Recent statistics show that the total cost of healthcare accounted 9.6{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} of GDP across all the EU countries, ranging from over 11{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} in France, Germany, and Sweden to the lowest ratio of 5{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} recorded in Romania. Even if health spending grew in the previous years in line with the economy in Europe, a continuous increase of such expenses could implicate a great financial burden not only on health systems, but also on social security programs [1] and, indirectly, on society in form of reduced employment and productivity [2]. Therefore, for all EU countries, irrespective of the type of healthcare system and financing arrangement, managing the increase of health services cost is a medium- and long-term strategic objective [3]. To support this approach, it is a priority to carry out specialized studies on the population health needs, the types and frequency of the demand of health services, the factors that determine the structure and dynamics of healthcare utilization, the profile of people using the healthcare services, etc. It is equally important to assess possible means of reducing healthcare expenditure not only for ensuring access to needed care, but also for strengthening the effectiveness and the resilience of health systems [1]. In this respect, important instruments to be considered, besides cost containment policies [4] and care management strategies [5], are those related to diseases prevention and health promotion [6].
As a response to the need to prevent and control diseases and to promote a healthier lifestyle, the literature emphasizes the positive influence of physical activity on the health status of the population. It is well known that regular physical activity (1) reduces the risks for non-communicable diseases, mainly cardiovascular diseases, various types of cancer, chronic respiratory diseases and diabetes [7], (2) provides protection against future depression [8], (3) reduces stress reactions and delays the effects of various forms of dementia [9], (4) prevents the obesity, given that it is a key determinant of energy expenditure [7]. Physical activity could be considered not only as a preventive measure but also as an alternative or complementary treatment for various physical or mental health conditions. For instance, some recent studies [10,11,12,13] find consistent evidences supporting that physical activity with moderate intensity is effective in alleviating or even treating the severe symptoms of depression in affected adolescents. Interventions involving physical activity are also an accessible way of reducing the symptoms of severe anxiety or mental illness among adults, including schizophrenia-spectrum disorders, major depressive disorder, and bipolar disorder [14,15,16,17,18]. The effects of physical activity as an additional or stand-alone treatment are sustained in the case of other medical conditions such as: alcohol use disorder [19,20,21,22,23]; functional outcome after stroke [24,25,26,27,28,29,30]; cardiovascular disease [31]; type 2 diabetes [32]; cancer [33]. This double role of physical activity [34] reflects its negative association with demand of health services, which could lead to lower spending on healthcare systems [3, 35,36,37].
Studies on the relationship between physical activity and healthcare utilization
Following our critical analysis of the literature on the relationship between physical activity and healthcare utilization, several observations are noteworthy to be mentioned. These remarks concern (1) the population for which the studies were performed, (2) the indicators used as measurements for healthcare utilization, (3) the methods and means of measuring physical activity, and (4) the control variables used in modelling the relationship between physical activity and healthcare utilization.
Types of population
The first observation results from the fact that most of the existing literature examines the link between physical activity and healthcare utilization just for certain segments of the population, which could depend on factors as age, gender, a particular disease, etc. A large part of such studies concentrates on older adults [36, 38,39,40,41,42,43,44,45,46], underlining that physical activity is strongly associated with lower usage of healthcare services. According to [38], reduced physical activity, such as walking activity, could be the most promising modifiable predictor of healthcare utilization as measured by the number of drugs and number of physician contacts over 12 months among older adults. The findings of [41, 43] indicate that being physically active might lead to beneficial results and a quicker recovery for hospitalized older adults. Analyzing only the category of older women, Silva [44] concludes that higher volumes of physical activity are significantly associated with lower usage of medications in women who are involved in a physical activity program. In this research direction, there are also strong evidence suggesting that the many benefits of physical activity for older adults extend beyond better health, improved physical function, reduced impairment, independent living, and increased quality of life to include significantly reduced healthcare costs and mortality [42,43,44,45,46,47]. Another range of studies reveals the role of regular physical activity interventions in lowering the usage of health resources and services and saving a substantial amount of healthcare expenditure among people with specific health conditions, such as asthma, cardiovascular disease, obstructive pulmonary disease, arthritis, and diabetes [42, 48,49,50,51,52], or those suffering from obesity problem [42, 50, 53,54,55,56]. However, it is noteworthy that the effects on healthcare utilization and costs are likely to be a result of long-time regular physical activity behaviour rather than a short-term behaviour change [56]. Of these studies, several focus on persons engaged in clinical trials fitness activity or in health program [42, 44, 45]. While their empirical evidences support that engaging in regular physical activity only involves health benefits and therefore reduced use of some health services as hospital admissions or medicine consumption, these studies have a restrictive ability to generalize to a larger population. By contrast, the literature on using representative sample from the general population is relatively limited. In this respect, a relevant, but not exhaustive enumeration of prior studies regarding the relationship between physical activity and healthcare utilization encompasses the analyses of Katzmarzyk et al. [57], Bertoldi et al. [58], Sari [59], Maresova and Vokoun [60], Rocca et al. [2], Fernandez-Navarro [61], and Kang and Xiang [37].
