University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Long-Term Healthcare Interruptions
Long-term healthcare interruptions among HIV-infected patients in Uganda
Global Health Sciences Literature Digest
Published February 25, 2013
Journal Article

Mills EJ, Funk A, Kanters S, Kawuma E, Cooper C, Mukasa B, et al. Long-term healthcare interruptions among HIV-infected patients in Uganda. J Acquir Immune Defic Syndr. 2013 Feb 12. [Epub ahead of print]

Objective

To better understand the patterns and predictors of loss to follow-up and unstructured treatment interruptions in HIV-infected patients receiving antiretroviral therapy (ART).

Setting

Mildmay Centre, a specialized clinic providing free HIV/AIDS care in urban Kampala, Uganda.

Study Design

Retrospective cohort study.

Population

Patients age ≥14 years, on ART, with at least a baseline CD4 count.

Main Outcome Measures

Health care interruptions ≥12 months, loss to follow-up.

Methods

Patients were eligible for treatment if they had a baseline CD4 count of ≥250 cells/µL (though this changed to ≥350 cells/µL in the last few months of recruitment) or presented with a relevant clinical indication. Data from patient visits and clinical tests were entered into a central database. Patients in care were expected to visit their physicians at least once every six months. Health care interruptions were defined as not accessing clinical HIV care at Mildmay for ≥12 months. Loss to follow-up was defined as not accessing care for ≥12 months and not returning to care at Mildmay by October 2011. The time to health care interruption was calculated as ending ≥12 months since the last contact. The difference between the date of treatment initiation and the date of last contact was calculated as time until loss to follow-up. A range of demographic, behavioral and clinical predictive variables were specified. Investigators used Fisher's exact test and logistic regression to detect any important differences between outcome groups with regard to these variables. They used Kaplan-Meier survival curves to show proportions of healthcare interruption and loss to follow-up over time, relative to the variables.

Results

Between January 2004 and April 2012, 6,970 patients entered ART care at Mildmay. Two-thirds (n=4,595, 65.9%) of patients were female. Median patient age was 36 years (interquartile range [IQR] 29 years to 43 years). Of these, 784 (11.2%) patients were absent from care for ≥12 months; 217 (3.1%) were lost to follow-up.

With respect to health care interruptions, patients initiating therapy with CD4 of =250 cells/µL were at a higher hazard of interrupting care (adjusted hazard ratio [aHR] 1.23, 95% confidence interval [CI] 1.04 to 1.47). Those with at least a secondary school education had a lower hazard of interrupting care (aHR 0.77, 95% CI 0.64 to 0.93). Compared to adults aged 20-49, there was no significant difference in the hazard for adolescents (ages 14-19) to interrupt care (aHR 1.12, 95% CI 0.89 to 1.40).

With respect to being lost to follow up, CD4 count was not a significant predictor (aHR 0.75, 95% CI 0.49 to 1.13). Patients with sexual partners had much lower odds of being lost to follow-up, (odds ratio [OR] 0.22, 95% CI 0.16 to 0.31) as did patients who were sexually active at baseline (OR 0.40, 95% CI 0.28 to 0.55). Adolescents were at a much higher hazard of loss to follow-up than adults aged 20-49 (aHR 1.94, 95% CI 1.38 to 2.73).

Conclusions

The authors conclude that patients who had healthcare interruptions tended to have higher baseline CD4 counts or lower levels of education. Patients who were lost to follow-up were more often adolescents and adults without sexual partners.

Risk of Bias

The risk of bias in this observational study is moderate. The study presents baseline data and includes a cohort. It does not include an external comparison group. The authors use statistical methods to control for confounders. Loss to follow-up was low.

In Context

The proportion of patients lost to follow up in this cohort is quite low compared to other published African studies. Many clinics, including Mildmay, make efforts to improve retention in care through engaging patients in group activities and social networks, and even through home visits. Patients at higher risk of interrupting care or being lost to follow-up could be further targeted through mobile phone text-messaging(1) or community support groups.(2) An additional interesting finding was that CD4 counts of 250 cells/µL was associated with greater hazards of temporarily discontinuing therapy. This suggests that patients who are less immunodeficient and less symptomatic may be going off ART and waiting until they become ill to reinitiate therapy.

Programmatic Implications

Clinics and hospitals should measure lost-to-follow-up rates as part of their strategy to retain patients in treatment. They may wish to implement pilot programs focused on patients at higher risk of interrupting care or being lost to follow-up.

References

  1. Horvath T, Azman H, Kennedy GE, Rutherford GW. Mobile phone text messaging for promoting adherence to antiretroviral therapy in patients with HIV infection. Cochrane Database Syst Rev. 2012 Mar 14;3:CD009756.
  2. Decroo T, Telfer B, Biot M, Maïkéré J, Dezembro S, Cumba LI, das Dores C, Chu K, Ford N. Distribution of antiretroviral treatment through self-forming groups of patients in Tete Province, Mozambique. J Acquir Immune Defic Syndr. 2011 Feb 1;56(2):e39-44.