Introduction: Obstructive sleep apnoea (OSA) is common and is caused by repetitive obstruction of the upper airway during sleep. OSA is associated with increased cardiovascular morbidity and mortality and is an independent risk factor for hypertension. The immediate physiological effects of OSA include intermittent hypoxia, repeated arousal from sleep and intra-thoracic pressure swings. The resulting activation of the sympathetic nervous system, systemic inflammation and oxidative stress may result in increased arterial stiffness and endothelial dysfunction, potentially explaining any causal link between OSA and cardiovascular disease (CVD). Continuous positive airway pressure (CPAP) therapy improves excessive daytime sleepiness (EDS) and in non-randomised studies, reduces cardiovascular mortality. Prior to starting this study, there was a limited amount of evidence suggesting that CPAP therapy improved arterial stiffness and endothelial function, but the effects in subjects without pre-existing CVD were unclear. Aims: i) to determine whether CPAP therapy has an effect upon measures of arterial stiffness and endothelial function in patients with OSA, in the absence of known CVD. ii) To compare arterial stiffness and endothelial function in a subset of patients with OSAHS (defined as OSA and EDS), with a group of well-matched control subjects. Methods: Fifty three patients with OSA, defined as an apnoea/hypopnoea index of ≥15, and without known CVD, entered a double-blind placebo-controlled crossover trial of 12 weeks CPAP therapy, of whom forty three completed the study protocol. Sham CPAP was used in the placebo arm of the study and vascular assessments were made at baseline and after each arm of the study. Arterial stiffness was determined by measuring aortic distensibility using cardiovascular magnetic resonance imaging and by measuring the augmentation index (AIx) and aortic pulse wave velocity (PWV) by applanation tonometry. Endothelial function was assessed non-invasively by measuring vascular reactivity after administration of salbutamol and glyceryl trinitrate. In a subset of twenty patients with OSAHS, arterial stiffness and endothelial function at baseline were compared to readings obtained from healthy control subjects, matched on a one-to-one basis for age, sex and BMI. Results: Patients with OSAHS (n=20) had increased arterial stiffness [AIx 19.3(10.9) vs. 12.6(10.2) %; p=0.017] and impaired endothelial function, measured as the change in AIx following salbutamol [-4.3(3.2) vs. -8.0(4.9) %; p=0.02] compared to controls. Twelve weeks of CPAP therapy had no significant effect upon any measure of arterial stiffness or endothelial function in patients with OSA (n=43). A trend towards a reduction in AIx following CPAP therapy was seen, but this was non-significant. There was a reduction in systolic blood pressure following CPAP therapy [126(12) vs. 129(14) mmHg]. Sub group analysis showed CPAP to have no effect on arterial stiffness or endothelial function in patients with EDS or in patients using CPAP for ≥4 hours per night. Conclusions: This study demonstrates that even in the absence of known CVD, patients with OSAHS have evidence of increased arterial stiffness and impaired endothelial function. However, in patients with OSA, free from CVD, CPAP therapy did not lead to an improvement in any measure of arterial stiffness or endothelial function after 12 weeks.

This thesis addresses the organisation of drive to human upper airway and inspiratory pump muscles. The characterisation of single motor unit activity is important as the discharge frequency or timing of discharge of each motor unit directly reflects the output of single motoneurones. Thus, the firing properties of a population of motor units is indicative of the neural drive to the motoneurone pool. The experiments presented in Chapter 2 measured the recruitment time of five inspiratory pump muscles (diaphragm, scalene, second parasternal intercostal, and third and fifth dorsal external intercostal muscles) during normal quiet breathing and quantified the timing and magnitude of drive reaching each muscle. Chapter 3 examined the EMG activity of a major upper airway muscle (the genioglossus). The single motor units of the genioglossus display activity that can be grouped into six types based on its association or lack of association with respiration. The types of activity are termed: Inspiratory Phasic, Inspiratory Tonic, Expiratory Phasic, Expiratory Tonic, Tonic, and Tonic Other. A new method is presented in Chapter 4 to illustrate large amounts of data from single motor units recorded from respiratory muscles in a concise manner. This single figure displays for each motor unit, the recruitment time and firing frequency, the peak discharge frequency and its time, and the derecruitment time and its frequency. This method, termed the time-and-frequency plot, is used to demonstrate differences in behaviour between populations of diaphragm (Chapter 2) and genioglossus (Chapter 3) motoneurones. In Chapter 5, genioglossus activity during quiet breathing is compared between a group of patients with severe OSA and healthy control subjects. The distribution of central drive is identical between the OSA and control subjects with the same proportion of the six types of motor unit activity in both groups. However, there are alterations in the onset time of Inspiratory Phasic and Inspiratory Tonic motor units in OSA subjects and their peak discharge rates are also altered. Single motor unit action potentials in OSA subjects showed an increased area. This suggests the presence of neurogenic changes and may provide a pathophysiological explanation for the increased multiunit electromyographic activity reported in OSA subjects during wakefulness.

