Comparative hospitalization cost and length of stay between patients with and without diabetes in a large tertiary hospital in Johannesburg, South Africa

Healthcare Resource Utilization in Controlled Versus Uncontrolled Adults Living With Type 1 Diabetes in the South African Public Healthcare Sector

OBJECTIVES

This study aimed to understand the cost implications of managing people living with type 1 diabetes mellitus in the South African public healthcare system.

METHODS

A multicenter, noninterventional retrospective chart review study was performed. Data on healthcare resource consumption, demographics, risk factors, clinical history, and acute events were collected. Direct medical costs were collected over a 1-year period, stratified by controlled versus uncontrolled patients. In addition, the costs in people with controlled (glycated hemoglobin < 7%) versus uncontrolled glycated hemoglobin (≥ 7%) at time horizons of 1, 5, 10, and 25 years were modeled using the IQVIA Core Diabetes Model.

RESULTS

The costs based on the retrospective chart review were $630 versus $1012 (controlled versus uncontrolled population). The modeled costs at various time horizons were as follows: at 1 year, $900 versus $1331; at 5 years, $4163 versus $6423; at 10 years, $7759 versus $16 481; and at 25 years, $16 969 versus $66 268. The largest cost in the controlled population was severe hypoglycemia requiring nonmedical assistance, severe hypoglycemia requiring medical assistance, and treatment costs. In the uncontrolled population, the largest cost was the cost of diabetic ketoacidosis, severe hypoglycemia requiring nonmedical assistance, severe hypoglycemia requiring medical assistance, and foot complications.

CONCLUSIONS

Strict glycemic control reduces healthcare resource use overall. Patients in the controlled group still experienced high resource use related to hypoglycemic events. The introduction of a structured patient education program and analog insulins may result in less episodes of hypoglycemia and potential cost savings.

The costs of interventions for type 2 diabetes mellitus, hypertension and cardiovascular disease in South Africa - a systematic literature review

BACKGROUND

In the context of a move to universal health coverage, three separate systematic reviews were conducted to summarise available evidence on the direct costs of interventions for type 2 diabetes mellitus, hypertension, and cardiovascular disease in South Africa.

METHODS

PubMed® and Web of Science was searched for literature published between 01 and 1995 and 27 October 2022. Additionally, reference and citations lists of retrieved articles and experts were consulted. We also tracked reference lists of previous, related systematic reviews. Eligible publications were cost analyses of clinical interventions targeted at adults age 15 + reporting primary estimates of in- and out-of-hospital costs from a provider perspective. Costs were extracted and converted to 2021 US dollars, and article methodological and reporting quality was appraised using the 2013 CHEERS checklist.

RESULTS

Of the 600, 1,172 and 1,466 identified publications for type 2 diabetes mellitus, hypertension, and cardiovascular disease, respectively, 10, 12, and 17 met full inclusion criteria. 60% of articles reported cardiovascular disease costs, 52% were of good reporting quality, and 10%, 50%, and 39% of type 2 diabetes mellitus, hypertension and cardiovascular disease papers reported private-sector costs only. Hypertension drug costs ranged from $2 to $85 per person-month, while type 2 diabetes mellitus drug costs ranged between $57 and $630 per person-year (ppy). Diabetes-related complication treatment costs ranged from $55 for retinopathy treatment to $25,193 ppy for haemodialysis, while cardiovascular disease treatment costs were between $160 and $37,491 ppy. Drugs and treatment of complications were major cost drivers for hypertension and type 2 diabetes mellitus, while hospitalisation drove cardiovascular disease costs.

CONCLUSION

The intervention costs of type 2 diabetes mellitus, hypertension and cardiovascular disease care have received more attention recently, particularly diabetes-related complications and cardiovascular disease. However, 39% of identified cardiovascular disease treatment costs used a private sector perspective, leaving significant research gaps in the public sector and the cheaper to treat hypertension and type 2 diabetes mellitus. This review fills an information gap regarding the intervention costs of these diseases in South Africa.

