Abstract
Rationale: Obstructive sleep apnea (OSA) is a prevalent disease. Often limited clinical resources result in long patient waiting lists. Simpler validated methods of care are needed.
Objectives: To demonstrate that a nurse-led model of care can produce health outcomes in symptomatic moderate–severe OSA not inferior to physician-led care.
Methods: A randomized controlled multicenter noninferiority clinical trial was performed. Of 1,427 potentially eligible patients at 3 centers, 882 consented to the trial. Of these, 263 were excluded on the basis of clinical criteria. Of the remaining 619, 195 met home oximetry criteria for high-probability moderate–severe OSA and were randomized to 2 models of care: model A, the simplified model, using home autoadjusting positive airway pressure to set therapeutic continuous positive airway pressure (CPAP), with all care supervised by an experienced nurse; and model B, involving two laboratory polysomnograms to diagnose and treat OSA, with clinical care supervised by a sleep physician. The primary end point was change in Epworth Sleepiness Scale (ESS) score after 3 months of CPAP. Other outcome measures were collected.
Measurements and Main Results: For the primary outcome change in ESS score, nurse-led management was no worse than physician-led management (4.02 vs. 4.15; difference, −0.13; 95% confidence interval: −1.52, 1.25) given a prespecified noninferiority margin of −2 for the lower 95% confidence interval. There were also no differences between both groups in CPAP adherence at 3 months or other outcome measures. Within-trial costs were significantly less in model A.
Conclusions: A simplified nurse-led model of care has demonstrated noninferior results to physician-directed care in the management of symptomatic moderate–severe OSA, while being less costly.
Clinical trial registered with http://www.anzctr.org.au (ACTRN012605000064606)
Obstructive sleep apnea (OSA) is a prevalent disease. Often limited clinical resources result in long patient waiting lists. Simpler validated methods of care are needed.
A simplified nurse-led model of care has demonstrated noninferior results to physician-directed care in the management of symptomatic moderate–severe OSA, while being less costly.
In the early 1990s the prevalence of symptomatic OSA in developed countries as assessed by an apnea–hypopnea index (AHI) greater than 5/hour, plus chronic daytime sleepiness in adults, was found to be approximately 3–4% for men and 2% for women (1, 7). Although more recent prevalence studies in these populations are lacking, it is almost certain that the prevalence of OSA has increased in parallel with the rise in obesity (8–10). The prevalence of OSA in developing countries has been shown to be similar to the early estimates for developed countries and will likely climb rapidly with the increasing adoption of Western lifestyles (11).
In many countries when OSA is suspected the usual practice is to refer the patient to a specialist physician at a sleep center for clinical assessment and attended overnight, laboratory-based polysomnography (PSG). If OSA is confirmed the patient typically returns to the laboratory for another PSG to determine the therapeutic CPAP level although, if the AHI is high, the therapeutic CPAP level may be tested in the second half of the diagnostic study, that is, “split study.” The patient is subsequently reviewed by a physician and CPAP nurse or technician to commence the treatment. Even in developed countries this clinical pathway is often quite drawn out because of a relative scarcity of qualified personnel and laboratory facilities. In 2004 it was reported that the mean wait from referral to CPAP provision for patients with OSA in the United Kingdom was 14 months, in Canada 24 months, and in Australia 7–8 months (12). In developing countries such as China and India the situation is considerably worse, with services being sparse or nonexistent in most regions. Scarce health resources and a large and rising burden of disease mean that unless simpler, more cost-effective management strategies are developed, many patients with OSA will be denied effective treatment.
A number of strategies to assist in the earlier detection and treatment of OSA have been suggested but have either not been properly evaluated or widely accepted. Home diagnosis based on simplified portable respiratory monitors, some combined with automated scoring algorithms to reduce labor-intensive data analysis, has been suggested as an alternative to PSG. However, the quality of many of the studies designed to evaluate the accuracy of such devices has been brought into question (13). Few studies have tested the reliability of these monitors in the home and virtually no studies have compared patient outcomes when clinical decisions were based on the findings of portable monitors versus PSG. Home-based, autoadjusting methods of CPAP titration have been more thoroughly assessed with respect to patient outcomes and have generally been found to produce similar results to in-laboratory manual titration (14–17). The use of autoadjusting positive airway pressure (APAP) to titrate CPAP has, however, not been universally adopted. Many studies have excluded patients with comorbid illness, raising questions about the generalizability of results (15, 16). Setting CPAP pressure empirically, using a predictive equation, has also been reported but without long-term patient outcome follow-up data (17, 18). Ways of overcoming physician manpower shortages in sleep apnea services have not been systematically studied. A nurse-led model of care in OSA management has been described (19), but was not subjected to randomized controlled trial evaluation. In other chronic diseases such as heart failure, protocol-based specialist nurse care models have proved to be highly effective (20) and anecdotal evidence suggests that the skills of sleep nurses and technicians can be harnessed to markedly improve CPAP treatment outcomes for patients with OSA (21).
