Implementing a smoking cessation intervention for people experiencing homelessness: Participants' preferences, feedback, and satisfaction with the 'power to quit' program.

BACKGROUND: Smoking rates among populations experiencing homelessness are three times higher than in the general population. Developing smoking cessation interventions for people experiencing homelessness is often challenging. Understanding participant perceptions of such interventions may provide valuable insights for intervention development and implementation. We assessed participants' satisfaction and preferences for the Power to Quit (PTQ) program.
METHODS: PTQ was a 26-week community-based smoking-cessation RCT among people experiencing homelessness. A total of 315 of the 430 enrolled participants completed the 26 week-study feedback survey. Overall program satisfaction was measured on a 5-point Likert scale by asking the question "Overall, how satisfied were you with the Power to Quit Program?" Analyses were conducted to identify factors associated with overall program satisfaction.
RESULTS: Participants were mostly male (74.9%), African American (59.0%), 40 years and older (78.2%), and not married or living with a partner (94.9%). Visa gift cards were the most preferred incentive followed by bus tokens and Subway restaurant coupons. The patch and counseling were the top-ranked intervention component, 55.3% rated the patch as very helpful; 59.4% felt counseling sessions was very helpful; 48.6% found reminder phone calls or messages most helpful for appointment reminders. Majority (78.7%) said they were very satisfied overall, 80.0% were very satisfied with the program schedule, and 85.4% were very satisfied with program staff. Race and age at smoking initiation were predictors of overall program satisfaction. African American/Black participants were 1.9 times more likely to be satisfied with the program compared to White participants.
CONCLUSION: Majority of the participants of PTQ were satisfied with the program. This study supports the acceptability of a smoking cessation program implemented in a population experiencing homelessness. The high rate of satisfaction among African American participants may be in part because of race concordance between participants, study staff, and community advisory board. Including staff that have a shared lived experience with participants in a smoking cessation study may improve the participant satisfaction within such studies.

Introduction

Smoking remains a leading preventable cause of morbidity and mortality globally and in the United States [1]. Smoking is particularly high among the approximately 610,000 individuals experiencing homelessness in the U.S [2-5]. About 73%-80% of individuals experiencing homelessness are smokers compared to 14% of the general population [6-8]. Tobacco smoking increases the risk for obstructive lung disease, cardiovascular disease, cancer, and other chronic conditions [9, 10]. Homelessness exacerbates these risks and reinforces the critical need for smoking cessation interventions in populations experiencing homelessness [11-16]. However, there is a concern that the design and implementation of effective interventions may be hampered by competing priorities that occur commonly among smokers experiencing homelessness. For example, the need for mental health services, drug and alcohol services, food and shelter, and the transient nature of this population [17-19].

Participant satisfaction has long been recognized as an important element of health services research, and a report by the Institute of Medicine (National Academy of Medicine) identified satisfaction as a key indicator of quality of care [20]. Studies have found an association between participant satisfaction and program-related outcomes. In a study of Korean smokers, participants who were satisfied with Quitline services were more likely to maintain cessation for up to a year compared to those who were not satisfied (14.7% vs. 2.8%). Even among participants who relapsed, those who were satisfied with the intervention reduced the number of cigarettes smoked daily compared to those who were not (i.e., 37.6% versus 18.4%). In addition, participants who were satisfied with the contents of counseling and coaching protocols were more likely to quit smoking compared to those who were dissatisfied [21].

In a longitudinal study of 502 adults who began treatment for substance use, Carlson and Gabriel (2001) found that individuals who reported high levels of satisfaction with program services were two times more likely to abstain from substance use than those who reported a low level of satisfaction [22]. Others have also found positive associations between participant satisfaction and use of program services, longer treatment retention, and completion of program [23, 24]. These findings suggest that understanding participant satisfaction may not only be important for short-term outcomes such as retention and completion but also have implications for long-term outcomes such as cessation and/or relapse.

Few studies have examined participant satisfaction with smoking cessation programs among populations experiencing homelessness. This paper uses data from the Power to Quit study to assess participants’ preferences for various aspects of the smoking cessation program, their opinions of the main components of the program, and their satisfaction with program components. In addition, we identify factors associated with greater program satisfaction among smokers experiencing homelessness. Results from this study may be helpful to others designing smoking cessation programs for populations experiencing homelessness.

Materials and methods

The power to quit program

The Power to Quit program (Trial Registration Number: NCT00786149) was a randomized-controlled trial of smokers experiencing homelessness in Minneapolis and St. Paul, Minnesota. A total of 430 participants were randomized into two groups. At baseline, both arms received a two-week supply of 21-mg NRT patches. Over an eight-week period, both arms received an additional two-week supply of the nicotine patch every two weeks. In addition, the intervention group received six 15–20-minute participant-led counseling sessions (motivational interviewing) while the control group received standard care (one brief counselor-led session lasting 10–15 minutes). The primary aim of the intervention was to evaluate the efficacy of motivational interviewing (MI) as well as nicotine replacement therapy (NRT) in smokers experiencing homelessness [25]. Our study assessed participants’ satisfaction with the patch and counseling sessions for both groups.

The study lasted 26 weeks and in total there were 15 counseling and retention visits. At each visit, participants received incentives as compensation for their time and effort. At longer visits that included surveys, participants received $20 gift cards and two bus tokens ($3 value). For attending brief retention visits and the week 8 end of treatment visit, participants received $10 gift cards, two bus tokens, and another small gift item. Small gift items included playing cards, tote bags, movie passes, water bottles, T-shirts, and personal care items (e.g., soap, toothbrush, washcloth). For attending the final 6-month visit, participants received a $40 gift card and a sweatshirt. For participants who attended all 15 sessions, the monetary incentives totaled $275 over the 6-month study period.

Participants

Eligible participants were recruited over 15 months (May 2009 to August 2010) from eight emergency homeless shelters and transitional homes in the Twin Cities (Minneapolis and St. Paul, Minnesota). Study sites were located in the downtown/city center easily accessible by public transportation such as city-operated buses and light rail. Researchers recruited participants by conducting health fairs, holding informational interviews, posting flyers, and announcements at homeless shelters. Participants also helped to recruit by word of mouth. Inclusion criteria was a confirmation of homelessness as defined by the Stewart B. McKinney Act [26] as ‘any individual who lacks a fixed, regular and adequate nighttime residence’; or ‘one whose primary nighttime residence is a supervised publicly or privately-operated shelter designed to provide temporary living accommodations, transitional housing, other supportive housing program or a public or private place not meant for human habitation.’ Participants were also classified as experiencing homelessness if they were without a home and had been staying with family or friends for up to three months [26, 27]. Individuals were also included if they 1) were a current smoker who had smoked at least 100 cigarettes in their entire lifetime; 2) reported smoking at least 1 cigarette every day over the past 7 days and CO score above 5; 3) ≥ 18 years of age; 4) lived in the Twin Cities for at least 6 months and planned to stay for the following 6 months of the study; 5) wished to use the nicotine patch for 8 weeks and 6) were willing to complete 15 total appointments over the 26-week study period. Participants were excluded if they 1) used a tobacco quit aid in the previous 30 days; 2) had a cognitive impairment; 3) had suicidal ideation in the last 14 days, or 4) had a major medical condition within the previous 30 days. Participants were also excluded if they scored greater than five on items assessing psychotic symptoms. Full details of the study design have been published [27].

Data collection

Validated questionnaires were used to collect survey data. Metric measurements of height and weight were collected to calculate body mass index. Carbon monoxide (CO) and saliva cotinine were assessed as biomarkers of tobacco use. The participants were asked to exhale into a carbon monoxide monitor. Patch adherence was measured by: ‘patch checks’ (visual verification of whether a participant was wearing a patch); ‘patch counts’ (documenting the number of patches left in the participant’s possession); and administration of the Morisky scale, a self-reported adherence scale modified to assess adherence to NRT patch.

