Characteristics and Effectiveness of Interventions That Target the Reporting, Communication, or Clinical Interpretation of Lumbar Imaging Findings: A Systematic Review

J.L. Witherow; H.J. Jenkins; J.M. Elliott; G.H. Ip; C.G. Maher; J.S. Magnussen; M.J. Hancock

doi:10.3174/ajnr.A7432

Article Figures & Data

Tables

View popup

Table 1:

Summary of included studies for aims 1 and 2

Study	Imaging Technique	Intervention vs Comparator	Study Inclusion
Study	Imaging Technique	Intervention vs Comparator	Aim 1 Intervention Characteristics	Aim 2 Intervention Effectiveness
1) Ash et al, 2008¹⁹	MR imaging	Patients and physicians were blinded to imaging results vs standard care	Yes	Yes
2) Karran et al, 2018²²	CT or MR imaging	Educational intervention vs standard spinal clinic consultation	Yes	Yes
3) Jarvik et al, 2020²¹ Marcum et al, 2021²⁵Suri et al, 2021²⁶	X-ray, CT, or MR imaging	Prevalence information in imaging report vs standard report	Yes	Yes
4) Rajasekaran et al, 2021²⁷	MR imaging	Patients reassured imaging findings were normal vs factual explanation of imaging findings	Yes	Yes
5) Weeks et al, 2020²⁴	MR imaging	Prevalence information in imaging report vs standard report	Yes	Yes
6) Fried et al, 2018²⁰	MR imaging	Prevalence information in imaging report vs standard report	Yes	No
7) McCullough et al, 2012²³	MR imaging	Prevalence information in imaging report vs standard report	Yes	No

View popup

Table 2:

Summary of findings for inserting prevalence information into imaging reports versus standard reports^{^a}

Study/No. of Participants	GRADE Rating	Follow-Up Period	OutcomeHealth Care Use or Cost, Outcome Measure	EffectOR, 95% CI, or Rate Ratio (P Value)
Jarvik et al, 2020²¹N = 238,886^{^c}	Moderate	Short-term	Written opioid prescription	OR = 0.95 (0.90–0.99)^{^b}
Marcum et al, 2021²⁵N = 170,680^{^c}	Moderate	Short-term	New prescription for nonopioid pain-related medications^{^d}	OR = 1.02 (0.97–1.08)^{^b}
Jarvik et al, 2020²¹N = 238,886^{^c}	Moderate	Long-term	Written opioid prescription	OR = 0.95 (0.91–1.00)^{^b}
Suri et al, 2021²⁶N = 238,886^{^c}	Moderate	Long-term	Nonsurgical procedures^{^e}Any spine surgery	OR = 1.01 (0.93–1.09)^{^b}OR = 0.99 (0.91–1.07)^{^b}
Weeks et al, 2020²⁴N = 6904	Very low	Long-term	Primary care visitsChiropractic care visitsPhysical therapy care visitsSpecialty care visitsNerve-conduction testsMR imaging testingNon-MR imagingFacet injectionOn an opioidOn a muscle relaxantNonfusion spine surgery	Rate ratio = 0.86 P = NS^{^f}Rate ratio = 1.37 P = .05^{^f}
				Rate ratio = 1.19 P = NS^{^f}
				Rate ratio = 0.95 P = NS^{^f}
				Rate ratio = 0.57 P = .05^{^f}
				Rate ratio = 0.89 P = NS^{^f}
				Rate ratio = 0.73 P = .04^{^f}
				Rate ratio = 0.71 P = .02^{^f}
				Rate ratio = 0.98 P = NS^{^f}
				Rate ratio = 0.82 P = NS^{^f}
				Rate ratio = 0.71 P = NS^{^f}
			Fusion spine surgery	Rate ratio = 0.76 P = NS^{^f}
Weeks et al, 2020²⁴N = 6904	Very low	Long-term	Cost, total spine-related per member per month expenditures	Rate ratio = 0.85 P = NS^{^f}

Note:—NS indicates not significant.
↵a Dichotomous outcomes are shown. A rate ratio of <1 represents an effect in favor of the intervention group.
↵b Adjusted for health system, clinic size, age range, sex, imaging technique, Charlson Comorbidity Index category, seasonality, and health-specific trends. Results of opioid prescription additionally adjusted for prior opioid use. Results of nonsurgical procedures additionally adjusted for nonsurgical use in the year preceding index imaging.
↵c Articles reporting outcomes from the same study.
↵d Nonopioid, pain-related medications including skeletal muscle relaxants, NSAIDs, gabapentenoids, tricyclic antidepressants, and benzodiazepines.
↵e Procedures include lumbosacral epidural steroid injection, facet joint injection, facet join radiofrequency ablation, or sacroiliac injection.
↵f Adjusted for age, sex, line of business, deductible, and forecasted risk score at the time of first MR imaging.

