Document pmJnMyrDdJN7dL7pE7kr0KqX

British Journal of Industrial Medicine 1982;39:338-343 Lung function and radiographic change in chrysotile workers in Swaziland MARGERY McDERMOTT,' MARGARET M BEVAN,' P C ELMES,' J T ALLARDICE,3" AND A C BRADLEY3'1' From the MRC Pneumoconiosis Unit,1 Cardiff, Turner and Newutt Ltd,' Manchester, and Havelock Mine,3 Swaziland abstract The effect on lung function and radiographic indices of exposure to chrysotile asbestos was investigated by cross-sectional studies in two groups of men at Havelock Mine, Swaziland. The first group consisted of 214 employees and ex-employees, mean age 52, who had been employed for at least 10 years, and whose dust exposure ranged from minimal for surface workers to very heavy for those in the grading and bagging sections of the mill. In this group 29% had category 1 or more simple pneumoconiosis and 4-5% category 2 or more. For surface and mine workers, the estimated annual deterioration in FEVi and FVC and the increase in category of pneumoconiosis was similar to that due to age alone, while the heaviest exposure almost doubled the decline in lung function and trebled the rate of progression of pneumoconiosis. The second group consisted of 224 men, mean age 33, all currently working in the mill and having been employed there for at least a year. In this group 30% had category 1 or more simple pneumoconiosis, and 2-7% category 2. Exposure in the dustiest sections of the mill more than doubled the estimated annual decline in lung function and doubled the rate of progression of pneumoconiosis. The production of chrysotile asbestos is usually con sidered to be less hazardous to health than that of crocidolite and amosite. Rossiter el al' found a low level of asbestosis among workers in Canadian chrysotile mines compared with that reported from Italy.3 The aim of our investigation was to determine the extent of occupational lung disease among mill, mine, and surface workers at Havelock chrysotile mine, Swaziland, from where little asbestos-related disease had been reported. The lack of a serious hazard should be confirmed if the continued produc tion of chrysotile is to be justified. The work force at Havelock consists of about 2000 Africans and 100 Europeans who form a community that is the third largest centre of population in Swaziland. The mine has been in full operation since the late 1930s, at which time most of the Africans were from Malawi, Mozambique, and South Africa; today, however, .they are mainly Swazis. 'Present address Employment Medical Advisory Service. Leeds ^Present address: Office of the Medical Officer of Health. Port Elizabeth. South Africa. We report two studies that took place about five months apart and have been treated separately. First study Mitt HODS An attempt was made to trace all men who had been employed for 10 years or more before 1 y70. A search of the mine records produced a list of 748 men, of whom 185 were still employed. 153 had left but still lived in Swaziland, 200 had moved to Malawi or Mozambique, and the remaining 210 had died. In the group of 338 living in Swaziland were 270 Swazis and 68 Europeans, of whom 235 Swazis and 36 Europeans attended the mine hospital for chest radiographs and examinations, lung function measurements, and to complete occupational his tories and respiratory questionnaires. The respirat ory symptoms are currently being analysed; the analysis described here is of lung function and radiographic data from the Swazis. Twelve of these, who were exposed to coal or silica dust in other occupations, were excluded as were nine with incomplete data or technically unsatisfactory results, leaving a group of 214 men. Figure 1 gives the dis tribution of age and years of employment. The mean LAM 019134 e SlOS ind. icon face :9% and ;ry ilOSt The ving lple han i of five >een i. A 748 I left J to had vert axis for lion hisiralihe and e sc. iher with ults. disican Lung function and radiographic change in chrysotile workers tn Swaziland 339 Fig 1 First study: distribution of (a) age and (b) years of employment. age was 52 years and mean duration of employment 25 years. The radiographs, after being placed in random order, were read by three experienced readers. The profusion of irregular and rounded small opacities was assessed according to the 12-point scale (categories 0/- to 3/ + ) of the ILO classification/ and the index of parenchymal change was taken as the median of the three readings. The categories 0/ -- to 3.'