2012 Cohort

Dr Sharon Cadogan

The design, implementation and evaluation of a laboratory based intervention to optimise serum immunoglobulin test use in primary care

Institution:  UCC, 2012 cohort

Supervisors:  Prof. Mary Cahill (UCC), Prof. John Browne (UCC), Prof. Colin Bradley (UCC)

Abstract:       

Laboratory testing plays a fundamental role in the screening, diagnoses and monitoring of many conditions. Given the increased pressures on the Irish health service, improving inefficiencies and reducing waste, while maintaining the quality of care is at the forefront of healthcare planning. Promoting optimal laboratory service utilisation could play a key role in reducing health expenditure, in particular by preventing the unnecessary use of costly downstream services that often arise as a result of testing.

The overall aim of this thesis was to design, implement and evaluate a behaviour change intervention for optimising serum immunoglobulin test use in primary care. The thesis objectives were as follows: 1. To conduct a systematic review of the existing literature on the effectiveness of previous interventions targeting primary care test use. 2. Identify the barriers and enablers of improving test ordering for serum immunoglobulins among General Practitioners (GPs), using semi-structured interviews. 3. Identify the intervention components (behaviour change techniques and mode(s) of delivery) that could overcome the modifiable barriers and enhance the enablers. 4. Determine which GP and practice characteristics are associated with higher serum immunoglobulin test ordering patterns in the South of Ireland. 5. To implement and evaluate a behaviour change intervention targeting GP serum immunoglobulin test use in the Cork-Kerry region.

The published literature to date was synthesised in a systematic review (Chapter 3). This review was conducted in accordance with the Effective Practice and Organisation of Care (EPOC) guidelines and quality appraised using the Cochrane Collaboration risk of bias tool. A theory-based paper identifying the modifiable barriers and enablers to test ordering behaviour change and the selection of intervention components to overcome these is presented in Chapter 4. This involved using a combination of behaviour change models including the Theoretical Domains Framework (TDF), the behaviour change wheel constructs; capabilities, opportunities, motivations of behaviour (COM-B) and Behaviour Change Techniques (BCTs) to identify intervention functions best suited to targeting GP test ordering behaviour. The GP and practice characteristics associated with higher test ordering patterns are described in Chapter 5. These were identified by performing a multi-level analysis of all GP test orders in the studied region for a one-year time period, using routine laboratory data. The design of the intervention material and details on the implementation plans are provided in Chapter 6. Results of the effect of the intervention using nine-month follow up data are described in Chapter 7. This was performed using interrupted time-series with segmented Poisson regression models to assess the pre-and post-intervention trend for serum immunoglobulin testing among GPs in the Cork-Kerry region of Ireland. Finally, a discussion of the key findings, strengths and limitations of the thesis and recommendations for future research are addressed (Chapter 8).

A number of different interventions were of variable efficacy at changing GP test ordering behaviour. However, generalisability across tests and methodological weakness were identified in these studies (Chapter 3). GP factors contributing to higher immunoglobulin test ordering in our sample included female gender and fewer years of clinical experience (Chapter 5). The lack of clear guidelines and knowledge on how to interpret the test results posed greatest problems for GPs. Four key intervention functions were identified for overcoming these modifiable barriers to effective test use; education, persuasion, enablement and environmental restructuring (Chapter 4). Following the introduction of a guideline and education-based strategy targeting the two key issues (by incorporating the four functions), test orders for serum immunoglobulins dropped significantly. A nine-month evaluation of the effectiveness of this intervention found a statistically significant 1.5% reduction in the fortnight-to-fortnight test ordering trend for serum immunoglobulins (Chapter 7).

This research provides an important overview of the behavioural factors influencing laboratory testing among GPs. The incorporation of behavioural theory, specifically the COM-B, TDF and BCT taxonomy, has supported the identification of factors such as xxii knowledge and the social and environmental context, which are key for understanding testing behaviours. Combining these context specific “mechanisms of change” with international evidence on what has previously worked, assisted in the development of an effective behaviour change intervention targeting serum immunoglobulin test use in primary care.

