Prognostic factors associated with changes in knee pain outcomes, identified from initial primary care consultation data. A systematic literature review

Figures & data

Table 1. Characteristics and Quality in Prognosis Studies (QUIPS) assessment for included studies.

Characteristics											QUIPS score per domain
	Participants, Setting, design and sample size	Outcomes		Prognostic factors			Missing data		Analyses		Participation	Attrition	Prognostic factors	Outcome measure	Adjustment for other prognostic factors	Statistical Reporting
	Description and study deign	Type and Number of outcomes	Duration of FU (months)	Number and type of prognostic factors	Timing of prognostic factor and outcome measurement	Handling of prognostic factors in the analysis	Number of participants missing data (%)	Handling of missing data	Modelling method	Sample size calculation
[24]	Practitioners: Not stated Participants: males (N = 134 and female (N = 571) recruited from the Cohort Hip and Cohort Knee study in the Netherlands. Design: Prospective Cohort Sample size: N = 705	Pain (NRS)	40	Demographics: Age, gender, BMI, ethnicity Health: medication, alcohol use, smoking history, vitamin or supplement use Medical history: asthma, chronic sinusitis, CVD, hypertension, gastric ulcer, gallstones, liver disease, diabetes, thyroid gland disease, epilepsy, cancer, severe skin disease, other MSK disease. Knee symptoms and signs: WOMAC pain, stiffness and physical functioning Physical examination: Pain in ipsilateral hip, morning stiffness <30 min, palpable warmth, joint space tenderness, bony enlargement, crepitus, positive refill test, pain during knee ROM, pain during hip ROM, Bouchard swelling, Heberden node knee pain, range of motion Radiographs: knee and hip Blood tests: ESR Psychosocial: Pain coping behaviour, education level	Prognostic Factors: Baseline Outcome measure: Every year up to 5 years	Not stated	38 (5)	Not specified	Univariate multinomial regression analyses followed by latent class growth analysis	No	Low	High	High	Low	High	High
[25]	Practitioners: 40 general practitioners Participants: males (N = 211) and females (N = 269) recruited from five municipalities in the Netherlands. Design: Prospective Cohort Sample size: N = 480	Persisting knee symptoms (Dichotomized into recovered symptoms and persistent symptoms)	12	Demographics: mean age, age >60 years, women, mean BMI, BMI >25 Health: Presence of comorbidity in skeletal system, presence of other comorbidities Patient characteristics: private insurance, paid employment >8 h/week, sport participation Knee symptoms and signs: pain level, duration of symptoms, history of non-traumatic knee symptoms, history of traumatic knee symptoms, presence of locked knee, bilateral symptoms, recurrent symptoms feeling of giving way, limited when walking stairs, WOMAC pain and stiffness and physical functioning Physical examination: warm, swollen, crepitus PROM, crepitus AROM, varus/ valgus alignment, PROM pain, AROM pain, anterior drawer, patella alignment assessment, joint swelling, pain on internal hip rotation, restriction of internal hip rotation, presence of: Heberden’s nodes; Bakers cysts; prepatellar bursitis, ITB pain. Psychosocial: education level, mean Kinesiophobia score, Kinesiophobia score >25′	Prognostic Factors: Baseline Outcome measure: 1 year except disability and pain; every 3 months over 1 year	Dichotomization of candidate factors with continuous data	69 (14)	Multiple imputation	Univariable logistic regression followed by multivariable logistic regression with backward variable selection	No	Low	Low	Mod	Mod	Mod	Mod