+ were represented by numbers 1--12 to facili tate numerical analysis. The forced expiratory volume in one second (FEVi) and forced vital capacity (FVC) were meas ured using a McDermott digital spirometer. Each subject's height was measured, and the FEVi and FVC expressed at a standard height of 1-7 m, using the regression model proposed by Cole/ where each reading is multiplied by the factor (1-7/height).; No accurate estimates of dust exposure arc poss ible, as monitoring of the airborne dust was not started until 1976, when a major dust suppression programme began. Discussions with reliable, longservice employees, however, suggest that dust con centrations, particularly in the mill, had been extremely high in the past. They have compared previous conditions with a present, unmanned sec tion in which the airborne dust concentration is about 50 fibres/ml (5 x 107 fibres/m1). From their recollections, it seems that in most areas of the null the reduction in visibility due to the airborne dust was similar to, or worse than, in this "reference" section, although, since about 1970, some attempts had been made to reduce the dust concentrations. The proportion of fibrous dust varies among differ ent areas of the mill, the lowest proportion being in the rock crushing plant. In the mine, which is wet, dust concentrations have always been considerably lower than in the mill. Asbestos is transported from Havelock on an aerial ropeway; it was originally packed in hessian sacks, which resulted in the work ers on the ropeway receiving a high dust exposure. Since about 1970, however, polyethylene bags have been used thereby considerably reducing the dust concentration. No section of the workforce may be described as "non-exposed," as in dry, windy weather dust from the mill and waste tips is blown over the whole area, so that even estate workers are exposed to the "background" concentration. This concentration, too, is now lower than in the past as the waste tips have been dampened since about 1970, The jobs listed in the men's occupational histones have each been allocated to one of six categories, J1-J6 in ascending order of dust exposure. As these divisions are based on crude estimates, there is no way of calculating the absolute dust concentrations in each location. Table 1 lists the categories. Table I First and second studies. Allocation ofjobs to category of dust exposure Jl All surface jobs except road sweeping Aerial rope wav since Id7u J 2 Mine V*jsie dump since I v7U JJ Aerial ropeway before 1^71) J-1 Road sweeping Mill rock crushing plant J5 Waste dump before l'^7<i Mill treatment, conditioning, and sample plants Jf> Mill: grading and recovers plants l ihre storage Magging For workshop and maintenance workers who entered the mill regularly, their exposure has been calculated as equivalent to one das a week in JS Results Figure 2 shows the distribution of parenchymal change--29% of the men had category 1 or more pneumoconiosis. 4-5% category 2 or more, and 1-5% category 3. A multiple regression analysis has been used to relate lung function and radiographic indices to age and exposure, fable 2 shows the numbers of men who worked at any time in each of the six categories of exposure, from which it may be seen that, except LAM 019135 340 McDcrmott, Bevttn, Elmes, Allardice, and Bradley Table 2 First study: number of men in each exposure category Category Jt J2 J3 J4 J5 J6 Men who worked at Men with Men with any time in category exposure in no exposure in no other category No Mean exposure except J1 (years) 143 16-4 102 14 7 26 ! IS 27 10-1 63 1 10 25 9-3 38 32 0 4 11 3 _ 65 12 10 24 9 for J1 and J2, few men worked in one category alone; even when J1, the background level, is dis counted, the numbers in all categories except for J2 are too small for separate analyses. As already men- tinned, il is impossible to relate numerically the exposure levels in the six categories; hence, each has been treated as a separate variable. Table 3 gives the results of the multiple regression analysis which shows that exposure in categories J1-J4 produces no significant deterioration in any of the indices, while all indices deteriorate significantly with age. Exposure in categories J5 and J6 significantly reduces FVC and increases the profu sion of small opacities, while FEVi is significantly reduced in J6. TTe FEVi/FVC% appears to be unaf fected by exposure, even at the highest level. This table also shows the effect of excluding categories J1-J4 from the analysis; some significance levels are increased and the effect of J5 on the FEVi has become significant. The annual decline in FEVi in category .15 is 40% greater than that due to age alone, and in J6 80% greater, while for FVC the figures are 60% and 90% respectively. The increase in profusion of small opacities in J5 is slightly more than in J6; in both categories it is over three times greater than that due to age alone. Using the regres sion equations, the effect on the FEVi and FVC of a man working from 20 to 60 years of age in either of the two dustiest cuteitories has been estimated (fig 3). The total number of men who were exposed to the highest dust concentrations in categories J5 and J6 is 77; fig 4 shows the distribution of parenchymal change in this group compared