All responses by L.N. Matheson, PhD
Physical Demand Characteristics of Work Chart and Its Historical Background
In the original Handbook for Analyzing Jobs (1972), published by the United States Department of Labor (DOL), the strength demands of work were presented in terms of five levels. Each level carried a connotation of maximum number of pounds to be lifted on a frequent basis. However, frequent and infrequent were not defined. In A Guide to Job Analysis (1982), published by the Materials Development Center at Stout Vocational Rehabilitation Institute at the University of Wisconsin Stout, several major improvements of this job analysis technique were introduced, including modification and definition of these terms. Instead of infrequent, the term occasional was substituted, defined as “under 20 percent of the time.” Frequent was defined as “between 20 percent and 80 percent of the time.” Constant was defined as “over 88 percent of the time.”
This is the format incorporated into the Physical Demands Job Analysis (1981) and in the 1982 supplement to the fourth edition of the Dictionary of Occupational Titles. In the 1986 supplement to the Dictionary of Occupational Titles, a further revision of the rating structure for strength was introduced. This is presented in the current PDC chart. Occasionally is defined as up to 33% of the time. Frequently is defined as 34% to 66% of the time. Constantly is defined as more than 66% of the time. In my opinion these references use the term frequency incorrectly. If we think of frequency as a qualifier for work demands in terms of human performance, the appropriate reference is the number of repetitive cycles over a period of time before a rest break. For example, I would define frequency in terms of number of repetitions per minute for the duration of a task, also expressed in minutes.
Unfortunately, the DOL took a position that is relevant for people who are performing job analysis but not the best position for people who are evaluating human performance. I argue that the job analysts should follow the lead of the human performance experts rather than the other way around. I may have contributed to the problem by incorporating these terms and definitions into the PDC Chart, but I thought it necessary to follow the lead of the Department of Labor. Before the DOL had defined these terms in this manner, I had been using the definitions from the National Institute of Occupational Safety and Health (NIOSH).
I am looking for some definitions of frequent, occasional work. The classic time-related Department of Labor definitions are well known. I am aware that rate based definitions are also used; i.e., so many lifts per hour or quarter hour. What I need is a reference to substantiate the existence of those rate based definitions.
The use of adjectives to describe degrees of frequent is a poor convention. The actual use of frequency in human performance measurement is on a times-per-minute basis for handling or times-per-hour for lifting.
The United States Department of Labor was ill-advised when they adopted their convention for the DOT and related publications. It is used nowhere else in the world, nor does it have any basis in science or ergonomics or industrial psychology.
The RPC-1 ranges were based on research that demonstrated the likely scores in each sector for a given total score. It is one more way that you can internally validate the evaluee’s responses.
It all depends. If I get scores that do not fall within the range for each PDC level, I ask the evaluee about item scores that seem inconsistent in the affected sector. Often, I hear a very good explanation that allows me to accept the response as accurate. In these cases, I report that “internal consistency checks identified an abnormally low score in (name the sector). On further inquiry, it was established that Mr. Jones’ balance is also affected, along with right-side weakness as a consequence of hemiparesis due to the stroke.” This enriched information can only come from post-test inquiry. I never miss such opportunities!
On the other hand, if the evaluee gives me a response that suggests less than full effort, I report that “internal consistency checks identified an abnormally low score in (name the sector). Review of selected item responses indicates that Mr. Jones’s responses are suggestive of less than full effort/over-limiting due to fear/discouragement and depression…,” etc. Of course, I would be sure to “constructively confront” the evaluee so that we can make progress in the latter case. Often, a few days after such a confrontation, RPC scores are a lot higher and the evaluee appears to be making much more rapid progress in the rehab program.
Use the item responses, not just the score; the evaluee is able to tell you much more than the score indicates. The HFS test experience is a therapeutic encounter that can be very powerful!
The original Physical Demand Characteristics of Work chart was developed by Leonard N. Matheson at the Work Preparation Center at Rancho Los Amigos Hospital in 1975 based on the Handbook for Analyzing Jobs (U.S. Department of Labor, 1972). Dr. Matheson developed the PDC chart to bridge between the Department of Labor’s job analysis system and methods that were being used at that time to evaluate the Rancho clients’ ability to handle the strength and energy demands of work. Dr. Matheson termed this process Work Capacity Evaluation (Matheson, 1982).
