Inzicht in de rol van diabetes in het artroseziekte- en behandelingsproces: een studieprotocol voor het Zweedse cohort Osteoartritis and Diabetes (SOAD)

Inzicht in de rol van diabetes in het artroseziekte- en behandelingsproces: een studieprotocol voor het Zweedse cohort Osteoartritis and Diabetes (SOAD)

januari 15, 2020 0 Door admin

Translating…


CBD Olie kan helpen bij artrose. Lees hoe op MHBioShop.com


Huile de CBD peut aider avec l’arthrose. Visite HuileCBD.be


 

Understanding the role of diabetes in the osteoarthritis disease and treatment process: a study protocol for the Swedish Osteoarthritis and Diabetes (SOAD) cohort

Loading
  1. Andrea Dell’Isola1,2,
  2. Johanna Vinblad3,4,
  3. Stefan Lohmander1,
  4. Ann-Marie Svensson RN, PhD5,6,
  5. Aleksandra Turkiewicz2,
  6. Stefan Franzén7,8,
  7. Emma Nauclér3,
  8. A W-Dahl9,10,
  9. Allan Abbott11,
  10. L Dahlberg1,
  11. Ola Rolfson3,4,
  12. Martin Englund2

  1. 1Faculty of Medicine, Department of Clinical Sciences, Orthopedics, Lunds University, Lund, Sweden

  2. 2Faculty of medicine, Department of Clinical Sciences, Orthopedics, Clinical Epidemiology Unit, Lund University, Lund, Sverige, Sweden

  3. 3Centre of Registers Västra Götaland, The Swedish Hip Arthroplasty Register, Goteborg, Sweden

  4. 4Department of Orthopaedics, Sahlgrenska Academy, University of Gothenburg, Institute of Clinical Sciences, Gothenburg, Sweden

  5. 5National Diabetes Register, Centre of Registers in Region Västra Götaland, Goteborg, Sweden

  6. 6Department of Molecular and Clinical Medicine, University of Gothenburg, Goteborg, Sweden

  7. 7National Diabetes Register, Centre of Registers Västra Götaland, Gothenburg, Sweden

  8. 8Health Metrics Unit, Sahlgrenska Academy, University of Gothenburg, Goteborg, Sweden

  9. 9Department of Clinical Sciences, Lund University, Lund, Sverige, Sweden

  10. 10The Swedish Knee Arthroplasty Register, Lund, Sweden

  11. 11Department of Medical and Health Sciences (IMH), division of physiotherapy, Faculty of Medicine and Health Sciences, Linkoping University, Linköping, Sweden
  1. Correspondence to Dr Andrea Dell’Isola; andrea.dellisola{at}med.lu.se

Abstract

Introduction Osteoarthritis (OA) is the most common form of arthritis and a leading cause of disability worldwide. Metabolic comorbidities such as type II diabetes occur with a higher rate in people with OA than in the general population. Several factors including obesity, hyperglycaemia toxicity and physical inactivity have been suggested as potential links between diabetes and OA, and have been shown to negatively impact patients’ health and quality of life. However, little is known on the role of diabetes in determining the outcome of non-surgical and surgical management of OA, and at the same time, how different OA interventions may affect diabetes control. Thus, the overall aim of this project is to explore (1) the impact of diabetes on the outcome of non-surgical and surgical OA treatments and (2) the impact of non-surgical and surgical OA treatments on diabetes control.

Methods and analysis The study cohort is based on prospectively ascertained register data on a national level in Sweden. Data from OA patients who received a first-line non-surgical intervention and are registered in the National Quality Register for Better Management of Patients with Osteoarthritis will be merged with data from the Swedish Knee and Hip Arthroplasty Registers and the National Diabetes Register. Additional variables regarding patients’ use of prescribed drugs, comorbidities, socioeconomic status and cause of death will be obtained through other national health and population data registers. The linkage will be performed on an individual level using unique personal identity numbers.

