Table 2 Column guide
From: A framework for indirect elicitation of the public health impact of gambling problems
Column | Description |
|---|---|
1. Disorder | The specific comorbid disorder |
2. Number of estimates & 3. Mean comorbid prevalence (%) | These two figures/columns should be interpreted in conjunction. Column 2 is a count of individual estimates that were used to estimate the mean prevalence of co-morbidity. For example, the mean co-morbid prevalence of ‘alcohol abuse’ was derived using nine studies. Column 3 is the actual mean comorbid prevalence. For example, the co-morbidity of ‘alcohol abuse’ among problem gamblers was estimated to be 18.2%. The figures in this column were derived from Dowling et al. [71, 73] who examined treatment seeking problem gamblers, and Lorrains et al. [72] who examined community samples of problem gamblers. |
4. Community prevalence (%) | This figure is the rate of the disorder observed in the general population irrespective of problem gambling status (e.g. 8.5% community prevalence for any alcohol use disorder) |
5. RR (SE) | Relative risk (RR) is the likelihood of having a specific co-morbid disorder for a problem gambler, compared to the general population. E.g. The rate of alcohol abuse is almost 4x higher among problem gamblers than in the general population. This calculation was based on estimates from previous research, and associated standard error (SE) rates are approximated by propagating uncertainty for both the numerator and denominator, using a first-order Taylor expansion \( {\sigma}_f\approx \left|f\right|\sqrt{{\left(\frac{\sigma_A}{A}\right)}^2+{\left(\frac{\sigma_B}{B}\right)}^2-2\frac{\sigma_{AB}}{AB}} \) where \( f=\frac{A}{B} \) and A and B represent the probability of a gambler and the general population to have the condition, respectively. We assume the covariance term to be zero. |
6. DW | Disability weights (DW) quantify the health loss associated with an outcome and are measured on a scale from 0 (indicating full health) to 1 (a state equivalent to death) [36] |
Other notes | • Where possible, 95% confidence intervals for estimates are presented in square brackets • A dash in any cell ‘-’ indicates that piece of information was not able to be obtained • The information was obtained from a wide range of sources. Due to methodological variations between studies (e.g. diagnostic tools used) the figures should be interpreted with caution when comparing. A discussion of these issues will follow later in this paper. |