Sheet Music

A report for the Department of Health Patient Safety Research Portfolio February PDF

Description
Economic evaluation of a pharmacist-led IT-based intervention with simple feedback in reducing rates of clinically important errors in medicines management in general practices (PINCER) A report for the
Categories
Published
of 28
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
Economic evaluation of a pharmacist-led IT-based intervention with simple feedback in reducing rates of clinically important errors in medicines management in general practices (PINCER) A report for the Department of Health Patient Safety Research Portfolio February 2013 Rachel A Elliott 1, Koen Putman 2, Matthew Franklin 1, Nick Verhaeghe 2, Lieven Annemans 2 Martin Eden 3, Jasdeep Hayre 4, Sarah Rodgers 5, Judith A Cantrill 3, Sarah Armstrong 6, Kathrin Cresswell 7, Julia Hippisley-Cox 8, Rachel Howard 9, Denise Kendrick 8, Caroline J Morris 10, Scott A Murray 7, Robin J Prescott 7, Glen Swanwick 10, Matthew Boyd 1, Lorna Tuersley 3, Tom Turner 10, Yana Vinogradova 8, Aziz Sheikh 7, Anthony J Avery 8 1 Division for Social Research in Medicines and Health, The School of Pharmacy, University of Nottingham, University Park, Nottingham, NG7 2RD, UK 2 Department of Medical Sociology and Health Sciences, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Laarbeeklaan 103 B-1090 Brussel, Belgium 3 Drug Usage & Pharmacy Practice Group, School of Pharmacy & Pharmaceutical Sciences, University of Manchester, Oxford Road, Manchester, M13 9PL, UK 4 National Institute of Health and Clinical Excellence, Level 1A, City Tower, Piccadilly Plaza, Manchester, M1 4BT 5 Research and Evaluation Team, Quality and Governance Directorate, NHS Nottinghamshire County, Birch House, Southwell Road West, Mansfield, Nottinghamshire NG21 0HJ 6 Trent Research Design Service, Division of Primary Care, Tower Building, University Park, Nottingham, NG7 2RD, UK 7 Centre for Population Health Sciences, University of Edinburgh, Teviot Place, Edinburgh, EH8 9AG, UK 8 Division of Primary Care, University of Nottingham Medical School, Queen s Medical Centre, Nottingham, NG7 2UH, UK. 9 School of Pharmacy, University of Reading, PO Box 226, Whiteknights, Reading, RG6 6AP, UK 10 Department of Primary Health Care and General Practice, Wellington School of Medicine and Health Sciences, University of Otago, Mein Street, Wellington South, New Zealand 10 Consumers in Research Advisory Group, c/o: Research and Evaluation Team, Quality and Governance Directorate, NHS Nottinghamshire County, Birch House, Southwell Road West, Mansfield, Nottinghamshire NG21 0HJ Corresponding author: Professor Rachel A Elliott Division for Social Research in Medicines and Health, The School of Pharmacy, University of Nottingham, University Park, East Drive, Nottingham. NG7 2RD 1 address: Telephone: Competing interests: none Trial registration: Current controlled trials ISRCTN Abstract Title Economic evaluation of a pharmacist-led IT-based intervention with simple feedback in reducing rates of clinically important errors in medicines management in general practices, based on a cluster randomised trial (PINCER). Authors Rachel A Elliott, Koen Putman, Matthew Franklin, Nick Verhaeghe, Lieven Annemans, Martin Eden, Jasdeep Hayre, Sarah Rodgers, Judith A Cantrill, Sarah Armstrong, Kathrin Cresswell, Julia Hippisley-Cox, Rachel Howard, Denise Kendrick, Caroline J Morris, Scott A Murray, Robin J Prescott, Glen Swanwick, Matthew Boyd, Lorna Tuersley, Tom Turner, Yana Vinogradova, Aziz Sheikh, Anthony J Avery. Background Medication errors in general practice are considered an important source of potentially preventable morbidity and mortality. There is also a usually implicit assumption that improving safety is a good thing even though most errors documented are minor and unlikely to affect patient outcome and associated cost. Initiatives to reduce medication errors are usually costly. In an increasingly financially constrained healthcare environment, it is essential to be clearer about the true economic impact of medication error reduction. Objectives The overall aim of this study was to determine the cost-effectiveness associated with a pharmacist - led IT-based intervention to reduce rates of potentially harmful prescribing and monitoring errors in general practices (PINCER). Methods The economic analysis compared the costs and health benefits of a pharmacist-led IT-based intervention (PINCER) with simple feedback in reducing rates of six clinically important errors in medicines management in general practices. An economic evaluation was carried out to determine the cost per extra quality-adjusted life-year (QALY) generated, from the perspective of the National Health Service (NHS). This analysis combined the results from the PINCER trial with error-specific projected harm and NHS cost to allow generation of estimates of overall patient benefit and NHS costs. Six error-specific treatment pathway Markov models were constructed to quantify the economic impact of the medication errors included in the PINCER intervention. Incremental cost effectiveness ratios, cost effectiveness acceptability curves and net benefit, whereby a monetary value to QALYs was assigned, were generated. Results from the base case analysis were tested using sensitivity and scenario analysis. 