Ethics and dissemination This study received ethical approval (2019-02570) from the Swedish Ethical Review Authority. Results from this cohort will be submitted to peer-reviewed scientific journals and reported at the leading national and international meetings in the field.

  • osteoarthritis
  • general diabetes
  • register
  • cohort
  • exercise
  • surgery

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

View Full Text

Statistics from Altmetric.com

Strengths and limitations of this study

  • This study will use a large nationwide population-based cohort based on data from national quality registers with high coverage and completeness to explore the relationship between diabetes and osteoarthritis (OA) and their related care process.

  • We will include data regarding both non-surgical and surgical treatments for patients with OA giving the possibility to capture the influence of diabetes across the whole spectrum of OA treatments.

  • We will include covariate information from several national registers that will allow to account for potential confounders and effect modifiers.

  • A limitation of register-based studies is that the variables available and the characteristics of the treatments provided are predetermined, that is, it is not possible to add covariates, exposures or outcomes (not in the registers) or to modify the interventions that have been given.

  • People included in the National Quality Register for Better Management of Patients with Osteoarthritis Register, the Swedish Hip Arthroplasty Register, the Swedish Knee Arthroplasty Register received an intervention due to OA. Due to the complexity of the OA disease, treatments are individualised based on patient’s and disease characteristics, which implies that selection bias and confounding by indication may bias our estimates.

Introduction

Osteoarthritis (OA) is the most common form of arthritis and affect mainly the knee and the hip joint.1 In Sweden, more than 25% of the population aged >45 years is estimated to suffer from OA-related pain symptoms and associated physical activity restrictions.2 The average annual cost for a person affected by OA is reported to exceed €2000, while the total European expense directly attributable to OA is estimated to be as high as €700 billion.3

In addition to the already huge health and societal burden of OA, recent studies suggest that OA patients are twice as likely to have comorbidities compared with controls of the same age, indicating that the co-occurrence of multiple conditions in OA patients is the norm rather than the exception.4 For instance, based on data showing a higher incidence of knee OA in overweight patients with metabolic disorders, a metabolic OA phenotype has been hypothesised.5–7 Among the metabolic disorders, diabetes seems to play a central role due to its high prevalence and the toxic effect that hyperglycaemia has on the cartilage and its cells, and to the motor and sensory system through peripheral neuropathy.8–10

According to data from the Swedish National Diabetes Register (NDR), approximately 5% of the Swedish population has diabetes, with type II diabetes accounting for approximately 90% of the cases.11 Persons with diabetes have a higher risk of developing cardiovascular diseases and have a twofold to fivefold increased risk of mortality compared with the general population.12

In OA patients the prevalence of diabetes has been reported to be nearly three times higher than in the general population.7 Obesity is a shared risk factor for OA of the knee and the hip and diabetes and can partially explain the association between these diseases.13 14 In addition to the mechanical overload caused by the excess weight, adipocytes release cytokines into the bloodstream promoting chronic low-grade inflammation and activating proteolytic enzymes which can trigger matrix degradation and initiate OA. At the same time, adipose-induced low-grade inflammation influences the metabolic dysregulation underlying several metabolic disorders, for example, diabetes type II.14

Once diabetes is initiated, it further promotes cartilage degeneration and joint inflammation causing enrichment of advanced glycation end products and matrix stiffening preventing optimal cushioning of the joint.7 15 This process leads to a worsening in OA symptoms promoting physical inactivity and weight gain and creating a vicious cycle that maintains the metabolic dysregulation and increases joint symptoms.13 16–18

The evidence-based first-line management for people with hip and knee OA includes education and exercise which are recommended regardless of OA disease severity, and weight loss for those overweight.19 Metabolic comorbidities may have a significant impact on the treatment, partially explaining the lack of response experienced by some patients.

