Anthony J Cannon et al

Statistical Analysis

Participant characteristics were compared according to level of spirituality using a chi-square test for categorical data and the Wilcoxon test for continuous data (Table 1). Multivariate logistic regression models were fitted to evaluate separately the relationship between (1) spirituality with patient-rated worry as the outcome, (2) spirituality with follow-up health-care utilization as the outcome, and (3) patient-rated worry with follow-up health-care utilization as the outcome. In the above models, the following covariates were forced into each model: age, sex, cancer type, time from last cancer-related treatment to study start time, income, and type of medical insurance. These models were also fitted using outcomes ascertained at both 6 and 12 months. Interaction models between patient-rated worry and level of spirituality were also evaluated for an association with follow-up health-care utilization at 12 months to explore the role of spirituality in the relationship between patient-rated worry and health-care utilization. A P value of at least 0.05 was considered statistically significant.

Results

Study Participation

Of the 2,000 participants invited, 1,881 were deemed eligible (minus those who died or had wrong addresses). Baseline questionnaires were returned by 939 participants (baseline response rate of 50%). Seventeen wanted to participate only in the baseline survey. Of the 922 baseline participants, 691 returned the 6-month survey at the time of the analysis for this study, for a response rate of 76% when adjusted for deaths (182 no response, 18 deaths, 25 declined, 12 returned with wrong address). At 1 year, 691 surveys were mailed, with 588 surveys returned (58 no response, 17 deaths, 14 declined, 13 returned with wrong address, and one in hospice); a response rate of 87% was achieved after adjusting for deaths. Thirty-seven participants had missing information on spirituality, leaving a total of 551 included in this study. No differences in age, sex, and type of cancer were noted between patients included and excluded in the current analysis.

Characteristics of Study Participants

Demographic characteristics of the 551 study participants included in this study are shown in Table 1. We found that cancer survivors with low or high spirituality were more similar than different in all but two characteristics: highly spiritual survivors were more likely to be Protestant and male.

Prevalence of Spirituality and Patient Worry

Within our population, 271 (49%) survivors reported low spirituality and 280 (51%) reported high spirituality (Table 1). Also, at baseline, 277 (51%) survivors reported high levels of recurrence/progression-related worry, 190 survivors (35%) reported high levels of new malignancy–related worry, and 178 survivors (33%) reported high levels of treatment-related complication worry. As some participants may have reported one or more types of worry, this translates to 322 (59%) reporting any type of worry. Highly spiritual survivors reported significantly lower levels of high worry concerning recurrence/progression (6-month 27% vs. 38%, P < 0.01; 12-month 21% vs. 38%, P < 0.01), development of a different type of cancer (6-month 22% vs. 31%, P = 0.03; 12-month 15% vs. 26%, P < 0.01), and complications from treatment (6-month 17% vs. 30%, P < 0.01; 12-month 16% vs. 26%, P < 0.01). Highly spiritual survivors reported significantly lower levels for any type of worry at both 6 and 12 months (6 months 37% vs. 54%, P <0.01; 12 months 28% vs. 47%, P < 0.01) (Table 2).

Relationship Between Spirituality and Patient Worry

At the 6- and 12-month time points, after adjusting for covariates, highly spiritual survivors were significantly less likely to have worries than survivors who reported lower spirituality regarding disease recurrence/progression at 6 months (odds ratio [OR] = 0.61, 95% confidence interval [CI] 0.42–0.89, P < 0.01) and at 12 months (OR = 0.43, 95% CI 0.29–0.63, P < 0.01), complications from treatment at 6 months (OR = 0.50, 95% CI 0.33–0.76, P < 0.01) and at 12 months (OR = 0.54, 95% CI 0.35–0.83, P < 0.01), and development of a different type of cancer at 6 months (OR = 0.65, 95% CI 0.44–0.97, P = 0.04) and at 12 months (OR = 0.50, 95% CI 0.33–0.77, P < 0.01) (Table 3A).