Visit Profile

Dr Nora Ann Donnelly

The role of carer stress in acute and long-term care utilisation by community-dwelling older people

Institution: RCSI, 2012 cohort

Supervisors:  Dr. Frank Doyle (RCSI), Prof. Anne Hickey (RCSI), Dr. Niamh Humphries (RCPI), Dr. Christine McGarrigle (TCD)

Abstract:       

In examining the sustainability of homecare, gerontological researchers have increasingly recognised how stressful caregiving can be. Indeed, several researchers have postulated that carer stress could increase the risk of institutional care utilisation by care recipients. However, this contention has not been critically analysed. Therefore, this thesis asks to what extent, if any, carer stress influences institutional care utilisation by community-dwelling older people.

A mixed methods approach was adopted. Study 1 systematically reviewed and meta-analysed the strength of the effect of carer stress on subsequent institutionalisation of community-dwelling older people. Study 2 qualitatively analysed healthcare professionals’ (n=22) and carers’ (n=16) perceptions of how carer stress and health system factors may influence long-term care (LTC) admissions. Study 3 compared different theoretically-informed models to determine factors that may influence institutional care utilisation by community-dwelling older people (n=205). This was a secondary analysis of The Irish Longitudinal Study of Ageing (TILDA).

Study 1 found that while carer stress has a significant effect on subsequent institutionalisation of care recipients, the overall effect size was negligible (N= 54 studies; standardised mean difference =0.05, 95% CI=0.04-0.07). Study 2 found the escalation of care recipients’ needs, to the extent that the carer or community care cannot meet these needs, drove both carer stress and LTC admissions. This suggests that carer stress is an epiphenomenon of the journey to LTC. This was represented in a proposed alternative theoretical model. This model was examined in Study 3, which found the effects of institutional care utilisation on carer stress were as convincing as the effect of carer stress on institutional care utilisation.

Carer stress is not a risk factor for care recipient institutionalisation. It appears to co-occur with the need for institutional care utilisation rather than driving this utilisation. The thesis proposed and tested an alternative theoretical model to represent this, which has considerable implications for the caregiving literature. Future research should further test this model in prospective structural analysis with other sufficiently-powered samples.

Visit Profile

Dr Sarah Fitzgerald

Evaluation of complex workplace dietary interventions

Institution: UCC, 2012 cohort

Supervisors:  Prof. Ivan Perry (UCC), Dr. Ann Kirby (UCC), Dr. Aileen Murphy (UCC), Dr. Fiona Geaney (Food Choice at Work Ltd.)

Abstract:       

The workplace has been identified as an ideal setting in which to promote healthy dietary behaviours. However, uncertainty surrounds both the effectiveness and cost-effectiveness of workplace dietary interventions and a dearth of evidence exists with regards to evaluating the implementation process of such interventions. The core aim of this thesis was to conduct both a process evaluation and an economic evaluation of the Food Choice at Work (FCW) complex workplace dietary intervention.

The FCW study was a pragmatic trial which measured the effectiveness of a complex workplace dietary intervention in ideal settings. Firstly, a cross-sectional analysis which employed a zero-inflated negative binomial (zinb) regression model to examine associations between objective health status outcomes, lifestyle characteristics and absenteeism was conducted. Secondly, a detailed process evaluation addressed the implementation of the complex workplace intervention. Interviews were conducted at baseline (27 interviews) and at 7-9 months follow-up (27 interviews) with a purposive sample of workplace stakeholders (managers, caterers and employees). Data were analysed using a thematic framework. Thirdly, micro-costing methods were employed to estimate the cost of implementing and delivering the complex workplace dietary intervention from an employer’s perspective. Finally, an economic evaluation of the complex workplace intervention was conducted. This consisted of 1) a baseline costutility analysis (CUA) which measured the cost-effectiveness of the interventions in terms of quality-adjusted life-years (QALYs), 2) sensitivity analyses to test the robustness of the QALYs which involved performing cost-effectiveness analyses (CEA) using clinical measures to measure health outcomes and 3) a cost-benefit analysis (CBA) where the monetary value of absenteeism was employed so as to report the net benefit of the interventions relative to the control.

