CNP Fit Testing Under OSHA's Updated Respiratory Protection Standard

(Reprinted with permission from: Respiratory Protection Update 9(2):1,13-18. ISSN 1048-6658, Respirator Support Services, 2028 Virts Lane, Jefferson, MD 21755, 1995.)

Clifton D. Crutchfield, Ph.D., CIH
cdcrutch@u.arizona.edu
Environmental and Occupational Health
University of Arizona
1435 N. Fremont Avenue
Tucson, AZ 85719

The controlled negative pressure (CNP) fit test method is one of three quantitative fit test (QNFT) methods included in OSHA’s recently published revision of its Respiratory Protection Standard.¹ In addition to recognizing CNP as an accepted QNFT protocol in the standard, OSHA also commented upon the conservative nature of CNP fit test results relative to those of aerosol-based fit test systems. Those comments speak to one of the principal advantages of CNP systems, which is their ability to directly measure respirator leakage.

A number of scientific studies support the fact that CNP systems consistently detect more respirator leakage during fit tests than aerosol systems.^2-9 CNP systems have also demonstrated the ability to measure known amounts of respirator leakage with efficiencies close to 100%.^6-7 By contrast, the ability of aerosol-based fit test systems to effectively measure respirator leakage has never been substantiated. The scientific literature contains much more information on aerosol related measurement biases than it has on aerosol measurement efficiency. Aerosol-based measurement efficiencies for known leaks introduced into respirators were well below 50% of the known leak values.^6-7 The ability to measure respirator leakage is the foremost requirement of a fit test system. In that regard, CNP represents the only proven technology included in the standard.

OSHA’s recognition of the capabilities of CNP technology was extremely gratifying, as was the number of favorable comments about CNP that were submitted to the official docket (OSHA Docket No. H049). Those comments came from a variety of sources including NIOSH. The ability to directly measure respirator leakage with unprecedented speed and accuracy has added much to our understanding of respirator fit and performance. As the capabilities of CNP technology become more widely known and accepted, we should be able to learn and do a great deal more. Some of the potential for improving respirator fit testing is mentioned in the preamble to the revised standard. We still have a ways to go to realize that potential, however. The following paragraphs spell out some minor corrections to the CNP portion of the standard, and offer some thoughts on where we need to go to realize the true potential of respirator fit testing.

Minor Corrections to Revised Standard

Appendix A, Part I, Paragraph C.4 of OSHA’s revised standard describes the protocol and requirements for conducting CNP fit tests. The sub-paragraphs listed below in italics contain two errors of a typographical nature that were included in Paragraph C.4. Corrections are indicated in brackets below each sub-paragraph.

C.4(a)(2) ) The CNP system defaults selected for test pressure shall be set at --1.5 mm of water (-0.58 inches of water)…

[The CNP default test pressure of –0.58 inches of water is equivalent to –15.0 mm of water, not –1.5 mm of water.]

C.4(a)(5) The test subject shall be trained to hold his or her breath for at least 20 seconds.

[Actual test time in the DNI Nevada FitTester 3000 is 8 seconds. Therefore a test subject need hold his/her breath for no more than 10 seconds]

Fit Test Exercise and Time Issues

A separate fit test protocol has been specified in the revised standard for CNP fit test systems. Specified fit test exercises and descriptions of how to accomplish them are shown in Table I. Both the number of exercises included in the CNP protocol and the time specified to accomplish them are surprising. In discussing its rationale for required fit test procedures, OSHA states (FR, p.1227):

The fit testing procedures and specific requirements in the QLFT and QNFT protocols in Sections B and C of part I reflect both the experience that has been gained in performing fit testing and the validation testing that was done initially in order for each method to be accepted by OSHA. The OSHA-accepted methods were evaluated by comparing their performance with that of another accepted fit test to demonstrate that each new method would reliably identify adequately fitting facepieces. The OSHA-accepted protocols reflect the specific procedures and equipment that were used in validation testing, and they must be followed to ensure minimum reproducibility. These elements in the OSHA protocols are not written in performance-oriented language, since any significant variation from the required protocols would invalidate the reliability testing that was performed initially to gain OSHA acceptance and would add uncertainty to the validity of fit test results.

