This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Interactive behavior change technology (eg, computer programs, Internet websites, and mobile phones) may facilitate the implementation of lifestyle behavior interventions in routine primary health care. Effective, fully automated solutions not involving primary health care staff may offer low-cost support for behavior change.
We explored the effectiveness of an electronic screening and brief intervention (e-SBI) deployed through a stand-alone information kiosk for promoting physical activity among sedentary patients in routine primary health care. We further tested whether its effectiveness differed between patients performing the e-SBI on their own initiative and those referred to it by primary health care staff.
The e-SBI screens for the physical activity level, motivation to change, attitudes toward performing the test, and physical characteristics and provides tailored feedback supporting behavior change. A total of 7863 patients performed the e-SBI from 2007 through 2009 in routine primary health care in Östergötland County, Sweden. Of these, 2509 were considered not sufficiently physically active, and 311 of these 2509 patients agreed to participate in an optional 3-month follow-up. These 311 patients were included in the analysis and were further divided into two groups based on whether the e-SBI was performed on the patient´s own initiative (informed by posters in the waiting room) or if the patient was referred to it by staff. A physical activity score representing the number of days being physically active was compared between baseline e-SBI and the 3-month follow-up. Based on physical activity recommendations, a score of 5 was considered the cutoff for being sufficiently physically active.
In all, 137 of 311 patients (44%) were sufficiently physically active at the 3-month follow-up. The proportion becoming sufficiently physically active was 16/55 (29%), 40/101 (40%), and 81/155 (52%) for patients with a physical activity score at baseline of 0, 1 to 2, and 3 to 4, respectively. The patient-initiated group and staff-referred group had similar mean physical activity scores at baseline (2.1, 95% confidence interval [CI] 1.8-2.3, versus 2.3, 95% CI 2.1-2.5) and at follow-up, (4.1, 95% CI 3.4-4.7, vs 4.2, 95% CI 3.7-4.8).
Among the sedentary patients in primary health care who participated in the follow-up, the e-SBI appeared effective at promoting short-term improvement of physical activity for about half of them. The results were similar when the e-SBI was patient-initiated or staff-referred. The e-SBI may be a low-cost complement to lifestyle behavior interventions in routine primary health care and could work as a stand-alone technique not requiring the involvment of primary health care staff.
Physical inactivity is acknowledged to be the fourth leading risk factor for global mortality [
Primary health care has been acknowledged as a strategic setting for lifestyle behavior interventions, as indicated by the rapid increase in the number of studies in this field during the last decade [
In the light of these implementation challenges, researchers have suggested that the use of interactive behavior change technology (eg, computer programs, Internet websites, and mobile phones) could facilitate the implementation of lifestyle behaviors interventions in primary health care [
Acceptability of computer-based interventions has been reported to be high among patients in primary health care [
An electronic screening and brief intervention (e-SBI) system has been developed by a research team at Linköping University, Sweden. The system consists of a screening questionnaire collecting lifestyle data and an immediate feedback system that reports patient risk level and provides tailored advice for lifestyle behavior change. The e-SBI can be set up to be performed as part of ordinary patient counseling in primary health care or as stand-alone computer stations with touch screens without staff referral. Since it was started in the fall of 2006, the e-SBI has been successively implemented in primary health care in Östergötland County, Sweden. Results describing different aspects of the implementation phase have been reported previously [
The study was conducted in Östergötland County, Sweden, which had approximately 420,000 inhabitants during the study period (2007-2009). There were 42 operating primary health care units within the county when the study was performed. The units differed with regard to number of listed patients aged 18 years and over (average 9500, range 4200 to 16,500) and the number of general practitioners (GPs), nurses, and other staff members employed.
The number of primary health care units offering patients the e-SBI was successively extended during the study period from 10 units in 2007 to 28 units in 2009. The included units were situated in both urban and rural areas. The e-SBI was performed anonymously as part of routine health care. Patients performing the e-SBI during a two-year period, from September 2007 to August 2009 and who were not considered to be sufficiently physically active according to the results of the physical activity screening (see the physical activity section below) were included in the study. The patients were further divided into two groups. The first group consisted of patients who performed the e-SBI on their own initiative, hereafter referred to as the
Primary health care units participating in the study were provided with one or two sets of computers, monitors, and printers depending on enrolled patient population size; all were included in stand-alone, touch-screen information technology (IT) kiosks (
The e-SBI touch screen IT kiosk.
The physical activity measure in the e-SBI concept included two separate questions based on the American College of Sports Medicine/American Heart Association recommendation from 2007 [
After completing an e-SBI but before receiving a personalized printout, each patient was invited to participate in an optional follow-up mail survey 3 months later. Those who accepted this invitation were asked to register their national identification number at the end of the test, and they received a questionnaire by mail 3 months later. Mail addresses were retrieved from the Swedish population register. The mailed questionnaire included the same questions about moderate and vigorous physical activity as used at baseline. A reminder was sent 2 weeks after the follow-up questionnaire to those who had not returned the questionnaire.