The present study was designed as a randomized controlled study in which a package of care incorporating each of these newer management strategies including simplified home diagnosis, CPAP titration with an autoadjusting positive airway pressure device to set a fixed CPAP pressure, and overall care supervised by a specialist nurse was compared with the more traditional physician-directed, in-laboratory PSG, hospital-based program of care (Figure 1). We used oximetry for home OSA diagnosis, as we had previously evaluated this and found it to have reasonably high diagnostic utility in “ruling in” moderate to severe OSA (22). Some of the results of these studies have been previously reported in the form of an abstract (22).
The study was conducted as a randomized controlled open-labeled noninferiority clinical trial at three separate academic sleep medicine services: Adelaide Institute for Sleep Health (Adelaide, South Australia, Australia), Alfred Hospital (Melbourne, Victoria, Australia), and John Hunter Hospital (Newcastle, New South Wales, Australia). The study was approved by the ethics committee at each site and registered with the Australian and New Zealand Clinical Trials Registry. The study was conducted independently of the sponsors (see the online supplement).
Patients referred with a clinical suspicion of OSA were interviewed to assess their eligibility for the trial. Inclusion criteria were (1) Epworth Sleepiness Scale (ESS) score of 8 or more, (2) history of snoring “most nights” or “every night,” (3) age 18–75 years, and (4) patient willing to trial CPAP. Exclusion criteria are provided in the online supplement. Patients were recruited between March 2004 and September 2006 and monitored for 3 months after randomization.
All patients who met the inclusion/exclusion criteria and consented to the study had overnight home oximetry (Masimo Radical oximeter; Masimo, Irvine, CA). Details of oximetry measurement are provided in the online supplement. If the patient had an oxygen saturation dip rate greater than 27/hour at the greater than 2% level they were randomized into one of the two arms of clinical care. This dip rate cut-point was established in a preliminary study (22) to give the highest diagnostic sensitivity for moderate to severe OSA (PSG-determined AHI ≥ 30) while keeping the false positive rate at 10% or less. This corresponded to a positive predictive value for the test of 92% and positive likelihood ratio of 9.3. Those who did not meet this entry criterion were returned to their original place in the sleep medicine physician waiting list and continued on the usual sleep medicine diagnostic and therapeutic pathway for their particular clinic.
Model A (the simplified nurse-led model of care) was supervised by a specialist nurse experienced in sleep disorders and the management of patients receiving CPAP. The nurses involved in the trial had worked in the field of sleep CPAP therapy for a mean of 8.3 years (range, 1–15 yr). Home autotitrating CPAP (Autoset T; ResMed, Sydney, Australia) was used for four consecutive nights in the patient's home. The autotitrating mode was set between 4 and 20 cm H2O. The CPAP machine computerized data were reviewed by the specialist nurse and the median 95th centile CPAP pressure for the 4 days was recorded. This was determined a priori to be the appropriate fixed CPAP pressure for model A. The nurse then converted the patient to a fixed CPAP device (S6 Elite lightweight; ResMed), with in-built compliance meter and optional humidifier, and set the pressure accordingly. The nurse dealt with CPAP complications as appropriate (e.g., initiated humidification or nasal steroids, or changed masks) under protocol (see the online supplement). Patients who failed or refused CPAP during the trial period were referred to a sleep specialist for further advice and initiation of alternative treatments as appropriate. The patients were seen routinely at setup and again at 1 and 3 months. Phone consultation and extra reviews were possible at the discretion of the nurse.
Model B (traditional physician-directed care) consisted of full laboratory PSG to confirm the diagnosis of OSA and to identify any additional sleep disorder (e.g., periodic limb movements of sleep). CPAP was the primary treatment recommendation and manual laboratory CPAP titration during PSG was undertaken the night after the diagnostic PSG. CPAP titration in all centers followed the same protocol of manual technician-observed titration of CPAP pressure to abolish snoring, oxygen desaturation, and apneas and hypopneas. Sleep specialists (12 overall and at least 3 in each center) supervised and reported the PSGs. All patients were seen in a full-length initial consultation by the specialist physician, who reviewed each patient's history, physical findings, and PSG reports and prescribed a CPAP pressure. The frequency and timing of physician follow-up visits after CPAP initiation were decided by the responsible physician. There was no restriction placed on the physicians and they were free to diagnose and treat comorbidities during the trial (e.g., PSG-diagnosed periodic limb movements of sleep) whereas the nurse in model A was restricted to protocol-based management. Usual nursing CPAP support was offered in model B (independent of the model A nurse) to set patients up on CPAP after their CPAP PSG titration, and patients were reviewed at 1 and 3 months.
