- Email: [email protected]
A further comparison of the thumb-counting and comb-counting techniques used to determine Ctenocephalides felis infestation levels on dogs
veterinary parasitology ELSEVIER
Veterinary Parasitology 56 (1995) 349-352
Short c o m m u n i c a t i o n
A further comparison of the thumb-counting and comb-counting techniques used to determine Ctenocephalidesfelis infestation levels on dogs L.M. Gregory*, M. Zakson, R.G. Endris, W.L. Shoop Merck Research Laboratories, R8OA- 12, P.O. Box 2000, Rahway, NJ 07065-0900, USA
Accepted 31 March 1994
Abstract A comparison was made to determine whether thumb-counting or comb-counting was more accurate for determining flea infestation levels on dogs when performed for equal periods of time. To accomplish this, ten beagle dogs were each infested with 100 adult fleas, Ctenocephalidesfelis. After the fleas were allowed to disperse for 1 h the dogs were examined using the thumb-counting method. The time required to cover each dog and the number of fleas counted were recorded. Thumb-counting times ranged from 3.0 to 4.8 rain. Each of the dogs was then examined by the comb-counting method for the same amount of time it had been thumb-counted. The thumb-counting method detected a geometric mean of five (range, 0-13) fleas per dog, while comb-counting recovered a mean of 73.5 (range, 57-87) fleas per dog. These results were significantly different (P<0.01), indicating that the differences in accuracy previously recorded for the two methods are independent of time. The standard deviations for both methods were also statistically significantly different, suggesting that comb-counting is also more precise than the thumbcounting method. Keywords: Dog; Ctenocephalidesfelis; Counting methods
1. Introduction Thumb-counting and comb-counting are two common ways of quantifying the number of fleas on dogs. The thumb-counting technique is a visual inspection of a dog's skin for fleas where the fleas are counted but are not removed. To avoid * Corresponding author. 0304-4017/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSD10304-4017 ( 94 ) 00679-2
350
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
re-counting the same fleas during thumb-counting, the procedure is usually performed quickly and the surface of the dog is covered only once. On the contrary, the comb-counting method utilizes a fine-toothed comb that removes fleas and can therefore be performed for any designated period of time. The accuracy of the two methods was compared previously by Heckenberg et al. (1994) who found that the comb-counting technique, which was employed for 8 rain, recovered significantly more fleas than did thumb-counting. The average duration of the thumbcounting technique in that study was approximately 3 min. The purpose of the present study was to compare these two methods when both were performed for the same amount of time on the same dogs.
2. Materials and methods
Ten beagle dogs including both sexes were used in this study. Each dog was housed individually throughout the study and was given a daily ration of food and access to water ad libitum. One hundred unfed, adult Ctenocephalidesfelis were placed on the back of each dog and allowed to disperse for 1 h before the thumb-counting procedure began. The fleas had emerged from puparia within the previous 3 days. The ten dogs were assigned randomly to a counting order which was maintained for both counting procedures. The thumb-counting and comb-counting techniques were performed as described by Heckenberg et al. (1994). In brief, dogs were placed on a table and an investigator 'thumbed' through the dog's fur, counting all fleas observed. The surface of the animal was covered once to avoid re-counting fleas. The comb-counting procedure was performed using a Twinco ® (Lambert Kay, Cranbury, NJ) extra-fine flea comb ( 11.3 teeth cm -~ ) to remove and count fleas. All ten dogs were first thumb-counted by a single investigator. The number of fleas observed and the time required to cover the surface of each animal once were recorded. After the ten dogs had been thumb-counted, they were then comb-counted by a second investigator for the same period of time they had been thumb-counted. The investigator performing the comb-counts was made aware of the time each animal had been thumb-counted, but not how many fleas were observed. The data were analyzed using a paired t-test on the log(count+ 1 ) transformation and Pitman's test for difference in variance for paired observations (Suedecor and Cochran, 1980). 3. Results
