Gordon King, Animal Science, University of Guelph

Sound livestock management should always includes some method to monitor continuously the reproductive performance of any herd or flock. Comprehensive procedures for assessing efficiency should measure the total number of viable offspring produced by all mature females in the breeding unit over a suitable time period. Since most domesticated species have interparturition intervals of less than twelve months, the calculations usually made to cover a yearly basis. A number of procedures exist to tabulate and summarize breeding data with microcomputers now the preferred tool for updating, analyzing and storing results. Regardless of the method used, it will only provide useful information if accurate records are maintained, the figures are calculated properly, the results are reviewed regularly to determine whether goals are being met, and the outcome of these assessments used as a basis for management decisions.

Producers occasionally omit primiparous females or animals that are kept for long periods but fail to get pregnant. Such summaries, based only on those animals that reproduce successfully, produce inflated results. Thus, performance figures calculated with such omissions might look impressive but do not illustrate the true reproductive performance.

Reproductive Efficiency in Dairy Herds

The fertility measures used most commonly in dairy herds are: Services per Pregnancy (sometimes called Services per Conception or Conception Rate), Calving Interval and Days Open; calculated as indicated in the accompanying figure. Each of these may provide useful information on reproductive performance but must be considered in context because none summarize the complete picture. Services per Pregnancy, when calculated in the usual manner, is deficient since it provides no information on the proportion of females that fail to get pregnant at all, or any indication of how early or late conception occurs in relation to the previous parturition. The Services per Pregnancy figure can be improved by using a somewhat more inclusive method of calculation. This involves totaling all of the inseminations or natural services to cows that conceived in the previous year plus all of those given to infertile cows that were eventually culled for reproductive failure. The resulting figure, representing the total services to both fertile and infertile cows, would be divided by the actual number of cows that were mated. There would be little difference between Services per Pregnancy figures calculated in either way in herds with good reproductive performance. However, differences would become substantially greater in proportion to the number of reproductive problems. Knowing the total number of inseminations required per pregnancy is also useful information when expensive semen must be purchased by the insemination dose.

Like the conventional Services per Pregnancy figures, Calving Intervals can only be calculated for those females that have successive parturitions. Thus, it excludes all of the heifers and any cows that may have been mated numerous times but failed to conceive. Except in problem herds, the numbers of infertile or sterile animals should always be low, but this proportion must always be determined before accepting apparently satisfactory values for Calving Interval or Services per Pregnancy as reliable measures of performance.

When calculating Days Open, every member of the herd that the farmer hoped would get pregnant must be included to obtain a true picture of breeding efficiency. The actual number of days from the previous calving to current conception is entered for all cows successfully re-mated. For re-mated cows that failed to conceive, or for animals that should have been re-mated but were never detected, the total days from calving until culling is used for those which have left the herd and appropriate figures added for each open female still remaining on inventory. If desired, even heifers can be included by calculating the number of days between selection as breeding stock to remain in the herd and conception. Provided that the Days Open calculation includes all mature members of the herd that the farmer hoped would get pregnant, it is an accurate indicator of female reproductive performance. A theoretical example of Days Open calculations, with and without inclusion of animals that failed to conceive, is illustrated in the following table.

Theoretical Reproductive Performance in a 120 Cow dairy Herd With 65% Estrus Detection Efficiency

¹Cows mated after a voluntary waiting period of 54 days.
² Assume 10% of cows are abnormal due to reproductive pathology/acyclicity.
³Assume that 65% of the normal, cycling cows are detected and mated at each estrus.
⁴ Cows that successfully conceive, complete gestation and calve normally from each insemination. The probability is a combination of 90% chance of conception and 75% chance of successful gestation/parturition. (P = 0.9 x 0.75 = 0.675)
⁵ Number of cows calving x median days postpartum for this insemination.

Mean Days Open (pregnant cows) = 10556/114 = 92.6

But six cows were still open after six inseminations and should be culled for reproductive failure. If these animals remained in the herd for a full year, this would add 6 x 365 = 2190 days open to the total.

Mean Days Open (all cows) = 10556 + 2190/120 = 106.2

In the example above, the calculated Days Open for the 114 cows that were re-bred successfully was 92.6 days. However, this value increases to 106.2 days when the 6 infertile animals are included. Performance similar to that illustrated in this example could only be obtained in herds with outstanding reproductive management. The exceptional feature in these results is that only six of 120 cows remained open after 6 inseminations and would presumably become involuntary culls due to reproductive failure. The proportion of females culled eventually for reproductive failure in most dairy herds would be considerably higher so their exclusion from the calculation could certainly mask problems.

