3.1. Search results and study selection
The comprehensive search identified a total of 5.967 articles across the five databases (MEDLINE, EMBASE, CINAHL, CENTRAL and WHO COVID 19) and 8 additional records through other sources. After duplicates were removed, 4.625 references remained for the initial screening by title and abstract. This screening resulted in a total of 4.571 excluded articles. The remaining 54 articles were screened by full text, of which 35 did not meet the inclusion criteria with following reasons: wrong study population (n= 13), wrong outcome (n= 10), no full-text available (n= 6), wrong follow-up time (n= 2), wrong study design (n= 2), duplicate (n= 1) and wrong language (n= 1). 19 studies met the inclusion criteria and were included in the systematic review [17, 33, 54-70].
3.2. Study characteristics
The characteristics of the 19 studies included are reported in Table 1. The studies were published between 2021 and 2023. Regarding study design, 12 studies corresponded to cohort studies and 7 to cross-sectional studies. The overall population included 21.155 patients (65.3% female; 34.7% male). Study sample sizes ranged from 42 to 11.955 patients with post-COVID-19 (mean: 1.113). Most studies included middle aged participants (on average 49.15 years of age). The mean follow-up time was 11 months (range 3 months – 24 months). In all, 14 studies included patients treated in the hospital (range: 6.5% - 100%), 8 studies patients treated in intensive care units (ICU) (range: 1.2% - 28.5%) during acute SARS-CoV-2 infection, 3 studies included only non-hospitalized patients and 2 studies did not provide information. The studies were conducted in 14 different countries, of which 5 were implemented in Germany, 3 in Sweden, 2 each in Italy and the United Kingdom and 1 each in Brazil, Canada, Spain, Switzerland, Australia, Denmark and France.
Table 1 Characteristics of the study population
|
Study
|
Country
|
Study design
|
Sample size (N)
|
Sex
(% female)
|
Mean (M) age (SD) or Median (Mdn) age (IQR)
|
Follow-up time
|
Acute COVID-19 severity
|
Amorim et al. 2022 [54]
|
Brazil
|
cohort study
|
780
|
317 (40.6%)
|
M: 48 (12)
|
3 months
|
352 (45.1%) hospitalized; 131 (16.8%) intensive care unit (ICU)
|
Brehon et al. 2022 [55]
|
Canada
|
cohort study
|
81
|
52 (64.2%)
|
M: 48.9 (10.5)
|
165.2 days
|
N/A
|
Buonsenso et al. 2022 [56]
|
Italy
|
cohort study
|
155
|
79 (50.9%)
|
M: 46.48 (7.3)
|
12 months
|
18 (11.6%) hospitalized
|
Davis et al. 2021 [33]
|
United Kingdom
|
cross-sectional study
|
3.762
|
2.961 (78.9%)
|
N/A
|
6 months
|
317 (8.4%) hospitalized
|
Delgado-Alonso et al. 2022 [57]
|
Spain
|
cross-sectional study
|
77
|
67 (87.0%)
|
M: 46.31 (7.97)
|
20.71 months
|
15 (19.5%) hospitalized; 3 (3.9%) ICU
|
Diem et al. 2022 [58]
|
Switzerland
|
cross-sectional study
|
309
|
249 (80.6%)
|
M: 44.6; range: 19-83 years
|
13.0 months
|
33 (10.7%) hospitalized; 14 (4.5%) ICU
|
Harvey-Dunstan et al. 2022 [59]
|
United Kingdom
|
cohort study
|
42
|
28 (66.7%)
|
M: 49 (10)
|
44 weeks
|
non-hospitalized patients
|
Hodgson et al. 2021 [60]
|
Australia
|
cohort study
|
160
|
63 (39.4%)
|
Mdn: 62 (55-71)
|
6 months
|
all patients were hospitalized
|
Kedor et al. 2022 [61]
|
Germany
|
cross-sectional study
|
42
|
28 (69.0%)
|
Mdn: 36.5; range: 22–57
|
6 months
|
3 (7.1%) hospitalized
|
Kisiel et al. 2022 [62]
|
Sweden
|
cross-sectional study
|
336
|
253 (75.3%)
|
M: 43.1 (13.4)
|
12 months
|
