INTRODUCTION Within 2 months of the first detection of SARS-CoV-2 in Honduras, its government promoted the nationwide implementation of multi-drug COVID-19 inpatient and outpatient treatment protocols. This was associated with a case fatality rate decreased from 9.33% to 2.97%. No decrease was seen in Mexico, a similar Latin American country that did not introduce multi-drug treatment protocols at that time.
OBJECTIVE The primary objective of the study was to use statistical process control to assess the likelihood that the decrease in the case of fatality rate in Honduras was due to chance, using Mexico as a control country.
METHODS Fourteen-day running average COVID-19 case fatality rates in Honduras and Mexico were used to create Shewhart control charts during the first 6 months of the epidemic. The date of implementation in Honduras of the inpatient and outpatient multi-drug COVID-19 protocols were plotted on control charts, with a Mexican COVID-19 case fatality control chart for comparison.
RESULTS The case fatality rate for COVID-19 in Honduras dropped below the lower control limit 9 days after implementation of an inpatient and outpatient multi-drug therapeutic protocol, from an average 9.33% case fatality rate to 5.01%. The Honduran COVID-19 case fatality rate again dropped below the lower control limit to 2.97%, 17 days after launching a substantial government program to make the protocol medications accessible to underserved areas. Shewhart control chart plots of case fatality rates in Honduras suggest a plausible temporal association between the implementation dates of both the initial protocol implementation on May 3, 2020, and the outreach effort on June 10, 2020, and statistically significant control chart anomalies. No control chart anomalies were seen during that time in Mexico.
CONCLUSION Decreases in COVID-19 case fatality rates in Honduras were associated with both the initial publication of a multi-drug COVID-19 therapeutic protocol and a subsequent outreach program.
Honduras is a low-middle-income country with 66 percent of its population in poverty in 2016, and with 20 percent of Hondurans living in extreme poverty—that is, on less than US$1.90 per day. Honduras experienced its first case of COVID-19 documented by SARS CoV-2 RT-PCR testing on March 10, 2020.
As the COVID-19 pandemic unfolded in New York in the spring of 2020, Honduras faced the potential collapse of its health care system from COVID-19, with but 19 critical care physicians and 125 intensive care units (ICU) beds to serve a population of 9.9 million. By contrast, New York City with a population of 8.9 million, had 1060 critical care physicians and 1644 ICU beds. Still recovering from a recent Dengue epidemic, the Honduran government responded proactively to the COVID-19 pandemic based on the recommendations of its critical care and infectious disease consultants. Honduran physicians theorized that a multi-mechanism approach (MMA) to treating COVID-19 could block the inflammation, immune system disruption, and hypercoagulation that initial reports suggested as the cause for mortality from SARS CoV-2 infections. They created the MMA protocol based on the hypothesis that COVID-19 mortality resulted less from direct viral infection and more from the self-perpetuating cytokine storm and thrombosis triggered by the virus that, once triggered, increased independently of viral replication (Figure 1). Given Honduras’ limited resources, the MMA protocol used repurposed, inexpensive medications already proven safe and effective for other indications, for which in vitro and clinical evidence suggested clinical efficacy against COVID-19. The MMA protocol also optimized oxygenation with high-flow O2 therapy and self-pronation rather than mechanical ventilation whenever possible, conserving limited intensive care resources. In mid-April 2020, Honduran physicians began treating COVID-19 patients with the MMA protocol immediately upon admission to the hospital, as well as in outpatient clinics on diagnosis with COVID-19. Decreased morbidity and mortality were observed in patients receiving the MMA inpatient protocol, documented in a peer-reviewed retrospective cohort study that showed a decrease in ICU length of stay by 5.4 days with a trend towards decreased mortality.5
The MMA protocol(Table 1A) medications have yet to be labeled by the FDA as effective for COVID-19 therapy, although they have been labeled as safe for non-COVID-19 indications. After the initial promising results, the protocol was promoted by the Honduran Health Department on May 3, 2020, in a nationally televised educational program for health care professionals, detailing recommendations on treating COVID-19 with MMA for inpatients and outpatients. The Honduran Health Department hosted additional Zoom™ video meetings to educate clinicians throughout May 2020. The inpatient MMA protocol was marketed as “CATRACHO,” an acronym of the protocol’s components that references General Florencio Xatruch, the Honduran leader who defeated aggressors from the U.S. in 1856. Hondurans refer to themselves as “catches,” a Nicaraguan pronunciation of the Catalan surname Xatruch.
Due to the urgency of the pandemic and the paucity of local healthcare resources, it was impractical to prospectively organize randomized, controlled, double-blinded trials to evaluate the MMA protocol’s efficacy against COVID-19 before its implementation in Honduras. An alternative methodology, statistical process control, is a well-validated approach initially developed to improve manufacturing outcomes at Bell Laboratories about a century ago. Statistical process control has gained increasing acceptance in health care applications. It is less resource-intensive, achieving statistical significance through frequent measurements over time rather than by large patient sample sizes, and represents results in easy to comprehend control charts.
The initial implementation of the MMA protocols was associated with a 6.36% decrease in the COVID-19 case fatality rate in Honduras, from 9.33% before May 3, 2020, to 2.97% after. This suggests the number needed to treat (NNT) is 16 patients to prevent one COVID-19 fatality for the combined inpatient (CATRACHO) and outpatient (MAIZ/MAIZ-AA). Recalculating the control chart limits using the 5.01% average case fatality rate on June 10, 2020, as baseline demonstrated a statistically significant drop below the lower control limit on June 27, 2020. This case-fatality rate decrease from 5.01% to 2.97% suggests the avoidance of 1 COVID-19 fatality for every 49 outpatients treated by the additional outreach initiative by Honduran medical brigades implementing the MAIZ/MAIZ-AA outpatient therapeutic protocol.
Our findings demonstrate the utility of statistical process control methodology for quickly and efficiently evaluating and monitoring the efficacy of therapies for COVID-19, which could be generalizable to other emergent conditions.