Effectiveness of an adapted diabetes nutrition education program on clinical status, dietary behaviors and behavior mediators in adults with type 2 diabetes: a randomized controlled trial

PURPOSE

This study evaluated the effectiveness of an adapted social-cognitive theory underpinned diabetes nutrition education program (NEP) on: clinical (HbA1c, BMI, blood lipids, blood pressure) and selected dietary behaviors (starchy foods and energy intake, vegetables and fruit intake) and behavior mediators (knowledge and diabetes management self-efficacy) in patients with type 2 diabetes mellitus (T2DM).

METHODS

A tertiary hospital outpatient adults (40-70 years) with poorly controlled (HbA1c ≥ 8 %) T2DM were randomized to either intervention group (n = 39: NEP, 7-monthly group education sessions, bi-monthly follow-up sessions, 15-minute individual session, workbook + education materials) or control group (n = 38: education materials only). NEP aimed to improve clinical status through improved dietary behaviors and behavior mediators. Outcomes and changes in diabetes medication were assessed at six and 12 months. Intention-to-treat analysis was conducted. ANCOVA compared the groups (baseline values, age, sex adjustments).

RESULTS

Forty-eight (62.3 %) participants completed the study. Intervention group compared to the control group had lower (-0.53 %), clinically meaningful HbA1c (primary outcome) at 6 months, albeit not sustained at 12 months. Compared to the control group, the intervention group had significantly lower: (i) systolic blood pressure at six and 12 months (ii) diastolic pressure at 12 months, (iii) energy intake at six-months, (iv) up-titration of insulin at six and 12 months and higher diabetes knowledge scores at six months.

CONCLUSIONS

NEP had limited effects on HbA1c, targeted dietary behaviors and behavior mediators but showed positive effects on blood pressure. The NEP health cost savings potential supports the need for improving program participation.

TRIAL REGISTRATION

ClinicalTrials.gov. number NCT03334773; 7 November 2017 retrospectively registered.

Magnitude and predictors of hospital admission, readmission, and length of stay among patients with type 2 diabetes at public hospitals of Eastern Ethiopia: a retrospective cohort study

BACKGROUND

Type 2 Diabetes (T2D) represents one of the leading causes for hospital admissions and outpatient visits. Hence, T2D continuously imposes a significant burden to healthcare systems. The aim of this study was to assess predictors of hospital admission, readmission rates, and length of hospital stay among T2D patients in government hospitals of Eastern Ethiopia from 2013 to 2017.

METHODS

This study utilized retrospective data from a cohort of T2D patients following their treatment in government hospitals in Harari regional state of Ethiopia. Predictor of hospital admission was determined using parametric survival analysis methods. The readmission rate and length of hospital stay were determined by Poisson regression and mixed effect Poisson regression, respectively. All association were performed at 95% confidence level. Significance of association with determinants was reported using the hazard rate for hospital admission, and the incidence rate for readmission and length of hospital stay. Optimal model for each outcome was selected by using information criteria after fitness was checked.

RESULTS

The hospital admission rate for T2D patients was 9.85 (95%CI: 8.32, 11.66) per 1000-person-year observation. Alcohol drinking, inactive lifestyle, being a rural resident, history of comorbidities, and experiencing chronic diabetes complications were predictors of hospital admission. Seventy-one (52.2%) of the admitted patients had a history of readmission. Readmission rate was increased by being female, duration of disease, inactive lifestyle, having BMI greater than 29.9 kg/m2, and higher blood glucose. The median time of hospital stay for admitted patients was 18 (IQR:7). The length of hospital stay was longer among females, patients with the history of insulin administration, and higher blood glucose.

CONCLUSION

Multiple and complex factors were contributing for high diabetes admission and readmission rates as well as for longer in-hospital duration among T2D patients in Harari regional state. Socio-demographic characteristics (sex, place of residence), behavioral factors (alcohol intake, lifestyle), and medical conditions (longer duration of disease, comorbidities, chronic diabetes complications, higher blood glucose level, and treatment modality) were significant determinants of hospital admission, readmission and longer hospital stay among T2D patients.