non-hospitalized patients
|
Kupferschmitt et al. 2023 [63]
|
Germany
|
cohort study
|
150
|
39 (76.5%)
|
M: 52.32 (9.16)
|
at least 3 months
|
N/A
|
Müller et al. 2023 [64]
|
Germany
|
cohort study
|
127
|
97 (76.4%)
|
M: 50.62 (10.74)
|
408.81 days
|
33 (25.9%) hospitalized; 10 (7.9%) ICU
|
Nielsen et al. 2022 [65]
|
Denmark
|
cross-sectional study
|
448
|
325 (72.5%)
|
M: 46.8 (12.6)
|
246 days
|
63 (14.1%) hospitalized
|
Peters et al. 2022 [66]
|
Germany
|
cross-sectional study
|
2.053
|
1.677 (81.7%)
|
Mdn: 51
|
3-15 months
|
133 (6.5%) hospitalized; 35 (1.7%) ICU
|
Rutsch et al. 2023 [67]
|
Germany
|
cohort study
|
221
|
145 (65.6%)
|
M: 52.4 (9.0)
|
5 months
|
82 (37%) hospitalized; 18 (8.1%) ICU
|
Sansone et al. 2022 [68]
|
Italy
|
cohort study
|
247
|
159 (64.4%)
|
M: 48.1 (10.5)
|
15 months
|
108 (43.7%) hospitalized, 3 (1.2%) ICU
|
van Wambeke et al. 2023 [69]
|
France
|
cohort study
|
45
|
28 (62.2%)
|
M: 49.6 (11.2)
|
15.1 months and 22.6 months
|
non-hospitalized patients
|
Wahlgren et al. 2023 [17]
|
Sweden
|
cohort study
|
165
|
61 (37%)
|
M: 61 (13)
|
24 months
|
100% hospitalized; 47 (28.5%) ICU
|
Westerlind et al. 2021 [70]
|
Sweden
|
cohort study
|
11.955
|
7.129 (59.6%)
|
M: 48 (11.3)
|
4 months
|
2.960 (24.8%) hospitalized
|
N/A = not available; SD = standard deviation; IQR = interquartile range
|
3.3. Impact of post-COVID-19 on work ability
Out of the 19 studies included in the review, 15 of them providing data on the impact of post-COVID-19 on work ability. The following analysis summarizes the key findings from each study regarding work status, sick leave, work ability, and limitations in work duties/hours and taking into account different follow-up periods (see Table 2).
3.3.1. Follow-up less than 12 months
Many post-COVID-19 patients experienced a prolonged recovery period after COVID-19, leading to temporary or long-term work limitations. Five studies had a follow-up time less than 12 months after the acute SARS-CoV-2 infection. The studies revealed that even individuals with mild or moderate acute SARS-CoV-2 infection required an extended period to recover their pre-illness work capacity. In the study by Davis et al. [33] 817 (23.3%) patients were not working 3-7 months after the SARS-CoV-2 infection. Additionally, 1.598 (49.3%) participants were working reduced hours, suggesting some limitations in their work capacity. On the other hand, 957 (27.3%) participants were able to maintain the same working hours as before the onset of infection. The cross-sectional study conducted by Nielsen et al. [65] with a follow-up of approximately 8 months reported a higher percentage of participants working the same hours as before the SARS-CoV-2 infection. Specifically, 39.4% of the participants in the study were able to continue working to the same extent as they did prior to the infection, while 215 out of 401 (53.6%) were on sick leave. The national registry-based study by Westerlind et al. [70] involved 11.955 patients (follow-up: 4 months) and reported that 1.592 (13.3%) were on sick leave for post-COVID-19. Kedor et al. [61] consisted a cross-sectional study of 42 post-COVID-19 patients with a follow-up of 6 months and highlighted that participants had a median Bell disability score of 40 (post-COVID/ME/CFS-group) and 50 (post-COVID/non-ME/CFS-group) indicating limited work ability (reduced working hours or inability to work). Kupferschmitt et al. [63] and Rutsch et al. [67] are two studies that evaluated work ability following rehabilitation. Kupferschmitt et al. [63] examined a sub-sample of 51 post-COVID-19 patients. Prior to admission, a significant proportion of participants (43.1%) were on sick leave for over 6 months, 13.7% for 3-6 months, and 41% for less than 3 months. 