Relationship Between Patient Worry and Follow-Up Health-Care Utilization

Survivors who were highly worried about disease recurrence/progression, development of another type of cancer, and/or complications from treatment were more likely to visit their providers for follow-up care when compared with survivors who were less worried at 6 months (OR = 1.81, 95% CI 1.04–3.12, P = 0.03) and at 12 months (OR = 1.49, 95% CI 1.00–2.22, P = 0.05). Similarly, survivors who were highly worried were also more likely to place phone calls to their follow-up providers for medical inquiries than survivors who were less worried at 6 months (OR = 2.21, 95% CI 1.48–3.31, P < 0.01) and at 12 months (OR = 1.74, 95% CI 1.20–2.53, P = 0.01). We did not observe differences in emergency room visits between survivors with low and those with high rates of worrying at both 6 and 12 months (Table 3B).

Relationship Between Spirituality and Health-Care Utilization

No significant differences were noted for the frequency of follow-up visits, changes in follow-up providers, and emergency room visits between the levels of spirituality at both 6 and 12 months. However, at 12 months, highly spiritual survivors were less likely to call their follow-up providers for medical inquiries compared to survivors with low spirituality scores (OR = 0.70, 95% CI 0.49–1.00, P = 0.04) (Table 3C).

Interaction Between Spirituality and Patient Worry With Health-Care Utilization

Interaction between patient-rated worry and level of spirituality as it relates to health-care utilization was not statistically significant (data not shown). This suggests that spirituality does not modify the effect of patient worry in producing change in follow-up health-care utilization.

Discussion

Our study examined the relationships between spirituality, patient-rated worry, and follow-up health-care utilization among cancer survivors and found that individuals who possess higher levels of spirituality tend to have less worry of disease recurrence/progression, development of treatment-related complications, and development of new cancers. These findings are consistent with previous research among patients with advanced or terminal cancers that consistently showed such correlations between spirituality and general measures of anxiety.^{[10], [15], [17], [19], [30] and [31]} Additionally, our study showed that a higher degree of worry about common concerns of cancer survivors is associated with more follow-up visits and calls to health-care providers. However, our data also showed that spirituality by itself is for the most part not associated with follow-up health-care utilization.

It has been documented that psychosocial factors like anxiety and spirituality can influence behaviors.^{[32], [33], [34], [35] and [36]} Our analysis showed that both discretionary and nondiscretionary indices of health-care utilization increased significantly among highly worried cancer survivors. However, these increases are independent of one's level of spirituality. These results suggest that cancer survivors with a high degree of worry about disease recurrence/progression, development of treatment-related complications, or development of a new cancer produce a change in care-seeking behavior and may concomitantly alter the health provider's need to see the patient. Our results also suggest that while spirituality has an impact on one's level of worry, being less spiritual does not necessarily alter a cancer survivor's care-seeking behavior.

Worried patients present a potential problem for clinicians in that they may need more attention during clinic visits,³⁷ may result in requests for more ancillary/diagnostic tests including imaging modalities,^{[38] and [39]} or may use more medications^{[40] and [41]} or resort to other alternative therapies^{[42], [43], [44] and [45]} available to reduce their worries. Given that cancer patients already receive many chemotherapeutic agents for their treatment, many of them are more inclined to undergo alternative therapies.^{[16], [43], [46], [47] and [48]} Spirituality-based interventions shown to be effective at reducing anxiety and increasing QOL may therefore have a role among cancer survivors. And because spirituality and religiosity are closely linked,²⁹ faith-based interventions may also benefit the patient.

Our study has several implications in the assessment of cancer survivors in multidisciplinary survivorship clinics. While much attention about assessing depression, anxiety, and QOL has been given to cancer survivors, our study shows that the evaluation of one's spirituality may have some merit as well. Participants with low spirituality and a high degree of worry may benefit from activities that enhance spirituality (e.g., yoga, meditation). Because of the increasing number of cancer survivors,^{[32] and [49]} development of clinic-based spiritual interventions to address common worries of cancer survivors may be appropriate. In addition to the implications for clinical practice, our study has implications for future research. While the literature has shown a correlation between spirituality and religiosity,²⁹ these two concepts are not the same.^{[1], [2], [50], [51] and [52]} It would have been interesting to compare outcomes by level of spirituality and religiosity, but our data revealed a high degree of correlation between these two concepts. Over 90% of individuals who are spiritual are also religious.^{[28], [53] and [54]} This may be the reason that some spirituality-based interventions have enhancement of religious activities as main approaches to improve spirituality.^{[28] and [53]}

