The Advisory Committee on Immunization Practices revised recommendations for serogroup B meningococcal vaccines in June 2015, stating that adolescents and young adults 16–23 years of age may be considered for a MenB vaccine series to provide short-term protection. The preferred age for MenB vaccination is 16–18 years, ideally right before entering college.1,2
Bivalent rLP2086, or Trumenba®, is a recombinant protein vaccine composed of two lipidated variants of factor H binding protein, a surface-exposed protein expressed in more than 97% of serogroup B Neisseria meningitidis (MenB).3,4 Before the development of bivalent rLP2086 and MenB-4C (Bexsero®), there were several barriers to creating a MenB vaccine. First, the polysaccharide-protein conjugate–based approach used to target N. meningitidis serogroups A, C, Y and W-135 was unsuccessful, as the capsular polysaccharide of MenB is similar to polysialic acid of human neural cells and poses a risk of autoimmune reaction. Second, the serologic correlate of protection against MenB disease, human serum bactericidal assay with human complement (hSBA), is a labor-intensive technique with considerable interlaboratory variation, and acquisition of human complement sources for all test strains is challenging.5–8 Finally, the incidence of MenB disease in the United States is low, and the large sample sizes needed for efficacy studies prohibit trials that use disease as an outcome from being performed.9,10
In 2015, the rate of MenB disease in 18–23 year olds was 0.14/100,000. Although the rate of MenB disease within this age group is similar between college and noncollege students, in the last several years the majority of outbreaks in the United States have occurred on college campuses.1 In college campus outbreaks of MenB disease in New Jersey and California, mass vaccination campaigns were held using the MenB-4C vaccine (Bexsero®).11 Bivalent rLP2086 and MenB-4C were offered to students in an Oregon college campus outbreak from January to May of 2015; however, vaccination was not mandated, resulting in a 30% coverage rate.12,13
At a college in Rhode Island, an outbreak of MenB disease occurred between February 5 and February 8, 2015. No epidemiologic link could be identified between two cases of a rare strain (ST-9069) of MenB, thus representing a 489-fold increase in MenB disease.14 Bivalent rLP2086 (Trumenba®) was chosen for a mandatory intervention, primarily because it was more readily available than MenB-4C at the time.15
There is limited experience with bivalent rLP2086 outside of clinical trials.16–28 The Rhode Island outbreak marked the first time bivalent rLP2086 was used as the sole intervention response and provided a unique opportunity to collect safety data on bivalent rLP2086 in a real-world population of college-age individuals. A 94% coverage rate was achieved at college X after the first dose vaccination clinic, and no further cases of MenB disease were observed to date.15
The present study examines the type and severity of adverse events experienced by subjects over age 18 at college X following one, two, or three doses of bivalent rLP2086 vaccine. We also aim to compare the results of our real-world study to those obtained in clinical trials.
MATERIALS AND METHODS
This study took place from February 2015 to May 2016. The sampling frame included all subjects eligible for bivalent rLP2086 vaccination. Subjects eligible for vaccination included all undergraduate students, all graduate students 26 years of age and under residing in dormitories, all persons 26 years of age and under who were in an intimate relationship with someone residing on campus, all graduate assistants and faculty 26 years of age and under, and all persons with medical conditions (e.g., complement disorders) that put them at increased risk for meningococcal disease. Exclusion criteria were subjects under 18 or over 26 years of age who had a recent illness diagnosed by a physician, as reported in the survey.1,2,13
Institutional review board approval was obtained from the Rhode Island Department of Health, Lifespan (affiliated hospital network), and the college (Fig. 1). The vaccination clinics were held at 0, 2, and 6 months, in February, April, and September 2015. Informed consent forms and deidentified vaccine safety surveys were distributed 2–4 months after vaccination episode (at the following vaccination visit). The design was longitudinal, where data were collected retrospective of the previous dose. Participants were asked to provide their name, date of birth, and an e-mail address on a form, which was separated from the survey. Participants were assigned a code number, which linked their separate identifying information to their deidentified questionnaires. Unique code numbers allowed us to track individuals’ responses over time.
