Annemieke Witteveen

Background: Consumer wearables are increasingly being integrated into health research for data collection. Although they are attractive to use, the accuracy of their photoplethysmography (PPG)-based measurements can be influenced by user characteristics such as sex, age, BMI, and skin tone. However, our knowledge regarding the validity of these measurements in certain populations, such as those with darker skin tones, seems limited. This is concerning as uncorrected differences in measurement accuracy can lead to health disparities when consumer wearable measurements are used more frequently. A potential cause for the gap in our knowledge regarding consumer wearable validity is the underrepresentation of certain population groups in studies validating PPG-based consumer wearables. Objective: This scoping review aimed to map the representation of different sex, age, BMI, and skin tone groups in studies assessing the validity of PPG-based pulse rate, heart rate variability, blood pressure, peripheral blood oxygen saturation (SpO ₂ ), and respiratory rate measurements of consumer wearables. Methods: A literature search was conducted in Scopus, PubMed, and IEEE Xplore in July 2025. Papers were eligible if they assessed the validity of consumer wearable PPG-based measurements, expressed as the agreement with a reference method. From the included papers, the study population distribution of sex, age, BMI, and Fitzpatrick scale was extracted. To evaluate the representation, percentages of people in specific age, BMI, and skin tone groups were estimated based on reported means and SDs. The median percentage of participants in each population group, as well as the total percentage, is reported. Results: After the removal of duplicates, 734 papers were screened for eligibility. Following title and abstract screening, 238 papers remained, of which 186 passed full-text screening and were included in the review. Most of the studies (n=160) focused on pulse rate. Sex, age, BMI, and Fitzpatrick scale were reported by 179 (96.0%), 178 (96.0%), 101 (54.0%), and 35 (19.0%) out of 186 studies, respectively. While the median representation was 0% (IQR 0%-8%) for both older adults (>65 y) and individuals with obesity (BMI>30 kg/m ² ; IQR 0%-13%), aggregate participation across all studies was higher (1290/6367, 20.0% and 473/3428, 14.0%, respectively). Individuals with underweight (BMI2 ) remained rare (median 3%, IQR 0%-7%), and the aggregate was 7.0% (225/3428). The median percentage of people with darker skin tones (Fitzpatrick type V and VI) participating in a study was 0%. Conclusions: Based on our results, it can be concluded that older adults and people with underweight, obesity, or darker skin tones are generally underrepresented in studies assessing the validity of consumer wearable PPG-based measurements. Future validation studies should focus more on the representativeness of the study population. This can be achieved by setting a benchmark for representativeness and including study population representatives during the study design process.

Background: Postoperative complications remain a major concern following cancer surgery, both during hospitalization and after discharge. Continuous monitoring of physiological signals such as heart rate and respiratory rate using wearable sensors is now feasible. However, interpreting these time series for early physiological deviations remains challenging, particularly when complication onset labels are unavailable.Method: This study proposes an unsupervised change point detection (CPD) method, the Past–Future Long Short-Term Memory Autoencoder (PF-LSTM-AE), and compares its performance with representative approaches for identifying physiological deviations associated with postoperative complications of interest (Clavien–Dindo grade ≥ 2). PF-LSTM-AE was evaluated and its performance compared with a rule-based Remote Early Warning Score (REWS), a standard Long Short-Term Memory Autoencoder (AE), and a patient-specific Adaptive LSTM-Autoencoder (ALACPD). Data were obtained from 21 cancer patients recovering from major abdominal surgery and monitored using a wearable sensor for a median of 13.92 days. Cross-patient and patient-specific evaluations were performed, Raw change points (CPs) and temporally clustered CPs using a 1-hour window were assessed using detection coverage, detection timing, CP frequency, and inter-change point interval.Results: All methods identified physiological deviations preceding postoperative complications, with early detection coverage ranging from 75% (3/4) to 100% (4/4). However, they differed in detection timing, CP frequency, and temporal clustering. Clustering substantially reduced additional CPs for AE and PF-LSTM-AE, improving temporal interpretability, while ALACPD remained largely unchanged.Conclusion: Unsupervised CPD methods may help identify early physiological deviations during postoperative recovery. However, further improvements in interpretability, robustness, and clinical validation are needed before these methods can be considered for clinical implementation.

