1 of 58

Claire E Wylie �BVM&S MSc PhD MRCVS��Rossdales Equine Hospital & Diagnostic Centre, Newmarket���claire.wylie@rossdales.com

Collaboration as a key feature of equine evidence-based research: a laminitis case study

​

2 of 58

Equine laminitis

​

​

​

​

​

​

​

​

​

​

​

​

3 of 58

Evidence-based research & epidemiology

Diagnosis

  • What clinical signs should I look for?
  • Should I take an x-ray?

​

Treatment

  • Will ‘bute’ be a sufficient painkiller?
  • Will ACP help prevent rotation?
  • Should I cool the legs?

Prevention

  • Should owner stop feeding?
  • Does the pony have PPID?

Epidemiology

  • Help to answer owner queries

​

  • “How common is laminitis?”
  • “Why has my pony got laminitis?”
  • “Will s/he get laminitis again?"
  • “What can I do to stop my other horses getting laminitis?

​

4 of 58

Questions of interest

  • Frequency
  • Risk factors

5 of 58

Evidence-base for laminitis

Systematic reviews of available evidence

Design of epidemiological tools

Veterinary-practice based cohort and nested case-control study

Publication of new evidence

Modification of epidemiological tools

Design of owner-based cohort study

Publication of new evidence

Adoption of evidence-based prevention and control

Evaluation of the delivery of, and welfare impact derived from, the guidelines

6 of 58

Evidence-base for laminitis

Systematic reviews of available evidence

Design of epidemiological tools

Veterinary-practice based cohort and nested case-control study

Publication of new evidence

Modification of epidemiological tools

Design of owner-based cohort study

Publication of new evidence

Adoption of evidence-based prevention and control

Evaluation of the delivery of, and welfare impact derived from, the guidelines

Methods to gather evidence-based data

Challenges and lessons we have learnt

Veterinary practitioner role in providing and using data

7 of 58

Key players

Epidemiologists

Charitable bodies

I.T. consultants

P.R. officers

Veterinary practitioners

Laminitis researchers

Horse owners

Practice administrators

Others

8 of 58

Step 1

9 of 58

Systematic reviews

Q: How common is laminitis?

  • Qs:
  • “Why has my pony got laminitis?”
  • “Will he get laminitis again?"
  • “What can I do to stop my other horses getting laminitis?

10 of 58

Key players

Epidemiologists

Practice administrators

Veterinary practitioners (Expert discussion)

Laminitis researchers

PhD student

11 of 58

Systematic reviews

  • 17 studies frequency
  • 69 studies risk factors
  • Subtle ‘collaboration’
  • In effect ~450 others involved

12 of 58

Systematic reviews

  • Rationale: Are there sufficient data already out there to answer my clinical questions?

​

  • “The two high quality publications regarding the general equine population suggested that laminitis affects between 1.5% and 23.8% of animals.”

​

  • “Many of the publications focussed on easily measurable factors, such as signalment or season.”

​

​

​

​

13 of 58

Step 1 to Step 2

14 of 58

Design of epidemiological tools

  • Frequency

​

  • Case definition
    • What is a laminitis episode

​

  • Numerator data

​

  • Denominator data

​

  • Risk factors

​

  • Signalment
  • Feeding
  • Stabling
  • Grazing
  • Farriery
  • Exercise
  • Health

15 of 58

Epidemiological tools: frequency

  • Numerator data:
  • Veterinary reporting
  • Active episodes of laminitis
  • Checklist of clinical signs
  • ‘Laminitis reporting form’

​

16 of 58

17 of 58

Epidemiological tools: frequency

  • Paper format preferred
  • One page of A4
  • Checklist options
  • No need for additional data
  • Option for ‘didn’t assess’
  • Anonymous reporting option
  • Already pre-identified
  • Reply-paid envelopes

18 of 58

Practice recruitment

  • Monthly denominator data
  • Where practice manager (+/- IT) became key player

​

 

Number of active laminitis cases reported in this month

Actual number of active laminitis cases attended by the practice – (please include any you may have forgotten to complete a laminitis reporting form for!)

Number of horses/ponies attended by the practice in this month*

*If you are unable to calculate the number of horses/ponies attended please state the number of equine clients attended by the practice in this month

Total number of horses/ponies registered with your practice**

**If you are unsure of the number of horses/ponies registered please state the total number of registered/active equine clients

July

1

 

 

 

 

 

Aug

0

 

 

 

 

 

19 of 58

Epidemiological tools: risk factors

20 of 58

21 of 58

22 of 58

Epidemiological tools: risk factors

  • Paper format
  • Logically structured, skip responses
  • Checklist options
  • As concise as possible!
  • Coloured paper
  • Professionally printed
  • Already pre-identified
  • Reply-paid envelope

23 of 58

Key players

Epidemiologists

Practice administrators

Veterinary practitioners (Expert discussion, LRF, conduit)

Laminitis researchers

Horse owners

PhD student

24 of 58

Step 2 to Step 3

25 of 58

Practice recruitment

  • Not feasible to collect data from every practice

​

  • Essential to recruit compliant practices rather than ‘targeting’ a truly random sample of equine practices

26 of 58

Practice recruitment: phase 1

    • 2008 Royal College of Veterinary Surgeons (RCVS) electronic practice directory

​

  • Each practice within GB registered as treating equine patients
    • equine-only or mixed-species

​

​

​

27 of 58

Practice recruitment: phase 1

  • All practices completing the questions included in a prize draw
  • Invited to respond by telephone, fax, email or post
  • Full contact details provided
  • Reply-paid envelopes not provided
  • Letters published in the veterinary press

​

28 of 58

Practice recruitment: phase 1

  • Response to the initial letter was low, despite having requested ‘no interest’ responses to be returned

​

  • No response: 85%

​

29 of 58

Practice recruitment: phase 1

  • Reasons for non-response may have included:
    • a lack of interest
    • poor timing
    • written communication
    • initial mailing addressed anonymously
    • reply-paid envelopes not provided

​

30 of 58

Practice recruitment: phase 2

  • Of those who registered an interest

​

  • Further letter with more detailed explanation of the requirements for each objective

​

  • Eight questions to determine suitability

31 of 58

Practice recruitment: phase 2

Question

Positive (n)

% 

Would you like to contribute data on laminitis cases for our study?

