Methodological quality of the review: Medium confidence
Author: Lagani V, Koumakis L, Chiarugi F, Lakasing E, Tsamardinos I.
Region: United States of America (USA), Europe, including United Kingdom (UK) and Sweden.
Sub-sector: Diabetes complications, risk assessment, statistical models, risk factors and prognosis.
Equity focus: None specified
Review type: Systematic review
Quantitative synthesis method: Narrative analysis
Qualitative synthesis methods: Not applicable
A risk assessment model consists of any type of algorithm or mathematical formula (for example, a set of rules, a decision tree, a weighted sum, etc.) for assessing the overall statistical probability of certain adverse outcomes to occur in the future. When risk assessment models are built upon data collected from large-scale, longitudinal clinical studies, they are able to perform predictions in the long term, that is, on a time horizon spanning up to a decade and beyond. These models are the backbone of risk assessment tools used in clinical practice.
Given the health and social burden caused by diabetes-related complications, it is not surprising that several scientific works have proposed risk assessment models able to evaluate the probability for diabetes patients of developing one or more complications on the long-term period.
‘To compare and summarize the most relevant risk assessment models for diabetes-related complications published in the literature.’
The authors identified six major studies upon which current available risk assessment models were built on, namely:
All these studies had duration over five years and most of them included some common demographic and clinical data strongly related to diabetic complications. A total of 13 articles which introduced a risk assessment model for diabetes complications based on large scale studies mentioned above were included in the review. The most common predictions for long-term diabetes complications were related to cardiovascular diseases and diabetic retinopathy.
Authors noted that researchers and medical practitioners should take into account that some limitations undermine the applicability of risk assessment models; for example, it was hard to judge whether results obtained on a specific cohort can be effectively translated to other populations.
Authors included prospective and retrospective studies with at least 1,000 subjects and five years of follow-up studies as well as Type I and Type II diabetes studies. Publications were considered if they reported the name of the study in the title or in the abstract.
The search of the six major studies included in this review was not explicitly stated within the review. In order to identify publications for risk assessment models based on these studies, related publications were searched for at the official websites and on the well-known publication search engine Google Scholar, PUBMED and SCIRUS. The search covered combinations of concepts for diabetes, risk assessment, risk model and risk factors. In the case of Cleveland and Sweden, the search included Cleveland/Sweden alongside the study name. It should be noted, that searches for publications were specifically performed related only to the studies included. The search was limited to articles written in English.
Authors conducted a narrative review summary of most relevant risk assessment models for diabetes-related complications based on the six studies mentioned above.
Authors noted that due to certain limitations of the included studies, the applicability of the results may have been affected. As included studies in this review lasted several years, treatment for diabetes and comorbidities largely changed over the time, along with nutritional habits. Given the evolution in clinical sensors and in electronic health care records some may claim that data from these studies are obsolete. Another factor which may have affected the applicability of this reviews results is the close relationship of diabetes and its complications with the geographical location of the patient. Thus, diabetes-complications risk assessment models derived on specific samples were not ensured to be applicable on different populations or locations.
The searches focussed on the countries included in the six large-scale studies selected as a basis for identifying publications for risk assessment models. Out of the six studies, five were conducted in high-income countries; the remainder was performed in 16 countries in Europe, although country names were not provided. Nevertheless, finding from high-income countries, as reported by the authors, may not be applicable on different populations or geographical locations.
This review was based on a search of relevant databases only for studies related to the six large medical studies only. Language bias was not avoided as only studies written in English were included in the review. It was not possible to determine if study selection and data extraction was appropriately conducted, as authors did not report if this was conducted by two authors independently. Reviewers did note some of the limitations of the included studies in the review and did not draw strong policy conclusions. As such, medium confidence was attributed to the conclusions about the effects of this review.