In addressing the CNP protocol specifically, OSHA further states (FR, p.1234):

This fit test protocol is the same protocol allowed by OSHA under a compliance interpretation letter issued in 1994 and based on various studies on the performance of the CNP method conducted by its developer, Dr. Cliff Crutchfield (Exs. 71, 54-436). These studies reported results that were validated by comparing them to results from the existing aerosol fit test systems. The data showed that the fit factors measured with CNP are always lower than the fit factors measured with an aerosol QNFT. OSHA had reviewed these studies before issuing its compliance letter. OSHA believes that the CNP method, based on Dr. Crutchfield’s validation data, constitutes adequate support for the method’s reliability in rejecting bad fits.

The validation studies referred to above did in fact clearly demonstrate the consistently lower, more conservative nature of CNP fit factors versus aerosol-based fit factors measured during the same mask donning on a range of subjects. On numerous documented occasions,^4,9 aerosol-based systems have reported very high fit factors for masks with so much leakage in them (> 5 l/min) that a CNP system could not even establish a challenge pressure. However, the CNP test protocol that is currently specified in the revised standard is not the test protocol that was used during the CNP validation studies referred to by OSHA. Those validation studies are based on leakage measurements made at the six head positions described in steps 1, 3, 4, and 1 of Table I. They were not preceded by the 60 sec head movement exercises that are currently specified. The scientific basis for requiring 60 sec of head movement prior to making a CNP measurement in any specified head position is not known. For a respirator that has even a remote chance of passing a CNP fit test, I can discern no difference between moving the head for one sec versus 60 sec prior to making the leakage measurement. The net effect of requiring 60 seconds of head movement prior to making CNP leakage measurements is to make a CNP fit test take as long as currently specified aerosol-based fit tests. The tremendous speed advantages of CNP technology, that are gained by using a measurement principle that operates at the speed of sound, are essentially negated.

Several participants in the rule-making procedure, including myself, complained that the number and length of exercises required to be performed are excessive. OSHA addresses those complaints on page 1228 of the Federal Register. The only stated rationale that I can find for test duration refers to the way an isoamyl acetate test was performed when it was evaluated for inclusion in the lead standard in 1982. I have never seen a scientific rationale for the specified fit test duration of 60 sec.

I have also not seen a scientific rationale that supports the number and type of fit test exercises that are specified in the revised standard. None of the accepted quantitative fit test methods were validated using the eight exercises currently specified. Merely stating that Los Alamos National Laboratory used a given exercise in 1970 or that another exercise is not onerous does not provide a rationale for including an exercise in the specified test protocol. If we spend the time doing exercises, we should have some basis beyond supposition for believing that they provide significant data regarding the adequacy of respirator fit.

In my experience, the currently specified exercises consume a great deal of time while contributing little to the fit test. Although aerosol fit factors may appear to vary substantially as a result of exercises, actual respirator leakage typically does not. You can verify this point by converting individual aerosol exercise fit factors into estimates of respirator leakage and comparing the differences in leakage caused by the exercises. Converting fit factors into leakage estimates in ml/min involves changing measured fit factors into penetration (1/FF), then multiplying penetration by a typical fit test inspiratory flow rate of 30,000 ml/min. For comparison purposes, the equivalent leak rate for a failing half-mask fit test (FF = 100) would be 300 ml/min.

In the revised standard, OSHA jettisoned the requirement to do three fit tests and use the most conservative result to make decisions regarding mask fit acceptability. A number of participant complained that three repetitions of an eight-exercise protocol was too time consuming. I certainly agree. The three fit test protocol specified in OSHA substance-specific standards could take up to 45 min to complete. Many such fit tests were completed with a single mask donning, so that the primary benefit from requiring three repetitions was lost. We have found that mask donning has a much greater impact on respirator fit than do fit test exercises. It is difficult to be comfortable with the idea that we will be making decisions about respirator fit based on a single mask donning. With only one donning measured, we can do little more than hope that the worker can don the respirator close to the same way, and achieve approximately the same level of fit the next time the respirator is donned in the workplace.

