PRIMARY HEALTH CARE SPECIAL INTEREST GROUP SESSION
Wednesday 9th September 2015 4.30 - 6pm
Royal Statistical Society Conference, Exeter University. All welcome. One day conference registration is necessary
PRIMARY HEALTH CARE - OUT OF PRACTICE A session taking a look at primary care in other contexts to the more traditional general practice setting.
Speakers:
"STATISTICIAN BEHIND BARS: BY DESIGN AND ON TRIAL" - Sheila Bird, MRC Biostatistics Unit, Cambridge
I shall describe how a quarter century of surveillance designs (with associated biological sample), record-linkage studies, and bespoke "questionnaires" - a phrase coined from Hill and Doll - have improved prisoners' access to harm reduction (eg Hepatitis B immunization) , contributed to changed policy in prisons (e.g. put an end to random mandatory drugs testing), got us barred, yet quantified a 7 times higher risk of overdose death soon after prison-release, and eventually enabled three musketeers to mount the pilot N-ALIVE Trial in England to test whether those randomized to receive naloxone-on-release have 30% fewer opioid-related deaths in the 4-weeks post-release than controls (prior estimate: 1 in 200).
Even before the N-ALIVE Trial's first randomization, however, Scotland became the first country to introduce take-home-naloxone as a funded public health policy. Wales followed in May 2011. I shall describe the trials and tribulations of Scotland's closely-monitored evaluation of its 2011-15 national naloxone policy, which is complicated because Scotland's policy was introduced against a still-rising trajectory of age-related opioid-deaths.
Is its impact for an RCT to be overtaken by the policy it seeks to inform?
INVESTIGATING PRESCRIBING ERRORS IN COMMUNITY PHARMACIES USING A "NATURALISTIC" STEPPED-WEDGE DESIGN - Matthew J Boyd, Sarah Armstrong, Rajnikant Mehta on behalf of the EPS Project team, Division of Social Research in Medicines and Health, School of Pharmacy, University of Nottingham, and NIHR Research Design Service (East Midlands)
Objectives: to compare prevalence and types of dispensing errors and pharmacists' labelling enhancements, for prescriptions transmitted using the NHS electronic prescription service (EPS) versus paper prescriptions in community pharmacy.
Method : dispensed prescription medicines were reviewed by a qualified researcher (pharmacist technician) across 15 English pharmacies and discrepancies between the items requested and those dispensed were recorded on a standardised proforma noting prescription type(electronic/paper based). The study used a stepped wedge design adapted to the natural (rather than research-led) deployment of the intervention.
Clinical Results: 16357 dispensed items were reviewed. Labelling and content errors were identified in 5.4% and 1.4% of items respectively; 13.6% had positive enhancements . Items sent via the EPS had a higher prevalence of labelling errors than paper prescriptions, but a similar rate of content errors and enhancements. The increase in labelling errors was mainly accounted for by one pharmacy omitting the indication from the label and a sensitivity analysis excluding these cases revealed no remaining difference.
Conclusions: items transmitted electronically had higher prevalence of labelling errors, but this was predominantly due to local practice in a single pharmacy. Approximately one in seven prescription items were enhanced by community pharmacists. Medicines supply professionals should work together to agree how items should be dispensed and labelled to best reap the benefits of electronically transmitted prescriptions.
Our experiences in using a stepped wedge in the context of the natural deployment of an intervention are discussed and practical advice for other research studies considering this approach is offered.
WE ONLY RANDOMISE TO TREATMENT HERE! IS THE "COHORT MULTIPLE RANDOMISED CONTROLLED TRIAL" A BETTER ALTERNATIVE TO THE PRAGMATIC RCT? - David Reeves (presenting author), Mark Hann and Peter Bower, Centre for Biostatics and Centre for Primary Care, University of Manchester
The "cohort multiple randomised controlled trial" (cmRCT) design is a fairly recent innovation that aims to overcome some of the limitations of traditional pragmatic RCTs. The design is based upon first recruiting a large cohort of patients who are followed up longitudinally at regular time-periods, then randomly selecting smaller subsets of these to be offered trial interventions. Outcomes in this subset after the intervention are then compared with outcomes in the remainder of the cohort, who constitute the control group. With a large cohort, multiple RCTs can be conducted using different random draws of patient subsets.
At Manchester we are using a cmRCT design as a means of creating a pool of participants for testing out a number of public health initiatives. We have recruited a cohort of 4,000 patients and are in the process of recruiting 250 of these into an embedded trial, "CLASSIC PROTECTS", of a centrally provided telephone-based health coaching scheme supported by a multidisciplinary health and social care team.
In the process, we have found that as well as having a number of appealing features the cmRCT design also raises a number of interesting and important methodological issues. In this talk we will describe our experience of the cmRCT and discuss these issues, including (i) power and sample size relative to a comparable pragmatic RCT; (ii) the impact of non-consent and attrition; (iii) the treatment effect being estimated and implications for analysis; (iv) other potential sources of bias.
Further details about the Royal Statistical Society 2015 Annual Conference can be found at: http://www.rss.org.uk/RSS/Events/RSS_Conference/2015_Conference/RSS/Events/Conference/2015_conference.aspx
Session organisers: Maurice Marchant ([log in to unmask]) and Mark Hann ([log in to unmask])
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|
