TABLE 30: RFR FROM VIDEO DISPLAY TERMINALS AND PREGNANCY OUTCOMES
| Authors | Effects Sought or Examined | Exposure Modality | Effects Reported | Notes & Comments | |||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Weiss and Petersen (1979)
Petersen et al. (1980) | Electromagnetic emissions from eight representative VDTs at Bell Labs. | Frequency bands: 10 kHz-18 GHz, 350-600 nm (visible & UV), leakage of X-rays. | With 2 exceptions, measurable RFR levels occurred only at frequencies up to about 150 MHz; small signals of a few hundred MHz for the exceptions. For all units, the total RMS electric-field strengths were well below the most stringent exposure standard in the world.
UV measurements at approximate normal viewing distance were well below those recommended in NIOSH (1972)*. In the 11 CRT types examined for X-ray leakage, only 1 model showed levels above the background (but less than the highest level allowable from TV receivers). Repair of a faulty high voltage power supply corrected the problem. | The RF/microwave results were tabulated; each table included the total RMS electric-field strength for each unit.
*NIOSH (1972): "Criteria for a recommended standard for occupational exposure to ultraviolet radiation" Stuchly et al. (1983)
| Electric-field levels emitted by 86 VDTs (57 models) from 25 firms.
| Measurements were made in the range 15-150 kHz (stray fields from fly-back transformers) at the operator position, at the keyboard, and at the locations of maximum level at the screen surface, top, sides, and back of the various VDT models.
| The levels at the operator position were less than 1 V/m for 37 models and ranged from 1 to 5 V/m for the other 20 models. The highest level at any keyboard was 7 V/m; the levels exceeded 5 V/m for only 4 models. Also, the keyboard levels were below 1 V/m for 33 models and between 1 and 5 V/m for the other 20 models. The highest levels at the VDT surfaces were: below 5 V/m for 10 models, in the ranges 5-60 V/m for 22 models, 60-200 V/m for 10 models, and exceeded 200 V/m for 15 models. For the 3 models that had maxima at the screen, the level at 10 cm from the screen dropped to less than 70 V/m.
| The authors also tabulated the spatial-average and maximum SARs in a prolate spheroidal human model exposed to plane waves at 10, 20, 50, 100, and 200 kHz at 1 V/m, with the long axis parallel to the field. The highest spatial-average and maximum SARs were 0.000017 and 0.005 W/kg, both at 200 kHz, with the maxima at the surface of the model facing the source.
| Nurminen and Kurppa (1988)
| Kurppa et al. (1985) Pregnancy outcomes of Finnish women in occupations involving use of VDTs during pregnancy, based on occupational titles.
| Average exposure times per workday were grouped as follows: 4 hours or more (33 women), less than 4 hours but at least 1 hour (10 women), and less than 1 hour (21 women).
| Of 239 mothers with possible VDT usage, 8.4% had symptoms of threatened abortion as compared with 9.8% of 805 mothers in non-office work (nonsignificant). Also not significant was the difference in percentages of those who had bleeding or pain during pregnancy. The mothers in office and non-office work had similar proportions of preterm, term, and prolonged pregnancies, and there were no significant differences in neonate birth weights. The authors concluded that they had found no indication of work-related reproductive problems in offices or in VDT work in particular.
| Case and control mothers were interviewed about their work conditions and their various occupational and leisure-time activities, but were not asked specific questions about VDT work. Instead, occupational titles were used to recognize those in office work with possible VDT usage. The authors recognized that the smallness of the VDT group precluded more detailed analyses for possible differences between groups.
| Ericson and Källén (1986a)
| Effects of VDT usage on stillbirths, early neonatal deaths, birthweights, and malformations in Sweden during 1976-1977 and 1980-1981.
| Usage of VDTs in various occupations was characterized as high, medium, or low.
| The only statistically significant results were for birth weights <1500 g in the 1980-1981 medium-exposure group (O/E = 2.2, CI: 1.1-4.4), birth weights <2500 g in the 1976-1977 medium-exposure group (O/E = 1.5, CI: 1.2-1.8), and birth weights <2500 g in the 1976-1977 high-exposure group (O/E = 1.2, CI: 1.0-1.4). The authors noted that comparisons for the two time periods within each group suggested slight trends, both upward and downward, with greater VDT usage in the later time period.
| The authors remarked: "The absence of registrable effects could be due to a strong dilution of the material--that only a small proportion of the women actually worked with video screen equipment during pregnancy in the High group." This led them to do a case-control study [Ericson and Källén (1986b)].
