ca‐Livingston 2013.

Study characteristics
Methods	Retrospective cohort ‐ vaccine effectiveness in households
Participants	2176 household residents Between 5 February 2010 and 8 April 2010, 473 index households were contacted for follow‐up. Data were collected using a standard script. An interviewer requested to speak with an adult, who provided information on each household member. A minimum of 3 call attempts were made to each household. During calls, the following information was requested: (1) whether each household contact slept at home on average at least 5 nights per week; (2) total number of bedrooms in the house; and (3) for each household contact: birth date, vaccination status, and whether they had been sick with either cheek swelling that had lasted for at least 2 days or a doctor‐diagnosed case of mumps since September 2009. Households with index cases identified through surveillance from 1 September 2009 to 31 December 2009 were eligible for study inclusion. Case households were excluded if: (1) the index case lived alone; (2) the index case did not live in the house (e.g. lived in a dormitory); (3) the index case did not sleep in the house on average at least 5 nights per week; (4) there was no English‐speaking adult in the household; (5) an adult in the household was not able to be contacted; or (6) an adult in the household refused to provide information on household contacts or provided incomplete information.
Interventions	Mumps vaccination status was based on documented, valid MMR doses (2 doses). Acceptable documentation included MMR doses recorded in the New York City Citywide Immunization Registry (CIR) or those obtained directly from individual medical provider.
Outcomes	A case of mumps was defined as 1 meeting the Council of State and Territorial Epidemiologist (CSTE) surveillance case definition or a compatible case identified via the phone interview. An index case was defined as the first case in a household to be reported to the DOHMH. Primary cases were those with the earliest onset of mumps in the household. Household members were defined as being exposed 2 days before parotitis onset of the primary case, which is the first day that the primary case was infectious. We defined co‐primary cases as those with onset within 9 days after the primary case’s symptom onset. Secondary cases were defined as those reporting onset of mumps 10 to 25 days after the primary case. Non‐secondary cases were defined as those occurring more than 1 incubation period (> 25 days) after the primary case. The clinical case definition is acute onset of unilateral or bilateral swelling of the parotid or other salivary glands, lasting 2 or more days, and without other apparent cause. Index cases in households were identified through mandated electronic reporting of positive test results by laboratories, or clinical reports of suspect disease by providers.
Funding Source	Government
Notes	In order to be valid, doses had to be administered in accordance with the recommended vaccination schedule guidelines, meaning the first dose had to be administered no earlier than 4 days before the first birthday and subsequent doses at least 28 days after a previous MMR dose. Individuals lacking MMR documentation from a medical provider and with a record in CIR with at least 1 reported vaccination, but no recorded MMR doses, were considered unvaccinated with MMR. Individuals with a valid provider recorder with no recorded MMR doses were also considered unvaccinated. Individuals lacking MMR documentation from a medical provider and with no recorded vaccinations in CIR were considered to have unknown MMR vaccination status. Vaccination coverage estimates are exclusive to households with known mumps disease, and coverage in the overall Orthodox Jewish community may differ. In addition, the study was conducted during a community‐wide outbreak, so exposure to mumps may have occurred in other settings besides the home. We did not investigate specific exposures during religious holidays and community celebrations when members of the affected community may have had close contact.
Risk of bias
Bias	Authors' judgement	Support for judgement
PCS/RCS ‐ exposed cohort selection	Low risk	Adequate ‐ secure record
PCS/RCS ‐ non‐exposed cohort selection	Low risk	Adequate ‐ secure record
PCS/RCS ‐ comparability	Unclear risk	Amongst secondary cases, 15% were reported by the head of household. These cases were not confirmed by investigation or medical record review and may not have fulfilled the CSTE case definition. The time between the index case onset and the follow‐up interview may have led to cases being missed due to poor recall.
PCS/RCS ‐ assessment of outcome	Low risk	Adequate
Summary Risk of Bias assessment	Unclear risk	We had concerns regarding at least 1 domain such that some doubt is raised about the results.