Pith. sign in

REVIEW 5 major objections 5 minor 32 references

Towards automated symptoms assessment in mental health

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This thesis claims that passive physical-activity and phone-use data can differentiate psychiatric diagnoses and clinical mood states with reported accuracies of 67–95.3%, and can predict personalised mood scores with errors of 1.36–3.32…

desk verdict A substantial thesis with a new longitudinal dataset and a sensible symptom-driven framework, but the headline state-discrimination accuracies rest on self-report labels whose missingness is likely state-dependent, so the clinical-readiness claims need to be dialed back. read the letter →

arxiv 1908.06013 v1 pith:QEMXMDSG submitted 2019-08-14 physics.med-ph cs.CEcs.LGstat.ML

Maxim Osipov This is my paper · ORCID
classification physics.med-phcs.CEcs.LGstat.ML
keywords mentalhealthactigraphyphysicalactivitybipolardisorderborderlinepersonalityschizophreniamoodstateclassificationmmonitoring
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This thesis sets out to show that continuous, passively collected measurements of physical activity, sleep, and phone use carry enough information about psychiatric state to serve as objective symptom markers. Using accelerometer data from wrist-worn devices and smartphones, together with questionnaire-based mood labels, it reports classifiers that separate healthy controls from bipolar disorder (67%), healthy controls from borderline personality disorder (70%), bipolar from borderline personality disorder (80%), and the bipolar mood states euthymia, mania, and depression from each other (80–90%). A separate schizophrenia-versus-controls analysis reports 95.3% accuracy under leave-one-out cross-validation, with heart rate adding substantial predictive power to activity-based features. The broader claim is that disorders and clinical episodes leave measurable behavioural signatures that could enable early detection of deterioration in ambulatory patients.

What carries the argument

The central object is a feature pipeline built from accelerometer time series. Tri-axial acceleration is converted into the Euclidean Norm Minus One (ENMO) metric for day-time activity, while epoch-based activity counts are used for sleep analysis. Non-stationarity is treated as signal rather than noise: the Bayesian Online Change Point Detection algorithm segments activity into stationary segments, and the durations and transitions of those segments become features. Sleep and wakefulness are identified with an Explicit Duration Hidden semi-Markov Model whose parameters are trained on device-based bed-time annotations. These features are then ranked with minimum Redundancy Maximum Relevance or LASSO, and classified with logistic regression or support vector machines under leave-one-out cross-validation.

What would settle it

Run the same feature pipeline on a cohort where clinical states are independently adjudicated by structured clinical interview rather than self-report, and check whether the classifier's sensitivity and specificity for clinician-confirmed mania and depression remain at the reported levels when questionnaires are missing, delayed, or contradicted by clinician ratings.

Watch

Extended reading notes

Core claim

The central claim is that objective features extracted from activity and behaviour time series can differentiate mental health diagnoses and clinical mood states with clinically useful accuracy. The features are organised around three symptom dimensions from a five-cluster symptoms model: psychomotor (activity level and intensity), disorganisation (multiscale entropy, activity persistence, and day-to-day pattern variability), and mood (sleep-wake segmentation, circadian amplitude, and non-parametric rest-activity characteristics). Personalised regression models predict mood scores with a mean absolute error of 1.36 to 3.32 points, which falls within the 4–5 point ranges that psychiatric questionnaires reserve for distinct identifiable mood states. Adding heart-rate features to locomotor features improves schizophrenia classification by almost 10% over activity alone and by almost 17% over heart-rate features alone. The thesis argues that these results support a framework for computational behaviour analysis that could identify clinical deterioration earlier than routine clinic visits.

Load-bearing premise

The ground-truth clinical states are defined by self-report questionnaire scores (QIDS-SR16, ASRM, and Mood Zoom), and the thesis concedes that questionnaire compliance can itself change with clinical state; if the labels are wrong or missing during episodes, the reported accuracies may reflect questionnaire response patterns rather than true clinical states.