The Department of Labor system for describing the strength demands of occupations had five levels ranging from Sedentary to Very Heavy. The first published PDC chart had eight levels. In clinical work, the Rancho team found that clients appreciated having intermediate levels as they progressed through the program. In response to this, the intermediate levels of Sedentary-Light, Light-Medium, and Medium-Heavy were added.
At the time the original PDC chart was developed, the Rancho team often estimated the safe MET level of work for each client with cardiac impairment. In order to bridge between the Department of Labor’s system and the Rancho system of categorizing safe work levels for cardiac patients, Dr. Matheson derived estimates of the MET range of each Physical Demand Characteristics level from studies of the metabolic and oxygen demands of various jobs and work tasks that had been conducted by others.
When he developed the WEST Standard Evaluation for lift capacity in 1979, the PDC Chart was used to interpret WSE lift capacity findings. Each evaluee’s performance was described in terms of PDC level. This system was used in the first functional capacity evaluation test battery, developed by Dr. Matheson in 1979.
During the late 1970s Dr. Matheson also worked as a consultant to facilitate the development of VOCOMP, the first computerized job matching system that used the Department of Labor job analysis system. In order to support the adoption of VOCOMP by the rehabilitation community, Dr. Matheson developed the Vocational Factors Profile (Matheson, 1981). The VFP included all of the Department of Labor’s factors for analyzing jobs, and a section for PDC level. The vocational factors profile was used to develop a Transferable Skills Analysis based on the occupations in the person’s work history using the Department of Labor job analysis profile for occupation. Because PDC level was so important, the computer logic of the VOCOMP system was adapted so that the match of the individual to the database was performed in a stepwise manner with PDC level as the primary focus. That is, matching based on interests, aptitudes, temperament, and other factors was taken within the context of PDC level.
In the early 1980s, others began to use the original eight-level PDC chart and Dr. Matheson’s five-level version of the PDC chart, usually with his permission and enthusiastic support. However, certain individuals presented the PDC chart as their own creation, and provided explanations of the reasoning about its use that have been, at times different from what Dr. Matheson intended. Because the plagiarizers did not attribute the PDC chart to Dr. Matheson, new users were unable to obtain source information, and made many errors. For example, the PDC chart is a method to quantify the strength and energy demands of work so that a bridge can be made to measures of physical capacity. Although it is true that a capacity for 30 pounds occasionally, 15 pounds frequently, and 10 pounds constantly describes a job at the Medium PDC level, when applied to a person’s abilities, it would indicate the capacity for Light work. The reason is that, in order to be certified as a worker at the Medium PDC level, the person must be able to handle up to and including 50 pounds occasionally, 25 pounds frequently, and 10 pounds constantly. If a person were to have the capacity of 30 pounds occasionally, 15 pounds frequently, and 10 pounds constantly, he or she would not be safe in all Medium PDC jobs. This reasoning is especially important when persons with cardiac impairment are considered. A person who can lift 50 pounds occasionally, 25 pounds frequently, and 10 pounds constantly, but is limited to 4.0 METS would not be cleared for Medium work; it would be beyond his or her safe cardiac capacity.
The terms frequent and infrequent that are used in the Dictionary of Occupational Titles and in the PDC Chart do not seem to conform to the usual meaning of those terms. How did that happen?
The issue of frequency of repetition of lifting tasks is useful to understand. In the original Handbook for Analyzing Jobs (U.S. Department of Labor, 1972), published by the United States Department of Labor (DOL), the strength demands of work were presented in terms of five levels. Each level specified the loads encountered in the job, including the maximum number of pounds to be lifted on a frequent basis. However, the term frequent was not defined.
The Materials Development Center at Stout Vocational Rehabilitation Institute at the University of Wisconsin Stout, made several major improvements of this job analysis technique, including modification and definition of these terms. Instead of infrequent, the term occasional was substituted, defined as under 20 percent of the time. Frequent was defined as between 20 percent and 80 percent of the time. Constant was defined as over 88 percent of the time. This is the format incorporated into the Physical Demands Job Analysis (Lytel & Botterbush, 1981) and in the next supplement to the fourth edition of the (U.S. Department of Labor, 1982).