While the importance of assessing the occupational consequences of insomnia and other sleep disorders is emphasised in clinical nosologies and research guidelines, there is little consensus on which aspects of occupational performance should be assessed, how such impairment should be measured, and how outcomes should be reported. The research programme described in this thesis aimed to address this issue. Chapter 1 presents a systematic review and methodical critique of studies reporting those aspects of occupational performance most impacted by (or most frequently associated with) insomnia symptoms and degraded sleep quality. Equivocal results, wide variations in reporting conventions, and the overall lack of comparability among studies, strongly indicated the need to develop a standardised metric able to quantify sleep related occupational performance and serve as an assessment and outcome instrument suitable for use in research and clinical settings. Informed by the literature review, Chapters 2-4 describe the development and validation of the Loughborough Occupational Impact of Sleep Scale ( LOISS ), a unidimensional 19 item questionnaire that captures sleep-related occupational impairment across a number of workplace domains over a 4-week reference period. Chapters 5-7 describe LOISS outcomes from: i) surveys in a random population sample; ii) a representative sample of the UK workforce; and iii) a clinical sample of patients with obstructive sleep apnoea (before and after treatment with CPAP). Overall, the scale showed strong internal consistency (Cronbach s alpha range=0.84-0.94) and test-retest reliability (r=0.77, r2=0.59, p<0.001), high levels of criterion validity (significantly discriminating between good and poor sleepers), and proved an effective outcome measure in OSA. From the survey data reported in Chapters 2-7, LOISS score distributions showed no consistent gender difference but did show a significant ageing gradient, with sleep-related occupational impairment declining with increasing age. In conclusion, the work presented here supports the usability, validity and reliability of the LOISS as an assessment and outcome instrument, and also demonstrates the utility of this instrument in exploring the dynamics of sleep-related occupational performance

This thesis addresses the organisation of drive to human upper airway and inspiratory pump muscles. The characterisation of single motor unit activity is important as the discharge frequency or timing of discharge of each motor unit directly reflects the output of single motoneurones. Thus, the firing properties of a population of motor units is indicative of the neural drive to the motoneurone pool. The experiments presented in Chapter 2 measured the recruitment time of five inspiratory pump muscles (diaphragm, scalene, second parasternal intercostal, and third and fifth dorsal external intercostal muscles) during normal quiet breathing and quantified the timing and magnitude of drive reaching each muscle. Chapter 3 examined the EMG activity of a major upper airway muscle (the genioglossus). The single motor units of the genioglossus display activity that can be grouped into six types based on its association or lack of association with respiration. The types of activity are termed: Inspiratory Phasic, Inspiratory Tonic, Expiratory Phasic, Expiratory Tonic, Tonic, and Tonic Other. A new method is presented in Chapter 4 to illustrate large amounts of data from single motor units recorded from respiratory muscles in a concise manner. This single figure displays for each motor unit, the recruitment time and firing frequency, the peak discharge frequency and its time, and the derecruitment time and its frequency. This method, termed the time-and-frequency plot, is used to demonstrate differences in behaviour between populations of diaphragm (Chapter 2) and genioglossus (Chapter 3) motoneurones. In Chapter 5, genioglossus activity during quiet breathing is compared between a group of patients with severe OSA and healthy control subjects. The distribution of central drive is identical between the OSA and control subjects with the same proportion of the six types of motor unit activity in both groups. However, there are alterations in the onset time of Inspiratory Phasic and Inspiratory Tonic motor units in OSA subjects and their peak discharge rates are also altered. Single motor unit action potentials in OSA subjects showed an increased area. This suggests the presence of neurogenic changes and may provide a pathophysiological explanation for the increased multiunit electromyographic activity reported in OSA subjects during wakefulness.

