Study 1: Cross sectional survey


The GP practices where we conducted the research were randomly selected following stratification according to socio-demographic profile, ethnic diversity and rurality/urbanity. A total of 16 practices took part in the research, and 1430 men completed the Provide questionnaire.


Part 1 of the PROVIDE Survey contains a number of sections which ask men about the following:

Part 2 of the survey allows much more detailed questions, including types of abusive emotional, physical, and sexual behaviours which respondents might have experienced and/or perpetrated. This gives us an opportunity to look at what types of behaviours respondents consider to be abusive.

The abuse and impact questions within the PROVIDE Survey were developed from the COHSAR survey devised by Hester and colleagues (Hester, Donovan & Fahmy, 2010; McCarry, Hester & Donovan, 2008) to measure and compare the extent and nature of domestic abuse experienced in same sex and heterosexual relationships. The survey does not assume that certain behaviours are experienced as abusive by respondents and uses impact measures to ascertain abuse thresholds as well as asking participants to define those experiences themselves. The approach is very statistically reliable and valid.

We used a standard measure of depression, alcohol intake, and drug taking, and used the LARA cost benefit analysis as an initial basis for the questions about services used.


Men who were surveyed were also asked if they would be willing to take part in an interview to ascertain their views on the survey, their experiences of potentially abusive behaviours, and their views on help-seeking, in particular the role that healthcare professionals might play.

Medical Records

Survey respondents who agreed to complete the survey were also asked to consent to giving the research team access to their medical records.  The purpose of accessing medical records is to examine whether incidents of potentially abusive behaviours have been recorded by practitioners, and to see how much respondents have used health services.

Edit this page