Applying multivariate clustering techniques to health data: the 4 types of healthcare utilization in the Paris metropolitan area

Using unsupervised clustering approaches to identify common mental health profiles and associated mental health-care service-use patterns in Ontario, Canada

Mental health is a complex, multidimensional concept that goes beyond clinical diagnoses, including psychological distress, life stress, and well-being. In this study, we aimed to use unsupervised clustering approaches to identify multidimensional mental health profiles that exist in the population, and their associated service-use patterns. The data source was the 2012 Canadian Community Health Survey-Mental Health, linked to administrative health-care data; all Ontario, Canada, adult respondents were included. We used a partitioning around medoids clustering algorithm with Gower's proximity to identify groups with distinct combinations of mental health indicators and described them according to their sociodemographic and service-use characteristics. We identified 4 groups with distinct mental health profiles, including 1 group that met the clinical threshold for a depressive diagnosis, with the remaining 3 groups expressing differences in positive mental health, life stress, and self-rated mental health. The 4 groups had different age, employment, and income profiles and exhibited differential access to mental health-care services. This study represents the first step in identifying complex profiles of mental health at the population level in Ontario. Further research is required to better understand the potential causes and consequences of belonging to each of the mental health profiles identified. This article is part of a Special Collection on Mental Health.

Augmenting K-Means Clustering With Qualitative Data to Discover the Engagement Patterns of Older Adults With Multimorbidity When Using Digital Health Technologies: Proof-of-Concept Trial

BACKGROUND

Multiple chronic conditions (multimorbidity) are becoming more prevalent among aging populations. Digital health technologies have the potential to assist in the self-management of multimorbidity, improving the awareness and monitoring of health and well-being, supporting a better understanding of the disease, and encouraging behavior change.

OBJECTIVE

The aim of this study was to analyze how 60 older adults (mean age 74, SD 6.4; range 65-92 years) with multimorbidity engaged with digital symptom and well-being monitoring when using a digital health platform over a period of approximately 12 months.

METHODS

Principal component analysis and clustering analysis were used to group participants based on their levels of engagement, and the data analysis focused on characteristics (eg, age, sex, and chronic health conditions), engagement outcomes, and symptom outcomes of the different clusters that were discovered.

RESULTS

Three clusters were identified: the typical user group, the least engaged user group, and the highly engaged user group. Our findings show that age, sex, and the types of chronic health conditions do not influence engagement. The 3 primary factors influencing engagement were whether the same device was used to submit different health and well-being parameters, the number of manual operations required to take a reading, and the daily routine of the participants. The findings also indicate that higher levels of engagement may improve the participants' outcomes (eg, reduce symptom exacerbation and increase physical activity).

CONCLUSIONS

The findings indicate potential factors that influence older adult engagement with digital health technologies for home-based multimorbidity self-management. The least engaged user groups showed decreased health and well-being outcomes related to multimorbidity self-management. Addressing the factors highlighted in this study in the design and implementation of home-based digital health technologies may improve symptom management and physical activity outcomes for older adults self-managing multimorbidity.

Health Service Use Among Migrants in the German National Cohort-The Role of Birth Region and Language Skills

Objective: To compare health service use (HSU) between migrants and non-migrants in Germany. Methods: Using data from the population-based German National Cohort (NAKO), we compared the HSU of general practitioners, medical specialists, and psychologists/psychiatrists between six migrant groups of different origins with the utilization of non-migrants. A latent profile analysis (LPA) with a subsequent multinomial regression analysis was conducted to characterize the HSU of different groups. Additionally, separate regression models were calculated. Both analyses aimed to estimate the direct effect of migration background on HSU. Results: In the LPA, the migrant groups showed no relevant differences compared to non-migrants regarding HSU. In separate analyses, general practitioners and medical specialists were used comparably to slightly more often by first-generation migrants from Eastern Europe, Turkey, and resettlers. In contrast, the use of psychologists/psychiatrists was substantially lower among those groups. Second-generation migrants and migrants from Western countries showed no differences in their HSU compared to non-migrants. Conclusion: We observed a low mental HSU among specific migrant groups in Germany. This indicates the existence of barriers among those groups that need to be addressed.