Background: Compliance with CPAP treatment for OSA is not reliably predicted by the severity of symptoms or physiological variables. I conducted a series of studies to examine a range of factors that may affect compliance with CPAP. Methods: I performed a retrospective study examining association of demographic factors and OSA severity with long-term CPAP compliance. In a prospective study, I looked at the correlation of short and long-term CPAP compliance with socio-economic status, education, type D personality, demographics, disease severity, mood and clinician's prediction. I undertook a prospective, cross-over trial comparing the impact of Bi-level PAP therapy in individuals with low tolerance of CPAP. Results: In a retrospective analysis, an improvement in subjective daytime somnolence was correlated with optimal compliance. In the prospective study, median compliance with CPAP at 6 months was 5.6 (3.4- 7.1) hours/night with 73% of subjects using CPAP ≥ 4 hours/night. Compliance with CPAP was not found to be associated with socio-economic class for people in work, type D personality, education, sex, age, baseline sleepiness (ESS score) or disease severity (ODI). The clinician's initial impression had no predictive value for individual patients. Subjects who were long-term unemployed or reporting mood disorders (High Beck's Depression Index scores) were likely to have poor compliance and sub-optimal CPAP usage (OR 4.6, p = 0.011 and OR 1.4. p=0.04 respectively). Subjects experiencing side effects after the first night on treatment showed lower acceptance and subsequent compliance. In the cross-over trial, changing to Bi-level PAP in individuals with suboptimal compliance due to pressure related intolerance, did not lead to an improvement in CPAP compliance. In post-hoc analysis, compliance and comfort were better in the subgroup that complained of difficulty with exhalation on CPAP. Conclusion: My research as presented in this thesis, did not find an association between disease severity (ODI), socio-economic status (for people in employment), education or personality type and CPAP compliance. My research demonstrated that subjects with long-term unemployment, mood disorders and those experiencing side effects on the first night of treatment were likely to have sub-optimal compliance. Changing to Bi-level PAP is only likely to be useful for a sub-group of subjects experiencing pressure related intolerance. More research is needed to explore whether intensive support to individuals with OSA and long term unemployment, as well as mood disorders, may improve compliance.

Tidigare forskning indikerar att individer med sömnapné får mer allvarliga skador i sambandmed vägtrafikolyckor jämfört med individer utan sömnapné. Det har även visats att merallvarliga skador genererar högre kostnader. Det finns dock ett kunskapsglapp i frågan om demer allvarliga skadorna som involverar individer med sömnapné har högresjukvårdskostnader i Sverige. Studier visar även att om patienter ska kunna få ut maximaltmed vård för de skattepengar som läggs på hälso- och sjukvården bör samhällseffektivakostnadsanalyser göras. En del i samhällseffektiva kostnadsanalyser är att identifiera,kvantifiera och värdera de kostnader som är relevanta vid ett specifikt olycksfall.Uppsatsens syfte är att framställa, jämföra och analysera de sjukvårdskostnader somuppkommer vid vägtrafikolyckor i Sverige orsakade av individer med och utan sömnapnéunder en uppföljningsperiod på ett, två och tre år. I sjukvårdskostnaderna inkluderas dekostnader som uppstår i slutenvården, öppenvården, samt läkemedelskostnader. Syftetuppfylls först genom identifiering, kvantifiering och värdering av sjukvårdskostnaderna. Meden ekonometrisk modell avser vi dessutom att förklara sambandet mellan de förklarandevariablerna ålder, kön och patientgrupp och den beroende variabeln sjukvårdskostnader. Denekonometriska