with that of the 137 men who worked only in the lower exposure categories J1-J4. Of the men in the low-exposure group, 22% had category I or more pneumoconiosis compared with 43%, in the high-exposure group. Table 3 First study: dependence of lung function and radiographic indices on age and years of exposure in all six dust categories and in the two dustiest categones Index Regression (oefficient r of index on age and years in category Constant Age Jt JJ JJ J4 In all six Just categories FEV -0-027 ` ' ` FVC' -U-024' *' FHV/FVC% -0-294 *'' Parenchymal -`-U-033*' Change7 In i*o dustiest categones FEV, -0 027 r 0-004 * ' ' -0-003 -0-007 + 0-040 -0-015 FVC -0-025 r 0-01)5*'* F!:V/FVC% -0-286 0 ti73" * Parenchymal +0-032 - 0-01 2* * Change* 0 -0-002 -0- 046 0 -0-01 1 -0-011 -0-151 -0-023 -0-005 -0-004 -0-040 -n-020 -0-014 -0-02 ! * -o-ooH 0 l)73`" -0-023* -0-028* -O-00I *- 0 06 1 * 3 97 4-95 84-88 1-23 -0-01 1 1 0-OII3* -0-02 1 r 0-000* -0-016 - (1 OHIO * -0-022 * 0-010* -0-0311 r (MIX4 0-120 t 0-150 -0 OKI : 0-0|4* * * *-0-068 r 0 024*'* 3-04 4 Ho X^-07 1-09 rCjiciiories 0/ -- In V+ rcprysenlcd by Nos 1-12 p <. (MKII. ' p < li t*I. p < (HI5. he tobacco consumption was certainly under lUg and KiS probably onlv about 4 g. When the effect of dust exposure is ignored, there are no significant differ on ences between smokers and non-smokers in FEV,. ies FVC, or parenchymal change. The FEV,/FVC% is of significantly lower, however, in smokers titan in rl\ non-smokers (p < 0-00I ), J6 The effect ol dust exposure in smokers is not ;u* significantly different from that in non-smokers for ily any of the lung function indices. In category J6, af- however, the increase in profusion of small opacities is significantly greater in non-smokers than in smok mg ers (p < 0-05). tee V, Age Second study .V, Fig 3 First study: effect ofage and dust exposure on FEV< METHODS ige and FVC. As the men working in the mill are exposed to a ihe considerably higher concentration of dust than those a sc in the mine or in surface jobs, the second group was >rc chosen to consist of ail current mill employees :tes except for those included in the first study. A total of es- 273 men attended the mine hospital and were tested if a as before. Twelve Europeans have again been of excluded, while of the 261 Swazis, 17 had incom led plete data or technically unsatisfactory results and a further 20 had been exposed to silica or had worked the in the mill for under a year. The mean age of the h is remaining 224 men was 33 years and mean duration nal of employment eight years, the distributions being : 37 shown in fig 5. The lung function was measured by ure the medical staff at the mine, using a Vitalograph. ure and the chest radiographs were read as before by the )MS same three readers. The jobs have again been allo cated to categories J1-J6 (table 1), each man being currently employed in one of the mill categories-- that is J4. 5, or 6. i It! sUl. I s Figure 6 shows the distribution of parenchymal change--30% of the men had category 1 or more pneumoconiosis and 2-7% category 2, while there were none with category 3. Table 4 shows the number of men who worked at any time in each of the exposure categories and the mean duration of exposure. As in the first study, lung function and radio- graphic indices have been related to age and exposure Fig 4 First study: distribution ofparenchymal change m high-exposure and low-exposure groups. by multiple regression analysis, using categories J4, 5, and 6 only, and again treating each as a separate variable. The results of the analysis are shown in EFFECT OF SMOKING table 5. Although the men are much younger, and Of the 214 men in the sample, 10% were ex- have worked in the dust for shorter periods than smokers and 38% smokers, mostly of pipes and those in the first group, the deterioration in the indi hand-rolled cigarettes, with some manufactured ces due to exposure is again clearly shown. Paren cigarettes. Only imprecise estimates of the amount chymal change, FEV,, and FVC are all significantly smoked could be obtained, but the average daily affected by exposure in J5 and J6, while the FEV,/ LAM 019137 342 McDermott, Bevan, Elmes, Allardice, and Bradley 120-j j | 100- 0 0/- 0/0 ' 3/' I VC ' 1/1 IJ2 | 20 2/2 ' 2ft 3/2 ' 3/3 3 I II2I3 0^jfTiocof*os2S category 3/* Fig 6 Second study: distribution ofparenchymal change. profusion of small opacities for men in J5 is no longer significantly different from that due to age alone, but for men in J6 it is doubled. Figure 7 shows the estimated effect on FEVi and FVC of a man working from age 20 to age 60 in either of the two dustiest categories. Fig 5 Second study: distribution of (a) age and lb) years of employment. Table 4 Second study: number of men in each exposure category FFFhCI OF SMOKI.NO In this sample 4% were ex-smokers and 33/c smok ers, with smoking habits similar to those