In the present work inspectors used Motivational Interviewing (MI) to promote environmentally sustainable behaviour in inspectees. MI is a counselling method with scientific support for various health behaviour changes. Inspectors (
It could be said that we are all responsible for taking care of environmental resources in a sustainable way, yet it is hard to know what should be done and by whom. In Sweden, overall responsibility for environmental policy implementation lies with the Swedish Environmental Protection Agency (Naturvårdsverket). This includes ensuring compliance with the Swedish Environmental Code, whose aim is “to promote sustainable environment for present and future generations” (
Each municipality and county administrative board employs environmental protection inspectors who are tasked with fostering good environmental behaviour among inhabitants. The inspectors’ primary task is to implement and achieve compliance with the Environmental Code. Inspectors carry out their role by implementing control measures, and providing information and guidance. Ideally, compliance with the Environmental Code could be reached through self-regulation, which would entail a minimal level of inspection and enforcement costs.
The present study evaluated conversations between environmental inspectors and inspectees, during both pre-notified and on-the-spot inspections. Specifically, the study evaluated the use of Motivational Interviewing (MI), a psychological counselling method, by inspectors during their interactions with inspectees. The aim was to explore the extent to which MI training of inspectors was successful, and to evaluate the experience of inspectors using MI in their professional interactions with inspectees. The underlying assumption was that training in a psychological counselling method would be helpful for the inspectors (
MI has been widely used with health behaviour problems, and has a solid research base with more than 200 randomised controlled studies, which in the main have shown significant low to moderate effect sizes in respect of, e.g., reducing or stopping problem drinking, stopping the use of illegal drugs and tobacco use, and completing a treatment program (
MI practice is focused on behaviour change. A short definition of MI is ‘a collaborative conversation style for strengthening a person’s own motivation and commitment to change’ (
One previous study has shown MI to be feasible to use in conversations with people about their environmental behaviour, and to increase pro-environmental verbal behaviours compared to controls (
The present work was carried out as part of the research program:
Municipality | All inspectors | Inspection target behaviour under the law and administration of | |||||
---|---|---|---|---|---|---|---|
Men | Women | ≤ 35 yrs | ≥ 35 yrs | National food |
National environmental |
National board of |
|
Ale | 0 | 8 | 1 | 7 | 4 | 4 | |
Nybro | 6 | 4 | 6 | 4 | 2 | 8 | |
Älmhult | 2 | 4 | 3 | 3 | 2 | 2 | 2 |
Östersund | 8 | 8 | 7 | 9 | 5 | 11 | |
Total | 16 | 24 | 17 | 23 | 13 | 25 | 2 |
The study was approved by the Regional Board of Ethics in Stockholm (2012/1:7). All inspectors were provided with oral and written information regarding the study. They were informed that their participation was voluntary and that no negative consequences would ensue, should they choose not to participate or withdraw later on. In turn, inspectors informed inspectees that they had received MI training, that they would like to audio record the conversations for use in the MI training, and that they could choose whether to participate and that, in the event of participation, they could withdraw later without any negative consequences.
Prior to the start of the study, a pilot MI training protocol was developed and tested in a fifth municipality (Eksjö) with six inspectors plus the head of the OIEAE and the final MI training model was adjusted based on these experiences. The MI training was divided into six days, each day having a different theme (
Training day | MI–theme |
---|---|
DAY 1 |
|
– To understand the meaning of MI | |
– To listen and engage in conversation | |
– To convey cooperation and equality | |
– To direct toward a target behaviour | |
DAY 2 |
|
– To recognize, elicit, and strengthen change talk | |
– To ask open-ended and exploring questions | |
DAY 3 |
|
– To inform as a dialogue | |
– To understand and implement positive and negative reinforcement | |
– To convey listening by reflecting | |
DAY 4 |
|
– To use empathic listening | |
– To convey that you are listening and trying to understand through reflections | |
– To use MITI-coding as feedback | |
DAY 5 |
|
– To explore readiness to change and ambivalence | |
– To meet and roll with resistance | |
– To avoid MI-non-adherent utterances | |
DAY 6 |
|
– To summarize what has been gone through at the training | |
– To form a personal plan for upholding MI-proficiency |
The theory and practice element of MI training was offered to all inspectors in each of the four municipalities. In total, 40 inspectors participated in this part of the training. Eight inspectors did not take part in the feedback element of training because they did not carry out inspections. Thirty-two inspectors contributed audio-recorded conversations (17 inspectors under the regulation of the National Environmental Protection Agency, 13 under the Codes of the National Food Administration, and 2 under the National Board of Housing, Building, and Planning).