2.0% were not employable. Out of the 51 individuals assessed, 28 (54.9%) were unable to work on admission. At the time of discharge, 18 participants (35.3%) showed an ability to work at least 6 hours per day. Additionally, 6 patients (11.8%) underwent gradual reintegration. In the study conducted by Rutsch et al. [67], the rehabilitation took place 5 months after SARS-CoV-2 infection. Among the participants, 88 out of 178 (49%) were on sick leave, with an average duration of 21 weeks. The study reported that 32% of participants experienced restoration of their work ability after rehabilitation. The mean Work Ability Score (WAS) of the Work Ability Index (WAI) was 4 on a scale of 0-10, indicating some limitations in work capacity. 41% perceive their work ability as permanently at risk.
3.3.2. Follow-up between 12 months and 18 months
In 15 studies that provided data on the impact of post-COVID-19 on work ability, there were 6 studies with a follow-up period between 12 months and 18 months. In the study of Buonsenso et al. [56] with a sample size of 154 participants, the majority (85.7%) maintained the same occupational status as before COVID-19. However, 22 patients (14.3%) experienced a change in their work status with following reasons: sick leave (n=7), loss of job due to ill health (n=3), shortening of working hours (n=3), fired (n=1), different reasons (n=7). Kisiel et al. [62] included 158 post-COVID-19 patients followed up for 12 months. Among the 158 participants, 35% were on sick leave during the follow-up period, with an average duration of 8.1 weeks. Patients with persistent symptoms at the 12-month follow-up reported a decrease in work ability. With a follow-up of 13 months, Diem et al. [58] highlighted that 168 patients (62.7%) were unable to work, and the average sick leave duration was 26.6 weeks. There was a significant association between inability to work and symptoms such as fatigue, pain, and sleep disturbances. Müller et al. [64] performed a study on a total of 127 patients, and had a median time between SARS-CoV-2 infection and beginning of rehabilitation of 408.81 days. Among the participants, 90 (72.5%) were unable to work after rehabilitation. The majority of patients reported poor (69.8%), 29.3% moderate, and only 0.9% good work ability measured by the WAI. WAI-scores (scale 7-49) before (Median (Mdn): 24.75) and after (Mdn: 24.75) rehabilitation did not show significant changes. The study by Peters et al. [66] involved a large sample of 1.406 post-COVID-19 patients. The WAS-scores (scale 0-10) decreased from 9.3 before COVID-19 to 6.8 at the time of the survey, indicating a decline in work ability over time. The authors showed that the work ability was significant different between patients with symptoms>3 months and patients without symptoms. Sansone et al. [68] conducted a study with 247 participants who were followed up for 15 months. The findings reveal that participants with symptoms lasting 200 or more days (Mean (M): 4.5 ±1.44) had significantly lower mean work ability-scores (scale 1-6) compared to those with symptoms lasting less than 200 days (M: 5.18 ±1.08; p<0,001).
