Development of a Multilingual Web-Based Food Frequency Questionnaire for Adults in Switzerland

Dietary assessment is a major challenge in epidemiological research and is associated with a high time and financial burden. Automated food frequency questionnaires (FFQs) have the potential to rapidly collect dietary intake data in large studies while reducing human error risk during data processing. We developed a semiquantitative, multilingual, electronic FFQ for real-time dietary intake assessment in the Swiss adult population, called "Swiss eFFQ". The iterative development process involved stages of content identification, construction, pretesting, translation, and adaptation of the FFQ. Using 24 h dietary recalls from 2085 participants aged 18-75 years from a nationally representative survey, we conducted a stepwise regression analysis to identify foods contributing to >90% of the variance in intakes of energy and six nutrients. All 118 foods identified in the overall cohort or in any of the Swiss linguistic regions were selected and standardized to define the comprehensive 83-item food list, covering >90% of the intake of key nutrients in the entire study population. Once validated, the Swiss eFFQ can be used to classify individuals based on their habitual diets. The methodology described in this paper enhances the transparency of the Swiss eFFQ and may help researchers to develop multilingual dietary assessment tools for other populations.

Blending citizen science with natural language processing and machine learning: Understanding the experience of living with multiple sclerosis

The emergence of new digital technologies has enabled a new way of doing research, including active collaboration with the public ('citizen science'). Innovation in machine learning (ML) and natural language processing (NLP) has made automatic analysis of large-scale text data accessible to study individual perspectives in a convenient and efficient fashion. Here we blend citizen science with innovation in NLP and ML to examine (1) which categories of life events persons with multiple sclerosis (MS) perceived as central for their MS; and (2) associated emotions. We subsequently relate our results to standardized individual-level measures. Participants (n = 1039) took part in the 'My Life with MS' study of the Swiss MS Registry which involved telling their story through self-selected life events using text descriptions and a semi-structured questionnaire. We performed topic modeling ('latent Dirichlet allocation') to identify high-level topics underlying the text descriptions. Using a pre-trained language model, we performed a fine-grained emotion analysis of the text descriptions. A topic modeling analysis of totally 4293 descriptions revealed eight underlying topics. Five topics are common in clinical research: 'diagnosis', 'medication/treatment', 'relapse/child', 'rehabilitation/wheelchair', and 'injection/symptoms'. However, three topics, 'work', 'birth/health', and 'partnership/MS' represent domains that are of great relevance for participants but are generally understudied in MS research. While emotions were predominantly negative (sadness, anxiety), emotions linked to the topics 'birth/health' and 'partnership/MS' was also positive (joy). Designed in close collaboration with persons with MS, the 'My Life with MS' project explores the experience of living with the chronic disease of MS using NLP and ML. Our study thus contributes to the body of research demonstrating the potential of integrating citizen science with ML-driven NLP methods to explore the experience of living with a chronic condition.

Feelings of loneliness, COVID-19-specific-health anxiety and depressive symptoms during the first COVID-19 wave in Swiss persons with multiple sclerosis

The aim of our study was to investigate whether self-reported feeling of loneliness (FoL) and COVID-19-specific health anxiety were associated with the presence of depressive symptoms during the first coronavirus disease 2019 (COVID-19) wave. Questionnaires of 603 persons of the Swiss Multiple Sclerosis Registry (SMSR) were cross-sectionally analyzed using descriptive and multivariable regression methods. The survey response rate was 63.9%. Depressive symptoms were assessed by the Beck Depression Inventory-Fast Screen (BDI-FS). COVID-19-specific health anxiety and FoL were measured using two 5-item Likert scaled pertinent questions. High scoring FoL (2.52, 95% confidence interval (CI) (2.06-2.98)) and/or COVID-19 specific health anxiety (1.36, 95% CI (0.87-1.85)) were significantly associated with depressive symptoms. Further stratification analysis showed that the impact of FoL on depressive symptoms affected all age groups. However, it was more pronounced in younger PwMS, whereas an impact of COVID-19 specific health anxiety on depressive symptoms was particularly observed in middle-aged PwMS. FoL and COVID-19-specific health anxiety were age-dependently associated with depressive symptoms during the first COVID-19 wave in Switzerland. Our findings could guide physicians, health authorities, and self-help groups to better accompany PwMS in times of public health crises.