modellen skapas utifrån data från olycksregistret Swedish Traffic AccidentData Acquisition (STRADA) och organisationen European Sleep Apnoea Database(ESADA).I resultatanalysen presenteras och analyseras sjukvårdskostnaderna utifrånsamhällsekonomisk teori för att avgöra om det existerar en skillnad i sjukvårdskostnader samtför att utreda hur den eventuella skillnaden är fördelad enligt Pareto- och Kaldor-Hickskriteriet. Studiens resultat visar på att det existerar en framträdande skillnad mellan de tvåpatientgruppernas sjukvårdskostnader. Marginalkostnaderna för patientgruppen medsömnapné är betydligt större än för patientgruppen utan sömnapné och de inkrementellakostnaderna visar också relativt stora kostnadsskillnader under respektive uppföljningsår. / Previous research indicates that individuals with the condition obstructive sleep apnoea(OSA) get more severe injuries after road traffic accidents, in comparison with individualswithout OSA. It has also, in previous studies, been shown that more severe injuries generatehigher costs. There is although a knowledge gap concerning whether the more severe injuriesthat involve individuals with OSA result in higher medical expenses. Furthermore, earlierresearch also implies that for patients to receive maximum healthcare from the tax moneyreimbursing the healthcare in Sweden, there should be socio-effective cost analysisconducted. An important part of socio-effective cost analysis is the identification,quantification and valuation of relevant costs associated with a specific causality.The aim of the study is to produce, compare and analyse the healthcare costs associated withroad traffic accidents in Sweden caused by individuals with and without OSA, during afollow-up period of one, two and three years. The healthcare costs include the costs that occurin the inpatient and outpatient care as well as pharmaceutical costs. The aim has been fulfilledthrough identification, quantification and valuation of the healthcare costs associated withroad traffic accidents, for individuals with and without OSA. In order to examine therelationship between the describing variables age, sex and patient group and the dependentvariable healthcare cost, we constructed an econometric model. The econometric model hasbeen assembled by data from the accident register Swedish Traffic Accident Data Acquisition(STRADA) and the organisation European Sleep Apnoea Database (ESADA).The result of the study presents and analyse the healthcare costs through socio-economictheory to decide whether there does exist a difference in healthcare costs and to investigate ifthe eventual difference in costs is distributed according to the Pareto- and Kaldor-Hickscriteria.The study’s result demonstrates a significant difference of the healthcare costs in thetwo patient groups. The marginal costs for the patient group with OSA is considerably higherthan the marginal cost for the patient group without OSA. The incremental costs also showrelatively large cost differences during the three follow-up years.

Barorezeptorsensitivität, Herzfrequenzvariabilität und Blutdruckvariabilität bei Patienten mit einem mild-moderaten Schlafapnoe Syndrom und bei gesunden Probanden Die Behandlung von Patienten mit einem mild-moderaten obstruktiven Schlafapnoe Syndrom (OSAS) wird von der klinischen Symptomatik und dem Herzkreislaufrisiko bestimmt. Wir konnten nachweisen, dass bei einem mild-moderaten OSAS unter der Beatmungstherapie neben der Beseitigung der nächtlichen Atmungsstörung auch eine Änderung der Mikrostruktur des Schlafes (Arousal) und der Müdigkeit am Tage zu verzeichnen ist. Das Herzkreislaufrisiko untersuchten wir anhand noninvasiver Parameter der sympathovagalen Balance. Dazu wurden die Herzfrequenzvariabilität (HRV), die Blutdruckvariabilität (BDV) und die Barorezeptorsensitivität (BRS) im Zeit- und Frequenzbereich bestimmt und diese Größen