In the 1986 supplement to the Dictionary of Occupational Titles (U.S. Department of Labor, 1986), a further revision of the rating structure for strength was introduced, described fully in the Revised Handbook for Analyzing Jobs (U.S. Department of Labor, 1991). This is presented in the current PDC chart. Occasionally is defined as up to 33% of the time. Frequently is defined as 34% to 66% of the time. Constantly is defined as more than 66% of the time.
Although Dr. Matheson disagreed with these definitions of the concept of frequency, he adopted them in order to remain consistent with the Department of Labor system. As he notes:
In my opinion these references use the concept of frequency incorrectly. If we think of frequency as a qualifier for work demands in terms of human performance, the appropriate reference is the number of repetitive cycles over a period of time before a rest break; the rate per unit time being pertinent because it produces a physiologic response that is significant. For example, I would define frequency in terms of number of repetitions per minute for the duration of a lifting task, also expressed in minutes. For a very light finger repetition task, I would define frequency in terms of the number of repetitions per second or 10 seconds for the duration of a fingering task, also expressed in seconds.
Unfortunately, the DOL took a position that may be correct for a simple job analysis system, but is not optimal for evaluation of human performance, which must take work physiology into consideration.
- Lytel, RB, & Botterbush, KF. (1981). Physical Demands Job Analysis: A New Approach. Menominee WI: Stout Vocational Rehabilitation Institute.
- Matheson, L. (1981). Vocational Factors Profile. Woodland Hills, CA: Innovative Software.
- Matheson, L. (1982). Work capacity evaluation for occupational therapists: Rehabilitation Institute of Southern California.
- U.S. Department of Labor. (1972). Handbook for analyzing jobs. Washington, DC: Manpower Administration.
- U.S. Department of Labor. (1982). Dictionary of occupational titles: Fourth edition supplement. Washington, DC.
- U.S. Department of Labor. (1986). Dictionary of occupational titles, classification of jobs (Vol. I and II). Washington, DC.
- U.S. Department of Labor. (1991). The revised handbook for analyzing jobs. Washington, DC: Author.
We performed the normative study with each subject standing at a table 29 inches high. We recommend a table between 27 inches and 31 inches. If you want to use a high shelf or low shelf or the floor to have a client perform an assembly task that is tied to his or her work demands, this would be acceptable, but the norms should only be used as a guideline.
If the purpose of the testing is to identify problems with executive dysfunction, the Situational Assessment is best administered without prior Work Sample test experience, but I find it acceptable to use the Work Sample on the first day of the evaluation, with the Situational Assessment on the second day in order to select the “just right challenge” for the latter, using Table 18 in the Professional Manual. If I select a sufficiently demanding level from this table, I find that I can easily identify executive dysfunction issues. Used in this manner, the evaluee’s performance on the work sample test can provide the evaluator with guidance in selecting the proper demand level for the Situational Assessment. If I have only one evaluation day and it is important to identify executive dysfunction if it is present, I use the Situational Assessment near the end of the evaluation day without a prior Work Sample, using other performance information from tests that have been completed to help me select a “just right challenge” level from Table 18. If executive dysfunction is not of concern and I am primarily interested in learning about the evaluee’s performance in terms of bi-manual dexterity and coordination, I use the Baby Bear Chair Work Sample test by itself. I often video-record clients’ performance for later analysis.
The Placement test is always performed before the Assembly test when the Work Sample is administered. The Placement test is never used with the Situational Assessment, nor is the parts mat used with the Situational Assessment because we want to require the client to apply whatever organization and problem-solving abilities are available. For example, there are extra parts in the Situational Assessment that are used as distractor parts. If the parts mats were used, the distractor parts would be immediately identified and would therefore not provide a challenge to the evaluee.
With regard to application of the normative data, because there is a learning effect with the assembly tests, the most valid norms for general application are the Baby Bear Chair Placement and Assembly tests. When I am working with clients on an extended basis, after collecting baseline data with the Baby Bear Chair, I can choose to do follow-up testing with any of the three chairs, at which time the appropriate normative comparisons can be made.
With regard to the time lapse between the first and second trials of Placement tests, 90 minutes was used for the normative data study. In a clinical setting, the only reason to wait to do a second trial of the Placement test would be to look at consistency of effort.
With regard to the Situational Assessment and integrating the audio materials, the reiteration of instructions occurs after the “Ready, Set, Go!” provided by the evaluator.