[26]	Practitioners: 40 general practitioners Participants: males (N = 277) and females (N = 272) were recruited from the research network HONEUR (part of a prospective observation cohort) in the Netherlands. Design: Prospective Cohort Sample size: N = 549	Unfavourable outcome (categorized as persistent knee symptoms at 6-year FU or having undergone knee replacement surgery during FU).	3,6,9,12 and 54	Demographics: age, sex, and comorbidity Health: Health-related quality of life, advice given by the GP, knee medication, visits to health care professionals and operations Activity and sports participation: Daily activities and physical exercise Knee symptoms and signs: History of previous knee injuries or operations, duration and recurrence of knee symptoms, knee pain level, WOMAC pain stiffness, and function domains. Physical examination: Knee alignment, joint effusion, palpation, temperature, collateral ligaments, a joint line tenderness, assessment of effusion, passive knee ROM in flexion and extension, meniscal tests and knee stability tests. Psychosocial: Composition of household, education level, coping, sick leave from daily activities and data on impact of the knee symptom as a hindrance in terms of daily activities.	Prognostic Factors: Baseline Outcome measures: 3,6,9,12 and 54 months	Dichotomization of candidate factors with continuous data	209 (38)	Multiple imputation	Univariable logistic regression followed by multivariable logistic regression with backward variable selection	No	Mod	High	Mod	Low	High	High
[27]	Practitioners: 40 General Practitioners Participants: Adolescents and adults with traumatic knee symptoms from general practice. Males (N = 90) and females (N = 82) were recruited from the research network HONEUR (part of a prospective observation cohort) in the Netherlands. Design: Prospective cohort Sample size: N = 172	Persistent knee symptoms (dichotomized into recovered symptoms and persistent symptoms)	3,6,9,12 and 54	Demographics: Age, gender, BMI >25 Health: Comorbidities of the skeletal system, other non-skeletal comorbidities, self-rated poor health, Sports: Sports hindrance and level of daily activities. Knee symptoms and signs: Duration >3 months, recurrent symptoms, bilateral symptoms, pain level, self-reported warmth of knee, self-reported knee swelling, self-reported knee crepitus, knee locking, knee instability, history of knee symptoms, WOMAC pain, stiffness and function domains Psychosocial: Education level, sick leave	Prognostic Factors: Baseline Outcome measures: 12 and 54 months	Dichotomization of candidate factors with continuous data	72 (41)	Multiple imputation	Univariable logistic regression followed by multivariable backward logistic regression	No	Mod	Mod	High	Mod	Mod	Mod
[28]	Practitioners: 40 General Practitioners Participants: Males (N = 185) and females (N = 143) were recruited from the research network HONEUR (part of a prospective observation cohort) in the Netherlands. Design: Prospective cohort Sample size: N = 328	Self-reported perceived recovery (dichotomized into recovered symptoms and persistent symptoms)	3,6,9,12 and	Demographics: Age, gender, BMI Health: Co-morbidities of skeletal system, other non-skeletal comorbidities, mental health, general health. Patient characteristics: employment type, sport participation. Knee symptoms and signs: injury history including trauma during sport, fall on the knee, weight bearing on knee, rotational trauma, foot/leg blocked, immediate pain at trauma, immediate effusion after trauma, popping sensation at trauma, whether continuation of activity was possible, past history of traumatic knee symptoms, history of non-traumatic knee symptoms,; symptoms at time of consultation including pain level, self-reported knee warmth, self-reported knee swelling, self-reported knee crepitus, locking of the knee, knee instability, limitations on work/study/ daily function. Physical examination: Patella alignment, pain during AROM flexion, pain during AROM extension, pain during PROM flexion, pain during PROM extension, laxity during valgus stress test 30 degrees, laxity during varus stress test 30 degrees, laxity during anterior drawer test, effusion of popliteal fossa, McMurray meniscal test, Apley grinding test, Apley traction test Psychosocial: Education level	Prognostic Factors: Baseline Outcome measures: 3, 6, 9, 12 and 54 months	Not stated	32 (10) lost at 1 year. 145 (44) lost at 6 years	Multiple imputation	Univariable logistic regression followed by backward logistic regression	No	Mod	Low	High	Mod	Mod	Mod
