Key words: Carpet weaver, Lung functions

This study took place in late 2001 in a carpet making community in Awardpura, Gwalior, India. In this community, carpet making is a female occupation: girls start work at around 8 years of age and work for 8 hours every day for 750 Rupees (£10) or less per month. The carpets are made from cotton and wool. The work place is cramped, enclosed and poorly ventilated.

This study used spirometry to look for possible impairments in lung function in teenagers. Main outcome measures were Forced Vital Capacity (FVC), Forced Expiratory Volume in one second (FEV1), FEV1/FVC ratio and Mid-Expiratory Flow rate (MEF), which is the mean forced expiratory flow during the middle half of the FVC.

The carpet making community in Awardpura is a socio-economically deprived group. Lung function is influenced by ethnicity(1) and may be reduced in malnourished children(2). In order to negate potential confounding variables, families were sought in which one or more girls aged 8-20 years worked as weavers and one or more girls aged 8-20 years did not and had never done so. Weavers were compared with their non working sisters by paired "t" test.

For any given pair, the elder sister was more likely to have started working. This introduced age as a confounding variable. Therefore, four extra non-working older teenage girls were recruited (in 2 "pairs") from the same community to replace the youngest sister controls (aged <12 years). This allowed us to make a second (age-matched) control group: average values for working and non-working girls were then compared by "t" test.

Ethical approval was obtained. Information leaflets in Hindi which described the study were read through individually with the cases and controls before the study started. After full explanation in Hindi, verbal informed consent was obtained from the children and written informed consent from their mothers. Illiteracy is a recognized problem in Indian carpet making communities(3). If mothers were unable to sign, a thumb print was obtained.

Exclusion criteria were respiratory tract infection in the last 2 weeks, history of tuberculosis, chest deformity (pectus excavatum, pectus carinatum, scoliosis), history of chest surgery and admitted cigarette smoking.

A MicroLab 3500 Spirometer was used to obtain FVC, FEV1 and MEF values. Calibration certificates were obtained immediately before and immediately after the study from the manufacturers (Micro Medical Ltd., Rochester, Kent, ME1 2AZ, UK). Disposable mouthpieces with filters were used. These cause a small underestimation of FVC and FEV1 (2-4%) for which a correction was made(4). Spirometry was performed in the sitting position. The best blow was used for analysis, based on the following criteria. Best was defined as the spirogram with the highest sum of FVC and FEV1. We took the best of 3 blows with variance <5% or best of 5 blows with variance >5%. No more than 5 blows were allowed although obvious very poor attempts were not counted. If one girl elected to stop early then her sister was stopped after the same number of attempts; if her sister had already been tested, then her sister’s blows after this point were not counted.

Height and weight was recorded to the nearest kilogram and centimeter respectively, by standard techniques. The same instructor (DG) measured heights and weights and explained and demonstrated the spirometer in a standard way in every case.

Prediction equations were used for calculating FVC, FEV1 and MEF, based on age and height in healthy North-Indian girls(5). Values obtained on spirometry were expressed as percentages of these predicted values (FVC%, FEV1% and MEF% respectively).

After work status was revealed, inquiry was made about chronic respiratory symptoms and duration of occupation (if applicable). Duration was correlated with lung function, using Spearman’s correlation coefficient.

In most families all the females worked as carpet makers. A minority of families contained one or more girls aged 8-20 years who worked as weavers and one or more girls aged 8-20 years who did not. All such families were identified. These amounted to 19 families. All were approached and invited to participate. In one family one of the girls did not wish to participate. In one family we could not analyse the results because the identity of the weaver was revealed before both girls had performed spirometry.

Seventeen carpet makers aged 12 to 20 years were compared with their sisters. Average ages were 15½ and 12 years respectively. The second (age-matched) control group had an average age of 15 years (n=13).

None of the girls had respiratory symptoms.

Most of the girls were small by Indian standards(6). Of the 17 carpet makers and their sisters, 26%, 35% and 53% fell below the 3rd, 5th and 10th centiles for weight and 53%, 59% and 68% fell below the 3rd, 5th and 10th centiles for height respectively. There was no significant difference in height or weight between the carpet makers and their sisters. Weights did not differ significantly from those in a recent larger study which included the same community(7).

