Announcement

**sharon sanders** · December 22, 2006, 10:39 AM

Jason got it before I did - here it is for reference

JVI Accepts, published online ahead of print on 20 December 2006

2
20

Abstract

21 Influenza seasonality is one of the most obvious and basic characteristics of the disease
22 process, yet remains the least understood. Most epidemiological studies take this annual
23 oscillation as a basic assumption and build from it to examine other questions, while only a few
24 papers have specifically focused on seasonal trends themselves. By isolating the particular
25 cause(s) of seasonality in influenza, it may be possible to determine effective strategies for
26 intervention in local epidemics, or even global pandemics. This paper provides a review of the
27 diverse array of proposed hypotheses for the underlying cause of seasonality in influenza,
28 including seasonal host health, social causes, and environmental factors such as climate and
29 temperature. Additionally, we provide preliminary epidemiological and mathematical evidence
30 supporting the need for a holistic perspective of all aspects of influenza dynamics within a
31 population. We highlight gaps in these explanations and propose that seasonality may be an
32 emerging property of the virus itself.
33
34

Influenza seasonality: virological and epidemiological perspectives

35 Influenza (or ?flu?) hospitalizes more than 200,000 people yearly and results in 36,000
36 deaths from flu or flu-related complications in the United States (14), striking both elderly and
37 infant populations particularly hard (22). Two members of the Orthomyxoviridae family,
38 influenza A and B primarily cause this acute viral respiratory disease. Both viruses are
39 characterized as enveloped viruses that contain 8 negative-stranded RNA segments that encode 9
40 structural and 2 non-structural proteins (influenza virus A) or 10 structural and one non41
structural proteins (influenza virus B). Because of the higher levels of morbidity and mortality

ACCEPTED

3
associated with influenza A, in part due to the large reservoir 42 of the virus in aquatic birds, we
43 will predominantly restrict ourselves to discussions of this virus.
44 Each influenza A virus particle is surrounded by host cell membrane where two out of
45 three surface proteins, hemagglutinin (HA) and neuraminidase (NA) are responsible for viral
entry into host cell, and are the targets of B-cell immunity 46 (12). While 15 subtypes for HA and 9
47 subtypes for NA exist in the wild, historically only HA subtypes 1, 2, and 3, and NA subtypes 1
48 and 2 have been responsible for stable human infections (48). Human variants of HA 1-3 have
49 high affinity to NeuA

α2.6Gal-containg receptors on mucosal lining of human bronchi50
pulmonary system, and are responsible for viral binding to the potential host cell. NA?s
51 enzymatic cleaving of sialic acid appears to have functions both in reducing the number of
52 ?decoy? receptors that may render viral attachment to epithelial cells of the bronchi-pulmonary
53 system (55), and critically in allowing the release of viral particles from infected cells (32).
54 Two processes allow the virus to quickly change in response to selection, and adeptly
55 evade B-cell immunity through neutralizing antibodies (immunoglobulins). One is ?antigenic
56 drift?, the remarkable ability to mutate rapidly, a function of the infidelity of RNA polymerases.
57 This mutation alters the major antigenic proteins, hemagglutinin (HA) and neuraminidase (NA),
58 which can maintain their function while still undergoing considerable amino acid substitution
59 (11,29). The mutation effect is more restricted for other proteins involved in viral RNA
60 replication and packaging of the viral genomes. The second process, ?antigenic shift? is the
61 ability for the virus to undergo reassortment of its genome when more than one virus has infected
62 a cell, a function of the segmented viral genome. The combination of these processes leads to
63 viruses capable of evading host B-cell immune responses (18,82). The pathogen originates in
64 avian host species and is traditionally thought to infect human populations only via intermediary

ACCEPTED

4
hosts (e.g. pigs), although there is now evidence that direct 65 bird-human transmission is also
66 possible (74).
67 Novel influenza viral strains can be the source of infrequent but devastating pandemics,
68 most famously the 1918 pandemic, which killed between 20 and 40 million people (42). These
69 pandemics are associated with major shifts in the HA and NA proteins that define viral strains.
70 More routinely, influenza also generates epidemics or large outbreaks. Epidemics can be traced
71 to a drift in the HA and NA proteins that circumvents sufficient pre-existing B cell reactivity as
72 to render the individual susceptible. Often influenza may produce sporadic localized outbreaks.
73 Such cases have been not been the focus of research into influenza epidemiological virology but
74 it is likely that they are also associated with a transient viral subpopulation with immuno-evasive
75 properties, which has been borne out with mathematical models (29,68,69).
76 In temperate climates, flu infections at whatever level of intensity are characterized by a
77 flu season. In these areas, the disease is thought to exist at a low level throughout the year, but
78 exhibits a marked seasonal increase, typically during the winter months. Influenza epidemics and
79 outbreaks occur in tropical areas as well, although the timing and impact is not as well defined
80 (15,71,81). Local epidemics begin suddenly, peak in 2 to 3 weeks and last for a total of 5 to 10
81 weeks. It is believed that in most cases the seasonal outbreaks of influenza originate in China,
82 spreading from there (17).
83 Infectious disease dynamics offer a wide variety of intriguing and unexplained
84 phenomena, yet none is as consistently observed while still remaining so poorly understood as
85 seasonality in influenza. There is a gap in how the diverse studies encompassing immunology,
86 mathematics, epidemiology and virology combine to form a complete picture of flu seasonality.
87 This may be due to the daunting complexity of seasonality itself, which is likely to reflect the

ACCEPTED

5
action of a vast multitude of variables. For whatever reason, only limited r 88 esearch has actually
89 focused on supporting or rejecting each of the proposed underlying causes of seasonality in
90 influenza and most of these, while thorough and well supported, are largely confined to a single
91 discipline. This paper will provide a review of the proposed theories from existing literature and
92 examine possible links between them that may suggest a more unified view of influenza
93 seasonality, as well as call to attention gaps between the differing ways of understanding
94 seasonality. Once actually understood, it may be that the mechanisms governing the seasonal
95 pattern in the incidence of influenza can provide greater insight into all aspects of the dynamics
96 of transmission and maintenance of this disease (59). The authors believe that the current
97 research has reached the stage such that advances in our understanding of influenza seasonality
98 will emerge from an enhanced conceptual understanding of the biological dynamics of infection,
99 not a breakthrough in mathematical methodology. As such, the bulk of the review is devoted to
100 the disease parameters being modeled, rather than modeling techniques, although these are
101 briefly examined in their potential applications.
102
103

Biological causes: seasonality as an emergent property of viral infection and host status

104
105

Viral evolution and the immune response to virus

106 As described above, influenza A virus is extremely adept both at evading the host
107 immune system and at achieving heightened virulence. The HA and NA antigenic epitopes of
influenza virus A have a very high rate of mutation (estimated at 6.7 x 10

-3 108 nucleotide
109 substitutions per site per year for HA). When properly positioned, these mutations can prevent
110 the immunoglobulins raised against the HA and NA from previously encountered strains from

