Association of diabetic foot care and amputation rates in Germany

Abstract

Podologists are nurses who care for the diabetic foot (orthotics, offloading devices, blisters, calluses, treatment of fungus infection and patient education). In contrast to podiatrists, they are not qualified to perform any surgical treatment or wound care. We analysed whether there is an association between the decrease in major amputations and the number of podologic foot care (PFC) visits prescribed in Germany. Detailed list of all major lower limb amputations (OPS 5‐864) performed from 2007 to 2011 was provided by the Federal Statistical Office. Data were separated for the 16 federal states in Germany. Detailed lists of the number of PFC treatments for each of the 5 years were derived from the federal report of the statutory health insurance. The total numbers of hospitalised cases per year having diabetes mellitus documented as an additional diagnosis were used to adjust for the different rates of people with diabetes in each federal state. Within a 5‐year time period, population‐based major amputations per 100 000 people dropped from 21·7 in 2007 to 17·5 in 2011 (−18·5%); whereas the number of PFC treatments per 1000 insured increased from 22 in 2007 to 60 in 2011 (+172·7%). The total number of major amputations divided by the total number of hospitalised cases with the additional diagnosis of diabetes mellitus (DM) shows an inverse correlation with the number of PFC treatments per 1000 insured (Pearson's correlation factor is −0·52049). The five countries with the highest increase in PFC compared with the five countries with the lowest increase (35·6 versus 15·4 per 1000 insured) will have only small differences in the decrease in major amputation rates in this period (−5·1 versus −3·4 per 100.000). There is a strong association between increasing utilisation PFC and decreasing major amputations in Germany. Further study is required to document the cost‐effectiveness of this service.

Introduction

In recent years, some data have been published presenting promising numbers of decreasing amputation rates in the USA 1, 2, 3. A small trend was reported from the UK 4 and Australia 5. In Germany total number of hospitalised cases treated by major amputations has decreased in the last years 6.

The causative pathway leading from a foot ulcer to amputation is well known. The progressive additive effects of neuropathy, minor trauma, ulceration, faulty healing, ischaemia, and infection leading to amputation were first characterised in 1990 7. Theoretically, the decrease in amputation rates may correlate to a time period where the benefits of endovascular or surgical procedures positively impacted the outcomes in patients with peripheral arterial disease (PAD). One possible factor for observing the improved outcomes at this time is the rapid improvements in endovascular technology. Guidewires and balloons have smaller profiles, stent options now include drug elution and prosthetic graft covering, and new generation thrombolytic agents are available 1, 2, 3.

On the other hand, diabetic foot care has been recognised as a relevant problem and a lot of local strategies have been implemented in the recent decade. Especially, in Germany diabetic foot networks have developed and built effective support. Diabetes nurses play a role in the education of prevention of diabetic foot ulcers, foot care and prevention of foot injury. Such specialist nurses are called podologists in Germany. In contrast to the podiatrists, podologists are not doctors and they are not allowed to use a scalpel or other instruments for active debridement of bleeding tissue. They care for the diabetic foot disease outside the normal medical setting.

If a patient has diabetes mellitus and is at a high risk of developing a foot lesion, his or her doctors can prescribe diabetic foot care performed by podologists. The health insurance providers are required to pay the reimbursement costs for this service. In the 5 years between 2007 and 2011, expenses for podologic foot care (PFC) have clearly increased. Thus, an unknown but probably high number of patients with diabetes presenting with an at‐risk foot are effectively treated before any surgical procedures become necessary. We analysed whether there is an association between the decrease in major amputations and the number of PFC visits prescribed in Germany.

Patients and methods

In Germany, PFC can be prescribed by the doctors and the service is reimbursed by the statutory health insurance (SHI). Roughly 90% of the population in Germany are statutorily insured and entitled to receive health promotion and disease prevention benefits to maintain and restore their health or to improve their state of health. The SHI is funded by the statutory health insurance funds which, being public law corporations, are financially and organisationally independent. They carry out the tasks assigned to them by the State independently of the interprofessional team model.