Editorial extensions

If this is right

  • If a patient's sensor stream can flag mood episodes between clinic visits, clinicians could be alerted to deterioration earlier than weekly self-report alone would allow.
  • The reported accuracy levels suggest sensor-based features could act as a screening layer that prompts targeted clinical interviews, rather than replacing clinician judgement.
  • Personalised mood models imply that each patient's behavioural baseline can be learned, making deviations from that baseline more informative than population-level thresholds.
  • The schizophrenia result, where fusing heart rate and activity improved classification, suggests multi-modal sensing may be needed for disorders whose behavioural signature is weaker.
  • A standardised feature framework tied to symptom dimensions gives future studies a common language for comparing objective mental-health monitoring results.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The accuracy figures should be read as separating questionnaire-labelled states, not clinician-validated episodes; if questionnaire compliance is itself state-dependent, the classifiers may partly be detecting response behaviour rather than the underlying episode, a limitation the thesis acknowledges.
  • A natural next test is an external cohort with clinician-rated episodes and dense sensor data to see whether the 67%–95.3% accuracies survive independent adjudication.
  • The framework implies that symptom dimensions, rather than diagnostic categories, are the more tractable prediction target; the same features that separate mood states could extend to other conditions with psychomotor or circadian disruption.
  • A testable extension is applying the pipeline to consumer wristbands with lower sampling rates, to see whether the accuracy gains persist outside research-grade accelerometers.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The manuscript, a DPhil thesis deposited on arXiv, proposes a computational framework for automated symptom assessment in mental health from physical activity and phone-use data collected in ambulatory settings. It describes the AMoSS study, data pre-processing and segmentation methods, a symptom-driven feature framework, and three application areas: differentiation between healthy controls, bipolar disorder, and borderline personality disorder; differentiation between euthymic, manic, and depressive states; and personalised prediction of mood questionnaire scores. It also applies the framework to a separate schizophrenia dataset with heart-rate fusion. The headline results reported under leave-one-out cross-validation are 67--90% accuracy for disorder/state discrimination, 95.3% accuracy for schizophrenia versus controls, and mean absolute errors of 1.36--3.32 points for personalised mood regression.

Significance. If the reported results hold, this work would strengthen the evidence that passive sensor data contain clinically usable information about psychiatric state, and the proposed symptom-driven feature framework is a useful organizing principle for mHealth psychiatry. The strengths of the manuscript include a comparatively large longitudinal ambulatory cohort, the use of consumer devices alongside research-grade sensors, a principled mapping from Liddle's symptom dimensions to objective features, and an independent external schizophrenia dataset. However, the headline claims are conditional on two unresolved issues: the validity of self-report questionnaires as ground truth for mood state, and the integrity of the cross-validated performance estimates in small, resampled cohorts. The diagnosis-level comparisons are less exposed to the label-validity problem, but the state-discrimination and personalised mood claims, which are central to the abstract, currently inherit it.