ACCEPTED

6
binding to the mutant (31,44). This results in the rapid turnover 111 of viral strains, hampering
112 (though not totally negating) the ability of previously generated B ?cell mediated immune
113 response to guard against re-infection, even within the span of only relatively few viral
114 generations. Small, subtle mutations in HA and NA are incapable of escaping the limited cross115
protection from closely related viral strains that the proliferation of B cells provides, preventing
116 the diversity of circulating strains from increasing explosively (29). This gives influenza its
117 characteristic epidemiological profile of a frequently shifting dominant subtype, rather than a
118 huge number of closely related subtypes which often appear in other rapidly mutating RNA
viruses 119 (20). Models incorporating this limited cross-protection driven by viral adaptation and
120 immune response have been shown to generate cyclical patterns closely resembling the seasonal
121 patterns observed in influenza (29).
Influenza virus B mutates at a much lower rate (3.2 x 10

-3 122 nucleotide substitutions per site
123 per year) (57), though still under similar selective pressure, and appears to rely less on the
124 dramatic genetic reassortment from which much of influenza A virus?s variability is derived.
125 While the complete mechanism of viral evolution in influenza B virus is not yet understood, this
126 slower and more erratic viral evolution may be the driving force behind both the less frequent
127 and less periodic emergence of Influenza B virus capable of infecting large numbers of people
128 (57). Because influenza B virus is restricted to its human host, the lack of an external reservoir in
129 which the virus can evolve independently may also play an important role in its decreased
130 virulence, although there is some suggestion that the virus may have a reservoir in seals (62).
131 While antibody mediated protection is generally considered the primary protection
132 against infection, another source of protection is mediated by CD8 T cells that kill already
133 infected host cells. This cellular immunity does not protect against infection but can rapidly

ACCEPTED

7
ameliorate infection. The immunological studies conducted on hum 134 an populations revealed an
135 interesting pattern where T-cell mediated immune response is targeted against most conserved
136 influenza proteins, such as matrix M1 and NP, which are important in viral particle assembly and
137 extrusion (budding). In fact, age-related studies concluded that be age 15 years children develop
138 anti-influenza CD8 T cell memory comparable with adults (47), and the frequencies of memory
139 cells might reach 0.11-0.56% of the total CD8 T cell pool (43). However, there is a report that
140 memory T cells specific to influenza antigens being at high frequency among CD8 T cells in the
141 peripheral blood might have diminished functional capacity such as interferon-g production and
142 antigen-induced proliferation in elderly individuals, hampering their ability to control the
143 magnitude of viral infections (21).
144 The role of CD4 T cells in the immune response to influenza is well described (13,65,72).
145 Mature B cell responses generally require T cell help, and thus a viral antigen must be able to
146 both stimulate the B cell (hapten) and provide a T cell epitope (carrier). While it might be
147 possible that very mature B memory cells may be able to dispense with T cell help, for most
148 responses the absence of T cell help could lead to an absence of B cell effector function. CD4 T
149 cells may also play a role in generating a strong CD8 memory response.
150 Cyclical patterns in other viral diseases with different immunological and virological
151 behavior can provide useful comparisons to influenza virus. Measles shows a clear cyclical
152 epidemic pattern, although outbreaks are often separated by years, rather than seasons (30,54).
153 However, in this disease, human immune memory provides a powerful, lifelong immunity
154 subsequent to initial exposure. This limits the potential of a population to sustain epidemic levels
155 of infection after an outbreak until the birth rate and immigration can once more provide a
156 sufficiency of susceptible hosts. Influenza virus?s ability to evade protective immunity via the

ACCEPTED

8
introduction of new strains replenishes that pool of available 157 susceptible hosts much more
158 rapidly, shortening the expected length of the epidemic cycle (37). What is lacking is a linkage
159 between viral evolution and the wide assortment of other potential factors in influenza
160 seasonality. Theoretical modeling of viral evolution, or empirical studies, when combined with
161 the examination of high level weather events, mass population mixing and other seasonality
162 factors could establish whether viral evolution is a cause or an effect of differing pathogen-host
163 relations and disease profiles, and how these interact on a microbiological level, where vaccines
164 and prophylactics can address the disease directly.
165
166

Seasonal host health

167 Seasonal variations in the health and physiological status of animals are quite common. It
168 is not surprising that the immune system may experience a pattern of ebb and flow that could
169 leave a host animal vulnerable to infection. It has been suggested that the observed seasonality of
170 influenza is the result not of sweeping waves of disease traveling across the globe, but rather of a
171 constant level of infection mediated differently by the host immune system over time.
172 Specifically it has been posited that light/dark cycles, manifesting as melatonin levels, may have
173 an impact on the immune system, rendering the host more susceptible to infection at different
174 times of the year from pathogens which are present year-round (23). While little direct evidence
175 linking seasonality to susceptibility to influenza infection has yet been published, there is a
176 strong case for the biological plausibility of such a relationship. For instance, photoperiod ? a
177 useful measure of seasonal variations in light exposure ? has been shown to have an impact on
178 the immunocompetence of Siberian hamsters and mice, raising certain immune responses and
179 lowering others (83).

ACCEPTED

9
Two possible intermediaries between photoperiod and i 180 mmunity are melatonin and
181 vitamin-D (25-hydroxy-vitamin D). Melatonin appears to work partly by regulating host immune
182 response via Interleukin-1

β (IL-1β) levels, which rise when melatonin is present, and can exhibit
183 a protective effect in some viral infections (9). Further studies in mice have shown that an IL-1β

184 deficiency results in a higher mortality upon influenza infection (46). Vitamin-D levels have a
185 strong effect on immunity by promoting CD4 T cell and mucosal antibody responses (40) and
186 vitamin D levels are directly related to the amount of sunlight. In both these cases photoperiod
187 effects may be considered as altering normal levels of immunity. The relative impact of each ?
188 and other factors ? is not yet quantified however, and provides ample opportunity for further
189 examination. Dowell (2001) discusses several studies demonstrating that subjects exposed to
190 influenza during the summer were less likely to develop the disease compared to those in the
191 winter, suggesting a host-based defense mechanism active at high levels during the summer. Left
192 unexplored was the possibility of seasonal fluctuations of the nature of the virus itself. A clear
193 and definitive relationship between photoperiod, host immune response and influenza infection
194 has yet to emerge, but the groundwork has been laid for continuing research.
195 Host nutrient intake may also contribute to the seasonal patterns of flu. Low host levels of
196 selenium (Se) lead to an increase in the rate of viral mutations, particularly in the coding of the
197 viral M1 protein that has been shown to increase virulence in mice (8, 60). Additionally, mice
198 fed a long-term diet rich in the antioxidant Vitamin E have been shown to have lower virus titers
199 in the lungs after challenge with influenza as well as lessening the anorexic symptoms of
200 influenza infection that lead to weight loss (39). As both vitamins are taken up in food, it has
201 been proposed that seasonal fluctuations in diet (due to the availability of certain foods) lead to a
202 decrease in antioxidant levels, an increase in oxidative stress, and a commensurate increase in

ACCEPTED

10
viral mutation and, therefore, infection rates. The solid and well supported b 203 iological plausibility
204 of the host?s physiological standing contrasts with a paucity of epidemiological evidence for the
205 same. Questions of how largely local variations in immune response or nutrient intake give rise
206 to global patterns of disease, and how changes in modern lifestyles with the widespread
207 availability of nutritional supplementation, artificial lighting and the like, as well as other
208 questions arise from our currently limited understanding of host-virus interaction. Further
209 research into seasonal immunological phenomena on a population level, especially with regard to
210 influenza, is needed to bridge the gap between disciplines and allow seasonal immunology to be
211 applied directly to the study and prevention of infectious disease.
212
213