PSGs were conducted using Compumedics E Series equipment (Melbourne, Australia) at the Adelaide and Melbourne sites and Sensormedics equipment (Sydney, Australia) at the Newcastle site, with data transferred to an EDF file to enable conversion to a Compumedics file for review. Details of PSG measurements and sleep study scoring are provided in the online supplement (23, 24).
In both arms of care extra nursing consultations and phone advice were possible. The time to complete these extra consults was recorded. The same fixed-pressure CPAP machine was used in models A and B. Additional investigations, treatments, and advice for OSA or other coexisting sleep disorders were recommended at the discretion of the treating specialist. The investigation and management of these two arms of care were conducted over equal time periods.
The primary end point was the change in ESS score as measured before and after 3 months of CPAP therapy (25, 26). Secondary outcome measures were as follows: changes in Short-form 36 Health Survey (SF-36) (27–29), Functional Outcomes of Sleep Questionnaire (FOSQ) (30), and executive neurocognitive function (number of maze trials successfully completed in allotted time and total errors made); objective CPAP adherence; Maintenance of Wakefulness Test (MWT) after 3 months of therapy (31); and general patient satisfaction with investigation and treatment (Visit-specific Satisfaction Questionnaire [VSQ-9]) (32). Full details of these measurements are provided in the online supplement.
Within-study resource use and costs over the 3-month follow-up period from randomization were estimated at a patient level in each arm, based on observed use of tests and time for visits in home and hospital settings. The 2008 prices for relevant resource use (Medical Benefit Schedule item for tests and health care worker time) were applied to estimate 2008 costs.
The study was powered to demonstrate noninferiority of nurse-led management compared with specialist-directed care with respect to change in ESS, the primary outcome measure (33). Details of the sample size calculation, method of randomization, allocation concealment, and blinding are provided in the online supplement.
Statistical analysis for primary and secondary clinical end points compared change in scores at 3 months between groups, using an independent samples t test. The lower limit of the two-sided 95% confidence intervals was used to determine noninferiority. Bootstrapping (resampling with replacement) on patients and their associated costs and effect (ESS score) was used to estimate 95% confidence intervals for incremental cost, and the joint distribution of incremental cost and effects (ESS score), allowing for covariance between costs and effects (34). Data were analyzed using intention to treat principles, given patients' assignment and observed compliance. Approximately equal proportions of patients dropped out in each arm of the trial: 10 of 100 (10%) in model A, and 11 of 95 (12%) in model B. The reasons for the dropouts are given in Table 1.
Model A (Nurse Led) | Model B (Specialist Led) | |
|---|---|---|
| Unwilling to continue with CPAP | 5 | 6 |
| Depressive illness | 1 | 1 |
| Reason not specified | 4 | 4 |
The patients' pathway through the clinical trial is summarized in Figure 2. Once patients were contacted and gave informed consent, 195 of 882 met inclusion and exclusion criteria, with 263 excluded on the basis of patient characteristics and, of the remaining 619, 424 were excluded on the basis of not having a greater than 2% oximetry dip rate of more than 27 dips/hour. All patients who were excluded were returned to their same position in the sleep clinic waiting list to be reviewed at a later date.
Figure 1. Design of clinical trial. APAP = autoadjusting positive airway pressure; CPAP = continuous positive airway pressure; ESS = Epworth Sleepiness Scale; FOSQ = Functional Outcomes of Sleep Questionnaire; MWT = Maintenance of Wakefulness Test; PSG = polysomnography; R/V = review; SF-36 = Short-form 36 Health Survey; VSQ-9 = Visit-specific Satisfaction Questionnaire.
[More] [Minimize]
One hundred and ninety-five patients were randomized to the two arms of care (104 in Adelaide, 54 in Newcastle, and 37 in Melbourne). Key anthropometric variables are summarized in Table 2. These results confirm that the characteristics of the patient population were typical of those with moderate–severe OSA and that the two groups were similar at baseline with respect to key anthropometric variables.
Model A (Nurse Led) (n = 100) | Model B (Specialist Led) (n = 95) | |
|---|---|---|
| ESS, mean ± SEM | 13.7 ± 0.4 | 13.4 ± 0.4 |
| BMI, mean ± SEM | 35.1 ± 0.7 | 34.0 ± 0.6 |
| >2% oxygen saturation dips/h, mean ± SEM | 49.2 ± 2.1 | 52.5 ± 2.7 |
| Apnea–hypopnea index (events/h of sleep), mean ± SEM | NA | 67.9 ± 2.82 |
| Age (yr), mean ± SEM | 49.9 ± 1.2 | 50.3 ± 1.3 |
| Neck circumference (cm), mean ± SEM | 44.1 ± 0.4 | 44.0 ± 0.4 |
| Males, % | 72 | 76 |
| Females, % | 28 | 24 |
If nurse-led management (model A) was not inferior to specialist-led management (model B), then we might expect the mean change in ESS score to be identical in both groups, or at least to be within close proximity to each other. We prespecified a difference of −2 on the change in ESS score between groups to be the furthest away from zero that would still allow us to declare model A to be noninferior, because a difference of this magnitude would not be deemed inferior from a clinical viewpoint. This difference is called the margin of noninferiority. From the results presented in Table 3, we found that the mean change in ESS score for nurse-led management (model A) was not inferior to the mean change in ESS score for specialist-led service (model B) because the lower limit of the two-sided 95% confidence interval for the mean difference did not include −2, the margin of noninferiority.