major comments (5)
  1. [§3.2.1.2, §4.3.1] The state-discrimination results in Chapter 7 are not yet interpretable as clinical-state discrimination because the ground-truth labels are derived from self-report questionnaires whose missingness is state-dependent. The manuscript itself states that 'compliance can be a function of the clinical state, i.e. the patient may stop responding during a manic or depressive episode', and §4.3.1 then imputes missing weekly questionnaire scores with the unconditional mean. Under missing-not-at-random nonresponse, this procedure assigns missing episodes an average or euthymic-like score, so the remaining labels may track ease of self-report or device wear rather than true mood. The authors should quantify questionnaire missingness by state, test whether missingness is associated with concurrent self-report scores, and re-run the Chapter 7 and Chapter 8 analyses using only observed labels or an external clinician-rated episode source.
  2. [§6.2.2, §6.3.1, §6.4.3, §7.4.3] The cross-validated accuracy numbers in Tables 6.5 and 7.5 are point estimates from a protocol that does not, as presented, nest under-sampling, SMOTE, and feature selection inside each training fold. Under-sampling the majority class and applying mRMR or LASSO on the full data subset before LOOCV can inflate accuracy because information from the held-out subject leaks into feature selection and resampling. Please clarify the exact order of operations; if these steps are not nested within folds, the analyses should be redone with a nested or fully independent cross-validation pipeline. In addition, report bootstrap confidence intervals or per-subject prediction tables, given the small cohort sizes.
  3. [§4.5.1.5–§4.5.1.6] The selection of BOCPD with an expected segment length of 60 seconds for human accelerometer data is based on synthetic RR tachograms generated by the McSharry--Clifford model, with the rationale that heart rate correlates with physical activity. This is an unvalidated assumption: the model generates heart-rate dynamics, not accelerometer segment statistics, and the paper does not demonstrate that the segment-duration distributions of the two signals are similar. Since the BOCPD segment features feed into the psychomotor and disorganisation feature sets used in later chapters, the algorithm choice should be validated against human activity data with annotated change points, or at least cross-checked against the Fitbit bed/wake annotations described in §4.5.3.
  4. [§8.4.2, §8.5, Table 8.5] The personalised mood regression results inherit the label-validity problem of the daily Mood Zoom scores. The reported mean absolute errors are computed against the same self-report scores that are subject to state-dependent missingness and subjective bias, so the errors are not necessarily errors in an external clinical state. In addition, Table 8.5 does not report how many days per participant had imputed versus observed labels, which matters because unconditional-mean imputation will artificially improve apparent agreement when compliance is low. Please provide observed-only results and a missingness analysis.
  5. [§9.4, Table 9.4] The schizophrenia classification result of 95.3% and the claimed 10--17% improvement from adding heart-rate features are presented without per-fold breakdowns or confidence intervals. With the small number of participants in the Nuffield and Proteus datasets, a few individuals can drive the difference between feature sets, especially when feature selection is performed on the full cohort. Please report leave-one-out predictions per participant for the activity-only, HR-only, and combined feature sets, and state explicitly whether feature selection and any resampling were performed inside each fold.
minor comments (5)
  1. [§4.5.1.3, §4.5.1.6] The BOCPD expected-segment-length hyperparameter is denoted λ in the equations and Table 4.2, but the text that selects the value refers to it as τ = 60 seconds; please unify the notation.
  2. [Tables 3.2, 3.5, 3.7] The statistical test is spelled 'Wilcox rank sum test' in several places; the correct name is the Wilcoxon rank sum test.
  3. [Throughout] There are numerous typographical errors that should be corrected in a copyedit, including 'biploar disorder' in Figure 3.3, 'primarely' in the Glossary, and inconsistent hyphenation and encoding artifacts in 'Na¨ıve'.
  4. [§4.5.1.4–§4.5.1.6] The evaluation of change-point algorithms reports TPR and FPR, but the captions do not state the number of generated tachograms, the exact definition of the tolerance interval δ, or how true change points were defined from the synthetic model; please add these details.
  5. [§3.1.3.1] The mobile application section notes that 'no formal comparison were performed between smart phone characteristics with and without AMoSS mobile application'; this usability limitation should be acknowledged in the conclusions as a potential source of battery- or performance-related non-adherence.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor in-sample HMM bed/wake validation; headline classification claims are not circular.

  1. fitted input called prediction [Section 4.5.3.3-4.5.3.4 (Model for bed/wake segmentation; Segmentation results)]
    "Fitbit bed time annotation was used to estimate parameters of HMM, including the state duration distributions pi(d) and observation probability distributions bj(OOO). ... Bed/wake segmentation was performed using the proposed Hidden Markov Model and the accuracy of segmentation evaluated using manual Fitbit bed/wake annotation ground truth."

    The 'ground truth' annotations are the very same Fitbit data used to estimate the HMM's state-duration and observation distributions. Reporting segmentation error against this ground truth is therefore an in-sample fit, not an independent validation: the model was constructed to reproduce the labels it is then evaluated on. This is a fitted-input-called-prediction pattern for the bed/wake component. It does not by itself force the headline classification accuracies, since diagnosis and mood-state labels are external questionnaire/clinical labels and the activity features are not derived from those labels.

