Document b5RMVXrkZK5QxMYeEqaLrqwdy

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES PUBLIC HEALTH SERVICE CENTERS FOR DISEASE CONTROL ATLANTA. GEORGIA 30333 OFFICIAL BUSINESS Z5 19898SNER17 7925 MR RONALD D SNEE DUPONT DE NEMOURS & CO ENGINEERING SVC 0IV 1007 MARKET ST IflilLMINGTON/ DE 19898 POSTAGE AND FES PAID U.S. DEPARTMENT OP HHS HHS 396 THIRD CLASS BLK. RT. N33776 1 d i 1 Results are presented for an interlaboratory proficiency study o.f blood lead m determinations. Samples were pooled from individuals occupationally exposed to lead. Performance of individual laboratories is compared over a two year period. With Increased emphasis in recent years on proficiency studies of this type, agreement between laboratories has been somewhat improved, but blood remains a difficult matrix in which to measure lead concentration. Interlaboratory comparison of blood lead determinations C. C. MAHER. Ph.D., D..M. ROETTGERS, M.S. and H. J. CONLON, M.p. Industrial toxicology Laboratory. West Allis Memorial Hospital, West Allis, Wl 53227 introduction Of the several indices used for monitoring environmental or occupational exposure to lead, blood lead (PbB) determination is the most widely applied, both for defining the extent of exposure and for following individuals in known exposures. Other measurements can be useful in either high or iow exposure situations to supply additional information. Free erythrocyte protoporphyrin (FEP) is a useful analyte in the screening of large numbers of children.11'21 It is fast, inexpensive, free from most contamination Where higher exposure is suspected, the amount of delta-aminolevulinic acid or coproporphyrin excreted in the urine (ALAU and CPU) are both useful. But CPU is not specific for lead poisoning and neither ALAU nor CPU are consistently elevated except in severe pl^mbism.'s, Therefore, despite its shortcomings elaborated elsewhere,**' PbB determination remains the most widely applicable and accepted method for monitoring lead exposure. problems and a sensitive indicator of moderate exposure to lead. It is most appropriate in establishing the existence of an exposure above that ofthe majority ofthe population. Once FEP is elevated it remains so for a long time. This is advantageous in identifying children who may have had an exposure to lead which has permanently or temporarily ceased prior to screening. However, it severely limits FEP as a practical method for following individuals on a regular basis. Inhibition of the enzyme delta-amino levulinic acid dehydratase (ALAD) is an even more sensitive marker for lead exposure. Several authors*have presented data which suggest PbB determinations have been carried out for many years.'7' Despite the long history, a regular complaint among analytical chemists has been the relatively poor interlaboratory agreement of all the available analytical methods. Several previous reportsl-u> of interlaboratory comparative studies and one monograph*131 have served to document the problem. In addition, because of the current emphasis on PbB by the various regulatory agencies, the problem is readily apparent to those interested.1 u"l<1 We report here a new interlaboratory proficiency study of PbB determinations which indicates that some improvement has opeured in recent years. that this enzyme is partiallyinhibitedeven by the level of lead exposure experienced by much of the urban population. A finding of either high FEP or high ALAD calls for further investigation and a PbB determination. Although several of the previous studies*9'11,1:1 did use human blood collected from individuals occupationally exposed to lead, others used spiked blood. Current proficiency programs run by the Center for Disease Control (CDC) and Copyright 1979. American Industrial Hygiene Association 230 Am. Ind. Hyi Assoc. J (40) March. J979 TEH 0531854 DUP050032350 >od lead 1 to lead, od. With jreement jit matrix i'< A i 3 ! >ected, the c acid or ne (ALAU "PU is not her ALAU except in despite its vre,'6' PbB >st widely monitoring ried out for y, a regular :s has been ueeinent of ds. Several' a b oratory aph,ut have n addition, PbB by the problem is d.':4'1M We proficiency h indicates d in recent udies,,',,'u> individuals liters used ocrams run . CDC) and `.'afcc 1979 .1 { j i i TABLE I Results from all Laboratories Blood Pool N Mean S.D. Intralaboratory Data N CV IV low high 16 48.6 16 56.9 5.9 16 5,1.. ,17 3.0% 7.2 V low medium higtl IB 40.7 ' 6.4 18 57.5 6.7 18 74.2 7.3 $ . .3.2 ' ' 8 1.7 8 2.8 VI