Model A (Nurse Led) | Model B (Physician Led) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | n | Mean | SEM | n | Mean | SEM | MD | SE (MD) | 95% CI (MD) | ||||
| ESS change | 90 | 4.02 | 0.52 | 84 | 4.15 | 0.47 | −0.13 | 0.70 | −1.52 to 1.25 | ||||
| FOSQ change | 89 | −13.61 | 2.02 | 81 | −13.22 | 1.96 | −0.38 | 2.83 | −5.97 to 5.20 | ||||
| SF-36 Vitality change | 89 | −16.12 | 2.17 | 81 | −15.31 | 2.06 | −0.81 | 3.01 | −6.75 to 5.12 | ||||
| SF-36 Mental Health change | 89 | −4.81 | 1.46 | 81 | −5.09 | 2.11 | 0.27 | 2.53 | −4.71 to 5.27 | ||||
| CPAP compliance | 94 | 4.11 | 0.28 | 83 | 4.56 | 0.30 | −0.45 | 0.41 | −1.26 to 0.36 | ||||
| MWT, min | 65 | 30.18 | 1.24 | 61 | 31.68 | 1.08 | −1.49 | 1.65 | −4.76 to 1.78 | ||||
| Executive maze change in trials successfully completed | 73 | −2.11 | 0.61 | 68 | −1.19 | 0.56 | −0.92 | 0.83 | −2.57 to 0.73 | ||||
| Executive maze change in errors made | 73 | −17.56 | 4.64 | 68 | −16.85 | 6.07 | −0.71 | 7.57 | −15.68 to 14.27 | ||||
There was no significant difference in mean sleep latency as measured by MWT protocol, providing confirmatory objective evidence that daytime sleepiness after CPAP was not different between the two groups.
There were no significant differences between model A and model B after 3 months of CPAP in any of the quality of life indices. These included total FOSQ score after treatment, and change in FOSQ score after therapy (post–pre). SF-36 scores were analyzed in a number of different ways (total score after 3 months of CPAP, and change in SF-36 score before and after 3 months of CPAP). Because the Mental Health and Vitality (energy) components of the SF-36 measurement have been shown to be the most responsive in previous CPAP trials in OSA (28, 29), these were analyzed separately and also were not significantly different between the two groups.
There were no significant differences between the two groups in the change from baseline to 3 months in the number of maze completions or errors (Table 3). Post-CPAP, the numbers of mazes completed in 8 minutes decreased by 2.1 and 1.2, respectively (18 and 12% reduction in models A and B, respectively), but this was accompanied by a marked reduction in the numbers of errors (33 and 27%, respectively).
There were no statistically significant differences in CPAP adherence at 3 months between model A and model B (Table 3). The mean adherence figures included data from eight patients (four in each group) who did not use CPAP during the 3 months of follow-up but returned for final assessment.
Two of these patients had a mandibular advancement splint made and one had a tonsillectomy to treat OSA. There was no significant difference in the CPAP pressure prescribed between the two groups (model A, 11.2 ± 0.2 cm H2O; model B, 10.4 ± 0.4 cm H2O; P = 0.06). Only six patients in model A and nine patients in model B had their CPAP pressure altered during the trial (three increased and three decreased in model A, four increased and five decreased in model B), and the alterations were never greater than 2 cm H2O. Only 2 of 95 patients had a central apnea index greater than 5/hour during the model B CPAP titration
The specialist nurse discussed 15 of 100 patients (15%) in model A with a sleep physician, leading to 12 of 100 (12%) being reviewed in person by a physician during the trial, 9 because of problematic nonadherence to therapy. One patient was reviewed because the patient expressed a wish to see a doctor, one because of problematic restless legs, and one because large tonsils were detected on clinical examination by the nurse. This patient went on to have a tonsillectomy instead of using CPAP long term. When considering total time spent (unscheduled and scheduled) CPAP nurses spent about 50 minutes extra with the patients in model A than in model B, but there was an average of 2.36 physician consultations per patient in model B compared with 0.18 in model A (Table 4). Additional diagnoses or treatments, if any, provided by physicians in model B were not systematically recorded. However, model B PSG recordings revealed no cases of Cheyne-Stokes breathing and during the CPAP titration PSG only 5 of 95 patients (5.3%) had a periodic limb movement arousal index exceeding 5/hour and 2 of 95 patients (2.1%) had a central apnea index greater than 5.
