Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Incarceration was measured through publicly available data on arrests, convictions, and periods of incarceration in Indiana. We searched each name in state prison and county jail records and identified them as incarcerated if their name appeared in the jail or prison data during the 12 month study window of their enrollment.

Note: At baseline, we confirmed with our community partner that each participant had been in prison or jail, or on work release, within the past 3 months. However, this did not match the publicly available data, indicating that the available data is likely incomplete.

Incarceration was measured through publicly available data on arrests, convictions, and periods of incarceration in Indiana. We searched each name in state prison and county jail records and identified them as incarcerated if their name appeared in the jail or prison data during the 12 month study window of their enrollment.

Note: At baseline, we confirmed with our community partner that each participant had been in prison or jail, or on work release, within the past 3 months. However, this did not match the publicly available data, indicating that the available data is likely incomplete.

The Self-Determination Scale (SDS) assesses individual differences in the extent to which people tend to function in a self-determined way. It is considered a relatively enduring aspect of people's personalities which reflects being more aware of their feelings and sense of self and feeling a sense of choice with respect to their behavior. The SDS is a 10-item scale, with two 5-item subscales: (1) awareness of oneself and (2) perceived choice in one's actions. We utilized only the second scale on perceived choice.

These 5-items present respondents with 2 statements: Statement (A) indicates more choice and Statement (B) indicates less choice.

Respondents answer on a five point scale: 1 "Only A feels true", 2 , 3 , 4 , 5 "Only B feels true."

For perceived choice, the 5 items are reverse scored so that higher scores on every item will indicate a higher level of self-determination. Range: (5-25)

The Self-Determination Scale (SDS) assesses individual differences in the extent to which people tend to function in a self-determined way. It is considered a relatively enduring aspect of people's personalities which reflects being more aware of their feelings and sense of self and feeling a sense of choice with respect to their behavior. The SDS is a 10-item scale, with two 5-item subscales: (1) awareness of oneself and (2) perceived choice in one's actions. We utilized only the second scale on perceived choice.

These 5-items present respondents with 2 statements: Statement (A) indicates more choice and Statement (B) indicates less choice.

Respondents answer on a five point scale: 1 "Only A feels true", 2 , 3 , 4 , 5 "Only B feels true."

For perceived choice, the 5 items are reverse scored so that higher scores on every item will indicate a higher level of self-determination. Range: (5-25)

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy.

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy (a better outcome).

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy (a better outcome).

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Respondents identify individuals who make up their social networks. Each individual is identified as an alter, and alters are totaled to determine network size.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Respondents identify individuals who make up their social networks. Each individual is identified as an alter, and alters are totaled to determine network size.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how close they are to each alter using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close.

Below is the difference in the fraction of network close to the subject. Results are presented for the fraction of the network identified as being "very close", as well as combined "very close" or "sort of close".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how close they are to each alter using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close.

Below is the difference in the fraction of network close to the subject. Results are presented for the fraction of the network identified as being "very close", as well as combined "very close" or "sort of close".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters. This component focuses on the closeness of the participants network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each alter pair, the respondent scores their closeness on a range of 1 to 3, with a lower score indicating more closeness.

Below is the difference in the network closeness. The range for this scale will differ, depending on the number of alters identified.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how often they see this person or talk to them on the phone or video chat using a 3-point scale (1) Often, (2) Occasionally, (3) Hardly Ever. Below is the difference in the fraction of network communicating with the subject. Results are presented for the fraction of the network identified as communicating "often", as well as combined "often" or "occasionally".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how often they see this person or talk to them on the phone or video chat using a 3-point scale(1) Often, (2) Occasionally, (3) Hardly Ever. Below is the difference in the fraction of network communicating with the subject. Results are presented for the fraction of the network identified as communicating "often", as well as combined "often" or "occasionally".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify types of support provided by alters. Examples are given (listed below) and subjects select all that apply. Below is the difference in the fraction network providing support to the subject.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify types of support provided by alters. Examples are given (listed below) and subjects select all that apply. Below is the difference in the fraction network providing support to the subject.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify whether the alter ever hassles them, causes them problems, or makes life difficult, using a 3-point scale: (1) A lot, (2) Sometimes, (3) No/Not Really. Below is the difference in the fraction of network causing the subject problems. For analysis, we combined "a lot" and "sometimes" into a single category and compared it to "no/not really".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify whether the alter ever hassles them, causes them problems, or makes life difficult, using a 3-point scale: (1) A lot, (2) Sometimes, (3) No/Not Really. Below is the difference in the fraction of network causing the subject problems. For analysis, we combined "a lot" and "sometimes" into a single category and compared it to "no/not really".