Central obesity was positively associated with absenteeism and increased the expected rate of absence by 72% (mean number of absences was 2.5 days). Consuming a highquality diet and engaging in moderate levels of physical activity were negatively associated with absenteeism and reduced the expected frequency by 50% and 36% respectively. The process evaluation revealed that managers’ desire to improve company image, stakeholder buy-in, organisational support and stakeholder cohesiveness facilitated the implementation of the interventions. Anticipation of employee resistance towards menu changes, workplace restructuring and target-driven workplace cultures were found to impede intervention implementation. With regards to the cost-analysis, 3 main cost categories were identified 1) set-up costs 2) maintenance costs and 3) physical assessment costs. The combined intervention xvi reported the highest total costs (€47,305), followed by nutrition education (€44,726), environmental modification (€24,474) and the control (€21,412). In the economic evaluation, the baseline CUA indicated that each intervention (education (€970/QALY) environment (€98/QALY) and combined (€2,156/QALY)) can be considered cost-effective when compared to the control. The CEACs demonstrated that the uncertainty in the incremental costs and effects translated into decision uncertainty for the environment intervention (50% probability of being cost-effective at €45,000/QALY threshold). However, at no point between a ceiling ratio of €0 to €100,000/QALY did the education and combined interventions have a higher probability of being cost-effective than the control. The results of the secondary CEA confirm the baseline CUA results for each intervention. The environment intervention reported the lowest ICERs for: BMI (€14/kg/m2), midway waist circumference (€3/cm) and weight (€7/kg). Furthermore, the environment intervention offers the highest net benefit for employers with a positive net benefit of €145.82 per employee reported.

This thesis provides critical evidence on the relationship between obesity, adverse lifestyle factors and absenteeism. Furthermore, findings also indicate that environmental modification strategies have the potential to offer a cost-effective approach for improving employee health outcomes, depending on the perspective taken. This thesis also highlights the importance of considering contextual factors such as workplace structures and cultures in the development and implementation of future workplace dietary interventions.

Visit Profile

Dr Ronan Grimes

Medicine Use in Type 2 Diabetes in Ireland: An Analysis of Treatment Patterns and Prescription Costs

Institution: TCD, 2012 cohort

Supervisors: Prof. Martin Henman (TCD), Prof. Kathleen Bennett (RCSI), Dr. Lesley Tilson (TCD)

Abstract: Not available

Visit Profile

Dr Conor Keegan

Consumer mobility in the Irish health insurance market: determinants, incentives and risk equalisation

Institution:  TCD, 2012 cohort

Supervisors:  Prof. Steve Thomas (TCD) , Dr. Brian Turner (UCC)

Abstract:  Not available

Visit Profile

Dr Jamie Madden

Exploring Circadian Blood Pressure Patterns

Institution:  UCC, 2012 cohort

Supervisors: Dr. Tony Fitzgerald (UCC), Prof. Patricia Kearney (UCC), Dr. Xia Lee (UCC)

Abstract:       

Blood pressure (BP) is one of the most extensively researched topics. It is well established that elevated BP is the most prevalent treatable risk factor for cardiovascular disease. Despite our comprehensive knowledge of the importance of reducing mean levels of BP, we are less informed about the benefits of reducing other parameters of BP, specifically BP variability (BPV), which refers to the amount of variation over a period of time. Recent evidence has suggested that BPV may be an additional clinical target to mean level. However, its full prognostic significance and definition remains in doubt. BP does not remain stationary throughout the day but is constantly fluctuating and follows a circadian rhythm. Short-term BPV refers to fluctuations over this circadian rhythm. Ambulatory blood pressure monitoring (ABPM), which can be used to obtain estimates of BP usually every 30mins over a 24h period, offers a powerful tool in the analysis of circadian patterns and short term BPV. Longitudinal circadian data with such a cyclical structure consisting of multiple repeated readings provides an opportunity to analyse BP in many different ways and an overview of possible approaches is outlined in this thesis.

