Transcript presentation ( format)

Updates in College Health: A
Review of the Literature
ACHA National Conference
Philadelphia, Pennsylvania
June, 2010
Objectives
• Discuss newly published important research
studies and their relevance to clinical practice
• Understand common research study designs
• Demonstrate evidence based medicine and its
application in College Health
Team
• Cheryl Flynn, MD, MS, MA
– Interim Medical Director, Syracuse University
– Family Medicine; epidemiology; family therapy
• David Reitman, MD, MBA
– University Physician, George Washington University
– Pediatrics and Adolescent Medicine
• Samuel Seward, MD
– Assistant Vice President, Columbia University
– Internal Medicine and Pediatrics
• Sarah Van Orman, MD
– Executive Director, University of Wisconsin-Madison
– Internal Medicine and Pediatrics
Process Overview
• Team members conducted literature review of
studies published during past 24 months
• Key search words:
– patient population-adolescent, college student,
university, young adult
• Avoid redundancy of topics presented in 2008
& 2009 Updates
Steroids for reducing throat pain
Hayward et al. Corticosteroids for pain relief in
sore throat: systematic review and metaanalysis. BMJ 2009; 339: b2976
Background and Question
• Sore throat common problem in primary care and
college health
– Most viral; ~10% Group A Strep
– SU experience: 8.7% of provider visits
• Question: Are systemic corticosteroids effective in
reducing symptoms of sore throat?
Study Design
• Systematic review with meta-analysis
– Only included placebo controlled randomized
controlled trials (RCTs)
– Mathematically combined data where possible
• Performed sensitivity analyses to assess robustness of
findings
Study Methods
• Population
– ambulatory setting only (ED or primary care)
– adults or children with acute tonsillitis/pharyngitis or
clinical syndrome of “sore throat”
– excluded studies of infectious mono, post-tonsillectomy or
intubation, or peri-tonsillar abscess
• Intervention
– systemic corticosteroids vs placebo
– (many concurrently received antibiotics &/or
acetaminophen)
Results
• 8 RCTs met inclusion criteria
– Population
• 743 patients, nearly balanced between adults/children
• 47% exudative ST; 44% Strep positive
– Intervention
• Betamethasone IM, dexamethasone IM or PO, prednisone PO
• All doses fairly equivalent; ~60mg PO prednisone
– Quality of included studies
• High; all with adequately concealed allocation
Results—quantitative (meta-analysis)
• Complete pain relief
– At one day: RR 3.16; NNT = 3.7
– At two days: RR 1.65; NNT = 3.3
• Mean time to onset pain relief
– Steroid group 6.3 hr earlier (p<0.001)
• Sensitivity analyses found no changes in
results
– Adult vs child; PO vs IM; Strep vs viral; exudative
vs not
Results--Qualitative
• Adverse effects (reported in only 1 trial)
– 5 hospitalized for IVF (3 steroid, 2 placebo)
– 3 developed peri-tonsillar abscess (1 steroid, 2 placebo)
• No difference or trend favoring steroids in
– Time to complete resolution of pain
– Time missed work/school
– Recurrent symptoms
Conclusion
• Addition of systemic corticosteroids
significantly reduces pain in patients with sore
throat
Limitations
• Possible confounding of antibiotic use
– Don’t know effect of steroids independent of antibiotics
• Relatively small number of RCTs
– Unable to assess publication bias
Clinical Bottom Line
• Consider adding steroids in patients with
severe sore throat in non-mono pharyngitis
– 60mg prednisone PO x 1 dose
LBP in Children & Adolescents
• Ahlqwist, A et al. Physical therapy treatment
of back complaints on children and
adolescents. Spine 2008; 33: E721-E727.
Background
• LBP is common in college health
• Risk factors:
– poor physical conditioning, intense exercise,
inadequate strength/impaired flexibility, family
history
• Question:
– How does individualized physical therapy compare
to a self-training program in adolescents with
lower back pain?