Postoperative complications remain a major concern following cancer surgery, both during hospitalization and after discharge. Continuous monitoring of physiological signals such as heart rate and respiratory rate using wearable sensors is now feasible. However, interpreting these time series for early signs of deterioration remains challenging, particularly in the absence of labeled complication onset. This study proposes a new unsupervised change point detection (CPD) method, the Past–Future Long Short-Term Memory Autoencoder (PF-LSTM-AE), and compares its performance with existing approaches for early detection of physiological deviations associated with postoperative complications of interest (Clavien–Dindo grade ≥ 2). Four methods were evaluated: a Remote Early Warning Score (REWS), a standard Long Short-Term Memory Autoencoder (AE), the proposed PF-LSTM-AE, and a state-of-the-art Adaptive LSTM-Autoencoder (ALACPD). Data were obtained from 21 cancer patients recovering from major abdominal surgery and monitored using a wearable sensor for a median of 13.92 days. Two analytical approaches were investigated: raw change points (CPs) and temporally clustered change points using a 1-hour aggregation window. Performance was assessed using detection coverage, detection time, change points per day, and inter-change point interval. All methods detected CPs preceding postoperative complications of interest. Clustering substantially reduced additional CPs for AE and PF-LSTM-AE, improving temporal interpretability, while ALACPD remained largely unchanged. These findings suggest that unsupervised AE-based methods, particularly PF-LSTM-AE, show promise for identifying early signs of postoperative complications using continuous physiological monitoring, and that simple post-processing such as clustering can improve interpretability while reducing additional CPs.

BACKGROUND Consumer wearables are increasingly being integrated into health research for data collection. Though they are attractive to use, the accuracy of their photoplethysmography (PPG)-based measurements can be influenced by user characteristics, such as sex, age, BMI, and skin tone. Yet, our knowledge of the validity of these measurements in some populations, such as populations with darker skin tones and older adults, seems limited. This is concerning as uncorrected differences in measurement accuracy can lead to health disparities when consumer wearables measurements are being used more frequently. A potential cause for the gap in our knowledge on consumer wearable validity is the underrepresentation of some population groups in studies validating PPG-based consumer wearables. OBJECTIVE he aim of this scoping review is to map the representation of different sex, age, BMI, and skin tone groups in studies assessing the validity of PPG-based pulse rate, heart rate variability, blood pressure, peripheral blood oxygen saturation (SpO2), and respiratory rate measurements of consumer wearables. With this overview, we identified gaps in representation, and recommendations can be made to close these gaps. METHODS A literature search was conducted in Scopus, PubMed, and IEEE Xplore in June 2024 and July 2025. Papers were eligible if they assessed the validity of consumer wearable PPG-based measurements expressed as the agreement with a reference method. From the included papers, the study population distribution of sex, age, BMI, and Fitzpatrick scale was extracted. To evaluate the representation, the percentage of people in specific age, BMI, and skin tone groups was estimated based on reported means and standard deviations. The median percentage of participants in each population group, as well as the total percentage, is reported RESULTS After removal of duplicates, 734 papers were screened for eligibility. After title/abstract screening, 238 papers remained, of which 186 passed full text screening and were included in the review. Most of the studies (n=160) focused on pulse rate. Sex, age, BMI, and Fitzpatrick scale were reported by 95%, 94%, 54% and 19% of studies, respectively. The median% of older adults (>65 years), people with underweight (BMI <18.5 kg/m2), people with obesity (BMI ≥30 kg/m2), and people with darker skin tones (Fitzpatrick type V and VI) participating in a study was 0%. CONCLUSIONS Based on our results, it can be concluded that older adults, and people with underweight, obesity, or darker skin tones are underrepresented in studies assessing the validity of consumer wearable PPG-based measurements. Future validation studies should focus more on the representativeness of the study population. This can be achieved by setting a benchmark for representativeness and including study population representatives in the study design process.