65

98.5%

How many vets are engaged in equine work at your practice?

Range 1-15

(mean 4, median 3)

Would all vets be happy to contribute data for each laminitis case they see?

57

87.7%

If not, how many would be willing to help with this study?

Range 40.0%-83.3%

Would we be able to receive owner contact details for laminitic cases?

62

95.4%

Would it be possible to send questionnaires to owners of non-laminitics as part of general practice mailings? (e.g. with invoices/practice newsletters)

57

87.7%

Would it be possible for your practice to provide us with information on the number of horses that are seen by your practice (denominator data)?

55

84.6%

Does your practice use computerised patient information systems that could provide this denominator data?

50

76.9%

32 of 58

Practice recruitment: phase 2

  • Selected 28 practices able to achieve all requirements

​

  • Geographically representative - cluster detection analysis
    • Moran’s I statistic
    • SaTScan statistic

​

​

​

33 of 58

Practice recruitment

  • Practitioners assisted with obtaining owner-derived case-control data
  • Several options were provided on the LRF to allow practitioners to distribute case questionnaires either directly, or through provision of owner contact details (with owners’ permission)

34 of 58

Practice recruitment

35 of 58

Key players

Epidemiologists

Practice administrators

Veterinary practitioners

Laminitis researchers

Horse owners

PhD student

36 of 58

Cohort study

Epidemiologists

Practice administrators

Veterinary practitioners

Laminitis researchers

Horse owners

PhD student

37 of 58

Cohort study

38 of 58

Cohort study

39 of 58

Cohort study

  • Most practices (81.3%) returning denominator data under-reported laminitis cases
  • Only three practices accurately reported the number of laminitis episodes for most of the study period
  • Varied from none (all cases reported by LRF) to an 11-fold under-reporting (12/132 cases reported) by one practice

40 of 58

Cohort study

Pink - monthly % agreement return LRF and ‘actual’ no laminitis cases

Blue - monthly return of denominator data expressed as a % of practices who returned denominator data from the total

41 of 58

Cohort study

  • Apparent overall decline in data returned
  • Possible reasons:
    • Felt already provided estimates at start of study
    • ‘Study fatigue’
    • Practices may have been less able to assist with external projects at certain times of year

42 of 58

Cohort study

  • Contacted at least x2/month
  • ‘Practice Update Newsletter’ initiated
  • ? increased workload at the end of winter
  • Avoid initiating equine practice-based epidemiological studies in these months

43 of 58

Nested case-control study

    • Epidemiologists

​

    • Practice administrators

​

    • Veterinary practitioners
    • Laminitis researchers
    • Horse owners
    • PhD student

44 of 58

Nested case-control study

  • Owner source of data
  • Laminitic horses
  • Non-laminitic controls

​

  • Sample size target:
    • 150 cases
    • 600 controls
  • Achieved:
    • 191 cases
    • 819 controls

​

45 of 58

Practice recruitment

46 of 58

Nested case-control study

  • Laminitis cases
  • 577 LRFs to 191 case Qs
    • Converting 33%

​

  • Owner declined
  • No incentives
  • Intermediate stage (vet) study design and data protection
    • Lost
    • Time delay

​

​

​

​

47 of 58

Practice recruitment

48 of 58

Nested case-control study

  • Non-laminitic controls
  • 4670 mailings to 819 Qs
    • Converting 17.5%

​

  • Cold calling
  • Lack of interest
  • No incentives
  • No follow-up
  • Intermediate stage (vet) study design and data protection
    • Lost
    • Time delay

​

​

​

​

49 of 58

Nested case-control study

  • Statistically significant difference (P<0.001) in return rates owners contacted directly by study (60.1%) and via practices (35.4%).

​

  • Questionnaires distributed directly to owners were accompanied by personalised covering letters which may have improved compliance.

​

50 of 58

Step 3 to Step 4

51 of 58

Published papers

52 of 58

Step 4 to Step 6

53 of 58

Owner based cohort study

Modified Laminitis Reporting Form

54 of 58

Owner based cohort study

    • Epidemiologists
    • Charitable bodies
    • I.T. consultants
    • P.R. officers
    • Veterinary practitioners
    • Laminitis researchers
    • Horse owners
    • Practice administrators

Others

55 of 58

CARE About Laminitis

56 of 58

CARE Challenges

  • Required 5000 horse-years-at-risk (e.g. following 3333 horses for 18 months on average)

57 of 58

Collaboration is key

Epidemiologists

Charitable bodies

I.T. consultants

P.R. officers

Veterinary practitioners

Laminitis researchers

Horse owners

Practice administrators

Others

58 of 58

Thank you for listening

​

​

​

    • Claire E Wylie

BVM&S, MSc, PhD, MRCVS

​

    • • claire.wylie@rossdales.com

​

    • ☎ 01638 577 754

​

​