What Factors Are Linked With Diabetes, Pediatric Patients, and Dialysis

Diabetes and Pediatric and Dialysis

The objective of this study is to analyze the incidence of diabetes among pediatric patients, with dialysis, by way of analysis of outcomes one year following diagnosis.

Initiatives concerned with investigating pediatric diabetes have previously been associated with varying outcomes. What factors are linked with diabetes, pediatric patients, and dialysis with regards to Glycated hemoglobin (HbA1c) determinants? This forms the groundwork of the paper.

Protection of Human Participants iii. Data collection

Data management and Study Design

Control of blood glucose, as shown by levels of hemoglobin A1c (HbA1c), serves as one of the clearest forecasters of lasting complications in adults as well as children suffering from type 1 diabetes mellitus (T1D). This includes diabetic nephropathy, proliferative retinopathy, age-adjusted mortality, and cardiovascular disease rates.

Hemoglobin A1c is the form of hemoglobin in the blood (red blood cells) that develops when elevated sugars are present. The sugar binds to the hemoglobin, which normally functions to deliver oxygen throughout the body, and leads to impaired oxygen-binding and release. Because HbA1c is directly correlated with blood glucose levels, physicians can use the HbA1c measurement to determine whether a given patients 'glucose control' is adequate or inadequate. Poor control of glucose during the initial years of type 1 diabetes (T1D) in children is expected to cause development of retinopathy and micro-albuminuria in adulthood; this is because of the impaired oxygen delivery resultant. Though reports exist on factors concerning HbA1c in individuals having established T1D, no published research can be found on the relationship of HbA1c in terms of socioeconomic and diabetes-specific factors, during the disease's early stages, in a sample representing children treated in United States (U.S.) pediatric diabetes centers (Redondoa, Connor & Ruedyb, 2014).

Significance of the study in Nursing Practice

The research of the Pediatric Diabetes Consortium (PDC) Type 1 diabetes New Onset (NeOn) aims to improve care provided to children having diabetes by sharing the best practices. This huge, ethnically and geographically diverse group of youth suffering T1D offers a unique opportunity to study possible factors that may be linked to metabolic control one year following Type 1 diabetes (T1D) diagnosis (Redondoa, et al., 2014). This work is significant to the field of nursing because it facilitates monitoring of the condition, as well as providing insightful information on new directions that may be useful in patient care.

Protection of Human Participants

Protection of Personal Information

In obtaining data on the participants for the purpose of the study, all sensitive details regarding the participants was kept private. The protocol that was followed was consistent with standards established by the Institutional Review Board (IRB). General information centered on gender and age; no specific details were disclosed.

Source of consent and voluntary Participation

The participants' consent was obtained, as is the IRB's requirement. All participants took part voluntarily in the study, so long as the age requirement of less than 19 years was met. The large number of participating individuals and their responsiveness demonstrated that a great willingness was present on the people's part to voluntarily contribute to the process.

Data Collection

Variables

Key variables were defined and identified clearly. Both dependent and independent variables were included. In case of the independent variable, age, all participants had to strictly be aged below 19 years and also have been Pediatric Diabetes Consortium (PDC) participants for at least three months. The dependent variable, body weight, was governed highly by age and gender (independent variable). Weight and height independently contributed to the determination of Body Mass Index (BMI) as a dependent variable. A one-year-duration was established for the analysis.

Method of Data Collection

Data on clinical characteristics, demographics, and socioeconomic status were obtained from interviews with patients/parents, and medical records.

Rationale used in Data Collection

Diabetes ketoacidosis (DKA), as a standard rationale, was defined as per the criteria of the Diabetes Control and Complications Trial (DCCT). This measurement is bicarbonate (HCO3-1) levels at

Study Period

The PDC type 1 diabetes NeOn research had 1052 participants with T1D patients joining up between July 2009 to April 2011.

Sequence of Data Collection

Follow-up appointments were conducted according to usual protocols of care. All visits in the patients' first year after diagnosis were, for the purpose of the study, entered in the form of standard electronic case reports. For this purpose, BMI at the time of diagnosis was calculated from the nearest weight and height within 14 days before or after diagnosis. The Z scores and percentiles of BMI, adjusted for gender and age, were computed using the Centers for Disease Control and Prevention's (CDCs) data from the population growth chart of the year 2000. If weight and height records were absent within + or -- 14 days of diagnosis of diabetes, or in case the patient was aged below 2 years during diagnosis, body mass index percentile was not computed (Redondoa, et al., 2014).