Pathogen survival and transmission: social and environmental causes

214 When influenza virus occasionally makes the transition from its natural reservoir of
215 wildfowl and spread to humans, it is transmitted entirely through person-to-person contact. The
216 viral particles replicate in respiratory epithelial cells and are subsequently excreted from the
217 respiratory tract as small-particle aerosols (many less than 2

μm in size) during coughing,
218 sneezing or breathing. Incubation of the disease is very short, typically between 1 and 4 days.
219 Spread from direct contact is also possible in some cases (82,34), although theories of
220 seasonality based on direct contact or large droplet spread are not as well developed as those
221 based on aerosol transmission. The classic SEIR (Susceptible-Exposed-Infected-Recovered)
222 model of disease spread is extremely sensitive to the underlying population to which that model
223 is being applied, as it provides both the pool of susceptible hosts who may come down with the
224 disease, as well as the already infected individuals (spreaders) who will pass the virus on to
225 them. A model of influenza must, therefore, examine the behavior of the infected population, and

ACCEPTED

11
the possibility that the patterns of seasonal epidemics that characterize 226 flu are a result of the
227 population itself.
228
229

Crowding

230 The person-to-person spread of virus-laden aerosol particles is greatly enhanced by
231 having a dense population of susceptible individuals surrounding each infective subject, thereby
232 maximizing the potential for the spread of infection. Crowding has therefore been implicated as a
233 risk factor for a wide range of viral and bacterial diseases including influenza (71,64,10,5,54).
234 The origin and spread of the 1918 influenza pandemic has attributed to the hyper-crowded
235 conditions among military bases during the First World War, allowing several theories
236 concerning the pandemic?s origin to emerge, one implicating the European Western Front (63),
237 and another a U.S. Army base in otherwise isolated and sparsely populated Haskell County,
238 Kansas (7). The crowded conditions of the base, as well as crude and under-supported medical
239 facilities, would have given the otherwise isolated flu outbreak a massive pool of susceptible
240 individuals, who were then shipped all over the country and abroad to fight in the Great War,
241 carrying the disease with them.
242 Given the profound effect crowding has on the spread of viral diseases and the emergence
243 of epidemics, it is not surprising that it is often forwarded as a potential source of seasonality.
244 Seasonal fluctuations in host-behavior might give influenza a greater opportunity to spread and
245 maintain itself at epidemic levels during the winter (58). On a fundamental level, it is plausible
246 that crowding and seasonal social/behavioral patterns are a source of seasonality, although no
247 studies directly examining this plausibly causal relationship have been published. Much of the
248 literature examining this theory treats crowding as a fundamental assumption from which to

ACCEPTED

12
examine further, unrelated questions (28,45,66), and it is an especially 249 frequent explanation of
250 seasonality in popular media. Only a few studies have directly examined this underlying
251 assumption (e.g. Dowell, 2001) addresses the plausibility of the theory critically, asking why ? if
252 crowding is the source of influenza seasonality ? are there not frequent epidemics at busy
253 international summer conventions? The answer might lie in crowding not as the driving cause of
254 seasonal incidence of influenza, but as a contributing factor, amplifying what would otherwise be
255 a subtle and perhaps less pronounced change in virus biology, transmission or host response.
256
257

Ambient temperature

258 Decreased temperature is an environmental variable frequently found to be associated
259 with high levels of seasonal influenza infection (18). Accepted as a basic assumption, this
260 association has been used to explain the decreased effect of seasonality in the tropics (67), and is
261 cited extensively? although not rigorously examined ? in articles examining the rate of acute
262 respiratory infections (15,19,73,75). No direct biological justification for this effect has emerged
263 and it is becoming an increasingly inadequate explanation as our view of seasonality is refined
264 (24).
265 It may be that ambient temperature is simply extremely strongly correlated to the actual
266 mechanism responsible for driving seasonality. These actual causes could be among those
267 already discussed, such as a decrease in temperature inducing behavioral changes such as
268 increased crowding. Or it may be another cause, as yet unexplored. Perhaps decreased ambient
269 temperature increases physiological stress and energy costs for thermoregulation. This could, in
270 turn, weaken the immune system thereby increasing susceptibility to infection from an unaltered
271 rate of exposure. Influenza in waterfowl is an enteric virus and has adapted to the higher

ACCEPTED

13
temperature associated with fowl basal metabolism. T 272 he shift to a human host may have
273 involved a cold adaptation step that is further enhanced at colder temperatures. Another possible
274 temperature related factor is that viral particles are capable of prolonged persistence in colder
275 environments. Given the unclear interactions among temperature, all of the myriad of other
276 correlated mechanisms proposed, and the biology of influenza virus itself, further examination of
277 this effect is clearly warranted.
278
279

Indoor heating

280 Paralleling the direct effects of temperature and harsh weather on the biology of either
281 host or pathogen, it is possible that human defenses against declining temperature may
282 themselves contribute to the seasonality of influenza. Indoor heating levels should increase as the
283 temperature drops, resulting in a continuously recirculated body of air with very low humidity.
284 These conditions are ideal for the persistence of viral particles in the environment, with the
285 typical furnace filter incapable of effectively filtering the very small particles to remove them
286 from the circulating air ? although this may be offset by inactivation of the virus at high
287 temperatures. A lack of concurrent increase in influenza infection in summer, when the use of air
288 conditioning systems is high, may be based largely on the mechanistic differences between the
289 two systems. AC lowers the absolute humidity of the air via condensation (potentially trapping
290 virus-bearing aerosols within the unit itself) while heaters lower only the relative humidity, and
291 never exposing the air to a wet condensing surface. Large-scale heating, such as that in
292 apartment buildings, offices and university dormitories would create a viral dispersion
293 mechanism resembling the unintentional one potentially responsible for the Amoy Gardens
294 SARS outbreak (49). Thus far, this theory, though consistent with other hypotheses governing

ACCEPTED

14
seasonality, has not been discussed in the literature or examined e 295 mpirically in the case of
296 influenza, although levels of indoor air pollution ? which would be continually recirculated in
297 the winter ? have been shown to be risk factors for lower respiratory infection (70).
298 Mathematical models have also been developed examining the risk of indoor, airborne infection
299 risks. Small changes in ventilation accounted for dramatic changes in the R

0 (reproductive ratio,
300 i.e. the number of secondary infections each primary infection gives rise to in an entirely
301 susceptible population) for influenza, although the authors examined risk of infections for single
302 outbreaks, and did not discuss mass ventilation changes ? such as those that may accompany the
303 onset of winter ? as a potential driver of influenza seasonality (50).
304
305 Air travel