The bootstrapped joint sample distribution of within-study incremental costs and effect (primary outcome, ESS score) for model A relative to model B is shown in Figure 3. The ESS is not suggested to differ between models of care, whereas within-study nurse-led management (model A) was 1,111 Australian dollars (A$1,111) per patient (95% confidence interval: A$1,084, A$1,137) less expensive than specialist-led management (model B). Hence the within-study analysis suggests that a nurse-led model of care saves considerable resources without compromising effects in patients diagnosed by oximetry as having a high likelihood of moderate to severe obstructive sleep apnea and consequently is cost-effective in these patients.
Figure 3. Bootstrapped joint distribution of incremental costs (Australian dollars per patient) and effects (Epworth Sleepiness Scale [ESS] score per patient) for treatment model A versus model B.
[More] [Minimize]Uncertainty around within-study cost effectiveness of model A (nurse led) in comparison with model B (specialist physician led) requires joint consideration of incremental costs and effects (model A vs. model B) under uncertainty, and hence their joint distribution. Figure 3 shows the sampled joint distribution of incremental cost and effect from bootstrapping patient level data on costs and effects, with resampling of patients to retain covariance between cost and effects along treatment pathways (34). This joint distribution shows that model A is clearly less expensive than model B while having equivalent effect (reflecting the statistical test of ESS equivalence). Hence within-study evidence suggests that model A is a more cost-effective approach to the management of symptomatic moderate–severe OSA than model B.
In implementing strategies in practice rather than a trial setting, an additional incremental cost of model A relative to model B related to diagnostic oximetry (A$98.50 per test) may arise in patients who are referred and not excluded by other means. This potential additional cost per diagnosed patient depends on the prevalence of patients with moderate to severe disease (a >2% oximetry dip rate of >27) diagnosed in a screened population. In the trial population this prevalence was 195 of 619, which may lead to an additional cost per diagnosed patient of A$98.50 × 619/195 = A$313 for model A. Consequently, the incremental cost within trial of model A relative to model B could be reduced by A$313 from A$1,111 to A$798 in implementing strategy model A in the trial population in practice. More generally, in any given population with prevalence of moderate to severe patients r, the incremental cost may reduce by A$98.5 × 1/r. Hence a threshold prevalence at which the cost of models A and B in practice may be equivalent can be estimated by letting A$1,111 = A$98.5/r. That is, r = 98.5/1,111 = 8.9%. Thus, in shifting practice to include model A as part of routine care, overall costs can be expected to be lower as long as the prevalence of moderate to severe obstructive sleep apnea remains greater than 8.9%.
Total patient satisfaction with treatment as measured by the VSQ-9 was not statistically significantly different between the two groups (Table 5). Each of the nine questions in the VSQ-9 is scored individually, with the overall satisfaction with treatment being one of those nine questions. Four of the nine questions showed that patient satisfaction with model A was greater than with model B (P < 0.05), including time waited, explanation, information provided, and time spent with the health professional.
Time Satisfaction Waiting | Impression of Time Wait | Impression Ancillary Staff | Impression Personal Manner of HPs | Impression Competence of HPs | Satisfaction Time Spent with HPs | Adequate Explanation | Sufficient Information from HPs for Choices | Rate Overall Satisfaction with Treatment | |
|---|---|---|---|---|---|---|---|---|---|
| Model A | |||||||||
| Mean | 3.517 | 3.730 | 3.764 | 3.843 | 3.831 | 3.820 | 3.764 | 3.685 | 3.730 |
| SD | 0.676 | 0.45 | 0.43 | 0.40 | 0.38 | 0.39 | 0.45 | 0.49 | 0.47 |
| n | 89 | 89 | 89 | 89 | 89 | 89 | 89 | 89 | 89 |
| SEM | 0.072 | 0.05 | 0.05 | 0.04 | 0.04 | 0.04 | 0.05 | 0.05 | 0.05 |
| Model B | |||||||||
| Mean | 3.4810127 | 3.506 | 3.646 | 3.759 | 3.709 | 3.570 | 3.544 | 3.468 | 3.759 |
| SD | 0.5514726 | 0.53 | 0.48 | 0.43 | 0.53 | 0.59 | 0.59 | 0.60 | 0.43 |
| n | 79 | 79 | 79 | 79 | 79 | 79 | 79 | 79 | 79 |
| SEM | 0.0620455 | 0.06 | 0.05 | 0.05 | 0.06 | 0.07 | 0.07 | 0.07 | 0.05 |
| P Value* | 0.706 | 0.004 | 0.095 | 0.196 | 0.092 | 0.002 | 0.008 | 0.011 | 0.676 |
This study has compared a simplified package of care (incorporating nurse-led home diagnosis and CPAP therapy) for patients with moderate–severe OSA with physician-led current best practice in OSA management. The main finding of the study was that the simplified model of care was not inferior to the usual specialist sleep physician–led, hospital-based model with respect to our primary outcome measure, the mean change in ESS score after CPAP treatment. The lower limit of the confidence interval of the difference between the two models was well within the a priori determined range of clinical significance for ESS (i.e., 2.0).
