Understanding and Appraising the New Medicine Service in England A project funded by the Department of Health
Background to the NMS • 15 million patients in England have a long term medical condition, 813.3 million NHS prescriptions dispensed in 2009-10 • Adherence is poor in key prevalent diseases: • COPD: 33% (Marsden, et al. 2009) • Schizophrenia: 52% (Llorca 2008) • Asthma: 67% (Cerveri, Locatelli et al. 1999) • Diabetes: 78% ( Ho, Rumsfeld et al. 2006) • 15% people receiving a new medicine take few, if any, doses sub-optimal medicines use (assuming appropriate Rx) inadequate management of the LTC and poor outcomes cost to the patient, the NHS and society • Estimated opportunity cost (NHS England) of lost health gain from non- adherence = £930 million p.a. in 5 diseases :(Trueman, Lowson et al. 2010) 2
Hello, it’s the pharmacist calling: an economic evaluation of an intervention to improve adherence • Intervention: pharmacist telephoned 2 weeks after new Rx for chronic illness to discuss medication • Patients: Already on >3 medications: >74 or stroke, cardiovascular disease, asthma, diabetes, RAs • Results at 1 month follow-up • Self-reported non-adherence: 8% versus 16% p=0.030 • medication related problems: 23% versus 34% p=0.019 • Mean total patient costs (NHS): £77.8 versus £113.9 p<0.05 Clifford S, Barber N, Elliott RA, Hartley E, Horne, R. P .W.S. 2006; 28: 165-170 3 Elliott RA, Clifford S, Barber N, Hartley E, Horne R. P .W.S. 2008; 30: 17-23
Cost effectiveness plane for adherence intervention SE quadrant 4
The New Medicine Service Service description This service will provide support to people who are newly prescribed a medicine to manage a long-term condition,* which will generally help them to appropriately improve their medication adherence. Aims and intended outcomes The service should: a) help patients and carers manage newly prescribed medicines for a LTC and make shared decisions about their LTC b) recognise the important and expanding role of pharmacists in optimising the use of medicines * asthma/COPD, hypertension, Type 2 diabetes or anticoagulation/antiplatelet therapy 5
What is the NMS? GP referral to community pharmacist for NMS 7-14 14-21 days days Patient agrees Patient Patient Patient follow-up to adhere to engagement consultation new medicine Patient agrees to adhere to new medicine or pharmacist to resolve Patient identified by medicines-related community issues pharmacist for NMS Refer to GP to Refer to GP to resolve medicines- resolve medicines- related issues related issues
NMS Evaluation Study cast list Project Team Nottingham: Tony Avery, Matthew Boyd (co PI), Loraine Buck, Chris Craig, Rachel Elliott (co PI), Georgios Gkountouras, Asam Latif, Rajnikant Mehta, Ndeshi Salema, Lukasz Tanajewski, Justin Waring, Deborah Watmough London: Nick Barber, James Davies PPI: Antony Chuter Additional Patient Representation: Ember Vincent, Clancy Williams NMS Evaluation Advisory Group: Nick Mays (chair), Alistair Buxton, Sarah Clifford, Ailsa Donnelly, Alan Glanz, Sally Greensmith, Jeanette Howe, Carmel Hughes, Danny Palnoch, Gil Shalom, Gary Warner 7
8
Primary objectives of appraisal Evaluate the impact of the new medicines service (NMS) on • patient medicines-taking behaviour, • patient outcomes, • and cost-effectiveness from an NHS perspective. 9
Technology Appraisal: RCT methods • 504 participants from 47 pharmacies (EMSY/London) • Aged >14, eligible for NMS, identified in a community pharmacy on presentation of prescription for a new medicine for asthma/(COPD), hypertension, type 2 diabetes or an anticoagulant/antiplatelet agent. • Interventions: Randomised to NMS or current practice. • Main outcomes: Adherence to new medicine 10 weeks post recruitment. • The NMS question: ‘Since we last spoke have you missed any doses of your new medicine, or change when you take it (prompt: when did you last miss a dose)?’ • Morisky Medication Adherence Scale (MMAS-8) • Also: EQ-5D 3L, NHS costs, BMQ • Analysis : ITT, outcome adjusted for pharmacy clustering, NMS disease category, age, sex and medication count, multiple imputation for missing 10 data.