306 The role of air travel in the modern epidemiology of influenza has been examined in a
307 substantial body of work, and its impact on the spread of the disease can be subdivided into two
308 categories. The first is the role of air travel on the geographic spread of the disease after an
309 epidemic or pandemic strain has emerged. Using data from the 1968 Hong Kong pandemic,
310 models based on air traffic originating in Hong Kong and carrying passengers with an emerging
311 strain of pandemic influenza show rapid and wide dissemination of the virus across both
312 hemispheres (35). Beyond the model?s implications for the spread of a major pandemic strain,
313 the extremely rapid and universal spread of the hypothetical virus is also applicable to influenza
314 seasonality. Flu season is characterized by near simultaneous appearance of influenza epidemics
315 hemisphere wide. Rapid dissemination of a virus via air travel provides a possible alternative to
316 the theory that exposure is the result of a continually seeded viral pool. Such an alternative
317 would have a substantial effect not only on the spread of a particular strain of the virus, but in the

ACCEPTED

15
global evolution of the disease and host?s immune responses. A 318 constantly exposed population
319 that becomes vulnerable to infection triggered by a seasonal change in virus biology,
320 immunocompetence or social habits is vastly different than a population who ? due to the rapid
321 flow of people and diseases across the globe ? is periodically bombarded by a new viral strain in
322 a short period of time over a wide geographic area. The air travel dissemination model poses
323 intriguing possibilities on the theoretical level, with the potential to shed light on the
324 fundamental issue of the virus?s passage through human populations. What is needed now is
325 epidemiological and virological evidence to confirm (or fail to confirm) this particular
326 mechanism of spread.
327 The second category of air travel studies show that it is possible to forecast the severity of
328 the influenza season based on air traffic patterns. Using a standard compartmental model of
329 influenza transmission coupled with data from U.S. air travel statistics, Grais et. al. were capable
330 of predicting the influenza season of major U.S. cities, and suggest that air travel has a role in the
331 spread of influenza or in the creation of seasonal epidemics, as air travel has a large spike in the
332 winter period, concurrent with influenza season. The model?s inability to accurately predict the
333 incidence of influenza in several cities, and the absence of a peak in influenza during the summer
334 ? a time of high air travel ? are detailed by the authors (36).
335
336

Bulk aerosol transport

337 Beyond the issues of crowding, the aerosolization of influenza virus particles from
338 infectious individuals may itself be directly responsible for the disease?s seasonality. Coughing
339 and sneezing (both symptoms of influenza) produce massive amounts of small-sized aerosol
340 droplets with very high viral titers, which travel through the air at speeds of nearly 100ft/second

ACCEPTED

16
(77). A single patient or population of patients could, over the course 341 of several days, represent a
342 significant source of aerosolized viral particles that could disperse over a wide area. Modeling of
343 the 2003 SARS outbreak at the Amoy Gardens housing complex in Hong Kong (which infected
344 329 people, killing 42) has very thoroughly examined the role of so-called ?bioaerosols? in the
345 transmission of disease (49). The non-uniform spatial pattern of the cases was found to be
346 attributable to air currents created by the architecture and ventilation of the complex, which
347 effectively circulated the virus particles produced by a single individual source to multiple
348 apartments in the complex.
349 On a population level, an entire city experiencing an epidemic could produce staggering
350 amounts of virus aerosols, yielding something not unlike the medieval concept of infective
351 miasma. It has been suggested that this mass of infective particles is responsible for the
352 seasonality of influenza by using global convective currents in much the same way as the
353 ventilation system of the Amoy Gardens housing complex. Originating in Asia during the winter,
354 aerosol particles may be conveyed into the upper atmosphere by frequently forming cyclogenic
355 systems. Here, the low temperature and relative humidity of the upper atmosphere may enhance
356 long-term survival of viral particles, allowing them to be picked up by a westerly air current and
357 transported to North America within the span of a few weeks. Once over the North American
358 continent they are forced lower by frequent cold fronts. In the summer, the atmosphere over
359 North America is less favorable to dispersion, and South Asian pressure systems are weaker and
360 shifts directions, severely depleting the flow of particles and would therefore yield the observed
361 seasonal patterns of infection (38).
362 Although dispersed, the virus particles that survive this trip would find themselves in a
363 favorable environment of cold temperatures, and dry air both inside and out ? both of which have

ACCEPTED

17
been shown to be favorable for viral survival (41). Aerosolized t 364 ransmission of influenza has
365 been shown to be very efficient, and so only a very small number of particles are needed to reach
366 the lungs of a susceptible individual to initiate an epidemic (16). While the movement of small
367 virus bioaerosols in the upper atmosphere is understandably difficult to measure, advances in the
368 sampling of air for airborne viruses show promise in elucidating the quantity of influenza being
369 carried on the wind (2,3,76). An expanded study of atmospheric patterns that would extend this
370 mechanism to the rest of the world has yet to be brought forward and would have to emerge,
371 accompanied by actual sampling data, to be useful in elucidating any possibility of seasonal
372 fluctuations in the global spread of the virus. This notion of transoceanic viral movement in
373 aerosolized form is extremely speculative and difficult to evaluate, however several reported
374 examples of transoceanic dust movements exist. Beyond the issues of the physical movement of
375 the viral particles involved, the added issues of viral survival for such a long period of time ? and
376 under increased exposure to UV radiation ? seem to make likelihood of an infectious virus
377 surviving a trans-Pacific journey somewhat unlikely.
378
379

El Ni?o

380 The El Ni?o Southern Oscillation (ENSO) is a semi-periodic, long term warming of the
381 upper ocean in the tropical eastern Pacific Ocean, and represents the largest signal of
382 atmospheric-oceanic variation and is capable of influencing climates across the globe (80). Links
383 between El Ni?o episodes and infectious disease have been widely reported for a number of
384 different diseases. Oscillations in climate due to ENSO events were associated with influenza
385 morbidity and mortality in France between 1971 and 2002, with a rise in both during cold
386 periods of the ENSO cycle (79). A second study of the hospitalization of women in California

ACCEPTED

18
for viral pneumonia between 1983-1998 showed a similar association f 387 or the city of Sacramento,
388 CA, although it?s results are much less clear, and no association was found for San Francisco or
389 Los Angeles (26). No biological mechanism for this association has been put forward, although it
390 has been suggested that atmospheric and climate variations caused by ENSO cycles may drive
391 bulk aerosol transmission or increase crowding due to inclement weather.
392
393

Mathematical modeling of influenza

394 While many mathematical models have examined theoretical aspects of seasonality in
395 host/pathogen systems, very few have tailored their investigations solely to the parameters and
396 variables of influenza in humans. Although these more abstract models are crucial to the
397 understanding of infectious disease dynamics a whole (including influenza), we will limit our
398 discussion primarily to models which have been strictly focused to the seasonality of influenza.
399 Within this scope, theoretical models that have incorporated an examination of seasonal trends in
400 disease incidence have chosen one of two main perspectives: Seasonal Forcing or Emergent
401 System Properties.
402 Seasonal forcing, altering a model parameter at varying points through the year, is the
403 mathematical method of expressing an underlying process that affects the etiological rates of the
404 disease based on a seasonal pattern in the occurrence of the process (51). This underlying process
405 can stem from any source and the motivations for including resulting changes in the parameters
406 used to describe the disease dynamics can come either from a desire to incorporate
407 mathematically some of the hypothesized theories governing seasonality already discussed (e.g.
408 variations in host immunocompetence) or to prevent the absence of seasonal oscillation in the
409 outcome from making it more difficult to validate model predictions using reported data.

