Sex related differences in therapy and outcome of patients with intermittent claudication in a real-world cohort

BACKGROUND AND AIMS

The prevalence of lower extremity artery disease (LEAD) is increasing worldwide and sex-related differences are a current matter of debate.

METHODS

We analysed claims data on unselected patients with in-patient treatment for LEAD with intermittent claudication (IC; Rutherford grade 1-3) from 01.01.2014 to 31.12.2015. Data files included diagnostic and procedural information from two years before index, and a five-year follow-up.

RESULTS

Our analysis comprised 42,197 IC patients, thereof 28,520 (68%) male. Male patients were younger (median: 66.4 years vs. 72.6 years) but presented with higher frequency of cardiovascular risk factors such as diabetes (40% female vs. 46% male), atrial fibrillation (13% vs. 17%), chronic coronary syndrome (41% vs. 53%), chronic heart failure (23% vs. 27%), or chronic kidney disease (29% vs. 32%; all p < 0.001; age adjusted). Revascularisation applied in 80% of patients, thereof endovascular approach predominantly in female and surgery in male patients. Concomitant pharmacotherapy with statins (74% at 2 years) and platelet inhibitors (75% respectively) were long lasting below guideline recommendation, under-use being more pronounced in women. Two years after index, one-third of IC patients had subsequent revascularisation, one-quarter progressed to chronic limb threatening ischemia (CLTI), and 2% underwent amputation. Male sex was an independent risk factor for long-term mortality (female HR 0.75; 95%-CI 0.72-0.79; p < 0.001) and CLTI (female HR 0.89; 95%-CI 0.86-0.92; p < 0.001) during follow-up.

CONCLUSIONS

The majority of in-patient treated patients for IC are male, presenting with worse cardiovascular risk profiles. In view of a general under-supply with statins and platelet inhibitors, women received somewhat less often preventive medication. Despite low LEAD stages at index, serious prognosis was observed in the long term. Particularly male patients were at high risk for all-cause mortality and the combined endpoint CLTI and death.

[Case volume and complications after thyroid gland surgery in Germany: an analysis of routine data from 48,387 AOK patients]

BACKGROUND

Many studies showed that hospital and surgeon volume have a significant influence on the complication rates of thyroid surgery. The present study investigates whether this relationship applies in subtotal as well as total lobe resections. Furthermore, it is still unclear which threshold for the hospital-related case volume can be determined, above which the risk of complications lies below the current national average.

MATERIAL AND METHODS

The study was based on nationwide routine data for persons insured with the Local General Sickness Fund (AOK) who had undergone thyroid surgery in 2014-2016. Permanent vocal cord palsy, bleeding and wound infection needing revision were recorded using indicators. The effect of the case volume on the indicators and the case number threshold was determined using logistic regression.

RESULTS

Permanent vocal cord palsy was observed in 1.3% and bleeding or wound infections needing revision in 1.6% and 0.3% of the cases. Compared to hospitals with >450 surgeries per year, the risk of permanent vocal cord palsy in hospitals with fewer than 201, 101 and 51 surgeries was significantly increased (OR [95% CI]: 1.5 [1.1-2.1]; 1.5 [1.1-2.1]; 1.8 [1.3-2.5]). The threshold needed to achieve a risk for permanent vocal cord palsy below the national average (1.3%) was 265 thyroid surgeries per year (95% CI: 110-420). For bleeding or wound infection in need of revision, no association between volume and outcome was found.

CONCLUSION

The present study showed that the risk of postoperative permanent vocal cord palsy decreased with increasing case volume. The broad confidence interval of the threshold makes clear case volume recommendation difficult. In order that the risk for a postoperative permanent vocal cord palsy is not likely above the national average, the annual case volume should reach 110 thyroid interventions.

Changes in incidence of severe hypoglycaemia in people with type 2 diabetes from 2006 to 2016: analysis based on health insurance data in Germany considering the anti-hyperglycaemic medication

AIM

To investigate the incidence of severe hypoglycaemia over the past 10 years, taking into account changes in anti-hyperglycaemic therapy.