[29]	Practitioners: Not stated Participants: males (N = 283) and females (N = 338) were recruited from three practices from the North Staffordshire Primary Care Research Consortium, UK. Design: Prospective cohort Sample Size: N = 766	Functional outcome using WOMAC-PF scores.	18	Demographics: Age, gender Health: Alcohol consumption, smoking status, self-rated health, number of selected comorbid health conditions Patient characteristics: Occupation, marital status Knee symptoms and signs: Pain location, pain level, chronic pain grade, whole leg pain, widespread pain, number of days in pain (last 6 months) and pain episode duration. Psychosocial: Anxiety, depression, sociodemographic characteristics social network index	Prognostic Factors: Baseline Outcome measures: 18 months	Not stated	198 (26)	Not stated	Univariable cox regression followed by backwards stepwise cox regression with backwards variable selection	No	Low	High	Mod	Low	High	High
[30]	Practitioners: Not stated Participants: Three practices from the North Staffordshire Primary Care Research Consortium in the UK. Design: Prospective cohort Sample size: N = 621	Functional outcome using WOMAC-PF scores.	18	Demographics: Age, BMI Knee symptoms and signs: Knee pain severity, first-degree relative with arthritis, previous meniscectomy, contralateral total knee replacement, bilateral knee pain, duration of morning stiffness, inactivity gelling, self-reported swelling in past month, incident knee pain, giving way and locking. Physical examination: Clinical osteoarthritis of the hand, observed intermalleolar gap in standing, observed intercondylar gap in standing, severity of knee joint effusion, fixed flexion deformity of the knee, pain provocation on PFJ compression, observed bony enlargement, local tender point count, anteroposterior instability mediolateral instability, range of hip internal rotation, range of knee flexion, maximal isometric knee extensor strength, maximal isometric knee flexor strength, palpation of crepitus, timed single leg standing balance, hip rotation PROM, knee flexion PROM Radiography: Severity of knee changes Psychosocial: Anxiety	Prognostic Factors: Baseline Outcome measures: 18 months	Not stated	124 (19)	Not stated	Univariate cox regression followed by multivariate cox regression with backwards variable selection	No	Low	High	Mod	Low	High	High
[31]	Practitioners: 49 General Practitioners Participants: Males (N = 51) and females (N = 200) were recruited from 61 general practices (97 GPs) in the Netherlands. Design: Prospective cohort Sample size: N = 251	Perceived recovery (dichotomized into yes or no groups), Pain intensity measured as NRS. Functional outcomes measured through WOMAC	3 and 12	Demographics: Age, sex, BMI Patient characteristics: Work status, marital status, number of children (<5 years) in household Health: Smoking status, quality of life, perceived general health and vitality, comorbidities, physical activity, ACSM position stand, social support Knee symptoms and signs: Symptom duration, location, history, severity, perceived cause of the complaint, presence of menopause, use of pain medication Psychosocial: Education, pain coping strategies, distress, Kinesiophobia	Prognostic Factors: Baseline Outcome measures: 3 and 12 months	Dichotomization of candidate factors with continuous data Categorical (several coping strategies, distress, and the 2 Kinesiophobia subscales. Continuous variables: (Age, duration of the knee complaint, pain, WOMAC pain, stiffness and function, vitality and social support	3 months: 223 (92%) 12 months: 203 (80%).	Not stated	Perceived recovery: Cox regression with backward variable selection Pain and function: Univariable analysis followed by multiple linear regression using backward variable selection	No	Low	High	Mod	Mod	Mod	High