full rationale

The central derivation—objective sensor features discriminating clinical groups and mood states—is not circular. The target labels are external (diagnostic labels for HC/BD/BPD and questionnaire-derived state labels), and the features are computed from accelerometer and phone data without using those labels. Leave-one-out cross-validation is used for the headline accuracy numbers. The main circularity-like issue is confined to a preprocessing component: the HMM bed/wake segmenter is fitted to Fitbit button-press annotations and then evaluated against the same annotations, so its reported minute-level errors are in-sample performance. This does not reduce the central classification claims, and the self-citations in the thesis are not load-bearing for the central predictions.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The main empirical results rest on domain assumptions: sensor-derived activity reflects psychiatric symptoms, self-report questionnaires are valid labels, and a segmentation algorithm chosen on synthetic heart-rate data transfers to human activity. Algorithmic parameters fitted inside the study (BOCPD segment length, HMM bed/wake parameters, missing-data threshold) are the main adjustable inputs. No new physical entities are introduced.

free parameters (4)
  • BOCPD expected segment length lambda = 60 seconds
    Selected as best on synthetic RR tachograms (Section 4.5.1.6), then applied to human activity segmentation.
  • HMM bed/wake duration and observation parameters = Estimated from Fitbit annotations
    Used to segment GENEActiv data into bed/wake periods; fitted on the same study population (Section 4.5.3.3).
  • Missing-data exclusion threshold for HMM segmentation = 2 hours
    Described as empirically identified threshold (Section 4.5.3.4).
  • Classifier and feature-selection hyperparameters = Not fully specified
    LASSO regularization, SVM settings, SMOTE and under-sampling choices are tuned per analysis in Chapters 6 to 8; exact values are not given in the provided text.
assumptions (4)
  • domain assumption Physical activity and phone-use patterns are valid observable correlates of psychiatric symptoms and clinical states.
    The entire framework maps symptom clusters (psychomotor, disorganisation, mood) to actigraphic features based on Liddle's dimensions and prior literature (Chapter 5).
  • domain assumption Questionnaire-based self-reports (QIDS-SR16, ASRM, GAD-7, Mood Zoom) provide sufficiently accurate ground truth for clinical episodes and mood.
    Label assignment in Chapters 6 and 7 uses weekly questionnaire scores and thresholds; the thesis notes recall bias and state-dependent compliance (Sections 3.2.1.2 and 3.2.1.3).
  • ad hoc to paper Synthetic RR tachograms generated by the McSharry-Clifford model have change-point statistics similar enough to human accelerometer activity to select the segmentation algorithm.
    BOCPD was chosen after evaluation on heart rate, not activity, based on the assumed HR-activity correlation (Section 4.5.1.5).
  • standard math Classical statistical and machine-learning machinery (HMM, SVM, LASSO, LOOCV, SMOTE) is appropriate for these small, unbalanced, non-independent longitudinal samples.
    Assumed throughout Chapters 4, 6, 7, and 8.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards automated symptoms assessment in mental health." pith.science (2026). https://pith.science/paper/QEMXMDSG

@misc{pith2026190806013,
  author       = {Pith},
  title        = {Pith review of: Towards automated symptoms assessment in mental health},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QEMXMDSG}},
  note         = {Machine review of arXiv:1908.06013}
}
read the original abstract

Activity and motion analysis has the potential to be used as a diagnostic tool for mental disorders. However, to-date, little work has been performed in turning stratification measures of activity into useful symptom markers. The research presented in this thesis has focused on the identification of objective activity and behaviour metrics that could be useful for the analysis of mental health symptoms in the above mentioned dimensions. Particular attention is given to the analysis of objective differences between disorders, as well as identification of clinical episodes of mania and depression in bipolar patients, and deterioration in borderline personality disorder patients. A principled framework is proposed for mHealth monitoring of psychiatric patients, based on measurable changes in behaviour, represented in physical activity time series, collected via mobile and wearable devices. The framework defines methods for direct computational analysis of symptoms in disorganisation and psychomotor dimensions, as well as measures for indirect assessment of mood, using patterns of physical activity, sleep and circadian rhythms. The approach of computational behaviour analysis, proposed in this thesis, has the potential for early identification of clinical deterioration in ambulatory patients, and allows for the specification of distinct and measurable behavioural phenotypes, thus enabling better understanding and treatment of mental disorders.

Figures

Figures reproduced from arXiv: 1908.06013 by the authors.