Study Methods
• Design
– Randomized controlled trial
– Concealed allocation; blinding not possible
• Setting
– Primary care
• Population
– 12-18 y.o., lumbar pain at least 2/10 on pain scale
– Excluded those w/serious physical or mental disease, or
those who had PT in prior month
– N = 45; baseline comparison between groups similar
Study Design
• Intervention
– Intervention group: individualized PT and exercise plus selftraining (PT 1x/wk, exercises 2x/wk)
– Control group: self-training only; 3x/wk
– Duration: 12 weeks
• Outcomes
– Measured using validated instruments perceived health,
disability, pain, flexibility/endurance
• Pre/post within groups
• Compared change scores between groups
Results
• Perceived health
(CHQ-CF)
– Both groups had
statistically significant
improvement in nearly
all sub-measures
pre/post
– No differences between
groups
• Disability (Roland & Morris
Disability Questionnaire)
– Both groups had
improvement pre/post
• PT -4.6; Control -2.7
– p = 0.016 between
groups
Results
• Pain (visual analogue scale 0-10)
– Drop in pain scores pre/post
• PT -3.6; Control -3.3
– No difference between
groups
• No difference in pain
duration or quality of pain
• Flexibility & muscle
endurance (back saver sit
and reach)
– Both groups had
improvement pre/post
– No differences between
groups
Conclusions
• Both groups improved on all parameters
measured
• Small additional benefit with addition of
physical therapy
– Perceived health status
– Disability ratings
Limitations
• Attribution error
– Improvement of health attributed to time or
interventions?
– Benefits of PT could be attributed to increased
“medical attention”
• Small #s
– Lack power to find differences between groups
Clinical Bottom Line
• The benefit of PT for adolescents with back
pain is modest at best
– If available, reasonable addition
– If not, most will improve anyway
Contraception and Weight
• Dinger et al. Oral Contraceptive effectiveness according to
body mass index, weight, age, and other factors. Am J Obstet
Gynecol 2009; 201: 263 e 1-9.
• Chi et al. Early weight gain predicting later weight gain
among depo medroxyprogesterone acetate users. Obst
Gynecol 2009; 114: 279-84
OCP effectiveness across BMI
• Research Question
– Are OCPs effective across varying BMIs?
• Design: Cohort
– Subset of prospective surveillance study
– Followed ~58K women Q6mo x 5 yr
– Contraceptive failure rate was an a priori
secondary outcome
Results
• Population
– 142,475 women years; avg duration follow-up 2.4 years
– Mean age 25.2; mean BMI 22.1; 20.4% first time OCP users
• Outcomes
– OCP failure rate 0.75% year 1  1.67% year 4
– NO DIFFERENCE in effectiveness across BMI range
• Limitations
– Lower than expected failure rates
– Did not enroll morbidly obese women
Predicting weight gain in DMPA users
• Research Question:
– Does early weight gain in depo-users predict
continued excessive weight gain?
• Design: Cohort
• 240 women 16-33 y.o. choosing depo followed Q 3-6
months for 3 yrs
• Depo-users divided into two categories
– Avg (<5% by 6mo) vs early wt gainers (>5% by 6mo)
– Predictors of excessive gain at 6 mo included
» past pregnancy (RR 2.2), BMI<30 (RR 4.0)
Results
12 mo
24 mo
36 mo
Avg (N=144)
0.63 kg
1.48 kg
2.49 kg
Early (N=51)
8.04 kg
10.86 kg
11.08 kg
• Adjusting for other factors, early gainers had 7.03 kg more wt
gain at 36mo vs avg group
• Limitations
– Small n; stats controlled for confounding
– Some who gained wt at 3 months dropped out
Clinical Bottom Lines
OCPs effectiveness/wt
Depo/wt gain
• OCPs are equally
• Significant weight gain
effective across
from depo use can be
weight/BMI spectrum in
predicted within the
women who are not
first two doses
morbidly obese
Treatment of
Irritable Bowel Syndrome (IBS)
• Ford AC, Talley NJ, Spiegel BMR, FoxxOrenstein AE, Schiller L, Quigley EM,
Moayyedi, P. Effect of fibre, antispasmotics,
and peppermint oil in the treatment of
irritable bowel syndrome: systemic review
and meta-analysis. BMJ, 2008; 337a:2313.