Construction and Validation of an Individual Deprivation Index: a Study Based on a Representative Cohort of the Paris Metropolitan Area

The association between health status and deprivation is well established. However, it is difficult to measure deprivation at an individual level and already-existing indices in France are not validated or do not meet the needs of health practitioners. The aim of this work was to establish a validated, easy-to-use, multidimensional, relevant index that was representative of the population in the Paris metropolitan area. From the SIRS 2010 cohort study, 14 socio-economic characteristics were selected: health insurance, educational background, socio-professional category, professional status, feelings of loneliness, emotional situation, household type, income, perceived financial situation, social support (support in daily life, financial and emotional), housing situation, and migration origin. In addition, a total of 12 health status, healthcare use, and nutrition-related variables were also selected. Content validity and internal validity of the index were explored. The 14 socio-economic indicators were associated to varying degrees with poorer health status, less use of healthcare, and poorer nutrition and were distributed across the 14 multiple-choice questions of the index. Each answer was rated from 0 to 2. The index value of 10 that isolates 20% of the most deprived individuals was used as threshold. "Being deprived," as defined with this value, was significantly associated with 9 of the 12 studied health variables. This index could be a relevant instrument in the assessment of deprivation and social inequalities of health.

Timing of Health Service Use Among Truck Drivers After a Work-Related Injury or Illness

Purposes Timely delivery of treatment and rehabilitation is generally acknowledged to support injury recovery. This study aimed to describe the timing of health service use by injured truck drivers with work-related injury and to explore the association between demographic and injury factors and the duration of health service use. Methods Retrospective cohort study of injured truck drivers with accepted workers' compensation claims in the state of Victoria, Australia. Descriptive analyses examined the percentage of injured truck drivers using health services by service type. Logistic regression model examined predictors of any service use versus no service use, and predictors of extended service use (≥ 52 weeks) versus short-term use. Results The timing of health service use by injured truck drivers with accepted workers' compensation claims varies substantially by service type. General practitioner, specialist physician, and physical therapy service use peaks within the 14 weeks after compensation claim lodgement, whilst the majority of mental health services were accessed in the persistent phase beyond 14 weeks after claim lodgement. Older age, being employed by small companies, and claiming compensation for mental health conditions were associated with greater duration of health service use. Conclusions Injured truck drivers access a wide range of health services during the recovery and return to work process. Delivery of mental health services is delayed, including for those making mental health compensation claims. Health service planning should take into account worker and employer characteristics in addition to injury type.

Are people most in need utilising health facilities in post-conflict settings? A cross-sectional study from South Kivu, eastern DR Congo

Background: The disruptive effect of protracted socio-political instability and conflict on the health systems is likely to exacerbate inequities in health service utilisation in conflict-recovering contexts.Objective: To examine whether the level of healthcare need is associated with health facility utilisation in post-conflict settings.Methods: We conducted a cross-sectional study among adults with diabetes, hypertension, mothers of infants with acute malnutrition, informal caregivers (of participants with diabetes and hypertension) and helpers of mothers of children acutely malnourished, and randomly selected neighbours in South Kivu province, eastern DR Congo. Healthcare need levels were derived from a combination, summary and categorisation of the World Health Organisation Disability Assessment Schedule 2.0. Health facility utilisation was defined as having utilised in the first resort a health post, a health centre or a hospital as opposed to self-medication, traditional herbs or prayer homes during illness in the past 30 days. We used mixed-effects Poisson regression models with robust variance to identify the factors associated with health facility utilisation.Results: Overall, 82% (n = 413) of the participants (N = 504) utilised modern health facilities. Health facility utilisation likelihood was higher by 27% [adjusted prevalence ratio (aPR): 1.27; 95% CI: 1.13-1.43; p < 0.001] and 18% (aPR: 1.18; 95% CI: 1.06-1.30; p = 0.002) among participants with middle and higher health needs, respectively, compared to those with low healthcare needs. Using the lowest health need cluster as a reference, participants in the middle healthcare need cluster tended to have a higher hospital utilisation level.Conclusion: Greater reported healthcare need was significantly associated with health facility utilisation. Primary healthcare facilities were the first resort for a vast majority of respondents. Improving the availability and quality of health service packages at the primary healthcare level is necessary to ensure the universal health coverage goal advocating quality health for all can be achieved in post-conflict settings.

