Sex Disparities in Cardiovascular Risk Factor Assessment and Screening for Diabetes-Related Complications in Individuals With Diabetes: A Systematic Review

Background Insight in sex disparities in the detection of cardiovascular risk factors and diabetes-related complications may improve diabetes care. The aim of this systematic review is to study whether sex disparities exist in the assessment of cardiovascular risk factors and screening for diabetes-related complications. Methods PubMed was systematically searched up to April 2020, followed by manual reference screening and citations checks (snowballing) using Google Scholar. Observational studies were included if they reported on the assessment of cardiovascular risk factors (HbA1c, lipids, blood pressure, smoking status, or BMI) and/or screening for nephropathy, retinopathy, or performance of feet examinations, in men and women with diabetes separately. Studies adjusting their analyses for at least age, or when age was considered as a covariable but left out from the final analyses for various reasons (i.e. backward selection), were included for qualitative analyses. No meta-analyses were planned because substantial heterogeneity between studies was expected. A modified Newcastle-Ottawa Quality Assessment Scale for cohort studies was used to assess risk of bias. Results Overall, 81 studies were included. The majority of the included studies were from Europe or North America (84%).The number of individuals per study ranged from 200 to 3,135,019 and data were extracted from various data sources in a variety of settings. Screening rates varied considerably across studies. For example, screening rates for retinopathy ranged from 13% to 90%, with half the studies reporting screening rates less than 50%. Mixed findings were found regarding the presence, magnitude, and direction of sex disparities with regard to the assessment of cardiovascular risk factors and screening for diabetes-related complications, with some evidence suggesting that women, compared with men, may be more likely to receive retinopathy screening and less likely to receive foot exams. Conclusion Overall, no consistent pattern favoring men or women was found with regard to the assessment of cardiovascular risk factors and screening for diabetes-related complications, and screening rates can be improved for both sexes.

Data source: Data extracted from 2008-2011 Medicaid claims from three states with relatively high rates of Medicaid enrolment and complete claims, including those with diabetes and major depressive disorder. Those included were required to have >1 inpatient diagnosis or >2 outpatient or emergency department diagnoses of major depressive disorder or diabetes during a single year in the study period, and >1 claim for the condition in each year. Individuals with serious mental illnesses were excluded as well as dual enrollees in Medicare

Administrative claims data
Primary aim: To determine the extent to which the diabetes care needs are met for a population with both intellectual and developmental disabilities and diabetes who are solely insured by Medicaid in five states.

Administrative claims data
Primary aim: To describe regional variation in the utilization of the four measures across small regions in Switzerland and to explore potential influencing factors.
Data source: Data extracted from health insurance claims provided one of the largest health insurance companies in Switzerland. Those enrolled with Helsena with diabetes who were prescribed any diabetes medication between 1/1/ 2014 and 27/12/ 2014 were included. Date of the first prescription of any diabetes medication in 2014 (incident diabetes) or January 1, 2014 (prevalent diabetes) was considered as the index date for each participant. Those with incomplete insurance coverage in 2014 or not surviving until the end of 2014 were excluded, as well as those living outside Switzerland, asylum seekers, Helsana employees, with incomplete address information, living in nursing homes with lump-sum reimbursement. Youn

Electronic medical records
Primary study aim: To describe quality indicator performance in diabetes care in Swiss primary care and to analyze associations of practice, general practitioner and patient covariates with quality indicator performance.
Data source: Baseline data extracted from an electronic medical record database collected within a cluster randomized controlled trial. The baseline assessment covered 12 months retrospectively using electronic medical records database of the Institute of Primary Care of the University of Zurich. Those diagnosed with diabetes ≤4 months before the baseline date were eligible for inclusion. Hirst

Electronic medical records
Primary aim: To examine whether both an individual's previous HbA1c and the reporting deadline at the end of the administrative year are associated with overfrequent or delayed HbA1c testing in national data in the UK, and whether there are regional disparities across the UK and whether other pre-defined participant or general practitioner practice level variables may be associated with very frequent or delayed HbA1c testing intervals.
Data source: Data extracted from those with diabetes randomly selected from the Clinical Practice Research Datalink (CPRD) over a 10-year period (1/1/2005-31/12/2014). CPRD is a governmental database providing anonymized data from UK primary care. For those with existing diabetes, baseline HbA1c test was defined as first HbA1c test after 1/1/2005. Included participants had ≥2 HbA1c tests prior to the baseline test and post diagnosis. People with incident diabetes during follow-up, and ≥3 HbA1c test post diagnosis, were included in the analysis. For those, the baseline test was the second test. Participants had ≥2 HbA1c tests for inclusion. Those with gestational diabetes, malnutrition related diabetes, maturity-onset diabetes of the young, <3 HbA1c measures in total, steroid-induced diabetes or haemochromatosis-related diabetes, cancer or end-stage renal disease, were excluded. Bakke