With regard to how much to intervene with cues and assistance, refer to Appendix G, Work Sample Employability Definitions. I generally do not intervene until it is clear that “Personnel Accommodation” is required. After the accommodation has been provided, it is possible to determine whether active assistance is necessary. My own preference is to allow the evaluee as much time as is necessary to complete the task, observing whether there is improvement in performance as the task proceeds; video-recording is helpful here. In the future, I would like to have our users’ group develop guidelines, perhaps using a nomogram or some other simple table to integrate the Employability Rating with the time to completion. I would welcome your help with this.
With regard to the pictures of the chairs, these were developed at the request of a clinician in California who wanted to provide additional visual cueing in order to compare use of the Labeled Photo with the Assembly Manual with the Assembled Model. I will leave it to the users’ group to provide more guidance with this and possible integration of the labeled photos on a formal basis in the future.
With regard to the DCAT Situational Assessment Calculators, in practice you usually will be able to collect data on all of the factors, but when this does not happen, you can use an average score, which the calculators provide. For example, with the Situational Assessment, it will happen that a client will focus on avoiding and identifying errors and perform those at a competitive level, obviating your ability to determine whether he or she can correct errors. This might happen if the Situational Assessment was too simple or the evaluee might have not been able to shift from one task to the other, sticking with the Telephone Message Recording task until it is completed and then returning to the Assembly task. As we mature in our sophistication of evaluation of executive dysfunction, we may resolve this issue, but I want to be careful to not over-structure these tasks so that the experience becomes insensitive.
I find it important to write under each number at least one behavioural example that supports my Employability Rating. If you are revising the forms, it would be helpful to have a small space under each number to provide behavioural example(s). I think this will force the evaluator to have rationale and support for each individual rating and improve reliability of the overall Employability Rating.
I fully endorse the discipline of at least one behavioral example for each rating. I actually video-record all of my clients and when I write the report, I have the video in front of me and also justified each of my ratings that indicate some difficulty with an example. I will revise the rating scales so that there is room for notation of an example behavior, and send the revised forms to you. I will post the revised forms in the CRC section of the EpicRehab.com website for others to use, as well.
I am not avoiding using the same example in two constructs and I would love to have a discussion with you about this because sometimes I find that the same behavior contains information that I can interpret as pertinent to more than one construct. Obviously, we need to develop a consensus on this, but I don’t want to adopt this as a formal decision rule unless we can actually develop a consensus. That will be for the months to come.
Yes, I score clients in this category based on the same behaviors as you, reflecting less than a “Competitive” rating but they do not require input from a person.
I have changed the term Environmental Accommodation in my most recent report to “Environmental Accommodation/Self Adjustment” as I find behaviours do not always reflect something related to the environment. For example, sometimes a client changes their method or makes a self adjustment, thus demonstrating learning as they proceed through the task. I have been scoring clients as requiring Environmental Accommodation if their behaviour does not support a “Competitive” rating, yet they did not require cueing or guidance (Personnel Accommodation) or Personal Assistance. I have changed the term Environmental Accommodation in my most recent report to “Environmental Accommodation/Self Adjustment” as I find behaviours do not always reflect something related to the environment. For example, sometimes a client changes their method or makes a self adjustment, thus demonstrating learning as they proceed through the task. Environmental Accommodation refers to adapting one’s physical environment to improve task performance. Self Adjustment refers to adapting one’s approach (or method of doing) during the task to improve performance (i.e. anticipating and attempting to prevent undesirable circumstances or problems from recurring or persisting).
I will have to give some more thought to “Self Adjustment” because I do not want to mark people down for learning, which would reflect executive function. What I mean to say is that if a person self-corrects or self-organizes, then there should be no penalty. If the person spontaneously adjusts to the demands of the task after having made an early error, I would not rate that as requiring “Environmental Accommodation”. An example of an Environmental Accommodation would be if the client were to need something like a pencil and paper to develop a checklist or need the parts mat as methods to promote organization. The question I ask myself as the client proceeds is, “would more external structure allow the person to proceed without having difficulty on this construct?” This reflects my earlier experiences in sheltered workshops with the highest functioning clients who, when the task was sufficiently organized, would not require input from others, but who could not work in the competitive labor market because the structure the person needed was somewhat more than would normally be available on a regular basis. The problem for us in these relatively brief situational assessments is that we need to have the client demonstrate first for us that he or she is at a less-than-competitive level before we suggest or make available these sorts of “props”. I think the first pass would be purely a professional judgment call; if confirmation is necessary, repeating the task or a parallel task at a later date with the prop available to find out whether this improved performance would be necessary.