| Ericson and Källén (1986b)
| Effects of VDT usage in 522 cases selected from the previous study for the period 1980-1981 versus 1,032 controls.
| VDT usage was high, medium, or low as in the prior study. Cases were selected from women who had had spontaneous abortions, still-births and infants dead after birth, infants with severe malformations or with birthweights <1500 g.
| Significant O/Es (exceeding 1.00, with 95% CIs above 1.00) were shown for birth defects, with a trend toward an increase in O/E with exposure duration. The O/Es for abortion were not significant, but the authors displayed a graph of O/E for that endpoint versus <5, 5-9, 10-14, 15-19, and 20+ hours per week of VDT use. The five O/Es were about 0.7, 1.1, 0.9, 1.7, and 1.3 (with no 95% CIs given), regarded by the authors as a tendency for a dose-related effect.
| The authors remarked that the effects were small, and they suggested several alternative explanations, with specific data indicative of possible selective non-response rates and for recall bias in those who did respond. They saw no specific pattern of birth-defect type among the infants of mothers with the most VDT use. They suggested that the strong association between stress, smoking, and VDT use rendered it very difficult to separate the effects, and that an attempt to do so for those who had high VDT use yielded a nonsignificant O/E.
| McDonald et al. (1986)
| Pregnancy outcomes after delivery or spontaneous abortion by 56,012 women in 11 Montreal hospitals who occupationally used VDUs [VDTs].
| No field exposure levels were given, but those with VDU usage for at least 15 hours per week were regarded as exposed.
| In a total of 8,805 current pregnancies comprising both users and nonusers of VDUs, there were birth defects in 311 pregnancies (3.5%); of 3,257 VDU users in the 8,805 pregnancies, there were birth defects in 108 pregnancies (3.3%). Similar results were shown for previous pregnancies. The authors concluded that in VDU users, there was no excess in the overall rate of defects or in any of the specific defect categories.
| Odds ratios for spontaneous abortions in current pregnancies versus the ranges of VDU usage (none, 1-6, 7-29, or >30 hours a week) were 0.89, 1.24, 1.25, and 1.12, but with a 90% confidence interval (CI) that spanned 1.00 for the greatest usage (>30 hours a week), rendering that odds ratio nonsignificant. In general, the authors found no association between VDU usage and the incidence of spontaneous abortion. Unclear is why they used 90% CIs instead of 95% CIs. Of the 56,012 then pregnant women (approximately 90% of all Montreal births), 25,418 were in paid employment for 30 or more hours per week at the start of pregnancy; of 48,608 women with prior pregnancies, 22,697 were similarly working. The authors remarked that one possible confounding factor was the rapid rise in VDU use during the study period (1982-1984), and therefore they used separate models for current and previous pregnancies; logistic regression was used to estimate the contributions of such possibly confounding factors as maternal age, number of prior pregnancies, and smoking on abortion risk.
| McDonald et al. (1988)
| The authors added 29 other cases to those considered in McDonald et al. (1986), and they analyzed incidences of stillbirth and low birthweight as well as of abortion and birth defects.
| VDU uses during prior and current pregnancies were shown as: 0, <15, and >15 hours per week for 6 broad occupational categories.
| None of the observed-to-expected ratios (O/Es) for spontaneous abortion in the previous pregnancies significantly exceeded 1.00. For current pregnancies, the only O/E that significantly exceeded 1.00 was in the clerical sector with >15 hours of usage per week: O/E=1.26, 90% CI: 1.10-1.44. Noteworthy, however, is that for the "managerial" occupational category, in which computer programmers were included, the O/E was 1.00 with a 90% CI of 0.69-1.42. Regarding the O/Es for stillbirths, no variation other than by chance was found, and the O/Es for preterm births (<37 weeks) and low birthweights (<2500 g) all had 90% CIs that spanned 1.00 (nonsignificant).
| In a summary of all of the adverse outcomes, the authors indicated that for spontaneous abortion in the (then) current pregnancies only was there any suggestion of possible higher risk from VDU usage, but they discounted that finding with the suspicion that it was due to recall bias. One can add that in the few results deemed significant, the use of 95% instead of CIs 90% probably would have rendered those results nonsignificant also.
| Goldhaber et al. (1988)
| Sought for women in the Kaiser Permanente Medical Care Program (KPMCP) who were pregnant during the period 15 September 1981 through 30 June 1982 was a possible association between pregnancy outcome and VDT use.
| Data were collected on work done during pregnancy and on VDT use, and the women were assigned occupation codes. VDT usage during the first trimester of pregnancy was subdivided into none, low (<5 hours a week), medium (5-20 hours a week), and high (>20 hours a week).
