Table 3.
Source and derivation of mean differences used in review.
| Studies in meta‐analysis | ||||||||
|---|---|---|---|---|---|---|---|---|
| Row | Study | Participants abstinent on NRT, providing longitudinal data and loss to follow up (n) | Body fluid & assay times (i.e. after starting NRT) | Type and dose of NRT | Outcomesb | Derivation of outcome data | ||
| Mean difference (SEMD)d (ng/ml) | Percentage substitution (%) | |||||||
| 1 | Hegaard 2004 29 | 18 analysed | Saliva cotinine, 2 weeks | 2 mg gum (n = 6) | 97 (51.5) | 26.5 | Mean differences taken directly from paper SEMs calculated from CI presented in paper | |
| Of 75 enrolled 40 were excluded: | 15 mg/16‐hour patch (n = 7) | 103 (21.5) | 40.5 | |||||
| 16 dropped out | ||||||||
| 13 stopped NRT | 15 mg/16‐hour patch and 2 mg gum (n = 5) | 141 (47.5) | 42.7 | |||||
| 11 smoked the day before second sample | ||||||||
| 17 were not analysed: | ||||||||
| 15 samples not collected or not treated properly or went missing | ||||||||
| 2 used a 10‐mg patch | ||||||||
| 2 | Oncken 2009 30 | 14 | Serum cotinine, 4 days | 15 mg/16‐hour patch (n = 7) | 63 (12.47) | 54.3 | Mean differences taken directly from paper | |
| Spray 24 dose/day = 24 mg/24 hours (n = 7) | 91 (14.36) | 30.0 | SEM calculated from SD of difference | |||||
| 3 | Bowker 2014 40 | 33 | Saliva cotinine, 4 weeks | 15 mg/16‐hour patch | 51.2 (10.77) | 60.0 | Mean difference and SEM calculated using original study data | |
| 4 | Berlin 2014 39 | 18 | Saliva cotinine, 2 weeksc | Patch, variable strength | 70.22 (20.07) | 49.3 | Mean difference and SEM of abstinent women only calculated by study author using original study data | |
| Narratively reported studies | ||||||||
|---|---|---|---|---|---|---|---|---|
| Studies with additional data from authors | ||||||||
| Row | Study | Participants abstinent on NRT, providing longitudinal data and loss to follow‐up (n) | Body fluid & assay times (i.e. after starting NRT) | Type and dose of NRT | Outcomesb | Derivation of outcome data/reason for non‐inclusion in meta‐analysis | ||
| Mean difference (SEMD)d (ng/ml) | Percentage substitution (%) | |||||||
| 5 | Oncken 2008 38 | 4 | Urine cotinine, 6 weeks | 2 mg gum (n = 4) | 130.00 (245.738) | 65.2 | Mean difference and SD of abstinent women using gum at 6 weeks calculated by author from original data | |
| Not included in meta‐analysis as urine rather than saliva cotinine value | ||||||||
| Studies where only published data used | ||||||||
|---|---|---|---|---|---|---|---|---|
| Row | Study | Participants abstinent on NRT, providing longitudinal data and loss to follow‐up (n) | Type & dose of NRT body fluid & assay times (i.e. after starting NRT) | Outcomesb (ng/ml unless stated otherwise) | Reasons for non‐inclusion in meta‐analysis | |||
| 6 | Gennser 1975 33 | Unclear: 12 participants smoked at baseline, 6 were later using NRT but it is not clear which of the original 12 these were | Gum, 2 & 4 mg Blood nicotine, 30 minutes | Mean ± SEM | Only group means and their standard errors presented in paper; standard error of mean difference could not be calculated | |||
| Cig. 19.6 ± 1.4 (n = 12) | ||||||||
| 2 mg gum 10.0 ± 0.7 (n = 6) | ||||||||
| Difference in means calculated by review team as (group mean value for smoking) – (group mean value for gum use) | ||||||||
| Difference in means = 9.6 | ||||||||
| Cig. 19.6 ± 1.4 (n = 12) | ||||||||
| 4 mg gum14.7 ± 1.3 (n = 6) | ||||||||
| Difference in means = 4.9 | ||||||||
| 7 | Oncken 1996 34 | 15 | Gum, 2 mg blood cotinine, 5 days | Mean ± SD | Only group means and their standard errors presented in paper; standard error of mean difference could not be calculated | |||
| Of 19 enrolled 4 dropped out: 2 due to non‐abstinence 2 suffered severe nausea | ||||||||
| Smoking 153 ± 18 | ||||||||
| 5 days gum 33 ± 8 | ||||||||