Healthcare services
The second observation concerns the dependent variables used in literature. Related to the measurement of healthcare utilization, the literature is not very explicit, but a classification of studies can be outlined. One stream focuses on obtaining an objective measure of different healthcare services through medical records kept by the family doctor, the generalist or specialist physicians [44, 45], while the second stream includes a subjective (self-related) health evaluation based on the respondents data obtained from questionnaires [2, 37,38,39,40, 42, 56, 59,60,61]. Within the second approach, the measures for healthcare utilization concern both service contacts [2, 39, 42, 44, 61] and volume of services [37, 38, 40, 42, 44, 45, 56, 58,59,60]. Usually, the literature presents four categories of healthcare utilization: medicine use, expressed in number of consumed and prescribed medication, inpatient (hospitalization and home health services), outpatient (use of generalist and specialist physicians’ services) and preventive services (dental check-up, flu shot, blood pressure check-up, cholesterol check-up, blood glucose test, immunological test).
According to literature, most of the studies concern the relationship between physical activity and one or a few healthcare categories. For instance, for the association between physical activity and medicine use there are findings to support both a significant and non-significant relationship. On the one hand, higher levels of physical activity are significantly associated with lower use of medication [27, 38, 44, 58, 61]. On the other hand, an insignificant link between physical activity and the number of medication consumed was found [27, 45]. The latest results could be attributed to the fact that these studies focused only on older adults, suggesting that other factors also should be engaged in discussions related to physical activity. Other findings from literature imply also that if people are more physically active, they will use significantly fewer inpatient services [42, 56, 59, 60] or outpatient services [38, 42, 56, 59, 60]. Having an opposite effect, physical activity appears to be a stronger predictor of all types of preventive services, emphasizing that active people may be more health conscious and thus may use precautionary measures more frequently compared to inactive persons [42]. In contrast to these results, there are studies that failed to find a significant association between physical activity and the number of days spent in hospital [38], the number of home consultations from a medical professional [45] or the number of physician’s visits [45]. In addition, the home healthcare services [45] appear not to be significantly explained by leisure time physical activity. In contrast, only few studies have analyzed the relationship between physical activity and multiple categories of healthcare utilization. For instance, Fisher et al. [39] have used both service contacts (services used versus services not used) and volume of general and specialist physician services, and hospital services, while Kang and Xiang [37] have added 10 measures of preventive services, outpatient visits, home visits, emergencies, and prescribed medicine. Their results are consistent with other studies mentioned above, but they allow to obtain a more in-depth analysis of the association between physical activity and different categories of healthcare utilization.
Measurements of physical activity
Another relevant remark is related to the use of different types and measurements of physical activity in relation to healthcare utilization. The physical activity is divided into four main classes, namely leisure time, household, transportation, and work. While a vast body of research focuses only on one dimension of physical activity, especially related to leisure time [2, 39, 40, 59, 61], a more narrow range of studies considers an indicator encompassing more types of physical activities [37, 56, 58, 60]. With respect to the type of physical activity, an important issue is linked to the various methods used to measure the indicator’s levels. In this matter, Dishman et al. [62], Miles [63], Sallis [64], and Sylvia et al. [65] distinguish between objective monitors (pedometers, accelerometers, heart rate monitors, armbands, and direct observations), physiological measures of energy expenditure (doubly labelled water), and self-reports (questionnaires or activity diaries). In addition, the analysis of the literature as a whole stresses the lack of studies measuring the level of physical activity by factors such as age, gender, body weight, or psychiatric and medical co-morbidities [66]. Most empirical studies evaluate and test the differences between physical activity patterns with regard to these type of factors [37, 40, 56, 61, 67,68,69,70,71,72,73,74,75] or explore their impact on the relation between physical activity and healthcare utilization [2, 39, 42, 45, 58, 60, 61, 76], but the authors do not integrate them into the indicator’s measuring level.