Figure 3.1
Figure 3.1. User interface of the AMoSS mobile application. Figure a) shows the main ac￾tivity visualisation screen, b) shows the Mood Zoom questionnaire prompt and c) presents the application configuration options. • For phone calls and texts, the real addressee number is replaced with a MD5 hash, thus eliminating the possibility of identification, but at the same time preserving the addressee’s uniqueness, and content of text… view at source ↗
Figure 3.2
Figure 3.2. Architecture of the AMoSS smart phone data acquisition platform. The password￾protected internal computing resources are highlighted in orange. Participants access the internal network using provided smart phones with in-house developed AMoSS application. and password (Franks et al., 1999). Each user is assigned a unique study identifier (user name) and a specially generated strong password. The study identifier and… view at source ↗
Figure 3.3
Figure 3.3. Boxplots of BIS-11 impulsivity scores of the AMoSS study participants, where TOTL refers to the total score, and MOTR, ATTN, NPLN are the motor, attentional and non￾planning 2 nd order factors respectively. BD, BPD and HC refer to biploar disorder, borderline personality disorder and healthy controls. required to be present for a certain period of time and to exceed a specific threshold (see [PITH_FULL_IMAGE:figure… view at source ↗
Figures from the paper (25 more)
Figure 3.4
Figure 3.4. Figure 3.4: Weekly QIDS-SR16 depression scores collected from the AMoSS study participants, demonstrating compliance to email mood prompts. All questionnaires are delivered simultane￾ously, so similar results are observed also for ASRM and GAD-7 responses. The colour of the plot…
Figure 3.5
Figure 3.5. Figure 3.5: Mood Zoom sadness self-rating collected from the AMoSS study participants. Ac￾cording to the application’s design, all Mood Zoom questions have to be answered to enable submission of results, so compliance is the same for all mood dimensions. The colour of the plot i…
Figure 3.6
Figure 3.6. Figure 3.6: GENEActiv acceleromenter, a) presenting the physical design and b) the orientation of acceleration measurement axes. Adapted from (Activinsights Ltd., 2012). The GENEActiv accelerometer provides a very high quality physical activity mea￾surement with configurable sam…
Figure 3.7
Figure 3.7. Figure 3.7: GENEActiv activity data collection sessions. Each bar represents a single assess￾ment session. The colour of the plot represents the mean daily acceleration in units of g at sea level, after removing 1 G as the background gravitational field, with deep blue indicatin…
Figure 3.8
Figure 3.8. Figure 3.8: Acceleration data collected from a smart phone. Each dot represents a day and the colour of the plot represents the proportion of seconds during a day with at least one sample of acceleration, deep blue indicating a very low sampling rate or missing data. 3.3 Summary…
Figure 4.1
Figure 4.1. Figure 4.1: GENEActiv accelerometer data plot of a sample subject. Green dots represent accelerometer samples during inactivity periods (identified as described in this section), black circles are projections of a 1 g sphere of perfectly calibrated device measurements and red cr…
Figure 4.2
Figure 4.2. Figure 4.2: The first step of RMDM algorithm. The input signal in normalised units (top plot), together with the Student’s t-statistic for the difference of mean between the left and right subsignals (bottom plot) and the candidate change point at tmax (red dashed line). Given t…
Figure 4.3
Figure 4.3. Figure 4.3: The significance P(τ ) of the change point as a function of the length of timeseries and the selected significance level τ (50%, 75% and 95% lines are presented). 73 [PITH_FULL_IMAGE:figures/full_fig_p096_4_3.png]
Figure 4.4
Figure 4.4. Figure 4.4: The results of BBLOCKS algorithm. The input signal in normalised units (top plot, black) and the resulting segmentation (top plot, red), together with iterative estimation of the segment fitness function (bottom plot, black), where the maximum fitness indicates the l…
Figure 4.5
Figure 4.5. Figure 4.5: The results of BOCPD algorithm. The input signal in normalised units (top plot, black) and the resulting segmentation (top plot, red), together with iterative estimation of segment length probability distribution (bottom plot, black), where the maximum indicates the …
Figure 4.6