Key: Checklist for critical appraisal and data extraction for systematic reviews of prognostic factor studies (CHARMS-PF); Quality in prognosis studies (QUIPS); Follow-up (FU); body mass index (BMI); Cardiovascular disease (CVD); MSK; Range of movement (ROM); Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC); Western Ontario and McMaster Universities Osteoarthritis Index Physical Functioning (WOMAC-PF); Erythrocyte sedimentation rate (ESR); Active range of movement (AROM); Passive range of movement (PROM); Iliotibial band (ITB); Activities of daily living (ADL’s); General Practitioner (GP); American College of Sports Medicine (ACSM); Patellofemoral joint (PFJ).

QUIPS score per domain: low risk of bias; moderate risk of bias; high risk of bias.

Table 2. Summary of synthesis of significant prognostic factors, or prognostic factors that were investigated in two or more with associated GRADE evaluation.

Summary of synthesis of prognostic factors									Adapted GRADE criteria
Specific outcome	Number of studies	Potential prognostic factors	Authors	Effect Measure	Univariable effect size (95%CI)	p Value	Multivariable effect size (95%CI)	p Value	Consistency of results	Effect size	Precision of results	Publication bias	Overall Quality
		Three-month follow up
Pain (NRS)	2	Sex BMI > 30 Duration of the knee complaint Cause overload during unusual activities Baseline pain PCI Distress Middle vs. low Coexisting MSK complaints ACSM position stand recommendations	[31] [31] [31] [31] [31] [31] [31] [31]	RC RC RC RC RC RC RC RC	0.70 (−1.41 to 0.01 0.84 (−0.13 to 1.81) −0.20 (−0.42 to 0.02) −1.65 (−2.99 to −0.32) 0.59 (0.46–0.73) 0.38 (−0.50 to 1.27) −0.73 (2.10 to 0.64) 0.73 (−0.21 to 1.67)	<0.20 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20	−1.01 (−1.60 to 0.42)0.86 (0.06–1.67)−0.25 (−0.44 to 0.07)−1.09 (−2.19 to 0.02)0.65 (0.53–0.78) −1.66 (−3.06 to −0.26) −1.20 (−2.36 to −0.03) 0.77 (−0.01 to 1.55)	<0.001 0.004 0.01 0.05 <0.001 0.02 0.04 0.05	- - - - - - - -	Small Small Small Small Small Small Small Small	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely Likely Likely	Very low Very low Very low Very low Very low Low Very low Very low
		12-month follow up
		Non-traumatic knee complaint history Baseline pain present PCI distraction high vs. low PCI distress high vs. low Vitality	[31] [31] [31] [31] [31]	RC RC RC RC RC	−1.19 (−1.97 to −0.41) 0.61 (0.46–0.75) −1.09 (−2.05 to −0.14) −3.68 (−5.82 to 1.54) 0.02 (0.0–0.05)	<0.20 <0.20 <0.20 <0.20 <0.20	−1.31 (−1.94 to −0.67)0.69 (0.55– 0.82)−1.02 (−1.80 to 0.24)−2.03 (−3.93 to −0.12) 0.02 (0.00–004)	<0.001 <0.001 0.01 0.04 0.03	- - - - -	Small Small Small Small Small	Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely	Very low Very low Very low Low Very low
		54-month follow up
		Higher BMI Lower level of education Greater comorbidity Higher activity limitation scores Joint space tenderness	[24] [24] [24] [24] [24]	RR RR RR RR RR	- - - - -	- <0.05 <0.05 <0.05 <0.05	1.10 (1.01–1.20)0.55 (0.23–1.31)2.65 (1.25–6.99) 2.03 (0.96–4.33) 1.08 (0.51–2.29)	- - - - -	- - - - -	Small Small Moderate Small Small	Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely	Very low Low Low Very low Very low