At the April 2015 clinic, subjects receiving a second dose of bivalent rLP2086 were asked to complete a paper survey about events following the first dose of the vaccine during a mandatory 15-minute observation period following vaccination. In a similar fashion, at the September clinic, subjects receiving their third dose of bivalent rLP2086 were asked to complete a vaccine safety survey about events following the second dose of the vaccine. Using Qualtrics® software, an identical electronic survey was created and e-mailed to subjects approximately 2 months after the third (September) vaccination clinic, to capture data on any adverse events that may have occurred after the third dose of vaccine.
The electronic survey was authenticated, meaning access was controlled and limited to a set number of participants. Only e-mail addresses previously provided by students on paper surveys were used. Each participant completing the electronic survey received a $1 Amazon Gift card within 24 hours. An e-mail was sent out a week before distribution of the electronic survey notifying subjects of the upcoming opportunity. The electronic survey was then sent out weekly on different days and at different times over a period of 3 weeks. Advertisements for the electronic survey were displayed during vaccination clinics. These strategies were employed to maximize response rates and secure a large number of potential participants.29–31
In September 2015, an opportunity to enroll additional patients in our study arose when vaccination with bivalent rLP2086 was mandated for all incoming freshmen. This was decided following the results of a meningococcal carriage study at college X, which identified the outbreak strain, and showed no decrease in carriage rate following vaccination with bivalent rLP2086,14 thereby identifying no evidence of herd protection for incoming freshmen.
A 46-question, closed-ended, multiple-choice survey (Fig., Supplemental Digital Content 1, http://links.lww.com/INF/C825) was designed using survey methodology as recommended by Dillman.29–31 Subjects were asked about side effects most commonly addressed in clinical trials:16–28 pain at the injection site, fatigue, headache, myalgia, fevers, and chills. Subjects were asked about recent illnesses diagnosed by a physician, since such diagnoses could explain symptoms independent of the vaccine. Those who replied “yes” were excluded from our study. The survey also addressed serious adverse events, including allergic reactions and recent hospitalizations.
A systematic literature review of bivalent rLP2086 safety was performed using the search terms “bivalent rLP2086” and “meningococcal serogroup B vaccine” on PubMed, Web of Science, and the Cochrane Library. Date range limitations were not applied. No attempt was made to obtain information about unpublished studies. Studies assessing safety profiles in a 1–6 month period following vaccination were included. Studies were included if the sample was adolescent or adult-age, 120 µg formulation of bivalent rLP2086 was used (standard for U.S. vaccines), data were described numerically and categorized by event severity, and at least 1 subgroup received three doses of bivalent rLP2086 without concomitant vaccine administration. Studies excluded were in a language other than English, letters, abstracts, and reviews.
In total, 13 published clinical trials were found assessing the safety of bivalent rLP2086 in humans from 2012 to 2016. All 13 trials assess vaccine safety as a secondary outcome, with immunogenicity as the primary endpoint.16–28 Six of these clinical trials were used for comparison. We obtained safety data using a standardized data extraction form in Microsoft Excel and analyzed data using SAS 9.4. The methods used in the six trials to monitor local and systemic reactions were consistent, in that each subject recorded adverse events in an electronic diary for 7 days after vaccination. However, monitoring of serious adverse events was inconsistent between trials: time of reporting/follow-up ranged from 1 to 6 months after third vaccination.16,19,22–24,26
Within these six trials, the study by Marshall et al24 dosed bivalent rLP2086 at 0, 1, and 6 months, differing from the 0-, 2-, and 6-month schedule of the present study. Thus, adverse events following second and third vaccinations were not directly comparable with ours.24 Results were ultimately included, as this study represented one of the few adult samples available. Reiner et al26 and Sheldon et al23 also provided small numbers of adults with which to compare our results. The trials by Richmond et al19 and Vesikari et al16 had samples of 11–18 year olds; it was unclear what proportion of subjects were 18 years old. Only the group receiving bivalent rLP2086 at the 0-, 2-, 6-month interval from the study by Vesikari et al16 was used for comparison. The study by Senders et al22 was an exclusively adolescent population but provided the largest n for comparison to our data.
