Anthony J Avery

The case study approach

The case study approach allows in-depth, multi-faceted explorations of complex issues in their real-life settings. The value of the case study approach is well recognised in the fields of business, law and policy, but somewhat less so in health services research. Based on our experiences of conducting several health-related case studies, we reflect on the different types of case study design, the specific research questions this approach can help answer, the data sources that tend to be used, and the particular advantages and disadvantages of employing this methodological approach. The paper concludes with key pointers to aid those designing and appraising proposals for conducting case study research, and a checklist to help readers assess the quality of case study reports.

The relationship between body mass index and health-related quality of life: comparing the EQ-5D, EuroQol VAS and SF-6D

Background:No other studies have compared the relationship between body mass index (BMI) and health-related quality of life (HRQL) on more than one utility measure. Estimating the HRQL effects of obesity on a (common) utility scale enables the relative cost-effectiveness of interventions designed to alleviate obesity to be estimated.Objective:To examine the relationship between BMI and HRQL according to the EQ-5D, EuroQol visual analogue scale (EQ-VAS) and SF-6D.Methods:Patients aged ⩾45 years at one UK general practice were asked to complete the EQ-5D, EQ-VAS, SF-36 questionnaire (used to derive the SF-6D), and information on their characteristics and co-morbidity. Body mass index was categorized according to the World Health Organization (WHO) recommendations. Regression analysis was used to compare the HRQL of normal BMI patients to the HRQL of patients in other BMI categories, while controlling for patient characteristics and co-morbidity.Results:A total of 1865 patients responded (67%), mean BMI 26.0 kg/m2, 16% obese (BMI⩾30). Patients with back pain, hip pain, knee pain, asthma, diabetes or osteoarthritis were also significantly more likely to be obese. After controlling for other factors, compared to normal BMI patients, obese patients had a lower HRQL according to the EQ-5D (P<0.01), EQ-VAS (P<0.001) and SF-6D (P<0.001). Pre-obese patients were not estimated to have a significantly lower HRQL, and underweight patients were only estimated to have a significantly lower HRQL according to the SF-6D. These results arose because, on the EQ-5D, obese patients were found to have significantly more problems with mobility and pain, compared to physical functioning, social functioning and role limitations on the SF-6D. Whereas, according to the SF-6D, underweight patients had significantly more problems on the dimension of role limitation.Conclusion:The EQ-5D, EQ-VAS and SF-6D were in agreement that, relative to a normal BMI, obesity is associated with a lower HRQL, even after controlling for patient characteristics and co-morbidity. These three measures are thereby sensitive to the HRQL effects of obesity and can be used to estimate the cost-effectiveness of interventions designed to alleviate obesity.

Implementation and adoption of nationwide electronic health records in secondary care in England: final qualitative results from prospective national evaluation in "early adopter" hospitals

Objectives To evaluate the implementation and adoption of the NHS detailed care records service in “early adopter” hospitals in England. Design Theoretically informed, longitudinal qualitative evaluation based on case studies. Setting 12 “early adopter” NHS acute hospitals and specialist care settings studied over two and a half years. Data sources Data were collected through in depth interviews, observations, and relevant documents relating directly to case study sites and to wider national developments that were perceived to impact on the implementation strategy. Data were thematically analysed, initially within and then across cases. The dataset consisted of 431 semistructured interviews with key stakeholders, including hospital staff, developers, and governmental stakeholders; 590 hours of observations of strategic meetings and use of the software in context; 334 sets of notes from observations, researchers’ field notes, and notes from national conferences; 809 NHS documents; and 58 regional and national documents. Results Implementation has proceeded more slowly, with a narrower scope and substantially less clinical functionality than was originally planned. The national strategy had considerable local consequences (summarised under five key themes), and wider national developments impacted heavily on implementation and adoption. More specifically, delays related to unrealistic expectations about the capabilities of systems; the time needed to build, configure, and customise the software; the work needed to ensure that systems were supporting provision of care; and the needs of end users for training and support. Other factors hampering progress included the changing milieu of NHS policy and priorities; repeatedly renegotiated national contracts; different stages of development of diverse NHS care records service systems; and a complex communication process between different stakeholders, along with contractual arrangements that largely excluded NHS providers. There was early evidence that deploying systems resulted in important learning within and between organisations and the development of relevant competencies within NHS hospitals. Conclusions Implementation of the NHS Care Records Service in “early adopter” sites proved time consuming and challenging, with as yet limited discernible benefits for clinicians and no clear advantages for patients. Although our results might not be directly transferable to later adopting sites because the functionalities we evaluated were new and untried in the English context, they shed light on the processes involved in implementing major new systems. The move to increased local decision making that we advocated based on our interim analysis has been pursued and welcomed by the NHS, but it is important that policymakers do not lose sight of the overall goal of an integrated interoperable solution.