Validation of a Visually Aided Dietary Assessment Tool to Estimate Dietary Intake in an Adult Swiss Population

Background

Accurately assessing dietary intake is crucial for understanding how diet affects a person’s health. In large cohorts, paper-based dietary assessment tools (DAT) such as food recalls or food frequency questionnaires have emerged as valid tools with a low burden for participants.

Objective

To validate a visually aided DAT for use in studies with Swiss adults against the gold standard of a weighed 7-day food record (7 d-FR).

Design

Fifty-one adults (n = 24 women, n = 27 males) participated in the study and were recruited within two age groups (20–40 and 50–70 y). Each participant filled out the visually aided DAT, then the 7 d-FR. The DAT was compared to the 7 d-FR for total energy intake, macronutrients, sugar, water, and portions of fruits and vegetables. Pearson correlation and Bland–Altman analyses were used for statistical analyses.

Results

Total correlations ranged from 0.288 (sugar, p < 0.05) to 0.729 (water, p < 0.01). The older age group showed higher correlations for total energy intake, protein, fats, carbohydrates, and sugar, but not for water (p < 0.05). Correlations were moderate at r > 0.5, whereas only water and protein reached those values in the young group. Both groups overestimated total calories in kcal (+14.0%), grams of protein (+ 44.6%), fats (+36.3%), and portions of fruits and vegetables (+16.0%) but strongly underestimated sugar intake (−50.9%).

Conclusion

This DAT showed that all macronutrients and total energy intake were estimated more accurately by the older age group and therefore might be adequate to capture dietary habits in older Swiss adults.

The Multiple Sclerosis Data Alliance Catalogue: Enabling Web-Based Discovery of Metadata from Real-World Multiple Sclerosis Data Sources

Background

One of the major objectives of the Multiple Sclerosis Data Alliance (MSDA) is to enable better discovery of multiple sclerosis (MS) real-world data (RWD).

Methods

We implemented the MSDA Catalogue, which is available worldwide. The current version of the MSDA Catalogue collects descriptive information on governance, purpose, inclusion criteria, procedures for data quality control, and how and which data are collected, including the use of e-health technologies and data on collection of COVID-19 variables. The current cataloguing procedure is performed in several manual steps, securing an effective catalogue.

Results

Herein we summarize the status of the MSDA Catalogue as of January 6, 2021. To date, 38 data sources across five continents are included in the MSDA Catalogue. These data sources differ in purpose, maturity, and variables collected, but this landscaping effort shows that there is substantial alignment on some domains. The MSDA Catalogue shows that personal data and basic disease data are the most collected categories of variables, whereas data on fatigue measurements and cognition scales are the least collected in MS registries/cohorts.

Conclusions

The Web-based MSDA Catalogue provides strategic overview and allows authorized end users to browse metadata profiles of data cohorts and data sources. There are many existing and arising RWD sources in MS. Detailed cataloguing of MS RWD is a first and useful step toward reducing the time needed to discover MS RWD sets and promoting collaboration.

Onset Symptom Clusters in Multiple Sclerosis: Characteristics, Comorbidities, and Risk Factors

Background: Multiple sclerosis (MS) symptoms are expected to aggregate in specific patterns across different stages of the disease. Here, we studied the clustering of onset symptoms and examined their characteristics, comorbidity patterns and associations with potential risk factors.

Methods: Data stem from the Swiss Multiple Sclerosis Registry, a prospective study including 2,063 participants by November 2019. MS onset symptoms were clustered using latent class analysis (LCA). The latent classes were further examined using information on socio-demographic characteristics, MS-related features, potential risk factors, and comorbid diseases.