The SHI‐remedies‐information system reports and regulatory development issues are discussed in the public domain. The reports are based on billing data from all health insurance companies. A federal report based on the unaudited, service provided billing records according to a German law regulation (§ 302 SGB V) is published every quarter. For data security, all data with personal references are de‐identified. These reports are freely available on the website www.gkv‐his.de. We extracted data regarding the general costs and numbers of PFC visits.

The national statistics [Diagnosis Related Group (DRG) statistics] published by the Federal Statistical Office includes data from all hospitals in Germany that use the DRG system, representing more than 99% of hospitals. These hospitals are legally obliged to deliver extensive data on hospital treatment, including demographic data, diagnoses, co‐morbidities, complications and procedures to the ‘Institute for the Hospital Remuneration System’ (InEK) which transfers the data for a yearly summary to the Federal Statistical Office. For the years 2007–2011 all diagnoses were coded with the International Statistical Classification of Diseases and Related Health Problems as defined in the tenth Revision (ICD‐10) with the German Modification (ICD‐10‐GM) in the corresponding annual version. We have selected the OPS‐codes for minor and major amputation.

Statistics

A detailed list of all major lower limb amputations (OPS 5‐864) performed for each of the 5 years were provided by the Federal Statistical Office (Table 1). The data were separated for the 16 federal counties in Germany. To focus on PAD and neurovascular disease (NVD) as the underlying disease associated with amputation, we excluded patients with the principal diagnosis of injury and toxicity (ICD‐Codes: S00‐T98), musculoskeletal disorder (M00‐M99), diseases of skin and subcutaneous tissue (L90‐L99) and cancer or malignant neoplasm (C00‐D48). The analysis is based on cases and not on individual patients. As a consequence, a patient could be included several times in the statistics if he or she had an amputation at two different times within one or subsequent years.

Table 1.

Incidence of major (OPS 5‐864) amputation for the 16 federal states in Germany

Years	Major Amputation absolute						Major Amputation per 100 000
	2007	2008	2009	2010	2 011	Delta	2007	2008	2009	2010	2 011	Delta
Germany	17 846	17 316	16 323	15 196	14 463	−19·0	21·7	21	19·9	18·6	17·7	−18·5
Baden‐Württemberg	1973	1949	1810	1682	1565	−20·7	18·4	18	16·8	15·6	14·5	−20·8
Bavaria	2788	2588	2485	2281	2134	−23·5	22·3	21	19·9	18·2	17·0	−23·8
Berlin	634	658	631	546	534	−15·8	18·6	19	18·4	15·8	15·3	−17·5
Brandenburg	640	585	571	566	571	−10·8	25·2	23	22·7	22·6	22·9	−9·3
Bremen	146	126	148	99	133	−8·9	22·0	19	22·4	15·0	20·1	−8·5
Hamburg	272	248	251	224	172	−36·8	15·4	14	14·1	12·6	9·6	−37·8
Hesse	1191	1228	1100	1068	972	−18·4	19·6	20	18·1	17·6	16·0	−18·5
Mecklenburg‐Western Pomerania	472	447	472	409	371	−21·4	28·0	27	28·5	24·8	22·7	−19·1
Niedersachsen	1685	1626	1527	1497	1405	−16·6	21·1	20	19·2	18·9	17·7	−16·0
North Rhineland Westphalia	3731	3509	3266	3215	3009	−19·4	20·7	20	18·2	18·0	16·9	−18·6
Rhineland‐Pfalz	717	739	683	631	629	−12·3	17·7	18	17·0	15·7	15·7	−11·2
Saarland	290	250	242	210	208	−28·3	27·9	24	23·6	20·6	20·5	−26·5
Saxony	1216	1225	1152	1030	967	−20·5	28·7	29	27·6	24·8	23·4	−18·7
Saxony‐Anhalt	890	917	867	747	805	−9·6	36·7	38	36·6	31·9	34·6	−5·5
Schleswig‐Holstein	469	492	470	401	400	−14·7	16·5	17	16·6	14·2	14·1	−14·7
Thuringia	732	729	648	590	588	−19·7	31·8	32	28·7	26·3	26·4	−17·0

Detailed lists of number of PFC treatments and derived costs for each of the 6 years were derived from the federal report of the SHI. The data were separated for each of the federal counties in Germany. These federal reports include total number of single treatment, total cost per 1000 insured in euros, number of prescriptions per 1000 insured and number of treatments per 1000 insured (Table 2).