Background and Question
• Primary care providers frequently treat
irritable bowel syndrome (IBS)
• Many studies lack sufficient power to
demonstrate efficacy of treatments
• Conflicting outcomes in various studies
• What effect, if any, do fibre, antispasmodics or
peppermint oil have on the treatment of IBS
symptoms?
Study Methods
• Meta-Analysis of randomized controlled trials:
• Peppermint Oil (4 studies)
• Antispasmodics (22 studies)
• Fiber (12 studies)
• Primary, Secondary and Tertiary care settings
• Population-not specified
• Could not have other GI diagnosis
Study Design
• Treatment initiated
• Follow-up 1 wk – 60 months
• Needed to report
• Global assessment of cure
• Improvement of symptoms
• 35 studies met criteria
Results – Peppermint Oil
(4 studies, 293 Patients)
Results- Antispasmodics
(22 studies, 12 drugs, 1778 Patients)
Results – Fiber (12 Trials, 591 Pts)
Conclusions
• Fiber, antispasmodics (e.g. scopolamine) and
peppermint oil each more effective than
placebo in treating IBS
• NNTT
• Fiber 11
• Antispasmodics 5
• Peppermint Oil 2.5
Clinical Bottom Line
• Of three interventions studied, peppermint oil
shows the highest promise for efficacy in
treating IBS
Sleep Quality
• Lund H, Reider B, Whiting A, Prichard, J. Sleep
Patterns and Predictors of Disturbed Sleep in
a Large Population of College Students. J
Adol Health 46 (2010) 124-142
Background / Question
• Much data exists re: consequences of poor sleep in
children/younger adolescents
• Relatively little data in college age group
• NCHA Data
• 53% reported sleep problems
• 37% sleep had negative impact on academics
• In college population…..
• What are the predominant sleep habits?
• Can quality of sleep hygiene predict physical or behavioral
symptoms?
• What physical, emotional and psychosocial factors predict poor
sleep quality?
Study Methods and Design
• Cross-sectional Online Study
• Setting: Midwestern University
• Population:
•
•
•
•
College students, age 17-24
1125 participants
27% 1st years, 27% Sophomores, 24% Juniors, 20% Seniors
420 male, 705 Female
• Asked to complete 5 validated surveys to rate
– sleep quality, sleepiness, mood, distress, and diurnal
symptom variability
Results: Sleep Quality and Quantity
• Mean total sleep time 7.02 hrs
• 25% < 6.5 hrs
• 29.4% ≥ 8hrs
• Quality Sleep (PQSI)
– 34% “good”
– 38% “poor”
• Sleepiness (Eppworth Sleepiness Scale)
– 25% scored >10 (significant daytime sleepiness)
Results: Sleep Quality, Mood, & Health
• Poor Quality Sleepers:
– Higher levels of weekday stress (p<0.001)
– SUDS : 70.7 vs. 49.9
– Self reported negative moods (p<0.001)
– e.g. POMS Depression: 10.66 vs. 7.01
– More physical illnesses (p<0.05)
– 12% missed class in a month 3x+
– Increased use of Rx, OTCs and recreational drugs
to stay awake and to fall asleep >1x/month
Results:
Predictors of Poor Quality Sleep
• Stress
– Stress about school (39%)
– Emotional stress (25%)
• Excess noise (33%)
• Sleeping Partners (7%)
• Talking with friends prior to sleep (6%)
Conclusions
• “Epidemic” of insufficient and poor-quality
sleep in college students
• Perceived stress tends to predict poor
quality/ quantity of sleep
• Consequences of poor quality sleep include
higher stress, poorer moods, increased
physical symptoms, missed classes
Limitations
•
•
•
•
One-time, non-longitudinal survey
Students were from one university
Self-report
No mention of role of ETOH/Drug use
Clinical Bottom Line
• Clinicians need to proactively focus on both
the quality as well as the quantity of sleep in
patient history
• Poor Quality Sleepers increased risk of mood
disorders, substance abuse disorders and
somatic complaints
Douching and STIs
• Tsai CS, Shepherd BE, Vermund SH. Does
douching increase risk for sexually
transmitted infections? A prospective study
in high-risk adolescents. Amer J Obstetrics
and Gynecology. January 2009. 38e1-e8.