The main aims of this thesis were to explore and identify circadian BP patterns between individuals and groups, and extract meaningful measures that describe these patterns while appropriately accounting for the inherent cyclical structure of ABPM data. Specifically, the thesis includes a systematic review which identifies summary measures of BPV, such as standard deviation, that can easily be obtained from the observed data without the need for more advanced modelling.

A meta-analysis exploring the correlation between short-term BPV and subclinical target organ damage (TOD), specifically left ventricular mass index, is included. The association between the identified summary measures and subclinical TOD is then explored in a group of middle aged adults. In an attempt to maximise the power of the repeated cyclical readings in ABPM and incorporate the data together in one model, different random-effects models were explored which allowed us to obtain estimates of both within and between-individual variation of model parameters.

A piece-wise linear mixed-effects model was considered as a simple but suitable approach to capture BP trajectory throughout the day. We attempt to relate factors such as the morning slope and within person variability (allowing this to be group-specific) to TOD (microalbuminuria).

Finally, a two-component cosinor random-effects model is outlined where derivatives of the model fit presents a novel alternative method to locate and quantify the magnitude of slopes at critical points along the trajectory. This is used to obtain a measure of morning BP surge. We compare the random effects from this model to principle component scores obtained through functional principle component analysis. Our motivating data comes from the Mitchelstown Study, a population based study of Irish adults where a subsample underwent 24h ABPM.

Visit Profile

Dr Frank Moriarty

Pharmacoepidemiology and economic evaluation of measures of potentially inappropriate prescribing

Institution:  RCSI, 2012 cohort

Supervisors:  Prof. Tom Fahey (RCSI), Prof. Kathleen Bennett (RCSI), Dr. Caitriona Cahir (RCSI)

Abstract:       

Several measures of potentially inappropriate prescribing (PIP) exist, however their validity has been under-researched. The aim of this thesis was to assess measures of PIP in older and middle-aged people in primary care in terms of their applicability and relevance in Ireland, effect on patient outcomes and economic impact.

This thesis focussed on community-dwelling adults in Ireland, aged ≥65 years (older adults) or 45-64 years (middle-aged adults). Measures of PIP, the Screening Tool for Older Persons’ Prescriptions (STOPP), the Screening Tool to Alter doctors to Right Treatment (START), and PRescribing Optimally in Middle-aged People’s Treatments (PROMPT) criteria, were applied to two national data sources, the General Medical Services (GMS) scheme dispensing database and The Irish Longitudinal Study on Ageing. Economic analysis was also conducted by developing Markov models of PIP.

The prevalence of PIP in older people rose from 1997 to 2012 (32.6%-37.3%), though the odds of having PIP decreased over time after accounting for the increase in medications prescribed. Long-term prescribing of maximal dose proton pump inhibitors grew sharply and was not consistently associated with expected risk factors for gastrointestinal bleeding. PIP was present in 41.6% of middle-aged GMS patients, with prevalent criteria similar to those in older people. For older adults, having ≥2 STOPP criteria was significantly associated with higher rates of emergency department and GP visits, while having ≥2 START omissions was also associated with increased healthcare utilisation, functional decline and reduced quality of life (QoL). In middle-aged people, there was no evidence of a relationship between PROMPT criteria and healthcare utilisation or QoL after controlling for confounders. Of the three PIP criteria evaluated relative to appropriate alternatives in Markov models, long-term benzodiazepine prescribing had the greatest cost and quality-adjusted life year impact, although long-term non-steroidal anti-inflammatory drug use was the most costeffective PIP to target.

This thesis demonstrates that PIP is prevalent and can impact on patient and economic outcomes. Optimising prescribing to reduce PIP may provide benefits for patients and the wider health system.

Visit Profile