This study shows that the CES-D and AUDIT are reliable and valid instruments to use among TB and HIV patients in primary care. Using a singular construct to test for internal consistency, we found that the Cronbach's alpha was 0.84 for the CES-D and 0.98 for the AUDIT. This indicates that the participants showed adequate consistency in their responses. These high estimates are similar to previous studies performed on the CES-D  and the AUDIT . The four-factor model for the CES-D did not fit the data well. This means that the latent four factors in the CES-D were mis-specified. The two-factor model for the AUDIT showed the desired goodness of fit. This indicates that the two factors in the AUDIT could be considered as subscales.
The AUROCs for the CES-D and AUDIT (for both women and men) were high in detecting current MDD and AUDs from non-cases respectively. These findings are in keeping with a validation study conducted among HIV-infected person in South Africa that found the CES-D and AUDIT performed well in accurately discriminating MINI-defined current MDD (AUROC curve 0.76) and AUD (AUROC 0.96) respectively . The better accuracy of the AUDIT in women agrees with other studies [39, 40].
A highly sensitive test is needed for screening examinations in routine clinical care to identify potential cases, while a highly specific test is best in a confirmatory role. Of the cases identified by a screening test, few should be false positives (i.e. have high PPV) so that the expense and morbidity of further evaluation of false positive results are reduced in settings that already have limited resources . In our study, both the CES-D and AUDIT met the criterion of having cut-off scores that simultaneously have moderate to high sensitivities and PPVs. At a cut-off of 22, the CES-D yielded a sensitivity of 73% and PPV of 76% for current MDD. Similarly, at a cut-of score of 24 for women and 20 for men, the AUDIT yielded a sensitivity of 60% and PPV of 60% for women, and a sensitivity of 55% and PPV of 50% men for AUDs. The sensitivities of the AUDIT were moderate (55% sensitivity for men; 60% for women), meaning that 55-60% of the true AUD cases were identified. Also, the PPVs of the AUDIT were moderate (50% for men; 60% for women), indicating that those who screened positive about half were actually cases. The cut-offs were high compared to the CES-D cut-off of 16  and AUDIT cut-offs of 8  found in the general population. This discrepancy may indicate that our study population may have a greater likelihood of having current MDD and hazardous alcohol drinking than the general population. The high cut-offs may also reflect greater severity of current MDD and alcohol problems among our study participants; these may need intensive interventions.
Despite the available infrastructure for psychiatric admissions and outpatient care, most health facilities in Zambia do not have adequate health workers to treat depression and alcohol use disorders. We therefore recommend that individuals with high AUDIT or CES-D scores in this setting be offered treatment in accordance with the WHO Mental Health Gap Action Programme (mhGAP) . The mhGAP is a tool designed by the WHO to be used in PHC settings where health workers have limited training in Psychiatry. The mhGAP guidelines for depression include offering psychoeducation to the patient on the importance of continuing activities that used to be interesting for them and maintaining regular sleep cycles; physical activity; social activity and scheduled visits with the primary care professional when thoughts of suicide or self-harm arise. The guidelines also indicate the need to address the current psycho-social stressors for the patient by giving them the opportunity to talk about what they think are the causes of the symptoms they have, and by identifying family members who could help them solve these stressors. Furthermore, they indicate the need to identify the patient's prior physical activities, so that if these activities are re-initiated, they would have the potential for providing psycho-social support. Lastly, the guidelines indicate that if cognitive behaviour therapy (CBT) is available, it should be used on patient during scheduled visits at the clinic.
The mhGAP guidelines for those with alcohol use disorders include discussing with the patient the short and long-term risks of continued use of alcohol; asking about other substance use; having a discussion about their reasons for alcohol use, and providing examples of ways that the harmful or hazardous use of alcohol could be reduced. If the patient fails to respond or is suspected to have alcohol dependence, they should be referred to a specialist for further diagnostic evaluation and possible treatment for alcohol dependence. For those who score lower on the AUDIT, a Brief Drinker Profile  can be performed which measures quality and frequency of drinking in the previous month, and advice given on the effects of alcohol consumption on medication.
Generalisability of our findings is limited to TB and HIV patients on treatment in PHC centres. Further measures of depression and AUDs at a general population level in Zambia may be needed so that the diagnostic accuracy of CES-D and AUDIT test results among patients with depression and AUDs can be compared to those without these disorders.