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify whether the alter uses alcohol, tobacco, or illicit drugs/medication for reasons other their prescribed purposes. Below is the difference in the fraction of the network using a substance(s).

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify whether the alter uses alcohol, tobacco, or illicit drugs/medication for reasons other their prescribed purposes. Below is the difference in the fraction of the network using a substance(s).

Subjects enrolled in standard services.

Total of all reporting groups

The Treatment Motivation Questionnaire (TMQ) is a 26-item instrument. The instrument uses a 7-point scale, ranging from (1) not at all true to (7) very true. The TMQ has four sub scales: internal reasons, external reasons, help seeking, and confidence.

Subjects enrolled in standard services.

Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Note that analysis was limited to subjects with available data at both time points.

Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Note that analysis was limited to subjects with available data at both time points.

Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Note that analysis was limited to subjects with available data at both time points

Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Note that analysis was limited to subjects with available data at both time points

Using the Frequency of Drug Use scale, we measured self-reported use of alcohol, sedative, tranquilizers, painkillers, stimulants, marijuana, cocaine, crack, hallucinogens, inhalants, heroin, and prescription medications. Respondents are asked to indicate the number of days having used each listed drug within the past 30 days.

The minimum value is 0, and the maximum value is 30.

Note that analysis was limited to subjects with available data at both time points

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Note that analysis was limited to subjects with available data at both time points.

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Note that analysis was limited to subjects with available data at both time points.

Stages of Change, Readiness, and Treatment Eagerness Scale (SOCRATES) is a 19-item instrument that measures readiness to change behaviors related to substance use, using a 5 point scale 1 (Strongly Disagree), 2 (Disagree), 3 (Undecided), 4 (Agree), 5 (Strongly Agree). This measure is best used to assess stages of change at baseline; subsequent changes might not be clinically relevant.

Subscales:

1. Recognition: A higher score indicates acknowledgement of the problem, possible harm, desires change. (Range: 7-35)
2. Ambivalence: A higher score reflects openness to reflection on drug use. A lower score may mean they "know" their drug use is causing problems (high Recognition), or they "know" that they do not have drug use problems (low Recognition). Thus, a low Ambivalence score should be interpreted in relation to the Recognition score. (Range: 4-20)
3. Taking steps: A higher score means steps have been taken, and they may already be experiencing success. (Range: 8-40)

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

One of the four items requests participants to report on their general health, with a scale of (1) excellent, (2) very good, (3) good, (4) fair, (5) poor.

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Note that analysis was limited to subjects with available data at both time points.

Using the 4-item Current Quality of Life Scale, we measured an individual's perceived general physical and mental health during the past 30 days. Self-perception of current physical and mental health is indicative of a person's current health and future health outcomes. The general assessment of health has been shown to be both a solid measure of self-reported health and a powerful predictor of mortality and morbidity.

Two of the four items requested the number of unhealthy days, physically or mentally, out of the past 30 days. These are reported as a sum.

One of the four items requested the number of days that usual activities are prevented due to poor physical or mental health are also reported out of the past 30 days.

Note that analysis was limited to subjects with available data at both time points.

Incarceration was measured through publicly available data on arrests, convictions, and periods of incarceration in Indiana. We searched each name in state prison and county jail records and identified them as incarcerated if their name appeared in the jail or prison data during the 12 month study window of their enrollment.

Note: At baseline, we confirmed with our community partner that each participant had been in prison or jail, or on work release, within the past 3 months. However, this did not match the publicly available data, indicating that the available data is likely incomplete.

Incarceration was measured through publicly available data on arrests, convictions, and periods of incarceration in Indiana. We searched each name in state prison and county jail records and identified them as incarcerated if their name appeared in the jail or prison data during the 12 month study window of their enrollment.

Note: At baseline, we confirmed with our community partner that each participant had been in prison or jail, or on work release, within the past 3 months. However, this did not match the publicly available data, indicating that the available data is likely incomplete.

The Self-Determination Scale (SDS) assesses individual differences in the extent to which people tend to function in a self-determined way. It is considered a relatively enduring aspect of people's personalities which reflects being more aware of their feelings and sense of self and feeling a sense of choice with respect to their behavior. The SDS is a 10-item scale, with two 5-item subscales: (1) awareness of oneself and (2) perceived choice in one's actions. We utilized only the second scale on perceived choice.

These 5-items present respondents with 2 statements: Statement (A) indicates more choice and Statement (B) indicates less choice.

Respondents answer on a five point scale: 1 "Only A feels true", 2 , 3 , 4 , 5 "Only B feels true."

