Article Text

Rare diseases in disabled children: an epidemiological survey
  1. P Guillem1,
  2. C Cans2,
  3. E Robert-Gnansia3,
  4. S Aymé4,
  5. P S Jouk5
  1. 1
    Register for Disabled Children and the Isère county Perinatal Survey, Grenoble, France
  2. 2
    Department of Medical Information and Data Processing, Grenoble University Hospital, France
  3. 3
    Register for Congenital Malformations from Centre and East of France, Lyon, France
  4. 4
    INSERM/Orphanet, Paris, France
  5. 5
    Department of Genetics and Procreation, Grenoble University Hospital, France
  1. Professor P S Jouk, Département de Génétique et Procréation, Centre Hospitalier Universitaire, BP 217, 38 043 Grenoble Cedex 09, France; PSJouk{at}


Aim: To estimate the contribution of rare diseases (RD) to severe impairment in 7-year-old children.

Methods: Data from a morbidity register of childhood impairments in a single French region were used. Impairments were classified as a mental, sensorial, neuromuscular (skeletal or movement-related) impairment (MSN_I) according to the International Classification of Functioning. Details of children born from 1980 to 1994 and resident in the county under study when they were 7 years old were recorded. A rare disease was defined as a prevalence rate of <1 per 2000 general population.

Results: 26% of children with severe MSN_I had a rare disease; in 36% the MSN_I was of unknown origin. The proportion of impairments that were due to a rare disease varied according to the type of impairment: 3.3% for severe psychiatric disorders; 16.0% for intellectual impairment; 37.2% for hearing impairment; 41.2% for neuromuscular, skeletal and movement impairment; and 81.1% for visual impairment. The overall prevalence rate of rare diseases was 2.1 per 1000 (459/218 283), and it increased significantly over time (p = 0.003). The latter increase was not associated with a decrease in the proportion of impairments of unknown origin, indicating an improvement in the survival of the children with a rare disease.

Conclusions: In this study, a rare disease was at the origin of 26% of cases of severe MSN_I. This proportion remained stable over time, whereas the prevalence rate, as well as the prevalence rate of MSN_I disability, increased over time.

Statistics from

Since 1983, French public health authorities have focused on rare diseases, which have also been under the spotlight in the USA with the adoption of the Orphan Drug Act1 and the establishment of an Office of Rare Diseases at the National Institutes of Health.2 In Europe, similar regulations were adopted in December 1999, and rare diseases are listed as a priority in both research and public health programmes. Two European countries have adopted a plan of action in the field of rare diseases: France ( and Italy (

A rare disease is defined as one that occurs at a rate of less than one case per 2000 inhabitants. To date, 7000 rare diseases have been identified, and new ones are regularly added to the list.3 Overall they affect a significant proportion of the population, and the yearly cumulative incidence of rare diseases is estimated at 1.3–2.2% of the general population for the 199 most common ones.4

These diseases appear either during childhood or later in life; they are often chronic and disabling, with different types of impairment and different severity levels. Most (∼80%) are of genetic origin (chromosomal, monogenic or complex genetic disorders),4 but some are of infectious origin (eg, maternofetal measles) or autoimmune (eg, juvenile chronic arthritis). Congenital malformations are genetically heterogeneous. Some may be monogenic, but they are often sporadic cases, and the lack of a familial history makes it impossible to identify a monogenic aetiology. Others are complex genetic disorders considered to be of oligogenic, polygenic or multifactorial determinism. At least some are of epigenetic aetiology.

The rate of each rare disease is difficult to estimate because determination may be incomplete, mainly because of difficulties with diagnosis. Problems linked to lumping or splitting have already been described,5 and lumping/splitting habits are highly dependent on the level of knowledge about the aetiology of the disease. Therefore they vary over time. However, recent progress in designing diagnostic tools has increased the reliability of recognising rare diseases, making epidemiological studies possible in the evaluation of the impact of rare diseases in terms of public health. Childhood disabilities affect 1% of children in primary school, and half of all cases have a prenatal cause.6 The aim of this study was to estimate the proportion of childhood disabilities that are due to rare diseases, using data from a local morbidity register.


