[[08-01-2024]] Source - [Correlations Between Objective Behavioral Features Collected From Mobile and Wearable Devices and Depressive Mood Symptoms in Patients With Affective Disorders: Systematic Review - PMC (nih.gov)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/) #Digital_Phenotype [[Depression]] [[Digital Phenotyping]] Aim of this paper: (1) Which objective features have been collected?  (2) What is the correlation between objective features and depressive mood symptoms?  (3) Are the correlations similar across studies collecting the same features? **There are inconsistencies; for example, Beiwinkel found that outcoming SMS text negatively correlates with HDRS, whereas Fauurhold-Jespsen found a positive correlation.** "Some studies have shown similar results, while others have shown contradicting results. For example, Beiwinkel et al [[22](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref22)] found a statistically significant negative correlation between the number of outgoing SMS text messages and the HDRS, whereas Faurholt-Jepsen et al [[6](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref6)] found a statistically significant positive correlation. Asselberg et al [[15](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref15)] found a negative correlation with mobile phone usage frequency and depressive symptoms, while Saeb et al [[7](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref7)] found the opposite." **Behavioural Objective Features** 7 Feature Category (social, physical activity etc) from 85 different sensors signal (**[Table - PMC (nih.gov)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/table/table3/?report=objectonly)**) Social Physical Activity Location Device Subject Environment Bio # Social - **Social category had the lowest percentage of statistically significant correlations, because social network is complex, moderated by many factors.** - "The _social_ category had the lowest percentage of statistically significant correlations, by vote counting, across studies (10/38, 26%). _Social_ included features such as _call duration_ and _number of conversations_, which can be accessed on Android phones, contrary to iPhones [[77](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref77)]. We did not find any research article that explains how social patterns change with depression, but the review article by Baker et al [[27](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref27)] on online social networks suggests a complex relation involving factors that mediate or moderate the correlation and increase the variability in the findings." - **Cho Et Al found that genders have different correlation pattern with call duration and call frequency (Male negative, Female Positive**) - Furthermore, Cho et al [[44 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref44)] found a direct opposite correlation between genders (male negative, female positive) in the _call duration_ and _call frequency_ features. This suggests that social-based features should be treated as a highly personalized feature that should be assessed in a within-subject analysis. - **Social Based Features should be used for within-subject trend detection. # Device - **There is a general tendency for participants to use their phones more but use communication apps less.** - "The feature category with the highest percentage of statistically significant correlation features across studies was _device_ (13/24, 54%). As an example, using data provided by the corresponding author [[14](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref14)], we observed statistically significant results in _communication app usage_ (_r_=−0.33, _P_=.007) calculated using a within-subjects analysis of covariance. The low variability with device-based features could indicate that there is a general tendency for participants to use their phones more, but at the same time, withdraw from the social context by lowering the _communication app usage_." - **Home Stay and Screen Active Duration have a strong positive correlation with depression**. (Maybe they stay home and scroll Tiktok mindlessly; or Because they stay home scroll tiktok that's why they are depressed?) - Nonclinical Samples of Participants As seen in [Figure 2](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/figure/figure2/), we found two features that have a strong positive (ie, close to 1 _wD_) correlation with depression: _home stay_ and _screen active duration_; both of these showed a large proportion of statistically significant correlations across studies. Moreover, all 4 studies with a positive correlation between _home stay_ and depression level also had a large average participant number. Individual studies have shown that the degree to which a person stays at home is associated with depression [[45](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref45)], and it is a general hypothesis that this relation is positive. We were able to verify this hypothesis by combining the results across the included studies in this review. - On the other hand, no prior hypothesis has been formulated regarding the relationship between general phone usage and depressive mood symptoms. However, studies have shown a statistically significant positive correlation between depressive symptoms and the feature _screen active duration_ [[78](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref78)]. Similarly, subjective-based mobile phone use has been studied in relation to depression, where Thomée et al found that high mobile phone use was associated with symptoms of depression [[79](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref79)]. These findings were replicated in this review, with only a single statistically nonsignificant contradictive result from a two-sample study by Mestry et al [[14](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref14)] (_r_=−0.03, _P_=.79). **Physical activity, Movement outside of the house observed when participants score lower on depression scale.** - On the left side in [Figure 2](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/figure/figure2/), we see several features that have a strong negative correlation to depression, including _location clusters_, _entropy_, and _sleep duration_. A majority of these features indicates that enhanced physical activity and more movement outside of the house are observed when participants score lower on the depression scale. This is consistent with the Actigraph systematic review papers by Scott et al [[80](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref80)], who revealed a consensus of lower mean activity levels associated with bipolar depression, and Burton et al [[81](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref81)], who revealed a pattern of lower daytime activity but higher nighttime activity in depression **The more people spend time outside at different locations outside home, the better the mood (negative correlations)** - _Entropy_ is the most prominent feature in the figure with many studies (n=6), all yielding a negative correlation and a high statistically significant proportion. The only case of nonsignificance was reported by Saeb et al [[7](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref7)] (_r_=−0.42, _P_=.082), who, however, did show a high negative correlation. ==_Entropy_ is a measure that captures the distribution of time spent at the different location clusters registered. Thus, a high _entropy_ would indicate that the participant spends time more uniformly across different location clusters. Because all studies consistently showed a negative correlation, this implies that a higher _entropy_ correlates with a better mood.== If a participant stays home for a longer time than usual, the _entropy_ will drop. Hence, there is a dependency between _entropy_ and _home stay_, which is also evident in the figure where they are almost mirrored, both with a large proportion of statistically significant findings. Both features can be collected via the location Application Programming Interface, which uses the GPS sensor typically embedded in all mobile phones or wearables. **Incoming and Outgoing calls and frequency should be separated. When they are depressed, they receive more calls and talk longer. When their mood is better, they make more outgoing calls and make them more frequent.** - Social-based features were more extensively investigated with clinical samples of patients. The two features _incoming call duration_ and _incoming call frequency_ reveal a strong tendency that participants tended to receive more calls and talk longer during these calls when depressed. On the other hand, the features _outgoing call duration_ and _outgoing call frequency_ tend to suggest that patients make more and longer calls when they are less depressed. This difference between incoming and outgoing calls highlights that these features should be kept separate, and it raises concerns with some of the results on _Call duration_ with nonclinical samples of participants as in a study by Wang et al [[51](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111148/#ref51)], who measured _call duration_ and _frequency_ across incoming and outgoing calls. ![[mhealth_v6i8e165_app7.png]]