For perceived choice, the 5 items are reverse scored so that higher scores on every item will indicate a higher level of self-determination. Range: (5-25)

Note that analysis was limited to subjects with available data at both time points

The Self-Determination Scale (SDS) assesses individual differences in the extent to which people tend to function in a self-determined way. It is considered a relatively enduring aspect of people's personalities which reflects being more aware of their feelings and sense of self and feeling a sense of choice with respect to their behavior. The SDS is a 10-item scale, with two 5-item subscales: (1) awareness of oneself and (2) perceived choice in one's actions. We utilized only the second scale on perceived choice.

These 5-items present respondents with 2 statements: Statement (A) indicates more choice and Statement (B) indicates less choice.

Respondents answer on a five point scale: 1 "Only A feels true", 2 , 3 , 4 , 5 "Only B feels true."

For perceived choice, the 5 items are reverse scored so that higher scores on every item will indicate a higher level of self-determination. Range: (5-25)

Note that analysis was limited to subjects with available data at both time points

The Self-Determination Scale (SDS) assesses individual differences in the extent to which people tend to function in a self-determined way. It is considered a relatively enduring aspect of people's personalities which reflects being more aware of their feelings and sense of self and feeling a sense of choice with respect to their behavior. The SDS is a 10-item scale, with two 5-item subscales: (1) awareness of oneself and (2) perceived choice in one's actions. We utilized only the second scale on perceived choice.

These 5-items present respondents with 2 statements: Statement (A) indicates more choice and Statement (B) indicates less choice.

Respondents answer on a five point scale: 1 "Only A feels true", 2 , 3 , 4 , 5 "Only B feels true."

For perceived choice, the 5 items are reverse scored so that higher scores on every item will indicate a higher level of self-determination. Range: (5-25)

Note that analysis was limited to subjects with available data at both time points

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy.

Note that analysis was limited to subjects with available data at both time points.

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy (a better outcome).

Note that analysis was limited to subjects with available data at both time points.

Using the General Self Efficacy Scale, we measured an individual's belief about whether problems or barriers can be confronted and addressed with a successful outcome. This scale can be used to determine how the respondent copes with daily hassles and adapts to stressful life events. Self-efficacy is indicative of subsequent behavior and can be used for clinical practice and psychological research.

The 10-item instrument uses a 4-point scale, ranging from 1 (not at all true) to 4 (exactly true), so the composite score ranges from 10 to 40 with higher score indicating greater self-efficacy (a better outcome).

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Respondents identify individuals who make up their social networks. Each individual is identified as an alter, and alters are totaled to determine network size.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Respondents identify individuals who make up their social networks. Each individual is identified as an alter, and alters are totaled to determine network size.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how close they are to each alter using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close.

Below is the difference in the fraction of network close to the subject. Results are presented for the fraction of the network identified as being "very close", as well as combined "very close" or "sort of close".

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how close they are to each alter using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close.

Below is the difference in the fraction of network close to the subject. Results are presented for the fraction of the network identified as being "very close", as well as combined "very close" or "sort of close".

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters. This component focuses on the closeness of the participants network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each alter pair, the respondent scores their closeness on a range of 1 to 3, with a lower score indicating more closeness.

Below is the difference in the network closeness. The range for this scale will differ, depending on the number of alters identified.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

Network closeness refers to the closeness of the participant's network (i.e. how close the alters identified are to one another).

After identifying alters within their network, respondents are asked to identify how close each alter is to one another using a 3-point scale: (1) Very close, (2) Sort of close, (3) Not very close. Thus, for each original alter pair, the respondent scores their closeness on a range of 1 to 3. A lower score indicates a stronger network.

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how often they see this person or talk to them on the phone or video chat using a 3-point scale (1) Often, (2) Occasionally, (3) Hardly Ever. Below is the difference in the fraction of network communicating with the subject. Results are presented for the fraction of the network identified as communicating "often", as well as combined "often" or "occasionally".

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify how often they see this person or talk to them on the phone or video chat using a 3-point scale(1) Often, (2) Occasionally, (3) Hardly Ever. Below is the difference in the fraction of network communicating with the subject. Results are presented for the fraction of the network identified as communicating "often", as well as combined "often" or "occasionally".

Note that analysis was limited to subjects with available data at both time points.

The Social Networks tool is an interviewer-administered questionnaire to assess whom respondents depend on for help in important general matters and specific health matters. This measure captures each of the respondent's network members and a complete assessment of network structure to construct the respondent's social network and whom they rely on for help with health-related and other matters.

After identifying their network, respondents are asked to identify types of support provided by alters. Examples are given (listed below) and subjects select all that apply. Below is the difference in the fraction network providing support to the subject.

Note that analysis was limited to subjects with available data at both time points.