Behaviour change modules
The web-based intervention is based on modules developed in earlier studies targeting smoking cessation and physical activity for the general population [11, 27]. We adapted these modules for COPD patients and for people at risk for COPD. We used the I-Change model as theoretical framework in our intervention . This model includes several health promotion theories, i.e.: the Attitude-Social influence-Self-efficacy model (ASE) , which can be thought of as an incorporation of ideas of the Theory of Planned Behavior , the Social Cognitive Theory , the Transtheoretical Model , the Health Belief Model , and Implementation and Goal setting theories [31, 34, 35]. We evaluated and improved the usability of the behaviour change modules in a previous study .
The smoking cessation and physical activity modules provide participants with tailored feedback messages. Tailoring algorithms consider demographical characteristics and psychosocial constructs to generate personalized messages. The questionnaires have been tested experimentally among Dutch adults in previous studies [16, 36, 37]. The messages consist of texts, graphs and illustrations. Smoking cessation and physical activity modules include each the 6 intervention components described below.
1) The health risk appraisal
The health risk appraisal measures the behaviour (smoking and physical activity) and provides feedback based on Dutch guidelines for health behaviour. The feedback contains an explanation of how participants’ behaviour compares to the level of compliance with the Dutch guidelines. This is illustrated with a traffic light system, where a green light refers to meeting recommendations, a yellow light to partly meeting the recommendations and a red traffic light indicating lack of compliance with the recommendations . The behavioural feedback can be tracked each time participants complete the health risk appraisal and ipsative feedback  will be offered regarding changes in behaviour over time. At last, a graph that shows specific behaviour changes over time is presented.
2) Motivational beliefs
The attitude toward a relevant behaviour is assessed and the perceived positive (e.g. health benefit) and negative consequences (e.g. weight gain for non-smoking) of the behaviour are addressed to confirm, correct or place participants’ beliefs in perspective.
3) Social influence
Social influences of participants’ partner, family, friends and co-workers on the specific behaviour are assessed. An explanation about how these influence the behaviour of the participants and how the participants influence others is provided. Information about how to cope with social pressure, the importance of social support and being or noticing a good role model is emphasized.
4) Goal setting and action plans
Participants can set a goal for behaviour change. They will be guided through questions to formulate the goal (e.g. starting next week I want to be physically active for 20 minutes a day, 3 times a week). Hereafter, they can choose from a list of action plans to achieve their goal and receive feedback with additional advice on how to increase the likelihood for their plans to succeed. One week after the goal is due, participants will receive an e-mail prompt to revisit this intervention component and indicate whether they achieved their goal, receive feedback and have the opportunity to adjust their goal and make new action plans.
5) Self efficacy
Barriers to perform the healthy behaviour are assessed, by asking participants which situations they find difficult. Moreover, participants are asked to indicate whether they have made plans to overcome these difficult situations (coping plans). Consequently, participants receive a list of the situations that they indicated to be difficult and are encouraged to follow through with their plans. Suggestions for plans are given if they had not already made a plan.
This component is similar to self-efficacy, but focuses on maintaining instead of changing the behaviour.
Prompts to promote technology use
The web-based intervention can be used ad libitum. Prompts will be sent to the experimental group to boost intervention use  with a two week time interval and some referring to new content on the website, since this may increase the number of follow-up visits .
If participants do not use at least one of the behaviour change modules within two weeks, they receive a prompt by e-mail to login to the website and visit a behaviour change module. Another prompt is sent out two weeks after the first prompt if participants fail to visit a behaviour change module. If participants use a behaviour change module, they receive a prompt to use the intervention again every month. This means that if participants visit both behaviour change modules, they receive a prompt every two weeks. Prompts are tailored for COPD or people at risk for COPD and refer to intervention components, new stories on the website and possible current problems with behaviour change, such as physical activity during the winter season. Participants can click on a personalised link embedded in the e-mail to access the website without having to log in.