bei Patienten mit einem mild-moderaten OSAS im Vergleich zu gesunden Probanden als auch der Effekt einer CPAP-Therapie - im Schlaf als auch am Tage - analysiert. Bei gesunden Probanden fanden wir eine HRV- und BRS-Abnahme sowie eine BDV-Zunahme im REM- gegenüber dem NREM-Schlaf. OSAS Patienten haben im Vergleich zu Gesunden eine niedrigere BRS im NREM und eine erhöhte BDV sowohl im REM- als auch NREM-Schlaf. CPAP führt beim OSAS zu einer Abnahme der Herzfrequenz und Zunahme der BRS, vornehmlich im NREM Schlaf und bei zusätzlich bekannter Hypertonie. Die HRV nimmt ab und die BDV zu, jeweils unabhängig vom Schlafstadium bzw. einer bestehenden Hypertonie. Am Tage zeigt sich nur ein Kurzzeiteffekt hinsichtlich Zunahme der BRS und HRV. Dieser Effekt ist vom Ausmaß des OSAS und dem Vorhandensein einer Hypertonie abhängig und nach 4 Wochen Therapie nicht mehr nachweisbar. Untersucht man den Effekt der Beatmungstherapie auf HRV, BDV und BRS bei gesunden Probanden im Akutversuch, dann findet man eine Erhöhung des Blutdruckes bei Abnahme der Herzfrequenz und Zunahme der BRS. HRV, BDV und BRS als Parameter für das kardiovaskuläre Risiko zeigen nachweisbare Veränderungen bei Schlafapnoe Patienten, auch wenn nur ein mild-moderates OSAS vorliegt. Ein Therapieeffekt lässt sich anhand dieser Parameter auch nachweisen, wobei ein vorhandener Akuteffekt von Überdruckbeatmung auf HRV, BDV und BRS unabhängig von einem OSAS zu berücksichtigen ist. / Baroreceptor sensitivity, heart-rate variability, and blood-pressure variability in patients with mild to moderate sleep apnoea syndrome, and in healthy controls The treatment of patients with mild to moderate obstructive sleep apnoea syndrome, OSAS, is determined by the clinical symptom complex and by the cardiovascular risk. In patients with mild to moderate OSAS who received therapy in the form of assisted ventilation, we succeeded in evidencing that it is possible to influence the microstructure of sleep (i.e., of arousal) as well as fatigue experienced during the day, in addition to eliminating nocturnal respiratory disturbance. We investigated the cardiovascular risk by examining non-invasive parameters for sympathovagal balance. Therefore we analyzed heart-rate variability (HRV), blood-pressure variability (BPV), and baroreceptor sensitivity (BRS) over time and frequency ranges in patients with mild to moderate OSAS, in comparison to healthy controls. We likewise assessed the effects of CPAP therapy on these parameters, both during sleeping as well as non-sleeping hours. Among healthy test subjects, we determined decreases in HRV and BRS, as well as increase in BPV, during REM sleep, in comparison to NREM sleep. In comparison to healthy controls, OSAS patients have lower BRS during NREM and increased BPV in both REM and NREM sleep. In OSAS patients, CPAP leads to a decrease in heart rate and increase in BRS, especially in NREM sleep and in patients for whom hypertension is also known. HRV diminishes and BPV increases, in both cases regardless of the sleep stage or presence or absence of hypertension. During the day, only a short-term effect becomes apparent with respect to increases in BRS and HRV. This effect depends on the extent of OSAS and on the existence of hypertension; after four weeks of therapy, the effect is no longer in evidence. Acute testing of the effect of assisted ventilation on HRV, BPV, and BRS among healthy controls discloses increase in blood pressure, accompanied by decrease in heart rate and increase in BRS. HRV, BPV, and BRS as parameters for cardiovascular risk reveal evidence of alterations in sleep-apnoea patients, even for those suffering only from mild to moderate OSAS. Therapeutic effects are also in evidence on the basis of these parameters, whereby an existing acute effect of positive-pressure ventilation on HRV, BPV, and BRS regardless of OSAS must also be taken into account.