Urban-Rural Differences in Long-Term Care Service Status and Needs Among Home-Based Elderly People in China

Background: Long-term care (LTC) needs for the elderly have become increasingly crucial policy concerns in rapidly aging Asia, especially in China, the most populous nation. However, very few studies have examined the cohort differences in terms of their existing and expected utilization of LTC services, above all urban–rural differences. This study aims to evaluate the differences of LTC current status and needs between urban–rural areas and age groups, and to identify influencing factors causing the different LTC needs. Methods: The data come from the Chinese Longitudinal Health Longevity Survey in 2014. A total of 7192 home-based elderly aged ≥65 years by multistage sampling were enrolled. The Andersen Model was applied to categorize the influential factors into three components including predisposing, enabling and needs. Multivariate logistic regression analysis was used to analyze the influential factors of the three levels of LTC needs. Results: A total of 6909 valid sample sizes were included in this study. The overall LTC needs of the elderly showed a rapidly increasing trend among which older people had the highest needs for bathing (27.29%) and toileting (15.8%). It was also demonstrated the aged cohort between urban and rural exerted an impact on all aspects of LTC status and needs to varying degrees (p < 0.05). Compared with urban areas, the LTC needs for the elderly in rural areas was more vigorous, but the supply was seriously inadequate. The elderly who were older, living in rural areas, unmarried, non-farming, with low income, in poor health and having less autonomy had higher anticipated needs for LTC services (OR > 1, p < 0.01). Compared with the young-old in rural areas, the young-old in urban areas were prone to live alone (OR = 1.61, p < 0.01). The elderly who were older, living in rural areas, farming, with low income, lonely and depressed had higher anticipated needs for community-based services (1 < OR < 1.69, p < 0.05). Conclusions: The aged cohort in urban–rural distinction were facing an increasing need for immediate care due to the inadequate support being provided, especially among rural elderly. The oldest elderly in rural areas had higher LTC needs, and different levels of needs were affected by age, economic level, family support, health status and other related effects. This study provides evidence-based recommendation for further improving the construction and development of the LTC system in China.

Patterns of health service use following work-related injury and illness in Australian truck drivers: A latent class analysis

OBJECTIVES

To identify patterns of health service use (HSU) in truck drivers with work-related injury or illness and to identify demographic and work-related factors associated with patterns of care.

METHOD

All accepted workers' compensation claims from truck drivers lodged between 2004 and 2013 in Victoria were included. Episodes of HSU were categorised according to practitioner type. Latent class analysis was used to identify the distinct profiles of users with different patterns of HSU. Multinomial logistic regression was used to examine the associations between latent class and predictors.

RESULTS

Four profiles of HSU were identified: (a) Low Service Users (55% of the sample) were more likely to be younger, have an injury that did not result in time off work and have conditions other than a musculoskeletal injury; (b) High Service Users (10%) tended to be those aged between 45 and 64 years, living in major cities with musculoskeletal conditions that resulted in time off work; (c) Physical Therapy Users (25%) were more likely to be aged between 45 and 64 years, live in major cities and have nontraumatic injuries that resulted in time off work; and (d) GP/Mental Health Users (10%) were more likely to be over 24 years of age, from the lowest socioeconomic band, be employed by smaller organizations and be claiming benefits for a mental health condition.

CONCLUSIONS

This study identified distinct categories of HSU among truck drivers following work-related injury. The results can be used to prioritize occupational health and safety promotion to maintain a healthy truck driver work force.

A fuzzy approach for analysing equitable and sustainable well-being in Italian regions

OBJECTIVES

Recently, the Italian Institute of Statistics (ISTAT) and the National Council for Economy and Labor (CNEL) have proposed a measure for the equitable and sustainable well-being called the BES ("Benessere Equo e Sostenibile"). This paper aims to propose an original application of the fuzzy k-means approach to providing an analysis of the Italian regions according to their BES.

METHODS

The fuzzy k-means algorithm was used for clustering the Italian regions according to BES data 2015. Afterwards, a principal component analysis was conducted to show and interpret the results.

RESULTS

There is a clear difference between the regions of the North and the South. The only exceptions are represented by Lazio and Abruzzo, which belong to both groups with almost equal degrees of truth. Moreover, Trentino-Alto Adige and Valle d'Aosta exhibit the best condition, whilst Molise is the worst region.

CONCLUSIONS

This study reveals that some Italian regions are in a state of backwardness regarding health, environment, minimum economic conditions, subjective well-being, education, employment conditions, social relationships, and working conditions. Therefore, institutions should consider local policies to address these issues.