Table 2.

Data describing the amount of PFC performed in diabetic patients paid by the SHI in the 16 federal states in Germany

Years	Total costs in thousands of euro			Total cost per 1000 insured in euro			Number of treatments per 1000 insured
	2007	2011	Delta	2007	2011	Delta	2007	2011	Delta
Germany	36 011	101 035	180·6	561	1568	179·5	22	60	172·7
Baden‐Württemberg	2932	8464	188·7	356	1030	189·3	13	37	184·6
Bavaria	5655	13 788	143·8	593	1427	140·6	22	51	131·8
Berlin	1117	4196	275·6	443	1616	264·8	18	62	244·4
Brandenburg	771	2473	220·8	377	1207	220·2	18	54	200·0
Bremen	160	535	234·4	380	1022	168·9	12	37	208·3
Hamburg	759	1991	162·3	601	1473	145·1	23	54	134·8
Hesse	1639	4842	195·4	351	1030	193·4	13	38	192·3
Mecklenburg‐Western Pomerania	762	2122	178·5	548	1554	183·6	26	71	173·1
Niedersachsen	2605	11 860	355·3	409	1871	357·5	15	68	353·3
North Rhineland Westphalia	4587	12 369	169·7	658	1758	166·9	24	64	164·3
Rhineland‐Pfalz	1738	4529	160·6	552	1442	161·2	21	53	152·4
Saarland	661	1795	171·6	810	2236	176·0	30	81	170·0
Saxony	4059	9617	136·9	1144	2755	140·8	54	127	135·2
Saxony‐Anhalt	1825	4154	127·6	910	2118	132·7	42	97	131·0
Schleswig‐Holstein	1029	2849	176·9	466	1270	172·5	17	47	176·5
Thuringia	1059	2982	181·6	545	1581	190·1	26	73	180·8

PFC, podologic foot care.

The 16 German Federal States are different in number of inhabitants, average age and prevalence rates of persons with diabetes. Unfortunately there are no valid population‐based data regarding the differences in diabetic population demographics. We compared the total number of hospitalised cases per year with diabetes mellitus documented as an additional diagnosis provided by the Federal Statistical Office to adjust for the variations in diabetes in each county (Table 3).

Table 3.

Incidence of hospitalised cases with the additional diagnosis diabetes mellitus for the 16 federal states in Germany

Years	2007	2008	2009	2010	2011	Delta
Germany	2 115 707	2 209 062	2 315 494	2 419 714	2 515 462	18·9
Baden‐Württemberg	212 968	222 315	234 545	243 237	255 006	19·7
Bavaria	305 512	318 596	336 718	346 537	357 962	17·2
Berlin	90 094	92 753	96 179	100 660	104 779	16·3
Brandenburg	86 097	89 872	95 356	99 074	102 716	19·3
Bremen	17 020	18 640	19 673	20 575	21 922	28·8
Hamburg	35 915	39 326	41 065	44 121	46 392	29·2
Hesse	139 749	143 977	150 362	158 500	163 871	17·3
Mecklenburg‐Western Pomerania	61 927	63 712	66 205	71 017	72 754	17·5
Niedersachsen	163 482	172 484	183 398	195 016	204 769	25·3
North Rhineland Westphalia	450 410	471 823	492 523	517 691	542 025	20·3
Rhineland‐Pfalz	105 071	110 714	115 947	119 421	123 534	17·6
Saarland	34 231	35 197	36 106	37 330	39 948	16·7
Saxony	165 400	174 194	180 575	188 253	191 999	16·1
Saxony‐Anhalt	96 910	99 829	104 579	107 805	109 753	13·3
Schleswig‐Holstein	54 295	56 970	59 671	64 246	67 823	24·9
Thuringia	96 626	98 660	102 592	106 231	110 209	14·1

Calculations were performed using Microsoft^® Excel 2003 and Microsoft^® Access 2003.