In the early 1990s claims were made that in all areas of health care, “30-40% of patients do not receive treatments of proven effectiveness”, and, “20-25% of patients have treatments that are unnecessary or potentially harmful”. Many such practices were diffused prior to the acceptance of modern evidence-based standards of clinical- and cost-effectiveness. I define disinvestment in the context of health care as the processes of withdrawing (partially or completely) resources from any existing health care practices, procedures, technologies or pharmaceuticals that are deemed to deliver little or no health gain relative to their cost, and thus are not efficient health resource allocations. Arguably disinvestment has been central to Evidence-Based Medicine(EBM) for well over a decade yet despite general advances in EBM, this topic remains relatively unexplored. This thesis examines the ongoing challenges that exist within the Australian context relating to effective disinvestment. Upper airway surgical procedures for the treatment of adult Obstructive Sleep Apnoea Syndrome (OSA) are used as a case study to contextualise these challenges. This thesis has six sections: 1. A review of the literature outlines developments in EBM broadly and provides a detailed background to OSA, including the numerous treatment options for the condition. This review examines evidence that highlights the importance of ‘highly effective treatment’ over ‘subtherapeutic treatment’ as a necessity to confer improved health outcomes in OSA. It is argued that claims of surgical success inherent in most published results of surgery effectiveness fail to assimilate contemporary evidence for clinically significant indicators of success. 2. Section two comprises the first reported meta-analysis in this area. It presents the pooled success rates of surgery according to various definitions. Specifically, when the traditional ‘surgical’ definition of success is applied the pooled success rate for Phase I (i.e. soft palate) surgical procedures is 55% (that is 45% fail). However, using a more stringent definition (endorsed by the peak international sleep medicine body), success is reduced to 13% (that is 87% fail). Similarly for Phase II (i.e. hard palate) procedures success rates decrease from 86% to 43% respectively when moving from a surgical to a medical definition of success. That various medical specialties differentially define treatment success, I argue, creates uncertainty for observers and non-clinical participants in this debate (eg policy stakeholders and patients). This represents a barrier to disinvestment decisions. 3. Results are presented from a clinical audit of surgical cases conducted as a component of this thesis. Both clinical effectiveness and procedural variability of surgery are reported. A unique methodology was utilised to capture data from multiple centres. It is the first time such a methodology has been reported to measure procedural variability alongside clinical effectiveness (inclusive of a comparative treatment arm). The observed cohort (n=94) received 41 varying combinations of surgery in an attempt to treat OSA. Results on effectiveness demonstrate an overall physiological success rate of 13% (according to the most stringent definition; phases I and II combined). This demonstration of procedural variability combined with limited effectiveness highlights clinical uncertainty in the application of surgical procedures. 4. Section four outlines how a qualitative phase of enquiry, directed at exploring the perspectives and experiences of surgery recipients, was approved by three independent research ethics review boards but was not supported by a small group of surgeons, resulting in the project being canceled. Potential consequences of this for impeding health services research (HSR) are discussed. 5. Two sets of results are reported from a qualitative phase of enquiry (semi-structured interviews) involving senior Australian health policy stakeholders. The first results are of policy stakeholders’ perspectives on the surgical meta-analysis and clinical audit studies in 2 and 3 above. The second results are from an extended series of questions relating to challenges and direction for effecting disinvestment mechanisms in Australia. Stakeholder responses highlight that Australia currently has limited formal systems in place to support disinvestment. Themes include how defining and proving inferiority of health care practices is not only conceptually difficult but also is limited by data availability and interpretation. Also, as with any policy endeavour there is the ever-present need to balance multiple interests. Stakeholders pointed to a need, and a role, for health services and policy research to build methodological capacity and decision support tools to underpin disinvestment. 6. A final discussion piece is presented that builds on all previous sections and summarises the specific challenges that exist for disinvestment, including those methodological in nature. The thesis concludes with potential solutions to address these challenges within the Australian and international context. Systematic policy approaches to disinvestment represent one measure to further improve equity, efficiency, quality of care, as well as sustainability of resource allocation. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1297655 / Thesis (Ph.D.) -- University of Adelaide, School of Population Health and Clinical Practice, 2007

