Study Design and Setting
This is a cohort study of subjects with type 1 diabetes from Southern Brazil, with data collected retrospectively from medical records. Participants with a previous diagnosis of type 1 diabetes, with regular follow-up at the endocrinology outpatient clinic of a tertiary care public hospital in Southern Brazil were selected by requesting a query in electronic medical records with keywords, identifying all patients with type 1 diabetes mellitus treated in the institution. Electronic medical records were used to select participants who met the inclusion criteria for the study. Scheduled appointments between January 1st and November 6th, 2020, were used to assess the impact of the pandemic on quality of care indicators in this population, in comparison with the data collected from the same group of patients in 2019. In Brazil, the first case of COVID-19 was diagnosed on February 26th, 2020 , and, during the pandemic, three main peaks occurred in the months of August 2020, December 2020 and March 2021 in the South region . In March 22nd, 2020, the city of Porto Alegre, where most studied participants reside, presented its first requirement that guides social distancing and regulates establishments. Despite flexibility of social restrictions at some points, the social distancing for high risk groups for severe COVID-19, such as diabetic patients, remained very prevalent throughout 2020 .
All patients aged over 18 years diagnosed with type 1 diabetes mellitus, who received outpatient care between January 2019 and November 2020 in our institution, were potentially eligible. For inclusion in the study, these patients should have at least one appointment scheduled between January and December 2020, regardless of whether this medical appointment was attended or not. Exclusion criteria were having a record of other type of diabetes – type 2 diabetes, MODY, LADA or uncertain type of diabetes –, pregnancy, death, outpatient discharge in 2019, or not having an outpatient appointment scheduled for the year of 2020.
To assess the impact of the pandemic on the quality of care indicators for type 1 diabetes, an evaluation of clinical and laboratory parameters of the same cohort of patients was conducted, from the research conducted in 2019, comparing the care provided during the pandemic COVID-19 in the year 2020. The information was collected for one year of follow-up before the last medical appointment in the period, whereas the data were collected using information from the electronic medical records.
In order to identify possible flaws and to reduce bias, the researchers performed simulations and then collected data. The registering procedures were performed using an online form. The sociodemographic characteristics included age, sex, race/ethnicity and scholarity. The comorbidities were evaluated based on records of cardiovascular events, dyslipidemia, arterial hypertension, nephropathy, neuropathy, foot injuries, amputations and psychiatric conditions.
Furthermore, it was obtained information regarding health appointments, including information on the number of outpatient visits (nutrition, nursing and physician), number of teleconsultations performed during the COVID-19 pandemic, body mass index (BMI), and blood pressure. The assessment of neuropathy was performed through the 10 g Semmes Weinstein monofilament evaluation, or the Ipswich Touch Test, or the vibration sensitivity evaluation records. The assessment of retinopathy was performed according to the last fundus examination or retinography recorded. The information registered was, preferably, extracted from the last appointment (medical, nutrition and nursing) of the patient at the institution.
Finally, laboratory test results were collected, including the measurements of glycated hemoglobin (HbA1c) (measured with NGSP, certified high performance liquid exchange chromatography with Merck-Hitachi L-9100 Analyzer, Merck, Darmstadt, Germany), creatinine (measured with the traceable Jaffe method), albuminuria (measured in a sample using immunoturbidimetry, Alinity - Abbott Diagnostics, Illinois, USA), lipid profile (total cholesterol, LDL-cholesterol and triglyceride, measured with enzymatic colorimetric methods, Alinity - Abbott Diagnostics, Illinois, USA, and HDL-cholesterol measured with the accelerator selective detergent methodology, Alinity - Abbott Diagnostics, Illinois, USA) and thyroid-stimulating hormone (measured with chemiluminescent microparticle immunoassay, Alinity - Abbott Diagnostics, Illinois, USA). When there were more tests than recommended by the guidelines [4, 22], the last ones performed were considered.
Quality Of Care Indicators And Outcome Measures
Quality of Care Indicators and Outcome Measures
The quality of care indicators chosen followed the guidelines of the American Diabetes Association and the Brazilian Diabetes Society [4, 22] and included:
HbA1c assessment: at least two annual measurements;
Retinopathy assessment: annually after five years of diagnosis;
Assessment of distal symmetric diabetic neuropathy: annual, through the 10 g Semmes-Weinstein monofilament or Ipswitch Touch Test or assessment of vibration sensitivity;
Evaluation of albuminuria and / or serum creatinine: having at least one measurement in the last year;
Lipid profile (total cholesterol, HDL cholesterol, LDL cholesterol, and triglycerides): have a measurement in the last three years;
Thyroid-stimulating hormone (TSH): a measure in the last two years;
Furthermore, it was checked if a glycemic target was registered in the patient’s online medical record. In daily clinical practice, patients who had a history of ischemic heart disease, records of frequent episodes of hypoglycemia, severe visual impairment, who underwent hemodialysis or peritoneal dialysis, and performed only two or less capillary blood glucose tests per day are considered for a flexible target (HbA1c ≤ 8.0%). For all other patients, strict glycemic control was considered adequate (target of HbA1c ≤ 7.0%).
The primary outcome was the presence of quality of care indicators (positive or negative) regarding the periods before and during the COVID-19 pandemic in the same cohort of patients. Secondary outcomes include the comparison between the two periods in relation to the presence of all minimum indicators of quality of care in diabetes and of each quality indicator individually, HbA1c levels, and lipid profile.
Sample size was calculated in the Power and Sample Size Health online version  to compare the percentage of individuals who had achieved the goal for quality indicators (HbA1c assessment, retinopathy assessment, distal symmetric diabetic neuropathy assessment, evaluation of albuminuria and / or serum creatinine, lipid profile and TSH) in 2019 but not in 2020, with the percentage of individuals who did not achieve the goal in 2019 but did in 2020. Considering 80% power and 5% significance level, the sample size necessary was at least 81 individuals .
The data were transcribed from the online platform Google Forms (Alphabet Co., Mountain View, CA) to the Statistical Package for Social Science (SPSS®) version 22.0 for analysis. Descriptive data are presented as frequency (%) or mean ± standard deviation (SD) if the assumption of normal distribution did not seem violated; otherwise, data were reported as median ± interquartile range (IQR). Normality was defined by the Shapiro-Wilk test. The last results of laboratory tests and the last clinical evaluation available in the medical records were considered for analyses. For the assessment of the primary outcome, a dichotomous outcome (yes/no) was considered for the presence of quality indicators previously defined. The difference between the two periods (prior and during the COVID-19 pandemic) was assessed using the McNemar test for categorical variables and the Wilcoxon test for continuous variables. An alpha value < 0.05 was used to determine statistical significance.
An exploratory post hoc analysis was performed comparing patients who were included in the study in relation to those excluded for not having a medical appointment scheduled during the pandemic period. The two groups were compared for age, gender, time of diagnosis of diabetes, HbA1c, and number of patients with glycemic control on the target.
The study was approved by the institution’s Research Ethics Committee (No. 20380919800005327). The researchers signed a term of commitment for the use of the institution’s data, and the research considered the announced prerogatives in Resolution 466/2012 of the National Health Council. This document follows the STROBE Statement Checklist of items that should be included in reports of observational studies.