Results

Within the 5‐year time period of the study, the major amputations decreased by 19% from 17 846 in 2007 to 14 463 in 2011 (Table 1). The decrease in major amputations varied between the different counties and was lowest in Bremen with −8·9% and highest in Hamburg with −36·8%. Overall population‐based major amputations per 100 000 people dropped from 21·7% in 2007 to 17·5% in 2011 (−18·5%).

Within this time period, expenses for PFC increased by 180·6% from €3·6 million in 2007 to €11·1 million in 2011 in Germany (Table 2). The number of treatments per 1000 insured increased from 22 in 2007 to 60 in 2011 (+172·7%). The increase in treatments per 1000 insured varied between the different counties and was lowest in Saxony‐Anhalt with 131·0% and highest in Niedersachsen with 353·3%.

In this period, the total number of German hospitalisations for persons with diabetes mellitus (DM) documented as an additional diagnosis increased by 18·9% (from 2 115 707 in 2007 to 2 515 462 in 2011). The increases varied between 13·3% in Saxony‐Anhalt and 29·2% in Hamburg (Table 3). To adjust for the burden of DM in the different counties, we divided the total number of major amputations by the total number of hospitalised cases with the additional diagnosis DM and plotted it against the number of PFC treatments per 1000 insured for each county. As the plot (Figure 1) shows there is an inverse association between DM‐adjusted decrease in major amputations and the increase in number of PFC treatments per 1000 insured in Germany in the period from 2007 to 2011.

Figure 1.

The ratio of total number of major amputations is divided by the total number of hospitalised cases with the additional diagnosis DM (y‐axis) and plotted against the number of podologic foot care (PFC) treatments per 1000 insured for each county (x‐axis).

To explore the hypothesis that the more PFC was prescribed was associated with lower amputation rates, we analysed the changes in the five counties with the highest increase in PFC (top five) and compared them with the five counties with the lowest increase (lowest five). As Figure 2 demonstrates, the lowest five counties with PFC services had a higher number of major amputations (30·6 versus 19·7 per 100 000 in 2007) compared with the five counties with higher rates of PFC (35·6 versus 15·4 per 1000 insured). In general, the increase in PFC from 2007 to 2011 (+54·2 versus 26·6 per 1000 insured) was associated with a small difference in amputations (−5·1 versus −3·4 per 100 000).

Figure 2.

The number of amputations per 100 000 people and the number of prescribed units of podologic foot care (PFC) in the five counties with the highest increase in PFC from 2007 to 2011 (top five) compared with the counties with the lowest increase (lowest five).

Discussion

This analysis is the first to demonstrate a decrease in major amputation and the increase in PFC. The system with PFC prescribed by the doctors and paid by the Social Health Insurance (SHI) is very unique to Germany. Once again it has to be pointed out that PFC does not include any wound treatment or surgical procedures. Podologists care for the diabetic foot at risk and the use of a scalpel is not mandatory for comprehensive prevention of lesions such as hyperkeratosis. Podologists may treat callus by filing, sanding and levelling with special tools. Soaking and softening callus simplifies the treatment and avoids tissue lesions. They try to prevent callus development by a good foot care to keep the skin soft and intact and by local pressure relief using a variety of plasters and paddings.

The pathophysiologic mechanisms underlying diabetic foot disease are multi‐factorial and include neuropathy, infection, ischaemia, and abnormal foot structure and biomechanics. It is, therefore, not surprising that the management of the diabetic foot is a complex clinical problem requiring an interdisciplinary team. Integrated care models have been created as trans‐sectoral approach to link in‐hospital and outpatient care. This is a very important step as in Germany SHI pays for in‐hospital treatment and outdoor treatment but in absolute different ways and out of two different budgets. Thus diabetic foot care has been recognised as a relevant problem and several local strategies have been implemented in the recent decade. Diabetic foot networks have developed and built effective support with PFC in these settings.