Study data was collected at baseline, week 1, week 2, week 4, week 6, week 8, week 10, week 12, week 14, week 16, week 18, week 20, week 24, and week 26. Because of the nature of the study population, timings for assessments and counselling sessions were flexible.

Data analyzed in this study was limited to the baseline and week 26 feedback survey. At the baseline survey, information collected included: demographic characteristics, housing, general health status, smoking history, quitting history, cigarette accessibility, nicotine dependence, confidence in quitting smoking, self-efficacy, mental health measures including depression and perceived stress, alcohol use, drug abuse/dependence, exhaled carbon monoxide, height, weight, and cotinine. At the 26-week feedback survey, information obtained included: data on preferred program components and incentives, adherence to nicotine replacement, difficulties experienced with using the patch, reasons for missing appointments, what helped the most as appointment reminders, motivations for keeping appointments, helpfulness of counseling sessions and program satisfaction. Outcome data such as smoking characteristics, exhaled CO, weight, and cotinine were repeated at the 26-week survey. Some of the information collected in weeks 2, 4, 6, and 8 surveys included information on housing, adverse effects, tobacco cessation, patch adherence, confidence in quitting smoking and exhaled CO.

Measures

Outcome variable

Overall program satisfaction was our key dependent variable. This was measured on a 5-point Likert scale. Participant satisfaction was measured by asking: “Overall, how satisfied were you with the Power to Quit Program?” The response options were: 1)Not satisfied at all; 2) Somewhat unsatisfied; 3) Neutral; 4) Somewhat satisfied; 5) Very satisfied. In univariate and multivariable analysis, the outcome variable overall satisfaction was dichotomized and given a value of 1 if “Very satisfied” and a value of 0 for all other responses.

Exposure variables

The main components of the intervention were the nicotine patch and counseling sessions. Incentives for participation included Visa gift cards, bus tokens, calendar/organizers, tote bags, restaurant coupons, movie passes, polo shirts, water bottles, soap and washcloths, back massagers, and a dental package. Preferences for program components were assessed by asking participants: “Which items were the most helpful to you?” Participants were asked to number program components on a scale of 1 to 3 in the order that they were most helpful -1(most helpful), 2(helpful), 3(least helpful). The response options were: 1) Reading materials; 2) Individual counseling sessions; 3) Nicotine patch; 4) Community mobilizer contacts; 8)Don’t know/don’t remember. Community mobilizers are research assistants on the study team who were either homeless at the time of the study or had recently experienced homelessness. Similarly, preferences for program incentives were assessed by asking participants; “Choose three of the items that you liked the most and number them from 1 to 3 in the order of preference -1(liked the most), 2(liked), 3(liked the least). The response options were:1) Bus tokens; 2) Tote bag; 3) Calendar/organizer; 4) Visa gift cards; 5) Restaurant coupons (Subway); 6) Movie passes; 7) Polo shirt; 8) Water bottle; 9) Soap and washcloth; 10) Back massager; 11) Dental package; 88)Don’t know/don’t remember. The first ranked choice of each respondent was used for our analysis. The “Don’t know/don’t remember” responses were dropped from the analysis.

Socio-demographic covariates

Respondents ages were recoded into five groups i.e., < 30 years, 30–39 years, 40–49 years, 50–59 years, ≥60 years. Race/Ethnicity was recoded as African American/Black, White, and Other. Marital Status was recoded as married/living with significant other, divorced/widowed/separated, and never been married. Education was recoded as less than high school and at least high school education. Employment was recoded as currently employed and currently unemployed. Monthly income was recoded as <$400, ≥$400.

Health and psychosocial covariates

Self-reported general health was assessed by asking: “In general, would you say your health is?” The response options were: 1) Excellent; 2) Very good; 3) Good; 4) Fair; 5) Poor. Depression was assessed with PHQ-9. Respondents were asked how often they have been bothered by a range of nine problems over the preceding 2 weeks. Response options for each question were: 0) Not at all; 1) Several days; 2) More than half the days; 3) Nearly every day. Responses were summarized and respondents with scores ranging from 0–4 were categorized as having no depression, scores of 5–9 were categorized as mild depression, scores of 10–19 were categorized as moderate depression, and scores of 20–27 were categorized as severe depression [28]. Stress was assessed using the perceived stress scale. Respondents were asked four questions to assess how often they experienced stress in their life in the past 30 days. Response options were: 0) Never; 1) Rarely, 2) Sometimes; 3) Often; 4) Very often. Responses were summarized and respondents with scores ranging from 0–5 were categorized as low perceived stress, scores from 6–10 were categorized as moderate perceived stress, and scores from 11–16 were categorized as high perceived stress [29, 30].

Homelessness covariates

The number of times participants experienced homelessness in the past 3 years was assessed by asking: “During the last three years, how many separate times have you been homeless or without a regular place to live?” Responses were recoded as once, twice, and three or more times.

Smoking-related covariates

Age at smoking initiation was grouped as ≤10 years, 11–20 years, 21–30 years, 31–40 years, and ≥41 years. Time to first cigarette was assessed by asking respondents: “How soon after you wake up do you smoke your first cigarette?” Responses were dichotomized as ≤5 minutes and >5 minutes. The number of past-year 24-hour quit attempts was assessed by asking: “In the last year, on how many times have you seriously tried to quit smoking for at least 24 hours?” This was recoded as 1–10 times, 11–20 times, 21–30 times, 31–40 times. Confidence to quit smoking was assessed by asking a series of numerically scored questions with higher responses indicating higher levels of confidence. Scores were reported in means and standard deviation.

Substance use and dependence covariates

Drug and alcohol dependence was assessed using the 3-item Rost-Burnam screener. Drug dependence was measured by asking participants: "Have you ever used one of these drugs on your own more than 5 times in your lifetime?", "Did you ever find you needed larger amounts of these drugs to get an effect or that you could no longer get high on the amount you used to use?”, “Did you ever have emotional or psychological problems from using drugs-like feeling crazy or paranoid or depressed or uninterested in things?” Response options were: 0) No; 1) Yes. Scores were summed and a score of 0 was categorized as not dependent and a score of ≥1 was categorized as dependent. Alcohol dependence was measured by asking participants: “Did you ever think that you were an excessive drinker?”, “Have you ever drunk as much as a fifth of liquor in one day?”, “Has there been a period of two weeks when every day you were drinking 7 or more beers, 7 or more drinks or 7 or more glasses of wine?”. Response options were: 0) No; 1) Yes. Scores were summed and a score of 0 was categorized as not dependent and a score of ≥1 was categorized as dependent.

Statistical analysis

Pearson’s chi-square test and Fisher’s exact test (as appropriate) were used to examine bivariate associations between study variables. Logistic regression analyses were used to examine the associations between overall satisfaction and program components and incentives. In multivariable analysis, variables significant at p<0.10 in the bivariate analysis were imputed in the logistic regression in order to include variables tending towards a positive significance [31]. In addition, we controlled for age, race/ethnicity, and gender as previous studies have shown that greater program satisfaction is associated with demographic characteristics [32, 33]. All of the statistical analyses were performed using Stata 16.0 and p-values <0.05 were considered statistically significant at multivariable analysis [34].

Ethical considerations

Study procedures were approved and monitored by the Institutional Review Board of the University of Minnesota Medical School (Study Number: 1307M39761). Written informed consent was obtained from participants prior to data collection.

