Predictive Factors for not Undergoing RNA Testing in Patients Found to Have Hepatitis C Serology and Impact of an Automatic Alert

Dalia Morales-Arraez; Ana Alonso-Larruga; Felicitas Diaz-Flores; Jose A. García Dopico; Antonia de Vera; Enrique Quintero; Manuel Hernández-Guerra


J Viral Hepat. 2019;26(9):1117-1123. 

In This Article

Patients and Methods

Study Design and Patients

The study was divided in two phases. Phase I included a descriptive analysis to establish the prevalence of not undergoing RNA testing rates in our health area among patients with a HCV-positive antibody test and the characteristics of the patients. In phase II, we implemented an electronic alert and assessed its impact (Figure 1).

Figure 1.

Diagram of the study design

This initiative emerged after the detection of an alarming low rate of complete HCV diagnoses in a retrospective exploratory study, and received approval from the hospital steering committee as a part of a pilot study designed to improve clinical practice and care.

Phase I. All HCV antibody tests requested from October 2011 to September 2014 were identified retrospectively in the Central Laboratory Department database of the Hospital Universitario de Canarias. This laboratory processes all HCV tests requested by referring physicians of the area and serves a health area of 380 000 people with access to our public health care system (Sistema Nacional de Seguridad Social). Patients with a positive test were identified from the registry and followed until June 2018 with a focus on RNA testing evaluations. Then, patients were divided into two groups according to whether or not they underwent RNA testing to assess individual characteristics.

Phase II. In October 2014, an automatic electronic alert message was introduced and is visible on patients' HCV reports after testing positive. A computerized information system (OpenLab®, Nexus, IT, Spain) allowed us to introduce the following text to remind physicians to test RNA and to refer the patient to specialist care: 'The WHO and current guidelines recommend measuring viral load and referring this patient to specialist care'.

To assess the impact of this strategy, a prospective one-year cohort (post-alert cohort, October 2014-September 2015) was compared with a previously analysed cohort of patients with a positive HCV antibody test (pre-alert cohort, October 2011-September 2014). All patients were followed up until June 2018 before evaluating the percentage of RNA requests in each cohort. Delays in RNA tests and the percentages of patients attending specialist care (Gastroenterology or Internal Medicine specialists) were also evaluated after reviewing of electronic medical records.


The electronic medical records available at hospital (SAP©) and primary care centres (DRAGO AP©, Gobierno de Canarias) were reviewed to collect demographic, clinical and biochemical variables. We collected information on the date of the HCV antibody request, past or current use of drugs, alcohol use, performance status, social support (good or lacking, defined by ICD-9 diagnoses), baseline value of transaminases (alanine aminotransferase [ALT] and aspartate aminotransferase [AST]; abnormal value: >35 U/L), psychiatric disease, and comorbidities expressed by Charlson comorbidity index adjusted for age.[22] In addition, the specialty of the physician who requested the HCV antibody tests was also registered. The specialties were divided into hospital-based specialists and primary care physicians responsible for the outpatients at clinics, prisons and drug addiction units.

To evaluate the availability of RNA testing, the date and result of the test were also recorded. In both phases, patients were excluded from the analysis if they were under 2 years old, had a previous viral load determination, they were lost to follow-up (without subsequent visits by physicians or without blood tests after the HCV antibody determination) or died one year after HCV serology during the follow-up period since these patients lost the possibility to be fully evaluated.

Statistical Analysis

The chi-squared test was used to assess qualitative variables, and Student's t test or Mann-Whitney U test was used for continuous variables based on whether the distribution of variables was normal. Independent predictors for not undergoing RNA testing were analysed using logistic regression. Actuarial rates of RNA testing were estimated by the Kaplan-Meier method and compared in both cohorts using the log-rank test. The Poisson variable was used to test the impact of the alert. P values < 0.05 were considered statistically significant. We used SPSS 15.0 for the statistical analyses.

Ethical Aspects

The study was conducted in accordance with the ethical principles of the Helsinki Declaration of October 2013, and approval was obtained from the Clinical Research Ethics Committee of the Hospital Universitario de Canarias concerning ethical, legal and methodological aspects. The collected data were handled confidentially in an encrypted database that was only accessed by the researchers involved in accordance with current laws (Ley Orgánica de Protección de Datos 15 1999).