| Tabulated were 9,623 fetal outcomes in 9,564 pregnancies, of which 65% were normal births, 31% of births had various types of anomalies, and 4% had unknown outcomes. The final response rates were 82.7% for miscarriages, 87.8% for birth defects, and 87.8% for controls (live births with no birth defects or growth retardation). The odds ratios (ORs) for the various anomalous outcomes were not significant except for miscarriages in those with high VDT use, but the overall miscarriage OR for any VDT use versus no VDT use was not significant.
| Two ORs, in which the controls were nonworking women instead of working women with no VDT use, were significant for miscarriage: in administrative support/clerical women with high VDT use and for women in technical/sales with no VDT use, results that suggest an occupational effect unrelated to VDT use. The authors also suggested that although high VDT usage may raise the risk of miscarriages, ergonomic and other nonexposure factors may be contributors.
| Bryant and Love (1989)
| Sought was a possible association with VDT use by women in Calgary hospitals for spontaneous abortion. One set of controls were prenatal women before 25 weeks of gestation; the other controls were women in the same hospitals following birth of a normal infant. Using specific criteria, the authors matched 333 cases with postnatal controls and 314 cases with prenatal controls.
| Cases and controls were said to have had VDT usage if they used VDTs at any time: between 3 months before their last menstrual period (LMP) and 4 months post-LMP, between 3 months pre-LMP and their LMP, between their LMP and 2 months post-LMP, or between 2 months post-LMP and 4 months post-LMP.
| The ORs and 95% CIs for each such VDT usage period relative to no VDT usage indicated no significant differences between cases and either of the control groups. Further quantification of VDT usage revealed that fully a third of those who reported such VDT usage had extremely low levels of "exposure", such as at home (casual use) and/or to video games at four or more feet from the screen. An analysis in which those women were reclassified as "unexposed" also yielded no significant differences between the cases and controls. For the cases and controls who reported non-casual VDT usage, their usage was segmented into <2, 2-10, 11-20, and >21 hours a week. The highest OR was for the cases, relative to postnatal controls, with usage for less than 2 hours a week: 1.49 and a 95% CI of 0.89-2.49 (not significant). Also, the findings for VDT usage remained negative when income, education, cigarette use, and alcohol use were considered.
| The authors presented an interesting discussion on the possible effects of recall bias, suggesting that cases are more likely to report past exposures than controls and that the data for controls are likely to reflect usage under-reporting. The authors noted specifically that such under-reporting may have occurred among the casual-user postnatal controls in their study.
| Schnorr et al. (1991)
| Pregnancy outcomes of directory-assistance operators (who used VDTs) versus general telephone operators (who used units that had light-emitting diodes or neon glow tubes for displaying telephone numbers).
| Each group worked 8½-hour days (minus 1 hour for lunch and two 15-minute breaks). Electric and magnetic fields in the VLF (3 to 30 kHz) and ELF (30 to 300 Hz) ranges were measured from 24 VDTs each from two manufacturers, as were emissions of X-rays.
| The crude rates of spontaneous abortions were respectively 14.8% and 15.9% for VDT-exposed pregnancies and unexposed pregnancies. The spontaneous-abortion rates for those who had 1 to 25 hours and for those who had more than 25 hours of VDT use per week during the first trimester did not significantly differ from the rates for those with no VDT use. In addition, a multiple logistic-regression model showed no association between VDT use in the first trimester of pregnancy and spontaneous abortion incidence (OR=0.93; 95% CI: 0.63-1.38). The authors also found no evidence of significant recall bias.
| X-ray measurements showed no differences from background. A tabulation of VLF emissions at abdominal level (with the operators present) indicated means of 0.5 ± 1.68 (SD) V/m and 17.4 ± 1.74 mA/m for the VDTs from one manufacturer; the corresponding levels from the other manufacturer were lower. The mean ELF levels from the first type were 0.8 ± 3.61 V/m and 62.3 ± 1.59 mA/m, and were 57.7 ± 2.12 mA/m from the second type. The time-rate-of-change of VLF magnetic flux density ranged from 9.0 to 38.0 mT/s.
| |
This is a Department of Defense computer system for authorized use only. DoD computer systems may be monitored for all lawful purposes, including to ensure that their use is authorized, for management of the system, to facilitate against unauthorized access, and to verify security procedures, survivability, and operational security. Using this system constitutes consent to monitoring. All information, including personal information, placed on or sent over this system may be obtained during monitoring. Unauthorized use could result in criminal prosecution.
POC: AFRL/HEDM, (210)536-6816, DSN 240-6816