPAP treatment10,21,22, as well as sleep habits12 have a key role. In our study, both HDU and AC were found to be positively associated with baseline AHI (p<0.01 and p<0.03, respectively), a finding similar to those of other investigators13,16,23.

One might have expected that complications associated with the use of the CPAP machine would have an inverse effect on compliance. In the majority of relevant reports, however, compliance appears to remain unaffected by complications. The incidence of complications and symptoms associated with the use of the CPAP machine in our study was similar to that reported in the literature and had no effect on HDU and AC. The 24 female patients who participated in our study reported a variety of problems, many of which were significantly more common in women than in men. Female patients were significantly older and had a higher BMI and lower educational level compared to male patients. Apnea-hypopnea index (AHI) was similar in both sexes, but females had a significantly lower mean apnea index. Although women reported that they used CPAP for more days per month and more hours per day compared to men, their adjusted compliance was significantly lower. Furthermore, reports of worsened spouse sleep and greater spouse disturbance by CPAP than by snoring were significantly more common among women than among men. Discrepancy between reported and actual HDU as significantly greater in women compared to men (p<0.01).

Apparently, women face important problems that have not been investigated in depth yet; these problems, however, lead to either non-acceptance of CPAP treatment from the very beginning9 or discontinuation of CPAP10 being more common in women than in men. Compliance is also worse among women. The reasons that lie behind providing inaccurate compliance data and using CPAP less than men do should be further investigated.

Undoubtedly, CPAP treatment is a major change in a couple's life. Many patients complain that the CPAP machine has alienated their partners, and others, mainly older patients, report that CPAP treatment had a negative effect on their sex life. The effect of the treatment on the relationship w