ACCEPTED

19
Mechanistically, seasonal forcing can be accomplished by altering 410 parameters governing
411 environmental exposure, the reproductive rate of the disease (R

0), or infectivity, or by
412 incorporating a separate variable into the model equations.
413 While seasonal forcing incorporates external, seasonally governed processes into
414 mathematical models, it is possible to examine internal mathematical properties of disease
415 dynamics in order to look for inherent patterns of oscillations in the processes themselves.
416 Closely related to the idea of true seasonal forcing is the idea of dynamical resonance (25).
417 While models of seasonal forcing are founded on the belief that annual oscillations are driven by
418 large-scale, observable changes, this theory proposes that many minute changes in R0, so small
419 as to be empirically undetectable, can act in concert with the duration of the infectious period
420 and partial immunity, together generating regular seasonal oscillations. In general, these sorts of
421 internally generated patterns are referred to as ?emergent system properties? and can also arise
422 for a variety of different reasons. The most commonly studied is the phenomenon of bifurcation:
423 the term used to describe systems that have a behavioral threshold beneath which they exhibit
424 one set of behaviors and above which they exhibit another. It has been shown many times that,
425 by examining multiple strains of influenza into a modeled system that incorporates imperfect
426 cross-protective immunity among strains, bifurcations can arise yielding periodic oscillations
427 (33,61). A number of cross-protective immunity models, while not specifically examining
428 bifurcation, have also incorporated rates of influenza strain mutation in order to produce seasonal
429 trends (4,68,69)
430 Some models have tried to induce seasonality as an endogenous property of within431
system dynamics without altering either external seasonal forces or internal strain mutations or
432 multi-strain cross protection based solely on theories of stochastic processes, the influence of

ACCEPTED

20
random noise in the system and cascading local effects (1). TRANSIMS 433

(53) and EpiSims (27)
434 are particularly ambitious in their attempt to capture all of the complexity of a functional society
435 and experimentally examine the impact of direct intervention strategies on disease spread
436 throughout a population. These models are now being adapted to directly examine the impact of
437 pandemic influenza (6). One recent study (78) has examined the spatial movement of individuals
438 on varying scales as a factor in determining the synchrony of disease incidence across large
439 distances. This spatial synchrony could then yield annual oscillations. These models attempt,
440 each in their own way, to incorporate much of the complexity of real-life population dynamics
441 and produce patterns of influenza as emergent properties of populations themselves. Social
442 interactions among different groups within a single population have also been shown to lead to
443 oscillations in the incidence of influenza. By parameterizing the social interaction model
444 presented in Fefferman & Naumova (in prep) using constant, age-specific values for the etiology
445 of influenza taken from Longini et al. (2004), we were able to produce incredible variation in the
446 resulting influenza incidence patterns by comparing different interaction rates among social
447 groups (see Figure 1). Based on these results, it is apparent that social interactions by themselves
448 can be responsible for periodic oscillations. This lends mathematical support to the theory that
449 many different facets must contribute to the resulting observed seasonal patterns and that, to
450 understand the whole, it is important to understand all components within a holistic context.
451
452 Towards a holistic view of seasonality

453 The myriad theories accounting for seasonality reviewed in this paper, as well as those
454 that will hopefully emerge as influenza continues to rise in prominence, suggest that the elegant
455 and predictable periodicity of non-pandemic influenza is caused by a less than straight-forward

ACCEPTED

21
interaction of many different factors. The authors suggest that recognition of 456 this complexity,
457 and the likelihood that seasonality arises from many different factors, is essential for the
458 continued examination and elucidation of seasonality.
459 Of particular note are the gaps between theories and between disciplines. While
460 epidemiologists, virologists, immunologists and mathematicians have all developed laudable
461 theoretical and empirical models to explain seasonality, none are so complete as fully and
462 adequately explain the phenomenon. Our decoding of the dynamics of non-pandemic influenza
463 must take place on biological, social and environmental levels, but more importantly must take
464 place between disciplines. Complex networks of interactions between individual patients and
465 their immune systems, society as a whole, global and local weather and the continual mixing and
466 adaptation of viral antigens to form new strains are likely responsible for seasonal flu infections,
467 and must be understood before steps in public health and medicine can be taken to positively
468 affect health outcomes.
469 Influenza has been a constant global health concern since the pandemic of 1918, if not
470 before, and yet, the most obvious trends in its incidence remain unexplained. Epidemiological
471 investigations have primarily incorporated seasonality as an underlying assumption, focusing
472 instead on other aspects of flu transmission and exposure. As discussed, the proposed theories of
473 causes of seasonal trends in flu incidence span many facets of epidemiology, ranging from
474 specific genetic properties of the virus, to the build-up of pollution in indoor heating systems, to
475 the paths of global wind streams. Alternatively, the answer may be as yet unconsidered. It is our
476 intention that this paper should provide a framework from which further empirical and
477 theoretical investigation specifically into the causes of seasonality in influenza may be
478 undertaken. Studies seeking novel mechanism and those to provide both biological plausibility

ACCEPTED

22
and epidemiological evidence for existing theories are needed. A 479 n understanding of what drives
480 seasonal trends may allow better understanding of transmission dynamics leading to better
481 methods of prevention of annual endemic outbreaks, of pandemics of already existing flu strains
482 and of novel emerging influenzas.
483
484

Acknowledgments

485 The authors wish to thank of the following funding agencies for their support: the National
486 Institute of Allergy and Infectious Diseases (U19AI062627, HHSN266200500024C), and the
487 National Institute of Environmental Health Sciences (R01ES013171) and the Tufts University
488 Undergraduate Research Fund. Finally, the authors wish to gratefully acknowledge our two
489 anonymous reviewers, whose thoughts and insight into earlier versions of this work were
490 exceedingly helpful.

ACCEPTED

23
491

References

492 1.

Abbas, K., A. R. Mikler, and R. Gatti. 2005. Temporal analysis of infectious diseases:
493 influenza. Proceedings of the 2005 ACM symposium on Applied computing 267-271.
494
495 2. Agranovski, I. E., A. S. Safatov, A. I. Borodulin, O. V. Pyankov, V. A. Petrishchenko,

496

Sergeev, A.N., Sergeev, A.A., Agranovski, V., and S. A. Grinshpun. 2005. New personal
497 sampler for viable airborne viruses: feasibility study. J Aerosol Sci 36:609-617.
498
499 3. Agranovski, I. E., A. S. Safatov, A. I. Borodulin, O. V. Pyankov, V. A. Petrishchenko, A.

500

N. Sergeev, A. P. Agafonov, G. M. Ignatiev, A. A. Sergeev, and V. Agranovski. 2004.
501 Inactivation of viruses in bubbling processes utilized for personal bioaerosol monitoring. Appl.
502 Environ. Microbiol. 70:6963-6967.
503
504 4. Andreasen, V. V., J. V. Lin, and S. A. V. Levin. 1997. The dynamics of cocirculating
505 influenza strains conferring partial cross-immunity. J. Math. Biol. 35:825-842.
506
507 5. Baker, M., A. McNicholas, N. Garrett, N. Jones, J. Stewart, V. Koberstein, and D.

508

Lennon. 2000. Household crowding a major risk factor for epidemic meningococcal disease in
509 Auckland children. Pediatr. Infect. Dis. J. 19:983-990.
510
511

ACCEPTED

24
6.