The results from the experiment are contained in Table 1. Without exception, comb-counting recovered a greater number of fleas from each dog than did thumbcounting. The thumb-counting method yielded a geometric mean of five (range, 0-13) fleas per dog, while comb-counting recovered a mean of 73.5 (range, 57-
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
351
Table 1 Numbers of Ctenocephalides felis found on dogs by thumb-counting and comb-counting for equal time periods (rain) Dog
Thumb-counting
Comb-counting
Time
No. of fleas
Time
No. of fleas
2 3 4 5 6 7 8 9 10
4:34 4:48 4:31 3:01 4:10 3:20 3:47 3:37 3:35 3:37
9 0 5 9 3 8 13 4 12 2
4:34 4:48 4:31 3:01 4:10 3:20 3:47 3:37 3:35 3:37
81 86 78 71 57 87 81 60 65 76
Mean
3:54 a
3:54 a
73.5 b'c
I
SD
5.0 b'c 0.81 d'e
0.15 d'e
Dogs were infested with 100 fleas 1 h before counting. "Arithmetic mean. bGeometric mean. *I~ne differences between thumb-counting and comb-counting were highly significantly different (P< 0.01 ) by a paired t-test on the log(count + 1 ) transformation. aStandard deviation (SD) on the log scale. CDifference significant ( P < 0.01 ) by Pitman's test.
87) fleas per dog. This difference was highly statistically significant ( P < 0.01 ). Furthermore, variability on the log scale (i.e. relative variability) was significantly smaller for comb-counting than for thumb-counting. The length of time each dog was thumb-counted and comb-counted also is shown in Table 1. The mean thumb-counting time was 3.9 rain with a range of 3.0-4.8 rain. The exact time the animal was thumb-counted was used as the time the animal was comb-counted. The time interval between the thumb-count and comb-count for each dog was approximately 1 h.
4. Discussion The comb-counting method was previously shown to be a more accurate method than thumb-counting for quantifying the number of fleas on dogs (Heckenberg et al., 1994). However, the techniques differed from those in the present study in several ways, the most striking of which was the length of time that each technique was performed. Heckenberg et al. used an 8 min comb-count time, selected through experience, and others have used various lengths of time. Thumb-counting is generally performed for a shorter amount of time to prevent re-counting the same fleas; the average thumb-count time from Heckenberg et al. was 3.2 min.
352
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
Because the timing for thumb-counting is dictated by how long it takes to cover the surface area of an animal once, we chose to thumb-count the dogs first and then adjust the comb-counting time period accordingly. Similar to the Heckenberg et al. data, the thumb-counting technique was neither more precise nor accurate than comb-counting, yielding a zero on one animal and an overall geometric mean of five, while the comb-counting technique recovered a mean of 73.5 fleas with a much tighter range. It is clear that, even at equal lengths of time, the comb-counting method is more precise and accurate at assessing a flea population. The order in which the counting methods were performed should not have influenced our results according to the cross-over analysis made by Heckenberg et al. (1994). That study indicated that there was no influence of one method on the other, regardless of the order in which the methods were performed. The time period for flea dispersal before and between the two counting procedures was approximately the same, and previous infestations can be ruled out as an influence since each dog was counted using both procedures. In summary, these two methods have been used in flea efficacy trials. However, the results of this study support the idea that comb-counting, for even short periods of time, is a more accurate and precise way to determine flea infestations on dogs than thumb-counting.
Acknowledgements We would like to thank James Bologuese for statistical guidance and analysis.
References Heckenberg, K., Costa, S.D., Gregory, L.M., Michael, B.F., Endris, R.G. and Shoop, W.L., 1994. Comparison of thumb-counting and comb-counting methods to determine Ctenocephalides felis infestation levels on dogs. Vet. Parasitol., 53:153-157. Snedecor, G.W. and Cochran, W.G., 1980. Statistical Methods. The Iowa State University Press, Ames, pp. 190-191.