low medium high 22 38.9 4.6 22 55,0 6.5 22 63.0 5.0 8 8 8 6.3 3.2 3.1 VII low 23 25.0 6.2 0 10.7 medium 23 41.0 5-5 9 4.6 high 23 67.9 6.5 9 4.3 VUI low high 37 36.4 37 70.1 7-8 9.0 9 9 4.8 2.7 IX low high 38 43.4 5.6 12 38 68.5 7.5 12 4.3 5.5 The results: of one participating laboratory have been omitted from this tabulation. They were continuously erratic. various states generally Use animal blood. While these samples are useful, the ideal sample remains human blood in which lead is incorporated into the red cell by "normal" processes. Thus, when this program was undertaken, it recognized that no suitable human blood standard was available that compared in quality to the various standard reference materials manufactured by the National Bureau of Standards. This laboratory performs approximately 30,000 PbB determinations annually on venous blood samples obtained primarily from workers iii the lead industry. The portion of each sample remaining after the determination is normally collected into six pools characterized by varying lead concentration. Cut points are set at 20,40, 60,80 ancl 100 micrograms oflead per deciliter of blood (jig/ dL). The pooling has been done in order to provide a suitable supply of quality control samples. In the'fall of 1974 the decision was made to offer portions of these pools to other laboratories for an interlaboratory comparison of PbB determinations. procedure Blood samples which arrive in this laboratory generally are already hemolyzed. If the sample has not for some reason been collected in a tube containing Triton X-100, then this hemolyzing agent is added immediately upon receipt of the sample. Hemolyzed blood remaining after analysis is pooled daily and frozen. This is accomplished by identifying one liter, acid washed, linear polyethylene bottles to serve for each pool. When needed for this study, each pool was thawed and filtered through a heavy layer of gauze using only gravity in order to avoid frothing. Each pool was then gently swirled in a large Erlenmeyer flask to provide homogeneity. After the pool was well mixed, it was divided into 10 roL portions in screw-top, acid washed, linear polyethylene vials and refrozen. Each pool was analyzed numerous times in this laboratory over a period of two weeks prior to shipment in order to assure homogeneity. Data summarizing the intralaboratory coefficient of variation (CV) for each pool are included in Table I. Finally, samples were mailed to the participating laboratories in insulated containers with enough dry ice to insure frozen delivery in two days. Generally, this is sufficient. However, each laboratory was asked to describe the condition of each sample when received, During the course of the study there has been no clear evidence to indicate sample degradation TABLE II Results from AIHA Approved Laboratories Blood Pool N Mean S.D, IV low high It 46.5 4.9 11 54.8 3.6 V low medium high 13 39.2 13 56.3 13 72.8 4.7 7.1 8.1 VI low medium high 15 38.8 15 64.3 15 62.3 4.3 6.6 5.1 VII low medium high 15 25.3 15 40.3 15 57.3 7.4 6.1 7.2 Vltl low high 20 38.5 8,9 20 69.5 9.4 IX low high 19 42.7 19 66.9 4.0 6.8 The results of one participating laboratory have been omitted from this tabulation. They were continuously erratic- American Industrial Hygiene Association JOURNAL (40) 3/79 231 TEH 0531855 DUP050032351 when the sample has been thawed upon arrival extraction, anodic stripping voltammetry and due to extended delivery time. emission spectrography. results This interiaboratory program has gone through nine rounds as of the end qf 1977. Some problems during the first year limited the usefulness of the available data. These included poor sample packaging in one round, inferior sample containers in another round and a small number of participants in the initial rounds. During the past two years these typical start-up problems have been eliminated. This report presents data obtained in the six rounds of 1976 and 1977, Table I is a summation of results from the participating laboratories. In three ofthe rounds (IV, VIII and IX) only high and low specimens were sent out. In the remaining rounds (V-VII.I) Where more than one or two laboratories had used a particular procedure, there was no significant bias which could be attributed to the method of analysis. Thus, from this data one must draw the conclusion that the three atomic absorption methods, the dithizone extraction and anodic stripping voltammetry are equally accurate. Precision of analysis depends more