Effect of pay for performance on the management and outcomes of hypertension in the United Kingdom: interrupted time series study

Objective To assess the impact of a pay for performance incentive on quality of care and outcomes among UK patients with hypertension in primary care. Design Interrupted time series. Setting The Health Improvement Network (THIN) database, United Kingdom. Participants 470 725 patients with hypertension diagnosed between January 2000 and August 2007. Intervention The UK pay for performance incentive (the Quality and Outcomes Framework), which was implemented in April 2004 and included specific targets for general practitioners to show high quality care for patients with hypertension (and other diseases). Main outcome measures Centiles of systolic and diastolic blood pressures over time, rates of blood pressure monitoring, blood pressure control, and treatment intensity at monthly intervals for baseline (48 months) and 36 months after the implementation of pay for performance. Cumulative incidence of major hypertension related outcomes and all cause mortality for subgroups of newly treated (treatment started six months before pay for performance) and treatment experienced (started treatment in year before January 2001) patients to examine different stages of illness. Results After accounting for secular trends, no changes in blood pressure monitoring (level change 0.85, 95% confidence interval −3.04 to 4.74, P=0.669 and trend change −0.01, −0.24 to 0.21, P=0.615), control (−1.19, −2.06 to 1.09, P=0.109 and −0.01, −0.06 to 0.03, P=0.569), or treatment intensity (0.67, −1.27 to 2.81, P=0.412 and 0.02, −0.23 to 0.19, P=0.706) were attributable to pay for performance. Pay for performance had no effect on the cumulative incidence of stroke, myocardial infarction, renal failure, heart failure, or all cause mortality in both treatment experienced and newly treated subgroups. Conclusions Good quality of care for hypertension was stable or improving before pay for performance was introduced. Pay for performance had no discernible effects on processes of care or on hypertension related clinical outcomes. Generous financial incentives, as designed in the UK pay for performance policy, may not be sufficient to improve quality of care and outcomes for hypertension and other common chronic conditions.

Investigation into the reasons for preventable drug related admissions to a medical admissions unit: observational study

Objective: To describe the drugs and types of medicine management problems most frequently associated with preventable drug related admissions to an acute medical admissions unit. Design: Observation study. Setting: Medical admissions unit in a teaching hospital in Nottingham, UK. Participants: 4093 patients seen by pharmacists on the medical admissions unit between 1 January and 30 June 2001. Main outcome measures: Proportion of admissions that were drug related and preventable, classification of the underlying causes of preventable drug related admissions, and identification of drugs most commonly associated with preventable drug related admissions. Results: Of the admissions seen by pharmacists, 265 (6.5%) were judged to be drug related and 178 (67%) of these were judged to be preventable. Preventable admissions were mainly due to problems with prescribing (63 cases (35%)), monitoring (46 cases (26%)), and adherence to medication (53 cases (30%)). The drugs most commonly implicated were NSAIDs, antiplatelets, antiepileptics, hypoglycaemics, diuretics, inhaled corticosteroids, cardiac glycosides, and beta-blockers. Conclusions: Potentially preventable drug related morbidity was associated with 4.3% of admissions to a medical admissions unit. In 91% of cases these admissions were related to problems with either prescribing, monitoring, or adherence.

Interventions in primary care to reduce medication related adverse events and hospital admissions: systematic review and meta-analysis