Technology Appraisal: RCT results • Follow up: At 10 weeks 85% patients contacted by telephone (n=443), 60% of questionnaires were returned (n=321), 53 patients withdrawn from study. • Adherence: In the unadjusted intention-to-treat analysis of 378 patients still taking the initial medicine: 115/190 (60.5%) and 133/188 (70.7%) (p=0.037) patients were adherent in the current practice and NMS arms, respectively, yielding an odds ratio (95% CI) of 1.58 (1.03, 2.42, p=0.037). In the adjusted analysis: Adherence yielded an odds ratio (95% CI) of 1.67 (1.06, 2.62, p=0.027), in favour of the NMS arm. 11
Technology Appraisal: RCT results Health status : • Mean (SD) EQ-5D at baseline and follow-up: • current practice: 0.73 (0.28) and 0.75 (0.26); • NMS 0.76 (0.28) and 0.77 (0.30). NHS costs at 10 weeks: • Mean (median, range) total NHS cost: • Current practice: £260.87 (121.2, 0-1668.45) • NMS: £215.16 (110.78,0-1458.7) • Difference: £45.71 (95% CI: -33.41- 124.84, p= 0.1281). • This difference reduces to £21.11 once the cost of the NMS intervention is included. 12
Using economic evaluation to determine the impact of a cross-therapeutic adherence intervention • Economic models can tell you the long term health consequences and costs incurred by diseases and treatments. • Need to understand (and therefore need data on): • Disease and treatment pathways • Probability of moving from one disease state to another, and the effect of treatment on that • The quality of life of a person in each disease state • The costs of treating the person in each disease state • Economic models are disease-specific • Safety and adherence interventions are often cross-therapeutic • Use of errors and adherence as proxy outcomes • OR………….. 13
Economic evaluation Markov model* State 2 State 1 Death Probability and resource use Probability, resource use and utility data from trial data from published sources 14 *number and type of health states will depend on the disease/drug group
NMS economic models • The resultant six treatment pathway models are: • Hypertension-amlodipine • Hypertension-ramipril • Asthma-inhaled corticosteroid (beclometasone) • COPD-tiotropium • Diabetes-metformin • Anticoagulants-aspirin • Lifetime time horizon, NHS perspective • Combined with • effect size, age, disease severity, drug being prescribed and health status from NMS RCT • Proportion of disease groups covered by NMS • Intervention costs ` 15
Hypertension-amlodipine model 16
Combining the data from the RCT and the treatment pathway models Cost & QALY caused by Adherence: 10-week ITT analysis non-adherence from models incorporating imputed missing values, for MMAS-8 composite adherence outcome: odds ratio, SD (NMS vs.current practice): 1.81 (1.07, 3.05). Model % NMS cohort p [adherence] NMS group : 63.6% p [adherence] current practice : 49.1% CCB* 25.3% ACE* 24.1% Aspirin 8.5% Cost of NMS intervention : £24.60 Asthma 17.5% COPD 5.8% Diabetes 18.9% Overall 100% 17 Composite economic evaluation
Incremental economic analysis • NMS generated a mean of 0.06 (95%CI: 0.00, 0.16) more QALYs per patient, at a mean reduced cost of -£190 (95%CI: -929, 87). • NMS dominates current practice, with an ICER (95% credibility range) of -£3 005 (-17 213, 4 543) 18
Incremental cost effectiveness ratio 19
Some qualitative findings (20 interviews) • NMS consultations were found to be mutually respectful and polite encounters with discussions generally centred on the new medicine within which issues of use and adherence featured alongside other health-related matters. • Consultations were led from the onset by the pharmacist who routinely dominated the discussion by asking most questions; patients were found to ask fewer questions. • For many pharmacists, their intention was to approach the NMS as an information providing exercise, to support patient use of new medicines. • Not all pharmacists used the NMS interview schedule, for example failing to ask about missed doses. As a consequence, opportunities to discuss adherence in-depth were not always taken. 20
Recommend
More recommend