While our study has the strengths associated with a prospective study in a relatively large number of cancer survivors treated in a single medical center, it has several limitations. Our participation rate at baseline was only 50%, although our retention rates at 6 and 12 months were on average 80%. Another limitation of our study is that the baseline surveys were conducted at different time intervals from last treatment, although this limitation also allowed us to include all kinds of cancer survivors in terms of disease and time interval from last cancer treatment. Analysis confined to patients who received treatment within the last 5 years (n = 371) showed essentially the same results. We also compared the baseline spirituality scores of the study participants according to time from last treatment to study participation (0–2, 2–5, >5) and showed no statistically significant differences. Additionally, we adjusted for time from last treatment to study participation in the multivariate analyses. Combining all the participants into one analysis allowed for our exploratory analyses to have stronger statistical power. Another limitation of our study is the crude measurement of patient worry. However, in the absence of validated instruments to measure these worries, we felt the measures reflected subjective ratings of common worries by cancer survivors. Health-care utilization would have been ideally measured continuously to better quantify the medical services utilized. However, because we included a heterogeneous group of cancer patients, this measure would be highly variable and depend on the type of disease and treatment received by the patient. Thus, type of disease and time period from last treatment were adjusted for in the multivariate analyses.

In summary, cancer survivors who possess higher levels of spirituality tend to have a lesser degree of worry over disease recurrence/progression, development of treatment complications, and development of new cancers. A higher degree of worry about the common concerns of cancer survivors is associated with more follow-up visits and calls to health providers. However, our data showed that, for the most part, spirituality is not associated with follow-up health-care utilization.

Acknowledgments

The authors thank Linda Bauer, Garrett Frost, and Gregory McFadden for their help in coordinating the study and processing the data. This work was supported by the University of Nebraska Medical Center–Eppley Cancer Center (Support Grant P30 CA 036727) and the Medical Student Research Program. The funding source had no role in the design, collection, analysis, and interpretation of the data or in the writing of the article.

Statistical Analysis

Participant characteristics were compared according to level of spirituality using a chi-square test for categorical data and the Wilcoxon test for continuous data (Table 1). Multivariate logistic regression models were fitted to evaluate separately the relationship between (1) spirituality with patient-rated worry as the outcome, (2) spirituality with follow-up health-care utilization as the outcome, and (3) patient-rated worry with follow-up health-care utilization as the outcome. In the above models, the following covariates were forced into each model: age, sex, cancer type, time from last cancer-related treatment to study start time, income, and type of medical insurance. These models were also fitted using outcomes ascertained at both 6 and 12 months. Interaction models between patient-rated worry and level of spirituality were also evaluated for an association with follow-up health-care utilization at 12 months to explore the role of spirituality in the relationship between patient-rated worry and health-care utilization. A P value of at least 0.05 was considered statistically significant.

Results

Study Participation

Of the 2,000 participants invited, 1,881 were deemed eligible (minus those who died or had wrong addresses). Baseline questionnaires were returned by 939 participants (baseline response rate of 50%). Seventeen wanted to participate only in the baseline survey. Of the 922 baseline participants, 691 returned the 6-month survey at the time of the analysis for this study, for a response rate of 76% when adjusted for deaths (182 no response, 18 deaths, 25 declined, 12 returned with wrong address). At 1 year, 691 surveys were mailed, with 588 surveys returned (58 no response, 17 deaths, 14 declined, 13 returned with wrong address, and one in hospice); a response rate of 87% was achieved after adjusting for deaths. Thirty-seven participants had missing information on spirituality, leaving a total of 551 included in this study. No differences in age, sex, and type of cancer were noted between patients included and excluded in the current analysis.

Characteristics of Study Participants

Demographic characteristics of the 551 study participants included in this study are shown in Table 1. We found that cancer survivors with low or high spirituality were more similar than different in all but two characteristics: highly spiritual survivors were more likely to be Protestant and male.

