Elevated Lead Contamination in Boat-caulkers’ Homes in Southern Thailand

Abstract

Surface-wipe lead loading was measured at various locations in the homes of 31 boat-caulkers and 31 location-matched controls to identify factors associated with household lead contamination. Data were obtained by observation checklist and questionnaire. Lead loading was significantly higher in caulkers’ than in control households. Median lead loadings (in µg/ft²) of various locations in caulkers’ homes were windowsill, 43.9; exterior entrance, 9.5; interior entrance, 21.1; living room floor, 9.8; and bedroom floors 15.6. Corresponding levels in control homes were all less than 0.2 µg/ft². Regression modeling indicated that lead loading was higher in caulkers’ homes that were closer to a boatyard, in which the caulker had a longer duration of boatyard work, and in which there were no children aged under 6 years resident. Exterior and interior entrance and living room floors had lower lead loading than windowsills. However, bedroom floors had significantly higher lead loading, similar to windowsills.

Although lead hazard to humans has been known since ancient times and many regulatory actions and reductions in lead risk have been achieved over the past century, lead contamination and exposure remain significant problems worldwide, especially among workers in developing countries. Exposure to lead from the workplace may not be limited to the job site. Workers can inadvertently carry lead home from work on clothes, skin, hair and tools, and in their vehicles.^1,2 These “para-occupational” or “take-home” exposures of non-workers, particularly workers’ families, may result in lead poisoning of household members.^3–8

In the boat repair industry in Thailand, lead oxide (Pb₃O₄) is used in the process of caulking wooden boats. There are reported to be 234 major boat construction and repair yards in Thailand, with 86 in the southern region.⁹ Boat-repair workers, especially caulkers, are readily exposed to lead but, as they are classified as informal workers, their health status is not routinely monitored and their total numbers not recorded. The only legal requirement regarding exposure in the workplace is that the 8-hour time-weighted average airborne lead should not exceed 200 µg/m³. Direct contact with caulking material containing lead oxide, however, is likely to be a major pathway of contamination of the bodies and clothes of caulkers. Our previous study revealed that 48% of all workers and 67% of caulkers had blood lead levels exceeding 40 µg/dl, and none of the workers took a shower or regularly changed their clothes prior to going home.¹⁰ Sixty-six percent of workers washed only their hands and face before going home but did not clean their shoes, and 60% of workers did not shower until after at least half an hour after work. These poor hygiene practices meant that the workers were likely to carry lead contamination to the home.

Lead contamination of children and households in communities close to boat-repair yards has been documented.¹¹ However, no existing data indicate the extent of take-home lead among boat repair workers in Thailand. The objectives of the study were to evaluate and determine the level of lead contamination in the homes of boat-caulkers and of a control group. An additional objective was to identify the relationship between household lead contamination and caulkers’ behaviors and other influencing factors.

MATERIAL AND METHODS

Study Population and Sample Size

Workers were selected to participate in this study from among the caulkers in two boat-repair yards in one area of Nakhon Si Thammarat province, Thailand, that had been the site of our previous study.¹⁰ Both boatyards in the study had a similar production process, number of workers, style of employment, working hours, facilities, and operation period, although the average productivity was somewhat higher in boatyard 2 (average of 10 boats per month) than in boatyard 1 (average of 7 boats per month). The following inclusion criteria were applied to caulkers:

1. had been employed in a boatyard for at least one year,

2. had lived in the current residence in Pakpanang district, Nakhon Si Thammarat province, while working in the boatyard, and

3. the male head of the household was willing for his home to be included in the study.

With these inclusion criteria, we enrolled 31 caulkers’ homes in the study.

Control households were selected by purposive sampling and matched 1:1 on location with each of the caulkers’ homes, with the following inclusion criteria:

1. located within 30 meters of, but not immediately adjacent to caulker’s homes,

2. the house was of the same general construction type as the caulker’s house,

3. no person living in the home worked in a job with known exposure to lead, and

4. the male head of the household was willing for his home to be included in the study.

Sampling Strategy

All home surveys were conducted during May 2005. In general, researchers visited homes only when caulkers were present, typically during workday evenings or weekends. Researchers gathered data on the study homes using an observation checklist and on-site survey and a face-to-face interview. A quantitative assessment of lead contamination inside the home of each caulker was made using a wipe method. Inside the homes, sampled locations were the interior and exterior floors adjacent to the entrance door normally used by the caulker when returning from work, the livingroom floor, and the caulker’s bedroom floor.

In addition, windowsills in the living room were sampled to evaluate possible environment sources of lead contamination in the household. In this setting, windows were generally left open, so no distinction was made between interior and exterior windowsill. Specimens were collected from similar locations in both caulker and control households. All specimens within each household were collected on one day.

Lead Sampling and Analysis

Surface lead specimens were collected using a standard wipe method. Wipe dust specimens were collected using Zefon lead wipes (Zefon International Inc.), which are stated to meet the American Society for Testing and Materials Standards’ ASTM E1792 requirement for lead dust sampling wipes and are listed in the American Industrial Hygiene Association (AIHA) policy on sampling materials for lead in surface dust. Both floor and windowsill dust wipe specimens were collected from a total area of 400 cm² (2 positions, each 10 cm × 20 cm, for floor dust specimens and 4 positions, each 5 cm × 20 cm, for windowsill specimens).

Dust wipes were placed over representative surfaces and pressed evenly over the entire surface with a gloved hand. All specimens were placed in pre-labeled, lead-free plastic tubes. To prevent contamination, the researcher performing the dust wipe collection used new, clean, powderless, disposable, natural rubber (latex) gloves for each wipe sample. In addition, all templates in this study were made from disposable plastic sticker tape. Sampling techniques followed the NIOSH Manual of Analytical Methods 9100/1996.¹² Two field blanks per set of caulker and control homes were used for quality control.

Lead wipes were then analyzed for lead content of the collected dust so that the lead loading (mass of lead per unit surface area) could be determined. This wipe method is similar to that used by US Department of Housing and Urban Department (HUD) ¹³ for measuring lead loading levels. All lead wipe samples were analyzed by flame atomic absorption spectrophotometer (AAS, Hitachi Model Z-8200) for which the limit of detection (LOD) was 0.01 ppm according to NIOSH 7082/1994.¹⁴ Lead analysis was performed by scientists at the Central Equipment Unit, Faculty of Tropical Medicine, Mahidol University, Bangkok. This laboratory has received certification for this type of analysis from the Department of Science Service, Ministry of Science and Technology. Re-analysis in the same laboratory was performed on a random 10% subset of specimens, and a further subset analyzed in a second laboratory. No major discrepancies were found. Lead loading was calculated and expressed in units of micrograms per square foot (µm/ft²) of sampled area in order to facilitate comparison with the majority of reports in the literature. All blanks submitted for analysis were below the limit of detection for flame AAS.

Statistical Analysis

Data were entered twice using EpiData 3.1¹⁵ and their accuracy verified. Descriptive and analytical studies were performed using Stata version 7.0.¹⁶ For wipe surface specimens with lead measurement below the analytical LOD, a lead mass equivalent to LOD/√2 was assumed for data analysis.¹⁷ Characteristics of study homes were summarized by percentage, mean, and standard deviation. Descriptive statistics were used to evaluate the distribution and test normality of data. Lead loading data were not normally distributed and logarithmic transformation of data also did not result in a normal distribution owing to the wide range of lead loading. Therefore, the nonparametric Wilcoxon matched-pairs signed-ranks procedure was used to test for differences in lead loading between households of caulkers and neighborhood controls. Relationships between lead loading and influencing factors were estimated using the distribution-free Cox proportional hazard regression analysis,¹⁸ in which the lead loading was used in place of the duration variable. Although it is somewhat unusual to use the Cox model as essentially a “non-parametric” model, owing to its development and main application for the analysis of time-to-event data, such a use has been described previously by the Macquarie University Statistical computing group.¹⁹ None of the observations were considered to be censored. Using the Cox model in this way, a hazard ratio significantly below unity indicates a higher level of lead loading compared with the reference category. To take account of the matched-pairs design, as well as the multiple samples from each household, the cluster option and robust estimates of variance were applied. Initially, all variables that showed a crude relationship with lead loading in univariate analysis with p-value < 0.2 were included in the model. The significance of each variable in the multivariate setting was then assessed using the change in log-likelihood of successive subset models in a backward elimination process until all variables had a p-value < 0.05 in the likelihood ratio test. Validity of the assumption of proportional hazards of each model was tested using the method of Grambsch and Therneau²⁰ as implemented in Stata.

RESULTS

Study Homes

All 31 caulkers who were currently working in the two boatyards agreed to participate and were included in the study. As none of them had the same residence, data were obtained from 31 caulkers’ homes and 31 neighborhood-matched control homes.

Demographic Characteristics of Caulkers and Controls

All caulkers were in the informal employment sector, worked nine hours a day plus one hour for rest, and received a daily wage. They were not provided with hand-washing and shower facilities, personal protective equipment (PPE), or annual health examination by the employers. All caulkers were male and Buddhist. The average age of caulkers was 49 years (range 24–82). Most caulkers (45%) were in the age 51–60 years of age. Most of the caulkers had only junior primary school education (65%), were of low socioeconomic status, and were married (77%). Approximately 42% of caulkers had children under the age of six. The average lifetime duration of working as a caulker was 22 years (range 6–52). Those caulkers who reported having had a previous job had all been farmers. Similarly, those who had an alternative job when no caulking work was available were all engaged in farm work.

The occupation of the male head of the control households included farmer (19 households), farm laborer (7 households), and casual laborer (5 households). Most (94%) had only primary school education and all households were of low socioeconomic status.

Behaviors Related to Take-home Lead

About half of the caulkers drove a motorcycle or rode a bicycle to and from the boatyard and had lunch at home. Three quarters (74%) washed only their hands and face before going home after work. About 97% of the caulkers never changed their clothes before going home; 71% percent did not clean their shoes; 81% did not shower until after at least half an hour after returning home in the evening; and 39% took their hand tools home. Most caulkers (97%) washed their clothes at home.

Characteristics of Study Homes

The mean age of the homes of caulker and control groups was 27 and 25.5 years, respectively. Seven sets of caulkers’ and control homes were located not more than 1 km from a boatyard and 24 were located further away. No lead-related hobbies or work was practiced in any of the homes. Approximately 77% of the houses among caulker and control groups were detached with a raised floor. The walls and floors of the majority of houses were made from wood. Fifteen of the caulkers’ homes and 9 of the control homes had no paint. Five households in each group had paint throughout the house. The remainder had some parts, such as window and door frames, painted. Many of the study homes had indoor and/or outdoor pets such as cats, dogs, and/or chickens, more commonly among the caulkers. In most of the study homes, the floor was cleaned with a frequency ranging from more than once a month to once a week, using a mop, feather duster, cloth, or broom (Table 1).

TABLE 1.

Characteristics of Caulkers’ and Control Homes

Items	Caulker (n = 31)	Control (n = 31)
Home location (distance from boatyard)
≤ 1 km	7	7
> 1 km	24	24
Age of the study home (years), mean (SD)	27.0 (± 17.1)	25.5 (± 17.2)
Lead-related hobbies practiced in home
No	31	31
Type of house
Detached house (elevated)	24	24
Detached house (not elevated)	4	6
Row house	3	1
Type of floor
Wooden	16	20
Wooden and linoleum	11	6
Concrete, tile, others	4	5
Type of walls
Wooden	25	22
Concrete	2	2
Wooden and palm leaves	2	2
Others	2	4
Painting of the house
No	15	9
Some parts (only door and window frames)	11	17
All parts	5	5
Have indoor/outdoor pet
No	9	18
Yes	22	13
Frequency of floor cleaning by broom
>1 time a month to 1 time a week	1	0
>1 time a week	30	31
Frequency of floor cleaning by mop
Never/≤1 time a month	6	5
>1 time a month to 1 time a week	16	18
>1 time a week	9	8
Frequency of dusting by feather duster
Never/≤1 time a month	7	5
>1 time a month to 1 time a week	15	19
>1 time a week	9	7
Frequency of dusting by cloth
Never/≤1 time a month	13	8
>1 time a month to 1 time a week	12	18
>1 time a week	6	7

Surface Lead Loading in Study Homes

Surface lead loadings in caulkers’ and control homes are shown in Table 2. The Wilcoxon matched-pairs signed-ranks test revealed that lead loading was significantly higher in the caulkers’ than control households for all sample locations. Median lead loadings (in µg/ft²) of various location in caulkers’ homes were: windowsill, 43.9; exterior entrance, 9.5; interior entrance, 21.1; living room floor, 9.8; and bedroom floor, 15.6. Corresponding levels in control homes were all less than 0.2 µg/ft². In caulkers’ homes, floor lead loadings of various rooms were moderately or strongly correlated, with Spearman correlation coefficients ranging from 0.46 to 0.68.

TABLE 2.

Surface Lead Loadings (µg/ft²) in Caulkers’ and Control Homes

		Percentile
Specimen Location	Group	25th	50th	75th	95th	Range	p-valuea
Window sill	Caulker	1.9	43.9	163.7	990.8	< 0.2–1059.8	0.0076
	Control	< 0.2	< 0.2	11.6	762.8	< 0.2–1070.5
Exterior entrance floorb	Caulker	< 0.2	9.5	54.1	306.8	< 0.2–475.4	0.0001
	Control	< 0.2	< 0.2	< 0.2	50.4	< 0.2–102.7
Interior entrance floorb	Caulker	< 0.2	21.1	38.6	243.3	< 0.2–273.1	0.0000
	Control	< 0.2	< 0.2	< 0.2	84.8	< 0.2–162.8
Living-room floor	Caulker	< 0.2	9.8	37.6	144.2	< 0.2–166.1	0.0001
	Control	< 0.2	< 0.2	< 0.2	38.2	< 0.2–51.6
Bedroom floor	Caulker	< 0.2	15.6	103.4	349.5	< 0.2–540.7	0.0004
	Control	< 0.2	< 0.2	< 0.2	102.4	< 0.2–127.7

^aWilcoxon matched-pairs signed-ranks test

^bAdjacent to the entrance door normally used when returning from work

The US Environmental Protection Agency (EPA) and Department of Housing and Urban Development (HUD) have set national standards for dangerous levels of lead in dust on floors and windowsills. The EPA and HUD hazard standards for dangerous levels of lead in settled dust are 40 µg/ft² for floors, including carpeted floors, and 250 µg/ft² for interior windowsills.^21–22 Among caulkers’ homes, contaminated dust in excess of US standards was found in 45% of bedroom floors, and in respectively 39% and 19% of external and internal floors adjacent to the entrance door. In addition, 23% of living room floors and 19% of windowsills exceeded these standards (Table 3). Using this floor standard, 61% of caulkers’ homes had at least one floor specimen in excess of the standard, in contrast to 13% of control homes.

TABLE 3.

Numbers of Homes Exceeding US EPA and HUD Lead Dust Standards

	Number of Homes with Specimen Exceeding US Standards (Percent)		Number of Homes with at Least One Floor Specimen Exceeding US Standards (Percent)b
Location	Caulker	Control	Caulker	Control
Windowsill	6 (19.4)	2 (6.5)	—	—
Exterior entrance floora	12 (38.7)	1 (3.2)	—	—
Interior entrance floora	6 (19.4)	1 (3.2)	19 (61.3)	4 (12.9)
Living-room floor	7 (22.6)	1 (3.2)	—	—
Bedroom floor	14 (45.2)	4 (12.9)	—	—

Note: US EPA and HUD standards are 40 µg/ft² for floors, 250 µg/ft² for interior windowsills.

^aAdjacent to the entrance door normally used when returning from work

^bExcludes windowsill

Surface Lead Loading in Caulkers’ Homes

Cox proportional hazard regression analysis, using lead loading as the time scale, was used to examine the association of lead loading in caulkers’ homes with location of the household and characteristics and behavior of the caulker. Initially, the model was applied to assess univariate associations (Table 4). Lead loading was significantly higher in homes that were closer to a boatyard, in homes with no young children, and in homes with no pet animal. It was also higher in homes of caulkers who worked in boatyard number 2, had worked longer in a boatyard, went home for lunch, did not wash before going home, and delayed showering and changing their clothes for between 10 and 30 minutes after coming home from work in the evening. There was no evidence of differences in lead loading with differences in house cleaning frequency, with taking hand tools home, or with washing shoes before going home.

TABLE 4.

Crude Associations of Lead Loading with Characteristics and Behaviors of Caulkers Using Cox Proportional Hazards Regression Models

Variable	Number of Caulker’s Homes	Hazard Ratio	95%CI	p-valuea
Sample location				0.002
Windowsillb	31	1
Exterior entrance floor	31	1.813	1.278, 2.573
Interior entrance floor	31	2.086	1.440, 3.021
Living-room floor	31	2.208	1.496, 3.260
Bedroom floor	31	1.487	0.913, 2.422
Boatyard (No. 2 / No.1b)	16/15	0.480	0.310, 0.743	0.001
Duration of work (≥20 / < 20 yrsb)	14/17	0.542	0.336, 0.874	0.012
Distance from boatyard (≤1 / >1 kmb)	7/24	0.415	0.256, 0.674	0.000
Have children under 6 years (No / Yesb)	18/13	0.530	0.341, 0.824	0.005
Place of lunch
(At home / Food shop-working areab)	16/15	0.577	0.363, 0.917	0.020
Type of vehicle
(Walk-others/Bicycle-motorcycleb)	16/15	0.659	0.418, 1.038	0.072
Washing body before going home
(No / Wash hand and faceb)	8/23	0.489	0.285, 0.839	0.009
Have indoor/outdoor pet (No/Yesb)	9/22	0.617	0.393, 0.969	0.036
Changing clothes and taking shower at home				0.005
Immediately after arriving homeb	7	1
> 10 to 30 min after arriving home	10	0.459	0.270, 0.779
> 30 min after arriving home	14	0.965	0.588, 1.582
Taking hand tools back home
(Sometime to everyday / Neverb)	12/19	0.781	0.481, 1.267	0.316
Washing shoes before going home (No / Yesb)	22/9	1.085	0.711, 1.656	0.706
Frequency of floor cleaning by mop				0.328
>1 time a weekb	9	1
>1 time a month to 1 time a week	16	1.424	0.839, 2.420
Never / 1 time a month	6	1.442	0.783, 2.657
Frequency of floor cleaning by a feather duster				0.335
>1 time a weekb	9	1
>1 time a month to 1 time a week	15	1.454	0.834, 2.535
Never / 1 time a month	7	1.379	0.799, 2.379
Frequency of floor cleaning by cloth				0.206
>1 time a weekb	6	1
>1 time a month to 1 time a week	12	1.580	0.895, 2.792
Never / 1 time a month	13	1.172	0.609, 2.255

Note: Variables indicated in bold type were included in the initial multivariate model.

^aP-value from likelihood-ratio test

^bReference group

A series of multivariate proportional hazard models was then constructed to identify the relationship of caulker behaviors and influencing factors with household lead loading among caulkers’ homes. Variables initially comprised sample location in the home, boatyard in which the caulker worked, distance of the home from the boatyard, duration of the caulker’s work at the boatyard, place of having lunch, type of vehicle used to travel to and from work, bathing before going home, time to taking a shower and changing clothes at home after work, having a child under age of six resident in the home, and having a pet. Backward elimination of variables not significantly contributing to the fit of model using the likelihood ratio test yielded a final model in which distance from the boatyard, sample location, duration of the caulker’s work in caulking, and having a young child in the home were independently and significantly associated with lead loading among caulkers’ homes (Table 5). The rather marked association between boatyard and levels of surface lead loading in caulkers’ homes seen in the univariate analysis was almost entirely accounted for by the distance of the caulkers’ homes from the boatyard. All of the caulkers in boatyard 1 lived further than 1 km from their place of work.

TABLE 5.

Best Fitting Model of Lead Loading in Caulkers’ Homes

Variable	Number of Caulkers’ Homes	Hazard Ratio	95%CI	P-value (from likelihood- ratio test)
Sample location	31			0.0031
Windowsill		1a
Exterior entrance floor		2.014b	1.275, 3.182
Interior entrance floor		2.311b	1.509, 3.541
Living-room floor		2.404b	1.454, 3.975
Bedroom floor		1.298a	0.686, 2.454
Resident children aged under 6 years				0.0023
Yes	13	1
No	18	0.587	0.391, 0.881
Duration of work				0.0011
< 20 yrs	17	1
≥20 yrs	14	0.568	0.373, 0.864
Distance from boatyard				0.0001
> 1 km	24	1
≤1 km	7	0.456	0.324, 0.640

^a,bHazard ratios for the variable “sample location” not having a superscript in common differ significantly (p < 0.05)

Living not further than 1 km from a boatyard, ≥ 20 years of working as a caulker, and the absence of a child under 6 years of age in the home were associated with increased lead loading in the home. External and internal entrance floors and living room floors had lower lead loading than windowsills. However, bedroom floors had significantly higher lead loading than other floors, with values similar to windowsills.

None of the regression models was found to violate the assumption of proportional hazards.

DISCUSSION

These results reveal elevated levels of lead contamination in the homes of boat-caulkers compared with con- trol homes. There is substantial evidence that the caulkers’ occupational exposure to lead is the primary source of the surface lead contamination in their homes. None of the caulkers in this study reported showering or regularly changing their work clothes before going home. Thus, it is likely that the main source of lead contamination being carried from the boat-repair yards is lead oxide on the skin and clothes.

The possibility of contributions of lead from sources other than the boatyard was considered in this study. These confounding sources included soil contamination from past motor vehicle emission and particulate fallout from local source of industrial air pollution. The study was designed as a matched-pairs cross-sectional comparative study to control for these potential confounders. In a number of studies, disintegration of lead-based paints has been implicated as a source of household lead loading.^2,23 In our setting, we found that the walls and floors of study homes in both groups were mostly made from wood with only some parts, such as window and door frames, being painted. Observation revealed no paint chips flaking from these surfaces; therefore they were unlikely to be contributing substantially to the measured levels of lead in the boatcaulkers’ or control homes.

Not only was there an evident difference between the levels of lead loading in the caulkers’ homes and those of control households, but the lead levels were high in each location of the caulkers’ homes. In general, most caulkers changed their work clothes on either the external or the internal floors adjacent to the entrance door normally used when returning from work. Some caulkers changed in the living room area that was also used for resting and drinking. These areas were likely to come in contact with the caulker’s feet and lead-contaminated clothes immediately after the caulker arrived home. However, the location where the lead contamination most frequently exceeded US standards was the caulkers’ bedroom floor (45%); some of the caulkers reported keeping their hand tools in this room, but the elevated levels in the sleeping area might also have been a result of a lower frequency of cleaning than in other parts of the home, although no data were obtained on the relative frequency of cleaning each room of the home. Furthermore, in most houses in the study, the “bedroom” was not a separate room but simply an area separated from the living room by a wardrobe or other furniture. Thus, dust could readily spread from other areas (such as external and internal floors adjacent to the entrance door or the living room floor) to the sleeping area. This is consistent with the correlation of floor lead loading values from various rooms of the house.

In the cultural setting in which our study was conducted, young children usually sleep in the same room or area as their parents. Thirteen of the 31 caulker’s households in our study had resident children under the age of six years. Children in this age range are most susceptible to lead’s toxic effects. Because of their normal hand-to-mouth behavior, young children may ingest lead from their environment and are considered to be at greatest risk for lead poisoning.²⁴ It is of interest in our study that the homes of caulkers with young children resident had somewhat lower lead levels than other caulkers’ homes. It is tempting to ascribe this differential to a more effective cleaning of the home where a young child lives, but our data failed to show any significant relationship of frequency of house cleaning, whether by wet mop, dry cloth, or feather duster, with presence of children or with lead loading.

Nevertheless, despite the lower levels of lead loading in households with young children, some of these children are still likely to be exposed to higher than acceptable levels of lead in the home environment. Although there is no current standard in Thailand to evaluate the health risk to young children from exposure to lead-contaminated dust in the home, 6 of the 13 homes with young children had lead loadings in excess of the US standard of 40 µg/ft², suggesting a significant risk to the children of caulkers. Several studies have reported similar situations in which transport of lead dust from parents’ occupation or hobby had been an important pathway of childhood contamination.^3,5,11,25

Not all the deposition of lead in all households, however, is likely to have been due to take-home lead. A number of control households had high levels of lead loading, and these were houses located within a short distance (< 1km) from a studied boatyard. Distance from a boatyard was a significant determinant of household lead loading not only among the caulkers, but also among the controls (results not shown). Furthermore, among the subset of locations close to a boatyard, there was no evidence of any difference in the lead loading of caulker and control homes for both windowsill and indoor levels, suggesting that the predominant pathway of household lead contamination in close proximity to a boatyard is not via take-home lead, but by other mechanisms, possibly transport of windborne lead-laden dust. Similar instances of fall-off of household lead levels with distance from source have been described in the area surrounding lead smelters^26,27 and in other communities close to boat-repair yards.¹¹

Nevertheless, overall, there was clear evidence of elevated lead contamination of caulkers’ compared with control homes, the difference being most marked among households at some distance from a boatyard, where most control homes had undetectable levels of lead loading. This difference is most reasonably explained by the take-home process. Furthermore, it seems that deposition of take-home lead might be a cumulative process as suggested by the higher lead loadings in homes of long-time workers. The low frequency of house cleaning among our study caulkers might explain this apparent cumulative effect over time of working.

Personal hygiene of the caulkers was generally rather poor. However, no single behavior could be identified as being associated with elevated household lead loading even though it might be expected that deposition of take-home lead could occur as a result of transfer of lead oxide on work clothes, hand tools, shoes, and vehicles, combined with low frequency of floor cleaning, not bathing before going home, and delaying taking a shower and changing clothes at home after work, etc. It is possible that the effects of poor hygiene behavior are cumulative, or that there was insufficient variation in hygiene behavior, whereby it would be difficult to identify any particular contaminating behaviors in a statistical model.

Identification of additional risk factors associated with take-home lead in caulkers’ homes may have been limited to some extent by the small sample size. Measurement of lead on the bodies, clothes, and vehicles of caulkers returning home after work would possibly have enabled the identification of the medium via which lead is carried into the home. Further studies addressing this issue in relation to the particular tasks and behaviors of caulkers during work time are currently being undertaken and will be the subject of a subsequent report.

CONCLUSIONS AND RECOMMENDATIONS

In conclusion, the study has revealed markedly higher lead loadings in the homes of boat-caulkers than in nearby homes of non-boatyard workers, a difference which is most reasonably explained by the caulkers’ take-home exposure pathway. Unfortunately, the factors identified in the multivariate analysis as independently associated with household lead loading in caulkers’ homes are all non-modifiable, and therefore do not in themselves point directly to any obvious means of intervention to reduce the problem. The apparent cumulative effect on household lead loading of years working as a caulker, however, suggests that caulkers’ homes may themselves be acting as an unrecognized depository of lead. Lead appearing on household surfaces at any one time may be the result of a redistribution of lead-laden dust within the house, a process in which the total amount of contaminant is augmented whenever the caulker returns home from work. Verification of this hypothesized scenario and clarification of the dynamics of dust-lead contamination in the household will require more detailed study, but may be necessary if a specific intervention to reduce the levels of household lead and consequent health hazards to the families of boat-caulkers is to be implemented. Other than that, it may be useful to recommend a general improvement in the hygiene behaviors of caulkers, such as showering and changing clothes at work before leaving for home, and a periodic thorough de-leading cleansing of the home. Ultimately, a non-toxic alternative to the leaded caulking material should be sought.

With more than 230 major boatyards throughout the country and over 6200 boatyard workers,⁹ among whom about 70% are estimated to be caulkers, the total population at risk of lead contamination associated with boat repair work is considerable. Observation of boatyards in several provinces in the south of Thailand has indicated similar conditions in each of these boatyards. The scale of this industry and the magnitude of lead contamination seen in this study indicate a need for regulations to be put in place and enforcement and educational efforts made to motivate both employers and workers to follow the necessary steps for preventing lead from leaving the boatyard and entering the home.^28–29 It is recommended that employers provide work clothing and facilities for showering and changing and a lead-free environment for eating lunch. Caulkers should shower and change clothes before coming home and remove shoes before entering their home to avoid tracking in lead from the ground of the boat repair yards. In addition, caulkers’ families should clean floors, windowsills, window frames, and other surfaces regularly, preferably using wet mopping or wet wiping, even in the bedroom, which might not be an obvious location for lead accumulation. Working clothes would best be laundered at the boatyard, or, if that is not possible, should be laundered separately from the rest of the family’s clothes.

Even if these recommendations are followed, the potential for take-home lead would remain. This indicates a need for routine monitoring of lead levels, not only of boat-caulkers themselves but also in their homes and in their family members, perhaps under a sentinel surveillance and tracking system, with appropriate action taken where contamination is found to exceed acceptable levels.