1.2.1

The survey

In the French nationwide Décennale Santé survey performed in 2002–2003, the sample comprised 16,449 households and a total of 40,867 individuals. The target population consisted of “ordinary households” in continental France and excluded communities (such as retirement homes, student residences, hostels, etc.).

The survey included three interviews (performed by specially trained staff) and a self-questionnaire, which notably included a French-language version of the LBP-specific Nordic questionnaire and the SF-36 questionnaire .

The survey data enable the individuals’ sociodemographic and economic characteristics to be cross-correlated against their healthcare status and health care consumption.

1.2.2

Study population

The study population was constituted by all the subjects in the 30 to 69 age category in the 2002–2003 Décennale Santé survey who were able to complete the self-questionnaire and had participated in the three face-to-face interviews with survey staff, i.e., a total of 17,792 people (8,403 men and 9,389 women).

We excluded 210 people unable to complete the self-questionnaire and 5,723 people who did not participate in all three face-to-face interviews.

The 30–69 age category was chosen in order to take account of the subjects’ professional status and in view of the known relationship between age and LBP, since LBP-related factors probably change beyond a certain age.

1.2.3

Variables describing LBP

In the present study, the main healthcare variable was consultation with a specified healthcare professional for LBP at some point during the 2-month survey period. We considered that a subgroup of survey subjects was more likely to seek care from healthcare professionals for LBP: those who had reported the presence of LBP in the previous 12 months in the self-questionnaire and those who had spontaneously mentioned LBP as a chronic healthcare problem during the interview. Three variables were used to characterize the severity of LBP in these subjects: its duration over the 12 months examined by the self-questionnaire, the pain site (sciatic radiation to the legs, etc.) and the SF-36 Physical Functioning score.

1.2.4

Other parameters taken into account

Of the many variables recorded during the Décennale Santé survey, we selected three categories in view of the literature in this field and our starting hypotheses (particularly with respect to potential geographic disparities):

• •
  

  sociodemographic and economic variables: gender, age, highest degree or qualification, marital status, living alone or not, income per unit of consumption, socioprofessional category (SPC), receipt of “minimum guaranteed income” state benefits (RMI), full public health insurance coverage (CMU), private health insurance cover, geographical region (ZEAT), the number of inhabitants in the locality of residence and, lastly, residency in a deprived area (ZUS) ;

  
  
• •
  

  individual variables: height, weight, body mass index (BMI, calculated as weight/[height] ² ) and the SF-36 Mental Health score ;

  
  
• •
  

  variables related to occupational exposure as noted on the self-questionnaire: carrying heavy loads during work, uncomfortable and tiring work.

1.2.5

Statistical methods

Our statistical analyses were performed separately on men and women; most analyses were restricted to the sub-group of subjects who were likely to seek care from a healthcare professional for LBP, i.e. those who had actually reported LBP.

First, univariate analyses were performed in order to study potential associations between consultation with a healthcare professional (physiotherapists and/or physicians) and the selected variables. We also sought to identify any factors related specifically to consultation with a general practitioner (GP) or with a physiotherapist (regardless of whether or not other healthcare professionals were also consulted).

In a second step, the individuals having reported LBP in the 12 months (in the self-questionnaire or spontaneously during the survey interview) were analyzed in stepwise backward multivariate logistic regression models by incorporating the significant ( p < 0.05) variables from the univariate analysis and for all three types of consultation (healthcare professionals in general, GPs and physiotherapists).

Significant variables ( p < 0.1) and age were included in the final models. This approach means that the variables included in the final models (and thus the adjustment factors) may differ for men and women. Variables significant at the p < 0.1 level but not the p < 0.05 level in the final model can be considered to be adjustment variables (like age).

In view of the results of a preliminary analysis and as briefly explained in the discussion, we decided to present results that do not take account of the ZEAT regional variable. In fact, this variable was significantly associated with the three types of consultation for both genders. However, no one region stood out clearly and the results were very difficult to interpret. Furthermore, our other findings were not affected by inclusion or non-inclusion of the ZEAT variable.

All statistical analyses were performed with SAS software, version 9.1 (SAS Institute Inc, Cary, NC, USA).

1.3.1

Frequency of lower back pain and seeking care

Of the 17,792 individuals in the survey as a whole, 44.9% stated (either in the self-questionnaire or spontaneously during the interview) that they had suffered from LBP in the previous 12 months.

A total of 808 subjects (4.54%) said that they had sought care from a healthcare professional for LBP at least once during the 2-month period (962 medical consultations and 245 consultations with a physiotherapist).