The thing we know for certain is that doing a lot of fit test exercises takes a lot of time. No one has produced an analysis that shows that the results of the exercises are actually worth the time spent. Most statistical analyses I have seen indicate that they do not. If we feel compelled to do exercises in the belief that they make fit tests more closely resemble work, we should at least decide which ones may be most indicative of poor mask fit, and concentrate on using the most predictive. We conducted an extensive study involving 14 subjects each wearing two different sizes of the same type mask and completing three fit tests per day with each mask using both a CNP and an ambient aerosol fit test system for five consecutive days. A total of 840 fit tests were completed during the study. Study results from both fit test systems clearly show that donning affects respirator fit much more significantly than exercises do. The exercises described in the excerpt from the revised standard below were used with the CNP system. Of the exercises used, the bending exercise was by far the most predictive in identifying poor fitting masks. For the ambient aerosol system, the talking exercise produced the lowest fit factors. (You may want to reconsider that finding in light of the fact that, during the talking exercise when the mouth is being moved, the subject is exhaling, the respirator is positively pressurized, and any leakage would be out of – not into - the respirator.)

The results of the study described above strengthened my conviction that we need to refocus fit testing. Most of the currently specified protocol is spent trying to measure transient respirator leakage associated with dynamic head movements. If they occur (probable) and current fit test systems could actually measure them (improbable; it is difficult to imagine how a transient leak caused by dynamic movement could last for more than a few milliseconds), we would find that transient leaks affect relatively few of the total breaths taken in the workplace. Three more important things we should try to assess during a fit test are 1) how well does the selected respirator basically fit the subject, since basic fit affects every breath taken in the workplace, 2) does the basic fit remain fairly constant as the worker moves around, or does it shift to a different, unmeasured position due to movement (would affect all remaining breaths in the workplace), and 3) can the subject achieve the same basic fit the next time the respirator is donned. The proposed CNP fit test protocol described in the excerpt below is designed to assess those factors. We are grateful that OSHA included a more expedient means for revising the final standard.

In addition to the OSHA-accepted CNP fit test protocol, Dr. Crutchfield (Tr. 254) testified about a new fit test protocol for the CNP method. This new protocol is substantially different from the OSHA- accepted protocol, which requires the performance of test exercises followed by CNP measurements. The new protocol was also described in detail in a letter from Senator John McCain of Arizona on behalf of Dr. Crutchfield (Ex. 54-460). The new protocol submitted after the close of the post-hearing comment period is described as consisting of three exercises and two redonnings. The first exercise measured "fundamental respirator fit" with the head facing forward. The second exercise was a bending exercise, with the respirator parallel to the floor. The third exercise consisted of vigorously shaking the head from side-to-side for three seconds, followed by a "fundamental fit" measurement. The respirator user then is required to remove and redon the respirator twice, with "fundamental fit" measured after each redonning. This protocol results in five CNP measurements, from which a harmonic mean fit factor is calculated and used to make a pass-fail determination for the fit test. The information on the new protocol was not submitted to the rulemaking docket in time to allow an opportunity for public comment. OSHA, therefore, cannot include it in this final standard. Appendix A, Part II establishes procedures by which OSHA will approve new fit testing protocols after allowing opportunity for public comment. A proponent of the revised CNP fit test protocol may submit it for approval in accordance with Appendix A, Part II.

References

29 CFR 1910.139, Respiratory Protection, Federal Register 63:1152-1300, January 8, 1998.

Crutchfield, C., R. Murphy, and M. Van Ert: "A Comparison of Controlled Negative Pressure and Aerosol Quantitative Respirator Fit Test Systems Using Human Subjects." Am. Ind. Hyg. Assoc. J. 54(1):10-14, 1993.