Prevalence of Spirituality and Patient Worry

Within our population, 271 (49%) survivors reported low spirituality and 280 (51%) reported high spirituality (Table 1). Also, at baseline, 277 (51%) survivors reported high levels of recurrence/progression-related worry, 190 survivors (35%) reported high levels of new malignancy–related worry, and 178 survivors (33%) reported high levels of treatment-related complication worry. As some participants may have reported one or more types of worry, this translates to 322 (59%) reporting any type of worry. Highly spiritual survivors reported significantly lower levels of high worry concerning recurrence/progression (6-month 27% vs. 38%, P < 0.01; 12-month 21% vs. 38%, P < 0.01), development of a different type of cancer (6-month 22% vs. 31%, P = 0.03; 12-month 15% vs. 26%, P < 0.01), and complications from treatment (6-month 17% vs. 30%, P < 0.01; 12-month 16% vs. 26%, P < 0.01). Highly spiritual survivors reported significantly lower levels for any type of worry at both 6 and 12 months (6 months 37% vs. 54%, P <0.01; 12 months 28% vs. 47%, P < 0.01) (Table 2).

Relationship Between Spirituality and Patient Worry

At the 6- and 12-month time points, after adjusting for covariates, highly spiritual survivors were significantly less likely to have worries than survivors who reported lower spirituality regarding disease recurrence/progression at 6 months (odds ratio [OR] = 0.61, 95% confidence interval [CI] 0.42–0.89, P < 0.01) and at 12 months (OR = 0.43, 95% CI 0.29–0.63, P < 0.01), complications from treatment at 6 months (OR = 0.50, 95% CI 0.33–0.76, P < 0.01) and at 12 months (OR = 0.54, 95% CI 0.35–0.83, P < 0.01), and development of a different type of cancer at 6 months (OR = 0.65, 95% CI 0.44–0.97, P = 0.04) and at 12 months (OR = 0.50, 95% CI 0.33–0.77, P < 0.01) (Table 3A).

Relationship Between Patient Worry and Follow-Up Health-Care Utilization

Survivors who were highly worried about disease recurrence/progression, development of another type of cancer, and/or complications from treatment were more likely to visit their providers for follow-up care when compared with survivors who were less worried at 6 months (OR = 1.81, 95% CI 1.04–3.12, P = 0.03) and at 12 months (OR = 1.49, 95% CI 1.00–2.22, P = 0.05). Similarly, survivors who were highly worried were also more likely to place phone calls to their follow-up providers for medical inquiries than survivors who were less worried at 6 months (OR = 2.21, 95% CI 1.48–3.31, P < 0.01) and at 12 months (OR = 1.74, 95% CI 1.20–2.53, P = 0.01). We did not observe differences in emergency room visits between survivors with low and those with high rates of worrying at both 6 and 12 months (Table 3B).

Relationship Between Spirituality and Health-Care Utilization

No significant differences were noted for the frequency of follow-up visits, changes in follow-up providers, and emergency room visits between the levels of spirituality at both 6 and 12 months. However, at 12 months, highly spiritual survivors were less likely to call their follow-up providers for medical inquiries compared to survivors with low spirituality scores (OR = 0.70, 95% CI 0.49–1.00, P = 0.04) (Table 3C).

Interaction Between Spirituality and Patient Worry With Health-Care Utilization

Interaction between patient-rated worry and level of spirituality as it relates to health-care utilization was not statistically significant (data not shown). This suggests that spirituality does not modify the effect of patient worry in producing change in follow-up health-care utilization.

Discussion

Our study examined the relationships between spirituality, patient-rated worry, and follow-up health-care utilization among cancer survivors and found that individuals who possess higher levels of spirituality tend to have less worry of disease recurrence/progression, development of treatment-related complications, and development of new cancers. These findings are consistent with previous research among patients with advanced or terminal cancers that consistently showed such correlations between spirituality and general measures of anxiety.^{[10], [15], [17], [19], [30] and [31]} Additionally, our study showed that a higher degree of worry about common concerns of cancer survivors is associated with more follow-up visits and calls to health-care providers. However, our data also showed that spirituality by itself is for the most part not associated with follow-up health-care utilization.

