Study design
In a cross-sectional study, RPs who were coming routinly for follow ups to the rheumatology clinic of Shahid Beheshti Hospital (Qom University of Medical Sciences, Qom, Iran)—the case group. and the healthy healthy individuals residing in Qom, Iran (the control group) were enrolled. The prevalence of COVID-19 was assessed in these two major groups. The subgroups of RDs and IS drug history was also assessed and compared to each other. The study protocol was implemented based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [9] and was approved by the Qom University of Medical Sciences and Health Services’ Research and Ethics Committee.
Setting and Data Source
The data of the patients of rheumatology clinic between December 1, 2019 and February 29, 2020, was obtained from the Shahid Beheshti Hospital records. For the control group, the data was collected from the records of the Qom Health Network, in the same manner as the case group. For the control group, residents of Qom were telephoned, and a survey of the signs, symptoms, exposure, and any history of COVID-19 was conducted. Besides, records of Qom Health Network for each person regarding the most recent hospital admission, clinical data, imaging and laboratory results, and history of COVID-19 and treatment were collected. The patient selection from the records was blinded utilizing the record number distribution in the period of the study.
Participants
To increase the power of analysis, the recruitment process resulted in 249 patients from the rheumatology clinic as the case group, and 207 residents of Qom city, categorized as the control group. The eligibility criteria comprised all patients of the rheumatology clinic, having an updated hospital record from December 1, 2019, to February 29, 2020, and residents of Qom city, who have been called by and had a record in Qom Health Network in the same date range. The recruitment of the control group was a blinded process in which participants were selected randomly from the records only using the record number—the other data were not visible to the selectors. For the case group, all eligible patients were included.
Variables
The demographic variables as well as the participants’ history of RD, hypertension, diabetes, hyperlipidemia, thyroid disorders, pulmonary disease, and cardiovascular disorders were included. Also, the symptoms of patients in the study time comprising cough, hemoptysis, shortness of breath, fever, chills, headache, myalgia, malaise, diarrhea, nausea, vomiting, constipation, abdominal pain, anosmia, sore throat, and sneezing were evaluated. Moreover, the participants’ history of admission, imaging, COVID-19, and any treatments for that were included. The subgroups for the case group were categorized based on the type of their RDs and the pharmacological/complementary treatments which they received. The diagnostic criteria for COVID-19 stood on the national guidelines of the Ministry of Health of the Islamic Republic of Iran, mainly by the signs and symptoms and chest computed tomography (CT) scan; in the non-conclusive results of the latter, the reverse transcription polymerase chain reaction (RT-PCR) test was considered.
Bias
The selection bias has been minimized by blinded recruitment for the control group. To reduce potential sources of bias the data were extracted after the selection of participants. The skilled staff of Qom Health Network performed the data gathering through telephone calls while they were completely unaware of this study.
Study Size
Because of the pandemic situation but uncommonness of COVID-19 (with an estimated prevalence of 0.025% by the Ministry of Health), the sample size calculated by the conventional protocols was unachievably large. Thus, to increase the test power, all 249 patients of the rheumatology clinic were included in the case group and 207 residents of Qom city in the control group.
Statistical Methods
The software used for statistical analysis was IBM SPSS Statistics for Windows, version 26.0 (IBM Corporation, Armonk, NY, USA). Descriptive analyses for all variables in all participants were performed and presented. To compare categorical data, Fisher’s exact test was executed, and a t-test was used to compare the means. All confidence intervals were set to 95%. Statistical significance was considered as p < 0.05.