Veterinary Parasitology 56 (1995) 349-352
Short c o m m u n i c a t i o n
A further comparison of the thumb-counting and comb-counting techniques used to determine Ctenocephalidesfelis infestation levels on dogs L.M. Gregory*, M. Zakson, R.G. Endris, W.L. Shoop Merck Research Laboratories, R8OA- 12, P.O. Box 2000, Rahway, NJ 07065-0900, USA
Accepted 31 March 1994
Abstract A comparison was made to determine whether thumb-counting or comb-counting was more accurate for determining flea infestation levels on dogs when performed for equal periods of time. To accomplish this, ten beagle dogs were each infested with 100 adult fleas, Ctenocephalidesfelis. After the fleas were allowed to disperse for 1 h the dogs were examined using the thumb-counting method. The time required to cover each dog and the number of fleas counted were recorded. Thumb-counting times ranged from 3.0 to 4.8 rain. Each of the dogs was then examined by the comb-counting method for the same amount of time it had been thumb-counted. The thumb-counting method detected a geometric mean of five (range, 0-13) fleas per dog, while comb-counting recovered a mean of 73.5 (range, 57-87) fleas per dog. These results were significantly different (P<0.01), indicating that the differences in accuracy previously recorded for the two methods are independent of time. The standard deviations for both methods were also statistically significantly different, suggesting that comb-counting is also more precise than the thumbcounting method. Keywords: Dog; Ctenocephalidesfelis; Counting methods
1. Introduction Thumb-counting and comb-counting are two common ways of quantifying the number of fleas on dogs. The thumb-counting technique is a visual inspection of a dog's skin for fleas where the fleas are counted but are not removed. To avoid * Corresponding author. 0304-4017/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSD10304-4017 ( 94 ) 00679-2
350
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
re-counting the same fleas during thumb-counting, the procedure is usually performed quickly and the surface of the dog is covered only once. On the contrary, the comb-counting method utilizes a fine-toothed comb that removes fleas and can therefore be performed for any designated period of time. The accuracy of the two methods was compared previously by Heckenberg et al. (1994) who found that the comb-counting technique, which was employed for 8 rain, recovered significantly more fleas than did thumb-counting. The average duration of the thumbcounting technique in that study was approximately 3 min. The purpose of the present study was to compare these two methods when both were performed for the same amount of time on the same dogs.
2. Materials and methods
Ten beagle dogs including both sexes were used in this study. Each dog was housed individually throughout the study and was given a daily ration of food and access to water ad libitum. One hundred unfed, adult Ctenocephalidesfelis were placed on the back of each dog and allowed to disperse for 1 h before the thumb-counting procedure began. The fleas had emerged from puparia within the previous 3 days. The ten dogs were assigned randomly to a counting order which was maintained for both counting procedures. The thumb-counting and comb-counting techniques were performed as described by Heckenberg et al. (1994). In brief, dogs were placed on a table and an investigator 'thumbed' through the dog's fur, counting all fleas observed. The surface of the animal was covered once to avoid re-counting fleas. The comb-counting procedure was performed using a Twinco ® (Lambert Kay, Cranbury, NJ) extra-fine flea comb ( 11.3 teeth cm -~ ) to remove and count fleas. All ten dogs were first thumb-counted by a single investigator. The number of fleas observed and the time required to cover the surface of each animal once were recorded. After the ten dogs had been thumb-counted, they were then comb-counted by a second investigator for the same period of time they had been thumb-counted. The investigator performing the comb-counts was made aware of the time each animal had been thumb-counted, but not how many fleas were observed. The data were analyzed using a paired t-test on the log(count+ 1 ) transformation and Pitman's test for difference in variance for paired observations (Suedecor and Cochran, 1980). 3. Results