Douching and STIs
• Question: Is there an association between douching and
Trichomonas, Chlamydia, Gonorrhea and Herpes
• Design: Observational Prospective (Longitudinal) Cohort
• Results:
– Assessed time to STI in women who never, sometimes, or always douched
– Average age 16.9 yrs. 73% Black. 65% HIV infected
– “Always douched” had a shorter STI-free time than those who “never
douched.” (2:1)
• Commentary/Limitations
– High risk adolescents, slightly younger than college age. 2/3 HIV
– Couching = independent risk factor for STI acquisition
• Clinical bottom line:
• Clinicians should counsel female patients about potential STI risks
with frequent douching
Antidepressant Treatment
• Fournier JC, DeRubeis RJ, Hollon SD, Dimidjian
S, Amsterdam JD, Shelton RC, and J. Fawcett.
Antidepressant Drug Effects and Depression
Severity. JAMA. 2010; 303(1):47-53.
Background and Question
• Most studies for antidepressant effectiveness
in severely depressed patients
• Majority of patients treated with
antidepressants have mild to moderate
symptoms
Question: What is the relative benefit of
antidepressant medication vs. placebo across
a range of depression symptom severity?
Study Methods
• Meta-analysis of 6 randomized placebo
controlled trials
• Setting-outpatient
• Population
– Adults > age 18 yrs
– 5 Major depression; 1 minor depression
Study Design
• Intervention
– Treatment Range 6-11weeks
– 434 Antidepressant vs. 284 placebo
– 3 imipramine and 3 paroxetine
• Outcome
– Hamilton Depression Rating Scale (HDRS)
• Mild to moderate < 18
• Severe 19-22
• Very severe > 23
– HDRS 3-point difference-clinically significant
Results
• Baseline HDRS 10-39
• Threshold for effect initial HDRS > 25
• Medium effect HDRS > 25
• Large effect HDRS > 27
• Drop-out rates 9-34%
Conclusions
• Antidepressant drug effect varies as function
of disease severity
• Antidepressant drug effect appears to be
negligible with mild to moderate depression
• Antidepressant drug effect was large only for
very severe depression
Commentary/Limitations
•
•
•
•
Few patients with HDRS scores < 13
Considered only acute treatment
Did not include newer antidepressants
Increased effectiveness of anti-depressant vs.
reduced effectiveness of placebo
Clinical Bottom Line
Effect of antidepressant therapy on mild to
moderate depression unclear
Hair Shedding
Kunz M, Seifert B, and RM Trueb. Seasonality of Hair Shedding
in Healthy Women Complaining of Hair Loss. Dermatology.
2009;219:105-110.
Hair Shedding
• Question: Is hair shedding seasonal?
• Study Design: Retrospective Case Study
• Results/Conclusions
– Assessed telogen rates
– 823 women, aged 18-78
– 79% with female pattern hair loss (FPHL)
– Telogen rates showed annual periodicity
with peak in July and April
Hair Shedding
• Commentary
– Most marked in women with FPHL
–Student population may have higher
rates of telogen effluvium
• Clinical Bottom Line: Seasonal variation
may be important to consider for patient
counseling and for response to treatment
Value of Family History
• Wilson B, Quresh N, Santaguida P, Little J,
Carroll J, Allanson J, and P Raina. Systematic
Review: Family History in Risk Assessment for
Common Diseases. Annals of Internal
Medicine. 151(12):878-887.
Background
• Family history associated with risk for many
common diseases
• Knowledge of family history may motivate
behavior change
• Collecting family history is associated with
risks and benefits
• Collecting family history requires clinician time
Questions
#1 Improved Health
What is the direct evidence that getting a family history will improve
health outcomes for the patient or family?”
#2 Harm
What is the direct evidence that getting a family history will result in
adverse outcomes for the patient or family?
#3 Key Elements
What are the key elements of a family history in a primary care
setting for the purposes of risk assessment for common diseases?
#4 Accuracy
What is the accuracy of family history, and under what conditions
does the accuracy vary?