| Difference in means = 120 | Difference in means calculated by review team as (group mean value for smoking) – (group mean value for gum use) | |||||||
| 8 | Oncken 1997 35 | 15 | Patch, 21 mg blood nicotine/hour | Mean ± SD (overlap: smoking group lower cotinine than patch group) | Only group means and their standard errors presented in paper; standard error of mean difference could not be calculated | |||
| Of 17 enrolled 2 were excluded: 1 single uterine artery 1 quit smoking after session 1 | ||||||||
| Difference in maximal plasma level calculated as (smoking maximal level) – (patch maximal level) | ||||||||
| Mean maximal plasma level | ||||||||
| Mean difference of areas under the plasma nicotine concentration time curve taken directly from paper | ||||||||
| Smoking 19.7 ± 8.09 | ||||||||
| Patch 16.0 ± 3.5 | ||||||||
| Difference in mean maximal plasma level 3.7 | ||||||||
| Time to reach max. | ||||||||
| Smoking 5 hours ± 2.4 | ||||||||
| Patch 3.2 hours ± 1.7 | ||||||||
| Area under the plasma nicotine concentration time curve | ||||||||
| Smoking 89 ng‐hour/ml | ||||||||
| Patch 93 ng ‐hour/ml | ||||||||
| Mean difference ± SEM of area under the plasma nicotine concentration time curve − 4.8 ± 10.3 ng‐hour/ml | ||||||||
| 9 | Wright 1997 23 | 6 | Patch, 21 mg saliva cotinine, 8 hours | aMean & range (μg/l) | Only group means and their standard errors presented in paper; standard error of mean difference could not be calculated | |||
| Smoking 100 (40–155) | ||||||||
| Patch 55 (20–100) | ||||||||
| Difference in means = 45 | ||||||||
| Difference in means calculated by review team as (group mean value for smoking) – (group mean value for gum use) | ||||||||
| 10 | Ogburn 1999 36 | 21 no reason given for missing data | Patch, 21 mg blood cotinine, day 1–4 | Mean ± SD | Only group means and their standard errors presented in paper; standard error of mean difference could not be calculated | |||
| Cotinine values | ||||||||
| Smoking 116 ± 54 (n = 21) | ||||||||
| Day 1 patch 142 ± 47 (n = 21) | ||||||||
| Day 2 patch 128 ± 38 (n = 20) | ||||||||
| Day 3 patch 123 ± 42 (n = 20) | ||||||||
| Day 4 patch 117 ± 38 (n = 20) | Difference in means calculated by review team as, (group mean value for smoking) – (group mean value for gum use); missing data ignored | |||||||
| Difference in means | ||||||||
| Days 1–26 | ||||||||
| Days 2–12 | ||||||||
| Days 3–7 | ||||||||
| Days 4–1 | ||||||||
| 11 | Wisborg 2000 37 | Unclear, as not all participants were abstinent when assays taken | Patch 15 mg, 10 mg Saliva cotinine, 8 & 11 weeks | Mean ± SD | Not all participants abstinent when assays occurred; only cross‐sectional data available in published report | |||
| Smoking (n = 124) 231 ± 125 8 weeks (15 mg) patch (n = 90) | ||||||||
| 153 Difference in means at 8 weeks, 78 | ||||||||
| 11 weeks (10 mg) patch (n = 83) 121. Difference in means at 11 weeks, 110 | ||||||||
| 12 | El‐Mohandes 2013 22 | Unclear, as not all participants were abstinent when assays taken | Patch, 14 mg, 7 mg Saliva cotinine, approx.2–4 & approx. 4–6 weeks | Mean ± SD (ng/ml) (n = 26) | Not all participants abstinent when assays occurred; only cross‐sectional data available in published report | |||
| Baseline 171 ± 143 14‐mg patcha 142. Difference in means 39 | ||||||||
| Difference in means calculated by review team as, (mean value for patch use) – (mean value for smoking) | ||||||||
| Baseline 171 ± 143 7‐mg patcha 129 Difference in means 42 | ||||||||
a
Data valued obtained using WebPlotDigitizer 24.
b
A negative mean difference/difference between means indicates higher cotinine/nicotine levels when smoking.
c
Data from one of two follow‐up times selected to avoid inclusion of non‐independent observations in meta‐analysis.
d
Standard error of the mean (SEM) difference. SD = standard deviation; CI = confidence interval; NRT = nicotine replacement therapy; BMI = body mass index.