Results: The LCA model with six classes (frequencies ranging from 12 to 24%) was selected for further analyses. The latent classes comprised a multiple symptoms class with high probabilities across several symptoms, contrasting with two classes with solitary onset symptoms: vision problems and paresthesia. Two gait classes emerged between these extremes: the gait-balance class and the gait-paralysis class. The last class was the fatigue-weakness-class, also accompanied by depression symptoms, memory, and gastro-intestinal problems. There was a moderate variation by sex and by MS types. The multiple symptoms class yielded increased comorbidity with other autoimmune disorders. Similar to the fatigue-weakness class, the multiple symptoms class showed associations with angina, skin diseases, migraine, and lifetime prevalence of smoking. Mononucleosis was more frequently reported in the fatigue-weakness and the paresthesia class. Familial aggregation did not differ among the classes.

Conclusions: Clustering of MS onset symptoms provides new perspectives on the heterogeneity of MS. The clusters comprise different potential risk factors and comorbidities. They point toward different risk mechanisms.

Major depressive disorder subtypes and depression symptoms in multiple sclerosis: What is different compared to the general population?

OBJECTIVE

To compare and characterize major depressive disorder (MDD) subtypes (i.e., pure atypical, pure melancholic and mixed atypical-melancholic) and depression symptoms in persons with multiple sclerosis (PwMS) with persons without MS (Pw/oMS) fulfilling the DSM-5 criteria for a past 12-month MDD.

METHODS

MDD in PwMS (n = 92) from the Swiss Multiple Sclerosis Registry was compared with Pw/oMS (n = 277) from a Swiss community-based study. Epidemiological MDD diagnoses were based on the Mini-SPIKE (shortened form of the Structured Psychopathological Interview and Rating of the Social Consequences for Epidemiology). Logistic and multinomial regression analyses (adjusted for sex, age, civil status, depression and severity) were computed for comparisons and characterization. Latent class analysis (LCA) was conducted to empirically identify depression subtypes in PwMS.

RESULTS

PwMS had a higher risk for the mixed atypical-melancholic MDD subtype (OR = 2.22, 95% CI = 1.03-4.80) compared to Pw/oMS. MDD in PwMS was specifically characterized by a higher risk of the two somatic atypical depression symptoms 'weight gain' (OR = 6.91, 95% CI = 2.20-21.70) and 'leaden paralysis' (OR = 3.03, 95% CI = 1.35-6.82) and the symptom 'irritable/angry' (OR = 3.18, 95% CI = 1.08-9.39).

CONCLUSIONS

MDD in PwMS was characterized by a higher risk for specific somatic atypical depression symptoms and the mixed atypical-melancholic MDD subtype. The pure atypical MDD subtype, however, did not differentiate between PwMS and Pw/oMS. Given the high phenomenological overlap with MS symptoms, the mixed atypical-melancholic MDD subtype represents a particular diagnostic challenge.

60/30: 60% of the Morbidity-Associated Multiple Sclerosis Disease Burden Comes From the 30% of Persons With Higher Impairments

Background: Multiple sclerosis (MS) is the most common chronic, non-traumatic, neurologic disease in young adults. While approximate values of the disease burden of MS are known, individual drivers are unknown.

Objective: To estimate the age-, sex-, and disease severity-specific contributions to the disease burden of MS.

Methods: We estimated the disease burden of MS using disability-adjusted life years (DALYs) following the Global Burden of Disease study (GBD) methodology. The data sources consisted of the Swiss MS Registry, a recent prevalence estimation, and the Swiss mortality registry.

Results: The disease burden of MS in Switzerland in 2016 was 6,938 DALYs (95%-interval: 6,018-7,955), which corresponds to 97 DALYs per 100,000 adult inhabitants. Morbidity contributed 59% of the disease burden. While persons in an asymptomatic (EDSS-proxy 0) and mild (EDSS-proxy >0–3.5) disease stage represent 68.4% of the population, they make up 39.8% of the MS-specific morbidity. The remaining 60.2% of the MS-specific morbidity stems from the 31.6% of persons in a moderate (EDSS-proxy 4–6.5) or severe (EDSS-proxy ≥7) disease stage.