Electronic medical records
Primary aim: To assess population, general practitioner, and practice characteristics associated with the performance of microvascular screening procedures and to propose strategies to improve type 2 diabetes care.
Data source: Data extracted from electronic health records from general practices located in five of Norway's nineteen counties with urban and rural areas participating in the ROSA 4 study, including adults with type 2 diabetes who had their main followup in general practice and a diabetes duration of ≥1 year. Those diagnosed with diabetes in 2014, new to the general practitioner, with main follow-up at by a specialist, in nursing homes, with unknown list-holding general practitioner, type 1 diabetes including LADA, and other i.e. MODY, pancreatitis, or undetermined were excluded. Dallo et al., 2019 10 United States 2015 6,622 (54%)

≥18
No eye exam during study period.

Administrative data
Primary aim: To estimate and compare the management of diabetes among Arab, Asian, non-Hispanic Black, and non-Hispanic Whites attending a large health system in metropolitan Detroit.

Self-reported
Primary aim: To examine trends and disparities in the quality of diabetes care among US adults with diabetes.
Data source: Data extracted from the National Health and Nutrition Examination Survey (NHANES) 1999-2016, including those with self-reported diabetes (not during pregnancy) based on questions about physician diagnosed diabetes and medication use, and levels of fasting glucose or HbA1c. Survey respondents were selected using a complex, stratified, multistage probability sampling design of the US noninstitutionalized civilian population. Survey data were gathered through in-home interviews, physical exams, and lab tests. An

Electronic medical records
Primary aim: To assess long-term adherence, in patients with diabetes, to the recommended regular eye exam guidelines, and to determine factors associated with non-adherence.
Data source: Patient data extracted from Kaiser Permanente Southern California (KPSC). KPSC is a non-profit, integrated health-care delivery organization in Southern California. KPSC provides integrated, comprehensive medical services within its own facilities, which include hospitals, outpatient facilities, and a centralized laboratory. All aspects of care and interaction with the health-care delivery system are captured in a continuously updated electronic Organization. Those with ≥2 outpatient-visits with a diagnosis code for diabetes between 1/1and 2009 and 31/12/2010 were included and the first diagnosis of diabetes or dispense date of an antidiabetic drug was defined as the index date. Those without continuous health-plan membership or drug benefit during the 12 months before and after the index date were excluded, as well as those with gestational diabetes. Ibáñez

Administrative claims data
Primary aim: To quantify persistent gender gaps in cardiovascular risk management and to assess the performance of routinely used commercial population health management tools in helping systems narrow gender gaps.
Data source: Anonymized data of medical and pharmacy claims, laboratory results, and enrolment data from one national health plan for commercial health plan members drawn from a population across Atlanta, Houston, New York City/Northern New Jersey and Southern California. Those with diabetes were included. Kreft

Administrative claims data
Primary aim: To investigate the influence of comorbidities on undergoing a diabetic eye exam in patients with newly diagnosed type 2 diabetes.
Data source: Data extracted from health insurance claims made between 1/2005 and 3/2013 using the database of Japan medical Centre Ltd. (Tokyo, Japan). This database consists of beneficiaries in health insurance unions across Japan in 2012, including individuals diagnosed with type 2 diabetes between 1/2005 and 3/2013 that had been prescribed antiglycaemic drugs with a 12-month follow-up from the index month. The index month was defined as the first month in which the study patients had been diagnosed with type 2 diabetes and received antiglycaemic drugs. Those who were not prescribed antidiabetic drugs after the index month were excluded, as well as those diagnosed with diabetes or prescribed antidiabetic drugs during the nine months after registration in the database, with diabetic retinopathy prior to the index month, those who had undergone eye exams, who had been diagnosed with eye diseases, or who had undergone an intervention for the eyes within the six months preceding the index month, in order to select patients who did not visit the ophthalmologist regularly. Lastly, those without information regarding the facility at which diabetes treatment took place in the index month were excluded. Varies per subtype and audit period, including ≥1 measurements during study period (15 months): HbA1c, blood pressure, cholesterol (triglycerides and another type of fat in the blood), creatinine, urine albumin, foot, smoking, BMI, combination (all eight care processes (excl. eye exam (<12 years only HbA1c)).

Administrative data
Primary aim: To measure the effectiveness of diabetes healthcare against NICE Clinical Guidelines and NICE Quality Standards, in England and Wales.
Data source: Administrative data extracted from participating general practitioners via pre-agreed extracts of their computer system and specialist diabetes service units in secondary care hospitals. This includes data from children being treated in adult care settings; but does not cover pediatric units. Both previously diagnosed and newly diagnosed individuals with type 2 diabetes during the audit period were included. General practices were invited to participate in the audit through their clinical systems.