Seven clinical trials were omitted from our comparison chart. Two focused on co-administration of bivalent rLP2086 with other vaccines. Neither trial contained a group receiving only bivalent rLP2086 and saline for all three doses, thus results were not eligible for direct comparison.17,21 A large multicenter phase III trial done at various sites in Denmark, the United States, and the United Kingdom by Ostergaard et al25 did not describe safety data per dose. Three phase I/II trials were excluded from comparison: two did not use 120 µg vaccine dosing, the standard formulation for bivalent rLP2086 in the United States,20,28 and the other trial was done in an infant population.27 Likewise, a phase I trial done in Australia was excluded as the sample was 18–36 months of age and 120 µg vaccine dosing was not used.18
Data and Statistical Analysis
Survey data from the present study were stored in REDCap.32 All analyses were conducted using SAS 9.4 unless otherwise specified. Rates of adverse events were modeled over the three time points using generalized mixed modeling assuming a binary distribution, where reported adverse events were nested within students; severity of adverse events were modeled in the same fashion, but assuming a binomial distribution between 1 and 3. For both, PROC GLIMMIX was used with a logit link and classic sandwich estimation to correct for possible model misspecification. In addition, confidence intervals for the results reported from clinical trials were calculated with the Clopper-Pearson method using R 3.133 statistical software and the BINOM package.34 All interval estimates were calculated for 95% confidence.
As seen in Figure 2, 3,525 of 3,745 subjects eligible for vaccination received dose 1 of bivalent rLP2086 in February 2015 (94% coverage rate), and in September 2015, 893 of 1,050 incoming freshmen received dose 1 of bivalent rLP2086 (85% coverage rate). In April 2015, 2,988 eligible subjects returned for dose 2 (80% coverage rate), and in November 2015, 543 freshmen received dose 2 of bivalent rLP2086 (52% coverage rate). In September 2015, 1,670 eligible subjects returned for dose 3 (56.7% coverage rate), and 454 freshmen received dose 3 in March 2016 (43% coverage rate).
In total, the survey response rate for the first dose was 45% (n = 1,736/3,865), and 49% (n = 1,395/2,124) for the second dose. A total of 1,712 students provided us with a legible e-mail address during vaccination clinics. The third dose response rate via electronic survey was 25% (n = 609/1,712). Subject responses were tracked over time, where 1,081 subjects completed one or more surveys. Four hundred seventy-four subjects answered surveys about doses 1 and 2,226 subjects answered surveys about doses 1 and 3, and 161 subjects answered surveys for doses 2 and 3. A total of 220 subjects completed all three surveys. Forty-nine subjects were excluded from our survey study, as they reported illnesses following vaccination diagnosed by a physician. The majority of illnesses were felt to cause similar events as those reported following vaccination and included pneumonia, gastroenteritis, bronchitis, strep throat, influenza, mononucleosis, and sinusitis.
Our sample was 67.4% female with a mean age of 19 years and 7 months (Table 1). None of the subjects reported pregnancy (recent or current) or breastfeeding.
Table 2 summarizes the percentage of subjects with local, systemic, and serious adverse events seen in our sample following each dose of vaccine; the mean and 95% CI for each dose are also provided. Severe adverse events were reported in a small percentage of subjects. The most commonly reported adverse event was injection site pain. Greater than 94% of subjects reported resolution of injection site pain, fatigue, headache, myalgia, fever, and/or chills within 7 days.
Six of 13 published clinical trials assessing the safety of bivalent rLP2086 in humans were used for comparison.16,19,22–24,26 Table, Supplemental Digital Content 2, http://links.lww.com/INF/C826, is a table describing characteristics of these trials in greater detail.
Table 3 describes the percentage of subjects with local and systemic adverse reactions following three doses of vaccine in this study as compared with the six clinical trials. We found an overall lower rate of headache within our sample relative to the rates in reported clinical trials. The rates of injection site pain, fatigue, thermometer-confirmed fevers (≥ 100.4°F) and chills were approximately similar in our study as in clinical trials. The rate of myalgia was higher for our sample than in clinical trials following the first and second doses of the vaccine, but similar after the third dose of vaccine (Fig., Supplemental Digital Content 3, http://links.lww.com/INF/C827 illustrates our findings as compared with clinical trials in graphical form).
Reported side effects from this sample were generally lower than those reported in the larger clinical trials,16,19,22,24 which may be due to attrition. Our results suggest that bivalent rLP2086 was well tolerated among subjects who returned for booster doses. Similar to published clinical trials of this vaccine,16,19,22–24,26 the majority of adverse events were local (pain at the injection site) or systemic events mild to moderate in severity. Most adverse events were transient, lasting less than 7 days after vaccination. Potentiation was not evident with subsequent dosing. More adverse events of moderate severity were experienced after the first dose, which is likely due to initial introduction of the antigen, with desensitization occurring in subsequent doses. Severe adverse events were relatively uncommon.