There were also no significant differences between model A and B after 3 months of CPAP therapy across a range of other measures including objective sleepiness, general and disease-specific quality of life measures, neurocognition, patient satisfaction, and CPAP adherence after 3 months.
The present study findings complement and extend those of an earlier study by Mulgrew and colleagues, which compared a similar simplified model of care involving home diagnosis and commencement of CPAP with laboratory PSG and sleep clinic-based care (35). Mulgrew and colleagues found no differences in patient outcomes between the two groups (simplified vs. best OSA practice), but the study was not powered to show that the simplified model of care they used was either equivalent or noninferior to standard care (36). The present study extends these earlier findings by showing that skilled nurses, working with a high degree of independence, can lead a simplified, ambulatory program of management for high-probability cases of moderate–severe OSA and produce patient outcomes that are not inferior to standard specialist-led care. Our results also showed that costs were significantly less in model A within the trial, and that cost savings would likely accrue if model A was implemented more generally in practice, even in clinical settings where the prevalence of moderate–severe OSA is relatively low.
It is important to note that the simplified model of care employed in the present study was underpinned by (1) knowledge of the pretest probability of OSA in our referrals; (2) the expertise of experienced CPAP nursing staff in OSA management; (3) tertiary sleep laboratory backup, in terms of interpretation and quality control for oximetry data and APAP data; and (4) sleep physician input if needed.
We believed it important for the specialist nurse to be able to cross-consult under circumstances in which they were uncertain about the management of the patient. Twelve of 100 patients had a sleep physician review as a result of unsatisfactory progress in model A. Nine of these reviews were a once-only consultation. Although this represents a small percentage of all patients managed by the nurse, it underscores the need for this simplified model of care to be ideally conducted either within a tertiary sleep medicine service or with patient access to same. We do not recommend that the simplified management approach occur autonomously. How this occurs in a rural or remote setting needs consideration, but the opportunity for Internet data transfer and teleconferencing/consultation makes it possible for a tertiary sleep medicine service to oversee this model of care in a remote setting.
It is also important to note that the patients included in this study had been referred to a sleep medicine clinic for assessment of OSA. They were symptomatic (chronic snoring plus ESS ≥ 8), and thus the pretest probability of moderate–severe OSA was likely to be high. We found it to be 36% in a similar previous study in our center, which fits with reports from other sleep disorders services. If such a simplified diagnostic and management approach was to be used in another setting (e.g., primary care), consideration would be needed as to which diagnostic test and cut-point would be appropriate, as the pretest probability of OSA would be lower as would the posttest probability if the same oximetry dip rate was used (37). The experience and skill of those providing clinical care must also be considered. Although nurses rather than sleep physicians provided care in model A, these were all highly skilled and expert in the management of OSA and CPAP use. Finally, it should be emphasized that these results pertain to a symptomatic, moderate to severely affected OSA population. The findings with respect to a nurse-led, ambulatory model of care may not necessarily translate to a less severely affected patient group.
Once informed consent was provided only 22% of patients met inclusion/exclusion criteria (Figure 2). Nevertheless, being able to triage and manage nearly one-quarter of patients referred with OSA could help considerably with OSA waiting lists. We targeted symptomatic patients with moderate–severe OSA as it is this group of patients most likely to be adherent to CPAP (38), and who have the greatest risk of adverse health outcomes related to OSA (2, 5).
At the conclusion of the trial we did not perform a PSG on patients in model A to assess the effectiveness of CPAP in controlling sleep-disordered breathing and oxygen desaturation. A potential concern could be that some patients might have had Cheyne-Stokes respiration or a mixed pattern of central and obstructive sleep apnea (so-called complex sleep apnea) that could not be adequately treated with CPAP. We think this is unlikely to be a significant limitation or concern. We included patients with a symptom complex that included chronic snoring and we specifically excluded patients with clinically significant heart failure. Also, among the patients randomized to model B only 2% had a central apnea index greater than 5 during their CPAP titration PSG when “complex sleep apnea” is usually first apparent and likely to be most problematic. Given that one of our inclusion criteria was an awake oxygen saturation of 92% or more, it seems unlikely that significant sleep hypoventilation as a result of obesity or other medical problems such as kyphoscoliosis or chronic lung disease would be present in our patient group, but it is possible that CPAP may not have normalized oxygen saturation in all of the patients.