Results

Sociodemographic, health, and substance use characteristics

Of the 430 participants enrolled in the study, 315 (73.3%) completed the week-26 feedback survey of which the majority were male (74.9%), African American/Black (59%), and 40 years or older (78.2%). More than half of the respondents (51.8%) were never married and 76.5% were high school graduates or had a General Educational Development (GED) qualification. Majority were unemployed (90.2%) and most (67.9%) had a monthly income of less than $400. Regarding their health, about a third (32.3%) reported having good health and an over half (57.2%) reported no depression or mild depression. Most respondents (77.4%) started smoking regularly between the ages of 11 and 20. For almost half of the respondents (46.4%), time to first cigarette was ≤5 minutes and many respondents (55.5%) had at least one unsuccessful 24-hour quit attempt in the past year. On a scale of 0 (not confident) to 10(extremely confident), the mean confidence to quit was 7.30 (2.41). About 10% of the respondents abstained from cigarette at the end of the study. More than half of the respondents (59.4%) had a positive screen for alcohol dependence, and majority (84.1%) had a positive screen for drug dependence (Table 1).

Table 1

Respondents’ demographic, health, psychosocial and smoking related characteristics.

	Frequency	Percentage
Demographic characteristics
Age (mean (SD)) (n = 314)	45.7(9.8)
Gender (n = 315)
Male	237	75.5
Female	78	24.8
Race/Ethnicity (n = 315)
White	106	33.7
African American/Black	186	59.0
Other#	23	7.3
Marital Status (n = 313)
Married/Living with significant other	16	5.0
Divorced/Widowed/Separated	135	43.1
Never been married	162	51.8
Highest Level of Education (n = 315)
< High school	74	23.5
≥ High school	241	76.5
Employment (n = 315)
Currently employed	31	9.8
Currently unemployed	284	90.2
Monthly income (n = 315)
<$400	214	67.9
≥$400	101	32.1
Homelessness characteristic
Number of times homeless in past 3 years (n = 313)
Once	124	39.6
Twice	83	26.5
Thrice or more	106	33.9
Health and psychosocial characteristics
Self-reported general health (n = 313)
Excellent	48	15.3
Very Good	89	28.4
Good	101	32.3
Fair	61	19.5
Poor	14	4.5
Depression PHQ9 ≥10 (n = 313)
None	97	31.0
Mild	82	26.2
Moderate	63	20.1
Moderately severe	50	16.0
Severe	21	6.7
Mean (SD)	8.97 (6.54)
Stress PSS4, past 30 days (n = 313)
Low stress	34	10.9
Moderate stress	235	75.1
High stress	44	14.0
Mean (SD)	2.03 (0.5)
Smoking-related characteristics
Age started smoking regularly (n = 314)
≤ 10 years	26	8.3
11–20 years	243	77.4
21–30 years	33	10.5
31–40 years	7	2.2
≥ 41 years	5	1.6
Mean (SD)	16.6 (6.2)
Time to first cigarette (n = 315)
≤5 minutes	146
>5 minutes	169	53.6
Number of 24-hour quit attempts past year (n = 310)
0 times	126	40.7
1–10 times	172	55.5
11–20 times	7	55.5
>20 times	5	2.3
mean (SD)	2.5 (4.9)
Confidence to quit, (mean(SD)) (n = 315)	7.3 (2.4)
Substance abuse variables
Had a positive screen for alcohol dependence (n = 313)
No	50	40.6
Yes	186	59.4
Had a positive screen for drug dependence (n = 314)
No	50	15.9
Yes	264	84.1

#Other categories include America Indian/Alaskan Native, Hispanic, Filipino, Jewish, Indian, Irish, Multi-racial

Preferred program components and program incentives

The patch was the top-ranked component followed by counseling sessions (Fig 1).

Fig 1

Participants preference for study components.

Of the incentives offered to participants for participation, Visa gift cards were the most preferred incentive followed by bus tokens and Subway restaurant coupons (Fig 2).

Fig 2

Participants preference for study incentives.

Participants’ feedback on the program components

Table 2 shows the participants feedback on the program components. Over half of the respondents (55.3%), rated the patch as very helpful. Of those that encountered some difficulties using the patch, the most common reason was remembering to put it on every day (26.6%). Other challenges included burning/itching side effect (21.0%) and difficulty sleeping (16.1%). Although less than half of the respondents continued the patch after the study ended (45.7%), 81.6% said they would have continued if provided. Similarly, more than half of the participants (59.4%) felt the counseling sessions were very helpful. Among those that missed appointments, the most common reason for missing appointments was that they forgot (39.7%). Reminder phone calls or messages (48.6%) and the desire to quit smoking (43.6%) motivated most participants to keep their appointments. Furthermore, nearly two-thirds (63.9%) of participants indicated they would like to continue with counseling.

Table 2

Participants’ feedback on the patch, appointments, and counseling sessions.

Variable	Frequency	Percentage
Patch
Number of weeks of self-reported adherence to patch Median (IQR) (n = 310)	6 (6–8)
Proportion of patch used (n = 314)
None	2	0.6
Less than one half	19	6.1
About one half	41	13.1
More than one half	39	12.4
All or nearly all	212	67.5
Don’t know/ Don’t remember	1	0.3
Experienced difficulties with using the patch
Yes	124	39.4
No	191	60.6
Primary difficulties encountered with the use of patch (n = 124)
Remembering to put it on every day	33	26.6
Side effect: Burning and/or itching	26	21.0
Side effect: Difficulty sleeping	20	16.1
Did not reduce urge to smoke	14	11.3
Side effect: Nausea and/or vomiting	12	9.7
Carrying around/Not having a place to store them	11	8.9
Finding a clean spot on my skin	8	6.4
Continued patch after study ended (n = 313)
Yes	144	45.7
No	169	53.99
Would have continued if given (n = 313)
Yes	257	81.6
No	169	53.99
Rated helpfulness of the patch (n = 313)
Very unhelpful	26	8.3
Unhelpful	9	2.9
Not sure how helpful	32	10.2
Helpful	73	23.3
Very helpful	173	55.3
Appointments
Primary reason for missing appointments (n = 189)
Forgot	75	39.7
Work	31	16.4
Another appointment	22	11.6
No transportation	18	9.5
Out of town	14	7.4
Sick	10	5.3
In the hospital	6	3.2
Looking for housing	5	2.7
School or other classes	5	2.7
Job hunting or interview	1	0.5
Had not quit smoking	1	0.5
Don’t feel like it, not in the mood to go	1	0.5
What helped the most as appointment Reminders (n = 315)
Reminder phone calls or messages	153	48.6
Paper reminder clips	59	18.7
I didn’t need help with this	20	6.3
Face to face reminders	13	4.1
All of them	38	12.1
None of them	5	1.6
Other	27	8.6
Primary motivations for keeping appointments (n = 314)
Wanted to quit smoking	137	43.6
Incentives (other than bus passes)	98	31.2
Liked the staff	27	8.60
Responsibility to see the counsellor	10	3.18
Community mobilizer encouragement	8	2.55
Enjoyed meetings/ sessions	8	2.55
Bus passes	5	1.59
Nothing else to do	3	0.96
Learned new information	1	0.32
Other	17	5.41
Counseling
Perceived helpfulness of counseling sessions (n = 313)
Very unhelpful	22	7.0
Unhelpful	4	1.3
Not sure how helpful	27	8.7
Helpful	74	23.6
Very helpful	186	59.4
Would like to continue with counseling (n = 313)
Yes	200	63.9
No	100	32.0
I don’t know	13	4.1

Program satisfaction

Majority of participants (78.7%) said they were very satisfied with the overall program Majority also said they were very satisfied with the location (81.3%) and with the counselors (86.4%). Most of the respondents (80.0%) were very satisfied with the program schedule, i.e., appointment times, frequency, and length of sessions. Majority (85.4%) were very satisfied with the program staff. On a scale of 0(least) to 5(most), 4.9(0.4) reported feelings of being treated with respect and feeling that the research was adequately explained (Table 3).