Based on their response to the invitation to participate in the follow-up mail survey and completion of the follow-up questionnaire, patients were further categorized into three groups:
Since the data collection was performed as part of routine health care and the data consisted only of responses to a written questionnaire provided by patients who had given informed consent, there was, according to Swedish law, no need for formal ethical approval at the time at which the data collection was started. However, since then―in June 2008―the regulations were changed due to uncertainty about how to distinguish between routine and research data collection. For new studies involving similar data collection methods, ethical approval would now be required.
Baseline data from nonparticipants, nonresponders, and responders were compared to determine the representativeness of participants in the follow-up (responders). Pearson’s chi-square test was used to analyze differences in terms of sociodemographic characteristics. Also, mean (95% confidence interval [CI]) and median (interquartile range [IQR]) physical activity scores were compared (
Among responders, physical activity score and physical activity score category at baseline and 3-month follow-up were compared between patients who performed the e-SBI on their own initiative and those who were referred to it by primary health care staff. Created were four physical activity score categories: 0, 1 to 2, 3 to 4, and greater than or equal to 5. Pearson’s chi-square test was used to analyze differences in physical activity score category at baseline and at 3-month follow-up, together with comparison of mean (95% CI) and median (IQR) physical activity scores (
A total of 7863 patients completed the e-SBI during the two-year sampling period (
In the patient-initiated group, the proportion of older patients at baseline was significantly higher among responders compared with nonresponders and nonparticipants. However, there were no significant differences in gender distribution or physical activity score among the three groups (
In the staff-referred group, the proportion of men was significantly higher among nonresponders compared with the other groups (
Patient-initiated e-SBI: baseline characteristics of nonparticipants, nonresponders, and responders
Nonparticipants |
|
Nonresponders |
|
Responders |
|
||
|
|||||||
Men | 716 (51) | 35 (44) | 55 (43) | ||||
Women | 678 (49) | 44 (56) | 74 (57) | ||||
Total | 1394 (100) | .25 | 79 (100) | .89 | 129 (100) | .07 | |
|
|||||||
18-20 | 125 (9) | 10 (13) | 3 (2) | ||||
21-30 | 233 (17) | 19 (24) | 18 (14) | ||||
31-40 | 360 (26) | 23 (29) | 22 (17) | ||||
41-50 | 237 (17) | 6 (8) | 12 (9) | ||||
51-60 | 216 (16) | 10 (13) | 31 (24) | ||||
≥ 61 | 211 (15) | 11 (14) | 42 (33) | ||||
Total | 1382 (100) | .15 | 79 (100) | < .001 | 128 (100) | < .001 | |
|
|||||||
Mean (95% CI) | 1.9 (1.8-2.0) | 1.8 (1.5-2.2) | 2.1 (1.8-2.3) | ||||
Median (IQR) | 2 (0-3) | 2 (0-3) | 2 (1-3) |
Staff-referred e-SBI: baseline characteristics of nonparticipants, nonresponders and responders
Nonparticipants |
|
Nonresponders |
|
Responders |
|
||
|
|||||||
Men | 319 (53) | 78 (63) | 92 (51) | ||||
Women | 283 (47) | 45 (37) | 90 (50) | ||||
Total | 602 (100) | .04 | 123 (100) | .03 | 182 (100) | .61 | |
|
|||||||
18-20 | 54 (9) | 4 (3) | 5 (3) | ||||
21-30 | 72 (12) | 12 (10) | 11 (6) | ||||
31-40 | 82 (14) | 16 (13) | 21 (12) | ||||
41-50 | 108 (18) | 27 (22) | 26 (14) | ||||
51-60 | 123 (21) | 35 (29) | 56 (31) | ||||
≥ 61 | 156 (26) | 28 (23) | 63 (35) | ||||
Total | 595 (100) | .11 | 122 (100) | .18 | 182 (100) | < .001 | |
|
|||||||
Mean (95% CI) | 2.0 (1.9-2.2) | 2.1 (1.8-2.3) | 2.3 (2.1-2.5) | ||||
Median (IQR) | 2 (1-3) | 2 (1-3) | 3 (1-3) |
Flowchart of the recruitment of patients.