In addition to their use as training material, the conversations recorded during the training period were also used to evaluate inspectors’ MI acquisition. Three conversations were chosen pre- and post-MI training in order to get an average estimate of the inspector conversation practice. During the MI training program, inspectors were asked to record one conversation prior to each training day. In order to have a sample of inspectors with accurate measures of MI competency that were relatively independent of the inspectee and inspection situation, two inspectors per municipality (three in one municipality) were asked to record three conversations prior to each training day. These nine inspectors formed a more intense recording group. In total, inspectors successfully recorded 76 percent of planned recordings (
Municipality | All 32 inspectors, including 9 intense recording group members | ||
---|---|---|---|
Number of planned |
Number of recorded and |
Per cent dropout |
|
Ale | 82 | 56 | 32 (26) |
Nybro | 94 | 67 | 29 (27) |
Älmhult | 85 | 57 | 33 (28) |
Östersund | 119 | 109 | 8 (10) |
Total | 380 | 289 | 24 (91) |
All recorded conversations were evaluated for MI proficiency by professional coders at the Motivational Interviewing Coding Laboratory at Karolinska Institute (MIC Lab) in accordance with the Motivational Interviewing Treatment Code (MITI) (
The MIC Lab coders, who coded the material for the present study, had worked at the MIC Lab since 2009 and 2010 and had undergone a four week initial training, followed by three-hour training sessions every week, including regular inter-rater checks. The reliability of the three coders was calculated by random recoding of 10% of the 289 recordings coded during the study period. The reliability for the five global variables was estimated as the percentage agreement between the coders and as intra-class correlations (ICC), calculated in a two-way mixed model, with absolute agreement and reported as single measures. For the seven behaviour count variables, only the ICC was calculated (
Intra-class-correlations (ICC) are calculated as mixed models, with absolute agreement, and reported as single measures. Also, percentage agreement (numbers).
MITI global |
Exact agreement |
One of the coders is one |
One of the coders is two |
ICC |
---|---|---|---|---|
Empathy | 50% (14) | 43% (12) | 7% (2) | 0.35 |
Evocation | 39% (11) | 54% (15) | 7% (2) | 0.36 |
Collaboration | 36% (10) | 57% (16) | 7% (2) | 0.47 |
Autonomy | 36% (10) | 57% (16) | 7% (2) | 0.53 |
MI-spirit | 0.67 | |||
Direction | 46% (13) | 46% (13) | 7% (2) | 0.21 |
Information | 0.44 | |||
MI-adherent | 0.52 | |||
MI-non-adherent | 0.70 | |||
Closed questions | 0.70 | |||
Open questions | 0.76 | |||
Simple reflections | 0.78 | |||
Complex reflections | 0.22 |
The ICC for the global variable Empathy was poor according to
To follow the inspectors’ experience of using MI during their routine inspections and to gather feedback on the MI training, they were asked to fill in a questionnaire on each training day (
Municipality | Number of |
Per cent attendance |
Submitted |
Per cent dropout |
---|---|---|---|---|
Ale | 8 | 60 (29) | 23 | 21 (6) |
Nybro | 10 | 86 (52) | 39 | 25 (13) |
Älmhult | 6 | 97 (35) | 31 | 11 (4) |
Östersund | 16 | 100 (117) | 101 | 14 (16) |
All municipalities | 40 | 233 | 195 | 16 (38) |
The inspector’s experience of the inspections during the yearlong period of training was monitored by an inspection questionnaire. The inspector filled in a web questionnaire (SurveyMonkey;
At the beginning of the visit, the inspectee had a positive attitude to the inspection;
I consider that following the visit the inspectee had enough knowledge about how his own activity influences the environment and/or health;
I consider that the inspectee clearly demonstrated that he/she understood the information I wanted to convey during the inspection;
I consider that the inspectee has carried out the required measures/demands toward reducing his/her influence on the environment/ensuring food safety;
I consider that the inspectee will need to take steps toward reducing his/her influence on the environment/ensuring food safety;
I consider that the inspectee will take steps toward reducing his/her influence on the environment/ensuring food safety;
I consider that the inspectee conveyed his/her own reasons and motives to take the necessary action during the conversation;
I consider that the inspectee showed an interest in contributing to sustainable development of the environment;
I was satisfied with my own work during the inspection;
I consider that the inspection was a positive experience for the inspectee.