Administrative claims data
Primary aim: To analyze compliance to HbA1c testing guidelines and explore associated individual and area-level determinants, focusing on regional variation.

Electronic medical records
Primary aim: To examine the extent to which variations in achievement to high impact indicators can be explained using routinely collected data.

Assumed to be selfreported
Primary aim: To investigate disparities with regard to foot self-care and lifestyle between men and women with diabetes.
Data source: Data extracted from a sample of individuals with type 2 diabetes living in the urban area of a large city in the South of Brazil. Study participants were drawn among those enrolled in the Hypertensive and Diabetics Individuals Registration System. Individuals selected were invited to participate in the study and data were collected at primary health care services through patient interviews and medical chart extraction. Those undergoing dialysis, with active ulcers in the lower limbs, without cognitive capacity, or not willing to participate were excluded. Follow-up eye exam after referral in the period between initial exam and interview.

Self-reported
Primary aim: To assess the proportion of patients not presenting for follow-up and the reasons for poor follow-up of diabetic patients after screening for retinopathy in Kilimanjaro Region of Tanzania.

Medical records
Primary aim: to compare the process of diabetes care of specialist diabetes clinics, and general medical clinics for different hospital level (regional, provincial, and community).

Self-reported
Primary aim: To determine which patient characteristics were associated with failure to receive indicated care for diabetes over time.

Medical records
Primary aim: To examine whether disparities in quality of care, intensity of lipidlowering medication regimen and medication adherence explain gender disparities in dyslipidemia.
Data source: Data extracted from the observational component of the R2D2C2 study, enrolling a sample of individuals from 7 outpatient clinics affiliated with an academic medical centre. The patient sample was drawn from a diabetes registry representing adults with a diagnosis of type 2 diabetes who had ≥1 encounters with family medicine, internal medicine or endocrinology within a 12 month period. All participants completed a baseline questionnaire. Medical records were abstracted for the 12-month period leading up to the date the questionnaire was completed.

Electronic medical records
Primary aim: To assess the strength of associations between patient factors and diabetes care processes and outcomes. Electronic medical records of the billing and accounting system Primary aim: To investigate eye care utilization among patients with diabetes who are seen in a county hospital clinic in the South that primarily serves high risk low income patients who are predominantly non-Hispanic African Americans.
Data source: Data extracted from two years of follow-up data, to examine eye care utilization among diabetes patients seen in 2007 at the internal medicine clinic of a large, urban, county hospital that serves primarily low income, non-Hispanic African American patients( Birmingham Alabama). The date of their first clinic visit in 2007 was defined as an index date. Follow-up (retrospectively) was carried out by linking patients' personal identifiers, i.e., medical record numbers, to electronic records of the hospital's billing and accounting system which included dates and procedures of patient encounters in the hospital's ophthalmology clinic. Those with ophthalmic complications were excluded.

Self-reported
Primary study aim: To determine the level of health care access for older Hispanics with type 2 diabetes living in a US-Mexico border area, and personal and health correlates to health care utilization.
Data source: Data obtained by community-based assessment conducted in 2008 at a clinic, senior centers, and colonias. Colonias are impoverished neighborhoods with substandard living conditions along the US-Mexico border. The health assessment included Hispanics with type 2 diabetes, living in Hidalgo County, Texas, at the Texas-Mexico border. To maximize recruitment, certified community health workers recruited participants in both clinical and community settings, including: a federally-qualified community health clinic that provided services for the uninsured and low-income individuals, and 2 nonclinical-based settings (community senior centers and colonias). Druss

Administrative claims data
Primary study aim: To evaluate the impact of a "piece-rate" pay for performance (P4P) program aimed at improving diabetes care processes, outcomes and related healthcare utilization for patients enrolled in a not-for-profit Medicaid-focused managed care plan.

Administrative claims data
Primary aim: To assess Kansas Medicaid data to determine the quality of diabetic care and the level to which individuals with physical disabilities' prevention and diabetes management needs are being met.
Data source: Data extracted from individuals with physical disabilities and diabetes who received medical benefits through Kansas Medicaid. Kansas Medicaid program provides insurance coverage for inpatient, outpatient, pharmacy, long term care and hospice coverage to adults with disabilities who qualify for Supplemental Security Income, have high medical needs, qualify for Medicare, or have a severe disability and are awaiting permanent federal disability status. Each of these programs has its own income qualifications. Persons with diabetes-related claims during a 12-month period (7/2007 -6/2008) were identified and quality of care was followed the subsequent 12 months. All individuals included were continuously eligible for the entire 24 months. Gold

Medical records
Primary aim: To determine whether patients with serious mental illness receiving care in Veterans Affairs mental health programs with collocated general medical clinics were more likely to receive adequate medical care than those in programs without collocated clinics based on a nationally representative sample.

