Phase 1: Quantitative findings

Table 1 below describes the investigated variables. Looking at the preferences for wearable sensors a total of 46% gave a positive rating. The negative ratings add up to 44% and 10% could not say whether they like or dislike the idea of wearable sensors.

Table 1 Participants demographic & sociodemographic characteristics

In Table 2 the results of the multivariable ordered logistic regression are presented. The differences concerning preferences were insignificant for age, education level and place of residence. On the other hand, being female was negatively associated with liking wearable sensors (P < 0.05, OR 0.75). In addition, participants not suffering from a chronic disease had decreased odds to like wearable sensors (P < 0.1, OR 0.75). To put it in other words, male and chronically ill individuals had increased odds to be in a more positive preferences category compared with female and not chronically ill individuals.

Table 2 Ordered logistic regression Phase 2: Qualitative findings

Through in-depth discussion with the interviewees various preferences were identified. Figure 1 illustrates a thematic map with the most relevant themes and corresponding sub-themes developed through the thematic analysis. First, the overall preferences including both scenarios are presented. Second, the preferences regarding scenario two (with chronic disease) are illustrated and in the last step the preferences regarding scenario one (without chronic disease) are discussed. The Appendix A.2 contains the complete thematic map followed by the corresponding explanations (Appendix A.3-A.9).

Fig. 1 Overall preferences

This section first covers the theme “use of sensors” which includes reasons why interviewees would use wearable sensors presented in the two scenarios. Furthermore, the intention to use wearable sensors was associated with several necessary features, which are also introduced in this section. Finally certain risks in connection with the use of wearable sensors were established, which are highlighted as a last theme of this section.

Use of sensors Increased personal safety

Increased personal safety was in most cases the first point mentioned when participants explained why they would use a wearable sensor. The biggest advantage was seen in the early detection of an emerging diseases, another acute health complications and the worsening of existing chronic diseases. For the interviewed persons the wearable sensor would provide reassurance by constantly measuring relevant health parameters. Deteriorations of vital parameters are not always directly noticeable and with the help of a wearable sensor, changes can be detected and treated at an early stage. Health predictions supported by algorithms and artificial intelligence are, according to some participants, better and faster than when performed by humans. The word that was often used to describe these benefits was “safety.“ Safety not only in the sense of early detection, but also the knowledge that someone will be informed and one is not left alone in emergency situations.

Physician knows relevant parameters

Another reason for using a wearable sensor would be that the primary care physician automatically has access to real-time information about a person’s health.

“Why should the doctor have to write all this down in his or her computer when he or she could actually have it already”.

This would save both the patient and the doctor valuable time. Ultimately, it would be an improvement and simplification for the healthcare system, as the patient would not have to go through multiple tests anymore.

Features of the sensors Not flashy

The design of wearable sensors was an often discussed topic and appeared to be one of the most relevant features. According to the interviewees the sensor should be small and not flashy. The sensor should also not have a medical appearance. The reasons for these points were partly aesthetic and partly social. Participants do not want to be asked what they are wearing and why they are wearing it.

No interference in normal life

Not interfering with normal life was a further important feature frequently mentioned by participants. Sensors should be compatible with everyday activities such as sports, should not disturb during the night, should be resistant to water and cold weather.

Battery power

The preferences in terms of battery power were very similar. A wearable sensor needs to last at least a full day to provide 24-hour monitoring. Anything less than a day would be annoying and not worth buying for the interviewees. A week, or as long as possible would obviously be desirable in terms of battery life. Establishing a routine was an important point mentioned by some respondents. If it is not less than once a day, but always at the same time, it does not matter if the sensor needs to be charged once a day or once a week. Nevertheless, there were skeptical thoughts when the sensor had to be charged every day. It becomes difficult when one is not at home at night and no power source is available.

Easy to use

The handling of the sensor should be, according to the participants, self-explanatory, require few steps to use, and little effort to avoid permanent removal of the sensor. Some would like the sensor and associated instructions to be provided by the primary care physician. With clear guidance from the primary care physician regarding the exact use of the sensor, application errors can be minimized. In addition, different instructions should be provided for specific target groups. For example, the instructions should be available in different languages and include pictures if reading is not possible.

Lifetime of sensors

Before the cost component could be discussed, the lifetime of wearable sensors had to be determined first. Due to technical progress, the participants expected lifetime of a wearable sensor should be between three and five years.

Costs

When the cost component was introduced, everyone assumed that it would be covered by health insurance. First, to ensure that everyone has access to a wearable sensor if they want it, and second, because health insurance companies should be interested in preventive healthcare. If the costs were not covered by health insurance, interviewees would be willing to pay around 85–700 Swiss francs (\(\overline\): 270.-) per year for a wearable sensor. The willingness to pay was very high for some people, as they consider health to be something very valuable and worth paying for.

Risk of use Data security

Interviewees all agreed that data security is a very important issue, but it was assessed differently. Some do not see any problems if their personal data gets disclosed and is used by external actors. They believe that their health parameters are not important to others. However, others see many potential harms if the data goes beyond the treating physician. External actors are enriching themselves with very personal data that can be used against the individual. The personal data could be used for tailored advertising, to exclude someone based on their health parameters, or it could cost someone’s job if health information gets to the employer. Although there is a certain risk of data misuse or breach, in the participants’ views, benefits of wearable sensors outweigh these risks.

Specific preferences with chronic diseases

This section focuses on the reasons why some interviewees would only use wearable sensors if they were chronically ill.

Use of sensors

Focus of measurement on chronic diseases

Those participants who are only interested in using a wearable sensor in the context of scenario two, explained that in such a situation it is clear which parameters need to be measured constantly. Individuals who are chronically ill have certain risk factors which need to be monitored to avoid complications.

Shared control by patient and sensor

Combined with the knowledge of important parameters to be measured, the use of a wearable sensor would allow patients to relinquish control. Since the measurement is limited to the relevant parameters, patients do not have to think about it all the time and can rely on the sensor.

Risk of life-threatening diseases

The use of wearable sensors would still be dependent on the type of chronic disease. The participants would only be willing to use a sensor if the chronic disease is fatal and life would likely be extended by its use. One chronic disease category for which a wearable sensor would be used, was a cardiovascular disease. In this case, the sensor could indicate early on when something is wrong, and premature death could be avoided.

Specific preferences without chronic disease

This section focuses on the reasons for not using wearable sensors in the absence of a chronic disease.

No use of sensors

Not at risk for health problems

Participants opposed to wearable sensors stated that the deciding factor for using a wearable sensor should be the risk assessment. In their perception, measuring various parameters in healthy individuals has no benefits as they are not at risk for health problems. Wearable sensors should only be used in individuals with a known risk, a hereditary predisposition or an existing chronic disease. The older generation in particular would benefit from wearable sensors, as they have multiple risk factors due to their age.

Fear of losing control

Furthermore, the use of wearable sensors is not necessary because the participants said they have a strong body awareness. They would notice when something is wrong and fear that using a sensor would take away this body awareness. They would lose the ability to notice changes in the body and relinquish control to the device. This decreased self-control would make them dependent on the wearable sensor and lead to decreased personal responsibility. Participants are unwilling to give up their personal responsibility and decision-making about their own bodies.