Figure 4.6. Figure 4.6: Performance of change points detection algorithms on artificial 24-hour tachograms. All algorithms were tested with one, five and ten second change point detection tolerance level and for a range of free parameter values. number of incorrect detections (FP). 4.5.1.6 …
Figure 4.7
Figure 4.7. Figure 4.7: Example segmentation of L5 and M10 periods. Activity time series in black, dark grey regions represent the most active 10 hours and light grey the least active 5 hours of activity. Red lines shows the beginning and end of the analysis period. The identification was p…
Figure 4.8
Figure 4.8. Figure 4.8: Probability distributions of bed- and wake- time activity observations and state durations, identified based on Fitbit bed/wake segmentation. 4.5.3.4 Segmentation results Bed/wake segmentation was performed using the proposed Hidden Markov Model and the accuracy of s…
Figure 4.9
Figure 4.9. Figure 4.9: Bed and wake time estimation error for each subject, representing offset of HMM￾estimated state change comparing to Fitbit reference. On each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers extend to the most …
Figure 4.10
Figure 4.10. Figure 4.10: Bed and wake time estimation error, representing offset of HMM-estimated state change comparing to Fitbit reference. and positive for wake time annotation, mentioned in Section 4.5.3. The reported bed time accuracy identification using sleep reports is 0.80 hours (L…
Figure 5.1
Figure 5.1. Figure 5.1: Example of Sample Entropy calculation for a specific template. Red, green and blue squares represent measurements of a specific template within a similarity threshold r (shown with dashed lines). The probability of matching the template of length 3, given the templat…
Figure 5.2
Figure 5.2. Figure 5.2: Example of Bayesian segmentation of one hour of locomotor activity data. On the top plot one hour of activity is presented in black, and the identified stationary segments as white and purple overlays. On the bottom plot the durations of identified segments are prese…
Figure 5.3
Figure 5.3. Figure 5.3: Example of Detrended Fluctuation Analysis calculation for a specific segment length. In each segment, the local trend is presented as a red line. Adapted from (Peng et al., 1994). If the resulting signal F 2 d (l) is proportional to the l α , the α is called the “sca…
Figure 5.4
Figure 5.4. Figure 5.4: Example of matching activity patterns of days X and Y using the Dynamic Time Warping. Picture a) represents mapping between the activity patterns and picture b) shows the distance matrix between activity levels on days X and Y (colour) together with the optimal match…
Figure 6.1
Figure 6.1. Figure 6.1: Decision boundary of SVM classifier. Filled and empty dots represent two different classes with support vectors selected in red. The solid line represents the maximum-margin hyperplane and dashed lines represent margins. The SVM classifier allows for more flexibility…
Figure 8.1
Figure 8.1. Figure 8.1: Distribution of self-reported mood scores between participant groups. Figure a) shows the ASRM scores, b) shows the QIDS-16SR scores and c) shows the GAD-7 scores. For each collected mood score the preceding week of phone activity data was analysed to identify object…
Figure 8.2
Figure 8.2. Figure 8.2: Example of LASSO lambda parameter selection using the MSE of cross-validated regression. The green dotted line represents the parametrisation with minimal MSE and blue dotted line represents the parametrisation with the least number of features and MSE within one SD …
Figure 8.3
Figure 8.3. Figure 8.3: Comparison of time series smoothing techniques for a single activity feature (LEVEL_AVG, defined in Section 5.3.1) of a single subject. Note, that in the case of LEVEL_AVG smoothing of daily average activity is equivalent to smoothing of the original activity signal.…
Figure 8.4
Figure 8.4. Figure 8.4: Example of results of mood regression model for a single subject. Regression was performed for all self-reported mood scores and smoothed activity features of a single subject. 2.13±1.63 for ASRM, 2.65±1.47 for QIDS-16SR and 2.51±1.31 for GAD-7 questionnaires as desc…
Figure 9.1
Figure 9.1. Figure 9.1: Adhesive activity and heart rate monitoring patch used in the study (Proteus Digital Health, Redwood City, CA, USA). Adopted from the company marketing materials. 9.3 Data analysis approach 9.3.1 Data pre-processing 9.3.1.1 The Nuffield data set Because actigraphic d…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