Barrett, C. L., S. G. Eubank, and J. P. Smith. 2005. If smallpox 512 strikes Portland... Sci. Am.
513 292:42-49.
514
515 7. Barry, J. M. 2004. The site of origin of the 1918 influenza pandemic and its public health
516 implications. J. Transl. Med. 2:3.
517
518 8. Beck, M. A., J. Handy, and O. A. Levander. 2004. Host nutritional status: the neglected
519 virulence factor. Trends Microbiol. 12:417-423.
520
521 9. Bonilla, E., N. Valero, L. Chacin-Bonilla, and S. Medina-Leendertz. 2004. Melatonin and
522 viral infections. J. Pineal Res. 36:73-79.
523
524 10. Burstrom, B., F. Diderichsen, and L. Smedman. 1999. Child mortality in Stockholm
525 during 1885-1910: the impact of household size and number of children in the family on the risk
526 of death from measles. Am. J. Epidemiol. 149:1134-1141.
527
528 11. Bush, R. M., W. M. Fitch, C. A. Bender, and N. J. Cox. 1999. Positive selection on the H3
529 hemagglutinin gene of human influenza virus A. Mol Biol Evol 16:1457-1465.
530
531 12. Cann, A. 2005. Principles of molecular virology. Elsevier Academic Press, Amsterdam ;
532 Boston.
533

ACCEPTED

25
13.

Carmichael, P., J. Copier, A. So, and R. Lechler. 1997. Allele-534 specific variation in the
535 degeneracy of major histocompatibility complex (MHC) restriction. Hum Immunol 54:21-29.
536
537 14. Centers for Disease Control and Prevention. 2005. Key Facts about Influenza and the
538 Influenza Vaccine. 2006:1.
539
540 15. Chew, F. T., S. Doraisingham, A. E. Ling, G. Kumarasinghe, and B. W. Lee. 1998.
541 Seasonal trends of viral respiratory tract infections in the tropics. Epidemiol. Infect. 121:121-
542 128.
543
544 16. Couch, R. B., T. R. Cate, R. G. Douglas Jr, P. J. Gerone, and V. Knight. 1966. Effect of
545 route of inoculation on experimental respiratory viral disease in volunteers and evidence for
546 airborne transmission. Bacteriol. Rev. 30:517-529.
547
548 17. Cox, N. J. and K. Subbarao. 2000. Global epidemiology of influenza: past and present.
549 Annu. Rev. Med. 51:407-421.
550
551 18. Cox, N. J. and K. Subbarao. 1999. Influenza. Lancet 354:1277-1282.
552
553 19. Crighton, E. J., R. Moineddin, M. Mamdani, and R. E. Upshur. 2004. Influenza and
554 pneumonia hospitalizations in Ontario: a time-series analysis. Epidemiol. Infect. 132:1167-1174.

ACCEPTED

26
20.

DeFilippis, V. and L. Villarreal. 555 2001. Virus Evolution, p. 353-370. In D. Knipe, P.
556 Howley, D. Griffin, M. Martin, R. Lamb, B. Roizman and S. Straus (ed.), Fields Virology, vol. 1.
557 Lippincott Williams & Wilkins, Philadelphia.
558
559 21. Deng, Y., Y. Jing, A. E. Campbell, and S. Gravenstein. 2004. Age-related impaired type 1
560 T cell responses to influenza: reduced activation ex vivo, decreased expansion in CTL culture in
561 vitro, and blunted response to influenza vaccination in vivo in the elderly. J. Immunol. 172:3437-
562 3446.
563
564 22. Dolin, R. 2005. Influenza--interpandemic as well as pandemic disease. N. Engl. J. Med.
565 353:2535-2537.
566
567 23. Dowell, S. F. 2001. Seasonal variation in host susceptibility and cycles of certain infectious
568 diseases. Emerg. Infect. Dis. 7:369-374.
569
570 24. Dowell, S. F. and M. S. Ho. 2004. Seasonality of infectious diseases and severe acute
571 respiratory syndrome-what we don't know can hurt us. Lancet Infect. Dis. 4:704-708.
572
573 25. Dushoff, J., J. B. Plotkin, S. A. Levin, and D. J. Earn. 2004. Dynamical resonance can
574 account for seasonality of influenza epidemics. Proc. Natl. Acad. Sci. U. S. A. 101:16915-16916.
575

ACCEPTED

27
26.

Ebi, K. L., K. A. Exuzides, E. Lau, M. Kelsh, and 576 A. Barnston. 2001. Association of
577 normal weather periods and El Nino events with hospitalization for viral pneumonia in females:
578 California, 1983-1998. Am. J. Public Health 91:1200-1208.
579
580 27. Eubank, S., H. Guclu, V. S. Kumar, M. V. Marathe, A. Srinivasan, Z. Toroczkai, and N.

581

Wang. 2004. Modelling disease outbreaks in realistic urban social networks. Nature 429:180-
582 184.
583
584 28. Fedson, D. S. 2003. Pandemic influenza and the global vaccine supply. Clin. Infect. Dis.
585 36:1552-1561.
586
587 29. Ferguson, N. M., A. P. Galvani, and R. M. Bush. 2003. Ecological and immunological
588 determinants of influenza evolution. Nature 422:428-433.
589
590 30. Finkenstadt, B., O. Bjornstad, and B. Grenfell. 2002. A stochastic model for extinction
591 and recurrence of epidemics: estimation and inference for measles outbreaks. Biostatistics 3:493-
592 510.
593
594 31. Fitch, W. M., J. M. Leiter, X. Q. Li, and P. Palese. 1991. Positive Darwinian evolution in
595 human influenza A viruses. Proc. Natl. Acad. Sci. U. S. A. 88:4270-4274.
596

ACCEPTED

28
32.

Gillian M. Air, W. G. L. 1989. The neuraminidase of influenza 597 virus. Proteins: Structure,

598

Function, and Genetics 6:341-356.
599
600 33. Gog, J. R. N. and J. N. Swinton. 2002. A status-based approach to multiple strain dynamics.
601 J. Math. Biol. 44:169-184.
602
603 34. Goldmann, D. A. 2000. Transmission of viral respiratory infections in the home. Pediatr.
604 Infect. Dis. J. 19:S97-102.
605
606 35. Grais, R. F., J. H. Ellis, and G. E. Glass. 2003. Assessing the impact of airline travel on the
607 geographic spread of pandemic influenza. Eur. J. Epidemiol. 18:1065-1072.
608
609 36. Grais, R. F., J. H. Ellis, A. Kress, and G. E. Glass. 2004. Modeling the spread of annual
610 influenza epidemics in the U.S.: the potential role of air travel. Health Care Manag. Sci. 7:127-
611 134.
612
613 37. Grenfell, B. T., O. G. Pybus, J. R. Gog, J. L. Wood, J. M. Daly, J. A. Mumford, and E.

614

C. Holmes. 2004. Unifying the epidemiological and evolutionary dynamics of pathogens.
615 Science 303:327-332.
616
617
618

ACCEPTED

29
38.