Table 3

Program satisfaction among respondents.

Variable	Frequency	Percentage
Overall program satisfaction
Not satisfied at all	2	0.7
Somewhat unsatisfied	1	0.3
Neutral–don’t feel strongly either way	13	4.1
Somewhat satisfied	51	16.2
Very satisfied	248	78.7
Satisfaction with program location
Not satisfied at all	7	2.2
Somewhat unsatisfied	5	1.6
Neutral–don’t feel strongly either way	14	4.4
Somewhat satisfied	33	10.5
Very satisfied	256	81.3
Satisfaction with program counselors (n = 314)
Not satisfied at all	5	1.6
Somewhat unsatisfied	2	0.6
Neutral–don’t feel strongly either way	11	3.5
Somewhat satisfied	24	7.6
Very satisfied	272	86.4
Satisfaction with the schedule, i.e., appointment times, frequency and length of sessions
Not satisfied at all	2	0.6
Somewhat unsatisfied	7	2.2
Neutral–don’t feel strongly either way	9	2.9
Somewhat satisfied	45	14.3
Very satisfied	252	80.0
Satisfaction with staff
Not satisfied at all	2	0.6
Somewhat unsatisfied	1	0.3
Neutral–don’t feel strongly either way	10	3.2
Somewhat satisfied	33	10.5
Very satisfied	269	85.4
Feelings of trust in staff
1-Least	2	0.6
2	2	0.6
3	7	2.2
4	30	9.5
5–Most	274	87.0
Mean (SD)	4.8 (0.6)
Felt comfortable asking questions
1-Least	0	0.0
2	1	0.3
3	5	1.6
4	25	7.9
5–Most	284	90.2
Mean (SD)	4.9 (0.4)
Felt treated with respect
1-Least	1	0.3
2	2	0.6
3	3	0.9
4	14	4.4
5–Most	295	93.7
Mean (SD)	4.9 (0.4)
Felt research was adequately explained
1-Least	3	1.0
2	1	0.3
3	2	0.6
4	19	6.0
5–Most	290	92.1
Mean (SD)	4.9 (0.5)
Felt information provided was kept confidential
1- Least	2	0.6
2	0.0	0.0
3	13	4.1
4	18	5.7
5–Most	282	89.5
Mean (SD)	4.8 (0.6)
Felt attention was given to special needs
1-Least	1	0.3
2	4	1.3
3	7	2.2
4	24	7.7
5–Most	277	88.5
Mean (SD)	4.8 (0.6)

Predictors of program satisfaction

Race/ethnicity, age started smoking and confidence to quit showed statistically significant association with participants satisfaction at p<0.10 (Table 4). Race was a predictor of overall programme satisfaction. African American/Blacks were 1.84 times more likely to be very satisfied with the program compared with White or other participants. Age at smoking initiation was also a predictor of participants satisfaction; participants who started smoking at a younger age were more likely to be very satisfied with the program (Table 5).

Table 4

The relationships between program satisfaction and other variables.

	Very Satisfied	Not very satisfied	Statistic	p-value
Demographic characteristics
Age (mean ±SD) (n = 314)	45.8±9.7	45.4±10.3	0.26T	0.794
Gender (n = 315)
Male	185(78.1)	52(21.9)	0.26	0.612
Female	63(80.8)	15(19.2)
Race/Ethnicity (n = 315)
White	76(71.7)	30(28.3)	5.80*	0.055
African American/Black	155(83.3)	31(16.7)
Other#	17(73.9)	6(26.1)
Marital Status (n = 313)
Married/Living with significant other	12(75.0)	4(25.0)	0.21*	0.900
Divorced/Widowed/Separated	106(78.5)	29(21.5)
Never been married	129(79.6)	33(20.4)
Highest level of Education (n = 315)
< high school	61(82.4)	13(17.6)	0.79*	0.374
≥ high school	187(77.6)	54(22.4)
Employment (n = 315)
Currently employed	26(83.9)	5(16.1)	0.54*	0.461
Currently unemployed	222(78.2)	62(21.8)
Monthly income (n = 315)
<$400	164(76.6)	50(23.4)	1.75*	0.186
≥$400	84(83.2)	17 (16.8)
Homelessness characteristic
Number of times homeless in past 3 years (n = 313)
Once	100(80.7)	24(19.3)	0.41*	0.815
One to three times(twice)	64(77.1)	19(22.9)
More than three times (≥3)	83(78.3)	23(21.7)
Health and psychosocial characteristics
Self-reported general Health (n = 313)
Excellent	38(79.2)	10(20.8)	0.81*	0.938
Very Good	68(76.4)	21(23.6)
Good	80(79.2)	21(20.8)
Fair	49(80.3)	12(19.7)
Poor	12(85.7)	2(14.3)
Depression PHQ9 ≥10 (mean±SD) (n = 313)	9.1±6.7	8.4±6.0	0.86*	0.391
Stress PSS4, past 30 days, (mean±SD)	2.0±0.5	2.0±0.5	-0.15T	0.603
Smoking-related characteristics
Age started smoking regularly (mean ±SD) (n = 314)	16.2±5.7	17.8±7.5	-1.80T	0.073
Time to first cigarette
≤5 minutes	117(80.1)	29(19.9)	0.32*	0.571
>5 minutes	131(77.5)	38(22.5)
Number of 24 hour quit attempts past year, (mean±SD) (n = 310)	2.7±5.3	1.8±3.0	1.27T	0.206
Confidence to quit, (mean±SD) (n = 315)	7.4±2.5	6.8±2.2	1.88T	0.060
Substance abuse variables
Had a positive screen for alcohol dependence (n = 313)
No	104(81.9)	23(18.1)	1.38*	0.240
Yes	142(76.3)	44(23.7)
Had a positive screen for drug dependence (n = 314)
No	35(70.0)	15(30.0)	2.66*	0.103
Yes	212(80.3)	52(19.7)
Top Ranked component(n = 315)
Reading materials	25(75.8)	8(24.2)	1.50*	0.681
Counselling sessions	73(78.5)	20(21.5)
The patch	117(81.3)	27(18.7)
Community mobilization	33(73.3)	12(26.7)
Top Ranked incentive(n = 315)
Visa gift cards	155(79.90)	39(20.10)	1.57*	0.456
Bus tokens	47(81.03)	119(18.97)
Others*	46(73.02)	17(26.98)

*Chi-square

T independent T-test

Overall satisfaction was dichotomized. not satisfied at all, somewhat unsatisfied, neutral and somewhat satisfied were coded as 0 (not very satisfied). Very satisfied was coded as 1(very satisfied)

#Other categories include America Indian/Alaskan Native, Hispanic, Filipino, Jewish, Indian, Irish, Multi-racial

Table 5

Predictors of program satisfaction among the respondents.

Variable	AOR(95% CI Lower limit, Upper limit)	p-value
Age (years)	1.005(0.977, 1.035)	0.695
Gender
Female(ref)
Male	0.832(1.035, 1.611	0.585
Race
White (ref)
African American/Black	1.847(1.027, 3.320)	0.040 *
Others#	1.018(0.356, 2.908)	0.974
Age at smoking initiation	0.957(0.919, 0.998)	0.040 *
Confidence to quit smoking	1.108(0.989, 1.243)	0.075

Pseudo R2 = 0.0370

Hosmer Lemeshow goodness of fit X^2 = 10.70 p = 0.220

#Other categories include America Indian/Alaskan Native, Hispanic, Filipino, Jewish, Indian, Irish, Multi-racial

*Significant at p<0.05

Discussion

This paper evaluates participants’ preferences and satisfaction with the components of a smoking cessation intervention among smokers experiencing homelessness. While participants were very satisfied with the Power to Quit program, being African American was a significant predictor of overall program satisfaction. The high satisfaction among African American participants in this program may be as a result of the involvement of racially concordant research staff in the design and implementation of the intervention. One of the two counselors that provided motivational interviewing was African American, two African Americans who had experienced homelessness in the past were recruited as participant mobilizers. Additionally, the members of the Community Advisory Board (CAB) included members who were familiar with the needs and desires of homeless people. Saha et al. found that provider-patient racial concordance can influence satisfaction with health care among African American and Hispanic populations [35]. Similarly, racial concordance in addition to shared experiences of homelessness may have positively influenced the study implementation and by extension, program satisfaction among the African American participants. We observed that age at smoking initiation may also be associated with overall program satisfaction. Younger age at smoking initiation has been associated with an increased likelihood of relapse [36]. This implies that adult smokers who started smoking at an earlier age might have more quit attempts and therefore may be more responsive to smoking cessation programs.

