Reporting of cohort outcomes

Clinical description

• Cohort analyses are needed to measure outcomes of patients in follow-up.
• In principle, the outcome status of any patient ever registered can change at any time, unless they

have died. Therefore, the records of all patients ever registered must be reviewed each time a cumulative cohort outcome analysis is done. Current outcome data cannot be obtained by addition from the previous quarterly outcome data.

• Patient outcomes are considered as of the last day of the reporting period. Any events (e.g., death)

that happened after that day are ignored in the respective cohort analysis but will be counted in the next report.

Primary follow-up outcome

• The primary outcome shows if a patient has been retained alive in care or if he has dropped out

and why.

• The primary outcome categories must add up to the total patients registered in the cohort.
• Table 27 lists the primary follow-up outcomes used for the different reports.
• For ART only, deaths are further classified according to the time after ART initiation. The categories

used are: death within 1st, 2nd, 3rd month after ART initiation or after 3rd month of ART initiation.

Secondary outcome

• Secondary outcomes are the latest treatment details among the patients retained alive in care.
• Secondary outcomes are counted directly from the cards of the patients retained alive in care,

usually by looking at the last visit before the end of the month or quarter evaluated. This visit might be several months before the end of the quarter, for example if the patient is on long ARV dispensing intervals (as long as the patient is still classified as ‘retained alive in care’ at the end of the quarter evaluated).

• Each set of secondary outcome categories must add up to the total number of patients retained

alive in care.

• Table 27 shows the secondary outcomes used for the different reports.

Definition of cohorts for different program reports

• 3 slightly different methods are used to define cohorts for outcome analyses:
• Cumulative cohort (ART): Follow-up status of all patients ever registered at the respective clinic.

The number of patients with adverse follow-up outcomes (death, default, etc.) inevitably increases over time. The number of patients retained in care is calculated by subtracting all patients with adverse follow-up outcomes from the total patient ever registered.

• Registration group cohort ‘Survival analysis’ in ART: Follow-up status of patients registered during

the quarters that ended 12, 24, 36, 48 and 60 months ago (ART). ANC cohort outcomes: final status as of the last ANC visit for the women who started ANC 6 months ago. This method standardises follow-up times and makes outcome data comparable between sites and over time.

• Birth cohort (HIV exposed child follow-up): Follow-up status of children who (would) have turned

2, 12 and 24 months old. Patient cards are filed in batches by month and year of birth (birth cohorts) and only the cards of children born 2, 12 and 24 months ago are pulled out for reporting.

Outcomes are counted separately for the 2-, 12- and 24-month birth cohort. Reporting is done monthly, and a different birth cohort is covered in each reporting month. This method standardises ages and is used for children enrolled in HIV exposed child follow-up.