Infertile cows, and thus involuntary culls, increase from a very small proportion with most cows inseminated at the proper time, as illustrated in the above table, to one-quarter of the herd with estrous detection efficiency below 50%. High culling rates for reproductive problems, in addition to disrupting production, reduce opportunity to select for desirable traits. [See similar tables with 50% or 40% estrus detection efficiency if interested in further examples].

For dairy herds, calculation of Services Per Pregnancy, Calving Interval and Days Open all provide information about reproductive performance. Whether any of these measures are useful or not, depend on accuracy of recording, proper calculation, regular updating, and frequent reviewing as a basis for management decisions. As mentioned earlier, measures like Services per Pregnancy are based only on females that get pregnant. Hence, it does not provide any indication of the proportion that fail to conceive or when matings occurred in relation to the previous calvings. Similarly, Calving Interval can only be calculated for successive pregnancies so all heifers and cows that did not conceive are excluded. Each of these measures might appear satisfactory in some instances even though a substantial number of animals were culled for reproductive failure. Provided that the numbers culled for sterility and the number of abortions are within acceptable limits, Calving Intervals below 13.5 or 14 months and Services per Pregnancy of 2.5 or less could be considered satisfactory. Most well managed herds should consistently record better performance. The Services per Pregnancy in the example herd illustrated above with 65% estrus detection efficiency is 2.19. Note how this measure increases with less efficient detection.

Days Open provides a much more comprehensive picture of dairy herd fertility, provided the calculation includes every mature female that the operator hoped would get pregnant. Unfortunately, many dairy herds obtain impressively short intervals by neglecting to include any correction for non-conceivers, and their breed associations or other supporting organizations often perpetuate these somewhat misleading figures in promotional material. Whenever all cows are included, Days Open values less than 125 or 130 days can be considered acceptable. Higher values indicate problems and the need for corrective action.

An inherent limitation of the usual performance measures used in dairy herds is that none provide information on calf viability. In addition to whatever calculations are used, one should always check that most of the calves are born alive and remain healthy until selection as herd replacements or disposal.

More Comprehensive Assessment of Reproductive Performance in Dairy Herds
Ideal performance measures would include some indication or correction for the number of females culled for failure to become pregnant in a reasonable period of time. Those responsible for decision making within any herd must appreciate the importance of and distinction between voluntary culling and involuntary culling. Animals that must be removed for reproductive failure represent a special class of involuntary culls and must be monitored closely to insure that numbers remain within acceptable limits.

The American Association of Bovine Practitioners propose using the Projected Minimum Average Calving-to-Pregnancy Interval as an estimate of future performance in dairy herds (Upham, 1991, Bovine. Practitioner 26:49). This measure incorporates the actual results for females that are bred and known to be pregnant combined with optimistic assumptions for recently mated animals and for those past the voluntary waiting period but not yet mated. Such an estimate could be accurate in herds with good reproductive efficiency but would be very unreliable in situations with a high degree of infertility or with poor estrus detection. It does, however, provide some insight into the current and immediate past situation. Users must be aware that it provides only a best estimate and that actual performance will almost always be worse than this.

The calculation of a Calving Interval-Adjusted Calving Rate combines calving interval and the reproductive culling rate into a single measure (Esslemont, 1992, Veterinary Record 131:209). This illustrates the actual proportion of a herd that calve again within a 12 month period. Since both the calving interval and the proportion of the herd that is represented are included in the calculation, this adjusted figure provides a somewhat more comprehensive picture of the situation. It can also be combined with other values to provide an economic indication of herd performance.

North American dairy farmers and extension workers use the figure for calving interval more than calving rate when assessing herd performance. Thus, an adjustment for proportion of animals that failed to reproduce as expected was incorporated into the interval calculation (Plaizier et al. 1998, J. Dairy Science). The resulting Adjusted Calving Interval is proposed as a better indicator of herd reproductive performance since it had a higher correlation with net revenue than did any of the other measures that were tested.

The adjustment factor (denominator) includes:
^1.Total cows culled for reproductive failure includes those animals that were re-mated one or more times but failed to conceive, plus any cows that the operator intended to re-breed but were eventually culled for acyclicity or any other reproductive problem.
^2. Total cows available for rebreeding represents all animals that should have been re-mated successfully. This number includes all of the cows that conceived and calved plus all those culled for reproductive failure (i.e.. total cows in the herd less any voluntary culls that the operator never intended to re-mate and any involuntary culls removed for reasons other than reproductive failure).

The Calving Interval and Adjusted Calving Interval for Dairy Herds

These values could be used as general guidelines when assessing reproductive performance in dairy herds.