		12-month follow up
Persisting knee symptoms	2	Age >60 years Education level Kinesiophobia Comorbidity of MSK system Non-traumatic knee history symptoms Bilateral symptoms >3-month symptom duration Crepitus of PROM extension	[25] [25] [27] [25] [25] [25] [25] [27] [25] [25]	OR OR OR OR OR OR OR OR OR OR	1.56 (1.08–2.24) Figure 2 Figure 2 1.99 (1.37–2.89) 0.83 (0.58– 1.19) 5.12 (2.97– 8.81) Figure 3 Figure 3 3.04 (1.98–4.65) 1.11 (0.75–1.66)	<0.20 <0.20 0.002 <0.20 <0.20 <0.20 <0.20 0.11 <0.20 <0.20	2.02 (1.30–3.13) Figure 2 Figure 2 1.85 (1.26– 2.72) 1.50 (0.99– 2.28) 4.30 (2.38–7.79) Figure 3 Figure 3 2.20 (0.9–5.6) 1.91 (1.01–3.63)	- - <0.001 - - - − 0.09 - -	- Yes Yes - - - Yes Yes - -	Small Small Large Small Small Moderate Moderate Moderate Small Small	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely	Very low Low Low Very low Very low Very low Low Low Very low Very low
		54-month follow up
Unfavourable outcome	1	Low/Middle education level Comorbidity skeletal system Poor mental Health (SF-36 score <50) >3-month symptom duration Bilateral knee symptoms Self-report warm knee History of non-traumatic knee symptoms Valgus Pain passive flexion Pain passive extension Bony enlargement of joint	[26] [26] [26] [26] [26] [26] [26] [26] [26] [26] [26]	OR OR OR OR OR OR OR OR OR OR OR	2.38 (1.47–3.85) 2.09 (1.34–3.27) 2.81 (1.16–6.83) 2.45 (1.51–3.98) 2.80 (1.73–4.51) 2.36 (1.61–3.68) 3.39 (2.03–5.65) 2.25 (1.38–3.67) 2.44 (1.51–3.94) 2.27 (1.40–3.71) 3.05 (1.38–6.72)	<0.01 <0.01 0.02 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.01	1.94 (1.18–3.19) 1.79 (1.12–2.87) 2.95 (1.16–7.48) 2.20 (1.27–3.78) 1.89 (1.11–3.19) 2.07 (1.27–3.36) 2.59 (1.52–4.41) 2.07 (1.24–3.48) 1.94 (1.17–3.21) 1.72 (1.01–2.92) 2.64 (1.17–5.96)	0.01 0.02 0.02 0.01 0.02 <0.01 <0.01 0.01 <0.01 0.05 0.02	- - - - - - - - - - -	Small Small Moderate Small Small Small Moderate Small Small Small Moderate	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely	Very low Very low Low Very low Very low Very low Low Very low Very low Very low Low
Self-reported perceived recovery		12-month follow up
	2	Age Poor general health (SF 36 < 50) History of non-traumatic knee symptoms Floating patella	[28] [28] [28] [28]	OR OR OR OR	1.03 (1.01–1.05) 2.64 (0.93–7.52) 1.94 (1.14–3.28) 0.52 (0.30–0.91)	<0.001 <0.07 0.01 0.02	1.03 (1.01–1.05) 3.10 (1.18–8.16) 1.94 (1.14–3.28)0.48 (0.27–0.84)	<0.001 0.02 0.01 0.01	- - - -	Small Moderate Small Moderate	Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely	Very low Very low Very low Very low
		54-month follow up
		Age BMI >27 non MSK comorbidity Self-reported crepitus History of non-traumatic knee symptoms	[28] [28] [28] [28] [28] [31]	OR OR OR OR OR HR	1.04 (1.01–1.06) 3.30 (1.72–6.32) 2.14 (0.98–4.69) 3.28 (1.65–6.49) 2.96 (1.53–5.73) 0.47 (0.30–0.74)	<0.01 <0.01 0.06 <0.01 <0.00 <0.20	1.03 (1.01–1.05) 2.86 (1.44–5.68) 2.40 (1.04–5.57) 2.22 (1.38–3.59) 2.28 (1.15–4.53) 0.51 (0.33–0.81)	0.02 <0.001 0.04 0.00 0.02 <0.001	- - - - Yes Yes	Small Moderate Small Small Small Moderate	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely	Very low Very low Very low Very low Low Low