It is possible that the favorable findings of the current study may have depended on the specific combination of diagnostic and therapeutic equipment used, and may not be replicated if different technologies are used. It has been found, for example, that different autotitrating CPAP devices when used according to manufacturers' recommendations to set a fixed therapeutic CPAP produce significantly different results (39). However, whether the effects of relatively small differences in technical performance of autoadjusting CPAP devices (and oximeters) would be sufficient to translate into important differences in long-term patient outcomes is uncertain. It is reassuring in this respect that although we used a different clinical screening algorithm and different devices compared with those employed by Mulgrew and colleagues, our study conclusions were in broad agreement with theirs.
In conclusion, this study has shown that a simplified diagnostic and management model to investigate and treat moderate–severe OSA can produce patient outcomes with respect to daytime sleepiness that are not inferior to current best practice that includes in-hospital PSGs and physician review. This simplified model of care could be used in existing sleep medicine clinics to reduce the PSG and physician waiting lists by 20–25% (a greater reduction is possible if an OSA “ruling out” algorithm is added) and reduce costs, thereby improving patient access to OSA services, perhaps the major challenge facing sleep medicine services around the Western and developing world. Simplified diagnostic and management strategies need to be carefully integrated into an overall package of care. If resources are not adequate to meet the clinical need, the inclusion of different approaches to OSA diagnosis and management may not reduce the wait between referral and CPAP delivery in the home. For example, a simplified diagnostic approach will have little impact if there are insufficient public funds to supply therapeutic equipment and too few skilled health professionals to fit CPAP and manage the therapy. Nevertheless, we believe this overall package of care involving simplified OSA diagnosis, APAP titration in the home, and the expansion of the sleep medicine workforce using skilled CPAP nurses working under protocol (with the backup of sleep medicine services if needed) has the potential to add significantly to the field of sleep medicine and improve access to care for those with OSA.
The authors acknowledge the contribution to data collection and analysis of Amanda Adams, Catherine Hansen, and Peter Catcheside (Adelaide Institute for Sleep Health); Teanua Roebuck, Pavlina Toman, Sally Ho, Belinda Miller, and Alan Young (Alfred Hospital); and Lucy Fong and Joe Donohue (Newcastle). The authors also acknowledge the contribution of the sleep physicians who participated in the care of the model B patients.
| 1. | Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230–1235. |
| 2. | Yaggi HK, Concato J, Kernan WN, Lichtman JH, Brass LM, Mohsenin V. Obstructive sleep apnea as a risk factor for stroke and death. N Engl J Med 2005;353:2034–2041. |
| 3. | Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378–1384. |
| 4. | George CF. Reduction in motor vehicle collisions following treatment of sleep apnoea with nasal CPAP. Thorax 2001;56:508–512. |
| 5. | Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea–hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005;365:1046–1053. |
| 6. | Pack AI. Sleep-disordered breathing: access is the issue [see comment]. Am J Respir Crit Care Med 2004;169:666–667. |
| 7. | Bearpark H, Elliott L, Grunstein R, Cullen S, Schneider H, Althaus W, Sullivan C. Snoring and sleep apnea: a population study in Australian men. Am J Respir Crit Care Med 1995;151:1459–1465. |
| 8. | Dal Grande E, Gill T, Taylor AW, Chittleborough C, Carter P. Obesity in South Australian adults—prevalence, projections and generational assessment over 13 years. Aust N Z J Public Health 2005;29:343–348. |
| 9. | Cameron AJ, Welborn TA, Zimmet PZ, Dunstan DW, Owen N, Salmon J, Dalton M, Jolley D, Shaw JE. Overweight and obesity in Australia: the 1999–2000 Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Med J Aust 2003;178:427–432. |
| 10. | Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002;288:1723–1727. |
| 11. | Sharma SK, Kumpawat S, Banga A, Goel A. Prevalence and risk factors of obstructive sleep apnea syndrome in a population of Delhi, India. Chest 2006;130:149–156. |
| 12. | Flemons WW, Douglas NJ, Kuna ST, Rodenstein DO, Wheatley J. Access to diagnosis and treatment of patients with suspected sleep apnea. Am J Respir Crit Care Med 2004;169:668–672. |