in the first sample. As before, estimates of the amount smoked are not accurate, but the average was probably 3-4 g daily. There are no significant differences between smokers and non-smokers m tiny of the indices, and the effect of dust exposure in smokers is not significantly different from that in non-smokers. Category Men who worked at any time in category Discussion \o Mean eipo.sure (yean) Ji IV 49 J2 17 4-5 J3 1 4-0 , J4 :o M) J 5 J47 tvX J(y 4h FVC% is again unaffected. In tile rock crushing plant (J4). which has the lowest proportion of fibrous dust, the effect of exposure is not significantly different from that of age for any of the indices. The regression coefficient of parenchymal change for J4, although greater than for J5, is not statistically significant as there are only 20 men in the category. Table 5 also shows the effect of excluding J4 from the analysis. Exposure in J5 almost doubles the annual decline in FEV, and more than doubles that in FVC, while in J6 the decline in FEV, is more than doubled and that in FVC trebled. The increase in The incidence of parenchymal change category I or more is similar in both studies, being 26% in the first, older group and 30% in the younger. Ihe younger group, however, were all employed in the mill, 63% having exposure in J5 and j6, while only 36% of the older group ever worked in these dus tiest areas. In Quebec mute and mill workers with 10-30 years' exposure the incidence of parenchymal change category I or more was 3%,' while Ghe/.zt cl til2 reported an incidence <>1 26-8% among work ers exposed for at least one year in an Italian chrysolile mine. In Corsica, of men examined 14 years after an 1 1 -year exposure period, 40-3% were found to have parenchymal change category 1/1 or more.5 In both groups of Swazis the increase in pro fusion of small opacities with age is statistically significant; this might be explained by the presence of the background dust over the area, to which, before the dampening of the tips, the whole popula tion was constantly exposed. Fh v exp of ; dele prof cu.",; w'w.j e\p% be*.': LAWI 019138 Lurig function and radiographic change in chrysolite workers in Swaziland 343 Table 5 Second study' dependence of lung function and radiographic indices on age and years of exposure in all three mill categories and in the two dustiest caiegones Index Regressu/n Loeffiaems of index on age and sears in tategorv Age J4 J5 J's Constant All three mill categories FEV FVC FEV FVC % Parent, hsmal change7 In two uusttesi caiegones FEV FVC FI-.V.FVC^c Parenchymal change* -0-029** * -0 01 ?** - 0-374 ** -o-u:4*** -0-014 -0-026 -0 181 -U 070 -0 031 r U-005*** -U-U20 i 0-005* " -0-353 r U-006*** + U-U62 r 0-01 2* ** ^Categories U< * **p < U-001 p < 0-01 ' p < U-U5 to 3/- represented by Nos 1-12. -0-029** it dip* * * -0-02 v -il-U4.V -links" - .i ft'.-* * ` -hi 4 58 4 97 c-4-3 < o 87 -0-026 r 0 009** - 0 030 r (l-tltbr * -0-06.3 t (J- 1 | 3 * 0-U29 r 0-020 'Mi'' : 'ii (,12" .,14 4 _ a u|2* ** .'i 1 -1: iin i r ;ii-o27* 4 61 ' 02 '14 III o 73 FEV, o Age ' JS ' j6 had been reported from Havelock, it would appear that past dust concentrations in the mill have been unacceptably high, probably at least 50 fibres'ml (5 O Age x 10' fibres/m'). Since IV76, however, measures have been taken to suppress the dust, and the levels of exposure have gradually been decreasing. A J5 programme of regular health checks on all 20U0 employees has recently started, which, together with j6 continuous monitoring of the airborne dust, should appreciably reduce the risk of asbestos-related dis ease in the future. 20 60 20 Age References Fig 7 Second study effect of age and dust exposure on )! FEV\ and FVC. 1C 1.R<issuer CL. Bristol [J, Cartier (M et at Radn(graphic changes in chrssoiile asbestos mine and null workers in Quebec. An h 1C The results of the first study have shown that ('.nvimn Health 1972.24 3SS-400 1C exposure in any part of the mill, with the exception Ghczzi I. Aresini G, Vtgliam 1C II riM.hu dt asbestos! m una iy :s- of the rock crushing plant, caused a considerable deterioration in lung function and an increase in the nunierj di anuaiito ensotilo. Wed Lav 1972,63 IH9-2I1 1 International Labour Olhee II U IVSO Internaitimul tlasstfuatiori of radiographs ofpm-tonot omoses Geneva ILO. lh profusion of small opacities. These findings are 19811. nil confirmed by the results of the second study, in ole IJ la near and proportional regressn mi models i n the pred .'./i which the men were much younger, and were iction of ventilator', function, Journal of the Staliunal Sotictv (series A) hv7s;|3K 297- 1 3"? k- exposed to dust for much shorter periods. Although, ' Vi.illai JR. lioiiim ( Radiographic. Guugcs in ehrvsotile mine ail before these studies, little asbestos-related disease and mill ex-workers in Corsica I.ant; 198(1,157 155-61 14 TO or (>- My .ce :h. la- LAM 019139