Data were extracted from the Registre des Handicaps de l’Enfant et Observatoire Périnatal (RHEOP) (Register for Disabled Children and Perinatal Survey), which is based in Isere county. This county has one million inhabitants, and ∼14 000 births per year are registered. Data are collected as soon as the diagnosis is made, and then updated when the child is 7 years old. Of the three aspects of handicap defined by the International Classification of Functioning, impairment was used as the inclusion criterion.7 Impairment is defined as (a) the absence of, or significant inability to fulfil, an organic function or (b) the loss of an anatomical structure. The registered impairments were classified according to the International Classification of Functioning chapters, mental, sensorial, neurological (skeletal and movement-related) impairments (MSN_I). The group of mental impairments includes intellectual impairment and psychiatric disorders. Severity criteria for inclusion of children in the register were as follows6:

  1. Presence of neuromuscular, skeletal and movement-related impairments that need specific aids and/or continuous rehabilitation; includes all children with cerebral palsy

  2. Severe psychiatric disorders coded F84.0, F84.1 and F84.5 for autism, F84.2 for Rett syndrome, and F84.8 or F84.9 for psychosis according to the International Classification of Diseases (10th revision) (ICD-10).

  3. Moderate, severe or profound intellectual impairment (IQ<50) and Down syndrome (any IQ)

  4. Bilateral hearing loss >70 db before correction or visual acuity of the best eye <3/10 after correction

The classification used to code the data was ICD (9th revision) until the 1990 birth cohort, then ICD-10 was used. Information on the aetiology of the main impairment is available in the text. We referred to the Orphanet8 database to confirm whether the disease responsible for the disability was a rare disease, and complementary information was extracted from OMIM.9 We describe rare diseases in the usual 10 groups: neurological disease (with an identified syndrome or with a structural anomaly of the central nervous system), syndromes with dysmorphy or malformation and intellectual impairment, congenital or hereditary severe vision loss, skeletal diseases (generalised or focalised), congenital or hereditary severe deafness, neuromuscular disease, hereditary metabolic disease, tumour, embryofetopathy and other rare diseases (see online supplementary material). Also the children with a unique severe MSN_I were described according to the type of impairment: neuromuscular, skeletal and movement-related impairment, intellectual impairment, hearing impairment, visual impairment and severe psychiatric disorders. Children with more than one severe MSN_I were combined as a separate group whatever the type of impairment. Other characteristics of the children, such as ability to walk and epilepsy, were obtained from the register. For all children in the study, characteristics were recorded at the age of 7. Data were analysed for differences between children with MSN_I and a rare disease (RD group) and children with MSN_I without an identified rare disease (WIRD).

The reported prevalence rate is the number of affected 7-year-old children living in Isère county and born between 1980 to 1994 divided by the total number of children born in the same period and living in the same area at the same age (218 283 children). This number allows us to observe, with a probability of 95%, at least one child whose rare disease associated with severe MSN_I has a frequency of 1/70 000 or more. This number was provided by the National Institute of Statistics and Economical Studies (INSEE). In the general population, 30% of children aged 0–7 years were born outside the county. The equivalent proportion of children in the RD group was 31%; the difference from the general population is not significant (p = 0.22).

Agreement was obtained from the National Committee of Registers (1996, 2000, 2006) and the National Commission of Computerization and Freedom (No 997086).

Statistical analysis of prevalence trends over three 5-year periods (1980–1984, 1985–1989, 1990–1994) was performed using the Armitage trend test.10


Children with at least one severe MSN_I

Of the children born in 1980–1994, 1739 had at least one severe MSN_I when they were 7 years old and resident in the county; 59% of them were male. One or more severe impairments was present in each child. Table 1 shows the distribution of the type of impairment according to the presence of a recognised rare disease. The mean number of severe impairments per child was 1.20.