There was an interesting discrepancy between reported fevers and thermometer-confirmed fevers in our study. When subjects were simply asked if they were febrile following vaccination, many answered “yes.” However, many subjects denied having taken their temperature with a thermometer, and/or that it was ≥ 100.4°F. Our rates of confirmed fever were similar to clinical trials, where all temperatures were thermometer-confirmed.
Our “in-person” survey method (following doses 1 and 2) of approaching students and physically handing them questionnaires, instead of mailing them out electronically, proved advantageous. It allowed subjects to ask us about survey questions they found confusing. Our 45–50% response rate for “in-person” surveys was attributed to several basic methods put forth by Dillman and other survey strategists.29–31 The drop in response rate seen in transitioning to electronic survey methods was expected, which was why the tangible reward (the $1 Amazon gift card) was offered. Incentives as small as $1–2 have been shown to increase both response quality and rates by 5–8%.35
While clinical trials of bivalent rLP2086 have stringent inclusion and exclusion criteria, our study was performed in a real-world setting where more than 90% of a college-age population was vaccinated with at least one dose. This is the first time safety data have been collected for this vaccine outside of clinical trials.
Our study had several limitations, including a potential nonresponse bias (i.e., differences in response between those who answered surveys and nonresponders). Subjects who received their first vaccine and did not return for subsequent doses (perhaps secondary to adverse events experienced) had no chance of being surveyed, and contact with students was limited to vaccination clinics. Although we could not completely avoid the limitation of selection bias (i.e., patients who experienced syncope after vaccination likely did not elect to participate in the survey), we do not anticipate an appreciable effect from what minimal selection bias exists within our sample. The time periods between vaccination clinics made recall bias another potential limitation. Recall bias may play a role in explaining lower rates for more subjective symptoms, such as fatigue or headache. Participants who did not carefully read questions, or were distracted by friends while taking a survey, were subject to response bias (giving inconsistent or blatantly false answers). Finally, a proportion of subjects reported redness and swelling at the injection site in the section of the survey asking about other symptoms; however, local reactions were not actively addressed with multiple choice questions.
Annuals variations of influenza, streptococcal sore throat, Epstein-Barr virus, and other illnesses mimicking adverse events of the vaccine (causing fever, fatigue, headaches, and myalgia) were a consideration, particularly as various barriers to seeking medical attention have been described in college-age students.36,37 Our response rate declined as the day went on; subjects who arrived earlier to vaccination clinics were more amenable to filling out questionnaires than latecomers. Reported rates of side effects may also be influenced by pluralistic ignorance, where subjects assume the symptoms of their peers to be rare and therefore underreport their own symptoms.38 Fatigue, one of the reportable adverse events of this vaccine, is also quite prevalent within the college-age population,39 thus falsely inflated reporting of this symptom within this sample is a potential limitation. This may help account for the high rate of severe fatigue seen within this population, especially following dose 2.
Our study represents the first collection of safety data for bivalent rLP2086 in a real-world, college-age population where more than 90% of eligible subjects were vaccinated. Overall, rates of local and systemic side effects reported following bivalent rLP2086 were lower or similar in our study than reported in clinical trials.
Koren Kanadanian, David Frinquelli, Ftr. Kenneth Sicard, Heidi Soeters, Manisha Patel, Melissa Clark, Michael Kraten, Todd Olszewski, Kerry LaPlante, Sarah Rhoad, Samantha DeAndrade, Christine Goulette, Steven Sears, Kristine Goodwin, Michael Smit, Ian Michelow, Zaid Alhinai, Kathryn Wilson, Sabina Holland, Allison Caldwell, Jason Mandell, Marianne Monahan, Gary Levy, Mark Fiorito, Matthew Fiorito, Ayesha Saya, Eric Chow, Jason Machan, Justin Cuomo, and the very kind students, faculty, and staff at college X.
1. MacNeil JR, Rubin L, Folaranmi T, et alUse of serogroup B meningococcal vaccines in adolescents and young adults: Recommendations of the Advisory Committee on Immunization Practices, 2015. MMWR Morb Mortal Wkly Rep. 2015;64:1171–1176.