3 Results In the probabilistic analysis, PINCER was cost-saving (mean ICER was per QALY gained (SD 97,460; median - 159; 2.5th percentile: - 23,939; 97.5th percentile 21,767). At a ceiling willingness to pay of 20,000, the PINCER pharmacist intervention reaches 59% probability of being cost effective. The probability of PINCER being cost effective does not increase beyond 59%. The net benefit statistic generated suggests a mean of 16 net benefit (SD 121; median 22; 2.5th percentile: - 218; 97.5th percentile 242), at a ceiling willingness to pay for a QALY of The mean cost per QALY generated suggested that PINCER increased health gain at a cost per QALY well below most accepted thresholds for implementation. However, the range around this ICER is extremely wide, reflecting the large degree of uncertainty around effect in some of the individual outcome models. If the PINCER intervention targeted one of the errors only, the mean (SE) costs per QALY generated were: NSAIDs prescribing: cost-saving ( , 94); Betablockers prescribing: cost-saving (- 2381, 3906); ACEI monitoring: ( 18008); Methotrexate monitoring: 2060 ( 4654); Lithium monitoring: cost-saving ( , ); Amiodarone monitoring: 475 ( 15). Targeting NSAID prescribing and amiodarone monitoring errors were the most cost effective activities within the PINCER intervention. These were also the models with the most data to support them. Varying the cost of the intervention or the practice size had a negligible effect on results. Conclusions This study estimated the economic impact of a safety-focused intervention in health care, which is known to be effective in reducing rates of key prescribing and monitoring errors in general practice. The intervention was more effective and less costly than the alternative but the huge levels of uncertainty present in the analysis meant that the PINCER intervention could not be considered cost effective with a large degree of certainty under current decision rules. However, correction of some errors has a larger clinical and economic effect, such that the PINCER intervention could be cost effective if the right errors are targeted. Conclusions from this economic analysis are hampered by the paucity of data around the real clinical and economic impact of medication errors. Better evidence on the impact of errors is required. Further work is required to address the economic impact of including other errors not included in the PINCER intervention. More importantly, given that reducing medication errors may produce non-health benefits such as trust and increased engagement with the health service, the role of cost effectiveness in allocating resources to safety-focused interventions in health care needs to be examined and explored. 4 List of Abbreviations ACE: ADE: CEAC: CHD: CPOE: CTU: DMEC: EMIS: GP: ICC: ICER: IMD: INR: IT: Li: MRC: NHS: NPSA: NSAIDs: ONS: OR: PCT: PPI: TFT: TPP: TSC: U&E: Angiotensin converting enzyme (inhibitor) Adverse drug event Cost effectiveness acceptability curve Coronary heart disease Computerised physician order entry Clinical Trials Unit Data Monitoring and Ethics Committee Egton Medical Information Systems (the name of a GP computer system) General practitioner (or family practitioner) Intraclass correlation coefficient Incremental cost effectiveness ratio Index of Multiple Deprivation International normalised ratio Information technology Lithium Medical Research Council The UK National Health Service National Patient Safety Agency non-steroidal anti-inflammatory drugs Office for National Statistics odds ratio Primary Care Trust proton pump inhibitor Thyroid Function test The Phoenix Partnership (the name of a GP computer system) Trial Steering Committee Urea and electrolytes 5 Table of contents 1 Background Estimating the true economic impact of medication error reduction What is the economic impact of medication errors? What is the economic impact of interventions to reduce medication error rates? 18 2 Work already completed by this research team Summary of PINCER trial methods Study sites and patient participants Study interventions Simple feedback Pharmacist intervention Study outcomes Summary of PINCER findings Within-trial PINCER economic analysis Methods Overall rationale Aims and objectives Methodology Model specification Sources of clinical outcome, health status and resource use data Incremental economic analysis Sensitivity and scenario analysis Results Results from outcome measure-specific models Patients with a past medical history of peptic ulcer who have been prescribed a non-selective NSAID and no PPI Patients with a history of asthma who have been prescribed a beta-blocker 35 6 4.1.3 Patients aged 75 years and older who have been