Table 1 Severe MSN_I in 1739 children born in 1980–1994, according to the presence of a recognised rare disease

Children with a severe MSN_I and a rare disease

Of the 1739 children with severe MSN_I, 459 had an identified rare disease. They form the RD group; this proportion (26.4%) did not change during the study period (p = 0.94). Table 2 shows the distribution of children in different groups of rare diseases. The most common (24.1%) were neurological diseases, including syndromes and central nervous system structural anomalies, which are mutually exclusive. Detailed information on these groups can be found in the online supplement. The least common were embryofetopathies, found in 4% of children.

Table 2 Children with severe MSN_I and an identified rare disease (n = 459)

The proportion of children with an identified rare disease varies according to the type of impairment (fig 1): 26.6% for more than one severe MSN_I, 3.3% for severe psychiatric disorders, and up to 81.2% for severe visual impairment.

Figure 1 Proportion of children with a rare disease according to the type of MSN_I.

In the RD group, the disease was acquired in 53 out of 459 children (11.5%). For instance, movement-related impairments can be attributed to osteochondritis or astrocytoma. In the WIRD group (n = 1280), the proportion of late acquired disease or accident, after the first month of life, was significantly lower: 6.5% (p = 0.001).

A factor responsible for the MSN_I other than a rare disease was identified in 649 children. In this group, the most common factor in isolated intellectual impairment was Down syndrome (211 out of 213), and factors involved in isolated neuromuscular, skeletal and movement-related impairments were related to perinatal risk (147 out of 208).

Prevalence rate of 7-year-old children with severe MSN_I and a rare disease

Whereas the prevalence rate of children with severe MSN_I was 8.0 per 1000 children resident in the same area (95% CI 7.6 to 8.3), the overall prevalence rate in the RD group was 2.1 per 1000 children resident in that area (95% CI 2.0 to 2.4), with a significant increase in this rate from 1.7 per 1000 to 2.5 per 1000 between the first (1980–1984) and last (1990–1994) period studied (p = 0.003). Over the same period, the prevalence rate of MSN_I of other known origin and unknown origin also increased significantly, from 1.7 to 2.4 per 1000 (p = 0.0017) and from 2.4 to 3.2 per 1000 (p = 0.0027), respectively. There was no significant decrease in the prevalence rate of Down syndrome (p = 0.40). However, in our data, the proportion of severe MSN_I of unknown origin remained stable over time, 35%, 38% and 35%, respectively, for 1980–1984, 1985–1989 and 1990–1994. The same was observed for the proportion of severe MSN_I of known origin in spite of a slight non-significant decrease in the proportion of Down syndrome (13.3% to 11.1%, p = 0.54) (table 3).

Table 3 Severe MSN_I in 1739 children born 1980–1994, according to the origin of the impairment

Activity (walking) limitation in children with a rare disease and severe MSN_I

The proportion of children with other impairments was similar in the RD and WIRD group (80.8% and 80.9%, respectively), whatever the severity of the impairment. The proportion of children with epilepsy did not differ between the two groups: 16.8% and 18.9%, respectively.

Information on ability to walk is used to define activity limitation when the child is 7 years old. Such data have been available for children with severe MSN_I since the 1987 birth year cohort (n = 998). Overall, 79.2% (791/998) of the children in that cohort could walk unaided, 8.7% could walk with aids, 7.9% used a wheelchair, and 4.2% were bedridden. Children in the RD group had more limitations than children in the WIRD group. This difference was significant in children with more than one severe MSN_I: 37% could walk unaided in the RD group versus 57% in the WIRD group (p = 0.02).


At least a quarter of the children with severe MSN_I were affected by an identified rare disease. In the RD group, the most common impairment was neuromuscular (skeletal and movement related) (41%). We observed some differences in walking ability between the RD and WIRD groups, with the more severe cases in the RD group.