METHODS

This retrospective population-based study used German health insurance data. All adults diagnosed with documented type 2 diabetes (extrapolated to the German population: 6.6 million in 2006; 7.9 million in 2011; 8.86 million in 2016) were screened for severe hypoglycaemia. Anti-hyperglycaemic agents were identified by Anatomical Therapeutic Chemical (ATC) code.

RESULTS

The event rate for severe hypoglycaemia was 460 per 100 000 people in 2006, 490 per 100 000 in 2011 and 360 per 100 000 in 2016. The proportion of people with severe hypoglycaemia receiving sulfonylureas, as well as receiving combination therapy of metformin and sulfonylureas decreased from 2006 to 2016 (23.6% vs. 6.2%) Among those with severe hypoglycaemia in 2006, there were no prescriptions for dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide-1 (GLP-1) receptor agonists or sodium-glucose co-transporter 2 (SGLT2) agonists. The proportions of people with severe hypoglycaemia receiving DPP-4 inhibitors, GLP-1 receptor agonists or SGLT2 agonists in 2011 and 2016 were low. The proportion of people receiving human insulin also decreased (from 11.3% in 2006 to 10.3% in 2011 and 4.3% in 2016); the proportion of people receiving insulin analogues increased from 5.4% in 2006 to 11.5% in 2016. Therapy with mixed insulins was used by 19.7% of people with severe hypoglycaemia in 2006, by 14.0% in 2011 and by 7.3% in 2016. People undergoing therapy with insulin analogues have the highest risk of severe hypoglycaemia adjusted by age, gender, nephropathy diagnosis and year of survey [odds ratio (OR) 14.4, 95% confidence interval (95% CI) 13.5-15.5].

CONCLUSION

The incidence of severe hypoglycaemic events in Germany increased between 2006 and 2011, and decreased in 2016.

Hospital volume and outcome in inguinal hernia repair: analysis of routine data of 133,449 patients

Introduction

Inguinal hernias are repaired using either open or minimally invasive surgical techniques. For both types of surgery it has been demonstrated that a higher annual surgeon volume is associated with a lower risk of recurrence. This present study investigated the volume-outcome implications for recurrence operations, surgical complications, rate of chronic pain requiring treatment, and 30-day mortality based on the hospital volume.

Materials and methods

The data basis used was the routine data collected throughout the Federal Republic of Germany for persons insured by the Local General Sickness Fund “AOK” who had undergone inpatient inguinal hernia repair between 2013 and 2015. Complications were recorded by means of indicators. Hospitals were divided into five groups on the basis of the annual caseload volume: 1–50, 51–75, 76–100, 101–125, and ≥ 126 inguinal hernia repairs per year. The effect of the hospital volume on the indicators was assessed using multiple logistic regression.

Results

133,449 inguinal hernia repairs were included. The incidence for recurrence operations was 0.95%, for surgical complications 4.22%, for chronic pain requiring treatment 2.87%, and for the 30-day mortality 0.28%. Low volume hospitals (1–50 and 51–75 inguinal hernia repairs per year) showed a significantly increased recurrence risk compared to high volume hospitals with ≥ 126 inguinal hernia repairs per year (odds ratio: 1.53 and 1.24). No significant correlations were found for the other results.

Conclusions

The study gives a detailed picture of hospital care for inguinal hernia repair in Germany. Furthermore, it was noted that the risk of hernia recurrence decreases in line with a rising caseload of the treating hospital.

Comparing routine administrative data with registry data for assessing quality of hospital care in patients with inguinal hernia

INTRODUCTION

Collecting clinical data is viewed as the gold standard for assessing health-care quality. However, considerable investment is needed if the quality of care is to be recorded over a long period of time. Accordingly, the question arises as to whether routine administrative data may be used for quality assurance purposes if certain methodological standards are followed. This present study now compares the outcomes of inguinal hernia repair from the Herniamed Hernia Registry with routine data from Germany's largest sickness fund, the Local General Sickness Fund "AOK".

MATERIALS AND METHODS

Included in the study were the hospital inpatient cases for the years 2011-2013 with inguinal hernia surgical repair whose data had been prospectively entered into the voluntary Herniamed Hernia Registry by 318 participating hospitals. These collectives were then compared, on the basis of the routine administrative data available, with patients from the AOK sickness fund who had been operated on during the same time period. The outcome criteria selected were the perioperative complication rates within 30 days as well as the recurrence rate and the pain rate requiring treatment at 1-year follow-up.

