Table 4.
Criterion-concurrent validity, criterion-predictive validity and construct validity of outcome measures
| Author, yr | Type of validity and OM | Missing values | Comparator OM or predicted outcome | Validation results |
|---|---|---|---|---|
|
Functional capacity |
|
|
|
|
| Jones 2012
[40] |
Crit-pred 6MWT |
Nil |
All-cause mortality |
Unadjusted HR p = 0.003; Compared to 6MWT <358.5 m adjusted HR = 0.61 (95% CI 0.34-1.07) if 6MWT 358.5-450 m; Compared to 6MWT <385.5 m adjusted HR = 0.48 (95% CI 0.24-0.93) if 6MWT >450 m |
| Pancieri 2010
[22] |
Crit-pred 6MWT |
NR |
Predicted post-op 6MWT = pre-op 6MWT x (FS – resected FS) ÷ FS |
r = 0.40, p<0.01 |
| Kasymjanova 2009
[25] |
Crit-pred 6MWT |
19pts dropped out |
Survival |
Compared to 6MWT ≥ 400 m mortality HR = 0.44 (95% CI 0.23-0.83) if 6MWT <400 m, p = 0.001 |
| Mao 2007* Miller 2005
[26,27] |
Crit-pred; construct 6MWT |
3pts not complete Ax |
1. Development of RP |
1. ROC area under curve = 0.41, p = 0.4 |
| 2. FEV1 |
2. r = 0.53, p<0.001 |
|||
| 3. FVC |
3. r = 0.44, p = 0.001 |
|||
| 4. DLCO |
4. r = 0.48, p<0.001 |
|||
| Saad 2007
[24] |
Crit-pred 6MWT |
9pts died; 30pts not complete Ax (rural) |
Predictors of improvement in pre-op to 180 days-post-op: |
1. GEE = 0.001, SE = 0.000, p = 0.003 |
| 1. SF-36 PF |
2. GEE = 0.001, SE = 0.000, p = 0.000 |
|||
| 2. SF-36 PR |
3. GEE = 0.001, SE = 0.000, p = 0.031 |
|||
| 3. SF-36 GH |
|
|||
| Parsons 2003
[39] |
Crit-pred 6MWT |
29pts |
LOS out of hospital < 30 days post-op |
Not significant |
| Pierce 1994
[23] |
Crit-pred 6MWT |
NR |
Post-op: |
1. p<0.05 |
| 1. respiratory failure |
2. p>0.05 |
|||
| 2. surgical POC |
3. p>0.05 |
|||
| 3. respiratory POC |
4. p>0.05 |
|||
| 4. cardiac POC |
5. p>0.05 |
|||
| 5. all POC | ||||
| Holden 1992
[41] |
Crit-pred 6MWT |
3pts not complete Ax |
Survival > 90 days post-op |
6MWT diff b/t groups with/without survival p<0.05; |
| 6MWT > 1000feet (305 m) pre-op sensitivity 100%, positive predictive value 85%, negative predictive value 100% for survival | ||||
| England 2012
[44] |
Construct ISWT |
Nil |
1. P max monitor (insp mm strength) |
1. r = 0.42, p = 0.01 |
| 2. dynamometry (peripheral mm power) |
2. r = 0.39, p = 0.01 |
|||
| 3. spirometry (% predicted FEV1) |
3. r = 0.22, p = 0.17 |
|||
| 4. spirometry (% predicted FVC) |
4. r = 0.21, p = 0.2 |
|||
| 5. CRDQ (mastery) |
5. r = 0.21, p = 0.18 |
|||
| Win 2006
[30] |
Crit-pred; crit-concurrent; construct ISWT |
Nil |
1. CPET (VO2peak) |
1. r = 0.67, p<0.001 |
| 2. CPET (VO2peak% predicted) |
2. r = 0.30 |
|||
| 3. spirometry (FEV1) |
3. r = 0.5 |
|||
| 4. 12 month survival |
4. ROC area = 0.7, p = 0.003 |
|||
| Win 2004
[64] |
Crit-pred, construct ISWT |
8pts |
1. poor surgical outcome (post-op death, MI, heart failure, resp failure, septicaemia, pneumonia, cardiac arrthymia) |
1. p = 0.6 between poor and sufficient outcome groups |
| 2. FEV1 |
2. r = 0.46, not significant |
|||
| Brunelli 2012
[43] |
Crit-pred SCT |
14 |
Median survival and 5-year survival |
Altitude >18 m independent predictor: HR = 0.5, p = 0.003 |
| Brunelli 2010
[31] |
Construct SCT |