Introduction. Many breast cancer survivors experience cancer-related fatigue (CRF), and several interventions to treat CRF are available. One way to tailor intervention advice is based on patient preferences. In this study, we explore preference heterogeneity regarding between-attribute and within-attribute preferences. In addition, we propose simple decision rules to match preferences to interventions. Methods. Nine attributes were included with dichotomized levels. Participants selected their preferred level per attribute and ranked the attributes using best-worst scaling. Between-attribute and within-attribute preferences were determined, together with their heterogeneity. Using decision rules, matching scores were calculated for a hypothetical intervention. Results. Sixty-seven breast cancer survivors completed the survey. They were on average 52 y old, 4.5 y after diagnosis, experienced CRF (6.5–7.2/10) on 3 dimensions (physical, mental, and emotional), and 43% already followed an intervention for CRF. Overall, participants ranked costs highest. Next to costs, proven effectiveness and type of intervention were also frequently ranked first. Only 13 participants (19%) shared the most common preference pattern of shorter interventions, daily sessions, shorter session time, a psychosocial intervention, no anonymity, and contact with a therapist and peers. Matching scores for a hypothetical intervention with attributes corresponding with the overall within-attribute preferences varied from 44% to 100%. Conclusion. A large heterogeneity in preferences of breast cancer survivors for CRF intervention attributes was demonstrated. Using simple decision rules, the effect of this heterogeneity on linking preferences to interventions with matching scores was demonstrated. Implications. Personalization of intervention advice is necessary due to preference heterogeneity. Tailored advice can result in higher involvement of patients in decision making, intervention adherence and satisfaction, and subsequently a potential higher quality of life after breast cancer. Highlights: Many breast cancer survivors experience cancer-related fatigue for which many interventions exist. Our results show large preference heterogeneity in breast cancer patients’ preferences for attributes of eHealth interventions. Based on this preference heterogeneity, intervention advice for cancer-related fatigue after breast cancer can be personalized, ultimately improving quality of life after breast cancer.

Background and objective:Cancer-related fatigue (CRF) affects the quality of life after breast cancer. In a previous study, we developed a 72-item questionnaire that assesses CRF from a holistic point of view; named the Holistic Assessment of CRF (HA-CRF) questionnaire. The current study assessed the face and content validity of the HA-CRF questionnaire.Methods:Using a mixed-method approach, ten breast cancer survivors (BCS) did a cognitive walkthrough of the HA-CRF via an app followed by a semi-structured interview about relevancy and essentiality (qualitative). In addition, ten health care professionals (HCPs) assessed the relevancy, clarity, and essentiality of each item via a questionnaire (quantitative).Results:BCS indicated minor textual improvement for four items and six items were not completely clear. The app was considered easy to use and the HC-CRF was on average completed in 18 minutes. The HA-CRF questionnaire provided openness about fatigue and gave the feeling of being heard. The items were helpful and induced self-awareness. HCPs indicated 71% of items being very clear or minor revisions proposed by the minority, with 64% of items being essential and 92% considered relevant.Conclusions:The HA-CRF showed good face and excellent content validity. Further research is needed to assess its ability to monitor in daily life.