In the early 1990s claims were made that in all areas of health care, “30-40% of patients do not receive treatments of proven effectiveness”, and, “20-25% of patients have treatments that are unnecessary or potentially harmful”. Many such practices were diffused prior to the acceptance of modern evidence-based standards of clinical- and cost-effectiveness. I define disinvestment in the context of health care as the processes of withdrawing (partially or completely) resources from any existing health care practices, procedures, technologies or pharmaceuticals that are deemed to deliver little or no health gain relative to their cost, and thus are not efficient health resource allocations. Arguably disinvestment has been central to Evidence-Based Medicine(EBM) for well over a decade yet despite general advances in EBM, this topic remains relatively unexplored. This thesis examines the ongoing challenges that exist within the Australian context relating to effective disinvestment. Upper airway surgical procedures for the treatment of adult Obstructive Sleep Apnoea Syndrome (OSA) are used as a case study to contextualise these challenges. This thesis has six sections: 1. A review of the literature outlines developments in EBM broadly and provides a detailed background to OSA, including the numerous treatment options for the condition. This review examines evidence that highlights the importance of ‘highly effective treatment’ over ‘subtherapeutic treatment’ as a necessity to confer improved health outcomes in OSA. It is argued that claims of surgical success inherent in most published results of surgery effectiveness fail to assimilate contemporary evidence for clinically significant indicators of success. 2. Section two comprises the first reported meta-analysis in this area. It presents the pooled success rates of surgery according to various definitions. Specifically, when the traditional ‘surgical’ definition of success is applied the pooled success rate for Phase I (i.e. soft palate) surgical procedures is 55% (that is 45% fail). However, using a more stringent definition (endorsed by the peak international sleep medicine body), success is reduced to 13% (that is 87% fail). Similarly for Phase II (i.e. hard palate) procedures success rates decrease from 86% to 43% respectively when moving from a surgical to a medical definition of success. That various medical specialties differentially define treatment success, I argue, creates uncertainty for observers and non-clinical participants in this debate (eg policy stakeholders and patients). This represents a barrier to disinvestment decisions. 3. Results are presented from a clinical audit of surgical cases conducted as a component of this thesis. Both clinical effectiveness and procedural variability of surgery are reported. A unique methodology was utilised to capture data from multiple centres. It is the first time such a methodology has been reported to measure procedural variability alongside clinical effectiveness (inclusive of a comparative treatment arm). The observed cohort (n=94) received 41 varying combinations of surgery in an attempt to treat OSA. Results on effectiveness demonstrate an overall physiological success rate of 13% (according to the most stringent definition; phases I and II combined). This demonstration of procedural variability combined with limited effectiveness highlights clinical uncertainty in the application of surgical procedures. 4. Section four outlines how a qualitative phase of enquiry, directed at exploring the perspectives and experiences of surgery recipients, was approved by three independent research ethics review boards but was not supported by a small group of surgeons, resulting in the project being canceled. Potential consequences of this for impeding health services research (HSR) are discussed. 5. Two sets of results are reported from a qualitative phase of enquiry (semi-structured interviews) involving senior Australian health policy stakeholders. The first results are of policy stakeholders’ perspectives on the surgical meta-analysis and clinical audit studies in 2 and 3 above. The second results are from an extended series of questions relating to challenges and direction for effecting disinvestment mechanisms in Australia. Stakeholder responses highlight that Australia currently has limited formal systems in place to support disinvestment. Themes include how defining and proving inferiority of health care practices is not only conceptually difficult but also is limited by data availability and interpretation. Also, as with any policy endeavour there is the ever-present need to balance multiple interests. Stakeholders pointed to a need, and a role, for health services and policy research to build methodological capacity and decision support tools to underpin disinvestment. 6. A final discussion piece is presented that builds on all previous sections and summarises the specific challenges that exist for disinvestment, including those methodological in nature. The thesis concludes with potential solutions to address these challenges within the Australian and international context. Systematic policy approaches to disinvestment represent one measure to further improve equity, efficiency, quality of care, as well as sustainability of resource allocation. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1297655 / Thesis (Ph.D.) -- University of Adelaide, School of Population Health and Clinical Practice, 2007