on the individual laboratory than the method of analysis. Precision will be discussed below. discussion The CDC, in its PbB proficiency programs, has established ad hoc guidelines for determining an acceptable range of reported results. This range ' is defined as 15% around the mean of PbB values greater than 40 pgJ dL, and + 6 /.ig/ dL for medium ranee^specimens accompanied the high PbB values of 40 y.g]dL or less. andJowOnesT --7;--------- Using these guidelines, it was possible to Although it was suspected that the data for all identify those laboratories which performed laboratories might fit a log-normal rather than a ^consistently well throughout their participation normal distribution, this was not the case. For In the program. Twelve laboratories were so all blood pools the geometric and arithmetic identified retrospectively using the following means..were virtually identical. criteria: Included m 'Table--H--are-the means and standard deviations of those laboratories in the study which have been accredited for lead determination by the American Industrial Hygiene Association (AIHA) as of February, 1. the laboratory must have participated in at least the last three rounds (one year). 2. the laboratory must have had zero or one determination outside the acceptable range in the last six rounds (two years). 1978- These laboratories must successfully determine lead in air filters sent regularly by AIHA. While the air filter is a significantly less complex matrix than blood, the accreditation does indicate proficiency in lead determination. Of the twelve selected laboratories, nine had perfect records and three had each missed one of fifteen determinations. This group is interesting for a number of reasons. First, these are laboratories which never Along with the data, the laboratories differed markedly from the mean of the larger identified the analytical method used in each population; in addition, they agreed very round. Table III lists the data subdivided by consistently with each other. If it is assumed that method. The most popular single procedure was there exist some laboratories which perform the Delves cup microanalytical method, PbB determinations more accurately and more although there is some indication that its precisely than others for whatever reason, then it popularity is waning. This was followed by the can be expected that the better laboratories extraction procedure using ammonium should be identified by this type of process. The pyrblHdine dithiocarbamate (APDC) and strength of this confidence rests on consistent methyl isdbutyl kefohe (MIBK)." These two performance over two years at various procedures combined with flameless atomic concentrations of lead. Second, if this process absorption made atomic absorption the does select out the better laboratories, then it overwhelming fayorite. Other methods used by should be possible to estimate the optimum, several laboratories include the classic dithizone interla boratory variation which can be expected ` ! * i , y pj ,232 ^ `. jf m Am. hid. Hyg. Assoc. J (40) Match. 1979 TEH 0531856 DUP050032352 imr&etry and oratories had lere was no ributed to the his data one three atomic le extraction y are equally ends more on ie method of :d below. rograms, has termining an s. This range tiean of PbB t6 fig/dL for > possible to h performed participation aries were so he following participated in ds (one year). idzeroorone se acceptable s (two years). ies. nine had missed one of a number of :s which never . of the larger agreed very assumed that tich perform cly and more eason, then it laboratories : process. The m consistent at various ; this process ories, then it :he optimum n be expected March. 1979 j j j j | j ! i l j j I i j j ! ! j ; i j ; ' f ; j' ! Blood Pool IV low high V low medium s high VJ low medium high VII low medium high VIII low high IX low high Blood Pool IV low high V low medium high VI low medium high Vli low medium high VIII low high IX low high TABLE 111 Results Subdivided by Method of Analysis Delves Cup APDCMiSK Flameless AA N Mean S.D. N Mean S.D, N Mean S.D. 9' 51 9 62 .S' " 7 5 5 45 53 3' 4 10 ' 43 10 60 10 76 6 5 37 5 5 51 5 5 71 3 2 45 1 4 2 62 16 9 2 83 10 11 41 11 57 11 64 .4 7 35 6 7 6t 6 7 62 2 1 33 3 1 38 3 1 51 * - - 13 24 4 7 24 13 43 13 7 40 13 55 9 7 51 6 2 23 s 2 43 2 6 6 2 58 12 12 34 12 66 3 11 5 11 35 14 72 18 8 41 8 70 12 21 10 43 10 65 3 12 5 12 43 69 6 9 45* 9 8 9 70* 11 Dithizone Anodie Stripping Emission Spec. N Mean S.D. N Mean S.D. N Mean S.D. 1 46 1 57 1 43 1 60 1 81 1 40 * - 1 38 1 55 - 1 52 J 67 - 1 61 1 29 1 51 1 63 1 21 - 1 38 - 1 62 1 58 - 1 54 , .. 