Replacement of the knee and of the hip is an effective treatment for patients with severe OA who do not sufficiently improve after non-surgical management.20 Due to the rising prevalence of OA and the growing demand for this procedure, the number of hip and knee replacements has dramatically increased. In Sweden (total population 10 million), 14 700 primary total hip replacements (THRs) and nearly 14 000 total knee replacements (TKRs) were performed in 2017 with OA as indication. These figures account for 81% and 97% of the annual hip and knee replacements, respectively, and translate in an annual incidence of nearly 150 procedures per 100 000 persons for both THR and TKR.21 22

Considering the association between diabetes and OA, surprisingly little is known regarding the influence that diabetes has on the outcome of OA treatments (both non-surgical and surgical).23 24 In addition, no evidence exists regarding the effect that OA treatments (both non-surgical and surgical) may have on diabetes control (both for types I and II). Thus, merging data from multiple Swedish registers will allow us to follow patients with knee and hip OA through the progress of their disease to understand how diabetes influences the OA disease process. This study cohort is created to increase knowledge of the influence that diabetes has on the outcomes of OA patients who have received non-surgical and/or surgical treatments for hip and knee OA, and the influence that hip and knee OA and its treatments have on the diabetes control.

Methods and analysis

Research questions

In order to understand how the coexistence of OA of the hip or of the knee and diabetes influences the treatment effects in these diseases, a series of research questions have been posed. The research questions cover two main thematic areas: (1) the impact of diabetes on the outcome of non-surgical and surgical OA treatments and (2) the impact of non-surgical and surgical OA treatments on diabetes control (consideration to type of diabetes (I or II) will be taken).

Area 1

  1. What is the prevalence of diabetes in people with OA undergoing a non-surgical intervention?

  2. Is the presence of diabetes, diabetes-related factors (eg, type of diabetes, diabetes-related medication, blood pressure, haemoglobin subunit alpha 1c (HbA1c)) associated with OA severity (eg, pain intensity, pain frequency, walking difficulties) of people with OA undergoing a self-management non-surgical intervention?

  3. Is the presence of diabetes and diabetes-related factors associated with the outcomes of a self-management non-surgical intervention for people with OA (eg, change in pain levels, pain frequency, walking difficulties)?

  4. Is the presence of diabetes and diabetes-related factors associated with the risk of joint replacement in people with OA who underwent a self-management non-surgical intervention?

  5. What is the incidence of re-operations and other adverse events such as thromboembolism, cardiovascular events and mortality following primary THR or TKR due to OA in people with or without diabetes?

  6. What diabetes-related factors are associated with the risk of re-operation and other adverse events following primary THR or TKR among person with diabetes?

Area 2

  1. How does a self-management non-surgical intervention for OA influence diabetes (type I vs type II) control (eg, change in diabetes drug intake after the intervention, change in HbA1c levels after the intervention) compared with people with diabetes who had not taken part in the intervention?

  2. How does primary THR or TKR influence the diabetes control compared with comparable persons with diabetes but with no history of hip or knee arthroplasty?

  3. What diabetes-related risk factors are associated with diabetes control following primary THR or TKR due to OA?

Main exposures and outcomes

The exposure and outcome measures are described in table 1. Potential confounding factors for main analysis and disease subanalysis are described with examples.

Table 1

Exposure and outcome for the study populations and examples of confounders and effect modifiers for the study analyses

The Swedish OA and diabetes cohort

This nationwide observational study cohort (Swedish OA and diabetes (SOAD)) will be based on prospectively obtained individual-level data from four main sources: the National Quality Register for Better Management of Patients with Osteoarthritis (BOA) Register, the Swedish Hip Arthroplasty Register (SHAR), the Swedish Knee Arthroplasty Register (SKAR) and the NDR. Data starting from the year of each register establishment will be merged using the unique personal identity number (PIN) issued to all legal residents in Sweden. Additional variables regarding patients’ use of prescribed drugs, comorbidities, cause of death and socioeconomic information will be obtained through the following population-based registers:

  • The Swedish Prescribed Drug Register held by the National Board of Health and Welfare.

