Current methods for the detection of contagious outbreaks ideally give contemporaneous information about the course of an epidemic, though, more typically, the indicators lag behind the epidemic.
However, the situation could be improved, possibly significantly, if detection methods took advantage of a potentially informative property of social networks: during a contagious outbreak, individuals at the center of a network are likely to be infected sooner than random members of the population. Hence, the careful collection of information from a sample of central individuals within human social networks could be used to detect contagious outbreaks before
they happen in the population-at-large.
A contagion that stochastically infects some individuals and then spreads from person to person in the network will tend, on average, to reach centrally located individuals more quickly than peripheral individuals because central individuals (as defined in various ways described below) are a smaller number of steps (degrees of separation) away from the average individual in the network (see ).
Indeed, although some contagions can spread via incidental contact, the duration of exposure between people with social ties is typically much higher than between strangers, suggesting that the social network itself will be an important conduit for the spread of an outbreak.
As a result, we would expect the S-shaped epidemic curve 
to be shifted to the left (forward in time) for centrally located individuals compared to the population as a whole (see ). This shift, if it could be observed, would allow for early detection of an outbreak.
Network Illustrating Structural Parameters.
Theoretical expectations of differences in contagion between central individuals and the population as a whole.
Prior modeling research suggests that vaccinating central individuals in networks could enhance the population-level efficacy of a prophylactic intervention 
and other work suggests that optimal placement of sensors in physical networks (such as water pumping stations) could detect outbreaks sooner.
However, mapping a whole network to identify particular individuals from whom to collect information is costly, time-consuming, and often impossible, especially for large networks.
We therefore explore a novel, alternative strategy that does not
require ascertainment of global network structure, namely, monitoring the friends of randomly selected individuals
. This strategy exploits an interesting property of human social networks: on average, the friends of randomly selected people possess more links (have higher degree) and are also more central (e.g., as measured by betweenness centrality) to the network than the initial, randomly selected people who named them.
Therefore, we expect a set of nominated friends to get infected earlier than a set of randomly chosen individuals (who represent the population as a whole). More specifically, a random sample of individuals from a social network will have a mean degree of μ
(the mean degree for the population); but the friends of these random individuals will have a mean degree of μ
plus a quantity defined by the variance of the degree distribution divided by μ
. Hence, when there is variance in degree in a population, and especially when there is high variance, the mean number of contacts for the friends will be greater (and potentially much greater) than the mean for the random sample. This is sometimes known as the “friendship paradox” (“your friends have more friends than you do”) 
While the idea of immunizing
such friends of randomly chosen people has previously been explored in a stimulating theoretical paper 
, to our knowledge, a method that uses nominated friends as sensors
for early detection of an outbreak has not previously been proposed, nor has it been tested on any sort of real outbreak. To evaluate the effectiveness of nominated friends as social network sensors, we therefore monitored the spread of flu at Harvard College from September 1 to December 31, 2009. In the fall of 2009, both seasonal flu (which typically kills 41,000 Americans each year 
) and the H1N1 strain were prevalent in the US, though the great majority of cases in 2009 have been attributed to the latter.
It is estimated that this H1N1 epidemic, which began roughly in April 2009, infected over 50 million Americans. Unlike seasonal flu, which typically affects individuals older than 65, H1N1 tends to affect young people. Nationally, according to the CDC, the epidemic peaked in late October 2009, and vaccination only became widely available in December 2009. Whether another outbreak of H1N1 will occur (for example, in areas and populations that have heretofore been spared) is a matter of some debate,
but many scholars have been studying the situation from biological and public health perspectives.
We enrolled a total of 744 undergraduate students from Harvard College, discerned their friendship ties, and tracked whether they had the flu beginning on September 1, 2009 (from the start of the new academic year) to December 31, 2009. This sample was assembled by empanelling two groups of students of essential analytic interest: (1) a sample chosen randomly from the 6,650 Harvard undergraduates (N
319), and (2) a “friends” sample (N
425) composed of individuals who were named as a friend at least once by a member of this random sample (see Supporting Information Text S1
for more details).
In addition, as a byproduct of empanelling the foregoing group of 744 students, we wound up having information about a total of 1,789 uniquely identified Harvard College students (who either participated in the study or who were nominated as friends or as friends of friends); we used this information to draw the social network of part of the Harvard College student body (see Supporting Information Text S1
for more details).
All subjects completed a brief background questionnaire soliciting demographic information, flu and vaccination status since September 1, 2009, and certain self-reported measures of popularity. We also obtained basic administrative data from the Harvard College registrar, such as sex, class of enrolment, and inter-collegiate sports participation.
We tracked cases of formally diagnosed influenza among the students in our sample as recorded by University Health Services (UHS) beginning on September 1, 2009 through December 31, 2009. Presenting to the health service indicates a more severe level of symptomatology, of course, and so we do not expect the same overall prevalence using this diagnostic standard as with self-reported flu discussed below. However, UHS data offer the advantage of allowing us to obtain information about flu symptoms as assessed by medical staff.
Beginning on October 23, 2009, we also collected self-reported flu symptom information from participants via email twice weekly (on Mondays and Thursdays), continuing until December 31, 2009. The students were queried about whether they had had a fever or flu symptoms since the last email contact, and there was very little missing data (47% of the subjects completed all of the biweekly surveys, and 90% missed no more than two of the surveys).
Self-report of symptoms rather than serological testing is the current standard for flu diagnosis. Similar to previous studies,
students were deemed to have a case of flu (whether seasonal or the H1N1 variety) if they reported having a fever of greater than 100° F (37.8°C) and
at least two of the following symptoms: sore throat; cough; stuffy or runny nose; body aches; headache; chills; or fatigue. We checked the sensitivity of our findings by using definitions of flu that required more symptoms, and our results did not change (see Supporting Information Text S1
). As part of the foregoing biweekly self-reports, in order to complement the UHS vaccination records, we also ascertained whether the students reported having been vaccinated (with seasonal flu vaccine or H1N1 vaccine or both) at places other than (and including) UHS.
To be clear, we are not suggesting that a person's precise position in the observed network, nor indeed whether he was nominated as a friend or not (and by whom), traces out the actual path by which he acquired (or did not acquire) the flu. The topological parameters we measured here, or indeed the fact that a person was deemed to be a member of the friend group, serve as proxies for the subject's actual location within what is an essentially unobservable social network (including real friends, relatives, casual contacts, and so on) through which the flu spreads by inter-personal means. Being a “friend” is a marker for a person's social-network position, whatever the path of infection to this person actually is. Of course, it is likely that measured friendship networks are related to contact networks more generally: for instance, people with more friends should come into greater contact with more strangers both directly and indirectly via their friends.