Crutchfield, C., and Van Ert, M.: "An Examination of Issues Affecting the Current State of Quantitative Respirator Fit Testing." J. Int. Soc. Resp. Prot. 11(2):5-18, 1993.

Oestenstad, R.K. and J.B. Graffeo: "Determination of Respirator Fit by an Aerosol Method and a Negative Pressure Method". J. Int. Soc. Resp. Prot. 11(3):6-14, 1993.

Crutchfield, C., Ruiz, A., and Van Ert, M.: A Validation Study of Respirator Fit Testing by Controlled Negative Pressure. Appl. Occ. Environ. Hyg. 9(5):362-366, 1994.

Crutchfield, C.D., Park, D.L. Hensel, J.L., et al.: "Determinations of Known Respirator Leakage Using Controlled Negative Pressure and Ambient Aerosol QNFT Systems." Am. Ind. Hyg. Assoc. J. 56(1):16-23, 1995.

Crutchfield, C.D., Park, D.L. Hensel, J.L., et al.: "Determinations of Known Respirator Leakage Using Controlled Negative Pressure and Ambient Aerosol QNFT Systems." Am. Ind. Hyg. Assoc. J. 56(1):16-23, 1995.

Crutchfield, C.D., and D.L. Park: "Effect of Leak Location on Measured Respirator Fit." Am. Ind. Hyg. Assoc. J. 58(6):413-417, 1997.

Crutchfield, C.D.: "Measuring Gas Mask Fit and Donning Effectiveness", Technical Report NAWCTSD TR97-002, Naval Air Warfare Center, Jacksonville, FL, March, 1997.

Table I. CNP Test Exercises Specified in Revised OSHA Standard^a

Normal breathing. In a normal standing position, without talking, the subject shall breathe normally for 1 minute. After the normal breathing exercise, the subject needs to hold head straight ahead and hold his or her breath for 10 seconds during the test measurement.

Deep breathing. In a normal standing position, the subject shall breathe slowly and deeply for 1 minute, being careful not to hyperventilate. After the deep breathing exercise, the subject shall hold his or her head straight ahead and hold his or her breath for 10 seconds during test measurement.

Turning head side to side. Standing in place, the subject shall slowly turn his or her head from side to side between the extreme positions on each side for 1 minute. The head shall be held at each extreme momentarily so the subject can inhale at each side. After the turning head side to side exercise, the subject needs to hold head full left and hold his or her breath for 10 seconds during test measurement. Next, the subject needs to hold head full right and hold his or her breath for 10 seconds during test measurement.

Moving head up and down. Standing in place, the subject shall slowly move his or her head up and down for 1 minute. The subject shall be instructed to inhale in the up position (i.e., when looking toward the ceiling). After the moving head up and down exercise, the subject shall hold his or her head full up and hold his or her breath for 10 seconds during test measurement. Next, the subject shall hold his or her head full down and hold his or her breath for 10 seconds during test measurement.

Talking. The subject shall talk out loud slowly and loud enough so as to be heard clearly by the test conductor. The subject can read from a prepared text such as the Rainbow Passage, count backward from 100, or recite a memorized poem or song for 1 minute. After the talking exercise, the subject shall hold his or her head straight ahead and hold his or her breath for 10 seconds during the test measurement.

Grimace. The test subject shall grimace by smiling or frowning for 15 seconds.

Bending Over. The test subject shall bend at the waist as if he or she were to touch his or her toes for 1 minute. Jogging in place shall be substituted for this exercise in those test environments such as shroud-type QNFT units that prohibit bending at the waist. After the bending over exercise, the subject shall hold his or her head straight ahead and hold his or her breath for 10 seconds during the test measurement.

Normal Breathing. The test subject shall remove and re-don the respirator within a tne-minute period. Then, in a normal standing position, without talking, the subject shall breathe normally for 1 minute. After the normal breathing exercise, the subject shall hold his or her head straight ahead and hold his or her breath for 10 seconds during the test measurement.

^aAppendix A, Part I, Paragraph C.4.(b)