Impact of Chemotherapy Stock-Out on Standard Therapy Delivery Among Cancer Patients in Botswana

Background: Cancer drug shortages represent a complex global issue with potentially adverse survival outcomes. Up to 98% of oncologists and pharmacists surveyed in North America reported at least 1 drug shortage in the prior year and 66% reported at least one patient who had clinical care impacted by the shortage. In low- and middle-income countries (LMICs), these shortages are even more frequent. No studies to our knowledge have evaluated the impact of chemotherapy stock-out on receipt of standard cancer therapy in LMICs. Aim: Quantify the association between the duration of chemotherapy stock-out and the risk of having a suboptimal therapy delivery event, compared with standard delivery of therapy among cancer patients in Botswana. Methods: Prevalent cohort study of patients with cervical, breast, prostate, esophagus, Kaposi sarcoma, head and neck cancers, lung, uterine, ovarian and colorectal cancers who received any systemic therapy between 01/01/16-12/31/16 at Princess Marina Hospital, Botswana. Primary exposure was stock-out duration per cycle interval calculated by generating a code for the six different patterns for chemotherapy stock-out, using stock data at the Central Medical Stores. Primary outcome was suboptimal therapy delivery defined as a dose reduction, dose delay or switch in intended therapy. We measured statistical associations using two sample t-test and mixed effects univariate and multivariate logistic regression models. Results: 378 patients were identified who met diagnostic criteria and received systemic chemotherapy in 2016. Of these, 293 received commonly prescribed standard regimens who contributed 1452 cycle intervals and were included in our analysis. Majority of the patients (48%) had breast cancer. The mean duration of stock-out for receipt of standard therapy without events was 3.2 days (95% CI: 2.8-3.7) compared with 7.8 days for patients who had a suboptimal therapy delivery event (95% CI: 6.6-9) ( P < 0.0001). Male sex, age < 65 and HIV-positive status were also significantly associated with an increased risk of experiencing dose reduction, change in therapy or switch in therapy. Adjusting for these factors in a mixed effects logistic regression, each week of stock-out was independently associated with an 80% increased risk of having a suboptimal therapy delivery event (OR=1.8 (95% CI: 1.6-2.0, P < 0.0001)). Conclusion: Chemotherapy stock-out is independently associated with an 80% increased risk of a patient experiencing dose reduction, change in therapy or delay in therapy. The risk increases with longer duration of stock out. Given prior data showing that these events lead to worse survival outcomes, our further analysis is focusing on quantifying risk of stock-out on survival outcomes in this population. to determine whether interventions promoting standard therapy delivery are warranted to optimize survival outcomes.