A new look at population health through the lenses of cognitive, functional and social disability clustering in eastern DR Congo: a community-based cross-sectional study

BACKGROUND

The importance of viewing health from a broader perspective than the mere presence or absence of disease is critical at primary healthcare level. However, there is scanty evidence-based stratification of population health using other criteria than morbidity-related indicators in developing countries. We propose a novel stratification of population health based on cognitive, functional and social disability and its covariates at primary healthcare level in DR Congo.

METHOD

We conducted a community-based cross-sectional study in adults with diabetes or hypertension, mother-infant pairs with child malnutrition, their informal caregivers and randomly selected neighbours in rural and sub-urban health zones in South-Kivu Province, DR Congo. We used the WHO Disability Assessment Schedule 2.0 (WHODAS) to measure functional, cognitive and social disability. The study outcome was health status clustering derived from a principal component analysis with hierarchical clustering around the WHODAS domains scores. We calculated adjusted odds ratios (AOR) using mixed-effects ordinal logistic regression.

RESULTS

Of the 1609 respondents, 1266 had WHODAS data and an average age of 48.3 (SD: 18.7) years. Three hierarchical clusters were identified: 9.2% of the respondents were in cluster 3 of high dependency, 21.1% in cluster 2 of moderate dependency and 69.7% in cluster 1 of minor dependency. Associated factors with higher disability clustering were being a patient compared to being a neighbour (AOR: 3.44; 95% CI: 1.93-6.15), residency in rural Walungu health zone compared to semi-urban Bagira health zone (4.67; 2.07-10.58), female (2.1; 1.25-2.94), older (1.05; 1.04-1.07), poorest (2.60; 1.22-5.56), having had an acute illness 30 days prior to the interview (2.11; 1.24-3.58), and presenting with either diabetes or hypertension (2.73; 1.64-4.53) or both (6.37; 2.67-15.17). Factors associated with lower disability clustering were being informally employed (0.36; 0.17-0.78) or a petty trader/farmer (0.44; 0.22-0.85).

CONCLUSION

Health clustering derived from WHODAS domains has the potential to suitably classify individuals based on the level of health needs and dependency. It may be a powerful lever for targeting appropriate healthcare service provision and setting priorities based on vulnerability rather than solely presence of disease.

Who self-medicates? Results from structural equation modeling in the Greater Paris area, France

OBJECTIVES

Our study aimed to describe the prevalence of self-medication among the Paris adult population and to identify the factors associated with self-medication.

MATERIALS AND METHODS

This cross-sectional study was based on data collected from the SIRS cohort (a French acronym for "Health, inequalities and social ruptures") in 2005 in the Paris metropolitan area using a face-to-face administration questionnaire among a representative sample of 3,023 French-speaking adults. Structural equation models were used to investigate the factors associated with self-medication in the overall population and according to income.

RESULTS

The prevalence of self-medication in the past four weeks was 53.5% in the Paris metropolitan area. Seven factors were directly associated with self-medication in the structural equation model. Self-medication was found more common among women, young people, in active employment or student, with a high income, but also among people with a health information seeking behavior, with a high daily mobility, and/or with a history of unmet healthcare needs due to economic reasons. When looking at these coefficients according to income, the association between self-medication and daily mobility appeared stronger in the bottom quartile of income whereas it was no longer significant in the rest of the survey population.

CONCLUSION

Self-medication is a frequent practice in the Paris metropolitan area. This study confirms the role of some factors found to be associated with self-medication in the literature such as age or gender and draws attention to other factors rarely explored such as daily mobility, especially among people with a low income, or health information seeking behavior.

Multiple brief interventions in police custody: The MuBIC randomized controlled study for primary prevention in police custody. Protocol and preliminary results of a feasibility study in the Paris metropolitan area, France

BACKGROUND

The 15- to 35-year-old population has little contact with the health care system and is exposed to risk factors. Several studies demonstrated the feasibility of brief interventions (BIs) in different settings, e.g., in addiction medicine during police custody, where arrestees are entitled to a medical examination. Approximately 700,000 individuals are detained in police custody in France annually, and custody is an opportunity for young people to be medically examined. The characteristics of the detainees and previous experience with BIs suggest that custody is an opportunity to contribute to primary prevention. We propose to investigate the feasibility of such a contribution.

