The Rufiji Health and Demographic Surveillance System (HDSS) was established in October 1998 to evaluate the impact on burden of disease of health system reforms based on locally generated data, prioritization, resource allocation and planning for essential health interventions.The Rufiji HDSS collects detailed information on health and survival and provides a framework for population-based health research of relevance to local and national health priorities. Monitoring of households and members within households is undertaken in regular 6-month cycles known as 'rounds'. Self-reported information is collected on demographic, household, socioeconomic and geographical characteristics.Verbal autopsies were done by trained Field interviewers to collect detailed data through structured and standardized INDEPTH Network verbal autopsy forms on symptoms and signs during the terminal illness, allowing assignment of cause of death following physician's review to a list of causes of death, based on the 10th Revision of the International Classification of Diseases.From 2008 to 2015 Rufiji HDSS recorded about 5500 deaths. About 90% of them were interviewed and assigned the underlying cause of death. The Ifakara Health Institute VA data portal will be periodically updated depending on the availability of new data from the field.

At the initial census (October 1998-anuary 1999), all individuals who were intending to be resident in the DSA for at least 4 months were eligible for inclusion. Verbal consent to participate in the census was sought from the head of every household.Definitions of several characteristics such as household, membership, migration and head of household are set in order to correctly assign individuals or households to events or attributes. A household in Rufiji HDSS is defined as a group of individuals sharing, or who eat from, the same cooking pot. A member of the HDSS is defined as someone who has been resident in the DSA for the preceding 4 months. New members qualify to be an in-migrant if s/he moves into the Rufiji HDSS and spends at least 4 months there.Women married to men living in the Rufiji HDSS and children born to these women qualify to be members of the Rufiji HDSS. In the case of multiple wives, the husband will be registered as a permanent resident in only one household. He will be linked to other wives by his husband identification number given to his wives. After the census, the study population is visited three times a year in cycles or updated rounds over February-May, June-September and October-January to update indicators.From July 2013 onwards, Rufiji HDSS switched to two data collection rounds per year, which happen in July-December and January-June. Mapping of households and key structures such as schools, health facilities, markets, churches and mosques was done by field interviewers using handheld global positioning systems (GPS). Updating of GPS coordinates has been an ongoing exercise especially for new structures and for demolished structures.In 2012 the population size of the DSA was about 103 503 people, residing in 19 315 households. There are several ethnic groups in the DSA. The largest is the Ndengereko; other groups include the Matumbi, Nyagatwa, Ngindo, Pogoro and Makonde. The population comprises mainly Muslims with few Christians and followers of traditional religions. The main language spoken is Kiswahili. English is not commonly used in the area.Around 75% of the population aged 7-15 years have attended primary education, 14% of those in age group 15-65 years have secondary education and only 1% of the population has tertiary education. Almost 50% of the adult population aged 15-65 are self-employed in agriculture, 28% engage in other small economic activities, 16% are selfemployed in small-scale business and 6% are unemployed. Fuel wood is the main source of energy for cooking and shallow wells are the main source of water for domestic use.The household heads in Rufiji HDSS are considered as breadwinners and most (67.3%) are male. Active community engagement programmes are in place which include key informants (KIs) days, where the HDSS team convenes meetings with KIs for presentations on recent findings to feed back to community and for distribution of newsletters to households. Community sensitization events are held at the time of introducing new studies. These initiatives have cemented good relationships with the community and eventually maintained high participation.

In Health and Demographic Surveillance System (HDSS), the follow-up of individuals aged 1559years was categorized into three periods: before ART (19982003), during ART scale-up (20042007), and after widespread availability of ART (20082011). Residents were those who never migrated within and beyond HDSS, internal migrants were those who moved within the HDSS, and external migrants were those who moved into the HDSS from outside. Mortality rates were estimated from deaths and person-years of observations calculated in each time period. Hazard ratios were estimated to compare mortality between migrants and residents. AIDS deaths were identified from verbal autopsy, and the odds ratio of dying from AIDS between migrants and residents was estimated using the multivariate logistic regression model.

Face-to-face interview