Among smokers experiencing homelessness, financial incentives may increase smoking abstinence and quit attempts [37]. In our study, Visa cards, bus tokens and Subway restaurant coupons were the most preferred program incentive; likewise, the patch was rated as the most preferred program component. These findings are not surprising given that individuals experiencing homelessness often have limited access to food, transportation, employment, and health insurance [38]. We observed that participants’ rankings of the various program components and incentives were not significantly associated with overall participant satisfaction. There may be several reasons as to why a significant association was not observed. First, approximately 25% of those initially enrolled in the study did not complete the week-26 feedback survey therefore subject attrition may have impacted the power to detect differences. Conversely, in previous studies, participants experiencing homelessness indicated high motivation and readiness to quit smoking [39, 40]. Therefore, incentives or specific components of the program may not have significantly affected participant motivation to participate in and complete the study, and subsequently their overall satisfaction with the program. Although most of the participants rated the patch as being ‘very helpful,’ many of the participants admitted that they often forgot to use the patch; phone calls and message reminders by the program staff were reported to be helpful. An understanding that smokers experiencing homelessness tend to have several competing priorities and building in novel ways to mitigate the effects of competing priorities into the design of smoking cessation programs for smokers experiencing homelessness should be considered in future iterations.

Limitations

This is one of the first studies to assess program satisfaction and participants preferences for the components of a smoking cessation program among smokers experiencing homelessness; however, it has some limitations. First, the sample was a convenience sample of individuals experiencing homelessness in the Twin Cities and may not be generalizable to persons experiencing homelessness living in other cities. Future studies may wish to assess program satisfaction in other parts of the United States. Second, the survey responses may reflect some level of social desirability bias. Individuals, especially those who are low-income and those with low educational attainment, as may be the case with persons experiencing homelessness, tend to respond favorably regardless of content [41, 42]. Our research team constituted persons, who had experiences with homelessness either directly or indirectly; having individuals who could identify with the participants’ background may have attenuated but not eliminated participants’ tendency to give favorable responses. Third, non -response bias which arises when respondents are systematically different from non-respondents was addressed by instituting measures to minimize attrition. Daily reminder calls were made to participants during the week prior to appointments, until the window for completing appointments was closed. Reminder slips were also given to the respondents at the time of setting the appointment. Fourth, we did not explore the possible influence racial concordance may have played in participant’s satisfaction. This should be considered in future research. Fifth, results are based on cross-sectional data and therefore we make no claims on causality based on the statistical design of the study. Sixth, our study did not examine the association between program satisfaction and program outcomes, primarily because the number of successful quitters in this study was too small to yield any meaningful statistically significant differences. Previously published results of the effectiveness of this intervention, adding motivational interviewing counseling to nicotine patch for smoking cessation among this population found no significant differences in verified seven-day abstinence rate at the end of follow up between the intervention group and the control group [43]. Future studies should assess this association to determine if program satisfaction impacts both short-term outcomes (i.e. retention) and long-term outcomes (i.e. cessation). Seventh, the instrument used to assess satisfaction has not been tested for reliability or validity in this population or any other population. Future studies should be aimed at creating a valid and reliable instrument to assess satisfaction in populations experiencing homelessness as their perceptions of care may be shaped by their experiences of homelessness and may be uniquely different from the general population [44]. Finally, 45.7% of participants continued to use the nicotine patch after the study concluded even though 81% of participants reported they would have continued to use the nicotine patch if they were given more. At the conclusion of the study, unused nicotine patches were donated to the medical clinic located within the largest shelter that served as a study site. Direct access to NRT has huge implications for long term smoking cessation efforts among populations experiencing homelessness. Unfortunately, logistical and procedural efforts to ensure NRT adherence once the study ended proved to be challenging. The mobile nature of this population made it difficult to monitor participant access to NRT and other quit aids once the study concluded. Future studies should consider ways to connect populations who experience homelessness to smoking cessation programs not only in the immediate area but the wider metropolitan area.

Conclusions

Visa cards and bus tokens seem to be the preferred incentives for participation among this group of smokers. Participants preferred the nicotine patch to the counseling sessions; however, reminders for consistent patch use were needed Preferences for program incentives were unrelated to overall program satisfaction. Satisfaction has implications for retention and long-term outcomes therefore future smoking cessation programs for people experiencing homelessness should be designed to enhance satisfaction.

14 Oct 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Marianne Clemence

Associate Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please ensure you have included the registration number for the clinical trial referenced in the manuscript.

3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper presents data on homeless participants’ satisfaction and preferences for the Power to Quit study from a survey conducted at 26 weeks. The questions asked in the survey are similar to a process evaluation of a trial which can be useful to help inform future studies and interventions. The resulting data will be useful feedback to those involved in the PTQ study, however, I have some concerns about the methods and whether the findings provide enough insight to make a significant contribution to the literature or for others designing future interventions with homeless populations. Additionally,

The data appears to be based on a 26-week feedback survey. Currently there is little description of the survey in the methods section. How were participants invited to take part, what was the method of administration/data collection, was there a participant incentive for completing the survey, how long was the survey? The methods would also benefit from a full description of what the survey covered. The results suggest that more was asked than is currently outlined in the methods.

Was all the data used in the paper collected at the 26-week survey? It would be useful to clarify this, i.e. was the demographic and other data collected at 26-weeks or was this collected earlier in the trial? This is important to highlight because some of the smoking data, if collected at 26 weeks, will likely have been influenced by participation in the trial, for example the % reporting an unsuccessful quit attempt in the last year and the high confidence to quit score could have been a direct result of the trial, and not therefore suggestive of homeless populations, and should be noted.

How was the survey (and importantly, the questions) developed? Were they tested for participants’ understanding and relevance? There are potential issues with some of the main questions asked. For example, the question, ‘Which items were most helpful?’ This question is open to interpretation. Did the authors mean which items were most helpful for reducing smoking or quitting? The inclusion of ‘reading materials’ or ‘community mobile contacts’ suggest that the question could also have been interpreted as ‘most helpful sources of information?’

The monetary/voucher incentives were particularly valued by homeless study participants. This is unsurprising, although they were not linked with programme satisfaction. The authors write that “incentives or specific components of the programme may not have significantly affected participant motivation to participate in and complete the study”. What were the incentives in the study used for, e.g. enrolment, attending follow-up appointments, motivation for smoking abstinence? This needs to be clarified as currently the paper doesn’t tell us very much about how incentives were used in the trial, and therefore the implications for future interventions are limited.

Reviewer #2: Thank you for the opportunity to review this paper. The manuscript describes satisfaction with the Power to Quit smoking cessation intervention in Minneapolis and St. Paul, Minnesota. I thought the manuscript was interesting and had practical recommendation for people conducting cessation trials with people experiencing homelessness. I did think that findings could probably be summarized as a brief report if that was an option. I have several other suggestions below:

1) I might consider changing the language to reflect people or populations experiencing homelessness or people experiencing homelessness who smoke as an indicator that this population is constantly changing and is dynamic based on economic and other circumstances, and move away from calling the population, "the homeless" or "homeless smokers" which signifies a static condition and potentially a character trait (which we know it is not). This might be viewed as less stigmatizing. I would suggest making this change throughout the manuscript.