3.3.3. Follow-up more than 18 months
Some individuals experienced a prolonged recovery period after COVID-19, leading to long-term work limitations. Three studies had a follow-up time of 15 months and longer. Delgado-Alonso et al. [57] involved 77 participants who were followed up for an average of 20.71 months. Out of the participants, 38 (49.4%) were working, while 39 (50.6%) were not working. Among those who were currently not working, 36 (92.3%) were on sick leave. A portion of the participants (16%) reported reduced working hours, and 23% required job adaptation (e.g., more breaks, telework, cognitive aids, or a position change). Factors contributing to work disability include higher levels of fatigue, and lower cognitive performance. The cohort study by van Wambeke et al. [69] included 45 participants who were followed up for 22 months. Among these participants, 18 (40%) patients were working full-time, 3 (6%) working 60%-70% of the time, 8 (18%) working half-time and the remaining individuals (36%) did not return to work. Among the mentioned studies, Wahlgren et al. [17] conducted the longest follow-up period with 24 months. At the 4-month follow-up, the majority (69.1%) of the participants were working, while a smaller proportion (23.4%) were on sick leave. At the 24-month follow-up, a similar percentage (66 out of 94 patients) were working, and a smaller proportion (16 out of 94 patients; 16%) were on sick leave.
Table 2 Impact of post-COVID-19 on work ability
|
Study
|
Sample size (n)
|
Follow-up time
|
Work status
|
Work ability
|
Limitations to work duties/hours
|
Kupferschmitt et al. 2023 [63]
|
51
|
at least 3 months
|
- 28/51 (54.9%) unable to work on admission
- sick leave before admission: 43.1% >6 months, 13.7% 3-6 months, 41% <3months
- 18/51 (35.3%) ability to work (at least 6h/day) at discharge
|
|
- 6/51 (11.8%) gradual reintegration
|
Westerlind et al. 2021 [70]
|
11.955
|
4 months
|
- 1.592/11.955 (13.3%) sick leave for at least 12 weeks
|
|
|
Rutsch et al. 2023 [67]
|
178
|
5 months
|
- 88/178 (49%) sick leave (M: 21 weeks)
- 32% restoration of their work ability after rehabilitation
|
- WAS (0-10) pre reha: M: 4
|
|
Kedor et al. 2022 [61]
|
42
|
6 months
|
|
- Bell disability score (0-100) Mdn.: 40 and 50 of 100 at the time of the survey
|
|
Davis et al. 2021 [33]
|
3.505
|
3-7 months
|
- 957/3.505 (27,3%) working same hours as before
- 817/3.505 (23.3%) not working
|
|
- 1.598/3.505 (49,3%) working reduced hours
|
Nielsen et al. 2022 [65]
|
401
|
246 days
|
- 158/401 (39,4%) working same hours as before
- 215/401 (53,6%) sick leave
|
|
|
Buonsenso et al. 2022 [56]
|
155
|
12months
|
- 132/154 (85.7%) patients same occupational status as before COVID-19
- 22/154 (14.3%) changed status
- 7 sick leave
|
|
- 3 patients shortening of working hours
|
Kisiel et al. 2022 [62]
|
336
|
12 months
|
- 55/158 (35%) sick leave during the follow-up period (M: 8.1 weeks ± 6.5)
|
- Work ability had decreased in patients with persistent symptoms at the 12-month follow-up
|
|
Diem et al. 2022 [58]
|
268
|
13 months
|
- 168/268 (62.7%) inability to work
- sick leave M: 26.6 weeks (95%CI 23,5-29.6)
|
- people with reported inability to work à more commonly fatigue (p <0.001), pain (p= 0.008), sleep disturbances (p= 0.030)
|
|
Müller et al. 2023 [64]
|
127
|
408.81 days
|
- inability to work: 86/127 (67.7%) pre reha; 90/124 (72.5%) after reha
- 69.8% of the patients reported poor, 29.3% moderate, and 0.9% good work ability
|
- WAI (7-49): total score pre reha: Mdn.: 24.75 (IQR: 21–28); post reha: Mdn.: 24.75 (IQR: 21–28)
|
|
Peters et al. 2022 [66]
|
1.406
|
3-15 months
|
|
- WAS (0-10): before COVID-19: M: 9.3 (±1.2); at the time of the survey: M:6.8 (±2.2)
|
|
Sansone et al. 2022 [68]
|
247
|
15 months
|
|
- adopted WAI (1-6): Group A (symptoms < 200 days): M: 5.18 (±1.08) at the time of the survey; Group B (200+ days): M: 4.5 (±1.44); Group A significantly higher (p < 0.001)
|
|
Delgado-Alonso et al. 2022 [57]
|
77
|
20.71 months
|
- 38 (49.4%) patients were working
- 36 (46.8%) sick leave M: 12.07 ± 8,07 months
|
