Nigeria - National Survey on Drug Use and Health 2016-2017, First round

A two-stage cluster sample design was adopted for the survey.
First stage selection (Selection of EAs) The first stage involves selection of primary sampling units. In each State 10 EAs were canvassed, Six (6) Urban and three (4) Rural EAs. This gives a total of 60 EAs to be covered in all the six (6) states. The second stage selection (selection of households) Eight (8) households (HHs) will be selected in each EA. A total of 80 HHs will be covered in each state and subsequently, a total of 480 HHs will be studied in all the six states. Conditions and procedures for replacement of EA In order to address some extreme cases of non coverage of the selected EAs, the following procedures will be adopted: 1) Extra five EAs will be provided for each state in case of any replacement 2) The supervisor and State officer must be contacted before any replacement is made 3) Replacement can be made if the EA is not accessible or if it is located in a conflict or crisis area, i.e. areas classified as high risk areas

The main survey of the National Health and Drug Use also had two (2) sampling procedures
First Stage Selection was the selection of EAs. In each State 70 EAs were canvassed giving a total of 2,590 EAs covered in all the States including FCT.

The Second Stage Selection involved selection of Housing Units (HUs). 15HUs were selected in each EA and 1 Household was interviewed per HU.
In a HU with more than one (1) household, the Kish grid method of selection was applied to determine the household to be interviewed. A total of 1050 household interviews were conducted in each state. A total of 38,850 HHs were covered nationwide.

The Sampling strategy for the problem drug users constituted a hidden and hard-to-reach population and the absence of a census-based sampling frame maked it difficult to estimate the parameters of this population in Nigeria. In view of this and other constraints, the snowball method (also known as chain referral sampling) was used to select the respondents for this study. Snowball sampling is particularly suited for a study of this nature because, among other things, it helps in tracing the diffusion of information within social networks and in delineating patterns of association.
Using the drug user questionnaire, data was collected from a sample of 200-250 high-risk drug users in each state, yielding a national sample of between 7400 and 9250 respondents.

For the key informant survey, a minimum of 50 -75 Key informants in each State were interviewed, yielding a national sample of between 1850 and 2775 key informants. On average, for each category of key informants, 9-10 interviews were conducted.


No deviation for the Household survey on drug use and health in Nigeria. In the problem drug users data collection process, It is important to note that the report from Ogun state indicates that in some instances venue-based sampling (VBS) was used instead of RDS and in Northeast Zone A. Some cannabis users were interviewed contrary to the mandate.


86% response rate was recorded for the Key Informant Survey while 98.1% was reported for the Problem Drug Use Assessment Survey.


Population and Sample weights were calculated for the Household data files.