2) In the abstract, I might include how satisfaction was qualitatively assessed. I might bring in the point that the rate of satisfaction was high among African American participants in part because there may have been race concordance between study staff, community advisory board and AA participants. I think that is the salient point here that it is helpful to have staff that have lived experiences of participant in the study to improve the effectiveness of the study.

3) Methods: I might include where the study sites were in Minneapolis and St. Paul - I recall they were in shelters in these cities. Did participants have to commute? What about timings for the assessment and counseling sessions -- were they flexible?

4) The authors alluded to this but was the validity of the satisfaction question in this sample -- could the authors conduct an internal validity of this question in this sample?

5) Page 7, Line 150 -- which item was most helpful to you -- is this in relation to smoking cessation or general life?

6) The authors described many covariates, e.g., health and psychosocial covariates, smoking-related covariates, substance use and dependence covariates -- while these are interesting, I wonder how much is relevant to the analysis on satisfaction as none of these variables are controlled for in the model. I might consider taking these out of this paper and referring to the original Power to Quit study findings for descriptive statistics. It does not seem that satisfaction differed based on mental health or substance use based on these findings and if so, might clarify and explain why that might be in the results and discussion.

7) A few questions on the results/discussion:

- I might consider discussing in the results, how satisfaction varied for participants who were abstinent at 6 months vs. not; for participants in the intervention vs control (I did not see this distinction and it seems relevant)

- one of the participants' feedback was on continuing patch and counseling if that option was available to participants -- and it has implications for extended duration interventions for PEH, perhaps it might be good to discuss this point in the discussion.

- when I read that African American/Black participants had higher satisfaction levels, I immediately thought this was because staff might have been African American/Black. Wonder if there was a way to create a variable on racially concordant staff/participant dyad/group variable vs. not and see if that is associated with satisfaction. I think this finding is really important for anyone doing work among PEH, as African American/Black folks are over-represented in populations experiencing homelessness in large urban cities, and calls for having study staff/teams that are representative of the lived experiences and gender/racial.ethnic diversity of PEH.

- can the authors comment on when should satisfaction be ascertained (e.g., at what time points -- all study visits), and how can intervention design be malleable to changing views on satisfaction during an intervention among PEH.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

17 Dec 2021

Professor Joerg Heber

Editor-in-Chief

PLOS ONE

26th November 2021

PONE-D-21-22291

Implementing a smoking cessation intervention for homeless smokers: Participants preferences, feedback, and satisfaction with the ‘power to quit’ programme

We would like to thank the academic editor and the reviewers for their constructive feedback and helpful comments on our manuscript titled ‘Implementing a smoking cessation intervention for homeless smokers: Participants preferences, feedback, and satisfaction with the ‘power to quit’ programme’. Please see below our responses.

JOURNAL REQUIREMENTS:

Comment 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Our Response (1): This manuscript meets PLOS ONE's style requirements

Comment 2. Please ensure you have included the registration number for the clinical trial referenced in the manuscript.

Our Response (2): Trial Registration number has been stated (Line 111)

Comment 3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

Our Response (3): Data will be made available on request. This is now being included in the cover letter.

Comment 4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

Our Response (4): Full name of the IRB as well as statement on informed consent has been stated. Study procedures were approved and monitored by the Institutional Review Board of the University of Minnesota Medical School (Study Number: 1307M39761). Written informed consent was obtained from participants prior to data collection. (Lines 306 - 308)

REVIEWER 1

Comment 1a: Currently, there is little description of the survey in the methods section. How were participants invited to take part.

Our Response (1a): ‘Researchers recruited participants by conducting health fairs, holding informational interviews, posting flyers, and announcing the study at homeless shelters. Participants also helped to recruit by word of mouth.’ This has been stated in the methods section (Line 148-150) The eligibility criteria is also stated (Lines 150 – 167)

Comment 1b: What was the method of administration/data collection,

Our Response (1b): Validated questionnaires were used to collect survey data. Metric measurements of height and weight were collected to calculate body mass index. Carbon monoxide (CO) and saliva cotinine was assessed as biomarkers of tobacco use. The participants were asked to exhale into a carbon monoxide monitor. Patch adherence was measured by: ‘patch checks’ (visual verification of whether a participant was wearing a patch); ‘patch counts’ (documenting the number of patches left in the participant’s possession); and administration of the Morisky scale, a self-reported adherence scale modified to assess adherence to NRT patch

Detailed method of data collection has been stated in the methods (Lines 170 – 196)

Comment 1c: was there a participant incentive for completing the survey, how long was the survey?

Our Response (1c): Incentives were given as compensation for participants time and effort. At each of the 15 visits, participants received incentives. For attending the final 6-month visit, participants received a $40 gift card and a sweatshirt. For participants who attended all 15 sessions, the monetary incentives totaled $275 over 6 months. The study lasted for 26 months. (Details in Lines 131-140)

Comment 1d: The methods would also benefit from a full description of what the survey covered. The results suggest that more was asked than is currently outlined in the methods.

Our Response (1d): The methods section has been improved upon and contains information on:

An overview of the study, aim of the study and the intervention components (Line 111-140)

Recruiting of participants and eligibility criteria ( Lines 148 -167)

Data collection (Lines 170 -196)

Comment 2: Was all the data used in the paper collected at the 26-week survey? It would be useful to clarify this, i.e. was the demographic and other data collected at 26-weeks or was this collected earlier in the trial? This is important to highlight because some of the smoking data, if collected at 26 weeks, will likely have been influenced by participation in the trial, for example the % reporting an unsuccessful quit attempt in the last year and the high confidence to quit score could have been a direct result of the trial, and not therefore suggestive of homeless populations, and should be noted.

Our Response (2): Study data was collected at baseline, week 1, week 2, week 4, week 6, week 8, week 10, week 12, week 14, week 16, week 18, week 20, week 24, and week 26. However, data analyzed in this study was limited to the baseline and week 26 feedback survey. The detailed information is provided in the Data collection section (Lines 179-196)

Comment 3: How was the survey (and importantly, the questions) developed? Were they tested for participants’ understanding and relevance? There are potential issues with some of the main questions asked. For example, the question, ‘Which items were most helpful?’ This question is open to interpretation. Did the authors mean which items were most helpful for reducing smoking or quitting? The inclusion of ‘reading materials’ or ‘community mobile contacts’ suggest that the question could also have been interpreted as ‘most helpful sources of information?’

Our Response (3):The survey tool was developed by the research team with the aim of assessing participants satisfaction with the research program and its components. Prior to the start of the survey, all questions were pretested for understanding and relevance. The survey was administered by trained interviewers to avoid possible misinterpretations of the questions.

Comment 4: The monetary/voucher incentives were particularly valued by homeless study participants. This is unsurprising, although they were not linked with programme satisfaction. The authors write that “incentives or specific components of the programme may not have significantly affected participant motivation to participate in and complete the study”. What were the incentives in the study used for, e.g. enrolment, attending follow-up appointments, motivation for smoking abstinence? This needs to be clarified as currently the paper doesn’t tell us very much about how incentives were used in the trial, and therefore the implications for future interventions are limited.

Our Response (4): Incentives were given as compensation for participants time and effort. Incentives were provided at each of the 15 visits. They were provided when participants completed surveys, attended brief retention visits, attended treatment visits, and at the follow-up 6 month visit. (Details in Lines 131-140)

REVIEWER 2

Comment 1: I might consider changing the language to reflect people or populations experiencing homelessness or people experiencing homelessness who smoke as an indicator that this population is constantly changing and is dynamic based on economic and other circumstances, and move away from calling the population, "the homeless" or "homeless smokers" which signifies a static condition and potentially a character trait (which we know it is not). This might be viewed as less stigmatizing. I would suggest making this change throughout the manuscript.