RESULTS

The data records examined comprised 64,748 cases from the Herniamed Registry and 130,121 AOK cases. Since in the Herniamed collective, the proportion of bilateral procedures was significantly higher, only the collectives of elective primary unilateral inguinal hernias in men (Herniamed n = 37,667; AOK n = 78,973) were compared. The most pronounced difference between these two collectives was in the proportion of laparo-endoscopic procedures (Herniamed 61.3% vs AOK 49.0%; p < 0.001). Accordingly, the Herniamed collective was found to have a significantly lower postoperative surgical complication rate (Herniamed 1.5% vs AOK 2.6%; p < 0.001) and surgical site infection (SSI) rate (Herniamed 0.3% vs AOK 0.6%; p < 0.001) within 30 days after the operation. On the other hand, the pain rates requiring treatment in the Herniamed collective were somewhat higher (Herniamed 3.0% vs AOK 2.6%; p < 0.001). No difference was found in the recurrence rate at 1-year follow-up (Herniamed 1.0% vs AOK 0.9%; ns).

CONCLUSION

Subject to critical evaluation of the limitations of data acquisition in this study, it does appear possible to use the routine administrative data from the AOK Sickness Fund for assessment of the quality of inguinal hernia surgery in Germany. Voluntary participation in the Herniamed Registry appears to be characterized by the fact that the participating hospitals conduct a higher proportion of laparo-endoscopic inguinal hernia repair. That could possibly explain the differences in outcome. However, in large patient collectives, statistical significance should not always be equated with clinical relevance. Univariate analysis does not take account of differences in the hernia findings, risk factors or operative details. Further efforts should be employed in future to improve the accuracy and granularity of routine administrative data for assessing the quality of care and to decrease the cost of gathering such data.

[One-year outcomes for proximal femoral fractures: Posthospital analysis of mortality and care levels based on health insurance data]

BACKGROUND

Proximal femur fractures are common and treatment is expensive. The aim of the present study was - after matching of hospital and health insurance data - to evaluate the influence of early operation on certain outcome parameters. Data from a German health insurance were used to identify then influence of the day of operation after admittance on the rate of mortality, decubitus, and revision surgery during the hospital stay and on the care level of the patients up to 1 year and in some cases longer after operation.

MATERIALS AND METHODS

In all, 7905 patients were included. The descriptive data, specifying the given population, described the hospital stay (occurrence, surgical procedures, early complications, secondary diagnoses, length of stay) and the course of patient recovery up to 1 year after the hospital stay (care level, late complications). The calculated data (analytical statistics) give correlations evaluating the influence of the length of the preoperative hospital stay on the outcome parameters mentioned above. Risk adjustment was performed by using secondary diagnoses.

RESULTS

The study included more women (mean age 81.5 years). Most common was the femoral neck fracture. Of the operations 77% were carried out on the first day after admission; dominating procedures were intramedullary nails and prostheses. Most common secondary diagnoses were diabetes, dementia, ischemic heart disease, and chronic heart insufficiency. Descriptive data revealed 6% early as well as late complications. In all, 50% of patients had a higher care level after operation. Almost 40% of patients changed from outpatient care to inpatient care. The time interval between admission and operation negatively influenced all outcome parameters. The relative risk to die, to develop decubitus, or to receive early revision was increased by approximately one third when patients were operated on later than the first day after admission. A total of 3172 patients died during the study period. Mortality after operation reached 9.9% within 30 days and 26.9% at 1 year. The mortality of patients operated after the first day was increased by more than 6% compared to patients treated within the first 24 h.

CONCLUSION

The present study clearly presents the importance of analysis of routine records after discharge and it demonstrates that longer periods up to 1 year and more can be evaluated. The data show that a longer time period between hospital admission and operation negatively influences all outcome parameters. The care data give impressive evidence for a significant loss of quality of life and the importance of intense postoperative rehabilitation.

