“When disorders are socially unacceptable, such as insanity they tend to be under-reported.” (Sackett et al. 1979) Unacceptable disease bias is a subtype of reporting bias in that the conditions, due to shame or embarrassment, are not entirely reported. This reduces the validity of data on such conditions and may make the comparison in studies difficult.
This bias is potentially responsible for discrepancies in cancer mortality rates across countries, whereby cultural factors influence the recording of causes of death and in some cases the recording of cancer at all. Unacceptable exposure bias is similar as individuals may under-report exposures regarded as socially undesirable.
In the Whitehall II cohort study, psychiatric reasons for absence from work were investigated. For a subgroup, these records were cross-checked with information from general practitioners records (Stansfeld et al. 1995). These showed an under-reporting of psychosis (2% in the work records versus 15% in the GP records) for sickness absence. The authors commented, “This major source of under-reporting is most likely related to the stigma of a diagnosis of psychotic illness especially in relation to work prospects.”
The Whitehall II study shows that impact of unacceptable disease bias can be large: only 13% of the population reported their psychiatric reasons for absence.
In questionnaires, unacceptable questions should come toward the end so that they will not affect other questions. (Choi BC et al. 2005) Anonymised responses (i.e. through mailed questionnaires) may overcome some issues that arise due to unacceptable disease bias.