The reported association between PFC and a reduction of major amputation might not be causal. Sixty treatments per 1000 insured in 2011 means that 60 of 1000 people can get one treatment a year or 5 people out of 1000 can get a regular PFC each month. Considering the high number of persons with diabetes in Germany, these numbers are very low. If the right patients receive PFC, this therapy has the potential to be very effective in the prevention of foot lesions. The right patients might be those with skin lesions in initial stage or patients with a history of ulcers or minor amputation as among the diabetic population those patients have the highest risk of developing foot lesions 8.

Finally it might not be the absolute number of PFC visits that are relevant. The increase in prescribed PFC might be a marker calling for an increased awareness in the doctors and the general population for foot problems in people suffering from diabetes. Perhaps these patients receive more attention from their doctors, more often consult specialised diabetic foot centres and have more or earlier revascularisation before amputation is required. Unfortunately we do not have data to support these aspects.

Diabetic foot care has been recognised as a relevant problem and several local strategies have been implemented in the recent decade. Diabetic foot networks have developed and built effective support. Thus an unknown but probably high number of patients suffering from foot lesions are effectively treated in outdoor settings.

One argument supporting a higher awareness might be that those top five counties with the highest increase in PFC from 2007 to 2011 had a much higher major amputation rate in 2007 than the lowest five. Thus the high increase in PFC might be triggered or at least the result of the higher amputation rates in these counties. The measured effect is small but with a higher increase in prescribed PFC, we would expect an even larger decrease in amputation rates as shown in Figure 2.

The effectiveness of diabetic foot prevention programmes has been shown before. There was a local lower‐extremity disease management programme consisting of screening and treatment protocols for diabetic members in a managed care organisation in San Antonio, Texas. Amputation incidence was tracked from 1 September 1998 to 31 December 2000 and compared with 12 months of historic data for this population prior to implementation of the treatment. It decreased by 47·4% from 12·89 per 1000 diabetic patients per year to 6·18 (P < 0·05) 11. In contrast we report a health policy decision to provide nationwide diabetic foot care by the SHI.

Our analysis is just descriptive and some limitation of the given data should be considered. Although routine data from the electronic patient record is frequently used for secondary purposes, there is currently no systematic analysis of coding quality in Germany 9, 10. Whether coding matches reality as a prerequisite for further use of the data in medicine and health policy requires quality assurance verification.

The data is subject to error as it is not based on personal records, but rather on DRG‐data and SHI‐remedies‐information system reports. The results of the distal vascular surgery and endovascular treatment also are likely to be a contributing factor. We have reported a 7% increase in absolute number of cural and pedal bypasses and a 60% increase of cural endovascular procedures in Germany from 2005–2008 12.

Population size in Germany is decreasing with a shift towards old and very old people and an increasing burden of DM. To cover these changes we adjusted for the number of hospitalised patients having DM documented as an additional diagnosis. We do not know how accurately this number represents the burden of DM in the different counties. We assume that this parameter is not only influenced by the number of diabetic people but also by the quality of treatment of the DM.

Diabetic foot ulcers predate lower limb amputations in 85% of cases 13. A systematic way of identifying the high‐risk foot with a screening tool and implementing a referral pathway for early intervention of vascular changes, deformity and plantar pressure redistribution with neuropathy will reduce the amputations. A prompt referral with an integrated and coordinated screening programme (e.g. simplified 60‐second tool) could further increase the effectiveness of podologic care 14.

In conclusion, there is an association between increasing expenses of PFC and decreasing major amputations in Germany. In this setting, PFC can play an even greater role with an integrated coordinated programme to identify all patients with a high risk of diabetic foot ulcers. In addition, identification of specific high‐risk foot screening tools, regular podiatric/foot care professional debridement along with the optimisation of haemoglobin A1c (HbA1c), blood pressure control and formalised interprofessional patient education may further decrease the incidence of amputations as documented in other published studies.