NR AEVF
AU Gravenor,M.B.; Cox,D.R.; Hoinville,L.J.; Hoek,A.; McLean,A.R.
TI Scrapie in Britain during the BSE years
QU Nature 2000 Aug 10; 406(6796): 584-5
PT journal article
VT
The experimental transmission of bovine spongiform encephalopathy (BSE) to sheep[1] raised the possibility that some sheep in the United Kingdom could have been infected during the 1980s after exposure to BSE-contaminated feed. In contrast to new diseases that have appeared in a number of feline species and wild ungulates[2], the symptoms of BSE in sheep are very similar to another transmissible spongiform encephalopathy called scrapie, which has been endemic in Britain for over 200 years. Although so far no cases of BSE in sheep have been found, these may have been misdiagnosed as scrapie. Here we present data describing the historical changes in scrapie incidence[3], and find no evidence for a peak in scrapie incidence before, during or after the BSE outbreak, making it unlikely that a substantial epidemic of BSE has occurred in the sheep population.
A postal survey of 11,554 sheep farms[4] requested information on the length of time farmers had been involved in farming sheep and the year in which scrapie had first appeared in their flock. The response rate was 61%, and 15% of the relevant responders reported ever having had scrapie. For each year we calculated how many responders were farming, how many had not yet had any scrapie, and how many acquired their first case that year. If the force of infection per farm (FI) is the rate at which farms acquire cases, we define the first-case force of infection (FI1) as the number acquiring a first case in a given year divided by the number of farms at risk of doing so.
Figure 1a shows FI1 for 1962-98, together with the FI1 for BSE in cattle herds for comparison. If an increase in spongiform encephalopathy in sheep was linked to the BSE epidemic, we might expect a substantial increase in new scrapie-affected farms during this time (or preceding the BSE epidemic by a few years owing to the shorter incubation period of BSE in sheep relative to cattle[5]). Instead, we see a steady increase from an estimated mean of 0.2% in 1970 to 0.4% in 1990, followed by a slight decline. Although a curve best fits the data, there is no sharp peak. This is apparent from the raw data. Further, the formal confidence interval for the date of maximum FI1 is very wide.
------------------------------------------
Figure 1 Analysis of results from scrapie survey among British farmers. a, Proportion of farms at risk of having a first scrapie case per year (FI1). The fitted quadratic curve (estimated using logistic regression) was not sensitive to smoothing of the high proportion of cases reported in 1970 and 1980 that probably resulted from digit preference. Also shown is the proportion of cattle herds with no previous BSE that experienced a first case in each year. b, Estimation of scrapie FI for all farms (not just first case). If the probability of acquiring any case in year n is theta(n) = theta (1+ alpha(n) + beta(n2)), the probability phi(n) of a first case in year theta(n), given no previous cases, can be expressed as (komplexe Formel). A gamma distribution for the probability density function of theta is assumed (with coefficient of variation=1/c approximated by 1.3, the current coefficient of variation in flock size). The parameters m, alpha (linear coefficient) and beta (quadratic coefficient), and hence the force of infection theta per year, can be estimated from the reported proportion of first cases each year (phi(n)). Parameter estimates (solid line) were m=0.0031 (0.0022, 0.0039), alpha=0.039 (0.015, 0.062), beta=6.610-5 (-0.0018, 0.0019), showing no significant departure from linearity. Dotted lines give the estimated FI assuming the coefficient of variation in theta is +/- 50% the coefficient of variation for flock size.
------------------------------------------
If some farms are more susceptible to scrapie than others, analysis of the farms with first cases only (FI1) will give a biased estimate of the true force of infection for all farms (FI). Variability in the probability of acquiring scrapie arises through a number of factors, including flock size and type[6]. We estimated the trend in FI from the survey data by expressing the probability of acquiring the first case in terms of the probability of acquiring any case and the variation in this probability between farms. FI was estimated to increase linearly over time (Fig. 1b). Factors that may underlie this trend include recall biases and increases in average flock size and scrapie awareness. Note that variation in farm susceptibility introduces a curved shape to FI1 (as in Fig. 1a) even when the underlying probability of infection is linear. The recent apparent decline in first-case farms may therefore be misleading.