prescribed an Angiotensin- Converting Enzyme Inhibitor (ACEI) long-term who have not had a recorded check of their renal function and electrolytes in the previous 15 months Patients receiving methotrexate for at least three months who have not had a recorded full blood count and/or liver function test within the previous three months Patients receiving lithium for at least three months who have not had a recorded check of their lithium levels within the previous three months Patients receiving amiodarone for at least six months who have not had a thyroid function test within the previous six months Summary of outputs for outcome-measure specific models Incremental analysis of PINCER intervention Deterministic incremental analysis Probabilistic incremental analysis Scenario and sensitivity analysis Discussion Key findings from individual models Key findings from composite error PINCER model Strengths and limitations Using economic evaluation to evaluate safety in health care Implications for policy makers and practitioners Priorities for future research Conclusions Source of funding Acknowledgements References Appendix 1: Patients with a past medical history of peptic ulcer who have been prescribed a non-selective NSAID and no PPI Introduction Aim of the study 7.3 Literature search Decision-analytic model for economic analysis The decision-analytic model Probabilities of moving from one state to another Required resource use and unit costs Utility weights for health states Appendix 2 Patients with a history of asthma who have been prescribed a betablocker Introduction Aim of the study Literature search Decision-analytic model for economic analysis The decision-analytic model Probabilities of moving from one state to another Utility weights for health states Required resource use and unit costs Appendix 3: Patients aged 75 years and older who have been prescribed an Angiotensin-Converting Enzyme Inhibitor (ACEI) long-term who have not had a recorded check of their renal function and electrolytes in the previous 15 months Introduction Aim of the study Literature search Decision-analytic model for economic analysis The decision-analytic model Defining Hyperkalaemia and Acute Renal Failure for the model Probabilities of moving from one state to another Utility weights for health states Required resource use and unit costs 10 Appendix 4: Patients receiving methotrexate for at least three months who have not had a recorded full blood count and/or liver function test within the previous three months Introduction Aim of the study Literature search The decision-analytic model Probabilities of moving from one state to another Derivation of probabilities for the not monitored group Derivation of probabilities for the monitored group Required resource use and unit costs (Table 14) Utility weights for health states Appendix 5: Patients receiving lithium for at least three months who have not had a recorded check of their lithium levels within the previous three months Introduction Aim of the study Literature search Decision-analytic model for economic analysis The decision-analytic model Modelling efficacy Modelling toxicity Varying definitions of relapse Model structure: Markov states How does monitoring affect outcome? Health state weights Resource use associated with each Markov state Appendix 6 - Patients receiving amiodarone for at least six months who have not had a thyroid function test within the previous six months Introduction Aim of the study 12.3 Literature search Decision-analytic model for economic analysis Model population Defining AIT and AIH Incidence of AIH and AIT Treatment of AIH Treatment of Type I and Type II AIT The decision-analytic model Markov states Effects of amiodarone not included in the model How does monitoring affect outcome? Transition probabilities for the model No Symptoms -- AIH or AIT (same value for error and non-error model) No Symptoms -- Death (same value for error and non-error model) AIT untreated -- AIT surgical management (different values for error and nonerror model) AIT untreated -- AIT medical management (different values for error and nonerror model) AIT untreated -- Death (same value for error and non-error model) AIT surgical management -- Post treated AIT (same value for error and nonerror model) AIT surgical management -- Death (same value for error and non-error model) AIT medical management -- Post treated AIT (same value for error and nonerror model) AIT medical management -- Death (same value for error and non-error model) Post treated AIT -- Post treated AIT (same value for error and non-error model) Post treated AIT -- Death (same value for error and non-error model) AIH untreated -- AIH medical management (different values for error and nonerror model) AIH untreated -- Death (same value for error and non-error model) AIH untreated -- AIH untreated (different values for error and non-error model) Post treated AIH -- Death (same value for error and non-error model) AIH treated -- AIH treated (same value for error and non-error