Incidence, socio-economic inequalities and determinants of catastrophic health expenditure and impoverishment for diabetes care in South Africa: a study at two public hospitals in Tshwane

BACKGROUND

Direct out of pocket (OOP) payments for healthcare may cause financial hardship. For diabetic patients who require frequent visits to health centres, this is of concern as OOP payments may limit access to healthcare. This study assesses the incidence, socio-economic inequalities and determinants of catastrophic health expenditure and impoverishment amongst diabetic patients in South Africa.

METHODS

Data were taken from a cross-sectional survey conducted in 2017 at two public hospitals in Tshwane, South Africa (N = 396). Healthcare costs and transport costs related to diabetes care were classified as catastrophic if they exceeded the 10% threshold of household's capacity to pay (WHO standard method) or if they exceeded a variable threshold of total household expenditure (Ataguba method). Erreygers concentration indices (CIs) were used to assess socio-economic inequalities. A multivariate logistic regression was applied to identify the determinants of catastrophic health expenditure and impoverishment.

RESULTS

Transport costs contributed to over 50% of total healthcare costs. The incidence of catastrophic health expenditure was 25% when measured at a 10% threshold of capacity to pay and 13% when measured at a variable threshold of total household expenditure. Depending on the method used, the incidence of impoverishment varied from 2 to 4% and the concentration index for catastrophic health expenditure varied from - 0.2299 to - 0.1026. When measured at a 10% threshold of capacity to pay factors associated with catastrophic health expenditure were being female (Odds Ratio 1.73; Standard Error 0.51), being within the 3rd (0.49; 0.20), 4th (0.31; 0.15) and 5th wealth quintile (0.30; 0.17). When measured using a variable threshold of total household expenditure factors associated with catastrophic health expenditure were not having children (3.35; 1.82) and the 4th wealth quintile (0.32; 0.21).

CONCLUSION

Financial protection of diabetic patients in public hospitals is limited. This observation suggests that health financing interventions amongst diabetic patients should target the poor and poor women in particular. There is also a need for targeted interventions to improve access to healthcare facilities for diabetic patients and to reduce the financial impact of transport costs when seeking healthcare. This is particularly important for the achievement of universal health coverage in South Africa.

Quantifying the hidden healthcare cost of diabetes mellitus in Australian hospital patients

BACKGROUND

Diabetes mellitus in hospital inpatients is most commonly present as a comorbidity rather than as the primary diagnosis. In some hospitals, the prevalence of comorbid diabetes mellitus across all inpatients exceeds 30%, which could add to complexity of care and resource utilisation. However, whether and to what extent comorbid diabetes mellitus contributes indirectly to greater hospitalisation costs is ill-defined.

AIM

To determine the attributable effect of comorbid diabetes mellitus on hospital resource utilisation in a General Internal Medical service in Melbourne, Australia.

METHODS

We extracted data from a database of all General Internal Medical discharge episodes from July 2012 to June 2013. We fitted multivariable regression models to compare patients with diabetes mellitus to those without diabetes mellitus with respect to hospitalisation cost, length of stay, admissions per year and inpatient mortality.

RESULTS

Of 4657 patients 1519 (33%) had diabetes mellitus, for whom average hospitalisation cost (AUD9910) was higher than those without diabetes mellitus (AUD7805). In multivariable analysis, this corresponded to a 1.22-fold (95% confidence interval (CI) 1.12-1.33, P < 0.001) higher cost. Mean length of stay for those with diabetes was 8.2 days versus 6.8 days for those without diabetes, with an adjusted 1.19-fold greater odds (95% CI 1.06-1.33, P = 0.001) of staying an additional day. Number of admissions and mortality were similar.

CONCLUSION

Comorbid diabetes mellitus adds significantly to hospitalisation duration and costs in medical inpatients. Moreover, diabetes mellitus patients with chronic complications had a greater-still cost and hospitalisation duration compared to those without diabetes mellitus.

Cost of diabetes mellitus in Africa: a systematic review of existing literature

Background

There is an increasing recognition that non communicable diseases impose large economic costs on households, societies and nations. However, not much is known about the magnitude of diabetes expenditure in African countries and to the best of our knowledge no systematic assessment of the literature on diabetes costs in Africa has been conducted. The aim of this paper is to capture the evidence on the cost of diabetes in Africa, review the methods used to calculate costs and identify areas for future research.