1 45 ' . - 1 68 - - 1 76 - 1 46 * 1 70 - 1 70 - - * 1 46 - 1 52 1 72 - 4 40 12 4 39 6 1 52 4 es 9 4 72 22 1 BO * - 4 43 4 71 4 4 43 9 4 70 3 5 'Results from one laboratory (low " 200. high = 234) omitted to avoid total distortion of data. at this time for PbB determination. Third, it might be interesting to identify the methods of analysis used by these successful laboratories. Table IV contains means and standard deviations for the twelve selected laboratories. A set of CDC acceptable ranges was also calculated based only on these twelve laboratories. Two of these did not participate in rounds IV and V. In addition, the three data points which were not within the CDC acceptance ranges were not included in this tabulation. Thus, Table IV probably contains the best available estimates of the concentrations for these fifteen samples. Although in comparing Tables I and IV the means are not very different, the variance (equal to the square ofthe standard deviation) are much smaller for the data in Table IV. The average coefficient of variation (CV) in Table IV is about Seven percent. This can be contrasted with a CV of an unselected group of laboratories of about fifteen percent. In comparing the means of Tables I and IV, thirteen of fifteen means (X2=4.03, P < 0.05) are higher in Table 1. This suggests a common bias present in the results from the laboratories which perform less well. Contamination at the laboratory or during the determination and old standards which may have lost a portion of the analyte to the container wails are likely explanations. In examining all results from all American Industrial Hygiene Association JOURNAL (401 3/79 233 TEH 0531857 DUP050032353 TABLE JV Results from Twelve Selected Laboratories Blood Pool N Mean s. CV Revised CDC Range IV low high 10 472 10 57.Q 2.5 3.6 5.2% 6-3 40.1 - 54,3 48.5 - 6,5.5 V low medium . high 10 39.8 10 66.1 9* 75.7 2.9 5.1 2.9 7.2 9.0 3.8 33.8 * 45.8 47.7 - 64.6 64.4 - 87.0 VI low medium high 11* 36.7 12 53.2 12 62 1 3.2 3.0 3.5 8.7 5.6 5.6 30.7 -42.7 45.3 -61.2 51.8-70.4 VII low medium high 12 23.1 3.1 13.4 12 40.2 2.9 72 12 673 3.7 6.4 17.1 - 29.1 34.2 - 46.2 48 7 - 65.9 VIII low high 12 34 5 12 678 3:0 3.3 7.4 4.8 28.5 - 40.5 57.7 - 77.9 IX low high 11* 42.1 12 66.6 3.1 4.3 8.8 6.4 35.8 - 48.4 56*6 - 76.6 ^Analyses in which one laboratory was outside the CDQ acceptance range* laboratories, high values are indeed more common than low values. However, extremely low reports did occur as well. Of these twelve laboratories, six regularly use the Delves cup procedure, five use the APDCMIBK extraction and one uses anodic stripping voltammetry. Most laboratories using flameless atomic absorption only participated in rounds VIII and DC and therefore were not eligible for inclusion in this selection. Figures 1 through 4 chart the performance of four participating laboratories relative to the results of the twelve laboratories in Table IV. Data have been normalized to the CDC range using the formula: (individual result) -- (mean of 12 selected labsl 6; or 15% of mean of 12 selected labs Laboratory 1 is a highly respected institution which has performed quite adequately in all but one round. This one inaccurate performance emphasizes the importanceof the extensive daily quality control procedures necessary to avoid occasional aberrant results. Laboratories2 and 3 were regularly inaccurate, one high and the other low. It appears that their performances have been improving recently, perhaps as a result of participation in this and other proficiency programs. Note that laboratory 3 has switched from regularly low results for all levels of Pb B t o disturbingly high results for the low controls only. Laboratory 4 is interesting because of its performance in relation to the method of analysis used. In rounds V through VII the Delves cup procedure was used. After initial problems, good results were obtained in round VII. Then the method of analysis was changed to flameless atomic absorption, and performance degenerated in round VIII only to improve somewhat, after a second apparent learning period, in round IX, Figures 2-4 suggest the possibility of a learning process which has takenplace as a result of proficiency testing. Among the twelve laboratories in Table IV one would not expect to find the effects of such a learning process; these laboratories have already achieved accurate