Conclusions: Morbidity has a larger influence on the disease burden of MS than mortality and is shared in a ratio of 2:3 between persons in an asymptomatic/mild and moderate/severe disease stage in Switzerland. Interventions to reduce severity worsening in combination with tailored, symptomatic treatments are important future paths to lower the disease burden of MS.

A step beyond the hygiene hypothesis-immune-mediated classes determined in a population-based study

BACKGROUND

Comorbidity patterns of childhood infections, atopic diseases, and adverse childhood experiences (ACE) are related to immune system programming conditions. The aim of this study was to make a step beyond the hygiene hypothesis and to comprehensively classify these patterns with latent class analysis (LCA). A second aim was to characterize the classes by associations with immunological, clinical, and sociodemographic variables.

METHODS

LCA was applied to data from the CoLaus|PsyCoLaus study (N = 4874, age range 35-82 years) separately for men and women. It was based on survey information on chickenpox, measles, mumps, rubella, herpes simplex, pertussis, scarlet fever, hay fever, asthma, eczema, urticaria, drug allergy, interparental violence, parental maltreatment, and trauma in early childhood. Subsequently, we examined how immune-mediated classes were reflected in leukocyte counts, inflammatory markers (IL-1β, IL-6, TNF-α, hsCRP), chronic inflammatory diseases, and mental disorders, and how they differed across social classes and birth cohorts.

RESULTS

LCA results with five classes were selected for further analysis. Latent classes were similar in both sexes and were labeled according to their associations as neutral, resilient, atopic, mixed (comprising infectious and atopic diseases), and ACE class. They came across with specific differences in biomarker levels. Mental disorders typically displayed increased lifetime prevalence rates in the atopic, the mixed, and the ACE classes, and decreased rates in the resilient class. The same patterns were apparent in chronic inflammatory diseases, except that the ACE class was relevant specifically in women but not in men.

CONCLUSIONS

This is the first study to systematically determine immune-mediated classes that evolve early in life. They display characteristic associations with biomarker levels and somatic and psychiatric diseases occurring later in life. Moreover, they show different distributions across social classes and allow to better understand the mechanisms beyond the changes in the prevalence of chronic somatic and psychiatric diseases.

How do patients enter the healthcare system after the first onset of multiple sclerosis symptoms? The influence of setting and physician specialty on speed of diagnosis

Background:

Diagnosing multiple sclerosis (MS) early is crucial to avoid future disability. However, potentially preventable delays in the diagnostic cascade from contact with a physician to definite diagnosis still occur and their causes are still unclear.

Objective:

To identify the possible causes of delays in the diagnostic process.

Methods:

We analyzed the data of the Swiss MS Registry. With logistic regression, we modeled the time from the first contact to the first consultation (contact-to-evaluation time, ⩽1 month/>1 month) and the evaluation-to-diagnosis time (⩽6 months/>6 months). Potential factors were health system characteristics, sociodemographic variables, first symptoms, and MS type.

Results:

We included 522 participants. Mostly, general practitioners (67%) were contacted first, without delaying the diagnosis. In contrast, first symptoms and MS type were the major contributors to delays: gait problems were associated with longer contact-to-evaluation times, depression as a concomitant symptom with longer evaluation-to-diagnosis times, and having primary progressive MS prolonged both phases. In addition, living in mountainous areas was associated with longer contact-to-evaluation times, whereas diagnosis after 2000 was associated with faster diagnoses.

Conclusion:

For a quicker diagnosis, awareness of MS as a differential diagnosis of gait disorders and the co-occurrence of depression at onset should be raised, and these symptoms should be attentively followed.

Factors associated with time from first-symptoms to diagnosis and treatment initiation of Multiple Sclerosis in Switzerland

BACKGROUND

Recent studies emphasise the importance of timely diagnosis and early initiation of disease-modifying treatment in the long-term prognosis of multiple sclerosis.

OBJECTIVES

The objective of this study was to investigate factors associated with extended time to diagnosis and time to disease-modifying treatment initiation in the Swiss Multiple Sclerosis Registry.