Culling Policies. Sound management also requires that some routine policy be followed to make decisions on whether or not individual cows are re-mated. One simple but effective procedure for herds with reasonably good reproductive efficiency is:

• Do not re-mate any second or later lactation cows with production below 80% of the herd average. Cull these animals at the end of lactation.
• Inseminate cows with production between 100 and 120% of herd average for up to 160 days postpartum. If not pregnant by that time, continue milking but cull when daily yield is no longer profitable.
• Cows producing more than 120% of the herd average can be inseminated up to 250 days postpartum.

This provides a variable format in which the poorest producers are replaced automatically without any expense of re-mating (voluntary culls). Also, the average cows have a reasonable time to become pregnant (160 days) while the better cows, who should persist longer at a profitable daily yield, are allowed extra time.

The 60-90 Day Non-Return Rate

Artificial insemination centres use a non-return rate to evaluate fertility of the entire operation as well as for individual sires and technicians. A Non-Return Rate is the percentage of animals inseminated for the first time within a certain period, usually a specific month, that have not returned for a subsequent service and so are presumed pregnant. The most widely used figure is the 60 to 90 Day Non-Return Rate calculated for all first services performed during the third preceding month. Thus, cattle who where inseminated at the start of that particular month had approximately 90 days in which to be detected in estrus and be re-inseminated if they failed to conceive to their first service, while cattle inseminated at the end of that month had 60 days. Non-Return Rates have limitations since they overestimate the true calving rate, may be biased by environmental factors, and only estimate what has gone on in the past. However, provided that the numbers of inseminations are large, they are useful measures for assessing the relative fertility of individual technicians and sires.

Heterologous semen trials or insemination of superovulated females has been proposed to measure accurately the fertilizing capacity of individual sires. Unfortunately, this technology is too expensive for routine use in the artificial insemination industry.

Reproductive Efficiency in Beef Herds

Any of the previous measures could be calculated in beef herds that have complete records obtained through artificial insemination or supervised individual matings. However, neither insemination nor hand mating is common practice in most beef operations. Also, the beef cow's reproductive task is not completed until she weans a calf, and a collective deficiency for all the previously described measures is that none give any indication of offspring viability or survival. For beef ranches or farms, the simple direct calculation of percentage of all mature heifers and cows that wean calves in any particular year provides a comprehensive picture of reproductive performance. This figure is only accurate if the number of calves weaned is divided by the total number of animals that were exposed to bulls and should have produced calves. Whenever production economics are of interest and calves can be weighed at weaning, the total weaning weight of all calves can be divided by the number of mature females in the breeding herd to give the actual kilograms of calf produced per cow. Similar calculations may also be used to assess reproductive efficiency in sheep flocks or goat herds with the lambing or kidding percentages usually exceeding 100%.

Reproductive Efficiency in Swine Units

For sows, true productivity must combine litter size, piglet survival and farrowing interval, so a useful measure is Piglets Weaned per Sow per Year. Many producers delude themselves, or try to impress their neighbors, by omitting culled females and gilts from the calculation. However, all members of the breeding herd are part of the inventory and overhead so each should be included in any efficiency calculation. Gilts may be added to the monthly inventory as soon as they pass market weight and are retained as potential breeding stock or when they are first mated. When the performance is calculated on an annual basis, proportional figures can be used for females kept for periods less than a full year. Litter sizes are sometimes calculated and even published in official production statistics with gilt litters excluded. Since gilts produce 30 to 50% of total litters in commercial piggeries, such figures may look impressive but are quite misleading. Realistic goals for intensive producers who include all mature female members of the breeding herd in their calculations are 18 to 22 piglets weaned per sow per year.

Another simple way to monitor productivity of a large continuously farrowing sow unit is by calculation of the Piglets Weaned or Finished Animals Marketed per Sow per Month. This is simply the actual number weaned or marketed divided by the total number of mature females on the herd inventory for every month of the year. A monthly average of 1.5 corresponds to 18 piglets per sow per year and this can be maintained or exceeded in most problem-free operations.

The Piglets Weaned per Sow per Year is a reliable measure of performance over the long term but, even if the calculation is revised monthly, it could take some time before a problem becomes obvious from this summary. The calculation of Piglets Weaned per Sow per Month should indicate any reduction in reproductive efficiency much sooner.

Reproductive Efficiency in Poultry

The Total Eggs per Hen per Year provides a general efficiency indicator in laying flocks but, since many groups may not be housed for a full 12 months, other indicators are needed. Figures such as egg mass in kilograms or simply Eggs per Hen-Day, when calculated weekly or monthly, continuously monitors output from commercial units.

Breeding flocks must also consider fertility and hatchability. The routine egg candling performed early in the incubation period indicates fertility rates. Similarly, the actual proportion of live chicks emerging from the hatcher demonstrates the overall reproductive performance. The reproductive efficiency would be expressed as the number of Chicks Hatched per Hen-Day or Hen Housed.