and precise PbB analyses. Indeed the CV does not improve significantly with time. Among other laboratories there does appear to be a tendency, not statistically significant, to improve in ability to measure PbB with time. Of the several previous interlaboratory evaluations of PbB published, onef9> was perhaps the most extensive and alarming. This study involved twelve blood samples in two rounds. The authors suggested three alternative methods for evaluating the performance of participating laboratories. The most attractive of the three was the regression of laboratory results on the true (substitute mean) values. Using this procedure they reported very disappointing results. Of 43 laboratories 234 Am. Ind. Hyg. Assoc. 3 (AO) March. 1979 TEH 0531858 DUP050032354 :!.o i.s- because of its e method of ough VII the . After initial ined in round vas changed to 1 performance y to improve irent learning ssibility of a )laceasaresult g the twelve d not expect to : process; these d accurate and i CV does not Among other be a tendency, ?rove in ability :eriaboratory d, one'9' was alarming. This ;mples in two iree alternative ;rformance of nost attractive of laboratory mean) values, -ported very laboratories ') March. 1979 Figure 1 --This laboratory, while normally accurate, had Figure 3 - This laboratory changed from regularly low one bad round. results to somewhat more adequate performance. Figure 2 - This laboratory changed from regularly high results to adequate performance. Figure 4 -- After mastering one procedure, this laboratory switched its procedure with disappointing results. participating in both rounds, only 18 had acceptable results in both (26 in the first and 36 in the second). Furthermore, of these 18 the slopes of the regression lines were significantly different in several cases for the two rounds. Acceptability was based arbitrarily on the correlation coefficient (r) ^ 0.95 and the standard error of the estimate (S,xX 15. Most of the laboratories which were not judged acceptable were characterized as having poor to negative correlation coefficients. In order to compare the two studies, r and Sy, have been calculated for all 24 laboratories which participated in at least three rounds ofthe present study. See the appendix for the formulas used. Table V contains the results. Despite three more blood samples and the fact that samples from six different rounds were calculated together, each of which might be expected to introduce extra variability, similar percentages of laboratories (46% vs. 42%) meet the same criteria of acceptability in each study. In addition, all but one of 24 laboratories in the present study have a very high (> +0.89) value for t . The range of the slope of the regression line was 0,78 - 1.34 in the present study. The earlier American Industrial Hygiene Association JOURNAL (40) 3/79. 235 TEH 0531859 DUP050032355 TABLE V Regression Results of Laboratory Data vs. Mean Values Laboratory N Slope Correlation Coefficient S,, 1 15 1.01 2 15 1.14 3 15 1.04 4 10 0.96 S 15 1.03 6 16 0.87 7 13 Ml 8 15 1.34 9 15 i.oi . 10 15 0.97 11 15 1.00 12 15 0.90 13 15 1.02 14 IS 0.90 15 15 1.11 16 15 1.09 17 15 0-80 18 10 0.99 19 10 0.94 20 to 1.04 21 7 1.08 22 13 0.78 23 13 1.09 24 11 097 099 0.99 0.99 0.98 0.94 0.92 0.90 0.31 0,99 0:98 0.99 0.99 0.95 032 0:99 0,94 0.91 1.00 0,95 0.98 0.90 0.89 0.96 0-95 5-5 7.9 4.7 11.5 28.9 29.2 74.7 3652. 3.2 10.4 4.1 4.5 22.9 29,5 6,5 33.0 27.4 0.3 Z4.1 24.0 91.8 34.0 21.3 11.9 study reported 14 of 36 slopes front their acceptable laboratories (two per laboratory) to be outside this range. conclusions During the last decade there has been some improvement in the ability of different laboratories to agree with each other on PbB determinations. With current methodologies, the optimum CV which can be expected for PbB is about 7%. It Is reasonable to expect at least twice this value in any broad survey of laboratories. It is more common to receive a falsely elevated PbB result, due most probably to a variety of causes including contamination at the sites of collection and analysis, or improperly maintained standards. However, artifically low results are occasionally reported as well. There are some indications in this data set that participation in proficiency programs of this type does aid a laboratory in improving accuracy and precision of analysis. appendix r= b l(x ~ x)2 Sfy - y)` s _ 2(y-y)z~b2X(x-x); Sr' n-2 where:. _ xy -- (SxZy)/n ._ 2(x2) - (2x)7n x = average of "true" concentra tions of all blood samples y -- average of observed concen trations of all blood samples NB: the formula for r is not that reported by Keppler, et at.; however, it is the one which must be used to obtain the results in their Table XI. references 1. Piomelli, S.: A Micromethod for Free Erythrocyte Porphyrins: the FEP test. J. Lab. Clin. Med. 37:932940(1973). 