The inspection questionnaire also contained 8 statements about the inspection and the inspectee’s activity:
(1) Was the meeting considered to be an inspection (from the inspector’s perspective)? (2) Was this the first visit to this inspectee? (3) Type of supervised activity? (4) Was the investigation notified in advance? (5) Do the supervised activities require a permit or notification? (6) Type of inspectee? (7) Size of activity/number of employees (8) Have you or a colleague previously visited this inspectee during the research programme?
MITI global variables (Empathy, Evocation, Collaboration, Autonomy support and Direction) were coded on a five-point ordinal scale and were thus considered as non-parametric. Therefore, chi square analyses and Mann–Whitney U-tests were conducted (
Number of recordings per scale value on five-point ordinal Likert scales per MI training occasion. Expected counts in parenthesis.
Recording MITI |
Scale value | 1st training | 2nd | 3rd | 4th | 5th | 6th | After training |
---|---|---|---|---|---|---|---|---|
Empathy | 1 | 19 (16.3) | 14 (10.7) | 17 (13) | 12 (13) | 11 (10.7) | 4 (9.6) | 7 (10.7) |
2 | 10 (10.7) | 3 (7) | 4 (8.5) | 11 (8.5) | 6 (7) | 11 (6.3) | 10 (7) | |
>=3 | 0 (1.9) | 2 (1.3) | 2 (1.5) | 0 (1.5) | 2 (1.3) | 2 (1.1) | 2 (1.3) | |
Chi-square (12) = 23.25, |
||||||||
Evocation | 1 | 24 (17.2) | 12 (10.7) | 19 (13.5) | 13 (13.7) | 8 (11.5) | 5 (10.1) | 7 (11.3) |
2 | 4 (9.4) | 5 (5.8) | 2 (7.5) | 9 (7.5) | 8 (6.2) | 11 (5.5) | 9 (6.2) | |
>=3 | 1 (2.4) | 1 (1.5) | 2 (1.9) | 2 (1.9) | 3 (1.5) | 1 (1.4) | 3 (1.5) | |
Chi-square (12) = 29.12, |
||||||||
Collaboration | 1 | 11 (7.4) | 4 (4.8) | 10 (5.9) | 7 (5.9) | 3 (4.8) | 2 (4.3) | 1 (4.8) |
2 | 16 (16.5) | 10 (10.8) | 8 (13.1) | 14 (13.1) | 12 (10.8) | 11 (9.7) | 14 (10.8) | |
>=3 | 2 (5.1) | 5 (3.3) | 5 (4.0) | 2 (3) | 4 (3.3) | 4 (3.0) | 4 (3.3) | |
Chi-square (12) = 18.01, |
||||||||
Autonomy | 1 | 6 (5.7) | 4 (3.5) | 8 (4.5) | 7 (4.5) | 2 (3.7) | 2 (3.3) | 0 (3.7) |
2 | 18 (14.7) | 8 (9.1) | 7 (11.7) | 11 (11.7) | 12 (9.6) | 8 (8.6) | 11 (9.6) | |
>=3 | 5 (8.6) | 6 (5.4) | 8 (6.8) | 5 (6.8) | 5 (5.6) | 7 (5.1) | 8 (5.6) | |
Chi-square (12)= 16.92, |
||||||||
Direction | 1 | 0 (0.4) | 1 (0.2) | 1 (0.3) | 0 (0.3) | 0 (1.1) | 0 (0.2) | 0 (0.3) |
2 | 3 (1.4) | 1 (0.8) | 2 (2) | 1 (1.1) | 0 (0.9) | 0 (0.8) | 0 (0.9) | |
>=3 | 26 (27.2) | 16 (16.9) | 21 (22.6) | 22 (21.6) | 19 (17.9) | 17 (16) | 19 (17.9) | |
Chi-square (12) = 10.94, |
Mean and (standard deviation/sd) per behaviour count variable in MITI and MI training occasion and after the MI training (
Recording MITI |
1st |
2nd | 3rd | 4th | 5th | 6th | After training | |
---|---|---|---|---|---|---|---|---|
( |
( |
( |
( |
( |
( |
( |
||
Information | mean (ds) | 17.6 (6.0) | 18.5 (8.1) | 15.2 (8.1) | 16.4 (6.0) | 16.5 (8.0) | 13.6 (3.4) | 13.2 (3.6) |
MI-adherent | mean (ds) | 0.1 (0.2) | 0.1 (0.3) | 0.1 (0.3) | 0.2 (0.3) | 0.3 (0.4) | 0.3 (0.5) | 0.3 (0.4) |