The results from the experiment are contained in Table 1. Without exception, comb-counting recovered a greater number of fleas from each dog than did thumbcounting. The thumb-counting method yielded a geometric mean of five (range, 0-13) fleas per dog, while comb-counting recovered a mean of 73.5 (range, 57-
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
351
Table 1 Numbers of Ctenocephalides felis found on dogs by thumb-counting and comb-counting for equal time periods (rain) Dog
Thumb-counting
Comb-counting
Time
No. of fleas
Time
No. of fleas
2 3 4 5 6 7 8 9 10
4:34 4:48 4:31 3:01 4:10 3:20 3:47 3:37 3:35 3:37
9 0 5 9 3 8 13 4 12 2
4:34 4:48 4:31 3:01 4:10 3:20 3:47 3:37 3:35 3:37
81 86 78 71 57 87 81 60 65 76
Mean
3:54 a
3:54 a
73.5 b'c
I
SD
5.0 b'c 0.81 d'e
0.15 d'e
Dogs were infested with 100 fleas 1 h before counting. "Arithmetic mean. bGeometric mean. *I~ne differences between thumb-counting and comb-counting were highly significantly different (P< 0.01 ) by a paired t-test on the log(count + 1 ) transformation. aStandard deviation (SD) on the log scale. CDifference significant ( P < 0.01 ) by Pitman's test.
87) fleas per dog. This difference was highly statistically significant ( P < 0.01 ). Furthermore, variability on the log scale (i.e. relative variability) was significantly smaller for comb-counting than for thumb-counting. The length of time each dog was thumb-counted and comb-counted also is shown in Table 1. The mean thumb-counting time was 3.9 rain with a range of 3.0-4.8 rain. The exact time the animal was thumb-counted was used as the time the animal was comb-counted. The time interval between the thumb-count and comb-count for each dog was approximately 1 h.
4. Discussion The comb-counting method was previously shown to be a more accurate method than thumb-counting for quantifying the number of fleas on dogs (Heckenberg et al., 1994). However, the techniques differed from those in the present study in several ways, the most striking of which was the length of time that each technique was performed. Heckenberg et al. used an 8 min comb-count time, selected through experience, and others have used various lengths of time. Thumb-counting is generally performed for a shorter amount of time to prevent re-counting the same fleas; the average thumb-count time from Heckenberg et al. was 3.2 min.
352
L.M. Gregory et al. / Veterinary Parasitology 56 (1995) 349-352
Because the timing for thumb-counting is dictated by how long it takes to cover the surface area of an animal once, we chose to thumb-count the dogs first and then adjust the comb-counting time period accordingly. Similar to the Heckenberg et al. data, the thumb-counting technique was neither more precise nor accurate than comb-counting, yielding a zero on one animal and an overall geometric mean of five, while the comb-counting technique recovered a mean of 73.5 fleas with a much tighter range. It is clear that, even at equal lengths of time, the comb-counting method is more precise and accurate at assessing a flea population. The order in which the counting methods were performed should not have influenced our results according to the cross-over analysis made by Heckenberg et al. (1994). That study indicated that there was no influence of one method on the other, regardless of the order in which the methods were performed. The time period for flea dispersal before and between the two counting procedures was approximately the same, and previous infestations can be ruled out as an influence since each dog was counted using both procedures. In summary, these two methods have been used in flea efficacy trials. However, the results of this study support the idea that comb-counting, for even short periods of time, is a more accurate and precise way to determine flea infestations on dogs than thumb-counting.
Acknowledgements We would like to thank James Bologuese for statistical guidance and analysis.
References Heckenberg, K., Costa, S.D., Gregory, L.M., Michael, B.F., Endris, R.G. and Shoop, W.L., 1994. Comparison of thumb-counting and comb-counting methods to determine Ctenocephalides felis infestation levels on dogs. Vet. Parasitol., 53:153-157. Snedecor, G.W. and Cochran, W.G., 1980. Statistical Methods. The Iowa State University Press, Ames, pp. 190-191.