2. Lamola. A. A.. M. Joselow and T. Yamane: Zinc Protoporphyrin (ZPP): a Simple, Sensitive. Fluorometric Screening test for Lead Poisoning. Clin. Chem. 21:93-97 (1975). 3. Roels. H., J. P. Buchet. R. Lauwerys, G. Hubermont, P. Bruaux, F. Claeys-Thoreau. A, Lafontaine and J. Van Overschelde: Impact of Air Pollution by Lead on the Heme Biosynthesis Pathway in School-age Children. Arch, environ. Hith. 37:310-316 (1976). 4. Hernberg, S. and J. Nikkanen: Enzyme Inhibition by Lead under Normal Urban Conditions. Lancet 1970 /7/63-64 (1970). 236 Am. tmi, Hyg. Assoc. J (40i March. 1979 TEH 0531860 DUP050032356 zJt i_n n te" concentrajd samples srved conceulood samples t that reported vever, it is the d to obtain the si. Free Erythrocyte tin. Med. 81:332- T. Yamane: Zinc ole. Sensitive, Lead Poisoning. Lauwerys, G, eys-Thoreau, A, de: Impact of Air ns Biosynthesis Arch. Environ. nzyme Inhibition ,nditions. Lancet h March, 1979 5. Goysr, R. A. and P. Munhak: Toxicology of Trace Elements, pp. 41-77. Halsted-Witey, New York 11977). 6. Beritic. T., D. Prpic-Majic, V. Karacic and S. Telismon: ALAD/EP Ratio as a Measure of Lead Toxicity. J. Occ. Med. 1$:661-557 (1977). 7. Meillere, G.: J. Pharm. Chim. 1:26-27 abstractedin Chem. Abstr. 7:2628 (1913). 8. Keenan, R. G., D. H. Byers, B. E. SalUman and F. L. -Hyslop: The 'USPHS' Method for Determining Lead in Air and in Biological Materials. Am. Ind. Hyg. Assoc. J. 24:481-491 (1963). 9. Keppler, J. F,, M. E. Maxfield, W. D. Moss, G. Tietjen and A. L. Linch: Interlaboratory Evaluation of the Reliability of Blood Lead Analyses. Am, Ind. Hyg. Assoc. J. 31:412-429 (1970). 10. Donovan, D. T,, V. M. Vought and A. B.Rakow: Laboratories which Conduct Lead Analyses on Biologic Specimens. Arch. Environ. Hlth. 23:1 11113(1971). 11. Berlin, A., P. Dei Castilho end J. Smeets: Environmental Health Aspects of Lead. pp. 1033- 1046. Commission of European Communites. Directorate General for Dissemination'of Knowledge, Centerfor Information and Documenta tion, Luxembourg (1973). 12. Browne, R. C.. R. W. Ellis and D. Weightman: Interlaboratory Variation in Measurement of Blood Lead Levels. Lancet 1974 (2J;1112-1113 (1974). 13. Pierce, J. O.. S. R. Koirtyohann, T. H. Clevenger and F, E. Lic.hte: The Determination of Lead in Blood: A Review and Critique ofthe State a!the Art. 1975. 76 pgs. International Lead Zinc Research Organization, Inc., New York (1976). 14. Anonymous: Subclinical Lead Poisoning. Lancet 1973 (If.87(1973). 15. Eckert, A. C.: Specie! Publication 422. pp. 275282. The National Bureau of Standards, Gaithersburg, Maryland (1976).. 16. Piscator, M,: Effects end Dose-Response Relationships of Toxic Metefs. pp. 172-1,83, Elsevier, Amsterdam (1976). Accepted September 13. 1978 Refresher course schedule and registration procedure changes . Two important modifications pertaining to the Refresher Courses, to be held in conjunction with the AIH Conference in Chicago, May 27-June 1, 1979, have been made. Courses will be presented Sunday afternoon. May 2:7 from 1 to 5 P.M. and Monday afternoon. May 28 from 1 to 5 P.M. (instead of Sunday afternoon and Monday morning, as in the past and previously announced for 1979). Attendance at any and all refresher course sessions must be signed up in advance, on a preregistration basis. There will be NO atconference registration for refresher courses. Be sure to plan ahead and pre-register. Refresher course monitor volunteers needed Volunteers are needed to serve as monitors and assisi: instructors conducting refresher courses at the 1979 AIH Conference in Chicago. In addition to the satisfaction of being of service, a monitor benefit is free admission to the course monitored. NOTE that theschedule has been changedfrom previous years and as initially announced for 1979. Monitors will serve on Sunday afternoon. May 27, from 1 to 5 P.M. and Monday afternoon. May 28, from 1 to 5 P.M. (NOT Monday Morning). To volunteer, contact Sam Kaplan, Refresher Course Committee Chairman, Enviro-Control, Ine., One Central Plaza, 11300 Rockville Pike, Rockville, MD 20852 (301) 468-2500, American Industrial Hygrene Associabon JOURNAL (40) 3779 237 TEH 0531861 DUP050032357