Hammond, G. W., R. L. Raddatz, and D. E. Gelskey. 619 1989. Impact of atmospheric
620 dispersion and transport of viral aerosols on the epidemiology of influenza. Rev. Infect. Dis.
621 11:494-497.
622
623 39. Han, S. N., M. Meydani, D. Wu, B. S. Bender, D. E. Smith, J. Vina, G. Cao, R. L. Prior,

624

and S. N. Meydani. 2000. Effect of long-term dietary antioxidant supplementation on influenza
625 virus infection. J. Gerontol. A Biol. Sci. Med. Sci. 55:B496-503.
626
627 40. Hayes, C. E., F. E. Nashold, K. M. Spach, and L. B. Pedersen. 2003. The immunological
628 functions of the vitamin D endocrine system. Cell. Mol. Biol. (Noisy-le-grand) 49:277-300.
629
630 41. Hemmes, J. H., K. C. Winkler, and S. M. Kool. 1960. Virus survival as a seasonal factor in
631 influenza and polimyelitis. Nature 188:430-431.
632
633 42. Hilleman, M. R. 2002. Realities and enigmas of human viral influenza: pathogenesis,
634 epidemiology and control. Vaccine 20:3068-3087.
635
636 43. Hoji, A. and C. R. Rinaldo Jr. 2005. Human CD8+ T cells specific for influenza A virus
637 M1 display broad expression of maturation-associated phenotypic markers and chemokine
638 receptors. Immunology 115:239-245.
639

ACCEPTED

30
44.

Ina, Y. and T. Gojobori. 1994. Statistical analysis 640 of nucleotide sequences of the
641 hemagglutinin gene of human influenza A viruses. Proc. Natl. Acad. Sci. U. S. A. 91:8388-8392.
642
643 45. Koch, A., P. Sorensen, P. Homoe, K. Molbak, F. K. Pedersen, T. Mortensen, H.

644

Elberling, A. M. Eriksen, O. R. Olsen, and M. Melbye. 2002. Population-based study of acute
645 respiratory infections in children, Greenland. Emerg. Infect. Dis. 8:586-593.
646
647 46. Kozak, W., H. Zheng, C. A. Conn, D. Soszynski, L. H. van der Ploeg, and M. J. Kluger.

648 1995. Thermal and behavioral effects of lipopolysaccharide and influenza in interleukin-1 beta649
deficient mice. Am. J. Physiol.

269:R969-77.
650
651 47. Lawson, T. M., S. Man, E. C. Wang, S. Williams, N. Amos, G. M. Gillespie, P. A. Moss,

652

and L. K. Borysiewicz. 2001. Functional differences between influenza A-specific cytotoxic T
653 lymphocyte clones expressing dominant and subdominant TCR. Int. Immunol. 13:1383-1390.
654
655 48. Lewis, D. B. 2006. Avial flu to human influenza. Ann. Rev. Med. 57:139-54.
656
657 49. Li, Y., S. Duan, I. T. Yu, and T. W. Wong. 2005. Multi-zone modeling of probable SARS
658 virus transmission by airflow between flats in Block E, Amoy Gardens. Indoor Air 15:96-111.
659
660

ACCEPTED

31
50.

Liao, C. M., C. F. Chang, and H. M. Liang. 2005. A probabilistic 661 transmission dynamic
662 model to assess indoor airborne infection risks. Risk Anal. 25:1097-1107.
663
664 51. Lloyd, A. L. and R. M. May. 1996. Spatial heterogeneity in epidemic models. J. Theor.
665 Biol. 179:1-11.
666
667 52. Longini, I. M.,Jr, M. E. Halloran, A. Nizam, and Y. Yang. 2004. Containing pandemic
668 influenza with antiviral agents. Am. J. Epidemiol. 159:623-633.
669
670 53. Los Alamos National Laboratory. 2002. TRANSIMS 3.0 Documentation Set.
671
672 54. Manfredi, P., E. Cleur, J. Williams, S. Salmaso, and M. Atti. 2005. The pre-vaccination
673 regional epidemiological landscape of measles in Italy: contact patterns, effort needed for
674 eradication, and comparison with other regions of Europe. Popul Health Metr 3:1.
675
676 55. Matrosovich, M.N., T.Y. Matrosovich, T. Gray, N.A. Roberts, and H.D. Klenk. 2004.
677 Neuraminidase is important for the initiation of influenza virus infection in human airway
678 epithelium. JVI 78: 12665-12667.
679
680
681

ACCEPTED

32
56.

McCarthy, M., M. K. Estes, and K. C. Hyams. 2000. Norwalk-682 like virus infection in
683 military forces: epidemic potential, sporadic disease, and the future direction of prevention and
684 control efforts. J. Infect. Dis. 181 Suppl 2:S387-91.
685
686 57. McCullers, J. A., G. C. Wang, S. He, and R. G. Webster. 1999. Reassortment and
687 insertion-deletion are strategies for the evolution of influenza B viruses in nature. J. Virol.
688 73:7343-7348.
689
690 58. National Research Council . Committee on Climate, Ecosystems, Infectious Disease, and

691

Human Health. 2001. Under the weather : climate, ecosystems, and infectious disease, p. 54-55.
692 In Anonymous . National Academy Press, Washington, D.C.
693
694 59. Naumova, E. 2006. Mystery of Seasonality: Getting the Rhythm of Nature. J. Public Health
695 Policy 27:2-12.
696
697 60. Nelson, H. K., Qi. Shi, P.V. Dael, E.J. Schiffrin, S. Blum, D. Barclay, O.A. Levander

698

and M.A. Beck. 2001. Host nutritional selenium status as a driving force for influenza virus
699 mutations. FASEB J. 10.1096/fj.01-0115fje.
700
701 61. Nu?o, M., Z. Feng, M. Martcheva, and C. Castillo-Chavez. Dynamics of two-strain
702 influenza with isolation and partial cross-immunity. SIAM J. Appl. Math. 65: 964?982

ACCEPTED

33
62.

Osterhaus, A.D.M.E., G.F. Rimmelzwaan, B.E.E 703 . Martina, T.M. Bestebroer and

704

R.A.M. Fouchier. 2000. Influenza B virus in seals. Science. 288: 1-51-1053.
705
706 63. Oxford, J. S., A. Sefton, R. Jackson, W. Innes, R. S. Daniels, and N. P. Johnson. 2002.
707 World War I may have allowed the emergence of "Spanish" influenza. Lancet Infect. Dis. 2:111-
708 114.
709
710 64. Reeves, W. C., M. M. Brenes, E. Quiroz, J. Palacios, G. Campos, and R. Centeno. 1986.
711 Acute hemorrhagic conjunctivitis epidemic in Colon, Republic of Panama. Am. J. Epidemiol.
712 123:325-335.
713
714 65. Rothbard, J. B., and M. L. Gefter. 1991. Interactions between immunogenic peptides and
715 MHC proteins. Annu Rev Immunol 9:527-565.
716
717 66. Sarubbi, F. A. 2003. Influenza: a historical perspective. South. Med. J. 96:735-736.
718
719 67. Shih, S. R., G. W. Chen, C. C. Yang, W. Z. Yang, D. P. Liu, J. H. Lin, S. C. Chiu, H. Y.