2. Folaranmi T, Rubin L, Martin SW, et alCenters for Disease Control (CDC). Use of serogroup B meningococcal vaccines in persons aged ≥10 years at increased risk for serogroup B meningococcal disease: Recommendations of the Advisory Committee on Immunization Practices, 2015. MMWR Morb Mortal Wkly Rep. 2015;64:608–612.
3. Gandhi A, Balmer P, York LJCharacteristics of a new meningococcal serogroup B vaccine, bivalent rLP2086
(MenB-FHbp; Trumenba®). Postgrad Med. 2016;128:548–556.
4. Serruto D, Bottomley MJ, Ram S, et alThe new multicomponent vaccine against meningococcal serogroup B, 4CMenB: Immunological, functional and structural characterization of the antigens. Vaccine. 2012;30 Suppl 2:B87–B97.
5. Budroni S, Kleinschmidt A, Boucher P, et alPooled-sera hSBA titres predict individual seroprotection in infants and toddlers vaccinated with 4CMenB. Vaccine. 2016;34:2579–2584.
6. Frosi G, Biolchi A, Lo Sapio M, et alBactericidal antibody against a representative epidemiological meningococcal serogroup B panel confirms that MATS underestimates 4CMenB vaccine strain coverage. Vaccine. 2013;31:4968–4974.
7. Borrow R, Carlone GM, Rosenstein N, et alNeisseria meningitidis group B correlates of protection and assay standardization—international meeting report Emory University, Atlanta, Georgia, United States, 16-17 March 2005. Vaccine. 2006;24:5093–5107.
8. Shea MWThe long road to an effective vaccine for meningococcus group B (MenB). Ann Med Surg (Lond). 2013;2:53–56.
9. Caesar NM, Myers KA, Fan XNeisseria meningitidis serogroup B vaccine development. Microb Pathog. 2013;57:33–40.
11. Schaffner W, Baker CJ, Bozof L, et alAddressing the challenges of serogroup B meningococcal disease outbreaks on campuses. Infect Dis Clin Prac. 2014;22:245–252.
13. Harrison LHVaccines for prevention of group B meningococcal disease: Not your father’s vaccines. Vaccine. 2015;33 Suppl 4:D32–D38.
14. Soeters HM, McNamara LA, Whaley M, et alCenters for Disease Control (CDC). Serogroup B meningococcal disease outbreak
and carriage evaluation at a college—Rhode Island, 2015. MMWR Morb Mortal Wkly Rep. 2015;64:606–607.
15. Fiorito T, Mihalakos A, Alexander-Scott N, et alRapid response to a Rhode Island college outbreak
of meningococcal serogroup B disease: Nation’s first widespread use of Trumenba®
vaccine. Open Forum Infect Dis. 2015;2:S447.
16. Vesikari T, Østergaard L, Diez-Domingo J, et alMeningococcal serogroup B bivalent rLP2086
vaccine elicits broad and robust serum bactericidal responses in healthy adolescents. J Pediatric Infect Dis Soc. 2016;5:152–160.
17. Vesikari T, Wysocki J, Beeslaar J, et alImmunogenicity, safety, and tolerability of bivalent rLP2086
meningococcal group B vaccine administered concomitantly with diphtheria, tetanus, and acellular pertussis and inactivated poliomyelitis vaccines to healthy adolescents. J Pediatric Infect Dis Soc. 2016;5:180–187.
18. Marshall HS, Richmond PC, Nissen MD, et alSafety and immunogenicity of a meningococcal B bivalent rLP2086
vaccine in healthy toddlers aged 18-36 months: a phase 1 randomized-controlled clinical trial. Pediatr Infect Dis J. 2012;31:1061–1068.
19. Richmond PC, Marshall HS, Nissen MD, et al2001 Study Investigators. Safety, immunogenicity, and tolerability of meningococcal serogroup B bivalent recombinant lipoprotein 2086 vaccine in healthy adolescents: A randomised, single-blind, placebo-controlled, phase 2 trial. Lancet Infect Dis. 2012;12:597–607.