  • The National Patient Register held by the National Board of Health and Welfare; information regarding in-hospital diagnoses and outpatient specialist care diagnoses, for example, interventions, adverse events such as thromboembolism or other comorbid conditions.

  • Swedish Cancer Register, The National Board of Health and Welfare.

  • The Cause of Death Register held by the National Board of Health and Welfare

  • Longitudinal integration database for health insurance and labour market studies (LISA) held by Statistics Sweden for data such as marital status, educational level and country of origin.

Data sources

BOA: The BOA register was started in 2008 and currently includes more than 100 000 individuals with OA who have registered for an evidence-based self-management programme. These patients sought treatment for knee and/or hip pain in primary healthcare in Sweden and were referred for standardised core treatment (education and supervised exercises) after a confirmed clinical/radiographic OA diagnosis in accordance with the recommendations for OA diagnosis from the Swedish National Board of Health and Welfare.25 These guidelines are in line with internationally accepted diagnostic criteria, and according to the guidelines, radiographic examination should only be used in uncertain cases, if the patient is not responding to treatment or when a surgical intervention is planned.26 27 BOA offers all the patients two education sessions focusing on the pathophysiology of OA and the benefit of exercise which are mandatory for participating in the second (exercise) session of the programme. A third, optional, session held by a trained OA communicator (a patient with OA who previously participated in BOA) is offered to provide a patient’s perspective on OA self-management and to teach about the lived experience with this condition, as well as his or her personal experience of non-surgical interventions. After the education, participants can take part in the exercise phase of BOA which consists of a face-to-face session with a physiotherapist. In this session, the patients receive a personalised intervention programme and the necessary instructions to perform it independently at home. Thereafter, participants are given the possibility to perform their exercise programme on their own or to participate in up to 12 supervised group exercise session with a physiotherapist provided two times per week for 6 weeks. Thus, the register contains two separate cohorts that performed, in addition to the education sessions, either home exercise or supervised exercise. The register has a data completeness of almost 90% and the BOA participants have answered validated and patient-relevant sociodemographic and outcome questionnaires at baseline, after the interventions (2–5 months) and at 1 year (12–15 months) (table 2).

Table 2

Description of the single variables collected from the BOA register

SHAR: Started in 1979, SHAR registers primary hip replacement operations and re-operations in Sweden, including individual patient data, surgical technique and type of implant used. Since 2002 patient-reported measures such as joint pain, Health Related Quality of Life (HRQoL) and satisfaction with treatment have also been collected before surgery and 1, 6 and 10 years postoperatively. The register encompasses 318 000 primary THRs due to OA and 61 500 re-operations after THRs where OA was the main reason for the primary surgery (at the end of 2018). The register has overall data completeness of 98.5% (2016) including all indications for THRs (table 3).

Table 3

Description of single variables collected from the SHAR

SKAR: The SKAR is a Swedish National Quality Register founded in 1975. The register collects individual patient data, surgical technique and type of implant used for patients who undergo knee replacement. The SKAR also collects information on re-operations/revision surgery. SKAR has completeness of 98.1% (2016) and has registered almost 270 000 primary knee replacements due to OA and more than 21 400 revisions at the end of 2018 (table 4).

Table 4

Description of single variables collected from the SKAR

NDR: NDR has been a Swedish National Quality Register since 1996 and collects data on clinical characteristics, risk factors, laboratory analyses, complications of diabetes and medications for patients 18 years of age or older with a diagnosis of diabetes (table 5). The completeness is 96.5% (2017) and the register has 750 004 (2017) unique individuals in their database. More than 95% of all individuals with type I diabetes mellitus (T1DM) and 90% of individuals with type II diabetes mellitus (T2DM) in Sweden are included in the NDR.