720

Chen, K. C. Tsao, C. G. Huang, Y. L. Huang, C. K. Mok, C. J. Chen, T. Y. Lin, J. R. Wang,

721

C. L. Kao, K. H. Lin, L. K. Chen, H. L. Eng, Y. C. Liu, P. Y. Chen, J. S. Lin, J. H. Wang,

722

C. W. Lin, Y. J. Chan, J. J. Lu, C. A. Hsiung, P. J. Chen, and I. J. Su. 2005. Laboratory723
based surveillance and molecular epidemiology of influenza virus in Taiwan. J. Clin. Microbiol.
724 43:1651-1661.

ACCEPTED

34
68.

Smith, D. J. 2006. Predictability and preparedness in i 725 nfluenza control. Science 312:392-
726 394.
727
728 69. Smith, D. J. 2003. Applications of bioinformatics and computational biology to influenza
729 surveillance and vaccine strain selection. Vaccine 21:1758-1761.
730
731 70. Smith, K. R., J. M. Samet, I. Romieu, and N. Bruce. 2000. Indoor air pollution in
732 developing countries and acute lower respiratory infections in children. Thorax 55:518-532.
733
734 71. Souza, L. S., E. A. Ramos, F. M. Carvalho, V. M. Guedes, L. S. Souza, C. M. Rocha, A.

735

B. Soares, F. Velloso Lde, I. S. Macedo, F. E. Moura, M. Siqueira, S. Fortes, C. C. de Jesus,

736

C. M. Santiago, A. M. Carvalho, and E. Arruda. 2003. Viral respiratory infections in young
737 children attending day care in urban Northeast Brazil. Pediatr. Pulmonol. 35:184-191.
738
739 72. Stern, L. J., J. H. Brown, T. S. Jardetzky, J. C. Gorga, R. G. Urban, J. L. Strominger,

740

and D. C. Wiley. 1994. Crystal structure of the human class II MHC protein HLA-DR1
741 complexed with an influenza virus peptide. Nature 368:215-221.
742
743 73. Straliotto, S. M., M. M. Siqueira, R. L. Muller, G. B. Fischer, M. L. Cunha, and S. M.

744

Nestor. 2002. Viral etiology of acute respiratory infections among children in Porto Alegre, RS,
745 Brazil. Rev. Soc. Bras. Med. Trop. 35:283-291.

ACCEPTED

35
74.

Subbarao, K. and J. Katz. 2000. Avian influenza viruses infecting 746 humans. Cell Mol. Life
747 Sci. 57:1770-1784.
748
749 75. Tan, J., L. Mu, J. Huang, S. Yu, B. Chen, and J. Yin. 2005. An initial investigation of the
750 association between the SARS outbreak and weather: with the view of the environmental
751 temperature and its variation. J. Epidemiol. Community Health 59:186-192.
752
753 76. Tseng, C. and C. Li. 2005. Collection efficiencies of aerosol samplers for virus-containing
754 aerosols. J Aerosol Sci 36:593-607.
755
756 77. Tyler, K. and N. Nathanson. 2001. Pathogenesis of Viral Infections, p. 199-241. In D.
757 Knipe, P. Howley, D. Griffin, M. Martin, R. Lamb, B. Roizman and S. Straus (ed.), , vol. 1.
758 Lippincott Williams & Wilkins, Philadelphia.
759
760 78. Viboud, C., O. N. Bj?rnstad, D. L. Smith, L. Simonsen, M. A. Miller, and B. T. Grenfell.

761 2006. Synchrony, Waves, and Spatial Hierarchies in the Spread of Influenza. Science

312:447-
762 451.
763
764 79. Viboud, C., K. Pakdaman, P. Y. Boelle, M. L. Wilson, M. F. Myers, A. J. Valleron, and

765

A. Flahault. 2004. Association of influenza epidemics with global climate variability. Eur. J.
766 Epidemiol. 19:1055-1059.

ACCEPTED

36
80.

Wang, H. J., R. H. Zhang, J. Cole, and F. 767 Chavez. 1999. El Nino and the related
768 phenomenon Southern Oscillation (ENSO): The largest signal in interannual climate variation.
769 Proc. Natl. Acad. Sci. U. S. A. 96:11071-11072.
770
771 81. Wong, C. M., K. P. Chan, A. J. Hedley, and J. S. Peiris. 2004. Influenza-associated
772 mortality in Hong Kong. Clin. Infect. Dis. 39:1611-1617.
773
774 82. Wright, P. and R. Webster. 2001. Orthomyxoviruses, p. 1533-1579. In D. Knipe, P.
775 Howley, D. Griffin, M. Martin, R. Lamb, B. Roizman and S. Straus (ed.), Fields Virology, vol. 1.
776 Lippincott Williams & Wilkins, Philadelphia.
777
778 83. Yellon, S. M., O. R. Fagoaga, and S. L. Nehlsen-Cannarella. 1999. Influence of
779 photoperiod on immune cell functions in the male Siberian hamster. Am. J. Physiol. 276:R97-
780 R102.
781

ACCEPTED

1
1

Figure Legends:

2
3

Figure 1: The different patterns of influenza incidence in a total population caused by the

4

different patterns of social interaction among 6 etiologically distinct age groups (children

5

<5 years, children 6-20 years, adults with young children, adults with older children, adults

6

with no children and elderly). All 12 modeled scenarios used the same population size and

7

demography. The only differences between those modeled scenarios that yielded constant

8

influenza incidence (Panel A), rapidly stabilizing oscillating influenza incidence (Panel B),

9

or long-term periodic oscillating influenza incidence (Panel C) were in the social interaction

10

rates among these etiological groups. All parameter values and interaction rates were held

11

constant throughout each scenario.

12

ACCEPTED

**sharon sanders** · December 22, 2006, 11:51 AM

Re: JV - Influenza Seasonality: Underlying Causes and Modeling Theories

See, everyone should move to Florida - The Sunshine State - to avoid "bird flu". I am renting out rooms - cheap.

For instance, photoperiod ? a
177 useful measure of seasonal variations in light exposure ? has been shown to have an impact on
178 the immunocompetence of Siberian hamsters and mice, raising certain immune responses and
179 lowering others (83).

9

Two possible intermediaries between photoperiod and i 180 mmunity are melatonin and
181 vitamin-D (25-hydroxy-vitamin D). Melatonin appears to work partly by regulating host immune
182 response via Interleukin-1
β (IL-1β) levels, which rise when melatonin is present, and can exhibit
183 a protective effect in some viral infections (9). Further studies in mice have shown that an IL-1β

184 deficiency results in a higher mortality upon influenza infection (46). Vitamin-D levels have a

185 strong effect on immunity by promoting CD4 T cell and mucosal antibody responses (40) and
186 vitamin D levels are directly related to the amount of sunlight. In both these cases photoperiod
187 effects may be considered as altering normal levels of immunity. The relative impact of each ?
188 and other factors ? is not yet quantified however, and provides ample opportunity for further
189 examination. Dowell (2001) discusses several studies demonstrating that subjects exposed to
190 influenza during the summer were less likely to develop the disease compared to those in the
191 winter, suggesting a host-based defense mechanism active at high levels during the summer.

Announcement

JV - Influenza Seasonality: Underlying Causes and Modeling Theories

JV - Influenza Seasonality: Underlying Causes and Modeling Theories

Comment

Comment