Table 5.
Summary of findings from meta-analysis of impact of SM interventions.
| Study | HRQoL | Hospitalization | ED visits | Mortality |
|---|---|---|---|---|
| Education/action plans | ||||
| Effing et al.12 | WMD −2.58 (−5.14 to 0.02) | OR 0.64 (0.47 to 0.89) | NR | NR |
| Tan et al.13 | WMD −3.78 (−6.82 to −0.73) | OR 0.55 (0.43 to 0.71) | NR | NR |
| Turnock et al.14 | WMD −0.32 (−3.34 to 2.70) | b | NR | NR |
| Walters et al.15 | WMD −0.54 (−3.05 to 1.98) | b | NR | NR |
| Complex interventions with an SM focus | ||||
| Dickens et al.17 | NR | NR | OR 0.68 (0.57 to 0.80) | NR |
| Kruis et al.19 | WMD −0.22 (−7.43 to 6.99)c | OR 0.59 (0.28 to 1.22) | OR 0.64 (0.33 to 1.25) | OR 0.45 (0.16 to 1.28) |
| Zwerink et al.20 | WMD −3.51 (−5.37 to −1.65) | OR 0.57 (0.43 to 0.75) | NR | OR 0.79 (0.58 to 1.07) |
| Jordan et al.21 | WMD −3.84 (−6.40 to −1.29) | HR 0.78 (0.52 to 1.17) | Not combined, low quality | HR 1.07 (0.74 to 1.54) |
| Pulmonary rehabilitation | ||||
| McCarthy et al.22 | WMD −6.89 (−9.26 to −4.52)d | NR | NR | NR |
| Telehealth | ||||
| Cruz et al.23 | SMD −0.53 (−0.97 to −0.09) | RR 0.72 (0.53 to 0.98) | RR 0.68 (0.38 to 1.18) | RR 1.43 (0.40 to 5.03) |
| Kamei et al.24 | NR | RR 0.80 (0.68 to 0.94) | RR 0.52 (0.41 to 0.65) | RR 1.36 (0.77 to 2.41) |
| Lundell et al.25 | NR | NR | NR | NR |
| McLean et al.26 | WMD −6.57(−13.62 to −0.48) | OR 0.46 (0.33 to 0.65) | OR 0.27 (0.11 to 0.66) | NR |
| Outreach nursing programme | ||||
| Wong et al.27 | WMD −2.60 (−4.81 to −0.39) | OR 1.01 (0.71 to 1.44) | NR | NR |
ED: emergency department; HRQoL: health-related quality of life; HR: hazard ratio; NR: not reported; NA: not applicable; OR: odds ratio; RR: relative risk; SMD: standard mean difference; WMD: weighted mean difference.
aPrimary outcomes of QoL, hospitalization, ED visits, unscheduled/urgent health care use and mortality are reported. All outcomes are presented in detail in Table S1. When summarizing results, we included the longest term results reported. In terms of QoL outcomes across varying scales, we have included the total results (e.g. SGRQ total). All outcomes are presented in detail in Supplementary Table S1.
bTurnock et al. and Walters et al. both included pooled estimates for hospitalizations, but these were not presented as relative risks. Neither found a statistically significant impact. Walters also reported a pooled estimate for ED visits (no significant impact), but no estimate of relative risk.
cSt. George’s Respiratory Questionnaire (SGRQ) for QoL at >12 months.
dSGRQ total for HRQoL, results for SGRQ symptoms, impact, activity and CRQ reported in Supplementary Table S1.