[Quality in Revision Arthroplasty: A Comparison between Claims Data Analysis and External Quality Assurance]

BACKGROUND

External quality assurance for revisions of total knee arthroplasty (TKA) and total hip arthroplasty (THA) are carried out through the AQUA institute in Germany. Data are collected by the providers and are analyzed based on predefined quality indicators from the hospital stay in which the revision was performed. The present study explores the possibility to add routine data analysis to the existing external quality assurance (EQS). Differences between methods are displayed. The study aims to quantify the benefit of an additional analysis that allows patients to be followed up beyond the hospitalization itself.

MATERIAL AND METHODS

All persons insured in an AOK sickness fund formed the population for analysis. Revisions were identified using the same algorithm as the existing external quality assurance. Adverse events were defined according to the AQUA indicators for the years 2008 to 2011.The hospital stay in which the revision took place and a follow-up of 30 days were included. For re-operation and dislocation we also defined a 365 days interval for additional follow-up. The results were compared to the external quality control reports.

RESULTS

Almost all indicators showed higher events in claims data analysis than in external quality control. Major differences are seen for dislocation (EQS SD: 1.87 vs. claims data [cd] SD: 2.06 %, cd+30 d: 2.91 %, cd+365 d: 7.27 %) and reoperation (hip revision: EQS SD: 5.88 % vs. claims data SD: 8.79 % cd+30 d: 9.82 %, cd+365 d: 15.0 %/knee revision: EQS SD: 3.21 % vs. claims data SD: 4.07 %, cd+30 d: 4.6 %, cd+365 d: 15.43 %). Claims data could show additional adverse events for all indicators after the initial hospital stay, rising to 77 % of all events.

CONCLUSIONS

The number of adverse events differs between the existing external quality control and our claims data analysis. Claims data give the opportunity to complement existing methods of quality control though a longer follow-up, when many complications become evident.

Volume-outcome relationship in pancreatic surgery

BACKGROUND

Volume-outcome relationships related to major surgery may be of limited value if observation ends at the point of discharge without taking transfers and later events into consideration.

METHODS

The volume-outcome relationship in patients who underwent pancreatic surgery between 2008 and 2010 was assessed using claims data for all inpatient episodes from Germany's largest provider of statutory health insurance covering about 30 per cent of the population. Multiple logistic regression models with random effects were used to analyse the effect of hospital volume (using volume quintiles) on 1-year mortality, adjusting for age, sex, primary disease, type of surgery and co-morbidities. Additional outcomes were in-hospital (including transfer to other hospitals until final discharge) and 90-day mortality.

RESULTS

Of 9566 patients identified, risk-adjusted 1-year mortality was significantly higher in the three lowest-volume quintiles compared with the highest-volume quintile (odds ratio 1·73, 1·53 and 1·37 respectively). A similar, but less pronounced, effect was demonstrated for in-hospital and 90-day mortality. The effect of hospital volume on 1-year mortality was comparable to the effect of co-morbid conditions such as renal failure.

CONCLUSION

Although mortality related to pancreatic surgery is influenced by many factors, this study demonstrated lower mortality at 1 year in high-volume centres in Germany.

[Re-evaluation of the AOK hospital navigator with a focus on total knee replacement]

BACKGROUND

One of the biggest health insurance companies in Germany (AOK, Allgemeine Ortskrankenkasse) has published new results focussing on process quality of total knee replacement in 2010. These results were published in the online portal "Weiße Liste", which is based on health insurance routine data. The German Association of Orthopeadic Surgery questions the credibility of the rating system of the "Weiße Liste". To prove the system an interdisciplinary task force was created.

MATERIAL AND METHODS

The task force identified patient-specific parameters, which influence the outcome of total knee replacement based on the literature and expert opinions. Out of 907 orthopaedic departments, 4 above average and four below average were identified. The AOK was asked to provide 80 data sets for each department. These anonymised data sets could be converted into patient-specific data sets in the identified departments. Statistical analysis was performed to answer the question of whether there are differences between the below and the above average groups.

RESULTS

625 cases could be investigated. We found an increased rate of postoperative complications in the below average group. There are differences between both groups in terms of factors influencing the procedure. In the below average group an increased rate of patients with one or more comorbidities and a preoperative extension lag of over 10° was found. The above average group has a higher rate of operations before the knee replacement.

CONCLUSION

The results need to be proven on a larger scale. Further, prospective investigations are planned.