Data Management and Study Design

Method used

The PDC type 1 diabetes NeOn research had 1052 participants with T1D joining up from July 2009 to April 2011. Protocol approval was obtained from the IRB at all 7 participating facilities. Informed permission was gained from participants aged 18 years and over, and from the parents of patients below 18 years. Assent of participants aged below 18 years was obtained, as per the requirements of the local IRB rules. To be eligible for participation in the research, the requirements were: [1] age limit of patients was below 19 years; and [2] patient being treated at any of the 7 participating centers of the PDC within three months of T1D diagnosis. The evaluations recorded herein contained data obtained from 857 study participants. Some 163 participants were omitted from the study because of non-availability of HbA1c measurement at one year (319 -- 455 days from diagnosis); 32 participants were omitted because of participating in an intervention research (Redondoa, et al., 2014).

Assessment of the relationship of HbA1c at one year from diagnosis of T1D, was determined, along with baseline (socioeconomic, clinical and demographic) data. The data included one-year patient records including total daily dosage of insulin, frequency of self-monitoring of blood glucose (SMBG), and the number of appointments with providers of diabetes management. The appointment record excluded visits in the first 3 weeks after diagnosis; this period includes initial diabetes education in facilities where patients aren't admitted routinely at diagnosis. Earlier models were formulated individually for every baseline factor. Later, a multivariate baseline-factor model was constructed, using selection stepwise, with p-values less than 0.10 necessarily incorporated into the model. Because of multiple comparisons, factors that only had p-values less than 0.01 were deemed statistically significant, though factors having p-values less than 0.10 were included to adjust for any potential confounding. Testing of interaction terms for every variable, with p-values less than 0.10 necessarily incorporated, in the ultimate multivariate model, was conducted. Another multivariate model served the purpose of adding the one-year factors to significant baseline elements (Redondoa, et al., 2014).

Linearity testing was carried out for continuous variables, by addition of a quadratic term into the model. In case of significant non-linearity being detected, the specific variable was split into categories, followed by analysis as separate variables. An indicator for missing values, in case of continuous variables, was incorporated into the model; missing covariates, in case of discrete variables, would be considered as a distinct category. Every reported p-value is two-sided (Redondoa, et al., 2014).

Software Application

Every analysis was carried out using Statistical Analysis System Software (SAS) v9.3, for ensuring a high accuracy level (Redondoa, et al., 2014).

Findings

During diagnosis, the mean +/- SD (standard deviation) age of the participating 857 individuals was 9.1+/-4.1 years. Some 51% of the participants were female; 68% had private health insurance coverage; 66% belonged to the category of Non-Hispanic Whites. Mean +/-SD for Hemoglobin A1c (HbA1C) recorded was 102 +/-25 millimole/mole (11.4% +/-2.3). Some 33% of participants had diabetic ketoacidosis (DKA); 19% were obese or overweight (with body mass index ? 85 thpercentile for gender and age). Some 95% of the 510 patients tested for each of the three auto-antibodies were positive for a minimum of a year following T1D diagnosis; mean+/- SD Hemoglobin A1c was 62+/-16 millimole/mole (7.8% +/- 1.5); median insulin dosage was 0.6 units/kilogram/day [interquartile range (IQR) 0.5 -- 0.8]. Median self-reported tests (frequency of self-monitoring of blood glucose (SMBG)) was 5/day (IQR 4 -- 6); median number of diabetes provider visits in the starting year of diabetes was 4 (IQR 4 -- 5). Where downloading of blood glucose meters was possible (n=469), IQR and median SMBG tests a day approximated to self-reports. Some 34% of the patients used insulin pump treatment at one year; while 59% patients took 3 or more insulin injections daily (Redondoa, et al., 2014).

Many baseline characteristics (characteristics at the time of diagnosis) were linked significantly, in univariate analyses, to lower HbA1c levels at one year. These included: clinical site, private healthcare insurance, higher education-level of parents, white race, better household income, diabetic ketoacidosis (DKA) absence during diagnosis, and residing with both parents (p-value less than 0.01 for all). In the baseline factors' multivariate analysis, status of health insurance, race, family structure and clinical site remained significant.

Interpretation

Impact of parent education and family income were seen to be confounded with other socioeconomic status factors, in a way that possible independent impacts of the above factors cannot be ruled out or confirmed. In case of multivariate analysis, no visible association could be found with age, gender, BMI, Tanner stage (scale of physical development), HbA1c at the beginning, DKA at time of diagnosis, or the quantity of positive anti-islet auto antibodies.