Our Response (1): The language has been changed to reflect populations experiencing homelessness or people experiencing homelessness.

Comment 2: In the abstract, I might include how satisfaction was qualitatively assessed. I might bring in the point that the rate of satisfaction was high among African American participants in part because there may have been race concordance between study staff, community advisory board and AA participants. I think that is the salient point here that it is helpful to have staff that have lived experiences of participant in the study to improve the effectiveness of the study.

Our Response(2): The assessment of satisfaction has been included in the abstract (Line 32).

Race concordance influencing satisfaction has been included in the abstract (Lines 49 – 52)

Race concordance influencing satisfaction has been included in the discussion ab-initio (Line 415 – 429)

Comment 3 Methods: I might include where the study sites were in Minneapolis and St. Paul - I recall they were in shelters in these cities. Did participants have to commute? What about timings for the assessment and counseling sessions -- were they flexible?

Our Response(3): Study sites were situated within homeless shelters and/or homeless service centers that are located in the downtown/city center easily accessible by public transportation such as city-operated buses and light rail. (Line 146 – 148)

Because of the nature of the study population, timings for assessments and counselling sessions were flexible. (Line 180 – 181)

Comment 4: The authors alluded to this but was the validity of the satisfaction question in this sample -- could the authors conduct an internal validity of this question in this sample?

Our Response(4): The importance of satisfaction is widely recognized in substance use treatment therefore future interventions should aim to assess the internal validity of this specific construct.(Line 484-489) Unfortunately, the Power to Quit study concluded in 2011 therefore we are unable to assess the internal validity of this question within our study sample. In addition, internal validation of the satisfaction question was not planned a priori therefore the satisfaction questions were not constructed in a way that would allow us to go back and run statistical analysis such as confirmatory factor analysis.

Comment 5: Page 7, Line 150 -- which item was most helpful to you -- is this in relation to smoking cessation or general life?

Our Response(5): We thank the reviewers for this comment. The question items that asked participants to state the items they found most helpful was with respect to smoking cessation. The survey was administered by trained interviewers who were able to provide additional context to questions where needed. The respondents were aware that the questions asked were with respect to smoking cessation.

Comment 6: The authors described many covariates, e.g., health and psychosocial covariates, smoking-related covariates, substance use and dependence covariates -- while these are interesting, I wonder how much is relevant to the analysis on satisfaction as none of these variables are controlled for in the model. I might consider taking these out of this paper and referring to the original Power to Quit study findings for descriptive statistics. It does not seem that satisfaction differed based on mental health or substance use based on these findings and if so, might clarify and explain why that might be in the results and discussion.

Our Response(6): We included information on the covariates that were used in the analysis of this paper. We have expunged information other covariates (self-efficacy to refrain from smoking) that were not included in the descriptive and bivariate analysis.

In multivariable analysis, variables significant at p<0.10 in the bivariate analysis were imputed in the logistic regression (Line 298-299 ) We feel it is important to show that some of the covariates did not show statistically significant association with satisfaction. Hence, we have included table 4 showing the bivariate analysis (Line 395).

Comment 7a: On the results/discussion, I might consider discussing in the results, how satisfaction varied for participants who were abstinent at 6 months vs. not; for participants in the intervention vs control (I did not see this distinction and it seems relevant)

Our Response(7a): The number of participants who successfully quit at the end of the survey was very small. As a result, such statistical analysis would not have yielded any meaningful statistically significant differences. We have highlighted this as a study limitation and highlighted it as an area for future research.(Lines 476-484)

Comment 7b: On the results/discussion, one of the participants' feedback was on continuing patch and counseling if that option was available to participants -- and it has implications for extended duration interventions for PEH, perhaps it might be good to discuss this point in the discussion.

Our Response(7b): Thank you for this recommendation. We have updated our results and discussion sections to include this discussion point. (Table 2, Line 489 – 500)

Comment 7c: On the results/discussion - when I read that African American/Black participants had higher satisfaction levels, I immediately thought this was because staff might have been African American/Black. Wonder if there was a way to create a variable on racially concordant staff/participant dyad/group variable vs. not and see if that is associated with satisfaction. I think this finding is really important for anyone doing work among PEH, as African American/Black folks are over-represented in populations experiencing homelessness in large urban cities, and calls for having study staff/teams that are representative of the lived experiences and gender/racial.ethnic diversity of PEH.

Our Response(7c): Thank you for this comment. While we agree that this would have enriched the study findings, we did not set out to assess this relationship in the initial study. As a result, we are unable to generate a variable for racially concordant staff/participant dyad/group and include this in our analysis. However, we have included this point as an area for future research.

(Line 472 – 474 )

Comment 7d: Can the authors comment on when should satisfaction be ascertained (e.g., at what time points -- all study visits), and how can intervention design be malleable to changing views on satisfaction during an intervention among PEH.

Our Response(7d): Satisfaction should be at the least assessed at the end of the study. The pros and cons of assessing satisfaction at multiple time points should be balanced. The pro of assessing satisfaction is the ability to obtain more information that will help to inform future studies. This would be appropriate in a study with a pragmatic design where design changes are built apriori on the study protocol. However, repeated measures of satisfaction may have unintended consequences. Participants may be more prone to giving socially desirable responses, the repeated surveys may become too burdensome for participants and may negatively impact attrition, and the repeated measures may become too burdensome for the research team. Furthermore, changing views on satisfaction should be incorporated into the study without changing randomization of the participants and without creating barriers to carrying out the study.

Yours sincerely,

Oluwakemi Ololade Odukoya

On behalf of the study investigators

Submitted filename: Response to reviewers.docx

Click here for additional data file.

15 Feb 2022

30 Mar 2022

Professor Joerg Heber

Editor-in-Chief

PLOS ONE

25th March 2022

PONE-D-21-22291

Implementing a smoking cessation intervention for homeless smokers: Participants preferences, feedback, and satisfaction with the ‘power to quit’ programme

We would like to thank the academic editor and reviewer for their constructive feedback and helpful comments on our manuscript titled ‘Implementing a smoking cessation intervention for homeless smokers: Participants preferences, feedback, and satisfaction with the ‘power to quit’ programme’. Please see below our responses.

EDITORS COMMENTS:

Comment 1. First, the Data Availability field states that Yes all data are fully available without restriction, but this does not appear to be the case per the cover letter. If data are only available on request, please state why there are legal or ethical restrictions on sharing data. The doi in dryad does not lead to public data set. Please review the PLOS ONE FAQ regarding data sharing: https://journals.plos.org/plosone/s/data-availability

Our Response (1): All data are fully available without restriction an can be accessed at https://www.openicpsr.org/openicpsr/project/165321/version/V1/view This has also been stated in the cover letter

Comment 2. Line 58: 20% is outdated if referring to U.S, which is now around 14%.

Our Response (2): The information on the prevalence of smoking in the general population in the US has been updated (Line 59). The reference has also been updated (Line 520-522).

Comment 3. Line 292-293: Please rephrase to clarify that participants had a positive screen for alcohol or drug dependence rather than identifying them as alcohol/drug dependent (given the relationship between screeners and actual diagnostic status of substance use disorder is far from 100%). Same guidance for Table 1.

Our Response (3): Thank you for your comment. The term “Alcohol/drug dependent” has been re-phrased as “had a positive screen for alcohol or drug dependence” (Line 297-298, Table 1 Line 307, Table 4 Line 369 )

Comment 4. Table 1: Please relabel rows to 1, 2, >3. It seems odd and minimizing to use the word “just” as a qualifier for the number of times experiencing homelessness.