Table 5

Description of single variables collected from the NDR

Data linkage

PIN: In Sweden, all legal residents are registered with a unique PIN that provides information on the date of birth and sex. Swedish law requires all documentation regarding healthcare contacts to be registered using the patient’s PIN.28 The PIN is also used for registration of data for statistics such as national population-based registers and healthcare quality registers.29 30 The system allows for linkage of data at an individual level between the different registers in Sweden with the possibility of creating merged research databases for epidemiological research on large populations, after the relevant ethical approval has been obtained. Data linkage for the current study will include all the data available in BOA, SHPR, SKAR and NDR and it will start from the first time point available in the registers. The data linkage process has been initiated and it is described in figure 1. Data linkage is expected to be completed by 2020. Estimated start and end dates for the project are 1 September 2020 and 1 September 2030, respectively.

Figure 1

The data linkage process. Data from the four national quality registers, BOA register, SHAR, SKAR and NDR, are safely transferred to statistics Sweden. Statistics Sweden will anonymise the data by replacing pin with serial numbers. Data will be extracted from LISA (longitudinal integration database for health insurance and labour market studies) and transferred to the entity principally responsible for the SOAD cohort research. The PIN and serial numbers will also be shared with national board of health and welfare which will return data from NPR, SPDR, CODR and SCR to the entity principally responsible for the research. The linkage key will be saved at statistics Sweden for 3 years to allow the possibility of adding more year cohorts or new variables to the research database if new research questions arise (with new ethical approval). BOA, Better management of Patients with osteoarthritis; CODR, Cause of Death Register; NDR, National Diabetes Register; NPR, National Patient Register; PIN, personalidentity number; SCR, Swedish Cancer Register; SHAR, Swedish Hip Arthroplasty Register; SKAR, Swedish Knee Arthroplasty Register; SOAD, Swedish Osteoarthritis and Diabetes; SPDR, Swedish Prescribed Drug Register.

Analysis plan

Data harmonisation will be performed, and the more reliable data source will be used to guarantee information quality and reliability across exposed and unexposed subjects. To establish the reliability of the source, we will consider how the measurement (eg, self-reported vs measured) and data quality (eg, percentage of missing) were performed. If the same variable (eg, BMI) will be present in the source deemed as most reliable at more than one time point, we will use the measurement closest to the time point of interest.

We will develop a specific statistical analysis plan for each specific study that will be conducted within SOAD. These will follow several general principles. We will aim for the inclusion of all available knee and hip OA and diabetic patients to limit potential selection bias. We will use multiple imputation methods to impute the missing data on exposures, outcomes and confounders, when relevant. The imputation model will be specific for each study and compatible with the chosen anal ysis model. De volledig voorwaardelijke specificatie (ook wel geketende vergelijkingen genoemd) kan bijvoorbeeld worden gebruikt om flexibele modellen mogelijk te maken voor een juiste imputatie van alle variabelen. In de statistische modellering zullen we streven naar schatting van causale effecten en statistische modellen zullen dienovereenkomstig worden gekozen met behulp van directe acyclische grafieken om juiste confounding controle mogelijk te maken. 31 Voor confounding control gebruiken we regressiemodellen of inverse kansweging. Voor analyse van paneldata (dat wil zeggen herhaalde longitudinale metingen van de deelnemers en / of gegevens geclusterd door zorgverlener) zullen we multilevel regressiemodellen gebruiken. Voor time-to-event-gegevens gebruiken we het Cox-regressiemodel met proportionele gevaren, of, indien van toepassing, parametrische modellen. Voor bemiddelingsanalyses zullen we, indien nodig, lineaire modellen of maximale waarschijnlijkheidsmodellen voor structurele vergelijkingen gebruiken. Voor categorische resultaten zullen we andere benaderingen gebruiken. 32 33 We zullen de resultaten van alle analyses rapporteren als relevante geschatte effectgrootte (zoals risicoverschillen, risicoverhoudingen van gevarenratio’s) met 95% CI’s en interpreteren deze voor klinische relevantie, ongeacht de statistische significantie. 34 35