Other determinants of healthcare utilization
In order to gain better insight into the relationship between physical activity and healthcare utilization, most studies include a set of variables such as demographic and socioeconomic factors, health status or health behaviour. The findings adjusted for these individual characteristics reveal that involvement in physical activity still reduces the use of healthcare utilization through its relationship with chronic diseases, physical and mental health status [38, 42, 44, 56, 61], personal health practices such as smoking and drinking [44, 58], body mass index [38, 44, 58], age [2, 38, 42, 44, 56, 58], gender – with a higher effect for men [2, 38, 42, 58, 61], educational level [2, 44], economic level [2, 58], employment status [39, 60].
Beyond the use of these factors as control variables in the relationship between physical activity and healthcare utilization, there is an extensive literature on their association with the use of healthcare services [76]. It is well known that people’s health status, including inherited diseases and conditions, requires medical care. More precisely, asthma, chronic conditions, and depression are frequently related to number of physician contacts and number of drugs. In particular, prescription drugs are most strongly associated with diseases such as coronary heart disease, diabetes, hypertension, thyroid problems, osteoporosis, and heart failure [38]. Outpatient health services are more likely to be used by those who have poor to good health status, are experiencing declining health, and have chronic diseases. Meanwhile, hospitalization is more likely among those people with poor health status or having a chronic disease. However, the prevalence of these medical conditions differs by gender, age, occupational status, and other factors. The role of age is essential since, as people age, they become more susceptible to disease and disability, which implies more frequent use of various healthcare services [77]. With regard to gender, there are wide evidence that women, having higher rates of disability and self-reported fair or poor health status than men, generally use more healthcare services than their counterparts [78]. In this respect, Salganicoff et al. [79] and NCHS [80] stress that women are more likely to have primary care visits, hospitalization or emergency visit, and to receive more diagnostic services, screening services, diet and nutrition counseling than men even though men generally have higher rates of obesity and cardiovascular problems. Individual behaviours such as smoking, excessive alcohol consumption, poor diet or obesity also cause conditions that require medical attention [81]. Concerning other socioeconomic determinants of health, the literature emphasizes that higher levels of education, having economic stability, being employed, or having community safety are correlated with better health status [81].
In summary, the relatively vast body of research on the topic of this study states that interventions aimed at increasing physical activity may result in significant reductions in healthcare utilization. In addition, most of the empirical studies outlines that this potential role of physical activity is better clarify in relation to other individual characteristics. Besides identifying the determinants and assessing their association with healthcare utilization, in the end, the empirical results of such studies must be analyzed in relation to a country’s public and/or private health system and have to serve as support for other countries by sharing successes or even failures and exchanging experiences to provide inspiration for further development, refinement and implementation of effective policies.
Physical activity in France: facts and policies
For the French population, the existing literature emphasizes a lack of physical activity and consequent sedentary behaviours, as well as a continuous degradation of these indicators in the last decades [82, 83]. Analyzing data from the ENNS study 2006–2007 and Esteban study 2014–2016, Verdot et al. [83] observe a decrease in the level of physical activity among all adult women (18–74 years old), from 63.2 to 52.7{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} people that are reaching the WHO recommendations on physical activity for health, while an increase is noticeable only for men (18–74 years old), from 63.2 to 70.6{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} [63]. The same study estimates that the prevalence of physical activity account only 50{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} for boys 6–17 years old and 33{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} for girls of the same age group. These percentages have not changed significantly between 2006 and 2016. Moreover, at the level of the EU, France is the country with the second highest prevalence of insufficient activity among school-going adolescents (86.2{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} in 2011 and 87.0{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} in 2016) [82]. For the adolescents between 11 and 14 years old it is recorded a decrease of physical activity prevalence from 38.1 to 33.7{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} for boys and from 23 to 20{e4f787673fbda589a16c4acddca5ba6fa1cbf0bc0eb53f36e5f8309f6ee846cf} for girls [83].