MI-non-adherent | mean (ds) | 3.8 (3.6) | 2.7 (3.3) | 3.5 (3.2) | 3.3 (2.5) | 2.4 (2.1) | 2.7 (2.6) | 2.0 (1.4) |
Closed questions | mean (ds) | 12.8 (6.8) | 14.2 (7.2) | 15.0 (7.5) | 13.7 (6.4) | 12.2 (5.6) | 13.0 (4.4) | 15.3 (4.7) |
Open questions | mean (ds) | 3.7 (3.6) | 4.0 (4.0) | 5.1 (4.0) | 4.0 (3.8) | 4.1 (2.9) | 4.3 (2.7) | 4.0 (3.3) |
Open question/question ratio | mean (ds) | .28 (.64) | .30 (.72) | .40 (.93) | .38 (.45) | .42 (.35) | .47 (.58) | .31 (.42) |
Simple reflections | mean (ds) | 6.4 (3.4) | 5.7 (3.9) | 7.6 (6.7) | 7.5 (4.9) | 6.7 (4.7) | 6.6 (1.5) | 6.3 (4.5) |
Reflection/Question ratio | mean (ds) | .46 (0.29) | .34 (.19) | .40 (.34) | .56 (.39) | .51 (.36) | .52 (.27) | .43 (.27) |
The inspectors significantly increased their competence in the Empathy global variable, calculated with chi-sqr (df 12) = 23.25;
The relatively low Empathy scores and MI competency as measured by MITI during the course of training might be explained by inaccurate MI training. The MI competency level reached by inspectors was too low, necessitating measures to increase MI skill acquisition by inspectors. The attained low level of MI competence might also be explained by the different inspection types selected for monitoring. In the selected sample, inspections targeted problematic behaviours, which were regulated by different legislative regimes. It was therefore difficult to know in which part of the conversations it was appropriate to use MI. Among the selected recordings there were many conversations in which MI was not appropriate to use e.g., where it was not clear which inspectee behaviour was targeted, or where inspectees were not in a position to make decisions to change current environmental behaviours.
More close data analyses revealed differences in Empathy scores related to type of inspection. On a Mann–Whitney U-test, the inspectors of the food administration demonstrated significantly higher scores in Empathy compared to environmental- and health protection inspectors (
Inspectors significantly increased their competence in the global variable, Evocation (chi-sqr (df 12) = 29.12;
In the MITI behaviour counts the number of utterances providing information had decreased, but not significantly (
In questions 1–3 of the questionnaire about the inspectors’ MI experience, inspectors considered each training day to be useful—approximately 5 on a 6-point scale, where 6 means very useful and 1 means not useful at all. There were no changes in these judgements over the yearlong training period or between municipalities. When the inspectors’ answers about the usefulness of the training day were analysed in detail, the theory sessions scored 5/6, and the feedback on the recorded conversations scored between 4.5/6 and 5/6. None of these judgements changed over time. In all, the inspectors had a positive evaluation of the training.