METHODS

We used retrospective data (diagnoses 1996-2017) of the survey-based Swiss Multiple Sclerosis Registry and fitted logistic regression models (extended time to diagnosis ≥2 years from first symptoms, extended time to disease-modifying treatment initiation ≥1 year from diagnosis) with demographic and a priori defined variables.

RESULTS

Our study, based on 996 persons with multiple sclerosis, suggests that 40% had an extended time to diagnosis, and extended time to disease-modifying treatment initiation was seen in 23%. Factors associated with extended time to diagnosis were primary progressive multiple sclerosis (odds ratio (OR) 5.09 (3.12-8.49)), diagnosis setting outside of hospital (neurologist (private practice) OR 1.54 (1.16-2.05)) and more uncommon first symptoms (per additional symptom OR 1.17 (1.06-1.30)). Older age at onset (per additional 5 years OR 0.84 (0.78-0.90)) and gait problems (OR 0.65 (0.47-0.89)) or paresthesia (OR 0.72 (0.54-0.95)) as first symptoms were associated with shorter time to diagnosis. Extended time to disease-modifying treatment initiation was associated with older age at diagnosis (per additional 5 years OR 1.18 (1.09-1.29)). In more recent years, time to diagnosis and time to disease-modifying treatment initiation tended to be shorter.

CONCLUSIONS

Even in recent periods, substantial and partially systematic variation regarding time to diagnosis and time to disease-modifying treatment initiation remains. With the emerging paradigm of early treatment, the residual variation should be monitored carefully.

The Swiss Multiple Sclerosis Registry (SMSR): study protocol of a participatory, nationwide registry to promote epidemiological and patient-centered MS research

Background

Multiple sclerosis (MS) is one of the most frequently observed neurological conditions in Switzerland, but data sources for country-wide epidemiological trend monitoring are lacking. Moreover, while clinical and laboratory MS research are generally well established, there is a gap in patient-centered MS research to inform care management, or treatment decisions and policy making not only in Switzerland but worldwide.

Methods

In light of these research gaps, the Swiss Multiple Sclerosis Society initiated and funded the Swiss Multiple Sclerosis Registry (SMSR) an open-ended, longitudinal and prospective, nationwide, patient-centered study.

The SMSR recruits adult persons with a suspected or confirmed MS diagnosis who reside or receive care in Switzerland. The SMSR has established a governance structure with clear rules and guidelines. It follows a citizen-science approach with direct involvement of persons with MS (PwMS), who contribute actively to registry development, operations, and research. Main scientific goals entail the study of MS epidemiology in Switzerland, health care access and provision, as well as life circumstances and wellbeing of persons with MS.

The innovative study design (“layer model”) offers several participation options with different time commitments. Data collection is by means of regular surveys and medical record abstraction. Survey participation is offered in different modes (web, paper & pencil) and in the three main national languages (German, French, Italian). Participants also receive regular data feedbacks for personal use and self-monitoring, contextualized in the whole population of study participants. Data feedbacks are also used to solicit data corrections of key variables from participants.

Discussion

The SMSR combines the advantages of traditional and novel research methods in medical research and has recruited over 1600 PwMS in its first year. The future-oriented design and technology will enable a response not only to future technological innovations and research trends, but also to challenges in health care provision for MS.

Trial registration

ClinicalTrials.gov NCT02980640; December 6, 2016; retrospectively registered.

A digitally facilitated citizen-science driven approach accelerates participant recruitment and increases study population diversity

QUESTION UNDER STUDY

Our aim was to assess whether a novel approach of digitally facilitated, citizen-science research, as followed by the Swiss Multiple Sclerosis Registry (Swiss MS Registry), leads to accelerated participant recruitment and more diverse study populations compared with traditional research studies where participants are mostly recruited in study centres without the use of digital technology.

METHODS

The Swiss MS Registry is a prospective, longitudinal, observational study covering all Switzerland. Participants actively contribute to the Swiss MS Registry, from defining research questions to providing data (online or on a paper form) and co-authoring papers. We compared the recruitment dynamics over the first 18 months with the a priori defined recruitment goals and assessed whether a priori defined groups were enrolled who are likely to be missed by traditional research studies.