  1. [1]

    On this questionnaire are groups of five statements; read each group of statements carefully

    work page

  2. [2]

    Choose the one statement in each group that best describes the way you have been feeling for the past week

    work page

  3. [3]

    occasionally

    Circle the number next to the statement you picked. Please note:The word "occasionally" when used here means once or twice; "often" means several times or more; "frequently" means most of the time. A.2 Questions

    work page

  4. [4]

    Task Force of the European Society of Cardiology the North American Society of Pac- ing Electrophysiology (1996)

    Clinical features and conceptualization.Schizophrenia Research, 110(1-3):1–23. Task Force of the European Society of Cardiology the North American Society of Pac- ing Electrophysiology (1996). Heart rate variability: Standards of measurement, physiological interpretation, and clinical use.Circulation, 93(5):1043–1065. te Lindert, B. H. W. and Van Someren,...

    work page 1996

  5. [5]

    (b) I occasionally feel happier or more cheerful than usual

    Positive mood: (a) I do not feel happier or more cheerful than usual. (b) I occasionally feel happier or more cheerful than usual. 229 (c) I often feel happier or more cheerful than usual. (d) I feel happier or more cheerful than usual most of the time. (e) I feel happier or more cheerful than usual all of the time

    work page

  6. [6]

    (b) I occasionally feel more self-confident than usual

    Self-confidence: (a) I do not feel more self-confident than usual. (b) I occasionally feel more self-confident than usual. (c) I often feel more self-confident than usual. (d) I feel more self-confident than usual. (e) I feel extremely self-confident all of the time

    work page

  7. [7]

    (b) I occasionally need less sleep than usual

    Sleep patterns: (a) I do not need less sleep than usual. (b) I occasionally need less sleep than usual. (c) I often need less sleep than usual. (d) I frequently need less sleep than usual. (e) I can go all day and night without any sleep and still not feel tired

    work page

  8. [8]

    (b) I occasionally talk more than usual

    Speech: (a) I do not talk more than usual. (b) I occasionally talk more than usual. (c) I often talk more than usual. (d) I frequently talk more than usual. (e) I talk constantly and cannot be interrupted

    work page

Show all 32 references
  1. [9]

    (b) I have occasionally been more active than usual

    Activity level: 230 (a) I have not been more active (either socially, sexually, at work, home or school) than usual. (b) I have occasionally been more active than usual. (c) I have often been more active than usual. (d) I have frequently been more active than usual. (e) I am c...

  2. [10]

    (b) I take at least 30 minutes to fall asleep, less than half the time

    Falling Asleep: (a) I never take longer than 30 minutes to fall asleep. (b) I take at least 30 minutes to fall asleep, less than half the time. (c) I take at least 30 minutes to fall asleep, more than half the time. (d) I take more than 60 minutes to fall alseep, more than hal...

  3. [11]

    (b) I have a restless, light sleep with a few brief awakenings each night

    Sleep During the Night: 232 (a) I do not wake up at night. (b) I have a restless, light sleep with a few brief awakenings each night. (c) I wake up at least once a night, but I go back to sleep easily. (d) I awaken more than once a night and stay awake for 20 minutes or more, ...

  4. [12]

    (b) More than half the time, I awaken more than 30 minutes before I need to get up

    Waking Up Too Early: (a) Most of the time, I awaken no more than 30 minutes before I need to get up. (b) More than half the time, I awaken more than 30 minutes before I need to get up. (c) I almost always awaken at least one hour or so before I need to, but I go back to sleep ...

  5. [13]

    (b) I sleep no longer than 10 hours in a 24-hour period including naps

    Sleeping Too Much: (a) I sleep no longer than 7–8 hours/night, without napping during the day. (b) I sleep no longer than 10 hours in a 24-hour period including naps. (c) I sleep no longer than 12 hours in a 24-hour period including naps. (d) I sleep longer than 12 hours in a ...

  6. [14]

    (b) I feel sad less than half the time

    Feeling Sad: (a) I do not feel sad. (b) I feel sad less than half the time. (c) I feel sad more than half the time. (d) I feel sad nearly all of the time

  7. [15]

    (b) I eat somewhat less often or lesser amounts of food than usual

    Decreased Appetite: 233 (a) There is no change in my usual appetite. (b) I eat somewhat less often or lesser amounts of food than usual. (c) I eat much less than usual and only with personal effort. (d) I rarely eat within a 24-hour period, and only with extreme personal effort ...