There was no statistically significant difference in physical activity score or physical activity score category between the patient-initiated and staff-referred groups at baseline or at the 3-month follow-up (
Physical activity score and category distribution at baseline and follow-up
|
Patient-Initiated |
Staff-Referred |
All Responders |
||
|
|||||
Mean (95% CI) | 2.1 (1.8-2.3) | 2.3 (2.1-2.5) | 2.2 (2.0-2.3) | ||
Median (IQR) | 2 (1-3) | 3 (1-3) | 2.5 (1-3) | ||
|
|||||
0 | 24 (19) | 31 (17) | 55 (18) | ||
1-2 | 49 (38) | 52 (29) | 101 (32) | ||
3-4 | 56 (43) | 99 (54) | 155 (50) | ||
≥ 5 | 0 (0) | 0 (0) | 0 (0) | ||
|
|||||
Mean (95% CI) | 4.1 (3.4-4.7) | 4.2 (3.7-4.8) | 4.2 (3.8-4.6) | ||
Median (IQR) | 3 (1-6) | 3 (2-6) | 3 (1-6) | ||
|
|||||
0 | 20 (16) | 21 (12) | 41 (13) | ||
1-2 | 26 (20) | 37 (20) | 63 (20) | ||
3-4 | 31 (24) | 40 (22) | 71 (23) | ||
≥ 5 | 52 (40) | 84 (46) | 136 (44) |
aThe physical activity score ranged between 0 and 18.9, and the cutoff for being sufficiently physically active was 5. No patients were categorized as sufficiently physically active at baseline according to inclusion criteria.
bχ2
2 = 3.99 (
cχ2
2 = 1.63 (
Change in physical activity score category from baseline to 3-month follow-up in all responders (n = 311)
Physical Activity Score Category at 3-Month Follow-up | |||||
Physical activity score |
n | 0 |
1-2 |
3-4 |
≥5 |
0 | 55 | 35 | 20 | 16 | 29 |
1-2 | 101 | 13 | 26 | 22 | 40 |
3-4 | 155 | 6 | 17 | 26 | 52 |
In this study, previously sedentary patients in primary health care improved their physical activity 3 months after performing an electronic screening and brief intervention (e-SBI). Overall, 44% of the patients became sufficiently physically active and the improvement in physical activity was similar when the e-SBI was patient-initiated or staff-referred.
These results suggest that the e-SBI is an effective method for promoting a short-term increase in physical activity in patients in primary health care. The e-SBI may be employed as a part of routine care, but there are several factors that need to be taken into account for implementation to be effective. These include staff expectations, perceived need for the innovation to be implemented, compatibility with existing routines, and implementation strategy [
The e-SBI could also be used as a stand-alone technique for promoting lifestyle behavior change, as it produces similar results without the involvement of primary health care staff. Posters informing the patients about the e-SBI can be placed in the waiting rooms. This would be an attractive, low-cost option for primary health care. Both patient-initiated and staff-referred e-SBIs may represent cost-effective complements to ordinary face-to-face interventions and may provide sufficient support to those patients who show better acceptance of this kind of technique. This may free up time for patients requiring face-to-face interventions. Although the initial costs of implementating the e-SBI might be high, the running costs, including technical support, would be far less than face-to-face counseling. The e-SBI would, therefore, deliver cost savings in the long run. Besides, the implementation costs for face-to-face interventions, including staff training, may also be high.
In the present study, there was a lower attrition rate at follow-up in the staff-referred group compared with the patient-initiated group. The extra attention/support experienced by the staff-referred group may have promoted continued participation in the study. However, this does not mean that all patients in the staff-referred group had sufficient motivation to improve their physical activity on their own. In the patient-initiated group, patients who remained in the study may have been those with higher internal motivation for behavior change. Hence, the extra attention/support provided to the staff-referred group versus the motivational characteristics of the remaining participants in the patient-initiated group may explain similarities in improvement in physical activity between the groups. The e-SBI (patient-initiated or staff-referred) may be adapted to meet the support needs of individual patients.
The results of the present study can be compared with those of our previous study of the effect of physical activity referral in routine primary health care in Östergötland County [
We are not aware of any comparable e-SBI physical activity study. Carroll et al performed a randomized controlled trial of a theory-based, computerized physical activity intervention in primary health care [
In the first phase of evaluating the effectiviness of the e-SBI in promoting improved behavior concerning physical activity (the present study) and alcohol consumption [
In the present study, a self-report measure of physical activity was used to assess change in physical activity following the intervention, as it is part of the e-SBI and may be the most feasible way of assessing physical activity in routine primary health care. However, self-report methods suffer from reporting bias, consisting of a combination of reactivity, recall bias, and social desirability [
There was a large attrition rate at follow-up in the present study, although patients participating in the follow-up were fairly representative of all patients who performed the e-SBI, reducing the risk of selection bias that may otherwise have affected the intervention outcome.
In conclusion, an electronic screening and brief intervention (e-SBI) implemented in routine primary health care improved physical activity for about half of the sedentary patients who agreed to participate in the follow-up. Similar results were obtained when the e-SBI was patient-initiated or staff-referred. The e-SBI may be a low-cost complement to lifestyle behavior interventions in routine primary health care and could work as a stand-alone technique not involving primary health care staff.
We are grateful to nutritionist Karolina Krus for providing the computers for the e-SBI to participating primary health care centers and for being responsible for on-site management of study questionnaires and data entry. Financial support was provided by the Swedish National Institute for Public Health.
Preben Bendtsen is a partner in a company that develops eHealth applications similar to the one described in this paper. The other authors declare no conflicts of interest.
confidence interval
electronic screening and brief intervention
general practitioner