		Three-month follow up
Poor functional outcome	3	Age Female Duration of knee complaint WOMAC pain WOMAC functioning PCI sub-scale [3]; distress high vs. lowest tertitle Complaints of upper and lower extremity vs. knee only complaint	[31] [31] [31] [31] [31] [31] [31]	RC RC RC RC RC RC RC	−0.29 (−0.48 to 0.09) - −3.74 (−5.57 to 1.91) 0.33 (0.19–0.48) 0.47 (0.35–0.59) - −6.72 (−13.26 to 0.18)	<0.20 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20	−0.21 (−0.36 to −0.06) −8.00 (−12.53 to −3.46)−2.58 (−4.01 to −1.15) −0.21 (0.39–0.04) 0.82 (0.66–0.99) −17.40 (−29.10 to −5.70) −5.19 (−10.36 to −0.03)	0.01 <0.001 <0.001 0.02 <0.001 <0.001 0.05	- - - - - - -	Small Large Moderate Small Small Large Large	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely Likely	Very low Very low Low Very low Very low Low Low
		12-month follow up
		Age Duration of knee complaint WOMAC stiffness WOMAC functioning PCI sub-scale [4]; retreating mid vs. low PCI sub-scale [2]; distraction high vs. low	[31] [31] [31] [31] [31] [31]	RC RC RC RC RC RC	−0.29 (−0.48 to 0.09) −3.74 (−5.57 to 1.91) 0.14 (0.02–0.25) 0.47 (0.35–0.59) 11.48 (3.48–19.48 −24.35 (−41.25 to 7.46)	<0.20 <0.20 <0.20 <0.20 <0.20 <0.20	−0.29 (−0.45 to 0.12) −2.71 (−4.19 to −1.24 −0.16 (−0.29 to −0.03 0.65 (0.50–0.80) 6.54 (0.18–12.89)−28.16 (−42.41 to −13.90)	<0.001 <0.001 0.02 <0.001 0.04 <0.001	- - - - - -	Small Moderate Small Small Large Large	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely	Very low Low Very low Very low Low Low
		18-month follow up
		Age 60–69 Age 70+ BMI 25–29.9 BMI > 30 Possible anxiety Probable anxiety Chronic pain grade II Chronic pain grade III Duration of morning stiffness <30 min Local tender point count 2 Local tender point count 3 Local tender point count 4–6 Single leg stand 10–29 s Single leg stand 4–9 s Single leg stand <4 s	[29] [29] [29] [29] [29] [29] [29] [29] [30] [30] [30] [30] [30] [30] [30]	RR RR RR RR RR RR RR RR RR RR RR RR RR RR RR	1.39 (1.08– 1.77) 1.34 (1.02–1.77) 1.58 (1.11–2.26) 1.82 (1.28–2.60) 1.40 (1.11–1.76) 1.52 (1.12–2.07) 1.39 (1.09–1.76) 1.80 (1.30–2.49) 1.68 (1.33–2.13) 1.51 (1.12–2.05) 1.66 (1.21–2.27) 1.63 (1.20–2.23) 1.34 (0.97–1.83) 1.61 (1.20–2.15) 1.65 (1.21–2.24)	0.012 0.038 0.011 0.001 0.005 0.007 0.008 <0.001 <0.001 0.008 0.002 0.002 0.072 0.001 0.001	1.38 (1.06–1.80)1.44 (1.08–1.92)1.50 (1.04–2.15)1.64 (1.14–2.38)1.43 (1.06–1.71)1.44 (1.04–1.98)1.34 (1.05–1.71)1.55 (1.10–2.11.47 (1.13–1.89)1.45 (1.06–1.96)1.54 (1.12–2.121.48 (1.07–2.04)1.27 (0.92–1.74)1.50 (1.12– 2.01)1.49 (1.09–2.04)	0.017 0.014 0.029 0.008 0.015 0.027 0.023 0.013 0.004 0.018 0.008 0.017 0.146 0.007 0.014	- - - - - - - - - - - - - - -	Small Small Small Small Small Small Small Small Small Small Small Small Small Small Small	Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise Imprecise	Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely Likely	Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low Very low