- not working associated with higher levels of fatigue (U= 372.50; p <0.001) and lower cognitive performance (U= 488.50; p = 0.010)
|
- 12/77 (16%) reduced working hours
- 18/77 (23%) job adaptation
|
Van Wambeke et al. 2023 [69]
|
45
|
22 months
|
- 18 (40%) working full-time; 16 (36%) no return to work
|
|
- 8 (18%) working half-time; 3 (6%) working 60%-70% of the time
|
Wahlgren et al. 2023 [17]
|
94
|
24 months
|
- 4 months: 65/94 (69.1%) working; 22/94 (23.4%) sick leave
- 24 months: 66/94 (70.2%) working, 15/94 (16%) sick leave
|
|
|
IQR = interquartile range
|
<Please insert Table 2 here.>
3.4. Impact of post-COVID-19 on Return-to-Work
The meta-analysis included eight studies that examined the RTW outcomes of patients previously infected with SARS-CoV-2 (Fig. 2). The random-effects meta-analysis estimated a pooled proportion of 0.609 (95% CI: 0.458-0.751), indicating that approximately 60.9% of post-COVID-19 patients were able to successfully RTW 12 or more weeks following the SARS-CoV-2 infection. In the Forest Plot, the dashed line represents the aggregated average RTW rate across all studies. Studies to the right of this line tend to indicate higher RTW rates, while those to the left suggest lower rates. Among the individual studies, Hodgson et al. [60] had the highest weight (13.1%) and reported a proportion of 0.886 (95% CI: 0.813-0.938), suggesting a high likelihood of successful RTW. The remaining studies had weights ranging from 11.5% to 13.0% and reported proportions ranging from 0.414 to 0.833. Heterogeneity analysis yielded an I2 index of 92% and a τ2 of 0,042 with p <0.01, indicating substantial variability and inconsistency. Visual inspection of funnel plot asymmetry for the RTW meta-analysis did not suggest the presence of publication bias (Appendix 5), and the Peters’ regression test (intercept = 1.111; standard error (SE) = 0.147; p = 0.146) was not statistically significant.
<Please insert Figure 2 here.>
3.5. Factors influencing the work ability and Return-to-Work of post-COVID-19 patients
Based on the information provided from various studies, several influencing factors for work ability and RTW of post-COVID-19 patients could be identified. The duration between symptom onset and the beginning of rehabilitation or treatment influences the likelihood of returning to work [55]. Early intervention and rehabilitation improve the chances of returning to work. Job adaptations and modified duties, such as reduced working hours, tasks with lower physical or mental strain, telework or flextime can positively affect work ability and facilitate the RTW [33, 55]. Economic factors and financial needs can force post-COVID-19 patients to continue working or RTW sooner despite ongoing symptoms [33]. An individuals' work ability can be significantly impacted by various psychological factors, among which include high levels of fatigue, depressive symptoms, and reduced cognitive performance. These factors are closely linked with diminished work capacity and overall effectiveness in the workplace [57]. In addition, Diem et al. [58] reported, that inability to work is commonly reported alongside symptoms such as fatigue, sleep disturbances, and pain. These symptoms act as significant barriers for post-COVID-19 patients, impeding their ability to engage in work-related activities and having a negative impact on overall performance and productivity. The presence of fatigue is also associated with a lower likelihood of returning to previous work hours [59]. Additionally, certain demographic and health-related factors have been associated with higher odds of not returning to work after SARS-CoV-2 infection. According to the study conducted by Westerlind et al. [70], factors such as older age, being male, having a history of sick leave before contracting COVID-19, and having received inpatient care are all associated with an increased probability of not returning to work. Overall, the influencing factors for work ability and RTW of post-COVID-19 patients are diverse and can vary between individuals, interacting in complex ways to determine work outcomes.