Methods

A desk search was conducted in Pubmed, Medline, Embase, and Science direct as well as through other databases, namely Google Scholar. The following eligibility criteria were used: peer reviewed English articles published between 2006 and 2016, articles that reported original research findings on the cost of illness in diabetes, and studies that covered at least one African country. Information was extracted using two data extraction sheets and results organized in tables. Costs presented in the studies under review are converted to 2015 international dollars prices (I$).

Results

Twenty six articles are included in this review. Annual national direct costs of diabetes differed between countries and ranged from I$3.5 billion to I$4.5 billion per annum. Indirect costs per patient were generally higher than the direct costs per patient of diabetes. Outpatient costs varied by study design, data source, perspective and healthcare cost categories included in the total costs calculation. The most commonly included healthcare items were drug costs, followed by diagnostic costs, medical supply or disposable costs and consultation costs. In studies that reported both drug costs and total costs, drug costs took a significant portion of the total costs per patient. The highest burden due to the costs associated with diabetes was reported in individuals within the low income group.

Conclusion

Estimation of the costs associated with diabetes is crucial to make progress towards meeting the targets laid out in Sustainable Development Goal 3 set for 2030. The studies included in this review show that the presence of diabetes leads to elevated costs of treatment which further increase in the presence of complications. The cost of drugs generally contributed the most to total direct costs of treatment. Various methods are used in the estimation of diabetes healthcare costs and the costs estimated between countries differ significantly. There is room to improve transparency and make the methodologies used standard in order to allow for cost comparisons across studies.

Electronic supplementary material

The online version of this article (10.1186/s12992-017-0318-5) contains supplementary material, which is available to authorized users.

Assessment of hospital length of stay and direct costs of type 2 diabetes in Hubei Province, China

BACKGROUND

The incidence of type 2 diabetes is increasing, creating a huge burden for China's social healthcare system. This study aimed to evaluate hospital length of stay (LOS) based on admission characteristics and direct costs correlated with various types of complications for type 2 diabetic inpatients in Hubei Province, China.

METHODS

A total of 1528 inpatients diagnosed with type 2 diabetes discharged between April 1, 2013, and March 31, 2014, were included in this study. Information regarding patients' admission and hospitalization were obtained from the hospital information system. The relationship between admission characteristics and LOS, distribution of total costs, and types of complications were described and analysed.

RESULTS

(1) The mean LOS was 11.65 days (median: 10 days). Multiple linear regression analysis demonstrated that inpatients with New Cooperative Medical Scheme (NCMS), aged 80 and above, had longer LOS than the reference group, and inpatients with chronic or acute + chronic complications had shorter LOS than those without. (2) Mean total costs per patient were US$159.72 ± 130.83 (median: US$135.33), US$240.60 ± 166.58 (median: US$192.09), and US$247.98 ± 166.22 (median: US$200.99) for inpatients with no complications, chronic complications, and acute + chronic complications, respectively. Total and individual costs were significantly less for patients without complications than for those with the two types of complications (p < 0.001). (3) Mean total costs per patient were US$225.40 ± 115.32 (median: US$200.34), US$221.25 ± 177.64 (median: US$170.05), and US$275.18 ± 193.14 (median: US$217.91) for inpatients with microvascular complications, macrovascular complications, and microvascular + macrovascular complications, respectively. Total costs were significantly higher for patients with microvascular + macrovascular complications than for those with other types of chronic complications (p < 0.001). (4) Drugs were the greatest expense for patients, and the least expensive treatment was nursing care.

CONCLUSIONS

Medical insurance status, age, and type of complication may help to predict LOS for patients with type 2 diabetes in Hubei Province, China. The total and individual costs for patients with complications were higher than for those without, and hospitalization expenses posed a heavy burden. Efforts should be made to reduce the financial impact on patients by integrating the medical insurance system of urban and rural areas, and by reducing the risk of complications, especially microvascular complications.