Overall, the data indicate that there is a gradual linear increase in the probability of farms acquiring a scrapie case over the past 36 years. Sensitivity analysis suggests that departures from this pattern would have been detected had the probability of having scrapie since 1981 increased to twice that expected from 1962-80. Different patterns in FI in a small subsample of farms may therefore be overlooked. Splitting the data by farm size does not reveal any such heterogeneity. A further subgroup to analyse would be farms that fed large amounts of meat and bone meal, but these farms were not identified in the survey.
The link between BSE and variant Creutzfeldt-Jakob disease in humans[7,8] raises the question of whether and to what extent BSE entered the British sheep population. Mass screening[9] to check for BSE has been proposed but is not yet feasible. The survey provides additional information that farms raising both sheep and cattle were not at a higher risk of having had scrapie, and regional scrapie and BSE incidences were not significantly correlated. Reports of case numbers in the past 12 months and previous 5 years were also consistent with a recent constant incidence of scrapie[10]. The data presented here cannot be used to determine whether any sheep became infected with BSE, but they do suggest that the change in sheep spongiform encephalopathy incidence has been a gradual process, with no large epidemic during the 1980s and 1990s.
References
1. Foster, J. D., Hope, J. & Fraser, H. Vet. Rec. 133, 339-341 (1993).
2. Kirkwood, J. K. & Cunningham, A. A. Vet. Rec. 135, 296-303 (1994).
3. Butler, D. Nature 395, 6-7 (1998).
4. Hoinville, L., McLean, A. R., Hoek, A., Gravenor, M. B. & Wilesmith, J. Vet. Rec. 145, 405-406 (1999).
5. Goldmann, W., Hunter, N., Smith, G., Foster, J. & Hope, J. J. Gen. Virol. 75, 989-985 (1994).
6. McLean, A. R., Hoek, A., Hoinville, L. J. & Gravenor, M. B. Proc. R. Soc. Lond. B 266, 2531-2538 (1999).
7. Collinge, J., Sidle, K. C., Meads, J., Ironside, J. & Hill, A. F. Nature 383, 685-690 (1996).
8. Bruce, M. E. et al. Nature 389, 498-501 (1997).
9. Hill, A. F. et al. Neurosci. Lett. 255, 159-162 (1998).
10. Gravenor, M. B., Cox, D. R., Hoinville, L. J., Hoek, A. & McLean, A. R. Proc. R. Soc. Lond. B (submitted).
IN Die Autoren schickten Fragebögen an 11.554 britische Schaffarmen und erhielten Antworten von 61% der Angeschriebenen. Von den Antwortenden hatten 15% schon scrapie gehabt. Aus den Angaben zum Jahr des ersten Auftretens berechneten sie mit einem komplizierten mathematischen Modell, dass die Zahl der jährlich neu betroffenen Farmen zwischen 1962 und 1998 keinen BSE-Peak aufweist. Aber der Anteil der jährlich neu betroffenen Farmen nahm kontinuierlich von etwa 0,2% um 1970 auf rund 0,4% um 1990 leicht zu und danach tendentiell leicht ab. Die Herdengröße scheint dabei keine Rolle gespielt zu haben. Auch gemeinsame Haltung von Schafen und Rindern führte nicht zu einem erkennbar erhöhten scrapie-Risiko. Außerdem fand man keine signifikante Korrelation zwischen den regionalen Häufigkeitsverteilungen von BSE und Scrapie.
MH Animal; Cattle; Creutzfeldt-Jakob Syndrome; Disease Outbreaks/veterinary; Encephalopathy, Bovine Spongiform/*epidemiology/transmission; Great Britain/epidemiology; Human; Incidence; Scrapie/*epidemiology/transmission; Sheep
AD Mike B. Gravenor und Angelar R. McLean, Institute for Animal Health, Compton, Berkshire RG20 7NN, UK; D.R. Cox, Department of Statistics, University of Oxford, 1 South Parks Road, Oxford OX1 3TG, UK ; Linda J. Hoinville; Alies Hoek, Epidemiology Department, Veterinary Laboratories Agency, New Haw, Addelestone, Surrey KT15 3NB, UK; e-mail: michael.gravenor@bbsrc.ac.uk
SP englisch
PO England
EA pdf-Datei und HTML-Version
ZF kritische Zusammenfassung von Roland Heynkes