Improving the impact of quality control programs

All involved in the commercial production of poultry recognize the need for a quality control program, and possibly a quality control laboratory. However, when accountants or new owners review expenses within a feed mill or poultry operation, they sometimes question the justification for the expenses involved in feed analysis. In fact these expenses can be quite high when one considers salaries, analyses, reagents, equipment, etc.

Those familiar with the practical production of poultry feed instinctively recognize the need for a quality control program. Among other justifications, the program enables us to interpret variations in nutrient levels, detect feed problems before birds are negatively affected, evaluate and compare ingredient suppliers, detect adulteration, and provide valuable information for interpreting possible feed effects on field problems. All of this is of course secondary to providing the nutritionist with reliable data upon which to construct and maintain a feed formulation matrix.

However, as in any industrial process, the comparison of cost and benefit must occasionally be made. Having said that, it is extremely difficult to precisely quantify the benefits of a quality control program, simply because problems that are avoided never actually occur, and thus their impact cannot be quantified.   Perhaps a more interesting approach is to ask a very basic question: What would happen if we simply stopped investing in quality control? Certainly in the short term very little negative effect would occur. Immediately, there would be considerable savings on salaries and laboratory expenses. As nutrient levels in ingredient rarely change over the short term, there would possibly be little if any rapid effect on bird performance. However, as time goes by, the negative impact of no quality control will become extremely obvious. First the formulation matrixes used by the nutritionists would agree less and less with the actual nutrient composition of feed ingredients arriving at the mill. Second, suppliers would quickly learn that the company no longer checks for quality, and would begin to send shipments of marginal quality ingredients, knowing that these would not be detected. Finally, the inevitable decrease in productive efficiency could not be traced to feed problems, as the ability to quantify the feed quality would have been lost.

What is essential, however, is to make sure that any investment in quality control is well justified. That is, routinely evaluating many characteristics of each ingredient is probably a waste of money. Rather, for each ingredient, the most critical nutrients and or characteristics must be noted and these followed in a conscientious way, while unnecessary analyses should be minimized or eliminated completely. It is not unusual to find laboratories conducting many unnecessary analyses, and these pointlessly increase the overall cost of the quality control program. Basically, several parameters should be considered in deciding whether a certain analysis is justified. First, is variation important, or is it likely? The ash content of grain, for example is neither important nor is it likely to vary. Therefore, there is no reason to conduct this analysis. The second point one might consider is whether a modest variation in the nutrient quality of an ingredient might have importance in the field. If not, this also can be eliminated. Finally, can our evaluation of nutrient levels affect future variation? That is, if we find a large variation, can we bring this to the attention of the supplier with the probability that they will improve the quality of future shipments?

During more than thirty years the author has had numerous opportunities to visit companies, review their nutrition and quality control programs, and make suggestions as to the implementation and modification of such programs. In retrospect, the author realizes that some of his ideas and prospectives have been correct and useful, while others, quite frankly have been in error. It is the purpose of this presentation to discuss aspects of both, with the objective of inviting others to reflect on their own experiences.

1) Involve others in the company. This is extremely important, partially because it involves no additional costs. It is valuable that others in the company be trained to make specific observations which may help us in evaluating feed quality. Not only will this give us additional perspective, but will improve the quality of feedback we get from those in the field, hatchery, or the processing plant. Many times we receive complaints about feed quality, but these are very inspecific and difficult to interpret. Those in the field may comment on an apparent increase in litter moisture while at the processing plant there may be comments about excess body fat. Egg shell quality is also quiet variable. What is very important is that before the problem exists, we need to train people to properly document the important parameters so as to determine what is normal. Very likely, those in quality control will have to work with field personnel to teach them how best to obtain, record, and report data so that observations have a valuable meaning. To illustrate, which comment is more useful: (A) the birds have a lot of abdominal fat, or (B) during the past two and a half weeks there´s been almost linear increase in the percent in abdominal fat from 2.8% to 3.75%. With the more sophisticated report, we are far better to review our diets to see if the feed may have increased in energy.