By restricting the analysis to the 8,792 individuals having stated (in the self-questionnaire or spontaneously) that they had suffered from LBP in the previous 12 months, nearly 9.19% stated that they had sought care for the condition from a healthcare professional at least once over the 2-month survey period.

1.3.2

The medical specialties of the physicians consulted for lower back pain

GP accounted for 77% of the 962 medical consultations for LBP ( Fig. 1 ). The second most consulted medical specialty for LBP was rheumatology (9%). Surgeons accounted for the third most consulted specialty for LBP, with nearly 3% of the total. Each of the other specialties corresponded to just a few consultations.

The medical specialties of the physicians consulted for lower back pain (962 consultations in all).

1.3.3

Factors associated with consultation for lower back pain

1.3.4

Consultation with a physiotherapist, severity of lower back pain and income

Fig. 2 provides a more detailed illustration of some of the results from the previous models – particularly those dealing with the impact of LBP severity and income. The figure shows the percentage of subjects consulting a physiotherapist (for subjects of both genders having reported LBP in the previous 12 months in the self-questionnaire or during the interview) as a function of income per unit of consumption and the SF-36 Physical Functioning score (with a higher score corresponding to less functional problems).

Percentage of subjects consulting with a physiotherapist as a function of income per unit of consumption and the SF-36 Physical Functioning score.

The lower the physical capacity score, the more the subjects consulted with a physiotherapist (regardless of the level of income). We also observed that the level of income appeared to be a determinant of consultation, regardless of the Physical Functioning score.

1.4.1

Study limitations

One important limitation of this study is that it did not enable us to document the links between the various consultations with healthcare professionals. A 2-month survey of the general population is not appropriate for describing patient pathways between professionals who differ in terms of their access conditions. In particular, the survey did not enable us to distinguish between direct consultations on one hand and referrals by another health professional on the other.

Comparing France with other countries poses specific problems, even when considering the definition of the medical specialties consulted; in some countries, consultations with a rheumatologist for low back pain are less frequent, and other healthcare professionals (such as chiropractors and massage therapists) play a significant role in LBP care.

The fact that the reference period differed according to whether we considered consultations with healthcare professional (2 months) or the stated presence of LBP (12 months) complicates the interpretation of our present results. Here, we decided to compare people having consulted a healthcare professional in the previous 2 months with people who had not consulted but had reported having suffered from LBP at the time of the survey or in the previous 12 months. An alternative would have been to compare people having consulted with the remainder of the whole survey sample and without taking into account self-reported LBP. However, the latter approach would have prevented us from distinguishing between factors related specifically to consultation and those related to the presence of LBP.

When the regional variable (ZEAT) was included in our analyses, the results were difficult to interpret; there was no apparent link with the density of medical professionals and neighbouring regions were not alike in terms of consultation patterns. Moreover, these regional “effects” differed for men and women. The absence of pertinent results might also be due to the arbitrary nature of the official ZEAT regions; for example, the so-called “Paris Region” includes the Burgundy, Centre, Champagne-Ardenne, Lower Normandy, Upper Normandy and Picardy regions.

Given this context and the lack of a pertinent regional variable, we did not investigate regional differences further.

In terms of seeking care, we considered only consultation with a physiotherapist or a physician (all specialties) and not medication intake. It is probable that patients suffering from acute LBP (even for a short duration) take pain relief medications. Hence, the use of over-the-counter (OTC) medication may equate to seeking care. The next issue would be to find out which factors are specifically related to use of OTC medications.

1.4.2

The main results

Some of the above-mentioned results were expected; it is not very surprising to see that subjects with severe LBP are more likely to consult a healthcare professional.

In the present study, the duration of LBP was a significant determinant of the decision to seek care (for both men and women and regardless of the type of healthcare professional consulted). The presence of sciatica was also a significant determinant (except for consultations with a physiotherapist).

Several studies in other countries have already confirmed the impact of the duration of LBP, the presence of sciatica and the subject’s physical capacities on the decision to seek care from a healthcare professional.

In the study by Szpalski et al., seeking care was related to the duration of the painful symptoms and the patient’s belief that his/her backache would be a long-term problem . The correlation between the duration of LBP and seeking care was also highlighted by Carey et al.’s study in South Carolina .

The link between sciatica and seeking care found in the present study agrees with Carey et al.’s findings .

Somewhat paradoxically, some LBP situations make the subject less likely to seek care; in particular, the presence of sciatic pain radiating below the knee is associated with less frequent consultation than sciatic pain that does not radiate below the knee. It is possible that certain long-term, chronic situations are poorly medicalized.