All the rare diseases listed in the online supplement are of low prevalence, the usual criterion of a rare disease. In most of the diseases encountered, a severe MSN_I is the common clinical feature. In some of the diseases in the list, the MSN_I is not the common feature but rather the result of complications, such as cerebral stroke in sickle-cell disease or Ehlers–Danlos syndrome. Other rare diseases such as cystic fibrosis and haemophilia are never expressed with a severe MSN_I in this age group, and thus are not found in our register.

The estimate we provide of the proportion and prevalence rate of children with severe MSN_I and a rare disease should be considered a minimum value, as some children with severe MSN_I and a rare disease may have died before the age of 7, ie, before the time of registration; we calculated prevalence rates, not incidence rates. The rate may also be underestimated because a rare disease may be present but not identified, and this is more likely when the disease is extremely rare. No aetiology could be recognised for 631 of the 1739 (36.3%) children studied, and these children may have an unrecognised rare disease. Furthermore, our results did not suggest an artefact linked to a decreasing prevalence or proportion of Down syndrome or increasing numbers of known diagnosis.

What is already known on this topic

  • Very little is known on the prevalence rates of rare diseases in childhood, even though such diseases generate great burden on the families affected.

  • Several studies have reported characteristics and prevalence rates of childhood impairments without taking rare diseases into account.

What this study adds

  • The proportion of severely disabled school-age children who have a rare disease has been determined for the first time.

  • The prevalence of rare disease leading to severe disability in children is 2.1 per 1000 and has increased over time (1980–1994).

There are several possible explanations for the increase in the prevalence rate in the RD group:

(a) Increased recognition of rare diseases. Our results for MSN_I of unknown origin do not support this hypothesis; their proportions remained stable over time, whereas their prevalence rate even increased

(b) Other reasons, such as an increase in maternal or paternal age at birth, need to be looked for in larger studies

(c) Improved survival of these children. This explanation seems the most plausible; the proportion of rare diseases remains stable over time, and the increase in the prevalence rate cannot be considered in isolation and has to be related to the overall increase observed in MSN_I of other known origin and of unknown origin.

To our knowledge, no other population database has been used to survey 7-year-old children with severe MSN_I and a rare disease. The prevalence rate of 2–3 per 1000 is an interesting first estimation. The level of confidence in our register was assessed for children born in 1985–1993; it varied between 80% and 87%, which can be considered fairly good,11 and there was no significant trend over time. Large national databases, constructed for health economy purposes (eg, DRG (diagnosis-related group, a system to classify hospital cases into one of approximately 500 groups), social insurance) are not very helpful when studying rare diseases.12 The overall prevalence rate of rare diseases is probably higher, as most become apparent later in life.

The proportion of children who could not walk was higher in the RD group, and greater severity of impairment may influence the type of care provided for these children. Godet et al13 emphasised that not only the management of rare diseases but also their diagnosis is more strongly influenced by personal initiatives than by healthcare system organisation. There is a need for research focused on organisation of care and factors that modify the attitude of families affected by a rare disease.

Although with some rare diseases most patients live as independent adults,14 children in the RD group with severe MSN_I seem to be more strongly affected than children in the WIRD group, a number of them being unable to walk unaided. It is important to ensure follow-up of these children to check their participation in adult life.


This work enables, for the first time, estimation of the prevalence of rare diseases causing MSN_I in 7-year-old children. It provides evidence that rare diseases are at the origin of 26% of severe MSN_I, with an increase in their prevalence rate over time in parallel with the overall increase in MSN_I disability.


We acknowledge the Isere County Council (CGI), the National Institute for Monitoring Sanitary Conditions (INVS), and the National Institute of Health and Medical Study (INSERM).


Supplementary materials

  • web only appendix 93/2/115

    Files in this Data Supplement:


  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.