| 13. | Flemons WW, Littner MR, Rowley JA, Gay P, Anderson WM, Hudgel DW, McEvoy RD, Loube DI; American Academy of Sleep Medicine, American College of Chest Physicians, American Thoracic Society. Home diagnosis of sleep apnea: a systematic review of the literature. Chest 2003;124:1543–1579. |
| 14. | West SD, Jones DR, Stradling JR. Comparison of three ways to determine and deliver pressure during nasal CPAP therapy for obstructive sleep apnea. Thorax 2006;61:226–231. |
| 15. | Stradling JR, Barbour C, Pitson DJ, Davies RJ. Automatic nasal continuous positive airway pressure titration in the laboratory: patient outcomes. Thorax 1997;52:72–75. |
| 16. | Berry RB, Parish JM, Hartse KM. The use of auto-titrating continuous positive airway pressure for treatment of adult obstructive sleep apnea: an American Academy of Sleep Medicine review. Sleep 2002;25:148–173. |
| 17. | Masa JF, Jimenez AJ, Duran J, Capote F, Monsterrat JM. Alternative methods of titrating positive airway pressure: a large multi-centre study. Am J Respir Crit Care Med 2004;170:1218–1224. |
| 18. | Stradling JR, Hardinge M, Paxton J, Smith DM. Relative accuracy of algorithm-based prescription of nasal CPAP in OSA. Respir Med 2004;98:152–154. |
| 19. | Tomlinson M, John Gibson G. Obstructive sleep apnoea syndrome: a nurse-led domiciliary service. J Adv Nurs 2006;55:391–397. |
| 20. | Stromberg A, Martensson J, Fridlund B, Levin LA, Karlsson JE, Dahlstrom U. Nurse-led heart failure clinics improve survival and self-care behaviour in patients with heart failure: results from a prospective, randomised trial. Eur Heart J 2003;24:1014–1023. |
| 21. | Engleman HM, Wild MR. Improving CPAP use by patients with the sleep apnoea/hypopnoea syndrome (SAHS). Sleep Med Rev 2003;7:81–99. |
| 22. | Antic N, Hansen CP, Mc C, Evoy RD. Utility of a new high sampling oximeter in the diagnosis of moderate–severe OSA [abstract]. Respirology 2004;9:A1–A75. |
| 23. | Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. EEG arousals: scoring rules and examples: a preliminary report. Sleep 1992;15:173–184. |
| 24. | American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. Sleep 1999;22:667–689. |
| 25. | Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep 1991;14:540–545. |
| 26. | Johns MW. Reliability and factor analysis of the Epworth Sleepiness Scale. Sleep 1992;15:376–381. |
| 27. | Baldwin CM, Griffith KA, Nieto FJ, O'Connor GT, Walsleben JA, Redline S. The association of sleep-disordered breathing and sleep symptoms with quality of life in the Sleep Heart Health Study. Sleep 2001;24:96–105. |
| 28. | Massie CA, McArdle N, Hart RW, Schmidt-Nowara WW, Lankford A, Hudgel DW, Gordon N, Douglas NJ. Comparison between automatic and fixed positive airway pressure therapy in the home. Am J Respir Crit Care Med 2003;167:20–23. |
| 29. | Jenkinson C, Davies RJ, Mullins R, Stradling JR. Comparison of therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised prospective parallel trial. Lancet 1999;353:2100–2105. |
| 30. | Weaver TE, Laizner AM, Evans LK, Maislin G, Chugh DK, Lyon K, Smith PL, Schwartz AR, Redline S, Pack AI, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep 1997;20:835–843. |
| 31. | Banks S, Barnes M, Tarquinio N, Pierce RJ, Lack LC, McEvoy RD. The Maintenance of Wakefulness Test in normal healthy subjects. Sleep 2004;27:799–802. |
| 32. | Ware JE Jr, Hays RD. Methods for measuring patient satisfaction with specific medical encounters. Med Care 1988;26:393–402. |
| 33. | Engleman HM, Martin SE, Kingshott RN, Mackay TW, Deary IJ, Douglas NJ. Randomised placebo controlled trial of daytime function after continuous positive airway pressure (CPAP) therapy for the sleep apnoea/hypopnoea syndrome. Thorax 1998;53:341–345. |
| 34. | Briggs A, O'Brien B, Blackhouse G. Thinking outside the box: recent advances in the analysis and presentation of uncertainty in cost effectiveness studies. Annu Rev Public Health 2002;23:377–401. |
| 35. | Mulgrew AT, Fox N, Ayas NT, Ryan CF. Diagnosis and initial management of obstructive sleep apnea without polysomnography: a randomized validation study. Ann Intern Med 2007;146:157–166. |
| 36. | Jones B, Jarvis P, Lewis JA, Ebbutt AF. Trials to assess equivalence: the importance of rigorous methods. BMJ 1996;313:36–39. |
| 37. | Flemons WW, Littner MR. Measuring agreement between diagnostic devices. Chest 2003;124:1535–1542. |
| 38. | McArdle N, Devereux G, Heidarnejad H, Engleman HM, Mackay TW, Douglas NJ. Long-term use of CPAP therapy for sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1999;159:1108–1114. |
| 39. | Sériès F, Plante J, Lacasse Y. Reliability of home CPAP titration with different automatic CPAP devices. Respir Res 2008;9:56. |