Key: Pain Coping Inventory (PCI (strategy number used); Body Mass Index (BMI); Musculoskeletal (MSK); Passive Range of Movement (PROM); Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC); Regression Coefficient (RC); Risk Ratio (RR); Odds Ratio (OR); Confidence Interval (CI). RC interpretation: Value >0 = greater reduction in pain/improved function; <0 = less reduction in pain or functioning. RC classification of effect size: Small if measures between -1.4 to 0 and 0 to 1.4, moderate -1.41 to -3.4 and 1.4 to 3.4, large > -3.41 and >3.41. OR/HR interpretation: Value >1 = Increased association; 1 = no association; <1 to a limit of 0 = reduced association. OR/HR classification of effect size >1: Small if measures between 1 to -1.49, moderate 2.5 to 4.24, large >4.25 (24). OR/HR classification of effect size <1: Small if measures between 0.99 to 0.66, moderate 0.65 to 0.32 and large if <0.32.

Effect size and confidence intervals in bold text indicate prognostic value (>moderate effect size with or without narrow CI's or small effect size with narrow CI's)

Figure 1. PRISMA flow chart outlining the literature search strategy and study selection process.

Figure 2. Graph comparing poorer education level as a prognostic factor for persisting knee symptoms in two at 12-month follow up – odds ratio analyses.

Figure 3. Graph comparing bilateral knee symptoms as a prognostic factor for persisting knee symptoms in two at 12-month follow up – odds ratio analyses.

Bastick AN, Wesseling J, Damen J, et al. Defining knee pain trajectories in early symptomatic knee osteoarthritis in primary care: 5-year results from a nationwide prospective cohort study (CHECK). Br J Gen Pract. 2016;66(642):e32–e39.

 PubMed Web of Science ®Google Scholar

Belo JN, Berger MY, Koes BW, et al. Prognostic factors in adults with knee pain in general practice. Arthritis Rheum. 2009;61:143–151.

 PubMed Web of Science ®Google Scholar

Kastelein M, Luijsterburg PAJ, Belo JN, et al. Six-year course and prognosis of nontraumatic knee symptoms in adults in general practice: a prospective cohort study. Arthritis Care Res (Hoboken). 2011;63(9):1287–1294.

 PubMed Web of Science ®Google Scholar

Kastelein M, Luijsterburg PAJ, Heintjes EM, et al. The 6-year trajectory of non-traumatic knee symptoms (including patellofemoral pain) in adolescents and young adults in general practice: a study of clinical predictors. Br J Sports Med. 2015;49(6):400–405.

 PubMed Web of Science ®Google Scholar

Kastelein M, Luijsterburg PAJ, Verhaar JAN, et al. Six-year course and prognosis of traumatic knee symptoms in general practice: cohort study. Eur J Gen Pract. 2016;22(1):23–30.

 PubMed Web of Science ®Google Scholar

Mallen CD, Peat G, Thomas E, et al. Predicting poor functional outcome in community-dwelling older adults with knee pain: prognostic value of generic indicators. Ann Rheum Dis. 2007;66:1456–1461.

 PubMed Web of Science ®Google Scholar

Thomas E, Peat G, Mallen C, et al. Predicting the course of functional limitation among older adults with knee pain: do local signs, symptoms and radiographs add anything to general indicators? Ann Rheum Dis. 2008;67:1390–1398.

 PubMed Web of Science ®Google Scholar

van der Waal JM, Bot SDM, Terwee CB, et al. Course and prognosis of knee complaints in general practice. Arthritis Rheum. 2005;53:920–930.

 PubMed Web of Science ®Google Scholar

Institute for Health Metrics and Evaluation. Findings from the Global Burden of Disease Study 2017 | Institute for Health Metrics and Evaluation (IHME). Findings from the Global Burden of Disease Study 2017.Seattle, WA: IHME, 2018.

 Google Scholar

Dey P, Callaghan M, Cook N, et al. A questionnaire to identify patellofemoral pain in the community: an exploration of measurement properties. BMC Musculoskelet Disord. 2016;17(1):237.

 PubMedGoogle Scholar

Kontis V, Bennett JE, Mathers CD, et al. Future life expectancy in 35 industrialised countries: projections with a bayesian model ensemble. Lancet. 2017;389(10076):1323–1335.

 PubMed Web of Science ®Google Scholar

Data availability statement

Data sharing is not applicable to this article as no new data were created or analysed in this study.