A comparative analysis of disability measures in Cameroonian surveys.

Background: Although identifying vulnerable groups is an important step in shaping appropriate and efficient policies for targeting populations of disabled people, it remains a challenge. This study aims to evaluate for the first time the comparability of the different disability measurements used in Cameroon. This is done by comparing them with the international standards proposed by the Washington Group (WG). It also evaluates the consistency of the association between the disability as measured by these surveys and the sociodemographic characteristics.

Method: We used data from the third Cameroonian Population and Housing Census (3RGPH) of 2005, the third Cameroonian Household survey (ECAM3) of 2007, the Demographic Health and Multiple Indicator Cluster Survey (DHS-MICS) of 2011 and a survey conducted on adults in Yaoundé (HandiVIH) in 2015 with the WG tool. The proportion and their confidence intervals, chi-square tests and multivariate logistic regressions are used for analyses.

Results: In the city of Yaoundé and for the 15-49 age group, disability prevalence was estimated at 3.6% (CI = [2.5, 5.1]), 2.7% CI = [2.1, 3.5]), 2.6% (CI = [2.4, 2.7]) and 1.0% (CI = [1.0, 1.10]), according to DHS-MICS, ECAM3, HandiVIH and 3RGPH, respectively. The prevalence of severe motor and mental disabilities in DHS-MICS (0.4% CI = [0.2, 0.8], 1.1% CI = [0.7, 1.8] and 0.5% CI = [0.2, 1.1], respectively) are not significantly different from the findings of HandiVIH (0.3% CI = [0.2, 0.3], 0.8% CI = [0.7, 0.9] and 0.5% CI = [0.5, 0.6], respectively). Only motor disability prevalence in ECAM3 (0.8%, CI = [0.5, 1.2]) is not different from that of HandiVIH. When the WG screening tool is used in HandiVIH, disability is positively associated with age, negatively associated with educational level, being in a union and socioeconomic status (SES) and it is not associated with sex. Severe disability, for its part, is not associated with SES and is positively associated with being a male. A different association trend is observed with 3RGPH, ECAM3 and DHS-MICS.

Conclusion: None of the instruments used in the nationally representative Cameroonian surveys produced both disability prevalence and association trends that are exactly similar to those obtained when using the WG disability screening tool, thus highlighting the necessity to include the WG questions in nationally representative surveys.