20. Richmond PC, Nissen MD, Marshall HS, et alA bivalent Neisseria meningitidis recombinant lipidated factor H binding protein vaccine in young adults: Results of a randomised, controlled, dose-escalation phase 1 trial. Vaccine. 2012;30:6163–6174.
21. Muse D, Christensen S, Bhuyan P, et alA phase 2, randomized, active-controlled, observer-blinded study to assess the immunogenicity, tolerability and safety of bivalent rLP2086
, a meningococcal serogroup B vaccine, coadministered with tetanus, diphtheria and acellular pertussis vaccine and serogroup A, C, Y and W-135 meningococcal conjugate vaccine in healthy US adolescents. Pediatr Infect Dis J. 2016;35:673–682.
22. Senders S, Bhuyan P, Jiang Q, et alImmunogenicity, tolerability and safety in adolescents of bivalent rLP2086
, a meningococcal serogroup B vaccine, coadministered with quadrivalent human papilloma virus vaccine. Pediatr Infect Dis J. 2016;35:548–554.
23. Sheldon EA, Schwartz H, Jiang Q, et alA phase 1, randomized, open-label, active-controlled trial to assess the safety of a meningococcal serogroup B bivalent rLP2086
vaccine in healthy adults. Hum Vaccin Immunother. 2012;8:888–895.
24. Marshall HS, Richmond PC, Nissen MD, et alA phase 2 open-label safety and immunogenicity study of a meningococcal B bivalent rLP2086
vaccine in healthy adults. Vaccine. 2013;31:1569–1575.
25. Ostergaard L, Lucksinger GH, Absalon J, et alA phase 3, randomized, active-controlled study to assess the safety and tolerability of meningococcal serogroup B vaccine bivalent rLP2086
in healthy adolescents and young adults. Vaccine. 2016;34:1465–1471.
26. Reiner DM, Bhuyan P, Eiden JJ, et alImmunogenicity, safety, and tolerability of the meningococcal serogroup B bivalent rLP2086
vaccine in adult laboratory workers. Vaccine. 2016;34:809–813.
27. Martinon-Torres F, Gimenez-Sanchez F, Bernaola-Iturbe E, et alA randomized, phase ½ trial of the safety, tolerability, and immunogenicity of bivalent rLP2086
meningococcal B vaccine in healthy infants. Vaccine. 2014;32:5206–5211.
28. Nissen MD, Marshall HS, Richmond PC, et alA randomized, controlled, phase ½ trial of a Neisseria meningitidis serogroup B bivalent rLP2086
vaccine in healthy children and adolescents. Pediatr Infect Dis J. 2013;32:364–371.
29. Dillman DA, Smyth JD, Christian LMInternet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method. 2014:4th ed. New York, NY: John Wiley & Sons, Inc.15–41.
30. Dillman DAThe design and administration of mail surveys. Ann Rev Soc. 1991;17:225–249.
31. Salant P, Dillman DAHow to Conduct Your Own Survey. 1994:New York, NY: John Wiley & Sons, Inc.33–77.
32. Harris PA, Taylor R, Thielke R, et alResearch electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
33. R Core Team R. A language and environment for statistical computing. R Foundation for Statistical Computing. 2014. Vienna, Austria. Available at: http://www.R-project.org
. Accessed October 30, 2016.
34. Dorai-Raj SBINOM: Binomial confidence intervals for several parameterizations. R package version 1.1–1. 2014. Available at: http://CRAN.R-project.org/package=binom
. Accessed October 30, 2016.
35. James JM, Bolstein RThe effect of monetary incentives and follow-up mailings on the response rate and response quality in mail surveys. Public Opin Q. 1990;54:346–361.
36. Davies J, McCrae BP, Frank J, et alIdentifying male college students’ perceived health needs, barriers to seeking help, and recommendations to help men adopt healthier lifestyles. J Am Coll Health. 2000;48:259–267.
37. Barth KR, Cook RL, Downs JS, et alSocial stigma and negative consequences: Factors that influence college students’ decisions to seek testing for sexually transmitted infections. J Am Coll Health. 2002;50:153–159.
38. Bjerring JC, Hansen JU, Pedesron NLOn the rationality of pluralistic ignorance. Synthese. 2014;191:2445–2470.
39. Hershner SD, Chervin RDCauses and consequences of sleepiness among college students. Nat Sci Sleep. 2014;6:73–84.