model) Health status valuations Resource use associated with each Markov state No Symptoms Untreated AIH Treated AIH Untreated AIT AIT Medical Management AIT Surgical Management AIT Post-treated Death List of tables Table 1 Characteristics of practices and patients at baseline by treatment arm Table 2 Prevalence of prescribing and monitoring problems at six months follow-up by treatment arm Table 3 Simple feedback and PINCER intervention arm costs and error rates and incremental economic analysis Table 4 Probabilities for the 3-month cycle Markov model in the error and non-error groups (NSAIDs) Table 5 Summary of utility weights and cost per health state for NSAID model Table 6 Probabilities for the 3-month cycle Markov model in the error and non-error groups (beta-blockers) Table 7 Summary of utility weights and cost per health state for beta-blocker model Table 8 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups for ACEI Table 9 Summary of utility weights and cost per health state for ACEI model Table 10 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups (methotrexate) Table 11 Summary of utility weights and cost per health state for methotrexate model Table 12 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups (lithium) Table 13 Summary of utility weights and cost per health state for lithium model Table 14 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups (amiodarone) Table 15 Summary of utility weights and cost per health state for amiodarone model Table 16 Summary of key cost and outcome parameters derived from each outcome measure-specific model Table 17 Summary of inputs and ICERs generated for deterministic incremental analysis of PINCER intervention versus simple feedback Table 18 ICERs, percentage ICERs in each quadrant and probability of cost effectiveness at λ for base case, sensitivity and scenario analyses Table 19 Probabilities for the 3 month-cycle Markov model in the error group for NSAIDs.. 77 Table 20 Probabilities for the 3 month-cycle Markov model in the non-error group for NSAIDs Table 3 Sources of unit costs for NSAIDs Table 4: Health states for Markov model (NSAIDs) Table 5 Probabilities for the 3-month cycle Markov model in the error groups for Betablockers Table 6 Probabilities for the 3-month cycle Markov model in the non-error groups for Betablockers Table 7 Health states for Markov model (Beta-blockers) Table 8 Sources of unit costs (Beta-blockers) Table 9 Cost per patient for each health state (Beta-blockers) Table 10 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups for ACEI Table 11 Derivation of transition probability from no symptoms to hyperkalaemia (ACEI) Table 12 Summary of resource use and cost in each ACEI health state Table 13 Probabilities for the 3-month cycle Markov model in the monitored and not monitored groups (methotrexate) Table 14 Summary of resource use and costs in each health state in methotrexate Table 15 Probabilities for the 3-month cycle Markov model in the error and non-error groups for lithium Table 16 Health status weights for lithium model Table 17 Costs of TDM carried out for regularly monitored lithium patients (lithium) Table 18 Healthcare professional resource use in a cycle without an adverse event (lithium) Table 19 Resource use and unit costs for stable (supra-therapeutic/therapeutic) state for lithium Table 20 Resource use and unit costs for stable (sub-therapeutic) state for lithium Table 21 Healthcare provider resource use in a cycle with a manic relapse (lithium) Table 22 Healthcare provider resource use in a cycle with a depressive relapse (OM7) Table 23 Enhanced Outpatient Care (EOC) resource use in lithium model Table 24 Healthcare professional resource use in a cycle with an adverse event (excluding an event that uses EOP) in lithium model Table 25 Resource use and unit costs for relapse in lithium model: manic state Table 26 Resource use and unit costs for relapse in lithium model: depressive state Table 27 Transition costs in lithium model Table 28 Summary of biochemistry and treatment for amiodarone-induced hyper- and hypothyroidism (amiodarone) Table 29 Transition probabilities for the error group (amiodarone) Table 30 Transition Probabilities that differ for the non-error group (amiodarone) Table 31 Health status valuations for each Markov state (amiodarone) Table 32 Resource use and unit costs for No Symptoms (amiodarone) Table 33 Resource use and unit costs for AIH-untreated (amiodarone) Table 34 Resource use and unit costs for treated AIH (amiodarone) Table 35 Resource use and unit costs for AIT-untreated (amiodarone) Table 36 Resource use and unit costs for AIT medical management (amiodarone) Table 37 Resource use and unit costs for AIT surgical management (amiodarone) Table 38 Resource use and unit costs for AIT Post-treated (amiodarone) List of Figures Figure 1 A decision analytic model of pharmacist intervention versus simple feedback in patients at risk of error Figure 2 Overview of economic mod
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks