Table 7 Individual studies reporting the association between occupational exposures and dementia

Peters et al., 2013 [60]

Aluminium (occupational exposure)

Survey information collected from miners living in Kalgoorlie, Western Australia. Data collected in 1961,’62,’74,’75 and 2000.

1894 ever underground gold miners linked with Western Australian Registrar General’s Mortality Database of whom 16 died with AD.

Retrospective cohort study: SMRs and Cox PH models. Exposure to aluminium dust ascertained by self-report.

Aluminium dust inhalation was associated with increased AD mortality (SMR, 95 % CI: 1.38, 0.69-2.75). There was also an association between inhaled aluminium dust and AD mortality in Cox models (HR, 95 % CI: 2.76, 0.88-8.82).

B

Salib et al., 1996 [61]

Aluminium (occupational exposure)

Individuals referred to a psychogeriatric unit in Warrington, UK

198 AD, 194 other dementia, 176 unmatched controls.

Cross-sectional study (case–control). Occupation based on self-report.

No association was found between working in the aluminium industry and AD (OR, 95 % CI: 0.98, 0.53-1.75). This was also the case for all dementias

B

Graves et al., 1998 [58]

Aluminium (occupational exposure)

Subsample of Kukull et al.’s cohort [71].

89 AD and 89 matched controls.

Cross-sectional study (case–control). Aluminium exposure rated blind based on occupational history.

There was an association between ever exposure to aluminium and AD (OR, 95 % CI: 1.46, 0.62-3.42) but not in models which took into account intensity, duration, or age at exposure.

B

Gun et al., 1997 [59]

Aluminium (occupational exposure)

Men and women recruited from Sydney hospitals 1986–1989.

170 probable or possible AD and 170 controls.

Cross-sectional study (case–control). Aluminium exposure was derived from an occupational interview.

Aluminium exposure was associated with a reduced odds ratio of AD (OR, 95 % CI: 0.33, 0.01-4.16). This study has very low statistical power because only four cases and controls were exposed. Probable and Possible AD cases considered together.

C

Tyas et al., 2001 [62]

Defoliants/fumigants

The Manitoba Study of Health and Aging – random sample from provincial health insurance records.

694 cognitively-intact older adults followed up for five years, during which time 36 developed probable or possible AD (clinically diagnosed).

Prospective longitudinal study. Exposures based on self-report.

Exposure to defoliants/fumigants was associated with an increased risk of AD (multivariable-adjusted RR, 95 % CI: 4.35, 1.05-17.90).

A

Koeman et al. 2015 [63]

Diesel motor exhaust (DME)

The Netherlands Cohort Study which consisted of 120,852 individuals aged 55–69 years in 1986.

682 men and 870 women who had died with non-VaD reported on their death certificates over 17 years follow up.

Case-cohort study: Cox PH models. Exposures based on self-report. Dementia status ascertained using death certification. Person-years were calculated from a randomly-selected sub-cohort.

Exposure to DME compared to no exposure was not associated with an increased risk of non-VaD mortality in men or women.

A

Koeman et al. 2015 [63]

Electromagnetic fields (ELF-MF and electric shocks)

The Netherlands Cohort Study which consisted of 120,852 individuals aged 55–69 years in 1986.

682 men and 870 women who had died with non-VaD reported on their death certificates over 17 years follow up.

Case-cohort study: Cox PH models. Exposures based on self-report. Dementia status ascertained using death certification. Person-years at risk were calculated from a randomly-selected sub-cohort.

Low or high exposure to ELF-MF compared to no exposure was not associated with an increased risk of non-VaD mortality in men (adjusted HR, 95 % CI 1.26, 1.01-1.57; 1.40, 0.92-2.14) but not in women. Conversely, an association was seen for electrical shocks for women (1.25, 0.85-1.84; 11.1, 3.84-32.2) but not men.

A

The hazard associated with cumulative ELF-MF exposure to showed no trend (P = 0.09).

Tyas et al., 2001 [62]

Excessive noise

The Manitoba Study of Health and Aging – random sample from provincial health insurance records.

694 cognitively-intact older adults followed up for five years, during which time 36 developed probable or possible AD (clinically diagnosed).

Prospective longitudinal study. Exposures based on self-report.

Occupational exposure to excessive noise was associated with a decreased risk of AD (multivariable-adjusted RR, 95 % CI: 0.12, 0.02-0.96). N.B. only one case was exposed to excessive noise.

B

McDowell et al., 1994 [66]

Glues and pesticides/fertilizers

Canadian Study of Health and Aging based in 10 provinces.

258 people clinically diagnosed with probable AD (less than three years since onset of symptoms) and 535 age-matched controls (stratified by study centre, community-/institution-dweller and clinically confirmed to be cognitively normal)

Cross-sectional study (case–control): logistic regression. Risk factor exposure gathered by questionnaire self-report.

Occupational exposure to glues and pesticides/fertilizers was associated with an increased risk of AD (multivariable-adjusted OR, 95%CI: glues 1.80, 0.99-3.27; pesticides/fertilizers 1.58, 0.81-3.10. Stratifying by education showed higher risk in those with less education.

B

Gauthier et al., 2001 [64]

Herbicides, insecticides and pesticides

1924 people aged ≥70 years old were screened and examined in the Saguenay–Lac Saint-Jean region (Québec, Canada).

122 people clinically diagnosed with AD and 122 age-matched controls (±2 years). 67 case–control pairs hade complete data and were included in the models.

Cross-sectional study (case–control): logistic regression. Exposure to pesticides derived from residential history and census data (1971–91) on herbicide and insecticide spraying in the area.

No increased risk of AD with exposure to herbicides, insecticides or pesticides (multivariable adjusted OR, 95 % CI: 1.07, 0.39-2.54); 1.62 (0.64-4.11); 0.97 (0.38-2.41).

B

Santibanez et al., 2007 [68]

Lead (occupational exposure)

Systematic review of studies linking occupational exposures and AD.

Twenty four studies: 3 cohort and 21 case–control studies.

Systematic review.

“For lead exposure there are no data supporting any association. All the studies are case–control studies, with a relatively low level of quality according to our classification.” (p. 730)

-

Koeman et al. 2015 [63]

Metals (occupational exposure)

The Netherlands Cohort Study which consisted of 120,852 individuals aged 55–69 years in 1986.

682 men and 870 women who had died with non-VaD reported on their death certificates over 17 years follow up.

Case-cohort study: Cox PH models. Exposures based on self-report. Dementia status ascertained using death certification. Person-years at risk were calculated from a randomly-selected sub-cohort.

Low or high exposure to metals compared to no exposure was associated with an increased risk of non-VaD mortality in men (adjusted HR, 95 % CI 1.21, 0.84-1.74; 1.35, 0.98-1.86) and women (4.55, 1.35-15.3; 1.78, 0.23-13.8).

A

The hazard associated with cumulative exposure to metals showed a significant trend (P = 0.01).

Hayden et al., 2010 [65]

Pesticides

The Cache County study (Utah, USA).

3,084 individuals aged ≥65 years without dementia followed up over 10 years during which time 500 people developed clinically-diagnosed dementia (344 AD).

Prospective cohort study: Cox PH models. Exposures based on self-report. Follow-up conducted at 3, 7 and 10 years from baseline.

Pesticide exposure was associated with an increased risk of dementia (adjusted HR, 95 % CI 1.38, 1.09-1.76) and AD (1.42, 1.06-1.91). A slightly greater risk associated with organophosphates for AD was reported.

A

Baldi et al. 2003 [70]

Pesticides

PAQUID cohort [110]. Male and female residents of Gironde, France aged ≥65 in 1987.

96 incident cases of AD from 1,507 contactable individuals between 5- and 10-year follow-up.

Prospective cohort study.

Occupational exposure to pesticides was associated with an increased risk of AD in men (adjusted RR, 95 % CI: 2.4, 1.0-5.6). This was not the case for other pesticide variables or in women.

A

Cumulative exposure to pesticides was calculated based on job history. AD was diagnosed by a neurologist.

Koeman et al. 2015 [63]

Pesticides

The Netherlands Cohort Study which consisted of 120,852 individuals aged 55–69 years in 1986.

682 men and 870 women who had died with non-VaD reported on their death certificates over 17 years follow up.

Case-cohort study: Cox PH models. Exposure based on self-report. Dementia status ascertained using death certification. Person-years at risk were calculated from a randomly-selected sub-cohort.

Exposure to any pesticides was not associated with an increased risk of non-VaD mortality in men or women. The same pattern was seen when disaggregating the effect of insecticides, herbicides, and fungicides in men. In women, high exposure to herbicides and fungicides was associated with an increased risk of non-VaD mortality (adjusted HR, 95 % CI 5.27, 1.30-21.4; 2.83, 0.87-9.16).

A

The hazard associated with cumulative exposure showed a significant trend for all pesticides and individual types (P = 0.01) but the trend was reversed with increasing exposure associated with a lower risk.

Povey et al., 2014 [67]

Pesticides

British farmers in the 1970s.

1,350 individuals born before 1958.

Retrospective cohort study: logistic regression. Exposure was based on self-reported levels of organophosphate exposure. Low dose chronic exposure was defined as handling organophosphate concentrate and acute exposure as having sought advice for pesticide poisoning. Dementia was identified by a screening questionnaire.

In people who had never sought advice for pesticide poisoning, handling pesticide concentrate was not associated with dementia. However, in those who had handled pesticide concentrate, seeking advice for pesticide poisoning was associated with an increased risk of dementia (adjusted OR, 95 % CI 4.27, 1.85-9.83).

B

Zaganas et al. 2013 [69]

Pesticides

Review of studies linking pesticide exposure and dementia.

Fourteen studies are reviewed (2 on cognitive performance, 1 mild cognitive dysfunction, 7 AD, 1 FTD, 2 VaD, and 1 PD).

Narrative review.

Of the 7 AD studies, 5 demonstrated an increased risk of AD (plus one which did so weakly) and one showed no association.

-

“Information from the literature [on VaD] is scant; however, occupational exposure to pesticides or fertilizers conferred a two-fold increased risk of developing vascular dementia in the Canadian Study of Health and Aging.” (p. 7).

“Due to the relative rarity of [FTD] at the population level, pesticide exposure has been studied as a contributing factor to FTD onset in relatively few studies and no association was found.” (p. 7).

Santibanez et al., 2007 [68]

Pesticides

Systematic review of studies linking occupational exposures and AD.

Twenty four studies: 3 cohort and 21 case–control studies.

Systematic review.

“For pesticides, research of greater quality and prospective design found increased and statistically significant associations with AD. … The two case control studies assessing risk associated with pesticide exposure and with GQI above the median found evidence of smaller and non-significant associations, supporting the hypothesis that potential biases might have affected these results, decreasing the associations towards the null.” (p. 729–31)

-

Tyas et al., 2001 [62]

Pesticides/fertilizers, inks/dyes, paints/stains/varnishes, gasoline/fuels/oils, solvents/degreasers, liquid plastics/rubbers, glues/adhesives, vibratory tools

The Manitoba Study of Health and Aging – random sample from provincial health insurance records.

694 cognitively-intact older adults followed up for five years, during which time 36 developed probable or possible AD (clinically diagnosed).

Prospective longitudinal study. Exposures based on self-report.

Other occupational exposures were not associated with an increased risk of AD (multivariable-adjusted RR, 95 % CI: pesticides/fertilizers 1.45, 0.57-3.68; inks/dyes 0.89, 0.28-2.81; paints/stains/varnishes 1.21, 0.46-3.21; gasoline/fuels/oils 0.79, 0.29-2.20; solvents/degreasers 0.88, 0.31-2.50; liquid plastics/rubbers 1.01, 0.12-8.38; glues/adhesives 1.41, 0.49-4.05; use of vibratory tools 1.03, 0.20-5.29).

A

Tyas et al., 2001 [62]

Radiation

The Manitoba Study of Health and Aging – random sample from provincial health insurance records.

694 cognitively-intact older adults followed up for five years, during which time 36 developed probable or possible AD (clinically diagnosed).

Prospective longitudinal study. Exposures based on self-report.

Occupational exposure to radiation was associated with an increased risk of AD (multivariable-adjusted RR, 95 % CI: 3.57, 0.38-33.38). N.B. only one case was exposed to radiation.

B

Kukull et al. 1995 [71]

Solvents

Individuals aged ≥60 years recruited to the Group Health Cooperative in Seattle-area clinics.

193 cases of AD and 243 unmatched controls.

Cross-sectional study (case–control): logistic regression. Risk factor exposure was based on self-report.

Solvent exposure was associated with an increased odds ratio of dementia in men (adjusted OR, 95 % CI: 6.3, 2.2-18.1) but not in women (0.6, 0.2-1.9). In men, a significant effect of exposure duration was observed.

B

Koeman et al. 2015 [63]

Solvents

The Netherlands Cohort Study which consisted of 120,852 individuals aged 55–69 years in 1986.

682 men and 870 women who had died with non-VaD reported on their death certificates over 17 years follow up.

Case-cohort study: Cox PH models. Exposures based on self-report. Dementia status ascertained using death certification. Person-years at risk were calculated from a randomly-selected sub-cohort.

High exposure to any solvents compared to no exposure was associated with an increased risk of non-VaD mortality in men (adjusted HR, 95 % CI 1.20, 0.90-1.61) but not women. High exposure to aromatic solvents was associated with increased non-VaD mortality in women (3.46, 0.59-20.5) but not men. High exposure to chlorinated solvents was associated with increased risk in men (1.33, 0.96-1.83) and women (2.08, 0.60-7.14), as was low exposure in men (1.25, 0.89-1.76).

A

The hazard associated with cumulative exposure to chlorinated solvents showed a significant trend (P = 0.01).

Santibanez et al., 2007 [68]

Solvents

Systematic review of studies linking occupational exposures and AD.

Twenty four studies: 3 cohort and 21 case–control studies.

Systematic review.

“For solvents, only two out of the 11 studies analysing this exposure found a significant association with AD.” (p. 730)

-