In every case, the girls laughed when spirometry was demonstrated. They appeared embarrassed and unwilling to blow with maximal effort. Only 8 of 38 cases had an FVC ³80. Average FVC was 70. In every case the FEV1/FVC ratio was normal. Average FEV1/FVC ratio was 98.7%. There was no significant difference in FVC, FEV1 and FEV1/FVC ratio between the 3 groups.

Values for MEF% are shown in Table I. Mean values and standard deviations (SD) were calculated for the 3 groups. Carpet makers had a mean MEF% of 97 (SD=17). Their sisters had a mean MEF% of 116 (SD=21). The age-matched control group had a mean MEF% of 115 (SD=22). There was no difference between the latter 2 groups. The carpet makers had a significantly lower mean MEF% than their sisters (p<0.01 by paired "t" test) and the age-matched control group (P<0.02 by "t" test).

Table I__MEF% of Carpet Makers and Their Sisters

Duration 
of work
(years)

MEF% of
sister

1

132

129

2

91

137*

3

122

139*

3

106

139

3

103

128

3

88

83*

4

115

101

4

96

105*

4

79

125*

5

113

90

5

103

106*

5

94

112*

5

63

109*

6

90

85

8

75

117

9

92

120

14

87

160

*Age <12 years: replaced in the age-matched control group by
 4 older teenage girls (MEF%: 123, 105,100 and 93)

Average duration of carpet making was 5 years. The scattergram (Fig 1) shows a moderate negative correlation between the duration of employment and MEF% (Spearman’s correlation = 0.5 with P <0.05).

Spirometry demands full co-operation and can be tiring and demanding. Our subjects were unable to give maximal effort and this was reflected in low FVC values. This was possibly due to cultural factors: all the girls were amused by the demonstration and all appeared embarrassed when asked to blow hard. However, MEF can be used to assess small airway narrowing even when FVC is reduced(8). In contrast to FVC (and FEV1), it is effort independent(9). Furthermore, it represents lung function in the smaller airways and so is effected in the mildest of airflow obstruction(9).

The weavers had lower MEF values than their sisters. Despite this, most of their values remained within the normal range. however, MEF has significant variation in normal populations(10). This diminishes its value in identifying mild expiratory airflow obstruction or "small airways disease" in an individual patient. However, if a study group is compared with a carefully selected control group then a statistically significant difference in MEF between the groups is of interest.

Carpet weaving is a home based industry. This raises the issue of "contamination" of the control group. The sisters had never worked as weavers but if we postulate that they might nevertheless have been inadvertently exposed to dust in the home then this effect, if present, would serve to spuriously reduce any measured difference between the subjects and controls. This source of error therefore would not detract from any differences we have been able to demonstrate; rather it would serve to underplay any detrimental effects on lung function.

The difference in MEF values demonstrated occurred after a mean duration of employment of only 5 years. Reduction in MEF was also shown to be correlated with duration of work.

There is very little information on the respiratory health of carpet weavers. They are exposed to cotton and to wool. A previous study has shown reduced MEF values in adult women weaving wool carpets compared with controls(11). We did not find a clinical difference between our groups - all the girls were asymptomatic. This paper’s contribution is to show that early spirometric changes have already occurred in child workers before they reach adulthood. A reduced MEF in an otherwise normal spirogram has been shown to be a useful indicator of the presence of bronchial hyper-responsiveness(10). It suggests early disease in the peripheral airways of asymptomatic patients(12). Such persistent airflow obstruction residing in peripheral airways may serve as a base for future recurrent episodes of asthma(13). Bronchial hyper-responsiveness is a feature of byssinosis(14). Byssinosis is seen after chronic deposition of cotton into small airways. It is recognised in workers exposed to unwashed cotton containing bract, rather than the form used by weavers. Ozesmi et al(15) showed that a large number of workers in a carpet weaving factory suffered from a disease indistinguishable from byssinosis, including a reduction in MEF values compared with controls, even though wool was used almost exclusively, the only cotton being the warp. The wool was contaiminated with endotoxin, which was blamed rather than the material itself.

Sambhav Social Services Organisations, 13-Balwant Nagar, Gwalior 474 002, Madhya Pradesh, India. Child Advocacy International, 79 Springfields Road, Trent Vale, Stoke on Trent, ST4 6RY, UK.

Contibutors: DA designed & conducted the study and wrote the paper.

Funding: None

Competing Interests: None declared.