Voor het huidige onderzoek hebben we geen vermogensberekening uitgevoerd. Dit omdat de kwestie van vermogen als secundair kan worden beschouwd in een dergelijke instelling waar de steekproefgrootte wordt bepaald door de beschikbaarheid van gegevens en niet a priori wordt besloten. Bovendien zullen we bij het interpreteren van resultaten het concept van statistische significantie niet gebruiken, maar we zullen onze interpretatie baseren op effectgroottes (en de onzekerheid eromheen) en klinische relevantie. 36 Daarom nemen we alle beschikbare gegevens op, met ~ 100.000 BOA-deelnemers, waarvan naar schatting 15.000 diabetes hebben. 37 Met betrekking tot de SHAR en SKAR zullen we gegevens hebben voor respectievelijk 240 000 en 320 000 gezamenlijke vervangingen. Op basis van eerdere studies verwachten we dat de prevalentie van diabetes ongeveer 8% en 14% zal zijn bij patiënten die respectievelijk THR en TKR ondergaan. 38 39 We verwachten dat dit nauwkeurige schattingen van de belangrijkste effecten van interesse mogelijk maakt.

Betrokkenheid van de patiënt en het publiek

Vertegenwoordigers van de patiënten waren niet betrokken bij de ontwikkeling van de onderzoeksvraag of de opzet van deze studie. Patiënten waren echter betrokken bij de oprichting van het BOA-ondersteunde zelfmanagementprogramma en droegen bij aan de ontwikkeling van de belangrijkste inhoud van het programma. 40 Patiënten zijn ook actief betrokken bij het BOA-programma en geven een verplichte opleidingssessie waarin het perspectief van de patiënt op OA-zelfmanagementbehandeling wordt verkend. De Zweedse heup- en knie-artroplastiekregisters hebben patiëntenvertegenwoordigers in hun respectieve stuurcomités.

Ethiek en verspreiding h2>

Opslag en beheer van gegevens

Een kopie van de volledige gegevensset wordt opgeslagen in het Center of Registers Västra Götaland, Göteborg, Zweden. Een tweede kopie van de volledige gegevensset wordt opgeslagen op de Lund University op het platform LUSEC (Lund informatiebeveiligingsplatform). De platforms zijn ontworpen om gegevens veilig op te slaan, te beheren en te analyseren in overeenstemming met de algemene verordening gegevensbescherming van de Europese Unie. Het proces van koppeling, opslag en beheer van gegevens, de rol van geïnformeerde toestemming in registergebaseerd onderzoek en het waarborgen van de integriteit van deelnemers aan de studie volgt de wettelijke en ethische kaders zoals beschreven door de Zweedse wet en ethische raden. Dit is beschreven door Ludvigsson et al . 28

div >

Verspreiding

De resultaten van deze studie zullen worden gepubliceerd in peer-reviewed wetenschappelijke tijdschriften en worden gepresenteerd in de toonaangevende nationale en internationale bijeenkomsten in het veld. De resultaten zullen ook worden verspreid via jaarverslagen die op de websites van de registers worden gepubliceerd om clinici te bereiken die werken met mensen met artrose en diabetes.

om mensen met OA te bereiken en diabetes, willen we de verbinding tussen BOA-mensen die zorg voor OA zoeken, benutten. In het kort zullen we ons richten op BOA-opvoeders die hen materiaal (via e-mail en e-mail) verstrekken met betrekking tot de voortgang en realisatie van het project, gericht op de impact die het naast elkaar bestaan ​​van artrose en diabetes op de behandeling heeft. Op deze manier zullen we de nieuwe patiënten bereiken die deelnemen aan BOA die beter geïnformeerd zijn over hun toestand en over de te ondernemen strategie om het voordeel van eerstelijnsinterventie te maximaliseren.