Study Methods
• Systematic Review, 1995-2009
– 137studies met criteria
– 69 reviewed
• Unable to perform meta-analysis
Results #1: Improved Health
• Studies
– 2 uncontrolled studies, high-study bias
• Outcomes
– No studies with direct health effects
– Increased uptake in breast cancer screening only
Results #2-Harm
• Studies
– 1 randomized controlled; 2 uncontrolled studies
– 2 generic family history; 1 cancer risk
• Outcomes
– 1 study found short-term increase in anxiety, gone at 3
months
– No long-term adverse effect found
Results #3-Key Elements
• Studies
– 20 longitudinal, 21 cross-sectional studies
– Cancer, coronary heart disease, stroke, diabetes
– 40 definitions of positive history
• Outcomes
– Sensitivity greatest
• Parents or other 1st degree relatives
– Specificity greatest
• Relative identified
• > 1 relative required
• Age of onset
Results #4-Accuracy
• Studies
– Specialized disease clinics
– 23-Longitudinal, Case-control, and Case series
– Patient report vs. relative’s medical records
• Outcomes
–
–
–
–
Informants disease status did not affect accuracy
Less accurate for 2nd and 3rd degree relatives
Widely varying sensitivity and specificity
Specificity (absence of disease) better than
sensitivity (presence of history)
Conclusions
• No evidence that family history leads to
improved health
– Insufficient evidence of changed health behaviors
•
•
•
•
No definitive evidence of lack of harm
Best method in primary care unclear
Best for 1st degree relatives
Accuracy often low, better for absence of
disease
Commentary/Limitations
• Few well done studies
• Most included patients with the condition of
interest leading to selection bias
• Many studies not done in primary care
settings
• Better studies are needed
Clinical Bottom Line
Family history collection
is considered to be a
standard of good care, but
value is unknown. If
collected, practitioners
should focus on 1stdegree
relatives.
Dexamethasone for Migraine
• Singh A, Alter H, Zaia B. Does the Addition of
Dexamethasone to Standard Therapy for
Acute Migraine Headache Decrease the
Incidence of Recurrent Headache for Patients
Treated in the Emergency Department? A
Meta-analysis and Systematic Review of the
Literature. Acad Emerg Med. Dec, 2008.
Background and Question
• Migraine is common
• Recurrent migraine after abortive therapy is
common:
– Up to 2/3 patients treated in ED within 48 hours
• Does addition of Decadron to standard
therapy decrease incidence of recurrent
migraine?
Study Methods
• Meta-analysis of RCT’s
• Inclusion criteria:
– Double blind
– Acute Migraine Dx
– Emergency Department (ED)
– Presence of control group
– Adequate follow-up
• 7 studies fulfilled criteria (n=742)
Study Design
• Intervention: “standard therapy” + Decadron
(IV or PO)
• Outcome: moderate or severe migraine at 24
to 72 hours
Results
• Modest but statistically significant benefit to
adjunctive Rx with Decadron
– ARR = 9.7%
– RR = 0.87
– 95% CI = 0.80 to 0.95
• Adverse side effects:
– 26% of Decadron pts
– 23% placebo pts
Conclusions
• Decadron, in this analysis, shown to decrease
rate of moderate or severe headache 24-72
hours after initial ED Rx
Commentary/Limitations
• “Standard Treatment” defined broadly—and,
in some studies, arguably unusually
• ED setting
• Not a large n
• Dose and route of Decadron
Clinical Bottom Line
• Decadron may have some value in abortive Rx
of acute migraine
X-rays and Harm
• Fazel R, Krumholz H, Yongei Wang SM, et al.
Exposure to Low-Dose Ionizing Radiation
from Medical Imaging. NEJM. August, 2009.