  8. [16]

    (b) I feel a need to eat more frequently than usual

    Increased Appetite: (a) There is no change from my usual appetite. (b) I feel a need to eat more frequently than usual. (c) I regularly eat more often and/or greater amounts of food than usual. (d) I feel driven to overeat both at mealtime and between meals

  9. [17]

    (b) I feel as if I’ve had a slight weight loss

    Decreased Weight (Within the Last Two Weeks): (a) I have not had a change in my weight. (b) I feel as if I’ve had a slight weight loss. (c) I have lost 2 pounds or more. (d) I have lost 5 pounds or more

  10. [18]

    (b) I feel as if I’ve had a slight weight gain

    Increased Weight (Within the Last Two Weeks): (a) I have not had a change in my weight. (b) I feel as if I’ve had a slight weight gain. (c) I have gained 2 pounds or more. (d) I have gained 5 pounds or more

  11. [19]

    234 (b) I occasionally feel indecisive or find that my attention wanders

    Concentration/Decision Making: (a) There is no change in my usual capacity to concentrate or make decisions. 234 (b) I occasionally feel indecisive or find that my attention wanders. (c) Most of the time, I struggle to focus my attention or to make decisions. (d) I cannot conce...

  12. [20]

    (b) I am more self-blaming than usual

    View of Myself: (a) I see myself as equally worthwhile and deserving as other people. (b) I am more self-blaming than usual. (c) I largely believe that I cause problems for others. (d) I think almost constantly about major and minor defects in myself

  13. [21]

    (b) I feel that life is empty or wonder if it’s worth living

    Thoughts of Death or Suicide: (a) I do not think of suicide or death. (b) I feel that life is empty or wonder if it’s worth living. (c) I think of suicide or death several times a week for several minutes. (d) I think of suicide or death several times a day in some detail, or ...

  14. [22]

    (b) I notice that I am less interested in people or activities

    General Interest: (a) There is no change from usual in how interested I am in other people or activities. (b) I notice that I am less interested in people or activities. (c) I find I have interest in only one or two of my formerly pursued activities. (d) I have virtually no int...

  15. [23]

    (b) I get tired more easily than usual

    Energy Level: (a) There is no change in my usual level of energy. (b) I get tired more easily than usual. 235 (c) I have to make a big effort to start or finish my usual daily activities (for example, shopping, homework, cooking or going to work). (d) I really cannot carry out m...

  16. [24]

    (b) I find that my thinking is slowed down or my voice sounds dull or flat

    Feeling Slowed Down: (a) I think, speak, and move at my usual rate of speed. (b) I find that my thinking is slowed down or my voice sounds dull or flat. (c) It takes me several seconds to respond to most questions and I’m sure my thinking is slowed. (d) I am often unable to resp...

  17. [25]

    (b) I’m often fidgety, wringing my hands, or need to shift how I am sitting

    Feeling Restless: (a) I do not feel restless. (b) I’m often fidgety, wringing my hands, or need to shift how I am sitting. (c) I have impulses to move about and am quite restless. (d) At times, I am unable to stay seated and need to pace around. 236 Appendix C The Generalised A...

  18. [26]

    (b) Several days

    Feeling nervous, anxious, or on edge: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day

  19. [27]

    (b) Several days

    Not being able to stop or control worrying: (a) Not at all. (b) Several days. 237 (c) More than half the days. (d) Nearly every day

  20. [28]

    (b) Several days

    Worrying too much about different things: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day

  21. [29]

    (b) Several days

    Trouble relaxing: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day

  22. [30]

    (b) Several days

    Being so restless that it is hard to sit still: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day

  23. [31]

    (b) Several days

    Becoming easily annoyed or irritable: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day. 238

  24. [32]

    (b) Several days

    Feeling afraid as if something awful might happen: (a) Not at all. (b) Several days. (c) More than half the days. (d) Nearly every day. 239

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.