2) Understand suppliers. It is obvious that we must determine the levels and variation of nutrients from different suppliers of feed ingredients. However, to understand a supplier means to understand where problems are most likely to occur. What does the supplier worry about when he wakes at 3:00 A.M.? If we understand the potential problems a supplier faces we are much better able to focus our analytical resources toward detecting potential problems. Often times, it is possible to work with a supplier to improve the quality of his product. A good quality control laboratory is in all probability superior to that of the supplier. As our goal is to establish long term reliable relationships, we must be able to relate to our suppliers in responsible and professional ways.

3) What do the analyses teach us? Simply accumulating analytical values is of limited importance unless we fully understand what these numbers imply. Determination of feed uniformity is an excellent example. Many believe that feed leaving the mixer should have a CV of less than 10%. However, CV percentages in excess of 25% may have little or no effect on bird performance. This is simply because additional mixing occurs at numerous places within the mill, during pelleting and crumbling, and the transfer from truck to bin to the interior feeder. As all these provide additional opportunities for feed mixing, are feed uniformity measurements irrelevant? No, but it useful to understand the value of our efforts. A high CV simply means there is something wrong with the mixer which should be fixed. It may have no major impact on bird performance.

4) Wasted efforts. As we know, every analysis costs money. Many laboratories have standard procedures by which each ingredient is routinely assayed for a number of parameters. Some of these may be important and others not. Certainly, determination of the entire proximate composition for each ingredient is terribly inefficient and useless. Rather, we should have a dynamic program by which first we establish those parameters of prime importance for each feed ingredient, and second, determine the frequency through which such analyses are made. If there is very little variation, the frequency of analysis can be reduced, and vice-versa. Periodically, the nutritionist and quality control director need to meet to discuss which analysis have been most valuable in either adjusting formulas or comparing suppliers. The program should never be static. Dynamic programs, in which the analysis performed vary according to need, are much less expensive and of far greater value to the nutritionist.

5) Be familiar with production. Occasionally, we find that some technicians have never actually seen a chicken. In fact, technicians are much more valuable to the company if they have an appreciation of the various aspects of production. It may be difficult to say in advance why this is so, but relating analytical values to field situations is clearly a good idea. The technicians who only understand milliliters of a reagent and changes in pH may not be able to appreciate the importance of what he or she is determining, and valuable perspective may be lost. Thus, technicians should periodically be given the opportunity to visit different parts of the company to really understand how chickens are produced.

6) In-Vivo Testing. Depending on the company, field research or in-vivo testing, sometimes falls under the responsibility of the quality control program. It is here where the author feels he has made major mistakes during his professional career. There is no question that properly controlled testing can be of enormous value to a poultry firm. However, unless those conducting the study are trained in research and can properly design and critically review results, such testing can be very detrimental. The author has seen numerous studies in progress which have been either poorly designed or are in facilities in which placement of pens, fans, and other variables probably affected experimental results more than the intended treatments. The author has seen numerous occasions in which decisions are made on the basis of a single experiment. If important, the experiment should be repeated. Both false positive and false negative results can lead to decisions which are incorrect and prove expensive for the company. We must bear in mind that the purpose of a poultry company is to produce massive quantities of chickens, not to conduct small scale tests and observations. Thus, the preparation of small volumes of several experimental diets may not done well by a large feed mill, while those in the production unit, lacking the professional perspective of a nutritionist, veterinarian, or quality control specialist, really have no particular interest in conducting a study. It is just one more detail, one more complication to add to a days´ work. In conclusion, while in-vivo data is extremely important and can be successfully used to improve a nutrition program, we must look very carefully at each study to be sure that the design and the conducting of the study are conducive to reliable results.

Summary: There is no question that quality control programs are expensive. While those not familiar with the feed industry may not recognize the value of such programs, those of us with years of experience know very well that it is impossible to maintain high quality feeds without analytical support. As the number of possible analyses far exceeds our ability to conduct or pay for them, a rigorous consideration of those analyses and how we utilize them, deserves periodic review.