Nil |
SCT (VO2max) |
Altitude: correlation coefficient = 0.7, p<0.0001 |
| Speed of ascent: correlation coefficient = 0.47, p = 0.005 | ||||
| Pancieri 2010
[22] |
Crit-pred SCT |
NR |
Predicted post-op SCT = pre-op SCT x (FS – resected FS) ÷ FS |
r = 0.66, p<0.001 |
| Brunelli 2008a
[34] |
Crit-pred SCT |
Nil |
POC < 30 days post-op |
Pre-op altitude: coefficient = −0.05, OR = 0.95 (95% CI 0.91-0.99), SE = 0.02, p = 0.045 |
| O2desat >4%: coefficient = 0.56, OR = 1.8 (95% CI 1–3), SE = 0.3, p = 0.05 | ||||
| Brunelli 2008b
[33] |
Crit-pred SCT |
Nil |
1. POC < 30 days post-op |
1. altitude: coefficient = 0.34, SE = 0.2, OR = 1.4 (95% CI 1.02-1.95), p = 0.04 |
| 2. Death < 30 days post-op |
2. altitude: coefficient = 0.91, SE = 0.4, OR = 2.5 (95% CI 1.1-5.5), p = 0.02 |
|||
| 3. Post-op hospital costs |
3. altitude: coefficient = 2160.2, SE = 573, p<0.001 |
|||
| Koegelenberg 2008
[29] |
Crit-concurrent SCT |
Nil |
CPET (VO2max) |
Altitude r2 = 0.06, Speed of ascent r2 = 0.77 (lung cancer only) |
| Nikolic 2008
[36] |
Crit-pred SCT |
Nil |
POC < 30 days post-op |
Best independent predictor = SpO2 after 40 steps and SCT duration for lobectomy group (60% sensitivity, 75% specificity cut off value 1.09) positive LR = 2.4 (95% CI 1.71-3.38), negative LR = 0.53 (95% CI 0.38-0.76) |
| Brunelli 2007
[20] |
Constuct SCT |
53pts at 3 months |
Post-op SCT (VO2peak) |
Pre-op SCT VO2peak directly associated with post-op SCT: regression analysis lobectomy F = 3.58, p<0.01; pneumonectomy F = 3.53, p<0.01 |
| Toker 2007
[37] |
Crit-pred SCT |
Nil |
POC (cardiac or pulmonary) |
SpO2 pre-SCT: OR = 0.74 (95% CI 0.58-1.00), p = 0.001 |
| Change SpO2 pre to post-SCT: OR = 1.59 (95% CI 1.21-2.10), p = 0.018 | ||||
| Brunelli 2005
[42] |
Crit-pred SCT |
13pts |
1. POC < 30 days post-op |
Inability to perform pre-op SCT: |
| 1. p = 0.7 | ||||
| 2. Death < 30 days post-op |
2. OR = 0.20 (95% CI 0.06-0.62), p = 0.005 |
|||
| Brunelli 2004
[35] |
Crit-pred SCT |
18pts |
POC < 30 days post-op |
Lower altitude pre-op independent predictor: coefficient = −0.18, p = 0.0015 |
| Brunelli 2003
[21] |
Crit-pred SCT |
Nil |
O2 desat during post-op SCT |
O2 desat during pre-op SCT independent variable: regression coefficient = 0.22, p = 0.0004 |
| Parsons 2003
[39] |
Crit-pred SCT |
29pts |
LOS out of hospital < 30 days post-op |
1. longer LOS correlated with speed of ascent r = 0.34, p≤0.02 |
| 2. workload achieved predicted LOS out of hospital r2 = 0.130 | ||||
| Brunelli 2002
[32] |
Crit-pred SCT |
Nil |
POC < 30 days post-op |
Altitude independent variable: p = 0.003 |
| Pate 1996
[38] |
Crit-pred SCT |
Nil |
POC < 30 days post-op |
Significant difference in pre-op SCT between pt who did and did not develop POC |
| Holden 1982
[41] |
Crit-pred SCT |
3pts |
Survival > 90 days post-op |
SCT diff b/t groups with/without survival p<0.05 |
| SCT > 44steps pre-op positive predictive value 91%, negative predictive value 80% for survival | ||||
|
Physical activity |
|
|
|
|
| Maddocks 2012
[47] |
Construct Acc |
Nil |
ECOG PS |
