Pharmacy research falls short

A systematic review has identified areas of pharmacy-related research that were not conducted to the highest standard or where reporting is inadequate

Improvements can be made to pharmacy research practice so findings can have greater validity and therefore impact, say academics from the University of Western Australia and the University of Aberdeen, UK.

In a new study, the researchers identified 50 papers through MEDLINE that reported results of pharmacy-related randomised controlled trials (RCTs), published between 2015 and 2017.

Pilot studies, non-English language papers and those not available online were excluded.

Each paper was read and analysed according to the Cochrane risk of bias tool and the CONSORT checklist—a comprehensive checklist that facilitates the reporting of transparent and high-quality RCT publications.

Most papers were from North America (n = 20), followed by Asia and Europe (both n = 12), Australasia (n = 4), and South America and Africa (both n = 1).

They covered a range of clinical (n = 25), medication management/use (n = 17), public health (n = 6) and continuing professional development (n = 2) topics.

The median score for risk of bias across the papers was 53% (interquartile range 38.5-68.5), with high risk of bias due to inadequate allocation concealment and blinding (66% and 56% of papers, respectively).

Evidence-based pharmacy practice requires a dependable evidence base.

Only one RCT out of 50 met all the quality standards in every domain of the Cochrane risk of bias tool.

Meanwhile the median level of compliance with the CONSORT checklist for RCTs was 64%.

The highest CONSORT score achieved by one paper was 96.9% while the lowest was 35.5%.

There was no correlation between Cochrane or CONSORT scores and journal impact factor.

Summary of CONSORT checklist findings:

    • 100% of the 50 papers included scientific background and explanation of rationale
    • 98% included specific objectives or hypotheses
    • 95% included interpretation consistent with results, balancing benefits and harms and considering other relevant evidence
    • 36% included description of trial design including allocation ratio, while 64% partially did so
    • 60% included how sample size was determined, but 38% did not
    • 54% did not include type of randomisation or details of any restriction such as blocking and block size
    • 54% did not specify who generated the random allocation sequence, who enrolled participants and who assigned participants to interventions
    • 70% did not describe the mechanism used to implement the random allocation sequence or describe any steps taken to conceal the sequence until interventions were assigned
    • 96% did not say why the trial ended or was stopped

“Evidence-based pharmacy practice requires a dependable evidence base. The evidence from RCTs is the usual basis for meta-analysis and systematic review, which informs policy and practice,” say the authors in the International Journal of Pharmacy Practice.

Their review “identified areas of pharmacy-related research which are either not conducted to the highest standard or where the reporting is inadequate”.

Researchers conducting RCTs on pharmacy services need to focus on including details of randomisation, sequence generation and allocation concealment, they say.

research graph results study
54% of RCTS analysed did not include type of randomisation or details of restriction

The results “highlight the need for identifying ways in which to update researchers on the standards expected, to allow pharmacy research findings to have the greatest validity and therefore impact,” they say.

“Future lines of enquiry could also explore associations with other potential predictors of research quality such as funding type (e.g. Research Council or commercial or none), or reporting of earlier feasibility and piloting.

“Qualitative work could also be conducted to understand the reason for deficiencies and introduce interventions to address them. For example, is it lack of expertise and expert advice, lack of funding and/or lack of time?”

The authors urge all researchers to access and apply the relevant research guidelines that exist.

Read the full article in the International Journal of Pharmacy Practice here

Previous ‘Massive and abrupt’ drop in codeine poisonings
Next Clinical tips: Bipolar disorder and impact on diabetes

NOTICE: It can sometimes take awhile for comment submissions to go through, please be patient.