3.6. Post-COVID-19 Symptoms
Studies [57-59] have underscored the impact of post-COVID-19 symptoms on an individual's work ability and RTW. Therefore, it is crucial to outline the prevalent post-COVID-19 symptoms reported in the included studies. 13 studies investigated self-reported post-COVID-19 symptoms in COVID-19 patients 12 or more weeks following diagnosis. The studies reported on a wide range of post-COVID-19 symptoms experienced by patients. Appendix 3 presents the five most commonly reported symptoms in the included studies, along with their respective prevalence rates. The prevalence of post-COVID-19 symptoms varied across studies, with estimates ranging from 12.2% to 100% of individuals who had recovered from the acute phase of the illness. Fatigue was the most commonly reported symptom, with prevalence rates exceeding 80% in many studies (mean prevalence: 72.9%). Other frequently reported symptoms included neurocognitive disorders such as concentration impairment, dizziness or memory problems. Estimates of neurocognitive symptoms prevalence ranged from 14% to 92% (mean prevalence: 59.5%). Most of the studies also reported physical ailments such as weakness, muscle pain or exercise intolerance with prevalence rates between 13% and 100% (mean prevalence: 56.2%). Other frequently reported symptoms included shortness of breath, headache, and sleep disturbances. Long-term follow-up studies indicated that patients with post-COVID-19 continued to experience symptoms for up to two years after the initial infection [17, 57, 68, 69]. For a comprehensive understanding of the full range of post-COVID-19 symptoms, it is recommended to refer to the original studies included in this systematic review.
3.7. Risk of bias
Of the 19 studies, more than half were assessed to be of moderate quality (n=10). Five studies were considered to be of high quality, and the remaining studies (n=4) were considered to be of poor quality. Taken together, the NOS rating of the component studies was moderate, evidenced by mean scores of 6.2 for cohort studies and 4.8 for cross-sectional studies. The NOS quality assessment results for cohort studies are summarized in Table 3, and quality assessment results for cross-sectional studies are summarized in Table 4.
3.7.1. Selection
Within the cohort studies (n=12), nearly all studies scoring a star for being either truly or somewhat representative of the average target population. Common methodological limitations were the failure to include a non-exposed group in cohort studies, and to ascertain whether outcomes were present prior to SARS-CoV-2 infection. The exposure (COVID-19) was usually measured using either objective measurement (e.g., polymerase chain reaction (PCR) test) or clinical judgment. Within cross-sectional studies (n=7), all but two studies had somewhat representative or truly representative samples (with selection bias). However, nonresponse characteristics (with non-response/self-selection bias); and a sample size justification were not provided or poorly described in all of the cross-sectional studies. 5 out of 7 studies used validated measurement tools.
3.7.2. Comparability
Cohort studies controlled for confounders in 12 of the 13 studies. However, only two studies controlled for age, sex, and an additional factor required to score two stars. One study scored zero stars, as it used unadjusted analyses. In the cross-sectional studies, 5 studies used adjusted analyses.
3.7.3. Outcome
Within the cohort studies, all studies used a validated objective assessment tool (e.g., WAI) or a structured/systematic interview conducted by a trained healthcare/research professional and were followed up after a sufficient duration (3 months). The follow-up cohort rate was inadequate in 3 studies, as no description of differences in responders and non-responders was provided, or less than 80% responded. Within the cross-sectional studies, 3 studies used a validated objective assessment tool or a structured/systematic interview conducted by a trained healthcare/research professional; therefore, they scored a star. 3 studies scored zero stars, as they used self-reported work ability measurements. All of the cross-sectional studies were considered to have used appropriate and clearly described statistical tests.