Our Response (4): The rows have been re-labelled as Once, Twice, Thrice (Table 1 Line 307, Table 4 Line 369 )

Comment 5: Table 3: Overall program satisfaction – please report frequency/percentage for all 5 response options.

Our Response (5): The frequency/percentage values have been reported for all 5 responses. (Table 3 Line 362)

Comment 6: Table 4: In table note please remind the reader of how ‘satisfied’ and ‘not satisfied’ were coded. Why was ‘somewhat satisfied’ coded as ‘not satisfied’? I can see a justification for combining response options 1-3, but ‘somewhat satisfied’ doesn’t seem like it fits under the umbrella of ‘not satisfied’. I won't require to re-code and rerun all analyses, but please provide a justification for this decision if you prefer to maintain this coding scheme, but adjust labels (e.g., ‘Very satisfied vs No very satisfied’.

Our Response (6): The options for satisfaction were 1)Not satisfied at all; 2) Somewhat unsatisfied; 3) Neutral; 4) Somewhat satisfied; 5) Very satisfied. There was no option stated as “somewhat satisfied” rather this was “somewhat unsatisfied” We believe that the options Not satisfied at all and somewhat unsatisfied both reflect some form of dissatisfaction, hence they were categorized together as “Not very satisfied” We have reminded the readers of how overall program satisfaction was re-coded. The labels have been adjusted as ‘Very satisfied vs Not very satisfied’ (Table 4 Line 369, Line 373-375)

Comment 7: Table 4: Please justify your alpha of 0.10 for Table 4.

Our Response (7): In multivariable analysis, variables significant at p<0.10 in the bivariate analysis were imputed in the logistic regression in order to include variables tending towards a positive significance. (Line 271-272)

Comment 8: Table 5: Could include the 95% CI in parentheses after the AOR (e.g., 1.005 (0.977, 1.035)). Please label as 95% CI in table note or column header (per Reviewer 3).

Our Response (8): The 95% CI is stated in parentheses after the AOR. The column header is labeled as AOR(95% CI Lower limit , Upper limit) (Table 5 Line 393)

Comment 9: Reviewer 3 provides useful suggestions that I recommend you incorporate into the revised manuscript. Given that this manuscript has already been reviewed by two peer reviewers, I would not require you to make the following changes recommended by Reviewer 3:

-Participants (line 150): readers can refer to your citation of the main design/outcomes papers

-Data Collection (Line 176-178): you already state only baseline/26wk are presented here.

-Table 1: do not need to recode age. Cosmetic changes to tables will occur during copy-editing, not necessary at this stage.

Our Response (9): Comments well taken

REVIEWER 3

Comment 1: Exposure variables

Line 195, coding labelling to be fully provided e.g. 1(most helpful), 2( ), 3(not helpful). Likewise Line 201 e.g. 1(liked the most), 2( ), 3(did not like it)

Our Response (1): Thank you for your comment. This has been done. (Line 197, Line 203-204)

Comment 2: Line 264, Analysis to be written as Statistical analyses.

Our Response (2): Thank you for your comment. This has been done. (Line 266)

Comment 3: Line 269, for the statement ‘In addition, we controlled for study arm,’ the reason to analyze as combined groups and not groups comparison to be clearly stated before displaying the results.

Our Response (3):We thank the reviewer for this comment. Study arm was not controlled for in the regression model. This has been corrected in the methods (Statistical analysis) (Line 272)

Comment 4: Results

Line 284, full name for GED to be provided.

Our Response (4): This has been done (Line 288-289).

Comment 5: Line 322. 21% to be replaced with 21.0%.

Our Response (5): This has been done (Line 327)

Comment 6: Line 324, Over 80% to replaced with 81.6%

Our Response (6): This has been done (Line 329)

Comment 7: Line 337, typo ‘thatthe’

Our Response (7): This has been done (Line 343)

Comment 8: Line 303 Table 1, title too brief. For monthly income, the symbol for the income category is incorrect (different to Line 214). For the homelessness characteristics ‘just once, One to three times(twice) and More than three times (≥3)’ are confusing Perhaps just state once, twice, thrice or more. Drug dependent and not drug dependent to be unbold.

Our Response (8): This has been done (Table 1 Line 307 - 308)

Comment 9: Line 303 Table 1 and Line 330 Table 2, all the data to be presented (i.e No to be included) while missing data to be denoted in the table footnote and n to be stated for all variables.

Our Response (9): All the data has been presented. However, missing data was not shown in table, rather ‘n’ was stated for each variable. (Table 1 Line 308, Table 2 Line 336)

Comment 10: Line 356 Table 3, the figure 4 or 5 in the subcategory to be spelled out or denoted in the table footnote. n to be stated and any missing data to be denoted.

Our Response (10): Thank you for this comment. This has been effected (Table 3 Line 362).

Comment 11: Line 360 Table 4, n to be stated for each variable. Actual symbol chi-square X^2 to be used. The chi square value to be reduced to 2 decimal points and the decimal points for p value to be standardized. All the statistical tests used in Table 4 to be denoted in the table footnote. For the age, depression, stress, age started smoking regularly, number of 24 hour quit attempts past year, confidence to quit variables, the data were presented as mean± sd. The statistical test to be stated. If chi-square test was employed, the categories of each variable and frequency to be displayed. For the self-reported general health category poor, 14.2% to be replaced with 14.3%. Please re check the chi-square value and p value for the variable ‘top ranked incentive’.

Our Response (11): The comments have been effected (Table 4 Line 370)

Comment 12: The word p value or p-value to be consistent with Table 5. # was mentioned in table footnote but the label nowhere found in the table.

Our Response (12): This has been done (Table 5 Line 393)

Comment 13: Line 373 Table 5, the model summary such as pseudo R^2 and goodness of fit test to be provided. 95%CI to be stated before lower and upper limit are stated.

Our Response (13): pseudo R^2 and goodness of fit test have been provided (Table 5 Line 393)

Comment 14: Line 367, the sentence ‘Race was with a predictor of overall programme satisfaction.’ requires revision.

Our Response (14): This has been corrected (Line 382)

Comment 15: Line 376, p=.05 to be replaced with p < 0.05.

Our Response (15): This has been corrected (Line 398)

Comment 16: Discussion

Line 427, 435 & Line 437, typos thesurvey, Fourth,, Fifth..

Our Response (16): This has been corrected (Line 447, 455, 457)

Comment 17: Was there any other possible bias arising from the interviewing process?

Our Response (17): There is also a possibility of non -response bias which arises when respondents are systematically different from non-respondents. Prior to data collection, measures were instituted to limit such biases and this has been included in the text. (Line 455-460).

Yours sincerely,

Oluwakemi Ololade Odukoya

On behalf of the study investigators

Submitted filename: Response to reviewers (2).docx

Click here for additional data file.

5 May 2022

Implementing a smoking cessation intervention for people experiencing homelessness: participants’ preferences, feedback, and satisfaction with the ‘power to quit’ program

PONE-D-21-22291R2

Dear Dr. Odukoya,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jesse T. Kaye, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for this important contribution to the literature and addressing the comments and suggestions from previous reviews.

In table 3 there appears to be a few typos in the response options where 'somewhat satisfied' is listed twice in several sections where I believe it should be 'somewhat unsatisfied' first and 'somewhat satisfied' second (on either side of neutral). Please correct this during the copyediting process.

Reviewers' comments:

24 May 2022

PONE-D-21-22291R2

Implementing a smoking cessation intervention for people experiencing homelessness: participants’ preferences, feedback, and satisfaction   with the ‘power to quit’ program

Dear Dr. Odukoya:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jesse T. Kaye

Academic Editor

PLOS ONE