[Complication rates and secondary interventions after coronary procedures in clinical routine: 1-year follow-up based on routine data of a German health insurance company]

BACKGROUND

Data on 1-year complication and follow-up intervention rates after coronary angiography (CA) and percutaneous coronary intervention (PCI) in German clinical routine are sparse. This analysis aims to determine these rates.

METHODS

The analysis uses 2009 AOK claims data. Patients were divided into 3 groups (CA, without cardiac surgery and without acute myocardial infarction (AMI) n=116.071; PCI with stenting, without AMI: n=36.685; PCI with stenting and with AMI: n=32.707). The frequency of the endpoints MACCE (mortality, AMI, stroke, TIA), CABG, PCI and CA was recorded for up to one year.

RESULTS

1-year MACCE rates were 8.1 % (CA), 9.9 % (PCI without AMI) and 17.9 % (PCI with AMI). Quality-relevant follow-up intervention rates in the CA group were 2.5 % for CABG (after 31-365 days), 1.7 % for PCI within 90 days and 3.5 % for follow-up CA within 1 year. In the PCI groups, the frequencies were 1.6 % (without AMI) and 2.7 % (with AMI) for CABG (after 31-365 days), and 10.2 % (without AMI) and 10.1 % (with AMI) for PCI after 91-365 days.

CONCLUSION

This is the first cross-sectoral routine analysis of cardiac catheters and sequential events up to one year in Germany. The actual medical care situation revealed information particularly with regard to the second and follow-up inventions, which cannot be derived directly from medical guidelines. Beyond clinical trials, knowledge can be gained which is important both for medicine as well as the politics of health services.

[The relationship of in-hospital and post-discharge complications and implications for quality measurement in hip replacement surgery - an analysis of AOK administrative data]

AIM

This study analyses the information gain achieved by additionally taking into account complications in the follow-up period instead of merely considering in-house events for a hospital-based quality measurement using the example of hip replacement.

METHOD

The analysis was performed with anonymous statutory health insurance data (AOK) for the years 2007-2009 within the framework of the quality measurement method "Quality Assurance with Administrative Data (QSR)". It included cases of hip replacement surgery due to osteoarthritis. In order to analyse hospital-related outcome quality, 6 quality indicators were formed (revision surgery within 365 days, surgical complications within 90 days, thrombosis/pulmonary embolism within 90 days, femur fracture within 90 days, mortality within 90 days and complication index). For each hospital, the adjusted SMRs (standardised mortality or morbidity ratio) with 95% confidence intervals were calculated. The relation between the in-hospital and the follow-up SMR was analysed by Spearman's rank correlation coefficient. Furthermore, the percentage consistency of hospital SMRs categorised into quartiles on the basis of in-hospital and post-discharge events was determined.

RESULTS

A total of 154 470 AOK patients from 930 hospitals were included in the analysis. The hospitals had a median overall complication rate of 11,22%. One quarter of the hospitals had complication rates of 8,18% or below. Another quarter of the hospitals had complication rates nearly twice as high (≥15,49%). Nearly one-third of all complications occurred after the initial hospitalisation. Regarding clinic-related complications, there was little correlation between the events in the initial case and during follow-up (r<0,3) for all indicators. The order of the hospitals defined by quartiles of SMR changed significantly by adding the complications in the follow-up for the indicators considered (min 21%, max 47% changes between quartiles). In particular, for the indicators revision and death, a change in the SMR quartile occurred in almost 50% of all hospitals.

CONCLUSION

Quality assessment of hip replacement surgery based exclusively on in-house events is quite unreliable. On the one hand, nearly a third of all complications occur in the follow-up period. On the other hand, predicting the occurrence of post-discharge events from in-house complications of a clinic is not considered acceptable for the indicators analysed in this study.

Risk of malignancies in patients with diabetes treated with human insulin or insulin analogues: a cohort study

AIMS/HYPOTHESIS

The aim of this cohort study was to investigate the risk of malignant neoplasms and mortality in patients with diabetes treated either with human insulin or with one of three insulin analogues.

METHODS

Data were provided by the largest German statutory health insurance fund (time-frame: January 1998 to June 2005 inclusive), on patients without known malignant disease who had received first-time therapy for diabetes mellitus exclusively with human insulin, aspart, lispro or glargine. The primary outcome was the diagnosis of a malignant neoplasm. Data were analysed by multiple Cox regression models adjusting for potential confounders.