Table 3 Newcastle-Ottawa Quality Assessment Scale criteria for cohort studies
|
Study
|
Selection (max. ★★★)
|
Comparability
(max. ★★)
|
Outcome (max. ★★★)
|
Total
|
Quality score
|
|
Representa-
tiveness of the exposed cohort
|
Selection of the non-exposed cohort
|
Ascertain-ment of exposure
|
Outcome not present at start of study
|
Comparability of cohorts on the basis of the design or analysis
|
Assessment of outcome
|
Follow-up long enough for outcomes to occur
|
Adequacy of follow-up of cohorts
|
|
|
Amorim et al. 2022 [54]
|
0
|
0
|
★
|
★
|
★
|
★
|
★
|
★
|
6★
|
moderate
|
Brehon et al. 2022 [55]
|
★
|
0
|
★
|
★
|
★
|
★
|
★
|
0
|
6★
|
moderate
|
Buonsenso et al. 2022 [56]
|
★
|
0
|
0
|
★
|
★
|
★
|
★
|
0
|
5★
|
moderate
|
Harvey-Dunstan et al. 2022 [59]
|
0
|
0
|
★
|
0
|
★
|
★
|
★
|
0
|
4★
|
low
|
Hodgson et al. 2021 [60]
|
★
|
0
|
★
|
★
|
★
|
★
|
★
|
★
|
7★
|
high
|
Kupferschmitt et al. 2023 [63]
|
★
|
★
|
★
|
★
|
★
|
★
|
★
|
★
|
8★
|
high
|
Müller et al. 2023 [64]
|
★
|
0
|
★
|
★
|
★★
|
★
|
★
|
★
|
8★
|
high
|
Rutsch et al. 2023 [67]
|
★
|
0
|
★
|
0
|
0
|
★
|
★
|
★
|
5★
|
moderate
|
Sansone et al. 2022 [68]
|
★
|
0
|
★
|
0
|
★★
|
★
|
★
|
★
|
7★
|
high
|
van Wambeke et al. 2023 [69]
|
★
|
0
|
★
|
★
|
★
|
★
|
★
|
★
|
7★
|
high
|
Wahlgren et al. 2023 [17]
|
★
|
0
|
0
|
0
|
★
|
★
|
★
|
★
|
5★
|
moderate
|
Westerlind et al. 2021 [70]
|
★
|
0
|
★
|
0
|
★
|
★
|
★
|
★
|
6★
|
moderate
|
Quality score: high=9-7 ★, moderate=6-5 ★, low=4 or fewer ★
|
Table 4 Newcastle-Ottawa Quality Assessment Scale criteria for cross-sectional studies
|
Study
|
Selection (max. ★★★)
|
Comparability
(max. ★★)
|
Outcome
(max. ★★★)
|
Total
|
Quality score
|
|
Representa-tiveness of the exposed cohort
|
Sample size
|
Non-respondents
|
Ascertainment of the exposure
(risk factor)
|
Comparability of subjects on the basis of design or analysis
|
Assessment of outcome
|
Statistical test
|
|
|
Davis et al. 2021 [33]
|
★
|
0
|
0
|
0
|
0
|
0
|
★
|
2★
|
low
|
Delgado-Alonso et al. 2022 [57]
|
★
|
0
|
0
|
★★
|
★
|
★
|
★
|
6★
|
moderate
|
Diem et al. 2022 [58]
|
★
|
0
|
0
|
0
|
★★
|
0
|
★
|
4★
|
low
|
Kedor et al. 2022 [61]
|
0
|
0
|
0
|
★★
|
★★
|
★
|
★
|
6★
|
moderate
|
Kisiel et al. 2022 [62]
|
★
|
0
|
0
|
★★
|
★★
|
0
|
★
|
6★
|
moderate
|
Nielsen et al. 2022 [65]
|
0
|
0
|
0
|
★★
|
0
|
★
|
★
|
4★
|
low
|
Peters et al. 2022 [66]
|
★
|
0
|
0
|
★★
|
★
|
★
|
★
|
6★
|
moderate
|
Quality score: high=9-7 ★, moderate=6-5 ★, low=4 or fewer ★
|