Medical records
Primary aim: To assess correlations addressing this central question, namely, how are changes in primary care panel size related to patient processes and satisfaction, and the amount of (waiting) time to be seen by a primary care doctor?

Medical records
Primary aim: To compare access and uptake of screening between groups of people with diabetes in each of three screening programs covering this area of southern England.
Data source: Data extracted from a patient-level dataset using data from general practices that refer to three diabetic retinopathy screening programmes and a combined health record. The Hampshire Health Record received data from approximately two thirds of general practices in the region, and from secondary care. It is used by clinicians to share information between primary and secondary care, and provides a rich source of contemporaneous data with potential for public health use. Multiple diabetes diagnosis codes were used in order to capture all registered people with diabetes. Diabetic retinopathy screening in England is provided by local programs with guidance and quality assurance oversight from the English National Screening.

Administrative claims data
Primary aim: To investigate the effectiveness of a pay-for-performance program to increase the receipt of quality care and to decrease hospitalization rates among patients with diabetes.
Data source: Demographic, pharmacy, inpatient, and outpatient administrative medical claims data from 1/1/1999, through 31/12/2006 were used. The study sample consisted of individuals with diabetes who saw Pay for Performance (PP4P)participating physicians or non-P4P-participating physicians exclusively. Those who saw both P4P-participating and non-P4P-participating physicians were excluded. P4P, implemented by a large provider of healthcare coverage in Hawaii, provides participating physicians with financial incentives to perform quality-of-care processes. Participation in the P4P is voluntary. Tomio

Administrative claims data
Primary aim: To assess the quality of diabetes care in two communities in Japan by using National Health Insurance claims data.

Medical record and/or selfreported
Primary aim: To determine the frequency and correlates of persistent long-term gaps in diabetes care.
Data source: Data extracted from patient surveys and reviews of medical records to assess preventive care services for previously diagnosed type 2 diabetes among those who were continuously enrolled in 10 US managed care plans from 1999 to 2002. Participants were considered eligible if they had been continuously enrolled in the health plan for at least 3 years, submitted at least 1 claim in the first 18 months, were not pregnant, and participated in follow-up survey. Those with probable type 1 diabetes were excluded. Whether HbA1c, lipid tests, and urine albumin tests were received was based solely on chart abstraction, while eye and foot exam were considered to have been received if they were self-reported or recorded in the medical record. Ng

≥45
Combination (HbA1c, eye exam and foot exam) in the 12 months prior to survey.

Self-reported
Primary aim: To examine the relation of age, gender and insurance status to quality of care among Americans with diabetes and cardiovascular conditions. Data source: Data extracted from nationally representative MEPS data (2004-2006 pooled). MEPS is a health survey developed to analyze health care use, expenditures and insurance coverage for the U.S. civilian noninstitutionalized population. The MEPS Household Component (MEPS HC) provides estimates of respondents' demographic and socioeconomic characteristics, access to care, health insurance coverage and effectiveness of care for an array of priority clinical conditions, including cardiovascular disease. The MEPS also collects information on diabetes care effectiveness separately through a self-survey, the MEPS Diabetes Care Supplement. Non-institutionalized individuals with self-identified diabetes were eligible for inclusion. Older adults who reported being "uninsured" were excluded.

Medical records or selfreported
Primary aim: To assess the use of eye care and its predictors among diabetic patients in China.
Data source: Between February and November 2009, those with physician-diagnosed diabetes were recruited from an urban tertiary and community hospitals and from a rural clinic in Guangdong, China. Subjects having been diagnosed less than 12 months previously or who were unable to cooperate with the interview were excluded. Outcomes were defined according to documentation in the patient's chart, and when this was unavailable or dates were not stated clearly, by the subject's self-report. Gulliford

Electronic medical records
Primary aim: To determine the extent of socioeconomic and ethnic differentials in diabetic retinopathy screening uptake and screening outcomes following the implementation of the screening programme.

Electronic medical records
Primary aim: To estimate the prevalence of diabetes by age, gender and ethnicity, to look at quality of care, and to investigate disparities in care.
Data source: Data extracted from three general practices in Hamilton (New Zealand), including those with type 2 diabetes (prevalent and newly diagnosed).

Electronic medical records
Primary aim: To measure quality of vascular risk factor measurement and control in people with type 2 diabetes after comprehensive pay-for-performance implementation and to examine variation by patient and practice characteristics.