Background: For hip fracture patients with a limited life expectancy, operative and palliative non-operative management (P-NOM) can yield similar quality of life outcomes. However, evidence on when to abstain from surgery is lacking. The aim of this study was to quantify the influence of patient characteristics on surgeons’ decisions to recommend P-NOM. Methods: Dutch surgical residents and orthopaedic trauma surgeons were enrolled in a conjoint analysis and structured expert judgement (SEJ). The participants assessed 16 patient cases comprising 10 clinically relevant characteristics. For each case, they recommended either surgery or P-NOM and estimated the 30-day postoperative mortality risk. Treatment recommendations were analysed using Bayesian logistic regression, and perceived risks were pooled with equal and performance-based weights using Cooke’s Classical Model. Results: The conjoint analysis and SEJ were completed by 14 and 9 participants, respectively. Participants were more likely to recommend P-NOM to patients with metastatic carcinomas (OR: 4.42, CrI: 2.14–8.95), severe heart failure (OR: 4.05, CrI: 1.89–8.29), end-stage renal failure (OR: 3.54, CrI: 1.76–7.35) and dementia (OR: 3.35, CrI: 1.70–7.06). The patient receiving the most P-NOM recommendations (12/14) had a pooled perceived risk of 30-day mortality between 50.8 and 62.7%. Conclusions: Overall, comorbidities had the strongest influence on participants’ decisions to recommend P-NOM. Nevertheless, practice variation and heterogeneity in risk perceptions were substantial. Hence, more decision support for considering P-NOM is needed.

Purpose: To understand the consequences of diseases and treatment such as cancer and thus the needs of patients for surveillance and care and to improve quality of life, patients should be assessed using a holistic approach. However, instruments to create such a holistic view do not exist and the development presents unique challenges. Therefore, this study presents a method for the development of a holistic assessment questionnaire using cancer‐related fatigue (CRF) as a case.Method: We started with (1) the definition of our construct of interest (CRF) on the theme level followed by (2) item selection, an iterative process of searching for validated questionnaires that together cover the full holistic construct. The construct definition on theme level (1) was too broad and was, therefore, redefined on the element level (construct > theme > element) based on interviews with relevant stakeholders. Hereafter, item selection (2) was performed on the element level based on a priority list, psychometric properties (e.g., discriminative parameter value) and consultation of experts and future users. Lastly, (3) items were reformulated.Results: Initial CRF construct definition (1) resulted in 110 relevant validated questionnaires with over three thousand items, requiring a construct redefinition on element level. Seventy‐two items from 21 validated questionnaires were included (2) in the preliminary holistic assessment questionnaire. For item reformulation (3), easy language was used to better suit the target population.Conclusion: Tailoring care to the individual requires a holistic view. This article presents a novel method to develop a holistic assessment questionnaire, including an example for CRF, with several recommendations for cancer‐specific instrument development. Although the development process of a holistic assessment questionnaire is time‐consuming, more late and long‐term effects of cancer are multidimensional and could benefit from a holistic approach in their assessment to enable personalised care, thereby improving quality of life and reducing societal impact.

Objective: Cancer- related fatigue (CRF) is one of the most reported long-term effects after breast cancer and severely impacts quality of life. To come towards optimal treatment of multidimensional CRF, the first step is to use a holistic approach to develop a holistic patient profile including the patient's experience and impact of CRF on their life. Methods and measures: Four semi- structured focus groups with twenty- seven breast cancer patients and fourteen interviews with healthcare professionals (HCPs) were held. Reflexive thematic analysis was used to define (sub)themes for the holistic patient profile. The themes of the interviews and focus groups were compared for validity. Results: Breast cancer patients and HCPs described the same five major themes, consisting of experience of CRF, impact and consequences, coping, personality, and CRF treatment. Experience of CRF consists of cognitive, emotional, and physical aspects. Impact and consequences include work, family, partner relation, social contact and hobbies, body, and misrecognition. Coping consists of twelve (mal)adaptive strategies. Personality and CRF treatment were summarised as themes. Conclusions: A first holistic patient profile was introduced for CRF for breast cancer. This profile can be conceptualized into a questionnaire to collect information for personalized treatment recommendations and monitoring of CRF over time.