Obstructive sleep apnoea hypopnoea syndrome (OSAHS) is a highly prevalent disease in which upper airways are collapsed during sleep, leading to serious consequences. The reference standard of clinical diagnosis, called Polysomnography (PSG), requires a full-night hospital stay connected to over 15 measuring channels requiring physical contact with sensors. The vast quantity of physiological data acquired during the PSG has to be manually scored by a qualified technologist to assess the presence or absence of the decease. The PSG is inconvenient, time consuming, expensive and unsuited for community screening. The limited PSG facilities around the world have resulted in long waiting lists and a large fraction of patients remain undiagnosed at present. There has been a flurry of recent activities in developing a portable technology to resolve this need. All the devices have at least one sensor that requires physical contact with the subject. Unattended systems have not led to sufficiently high sensitivity/specificity levels to be used in a routine home monitoring or a community screening exercise. OSAHS is a sleep respiratory disorder principally caused by functional deficiencies occurring in the upper airways during sleep. These conditions and the reduced muscle tone during sleep, cause the muscles in the upper airways to collapse partially or completely thus resulting in episodes of hypopnoea and apnoea respectively. During the process leading to collapse of upper airways, upper airways act as an acoustic filter frequently producing snoring sounds. The process of snore sound production leads us to hypothesise that snore sounds should contain information on changes occurring in the upper airways during the OSAHS. Snoring almost always accompanies the OSAHS and is universally recognised as its earliest symptom. At present, however, the quantitative analysis of snore sounds is not a practice in clinical OSAHS detection. The vast potential of snoring in the diagnosis/screening of the OSAHS remains unused. Snoring-based technology opens up opportunities for building community-screening devices that do not depend on contact instrumentation. In this thesis, we present our work towards developing a snore–based non-contact instrumentation for the diagnosis/screening of the OSAHS. The primary task in the analysis of Snore Related Sounds (SRS) would be to segment the SRS data as accurately as possible into three main classes, snoring (voiced non-silence), breathing (unvoiced non-silence) and silence. A new algorithm was developed, based on pattern recognition for the SRS segmentation. Four features derived from the SRS were considered to classify samples of the SRS into three classes. We also investigated the performance of the algorithm with three commonly-used noise reduction (NR) techniques in speech processing, Amplitude Spectral Subtraction (ASS), Power Spectral Subtraction (PSS) and Short Time Spectral Amplitude (STSA) Estimation. It was found that the noise reduction, together with a proper choice of features, could improve the classification accuracy to 96.78%. A novel model for the SRS was proposed for the response of a mixed-phase system (total airways response, TAR) to a source excitation at the input. The TAR/source model is similar to the vocal tract/source model in speech synthesis and is capable of capturing the acoustical changes brought about by the collapsing upper airways in the OSAHS. An algorithm was developed, based on the higher-order-spectra (HOS) to jointly estimate the source and the TAR, preserving the true phase characteristics of the latter. Working on a clinical database of signals, we show that the TAR is indeed a mixed phased signal and second-order statistics cannot fully characterise it. Nocturnal speech sounds can corrupt snore recordings and pose a challenge to the snore-based OSAHS diagnosis. The TAR could be shown to detect speech segments embedded in snores and derive features to diagnose the OSAHS. Finally presented is a novel technique for diagnosing the OSAHS, based solely on multi-parametric snore sound analysis. The method comprises a logistic regression model fed with a range of snore parameters derived from its features — the pitch and Total Airways Response (TAR) estimated using a Higher Order Statistics (HOS) based algorithm. The model was developed and its performance validated on a clinical database consisting of overnight snoring sounds simultaneously recorded during a hospital PSG using a high fidelity sound recording setup. The K-fold cross validation technique was used for validating the model. The validation process achieved an 89.3% sensitivity with 92.3% specificity (the area under the Receiver Operating Characteristic (ROC) curve was 0.96) in classifying the data sets into the two groups, the OSAHS (AHI >10) and the non-OSAHS. These results are superior to the existing results and unequivocally illustrate the feasibility of developing a snore-based non-contact OSAHS screening device.