Tot slot erkennen we in SOAD het belang van het bereiken van een breed publiek en om deze reden zullen we de Twitter- en Facebook-netwerken van de auteurs gebruiken om het grote publiek bewust te maken van het belang van het probleem. De sociale media zullen ons helpen om de wetenschappelijke praktijk gemakkelijk toegankelijk en begrijpelijk te maken voor een publiek van niet-specialisten.

Discussie

Dit onderzoekscohort biedt unieke inzichten in de relatie tussen diabetes en artrose. Door gegevens uit de BOA-, SHAR-, SKAR- en NDR-registers te gebruiken, kunnen we de invloed van diabetes op de uitkomst van niet-chirurgische en chirurgische OA-interventies onderzoeken, evenals op het effect van OA-behandelingen op diabetescontrole. p>

Voor zover wij weten, zal dit de grootste dataset zijn die gegevens over OA en diabetesmanagement combineert. Vanwege de grote steekproef op nationaal niveau zullen de resultaten van dit onderzoek waarschijnlijk een hoge externe validiteit en generaliseerbaarheid hebben. In de nationale registers verzamelde behandelingsgegevens zullen echter waarschijnlijk worden beïnvloed door de regionale verschillen in behandelingsprotocollen en gegevensverzameling die verschillende klinische omgevingen kenmerken in vergelijking met bijvoorbeeld zeer gestandaardiseerde klinische onderzoeken.

Concluderend is het voor het optimaliseren van behandelingen voor artrose en diabetes en voor een gepersonaliseerde aanpak belangrijk om factoren en comorbiditeiten te identificeren die de uitkomst van de interventies negatief kunnen beïnvloeden. Het naast elkaar bestaan ​​van verschillende aandoeningen creëert een complexere ziektestatus die aanvullende overwegingen vereist en zorgt voor de patiënt om het gewenste voordeel van de geboden interventies te ervaren. Het SOAD-cohort zal ons helpen deze patiënten met complexe behoeften te identificeren en een locatie te openen voor de ontwikkeling van betere behandelmethoden. Uiteindelijk kan het cohort invloed hebben op de manier waarop OA wordt beheerd wanneer andere comorbiditeiten naast elkaar bestaan, waardoor de enorme last van deze ziekte mogelijk wordt verminderd.

div> Abstract bekijken

      Div>

           

Machtigingen aanvragen

          

Als u een of meer van dit artikel opnieuw wilt gebruiken, gebruikt u de onderstaande link om naar de RightsLink-service van het Copyright Clearance Center te gaan. U krijgt een snelle prijs en directe toestemming om de inhoud op veel verschillende manieren opnieuw te gebruiken.

      Div>

           

Copyrightinformatie:

          

    

© Auteur (s) (of hun werkgever (s)) 2019. Hergebruik toegestaan ​​onder CC BY. Gepubliceerd door BMJ. Dit is een open access artikel gedistribueerd in overeenstemming met de Creative Commons Attribution 4.0 Unported (CC BY 4.0) licentie, waarmee anderen kunnen kopiëren, herdistribueren, remixen, transformeren en voortbouwen op dit werk voor welk doel dan ook, op voorwaarde dat het originele werk correct wordt geciteerd, een link naar de licentie wordt gegeven en wordt aangegeven of er wijzigingen zijn aangebracht. Zie: https://creativecommons.org/licenses/by/4.0/ . span> span> div> div> div> div>       Div> Div>    Div> Div>

      Div>

           

Lees de volledige tekst of download de PDF:

          

    

          

                        

      

        

               

    

  

    

               

    

Log in met uw gebruikersnaam en wachtwoord

      Div> Div>    Div> Div>    Div>       Div> Div>        Div>      Div>       Div>         Div>    Div>       Div> Div>        Div>             Div>                 Div>            Div>    Div> Div>    Div>            Div>          Div>                          Div>      Div>    Section> Div>
Lees Meer