Functional impairment has appeared to be a significant factor in most of the studies to have addressed this aspect . This is hardly surprising because one would expect individuals to consult more readily when they feel very handicapped from a physical standpoint.

Our most unexpected findings concern the factors associated with consultation with a physiotherapist.

We do not have an obvious explanation for the less frequent consultation with a physiotherapist in large towns and the Paris urban area by male subjects; the greater diversity of health services in larger towns than in the countryside is one hypothesis.

For both men and women, we found that two societal markers were associated with consultation with a physiotherapist for LBP: income per unit of consumption and the SPC.

One hypothesis was that people were more likely to consult a physiotherapist if their LBP causes problems at work. In this respect, workers should consult a physiotherapist more frequently. In fact, managers and the learned professions were most likely to consult.

The fact that high-income individuals consulted a physiotherapist more frequently suggests that the cost of care limits recourse to this type of treatment.

The fact that the consultation with a physiotherapist increases with income per unit of consumption has already been mentioned in a 1998 French study covering health problems in general . In terms of private health insurance, the latter study showed that workers tended to have individual contracts, which provide less coverage and lower reimbursements that the company health insurance packages held by senior managers.

Our results raise a number of questions. Is the more frequent consultation with a physiotherapist by subjects in the most affluent households related to the patients themselves (who might consult more readily on their own initiative) or their referring physicians? Do the up-front honoraria requested by physiotherapists dissuade less affluent subjects? Lastly, is consultation with a physiotherapist related to the type of private health insurance contract held (in addition to the presence or absence of private health coverage per se).

2.2.1

L’enquête

L’échantillon national de l’EDS 2002–2003 comporte 16 449 ménages, soit 40 867 individus. La population cible est l’ensemble des « ménages ordinaires » résidant en France métropolitaine, ce qui exclut par conséquent les ménages collectifs (maisons de retraite, foyers de travailleurs…).

L’enquête comporte trois entretiens réalisés par des enquêteurs spécialisés et un questionnaire autoadministré incluant entre autres un questionnaire spécifique aux lombalgies (version française du questionnaire nordique) et le questionnaire SF36 .

Les données recueillies permettent de croiser de nombreuses caractéristiques sociodémographiques et économiques des individus avec leur état de santé et leur consommation de soins.

2.2.2

Population d’étude

La population retenue pour cette étude est constituée de toutes les personnes interrogées dans l’EDS, âgées de 30 à 69 ans, aptes à répondre à l’autoquestionnaire et ayant participé aux trois visites de l’enquêteur, soit 17 792 personnes (8403 hommes et 9389 femmes).

Ont donc été exclues 210 personnes non aptes à répondre à l’autoquestionnaire et 5723 personnes qui n’ont pas participé aux trois visites de l’enquêteur.

La classe d’âge de 30–69 ans a été choisie afin de pouvoir prendre en compte les situations professionnelles des personnes et aussi le fait de la relation entre lombalgie et âge. En effet, à partir d’un certain âge, les facteurs liés à la lombalgie sont probablement différents.

2.2.3

Variables décrivant les lombalgies

Dans cette étude, la variable de santé principale était le recours à un professionnel de santé pour lombalgie dans les deux mois de l’enquête, avec une précision sur le type de professionnel. Nous avons considéré qu’un sous-ensemble des personnes de l’enquête était susceptible de recourir à un professionnel de santé pour lombalgie, ceux qui avaient signalé la présence d’une lombalgie (dans les 12 mois précédents) dans l’autoquestionnaire, et ceux qui avaient déclaré spontanément à l’enquêteur une lombalgie parmi les problèmes de santé chroniques. Parmi ces personnes, et pour caractériser la sévérité de la lombalgie, trois variables ont été utilisées : la durée de celle-ci sur 12 mois déclarée dans l’autoquestionnaire, la localisation de la douleur (présence d’une irradiation à la jambe…) et le score au SF36 Physical Functioning (capacité physique).