Objective: To identify and evaluate studies of interventions in primary care aimed at reducing medication related adverse events that result in morbidity, hospital admission, and/or mortality. Methods: Fourteen electronic databases were systematically searched for published and unpublished data. Bibliographies of retrieved papers were searched and experts and first authors contacted in an attempt to locate additional studies. There were no restrictions on language of publication. All interventions applied in primary care settings which aimed to improve patient safety by reducing adverse events resulting from medication overuse or misuse were considered. Randomised controlled trials, controlled trials, controlled before and after studies, and interrupted time series studies were eligible for inclusion. Study quality assessment and data extraction were undertaken using the Cochrane Effective Practice and Organisation of Care data collection checklist and template. Meta-analysis was performed using a random effects model. Results: 159 studies were initially identified, of which 38 satisfied our inclusion criteria. These were categorised as follows: 17 pharmacist-led interventions (of which 15 reported hospital admissions as an outcome); eight interventions led by other primary healthcare professionals that reported preventable drug related morbidity as an outcome; and 13 complex interventions that included a component of medication review aimed at reducing falls in the elderly (the outcome being falls). Meta-analysis found that pharmacist-led interventions are effective at reducing hospital admissions (OR 0.64 (95% CI 0.43 to 0.96)), but restricting analysis to the randomised controlled trials failed to demonstrate significant benefit (OR 0.92 (95% CI 0.81 to 1.05)). Pooling the results of studies in the other categories did not demonstrate any significant effect. Conclusions: There is relatively weak evidence to indicate that pharmacist-led medication reviews are effective in reducing hospital admissions. There is currently no evidence for the effectiveness of other interventions which aim at reducing admissions or preventable drug related morbidity. More randomised controlled trials of primary care based pharmacist-led interventions are needed to decide whether or not this intervention is effective in reducing hospital admissions.

Implementation and adoption of nationwide electronic health records in secondary care in England: qualitative analysis of interim results from a prospective national evaluation

Objectives To describe and evaluate the implementation and adoption of detailed electronic health records in secondary care in England and thereby provide early feedback for the ongoing local and national rollout of the NHS Care Records Service. Design A mixed methods, longitudinal, multisite, socio-technical case study. Setting Five NHS acute hospital and mental health trusts that have been the focus of early implementation efforts and at which interim data collection and analysis are complete. Data sources and analysis Dataset for the evaluation consists of semi-structured interviews, documents and field notes, observations, and quantitative data. Qualitative data were analysed thematically with a socio-technical coding matrix, combined with additional themes that emerged from the data. Main results Hospital electronic health record applications are being developed and implemented far more slowly than was originally envisioned; the top-down, standardised approach has needed to evolve to admit more variation and greater local choice, which hospital trusts want in order to support local activity. Despite considerable delays and frustrations, support for electronic health records remains strong, including from NHS clinicians. Political and financial factors are now perceived to threaten nationwide implementation of electronic health records. Interviewees identified a range of consequences of long term, centrally negotiated contracts to deliver the NHS Care Records Service in secondary care, particularly as NHS trusts themselves are not party to these contracts. These include convoluted communication channels between different stakeholders, unrealistic deployment timelines, delays, and applications that could not quickly respond to changing national and local NHS priorities. Our data suggest support for a “middle-out” approach to implementing hospital electronic health records, combining government direction with increased local autonomy, and for restricting detailed electronic health record sharing to local health communities. Conclusions Experiences from the early implementation sites, which have received considerable attention, financial investment and support, indicate that delivering improved healthcare through nationwide electronic health records will be a long, complex, and iterative process requiring flexibility and local adaptability both with respect to the systems and the implementation strategy. The more tailored, responsive approach that is emerging is becoming better aligned with NHS organisations’ perceived needs and is, if pursued, likely to deliver clinically useful electronic health record systems.

GPs' views on computerized drug interaction alerts: questionnaire survey

Background: There is evidence that patients are being prescribed potentially hazardous drug–drug combinations in general practice despite the use of computerized drug interaction alert systems. One reason for this may be that general practitioners (GPs) are overriding these alerts without properly checking them.

A review of human factors principles for the design and implementation of medication safety alerts in clinical information systems

The objective of this review is to describe the implementation of human factors principles for the design of alerts in clinical information systems. First, we conduct a review of alarm systems to identify human factors principles that are employed in the design and implementation of alerts. Second, we review the medical informatics literature to provide examples of the implementation of human factors principles in current clinical information systems using alerts to provide medication decision support. Last, we suggest actionable recommendations for delivering effective clinical decision support using alerts. A review of studies from the medical informatics literature suggests that many basic human factors principles are not followed, possibly contributing to the lack of acceptance of alerts in clinical information systems. We evaluate the limitations of current alerting philosophies and provide recommendations for improving acceptance of alerts by incorporating human factors principles in their design.

Self-reported medication use for older people in England and Wales

Objectives: To describe the patterns of self‐reported medication use, including both prescription and non‐prescription drugs, for older people in five areas in England and Wales, and to identify the effects of socio‐demographic features on medication use.