Let’s talk more about the “Self Adjustment” option; perhaps we can have a debriefing after the conclusion of the task to ask the client, “If you were to do this again, are there any tools or guides that you would find useful?” If the person recognizes this, a self adjustment rating would be appropriate. If necessary, confirmation could be provided through a repeat of the task or a parallel task with the accommodation.
Upon further reflection, scoring (weighting) would not reflect a client showing a recurring or persist behaviour of this nature. I will mention the Assessment of Motor and Process Skills tool again here to illustrate this concept, as I don’t want to take credit for concepts that I did not develop. With the AMPS, clients are scored down further if the same types of behaviours persist throughout the task and also considers impact on task outcome (e.g. undesirable versus task breakdown).
Yes, such a decision process as the AMPS employs makes a lot of sense to me. We need to consider both the behavior and the effect of the behavior on task outcome, always asking ourselves how this would show up in the competitive work environment and whether the productivity that the person is displaying would be acceptable.
Other Clinical Questions
In response to this question, explain to the evaluee that “This is an indicator of your current level of ability to perform the task in the written description. You do not have to do the task exactly as the drawing.”
Notice that the tasks are not described in such a way as to completely specify the manner in which the person is to do the task. This ambiguity is intentional, and allows us to get at the person’s perception of how their functional impairments have produced disability.
In addition, some of the items are designed in such a way that even a person who has mid-cervical quadriplegia is able to perform the task at some level. The example I like to use is HFS Item No. 10: “Turn a lever knob to open a door.” It does not read, “Turn a lever knob to open a door with your right hand.” It is also a lever knob, which was selected over a round knob because a person with mid-cervical quadriplegia normally uses this type knob to open a door. So a person who has a severe right hand injury who indicates that they are “unable” is quite different from a person who has the same severe right hand injury who indicates that they are “able” or “slightly restricted.” This also gets at the person’s perception of how their functional impairments have produced disability.
The HFS approach is based on a subtle but important emphasis in how we look at our patients’ impairment, and its impact on their ability to work. Are we more interested in finding out the specifics of the impairment, or whether the impairment has convinced him/her that he/she is unable to work? I have taken the position that the latter is more important. If a person with a severe right hand injury says that he is “able” on Item No. 10, I know that the person is working around his impairment in order to minimize his disability. If, on the other hand, the person says that he is “unable” on Item No. 10, I know that he is taking the opportunity to communicate how disabling the impairment is, without regard to whether or not there are alternative approaches to open the door that would not impact him very much. This is one important indication of symptom magnification.
In practice, I find out if the person is prone to symptom magnification in several ways. One of the first indicators is if a person asks me during the administration of one of the Sorts, “Do I have to do this exactly as the picture shows?” This tells me that the person is thinking in terms of using the instrument to describe his impairment rather than how he is able to perform tasks using whatever strategy works for him. My response to this question is, “Respond like you normally would in your life. I want to find out how you are doing.”
Another way I find out if the person is prone to symptom magnification is by reviewing items afterwards with the person that appear to be more restricted than I expect and asking, “Is there another way that you can do this task that doesn’t restrict you so much?” Let me give you an example.
Mr. Smith responds “unable” to item No. 10, “Turn a lever knob to open a door.” He also responds “unable” to Item No. 2, “Sort a deck of playing cards.” and “unable” to item No. 13, “Drink from a bottle of juice.”
Now, I see sitting in front of me a person who is well fed, with his fly zipped up and his shirt buttoned and his shoes tied. Unless he had his butler dress him that morning, I know that he can use both hands and I surmise that he is able to drink from a bottle of juice and open a door. I may even have observed him doing these things in the clinic; using a paper cup to get water from the fountain or going in and out of the bathroom door or front door of the clinic. So, I challenge him by asking why he gave me “unable” on the item that is likely to be easiest, usually item No. 10. In the normal course of events, the patient will back down and admit that this is something that he is able to do with little or no limitation as long as he uses his unimpaired left-hand. I then take the next most easy item, often No. 13 and ask him same question. Again the patient will back down and admit that this is able to be done with little or no restriction. Item No. 2 is placed where it is because it actually is fairly difficult to sort a deck of playing cards unless you have two pretty functional hands. So next we get to that item in my debriefing with the patient, and he says that this is actually difficult because he has a severe right hand injury. I point out that we are interested in what he can do, rather than what he can’t do. We talk about this and all of the other patients in the clinic have had the same talk in one form or another, so the patient knows that we are different from the clinic down the block that wanted to know every little detail about what was WRONG and we are different from his attorney who wants him to point out every little detail about what he is UNABLE TO DO. We are interested in having him do the most with what he has left.