2.2.4

Autres variables prises en compte

Parmi les nombreuses variables disponibles dans l’enquête, nous avons sélectionné trois catégories de variables, retenues selon la littérature existante sur le sujet, et en fonction d’hypothèses à tester, en particulier sur d’éventuelles disparités géographiques :

• •
  

  les variables sociodémographiques et économiques : sexe, âge, diplôme, statut matrimonial, vie en couple, revenu par unité de consommation, occupation actuelle, profession et catégorie socioprofessionnelle (PCS), revenu minimum d’insertion (RMI), couverture maladie universelle (CMU), couverture complémentaire, zone d’étude et d’aménagement du territoire (Zéat), taille de la commune, classement en zone urbaine sensible (ZUS) ;

  
  
• •
  

  les variables individuelles : indice de masse corporel (IMC) calculé comme le rapport poids/(taille) ² , taille, score au SF36 Mental Health (évaluant la santé psychique) ;

  
  
• •
  

  les variables d’expositions professionnelles, issues de l’autoquestionnaire : exposition au port de charges lourdes durant le travail, postures pénibles et fatigantes durant le travail.

2.2.5

Méthodes statistiques

Les analyses ont été menées séparément chez les hommes et les femmes, en se restreignant pour la presque-totalité des analyses au sous-échantillon des sujets qui étaient susceptibles de recourir à un professionnel de santé pour lombalgie, c’est-à-dire ceux qui avaient signalé une lombalgie.

Dans un premier temps, les analyses ont été réalisées de façon univariée. Les associations entre le recours à un professionnel de santé (kinésithérapeutes et/ou médecins) et les variables sélectionnées ont été étudiées. Les facteurs associés au recours au médecin généraliste (exclusivement ou non) et les facteurs associés au recours au kinésithérapeute ont également été recherchés.

Dans un second temps, parmi les individus ayant déclaré une lombalgie dans les 12 mois (dans l’autoquestionnaire ou spontanément à l’enquêteur), des modèles de régression logistique pas à pas descendants ont été utilisés à partir des variables significatives à un seuil de 20 % en mode univarié, cela pour les trois types de recours (professionnel de santé, médecin généraliste, kinésithérapeute).

Dans chacun de ces modèles, les variables significatives au seuil de 10 %, ainsi que l’âge, ont été conservés dans le modèle final. La démarche d’analyse retenue peut amener à ce que les variables retenues dans les modèles finaux soient différentes pour les hommes et les femmes, donc que les variables d’ajustement ne soient pas les mêmes. De façon générale, les variables significatives au seuil de 10 %, mais non au seuil de 5 %, dans les modèles finaux, sont considérées avant tout comme des variables d’ajustement, au même titre que l’âge.

Au vu des résultats d’analyses préliminaires (qui seront brièvement évoqués dans la discussion), nous avons décidé de présenter ici des résultats ne prenant pas en compte la variable régionale Zéat. En effet, chez les hommes comme chez les femmes, cette variable intervenait significativement dans les trois types de recours étudiés. Cependant, aucune région ne se distinguait nettement, et les résultats étaient très difficilement interprétables. Par ailleurs, inclure ou non la variable n’avait pas de conséquence pour les autres résultats.

Les analyses statistiques ont été réalisées avec le logiciel SAS version 9.1.

2.3.1

Fréquence du recours aux soins

Parmi les 17 792 individus de l’étude, 44,9 % déclarent avoir déjà souffert de lombalgie dans les 12 derniers mois soit à l’autoquestionnaire, soit spontanément à l’enquêteur.

Un total de 808, soit 4,54 %, déclarent avoir eu recours au moins une fois à un professionnel de santé pour lombalgie sur la période de deux mois, ce qui correspond à 962 consultations médicales et 245 recours à un kinésithérapeute.

En se restreignant aux individus ( n = 8792) ayant déclaré une lombalgie dans les 12 derniers mois dans l’autoquestionnaire ou spontanément, près de 9,19 % déclarent avoir eu au moins un recours à un professionnel de santé durant la période de l’enquête, soit deux mois.

2.3.2

Spécialités médicales consultées pour lombalgie

Le médecin généraliste représente 77 % des 962 consultations médicales pour lombalgie ( Fig. 1 ). La seconde spécialité médicale la plus consultée pour lombalgie est la rhumatologie (9 %). Les autres spécialités correspondent à peu de consultations, mis à part les chirurgiens qui représentent la troisième spécialité la plus consultée pour lombalgie avec près de 3 % de recours.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Physical and rehabilitation medicine in Europe, from the White Book to the eBooks A study of isokinetic trunk and knee muscle strength in patients with chronic sciatica Non-steroidal anti-inflammatory drugs for athletes: An update Assessment of compliance with prescribed activity by hemiplegic stroke patients after an exercise programme and physical activity education Influence of electrical stimulation frequency on skeletal muscle force and fatigue Analyse de livre

Stay updated, free articles. Join our Telegram channel

Join