Statistically significant difference in mean daily step count, time spent sitting/lying, upright, standing or stepping between PS 0, 1 and 2 p<0.05 but not mean daily sit-to-stand transitions |
| Grutsch 2011a, 2011b; Du-Quiton 2010
[48-50] |
Construct Acc |
16pts Acc, 16pts questionnaires |
1. HADS |
1. Outpt: depression and activity r = −0.41, p = 0.04 |
| 2. Ferrans and Power QLI Cancer Version III |
2. Daytime activity and QLI domains of health/functioning r = 0.51, p<0.01; social/economic r = 0.38, p = 0.048; psychological/spiritual r = 0.45, p = 0.02; family r = 0.45, p = 0.02; overall QLI r = 0.57, p<0.01 |
|||
| 3. EORTC |
3. Inpt: daytime activity and loss of appetite r = −0.41, p = 0.005 |
|||
| 4. PSQI |
4. Outpt: lower sleep medication use and activity r = −0.58, p<0.01 |
|||
| Novoa 2011
[51] |
Construct Pedom |
13 pts – unable to perform exercise test |
CPET (VO2max) |
Mean daily total steps r = 0.4 |
| Mean daily aerobic steps r = 0.16 | ||||
| Mean daily time of aerobic capacity r = 0.11 | ||||
| Mean daily distance walked r = 0.44 | ||||
| Maddocks 2010
[46] |
Construct Acc |
2 pts withdrawn |
Estimated EE (stepping and non-stepping) measured from acc |
Non-stepping EE and daily step count r = −0.91, p<0.01 |
|
Muscle strength |
|
|
|
|
| Brown 2005
[53] |
Construct CST |
Nil |
1. KPS |
1. r2 = 0.565, p < 0.001 (ca group) |
| 2. FACIT-fatigue | 2. with incr fatigue, lower CST p<0.01 (ca group) |
Abbreviations: 6MWT, six minute walk test; 95% CI, 95% confidence intervals; acc, accelerometery; Ax, assessment; b/t, between; ca, cancer; CPET, cardio-pulmonary exercise testing; CRDQ, Chronic Respiratory Disease Questionnaire; crit, criterion; CST, chair-stand test; desat, desaturation; diff, difference; DLCO, diffused capacity for carbon monoxide; ECOG, Eastern Cooperative Oncology Group; EE, energy expenditure; EORTC, European Organisation for Research and Treatment of Cancer quality of life questionnaire; FACIT-fatigue, Functional Assessment of Chronic Illness Therapy- Fatigue scale; FEV1, force expired volume in one second; FS, lung functioning segments; FVC, forced vital capacity; GEE, generalized estimation equations; HADS, Hospital Anxiety and Depression Scale; HR, hazard ratio; insp, inspiratory; inpt, inpatient; ISWT, incremental-shuttle walk test; KPS, Karnofsky performance status; LOS, length of stay; LR, likelihood ratio; m, meters; max, maximum; MI, myocardial infarction; mm, muscle; NR, not reported; OM, outcome measure; OR, odds ratio; outpt, outpatient; P, pressure; pedom, pedometer; POC, post-operative complication; post-op, post-operative; pre-op, pre-operative; pred, predictive; PS, performance status; pt, participant; PSQI, Pittsburgh Sleep Quality Index; QLI, Quality of Life Index; r, correlation coefficient; ROC, received operating characteristic curve; SCT, stair-climb test; SE, standard error; SF-36, Short Form 36 physical functioning/physical role/general health domain; resp, respiratory; RP, radiation pneumonitis; Rx, treatment; SpO2, oxygen saturation; VO2max, maximum oxygen consumption; VO2peak, peak oxygen consumption; yr, year published.
* results presented from most recent publication.