OBJECTIVES

The aim of this article is to present a study protocol and some preliminary results. The primary research objective is to assess the feasibility of performing brief interventions without a specific topic in police custody settings in arrestees aged 15-35 years. The secondary research objectives include i) testing four strategies for engaging in BIs that maximize the chances of success of the BI; ii) identifying the determinants that can orient the practitioner's choice to use a specific strategy over another one; and iii) analysing the differences between individuals who engage in BIs and those who do not and, in those who do engage, the determinants of success of the intervention.

METHODS

A two-step randomized and prospective study: i) randomization of eligible patients into 4 groups of 500 patients each; analysis of the response rates for each strategy; performance of the BI; and analyses of the associated factors and ii) a real-life, full-scale phase study evaluating the effectiveness of BIs performance of the BI; and analyses of the interventions. Analyses of the determinants of a positive response to BI, of success and of the topic of intervention will be conducted.

EXPECTED RESULTS

The rates of BI performed, rates of success, and characteristics associated with response and with success are the main expected results. Additionally, the development and assessment of filter questions and an improved BI dedicated to primary prevention for police custody settings will be attained.

Healthcare-seeking behaviour in case of influenza-like illness in the French general population and factors associated with a GP consultation: an observational prospective study

Background

GP consultation rates for influenza-like illness (ILI) are poorly known in France and there is a paucity of literature on this topic. In the few articles that have been published, the results are heterogeneous.

Aim

The aim of the present study was to estimate the proportion of ILI inducing a GP consultation, and to assess its determinants.

Design & setting

Participants of a French web-based cohort study who reported ≥1 ILI episode between 2012 and 2015 were included. Sociodemographic characteristics, access to health care, and health status variables were collected.

Method

Healthcare-seeking behaviour was analysed and factors associated with a GP consultation identified using a conditional logistic regression.

Results

Of the 6023 ILI episodes reported, 1961 (32.6%) led to a GP consultation, with no difference between those at risk of influenza complications and those not (P = 0.42). A GP consultation was more frequent for individuals living in a rural area (odds ratio [OR] = 1.21, 95% confidence interval [CI] = 1.02 to 1.43); those with a lower educational level (OR = 1.43, 95% CI = 1.18 to 1.74); those using the internet to find information about influenza (OR = 1.63, 95% CI = 1.30 to 2.03); patients presenting with worrying symptoms (fever, cough, dyspnoea, sputum, or asthenia); patients having a negative perception of their own health status (OR = 1.51, 95% CI = 1.07 to 2.13; and those having declared a personal doctor (OR = 2.86, 95% CI = 1.72 to 4.76). A GP consultation was less frequent for individuals using alternative medicine (OR = 0.68, 95% CI = 0.58 to 0.78).

Conclusion

This study allows the identification of specific factors associated with GP consultation for an ILI episode. These findings may help to coordinate health information campaigns and to raise awareness, especially among individuals at risk of influenza complications.

The effect of health status and living arrangements on long term care models among older Chinese: A cross-sectional study

Background

Currently, there are many studies focusing on the influencing factors of the elderly people’s living arrangements or health status, but little is known about the relationship between living arrangements or health status and long-term care models for the old-age, especially the joint effects.

Objective

We aimed to assess the effects of health status and living arrangements on long-term care models (LTCM) among the elderly of Xiamen, China, especially their cumulative joint effects.

Methods

A total of 14,373 participants aged ≥ 60 years by multistage sampling in Xiamen of China were enrolled. Multinomial logistic regression was used to estimate the Odds ratios (ORs) regressing LTCM on health status and living arrangements using the Anderson model as theoretical framework.

Results

Totally, 14,292 valid questionnaires were obtained, of which 86.37% selected home care. With the increase of disability degree, older people are more likely to choose institutional care, compared to living alone (ORs = 1.75, 2.06, 4.00, 4.01 for the “relatively independent’, “mild disability’, “moderate disability’, and “total disability’, respectively, in comparison with “completely independent’). The elderly living with children and other family members preferred to choose home care. (ORs = 0.50, 0.39, 0.40, and 0.43 for the “living with children’, “living with spouse’, “living with children and spouse’, and “living with others’, respectively, in comparison with “alone’). Additionally, residence, number of children, education level, and feelings of loneliness were the determinants of the choice of social pension.

Conclusion

A multitude of older people are trended to choose home care in Xiamen of China. There was an interaction and joint effect between the degree of disability and the living arrangements on LTCM. Therefore, policymakers should pay close attention to care for those living alone, childless, and disabled elders to meet their care needs, especially in home care. In addition, the social construction of facilities for elders in rural areas should be strengthened.