Development and Internal Validation of a Clinical Risk Prediction Model for Opportunistic Infections Among People Living with HIV in Nigeria: A Retrospective Cohort Study

References

[1] World Health Organization, Consolidated Guidelines on HIV Prevention, Testing, Treatment, Service Delivery and Monitoring: Recommendations for a Public Health Approach, Geneva, Switzerland: WHO, 2021.

[2] Federal Ministry of Health, National Guidelines for HIV Prevention, Treatment and Care, 4th ed., Abuja, Nigeria: Federal Ministry of Health, 2022.

[3] National Agency for the Control of AIDS, National HIV and AIDS Strategic Plan 2023 - 2027, Abuja, Nigeria: NACA, 2023.

[4] E. W. Steyerberg and Y. Vergouwe, “Towards better clinical prediction models: seven steps for development and an ABCD for validation,” Eur. Heart J., vol. 35, no. 29, pp. 1925–1931, 2014.

[5] G. S. Collins et al., “TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods,” BMJ, vol. 385, p. e078378, 2024.

[6] G. S. Collins, J. B. Reitsma, D. G. Altman, and K. G. M. Moons, “Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement,” Ann. Intern. Med., vol. 162, no. 1, pp. 55–63, 2015.

[7] R. D. Riley et al., “Calculating the sample size required for developing a clinical prediction model,” BMJ, vol. 368, p. m441, 2020.

[8] E. W. Steyerberg, Clinical Prediction Models: A Practical Approach to Development, Validation, and Updating, 2nd ed. Cham, Switzerland: Springer International Publishing, 2019.

[9] E. von Elm et al., “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies,” Lancet, vol. 370, no. 9596, pp. 1453– 1457, 2007.

[10] J. P. Vandenbroucke et al., “Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration,” PLoS Med., vol. 4, no. 10, p. e297, 2007.

[11] R. F. Wolff et al., “PROBAST: a tool to assess the risk of bias and applicability of prediction model studies,” Ann. Intern. Med., vol. 170, no. 1, pp. 51–58, 2019.

[12] A. Maheshwari, D. S. Meena, D. Kumar, G. K. Bohra, and R. Gadepalli, “Predictors of opportunistic infections among people living with HIV: a prospective cohort study from a tertiary care setting in India,” Sci. Rep., vol. 16, p. 5901, 2023.

[13] D. Firth, “Bias reduction of maximum likelihood estimates,” Biometrika, vol. 80, no. 1, pp. 27–38, 1993.

[14] G. Heinze and M. Schemper, “A solution to the problem of separation in logistic regression,” Stat. Med., vol. 21, no. 16, pp. 2409–2419, 2002.

[15] E. W. Steyerberg, F. E. Harrell Jr., G. J. Borsboom, M. J. Eijkemans, Y. Vergouwe, and J. D. Habbema, “Internal validation of predictive models: efficiency of some procedures for logistic regression analysis,” J. Clin. Epidemiol., vol. 54, no. 8, pp. 774–781, 2001.

[16] D. Damtie, G. Yismaw, D. Woldeyohannes, and B. Anagaw, “Common opportunistic infections and their CD4 cell correlates among HIV-infected patients attending at antiretroviral therapy clinic of Gondar University Hospital, Northwest Ethiopia,” BMC Res. Notes, vol. 6, p. 534, 2013.

[17] D. Weissberg, F. Mubiru, A. Kambugu, J. Fehr, A. Kiragga, A. von Braun, A. Baumann, M. Kaelin, C. Sekaggya-Wiltshire, M. Kamya, and B. Castelnuovo, “Ten years of antiretroviral therapy: incidences, patterns and risk factors of opportunistic infections in an urban Ugandan cohort,” PLoS One, vol. 13, no. 11, p. e0206796, 2018.

[18] I. O. Abah, V. B. Ojeh, J. Musa, P. Ugoagwu, P. A. Agaba, O. Agbaji, and P. Okonkwo, “Clinical utility of pharmacy-based adherence measurement in predicting virologic outcomes in an adult HIV-infected cohort in Jos, North Central Nigeria,” J. Int. Assoc. Provid. AIDS Care, vol. 15, 2016.

@article{1722953,
      author = {Damion Oche, Otache Adah Emmanuel, Judith Sally Chuhwak, Omale John Oche},
      title = {Development and Internal Validation of a Clinical Risk Prediction Model for Opportunistic Infections Among People Living with HIV in Nigeria: A Retrospective Cohort Study},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {10},
      number = {3},
      pages = {1616-1623},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1722953.pdf},
      abstract = {Background: Opportunistic infections (OIs) remain a primary cause of morbidity and mortality among people living with HIV (PLHIV). Combining routinely captured clinical variables into an individualized risk-prediction tool can optimize clinical risk stratification within resource-constrained care settings. 
Methods: We conducted a retrospective cohort study of 139 adults living with HIV enrolled between January 2022 and December 2023 at treatment centers in North Central Nigeria. Four candidate predictors (CD4 count category, antiretroviral therapy [ART] status, ART adherence, and WHO clinical stage) were prespecified. Multivariable binary logistic regression using Firth’s penalized maximum likelihood was fitted to mitigate quasi-complete separation and parameter distortion. Discrimination was assessed via the Area Under the Receiver Operating Characteristic Curve (AUROC), and calibration was evaluated using the Hosmer-Lemeshow test. Internal validation was performed using 1,000 bootstrap resamples. A simplified point-based score was subsequently constructed. 
Results: Among 139 participants (mean age 36.4±9.8 years; 58.3% female), 38 (27.3%) developed an OI within 12 months. CD4 count category was the primary independent predictor of OI development (Firth Adjusted Odds Ratio ["AOR" ]=0.124, 95% CI: 0.038" - " 0.342, p<0.001). The apparent model AUROC was 0.981 (95% CI: 0.967"- " 0.995), driven by clear separation across extreme CD4 tiers; bootstrap validation demonstrated robust optimism-corrected discrimination ("AUROC"=0.942). Goodness-of-fit was acceptable (χ^2=4.21,p=0.837). A 0"--" 7 point risk score classified patients into low (0 "- " 2), moderate (3"-" 4), and high-risk (5" - " 7) tiers, exhibiting observed OI incidence rates of 0.0%, 30.0%, and 72.7%, respectively. 
Conclusions: The four-variable risk score provides accurate, reproducible risk stratification for OIs using routine clinical data. Following external validation in broader cohorts, this tool can serve as a point-of-care decision-support aid in resource-limited HIV care programs.},
      keywords = {HIV; Opportunistic Infections; Clinical Risk Model; CD4 Count; Firth Logistic Regression; Nigeria},
      month = {September},
  }