For this study I am particularly interested in measuring;
Quantitative
- the frequency of adherence/non-adherence
Qualitative
- patterns in non-adherence
- causes for non-adherence
- barriers to adherence
Direct observation would be the most precise way of measuring medication adherence. However, this method would not be feasible for this study. Direct observation would also likely influence the participant’s behaviour. Weighing up the cost/benefits, it has been decided to look at other ways of measuring the outcome.
Potential methods
- Self-reporting
- Medical Adherence Questionnaire
- Visual analogue scale
- pill count
- Likert scale
Asking the participant to inform me whether or not they have taken their medication during the trial could act as another prompt or cue; influencing their behaviour.
Automatic pill count
In an ideal world, I’d use a Medication Event Monitoring System (MEMS) to measure the response to the trigger. The below example shows a pill bottle (more commonly used in the US) that logs the time and frequency of when it was opened. This would allow the researcher to evaluate whether or not their trigger resulted in immediate action by the participant by noting if the desired action was executed after the cue was received.
Image source – medscape.org
I, however, do not have access to this type of technology.
Manual pill count – how many are left over
Counting the number of remaining pills after the experiment is completed would allow me to see how many times it was missed. This method on its own is imprecise as there is a possibility that the participant might double dose to complete their medication, especially as they know the placebo is not a danger to them, unlike real medicine.
Post study survey
Many adherence scales and rating tools have been developed in the medical industry to assess patients adherence. The most relevant for my study are Brief Medication Questionnaire 1 (BMQ1), Morisky Medication Adherence Scale (MMAS-8), Medication Adherence Rating Scale (MARS), Medication Adherence Questionnaire (MAQ), Hill-Bone Compliance Scale, Brief Adherence Rating Scale (BARS) and the Simplified Medication Adherence Questionnaire (SMAQ). These tools use either likert scales or answers that can be codified.
Ranges vary between 3 point scales and 5 point scales. Examples include;
- None, a little, a lot
- not at all confident, somewhat confident, very confident
- never, rarely, usually, often, always
- never, 1-2 times, 3-5 times, 6-10 times, more than 10 times
- always, almost always, frequently, sometimes, rarely, never
- none of the time, some of the time, most of the time, all the time, not applicable, don’t know
- true, false
- no, yes
Example questions
(Adapted BMQ1) – Did you take any of your prescription?
(Adapted BMQ1) – How many days did you take your prescription?
(MMAS-8) – People sometimes miss taking their medication for reasons other than forgetting. Thinking over the last week were there any days when you did not take your medicine?
(MMAS-8) – How often do you have difficulty remembering to take all your medicine?
(MMAS-8) – How often do you have difficulty remembering to take all your medicine?
(BMQ1) – How many times did you miss taking a pill?
Post study interview
An interview asking about the participants behaviour would help to learn;
- If there were any leftover pills;
- why
- what caused them to miss pills
- when they missed the pills (time, date, occasion, etc.)
- how the system could be adapted to persuade users to take their medication
- how effective they found the intervention
- how efficient they found the intervention
- the overall satisfaction with the intervention
Final adherence measurement instrument
I created a guided diagram form to be completed during the interview which would allow further discussion about the participants adherence instead of using the medication adherence surveys such as BMQ1 or MMAS-08. This option was preferable because it allows me to collect qualitative and quantitative data about adherence.
The visual representation of their adherence allows us to see when they took their medication and discuss why they missed certain days and times.
Combined with the scale of “easy/hard to do”, we can understand the context of their medication regime.
Design for scale
I looked at online surveys while considering how the study could scale-up without exponentially increasing the amount of time it takes to analyse the results. The examples below show example variations of the adherence questions.
The iconography in the single choice option isn’t quite right but the concept was to allow the participant to make one single selection instead of multiple clicks.