An arbitrator’s decision in an Ontario case revolving around job simulation functional testing supported the right of the company to do specifically job-related testing when an employee moves from one job to another, provided that the physical demands of the jobs are “significantly” different. However, the arbitrator also ruled that the employee does not have to participate in any prescreening, such as resting blood pressure, completing a PAR-Q, or even seeing his or her doctor. This leaves me and many others in a bit of a bind. As certified kinesiologists we always do a prescreening; doing so is taught in our courses, including the Matheson system.
The basic issue is to separate the need to maintain safety in the evaluation from using medical information to determine the person’s ability to work. These must be kept separate.
The standard of care throughout the United States is to screen for basic risk factors that can put a person in jeopardy in a test that is physically demanding. We routinely collect resting heart rate and resting blood pressure and have guidelines that allow us to objectively determine whether or not a person is able to be tested. To not collect such data would leave one open to malpractice liability in this country, notwithstanding the rules and procedures that various jurisdictions may have about prohibiting use of “medical tests” in preplacement screening and other such testing.
It depends on which functional capacity evaluation you’re using, or which test you’re referring to, and whether or not the diagnosis is pertinent to the factors that you’re testing. For example, with a test of grip strength, whether the person has a back injury or not would not be important, but whether the person has a hand injury would be important.
In every good FCE there is a resting HR limit or guideline; it depends on the FCE. The basic idea is that if you have a too-high resting HR, you are at greater risk for a “cardiovascular event” or other similar problem, such as a stroke, and extra care must be taken to keep you safe.
In FCEs that use the EPIC Lift Capacity test, the resting HR limit is 90 bpm. This is derived from the American College of Sports Medicine guidelines for “non-physician supervised tests”.
The ACSM publishes new editions of their guidelines every few years. Check on their website for the latest.
With my FCE testing as of late, I have come across several situations that involve the relationships between HR, blood pressure, and subjective pain ratings. I have had discussions with Workers’ Compensation Case Managers suggesting that I should report an FCE invalid due to the fact that the patient continually reports high levels of pain, despite HR and blood pressure being 'normal'. Is there any data out there that would help in this situation? To my knowledge, significant research does not exist that suggests that HR and BP should get to a certain level to consider pain and/or effort valid as an entire rating for the FCE. I know that during dynamic lifting, there is the 10 bpm cuttoff to empasize whether the lift is considered valid or invalid, but how does/can this play a role in circumstances where the patient reports that he cannot perform a lift and declines?
I wish it were so simple.
We need to look at many characteristics of the person and of his/her test data to make a good decision concerning validity. For instance, the chronicity of the person’s pain is important. Often, people with high levels of chronic pain have blunted HR responses to nociceptive stimuli (such as FCE demands), while, in contrast, people whose pain is of recent vintage experience a fear avoidance response in response to the same stimuli.
I look at all of the patient characteristics and also their performance data before offering an informed opinion.
Always remember that the amateurs want you and me to come up with a simple system even they can use when, in reality, it takes seasoned judgment to make good decisions in this area.
Question #33 on the EpicRehab health questionnaire is unclear to me. Some people find it difficult to find time to perform regular exercise. I am not sure whether the intent of the question is to include these people or whether it is to focus on people who have COPD or Fibromyalgia/chronic fatigue syndrome that can’t perform regular exercise.
Use this as an indicator of likely fitness.
The Cal-FCP integrates the ELC with a good carry test. Dr. Matheson’s ToolKit, which has the Cal-FCP video and protocol at no extra charge, can be purchased here. The relationship between IM strength and lift capacity is weak but significant for vertical pulls only, not for horizontal push or pull. For the latter, there is no dependable relationship.