View popup

Table 3:

Summary of findings for withholding MR imaging results for 6 months versus results received within 48 hours^{^a}

Follow-Up Period/Outcome	Outcome Measure (Scale)	Effect Size^{^b}Mean Difference (95% CI)
Short-term
Pain	VAS (0–10)	−0.54 (−1.35−0.27)
Disability	RMDQ (0–24)	−1.00 (−2.63−0.63)
Absenteeism	Mean No. sick days	0.20 (−0.47−0.87)
Quality of life	SF−36 (0–100)	PF: −7.00 (−13.99 to −0.01); RP: −9.70 (−21.74−2.34)BP: −8.50 (−15.51 to −1.49); GH: −3.10 (−8.35−2.15)VT: −4.80 (−10.97−51.37); SF: −0.70 (−7.00−5.60)RE: −7.30 (−17.29−2.69); MH: −9.30 (−15.44 to −3.16)
Fear of movement	FABQ: PA (0–24)	−0.40 (−2.29−1.49)
Fear of movement	FABQ: W (0–42)	−1.30 (−4.57−1.97)
Long-term
Pain	VPAS (0–10)	−0.10 (−1.0−0.80)
Disability	RMDQ (0–24)	−0.70 (−2.54−1.14)
Absenteeism	Mean No. sick days	0.07 (−0.74−0.88)
Quality of life	SF −36 (0–100)	PF: −0.70 (−9.30−7.90); RP: −1.00 (−14.68−12.68)BP: −0.60 (−9.82−8.62); GH: −5.00 (−12.57−2.57)VT: −3.50 (−11.09−4.09); SF: −0.70 ( −7.00−5.60)RE: −1.60 (−12.26−9.06); MH:–6.70 (−13.15 to −0.25)
Fear of movement	FABQ: PA (0–24)	0.60 (−1.74−2.94)
	FABQ: W (0–42)	−0.50 (−3.04−4.04)

Note:—FABQ indicates Fear-Avoidance Beliefs Questionnaire; PA, physical activity; W, work; RMDQ, Roland Morris Disability Questionnaire; SF, short form−36; PF, physical functioning; RP, role-physical; BP, bodily pain; GH, general health; VT, vitality; RE, role-emotional; MH, mental health; VAS, visual analog scale.
↵a Ash et al, 2008,¹⁹, n= 246, GRADE = very low. Negative values of the mean difference estimate represent an effect in favor of the intervention group.
↵b Only mean difference effects are presented.

View popup

Table 4:

Summary of findings for an educational intervention versus standard, spinal clinic consultation^{^a}

Follow-Up Period/Outcome	Outcome Measure (Scale)	Effect Size, Mean Difference (95% CI)
Intermediate
Pain	NRS (0–10)	1.20 (−1.00−3.40)
Disability	NRS (0–10)	1.40 (−1.46−4.26)
Fear of movement	TSK−11 (11–44)	6.70 (−2.12−15.52)

Note:—NRS indicates numeric rating scale; TSK-11, Tampa Scale of Kinesiophobia−11.
↵a Karran et al, 2018,²²n = 31. GRADE = very low. Negative values of the mean difference estimate represent an effect in favor of the intervention group.

View popup

Table 5:

Summary of findings for an intervention involving reassurance that MR imaging findings are normal versus a factual explanation of MR imaging report^{^a}

Follow-Up Period/Outcome	Outcome Measure (Scale)	Effect Size, Mean Difference (95% CI)
Short
Pain	VAS (0–10)	3.76 (−4.55 to −2.97)
Pain self-efficacy	PSEQ–2 (0–12)	4.68 (−5.62 to −3.74)
Quality of life	SF-12 (physical) (0–100)	8.46 (−13.12 to −3.80)
	SF-12 (mental) (0–100)	10.48 (−14.76 to −6.20)

Note:—VAS indicates visual analog scale; PSEQ –2, Pain Self-Efficacy Questionnaire–2; SF-12, 12-item Short Form Health survey (physical and mental dimensions).
↵a Rajasekaran et al, 2021,²⁷ n = 44. GRADE = low. Negative values of the mean difference estimate represent an effect in favor of the intervention group.