Can Specific Physician Orders about Pain Medications Improve Pain Management in LTC?

1
Can Specific Physician Orders about Pain
Medications Improve Pain Management in LTC?

• Ralf Habermann, MD, CMD (Co-PI)
• Sumi Misra, MD, MPH, CMD (PI)
• Tracy Porchak, Project Coordinator
• Jamie Spicer, GNP
• Vanderbilt University

2
Faculty Disclosures

• Dr. Habermann and Dr. Misra have disclosed that
  they have no relevant financial relationships.

3
Acknowledgements

• Vanderbilt University, School of Medicine,
  Center for Quality Aging
• John F. Schnelle, PhD
• Sandra F. Simmons, PhD
• AMDA/Pfizer Quality Improvement Award

4
Background

• Pain is prevalent in LTC
• Pain assessment is required as a 5th vital sign
• Pain is often undetected and untreated, even
  though most (gt90) LTC residents are able to
  reliably answer pain questions
• LTC Residents are often not given the choice of
  PRN pain medication
• References
• 1. Chu L. et.al. JAGS 2004, 522057-2061.
• 2. Cadogan M et.al. J GeronMed Sci, 2004,
  59281-285.

5
Background

• Specific physician orders changed nursing home
  staff behavior for supplement delivery
• - Between meals instead of w/ meals
• May be able to impact other aspects of care
• References
• Simmons et.al. 2006 JAGS 54(9)1372-1376.
• Whiteman et.al. JNHA (in press).

6
Purpose

• To determine if specific physician orders about
  pain medications improve pain management in LTC
• How can Physicians improve daily care Quality

7
New Physician Order

• Ask resident during medication pass
• Do you have pain now?
• If resident responds yes, ask Would you like
  some medication for it?
• If resident responds yes, offer choice of PRN
  pain medication (if necessary in addition to
  scheduled pain medication)
• Note Order was implemented for all med pass
  delivery periods but night pass was not observed
  by research staff

8
Methods

• One nursing home
• Study Inclusion Criteria
• Able to self consent (as per IRB)
• Order for Pain Medication (Scheduled/PRN)
• 53/104 eligible (51 Consent Rate)

9
Demographics (n53)

• Average Age 83 years ( 9)
• 68 Female
• 70 Caucasian
• Average Length of Stay 1.9 years ( 3.0)
• 72 Long-Term Care (remainder sub-acute)
• Chart Order for Pain Medications
• 63 Order for Scheduled
• 96 Order for PRN

10
Most Common Pain-Related Diagnoses

1. Arthritis 49
2. Stroke 36
3. Osteoporosis 34
4. Cancer 28

Other Common Diagnoses

• Depression 51
• Dementia 36

11
Methods Data Collection

• 2 Days (morning, afternoon, evening) med pass
• Total of 6 observations per person baseline/post
• Same Days as Observation
• - Resident Interviews about pain
• - Chart Review (pain medications given)

12
Methods Timeline

• Baseline 2 observation days week one
• New Physician Order immediately after baseline
• Post 2 observation days
• - Day 1 immediately after order
• - Day 2 one week after order
• Follow-up (in progress) one month after order
  d/cd

13
Results Observations
Medication Passes 2 days pp (morning, afternoon, evening) Baseline Post
Did nurse ask if resident had pain? 20 67
Did resident say they had pain? 12 23
plt0.05 Note Numbers reflect percent of
observations that question was asked or response
was yes.
14
Results Observations
Medication Passes 2 days pp (morning, afternoon, evening) Baseline Post
If yes to pain, did nurse ask if resident wanted medication? 12 37
Did resident want pain medication? 11 20
plt0.05 Note Numbers reflect percent of
observations that question was asked or response
was yes.
15
Results Chart Review
Pain Medication Delivery Per Resident / 2-days (6 passes) Baseline Post
Average Total PRN given per resident 0.53 0.84
Percent of residents with 1 or more PRNs given 37 51
Percent of residents who wanted received PRNs 93 (13/14) 90 (18/20)
16
Results Resident Interview
Interview Question Baseline Post
Do you tell the nurse about your pain? 69 77
Does the nursing staff ask you about your pain? 58 80
Do you prefer to take medication when you feel pain? 77 72
plt0.05 Note Numbers reflect percent of yes
responses.
17
Results Observation
Time nurse spent with each resident per medication pass Mode Range Minutes Seconds
Average Total Time 1min lt 1 to 10 min
Nurse asked if resident had pain 1min lt 1 to 10 min
Nurse asked if resident had pain offered pain med 2min lt 1 to 12 min
Note Total time spent with resident includes
care activities beyond med pass
18
Results Summary

• Nurses asked residents questions about pain
  during medication passes more frequently
  following the new order.
• Residents expressed pain more frequently in
  response to nurses questions.
• Residents asked for pain medication more
  frequently in response to nurses questions.

19
Results Summary

• It required one extra minute per resident/pass to
  ask questions and provide medication.
• Residents noticed the difference in nurse
  behavior.
• Most residents expressed a stable preference to
  take medication for pain.
• The number of PRN pain meds increased

20
Implications

• Specificity of physician orders may improve care
  in other areas
• Trial of toileting assistance for incontinence
• Walking assistance to/from dining room
• Offering residents choice during daily care
• CHF management i.e. diuretic dose depending on
  daily weight

21
Barriers to Improvement

• Inadequate staffing to provide care consistent
  with order
• Staff Resistance
• Variance in nurse pain assessment and treatment
• - Limits care to specific time of day or
  protocol
• Survey Compliance Risk
• - Documentation will show care was not provided
  as ordered

22
Study Limitations

• Only one nursing home site.
• Short time frame for evaluation (lt 1 month)
• Residents most at risk for undetected and
  untreated pain (more cognitively impaired) were
  not included even though these residents are able
  to respond to nurse questions
• ( due to IRB concerns)

23
Future

• Development of specific facility wide protocol
  order set
• Inclusion of all residents in this program
• Reduction of potential for nurses bias regarding
  pain medication

24
Conclusion

• YES
• Physician orders have an impact and can change
  institutional behavior