Background
• Patients, over their lifetimes, are getting
increasing #’s of studies with radiation
• Ongoing concern about link between lowdose radiation and
– solid tumors
– leukemia
• Patients often unaware of potential risk
• Not all imaging procedures evidence-based
Background
Growth in Use of Advanced Imaging under Medicare, 1995–2005
Background
Imaging Procedure
Plain Film
Average Effective Radiation Dose
0.01 – 10 mSv
CT
2 – 20 mSv
Nuclear
0.3 – 20 mSv
Interventional
5 – 70 mSv
Study Methods
• Retrospective cohort study
• United Healthcare enrollee records
Study Design
• January 1, 2005 to December 31, 2007
• Ages 18 – 64 yo
• 5 sites
– Arizona, Dallas, Orlando, South Florida, Wisconsin
• CPT codes
• Standard definitions of “Effective Dosing” for
radiation
Results
• 1M total patients
– 68.8% at least one imaging procedure
– 655,613 patients
• Higher in older age groups:
– 85.9% of 60-64 years
– 49.5% 18-34 years
• Higher by gender:
– Women: 78.7% and Men: 57.9%
– Mean: 1.2 +/- 1.8 procedures/patient/year
– Median: 0.7 procedures/patient/year
Results
Distribution of Annual Effective Doses of Radiation Stratified by Gender
Conclusions
• Nearly 70% study population underwent ≥1
imaging procedure during 3-year study period
• Gender-specific findings bear further study
Commentary/Limitations
• Imagining procedures are a source of radiation
exposure in the United States
– can result, over time, in high cumulative effective
doses
– Young people included
• Challenge = balancing immediate clinical need
with LT dose effect
• Selection bias
• Claims data
Clinical Bottom Line
• More is being discovered about imaging
practices in the U.S. and their potential
negative relation to long-term health effects.
• Primum non nocere
Radiologic Work-Up for Acute Abd Pain—What
Helps Nail the Dx?
• Lameris W, van Randen A, van Es HW, et al.
Imaging strategies for detection of urgent
conditions in patients with acute abdominal
pain: diagnostic accuracy study. BMJ. March
2009.
Background
• Abdominal pain is common:
– 5-10% ED visits
– Columbia AY 08-09: 830 cases
• CT and U/S have both been shown to
– Positively effect diagnostic accuracy
– Impact management decisions
• Both costly
• CT = radiation exposure
Question
• What is the optimal imaging strategy for
accurate diagnosis of urgent conditions
related to acute abdominal pain?
Study Methods
• Prospective, paired diagnostic accuracy study
• 6 academic medical centers
– Adults ≥ 18yo with acute non-traumatic
abdominal pain
– ED setting
– Exclude:
•
•
•
•
Pregnancy
Shock
Ruptured AAA
Patients for whom no imaging indicated
Study Design
• Diagnostic Strategies:
1.
2.
3.
4.
5.
Diagnosis following clinical evaluation (CE)
CE + plain films
CE + U/S
CE + CT
CE + U/S + CT (if U/S negative or inconclusive)
Results
• 1021 patients
– ED/Urgent Care settings
– Mean age: 47 years
– 55% female
– Ethnicity/Race not specified
• 66% of patients hospitalized following ED
evaluation
• 47% required surgical procedure
Results
Final Diagnoses: URGENT
Diagnosis
No (%)
Appendicitis
284 (28)
Cholecystitis
52 (5)
Gynecological
27 (3)
Urological
22 (2)
Pneumonia
11 (1)
Total
661 (65)
Results
Final Diagnoses: NON-URGENT
Diagnosis
No (%)
Non-specific Abdominal Pain
183 (18)
IBD
30 (3)
Gynecological
9 (1)
Total
360 (35)
Conclusions
Imaging Strategy
Sensitivit
y
Specificity
False
Negatives
False
Positives
1) Clinical Exam
88
41
12
27
2) CE + Plain Films 88
43
12
26
3) CE + U/S
70
85
30
11
4) CE + CT
89
77
11
12
6
16
5)values
CE + in
U/S
+/- CT (95%
94 confidence68
All
percentages
intervals)
 Strategy 5 approximately ½ total number of CTs
completed in strategy 4
Commentary/Limitations
• More than one way to peel a banana
• Stepped approach to w/u may or may not be
practical depending on patient/sx severity
– And, if time allows, U/S before CT has merits
• Non-randomized
• Most patients referred to ED—selection bias
Clinical Bottom Line
• As a single imaging strategy, CT is overall
better than U/S for urgent conditions
• A conditional strategy with CT reserved for
-/inconclusive U/S provides:
– highest sensitivity
– reduced population-based radiation exposure