A crucial question in the questionnaire was whether the inspector considered MI to be useful when carrying out inspections. The responses showed that the inspectors judged MI to be useful in inspections, approximately 5 on the 6-point scale. These evaluations did not change over time. The responses indicate that inspectors in four different municipalities considered MI to be useful in the daily inspection routine throughout the 10–13 month long study period. MI’s usefulness to inspectors seemed to be maintained after the novelty value of taking part in MI training had decreased.
The inspection questionnaire had been filled in before the training only in one of the municipalities, Östersund. The other municipalities started to fill in the web questionnaire after training day 1. Seven inspectors did not file any questionnaires two inspectors completed the most questionnaires, with 116 questionnaires and 95 completed respectively. These two inspectors accounted for a little more than 40% of the questionnaire material. However, their satisfaction did not increase over time; their satisfaction with their own inspection practice was constantly high, and therefore did not contribute to an increase in satisfaction for the group. On the question whether inspectors were satisfied with their work during the inspections, the proportion who had answered “very well” increased from 26.1% to 51.3% while the part who had answered “very bad” or “not that well” or “neither well or bad” decreased from 31.5% to 3.2%. The results indicate that the inspectors’ satisfaction with their inspections increased over the study period, which may be related to acquisition of MI competency.
The inspectors estimated how inspectees viewed the inspections by the question: “I consider that the inspection was a positive experience for the inspectee”. In the responses to this question, inspectors more often (37%; 19/51) considered that their inspections had been a positive experience for the inspectee after MI-training compared with before (not significant), which may indicate that by using MI, inspectors had been able to improve their interactions with interviewees.
A strength of the study was that the inspectors used MI during a long time period (10–13 months), during which their competency in MI and experience of MI were continuously monitored. The evaluation period allows us to be reasonably confident in the inspectors’ evaluation of MI as well as how well MI competency was attained. A limitation of the study was that municipalities were not randomized, thus the results cannot be generalized to all Swedish municipalities. However, the four selected municipalities represented different parts of Sweden, and included large as well as small municipalities, which increases the representativeness of the results. Another limitation was a large dropout rate in returning inspection questionnaires among inspectors, which means that the results pertaining to these questionnaires should be interpreted with caution. Some inspectors also failed to audio record inspection conversations. However, these missing recordings may not have impaired the results, since the MI competence of the inspectors appeared to be stable for the group as a whole. The recordings in both the intense recording group of inspectors (who made three recordings prior to each training day) and the remaining inspectors (who had one recording prior to each training day) reached the same levels of MI competency. Another strength of the study was that the assessment of inspectors’ MI competency was reliably monitored throughout the long study period, except for the MITI variable, complex reflections.
Inspectors considered MI useful in environmental inspection and enforcement interactions with inspectees, which indicates that MI facilitated the work of inspectors. Future research should examine whether MI does affect inspectee environmental behaviour. It may also be important to explore more efficient MI training and to adapt the training to the type of inspections for which it is appropriate to use MI.
The authors would like to thank the inspectors in the municipalities of Ale, Nybro, Eksjö, Älmhult and Östersund. Helena Lindqvist at MIC Lab organized the coding of audio recordings. In addition, the considerable assistance of Jason Dale and Isra Black is gratefully acknowledged.
The authors declare there are no competing interests.
The following information was supplied relating to ethical approvals (i.e., approving body and any reference numbers):
The Regional Board of Ethics in Stockholm (2012/1:7).
The following information was supplied regarding the deposition of related data:
Raw data is kept at the MIC Lab research group at the Department of Clinical Neuroscience, Karolinska Institutet, Sweden and will be made available upon email request: Lars Forsberg,