In response to this alarming reality, France was concerned to implement several national physical activity plans that include components for increasing physical activity in different sectors such as health, education, sports, transport, and workplace. In France, the integration of physical activity into public health policy dates back to the 2000s. These policies target a wide range of the population, including the people with disabilities, those suffering from chronic diseases, the elderly, the adolescents, the migrants, and other low socioeconomic groups for which specific physical activity programs are either at low cost or completely free of charge [3]. The French National Nutrition and Health Program (PNNS – Programme National Nutrition Santé), which was launched in 2001, is a public health plan that aims to improve the health status of the population by acting on one of its major determinants: nutrition. For the PNNS, nutrition is understood as the balance between food intake and physical activity. The Health Act 2004–806 also establishes certain objectives for public health policy to reduce sedentary lifestyles and increase physical activity among the French population. Another example is the accession of French specialists and institutions to the European Network for the Promotion of Health-Enhancing Physical Activity (HEPA) in 2006, one year after its launch. It should also be noted that France has taken over in various forms the guidelines formulated by The Toronto Charter for Physical Activity which was adopted in 2010 by the Global Advocacy Council of Physical Activity, International Society for Physical Activity and Health. Last but not least, in France the idea of prescribing physical activity as a treatment according to the patient’s condition, physical ability and medical risk has been formulated several times, and the idea will be implemented through the Health Act of 2016. Another successful action, called “Medicosportsanté”, is taken by the national sports federation who provides guidance on adapting sports programs for participants with chronic diseases or for the elderly. As for promoting physical activity among children and young people, an effective national intervention based on a socio-ecological approach was implemented [3]. This intervention encourages them to engage in physical activities during and outside school hours by receiving social support from parents, teachers and sports instructors. Besides the strategies countering insufficient physical activity, other recent and equally important measures to prevent diseases and promote health at the national level refer to the campaigns on tobacco and alcohol consumption and obesity among young people, raising alcohol and tobacco taxes, assessing programs and reducing work-related risks [84].
Objective and motivation
In the EU context, all member states, including France, are involved in different projects and programs in order to promote physical activity and to evaluate its relationship with population health, and health systems. The WHO strategy for physical activity underlines as major future aims the surveillance and evaluation of policy initiatives and also the strengthening of the evidence base for physical activity and health for the EU countries [85]. Such strategy requires strengthening empirical evidence and highlighting the specificity of the relationships between physical activity, healthcare, health status, and other health risk factors in the EU context for different population groups depending on gender, age, profession or geographical area. Thereby, the implementation and the efficiency of public policies promoting physical activity and population health depend to a large extent on the health system of a country, the population structure, and a number of cultural and educational factors that can cause changes and behaviours regarding the individuals’ lifestyle and health [86].
The existing literature underlines the relevance of the association between physical activity and healthcare utilization. The increase of healthcare costs and the rising pressure on health insurance and health systems determined companies and governments to recommend physical activity as well as as complementary treatment, which in the end impacts the cost of healthcare [87]. To the best of our knowledge, in the case of French population, the research on the association between physical activity and different types of healthcare utilization is still insufficiently developed. In this regard, the outcomes of Gasparini et al. [88] and Lanhers et al. [87] should be outlined, as the authors have related the lower number of medical prescription for chronically ill patients and a lower cost of medication for type 2 diabetes in older adults to high volume of physical activities. But both studies were conducted on small and restrictive samples. Despite the generalization of their findings to the entire population, Nichèle and Yen [89] limit their study to an investigation of the role of physical activity, besides other socioeconomic characteristics and lifestyle, in the link between obesity and mental health for French adults.
Moreover, while a large body of literature provides strong evidences on the impact of physical activity and health status over healthcare utilization, only a few studies address the problem of endogeneity of these two determinants. This implies that physical activity can be itself influenced by healthcare utilization, which leads to the problem of reverse causality between the two variables. For example, as physical inactivity increases the duration of hospitalization, longer stays in hospital may also be related to the likelihood of being inactive [90]. As for the relation between healthcare utilization and health status, Bilgel and Can Karahasan [91] argue that health status is endogenous for the fact that individuals may receive healthcare and observe health status. Moreover, as Sari [59] states, it is also plausible that individuals with certain health conditions can be physically inactive and, at the same time, use more healthcare services.
In compliance with all the above underlined coordinates on the existing literature and with the EU strategy for physical activity, we aim at analyzing the association between physical activity and healthcare utilization, controlled by a set of socioeconomic and demographic factors, for a French representative sample. The contribution of this paper to the existing literature is threefold. Firstly, it provides an overall analysis of the context of healthcare utilization in relation to physical activity at the national level of France. To the best of our knowledge, no such studies have been conducted using a complex set of data provided by the European Health Interview Survey (EHIS) and the Health and Social Protection Survey (ESPS) 2014. Thus, our study provides valuable insights for policy-makers on how to improve solutions or developing programs to promote physical activity for a healthy life style in France. Secondly, following the WHO global recommendations on physical activity for health, in our paper we develop a more general measurement of physical activity that includes more components/dimensions of the indicator and also considers the age group. Hence, a more accurate classification of the population depending on the type and intensity of physical activities and age is obtained, which would be further reflected in its association with healthcare utilization. Thirdly, the methodological approach employed in the empirical analysis enables to cope with the problem of endogeneity caused by unobserved heterogeneity and possible reverse causality of healthcare utilization in relation to health status and physical activity by using instrumental variables provided by the EHIS-ESPS 2014 survey.