RESULTS

A total of 127,031 patients were included, with a mean follow-up time of 1.63 (median 1.41, maximum 4.41) years. A positive association between cancer incidence and insulin dose was found for all insulin types. Because patients receiving combined therapy with insulin analogues and human insulin were excluded, the mean daily dose was much lower for glargine than for human insulin, and a slightly lower cancer incidence in the glargine group was found. After adjusting for dose, a dose-dependent increase in cancer risk was found for treatment with glargine compared with human insulin (p < 0.0001): the adjusted HR was 1.09 (95% CI 1.00 to 1.19) for a daily dose of 10 IU, 1.19 (95% CI 1.10 to 1.30) for a daily dose of 30 IU, and 1.31 (95% CI 1.20 to 1.42) for a daily dose of 50 IU. No increased risk was found for aspart (p = 0.30) or lispro (p = 0.96) compared with human insulin.

CONCLUSIONS/INTERPRETATION

Considering the overall relationship between insulin dose and cancer, and the lower dose with glargine, the cancer incidence with glargine was higher than expected compared with human insulin. Our results based on observational data support safety concerns surrounding the mitogenic properties of glargine in diabetic patients. Prospective long-term studies are needed to further evaluate the safety of insulin analogues, especially glargine.

[Annual patient volume and survival of very low birth weight infants (VLBWs) in Germany--a nationwide analysis based on administrative data]

BACKGROUND

Despite numerous international analyses addressing survival of very low birthweight infants (VLBWs) in relation to patient volume, only few regional analyses from Germany exist. Thus, the aim of this study was to perform analyses on a nationwide database relating to this topic.

MATERIALS AND METHODS

We used nationwide hospital claims data according to data exchange processes designated in section sign 301 Social Security Code, of patients insured with the local healthcare insurance fund (AOK), with a hospital admission weight of 300-1 499 g, an age of less than 29 days, and a hospital discharge date between 1 Jan 2002 and 30 June 2006. Mortality within 30 days of hospital admission was used as end point. Crude and risk adjusted analyses were performed using logistic regression.

RESULTS

A comparison with federal statistics data showed that our dataset covers 28 % of all VLBWs born in Germany during the study period. 9 487 VLBWs could be included in risk adjusted volume-outcome analyses. Significantly increased risk adjusted 30-day-mortality rates in smaller units were observed using different thresholds. The relationship was more pronounced in the most recent time period starting from 2004.

CONCLUSION

Adoption of a minimum VLBW patient volume in addition to the structural requirements consented by the Federal Joint Committee (G-BA) seems reasonable, to maintain or improve the quality of VLBW health care provision in Germany.

[How frequent are diabetes-related amputations of the lower limbs in Germany? An analysis on the basis of routine data]

BACKGROUND AND OBJECTIVE

The total number of lower limb amputations in Germany as well as the frequency of lower limb amputations attributable to diabetes is unknown. Currently available estimates are based on regional studies resulting in different totals. The aim of this study was to perform a census of all lower limb amputations in Germany in order to obtain valid estimates of the number of amputations attributable to diabetes mellitus in Germany.

PATIENTS AND METHODS

Hospital performance and expenditure statistics were used to obtain a comprehensive count of lower limb amputations. The number of amputations in diabetic patients as well as the number of diabetes-related amputations were assessed by calculating the proportion of diabetes among amputees as well as attributable amputations standardized to the age and gender distribution in Germany in 2001, using routine data from the Local Health Insurance Funds (AOK) as well as previous analyses from Germany.

RESULTS

In 2001, 43,544 lower limb amputations and additional 3,981 revisions of amputations were performed in Germany. Among these almost 29,000 lower limb amputations were estimated to be performed in diabetic patients. More than 26,000 lower limb amputations per year were estimated to have been attributable to pre-existing diabetes mellitus.

CONCLUSIONS

Lower limb amputations as a complication of diabetes mellitus are more frequent in Germany than previously estimated. Intensified prevention and therapy--as intended in disease management programs--are urgently needed to decrease the complications of diabetes and ultimately to reduce diabetes mellitus-related costs.