RESULTS

The goal to recruit 400 participants in the first year was reached after only 20 days, and by the end of 18 months 1700 participants had enrolled in the Swiss MS Registry, vastly exceeding expectations. Of the a priori defined groups with potential underrepresentation in other studies, 645 participants (46.5%) received care at a private neurology practice, 167 participants (12%) did not report any use of healthcare services in the past 12 months, 32 (2.3%) participants lived in rural mountainous areas, and 20 (2.0% of the 1041 for whom this information was available) lived in a long-term care facility. Having both online and paper options increased diversity of the study population in terms of geographic origin and type and severity of disease, as well as use of health care services. In particular, paper enrolees tended to be older, more frequently affected by progressive MS types and more likely to have accessed healthcare services in the past 12 months.

CONCLUSION

Academic and industry-driven medical research faces substantial challenges in terms of patient involvement, recruitment, relevance and generalisability. Digital studies and stakeholder engagement may have enormous potential for medical research. But many digital studies are based on limited participant information and/or informed consent and unclear data ownership, and are subject to selection bias, confounding and information bias. The Swiss MS Registry serves as an example of a digitally enhanced, citizen-science study that leverages the advantages of both traditional medical research, with its established research methods, and novel societal and technological developments, while mitigating their ethical and legal disadvantages and risks.

Relative validation of a food frequency questionnaire to estimate food intake in an adult population

Background: Scientifically valid descriptions of dietary intake at population level are crucial for investigating diet effects on health and disease. Food frequency questionnaires (FFQs) are the most common dietary tools used in large epidemiological studies.

Objective: To examine the relative validity of a newly developed FFQ to be used as dietary assessment tool in epidemiological studies.

Design: Validity was evaluated by comparing the FFQ and a 4-day weighed food record (4-d FR) at nutrient and food group levels, Spearman’s correlations, Bland–Altman analysis and Wilcoxon rank sum tests were used. Fifty-six participants completed a paper format FFQ and a 4-d FR within 4 weeks.

Results: Corrected correlations between the two instruments ranged from 0.27 (carbohydrates) to 0.55 (protein), and at food group level from 0.09 (soup) to 0.92 (alcohol). Nine out of 25 food groups showed correlations > 0.5, indicating moderate validity. More than half the food groups were overestimated in the FFQ, especially vegetables (82.8%) and fruits (56.3%). Water, tea and coffee were underestimated (–14.0%).

Conclusions: The FFQ showed moderate relative validity for protein and the food groups fruits, egg, meat, sausage, nuts, salty snacks and beverages. This study supports the use of the FFQ as an acceptable tool for assessing nutrition as a health determinant in large epidemiological studies.

Cancer co-occurrence patterns in Parkinson's disease and multiple sclerosis-Do they mirror immune system imbalances?

BACKGROUND

To examine the site-specific cancer mortality among deaths registered with Parkinson's disease (PD) and multiple sclerosis (MS). We focused on the patterns related to the most frequent cancers.

METHODS

We analyzed Swiss mortality data over a 39-year period (1969-2007), using a statistical approach applicable to unique daabases, i.e. when no linkage with morbidity databases or disease registries is possible. It was based on a case-control design with bootstrapping to derive standardized mortality ratios (SMR). The cases were defined by the cancer-PD or cancer-MS co-registrations, whereas the controls were drawn from the remaining records with cancer deaths (matching criteria: sex, age, language region of Switzerland, subperiods 1969-1981, 1982-1994, 1995-2007).

RESULTS

For PD we found lower SMRs in lung and liver cancer and higher SMRs in melanoma/skin cancer, and in cancers of breast and prostate. As for MS, the SMR in lung cancer was lower than expected, whereas SMRs in colorectal, breast and bladder cancer were higher.

CONCLUSIONS

A common pattern of associations can be observed in PD and MS, with a lower risk of lung cancer and higher risk of breast cancer than expected. Thus, PD and MS resemble other conditions with similar (schizophrenia) or reversed patterns (rheumatoid arthritis, immunosuppression after organ transplantation).