Randomised Controlled Trial on the effect of MI on hearing aid use
A pilot RCT study was conducted by Aazh (2016a); (Aazh 2016b) at Department of Audiology Royal Surrey County Hospital, Guildford, in collaboration with London School of Hygiene and Tropical Medicine in order to explore feasibility and preliminary effect of audiologist-delivered MI on hearing aid use.
This was a pilot single-blind, randomised parallel-group study where participants were randomly assigned to one of the two groups: MI combined with Standard Care (MISC) (n=20) or Standard Care only (SC) (n=17). SC involved adjusting the hearing aids, and providing instructions and advice. MISC involved MI combined with SC tasks (further details of the interventions are provided later on in the text).
The outcome measurement tools utilised in the pilot RCT were (1) the data logging system on hearing aids which measures the average number of hours that a hearing aid is used per day, and (2) a range of validated self-report questionnaires.
The mean number of hours per day that people used their hearing aid(s) increased from 1 hour per day (SD=1.5) at the baseline to 7 hours per day (SD=3.7) one month after the intervention in the MISC group (n=19, p<0.001) and from 1.3 hours per day (SD=2) to 4 hours per day (SD=3.6) in the SC group (n=17, p<0.001). There was a large between-groups effect size of Cohen’s d = 0.98 (95% CI: 0.3 to 1.7) for hearing aid use.
Scores on self-report questionnaires showed improvement in both groups, however the between-groups effect sizes were small (see Aazh (2016a) for details). It is very important to establish whether the intervention in the MISC group was consistent with MI and the intervention in the SC group did not involve any aspects of MI. Therefore, all sessions were coded by independent coders using motivational interviewing treatment integrity (MITI 3.1.1) (Moyers et al. 2010). MISC sessions were coded as consistent with MI while SC did not involve any aspects of MI. See Table 1 for further details with regard to the difference in MITI scores between the MISC and SC groups.
Table 1. The Motivational Interviewing Treatment Integrity Coding System and the Difference between the MISC and SC Groups.
In order to see how well or poorly MI was delivered, all sessions were audio recorded and coded using MITI 3.1.1 (Moyers et al. 2010). Coding was conducted by an independent coder who is an MI expert and was not involved in any aspect of this research or training of the author.
The MITI is a behavioural coding system that assesses both global scores and behaviour counts within a single review of a random 20-minute segment of an MI session recording. MITI global ratings are in the form of Likert scales of 1 to 5 and an average of 4 is recommended for competency (Moyers et al. 2010). Global ratings are designed to capture the assessor’s overall judgment of the interviewer’s behaviour with regard to Evocation, Collaboration, Autonomy/support, Empathy, and Direction. The interviewer’s utterances are assessed using behaviour counts. The utterances that receive behaviour codes comprise (1) reflections, (2) questions, (3) giving information, (4) MI adherent (e.g., asking permission, affirmation, emphasize control, support), and (5) MI non-adherent (e.g., advise without permission, confront, direct). Reflection to question ratio of 2, 70% open questions, 50% complex reflections, and 100% MI-adherent behaviours are recommended for adequate MI competency (Moyers et al. 2010).
MI global mean scores for MISC sessions were all at competency level of 4/5 or above. As expected there was a significant difference in MITI global scores between MISC and SC where no MI was provided. The main contrast between MISC and SC sessions was related to the differences in Evocation, Collaboration and Empathy. For Evocation 50% of sessions were ranked as 4 and 50% as 5 in MISC group and 59% ranked as 1 and 41% as 2 in SC group. For Collaboration 40% of sessions were ranked as 4 and 60% as 5 in MISC group and 53% ranked as 1, 41% as 2, and 6% as 3 in SC group. For Empathy 75% of sessions were ranked as 4 and 25% as 5 in MISC group and 76% ranked as 1, 12% as 2, and 12% as 3 in SC group.
Group differences in Autonomy Support were relatively smaller. Twenty percent of the sessions were ranked as 4 and 80% as 5 in MISC group and 12% ranked as 2, 59% as 3, and 29% as 4 in SC group. Finally there was no significant group difference in Direction where 5% of sessions were ranked as 4 and 95% as 5 in MISC group and 100% of the sessions were ranked as 5 in SC group. This indicates that the sessions in both groups were very focused with a clear direction on consistent use of hearing aids.
MISC sessions successfully met the threshold for competency with regard to behaviour counts as measured via MITI. Mean percent open questions was 15% (SD=9) in the SC group and 75% (SD=13) in the MISC group (p<0.001) (based on MITI 3.1.1, 70% or more open questions is needed for acceptable fidelity to MI). Mean percent complex reflections was 38% (SD=42) in the SC group and 66% (SD=8) in the MISC group (p=0.003) (based on MITI 3.1.1, 50% or more complex reflections is needed for acceptable fidelity to MI). Mean percent MI-adherent behaviour was 63% (SD=0.3) in SC group and 95% (SD=0.2) in the MISC group (p=0.0004).
Aazh, H. (2016a). Feasibility of conducting a randomised controlled trial to evaluate the effect of motivational interviewing on hearing-aid use Int J Audiol, 55, 149-156.
Aazh, H. (2016b). Patients’ experience of motivational interviewing for hearing aid use: A qualitative study embedded within a pilot randomised controlled trial. J Phonet and Audiol, 2, 1-13.
Moyers, T., Martin, T., Manuel, J., et al. (2010). Motivational interviewing treatment integrity 3.1. 1 (MITI 3.1. 1). USA: University of New Mexico: Center on Alcoholism, Substance Abuse and Addictions (CASAA).