It has been documented that psychosocial factors like anxiety and spirituality can influence behaviors.^{[32], [33], [34], [35] and [36]} Our analysis showed that both discretionary and nondiscretionary indices of health-care utilization increased significantly among highly worried cancer survivors. However, these increases are independent of one's level of spirituality. These results suggest that cancer survivors with a high degree of worry about disease recurrence/progression, development of treatment-related complications, or development of a new cancer produce a change in care-seeking behavior and may concomitantly alter the health provider's need to see the patient. Our results also suggest that while spirituality has an impact on one's level of worry, being less spiritual does not necessarily alter a cancer survivor's care-seeking behavior.

Worried patients present a potential problem for clinicians in that they may need more attention during clinic visits,³⁷ may result in requests for more ancillary/diagnostic tests including imaging modalities,^{[38] and [39]} or may use more medications^{[40] and [41]} or resort to other alternative therapies^{[42], [43], [44] and [45]} available to reduce their worries. Given that cancer patients already receive many chemotherapeutic agents for their treatment, many of them are more inclined to undergo alternative therapies.^{[16], [43], [46], [47] and [48]} Spirituality-based interventions shown to be effective at reducing anxiety and increasing QOL may therefore have a role among cancer survivors. And because spirituality and religiosity are closely linked,²⁹ faith-based interventions may also benefit the patient.

Our study has several implications in the assessment of cancer survivors in multidisciplinary survivorship clinics. While much attention about assessing depression, anxiety, and QOL has been given to cancer survivors, our study shows that the evaluation of one's spirituality may have some merit as well. Participants with low spirituality and a high degree of worry may benefit from activities that enhance spirituality (e.g., yoga, meditation). Because of the increasing number of cancer survivors,^{[32] and [49]} development of clinic-based spiritual interventions to address common worries of cancer survivors may be appropriate. In addition to the implications for clinical practice, our study has implications for future research. While the literature has shown a correlation between spirituality and religiosity,²⁹ these two concepts are not the same.^{[1], [2], [50], [51] and [52]} It would have been interesting to compare outcomes by level of spirituality and religiosity, but our data revealed a high degree of correlation between these two concepts. Over 90% of individuals who are spiritual are also religious.^{[28], [53] and [54]} This may be the reason that some spirituality-based interventions have enhancement of religious activities as main approaches to improve spirituality.^{[28] and [53]}

While our study has the strengths associated with a prospective study in a relatively large number of cancer survivors treated in a single medical center, it has several limitations. Our participation rate at baseline was only 50%, although our retention rates at 6 and 12 months were on average 80%. Another limitation of our study is that the baseline surveys were conducted at different time intervals from last treatment, although this limitation also allowed us to include all kinds of cancer survivors in terms of disease and time interval from last cancer treatment. Analysis confined to patients who received treatment within the last 5 years (n = 371) showed essentially the same results. We also compared the baseline spirituality scores of the study participants according to time from last treatment to study participation (0–2, 2–5, >5) and showed no statistically significant differences. Additionally, we adjusted for time from last treatment to study participation in the multivariate analyses. Combining all the participants into one analysis allowed for our exploratory analyses to have stronger statistical power. Another limitation of our study is the crude measurement of patient worry. However, in the absence of validated instruments to measure these worries, we felt the measures reflected subjective ratings of common worries by cancer survivors. Health-care utilization would have been ideally measured continuously to better quantify the medical services utilized. However, because we included a heterogeneous group of cancer patients, this measure would be highly variable and depend on the type of disease and treatment received by the patient. Thus, type of disease and time period from last treatment were adjusted for in the multivariate analyses.

In summary, cancer survivors who possess higher levels of spirituality tend to have a lesser degree of worry over disease recurrence/progression, development of treatment complications, and development of new cancers. A higher degree of worry about the common concerns of cancer survivors is associated with more follow-up visits and calls to health providers. However, our data showed that, for the most part, spirituality is not associated with follow-up health-care utilization.

Acknowledgments

The authors thank Linda Bauer, Garrett Frost, and Gregory McFadden for their help in coordinating the study and processing the data. This work was supported by the University of Nebraska Medical Center–Eppley Cancer Center (Support Grant P30 CA 036727) and the Medical Student Research Program. The funding source had no role in the design, collection, analysis, and interpretation of the data or in the writing of the article.

Statistical Analysis

Participant characteristics were compared according to level of spirituality using a chi-square test for categorical data and the Wilcoxon test for continuous data (Table 1). Multivariate logistic regression models were fitted to evaluate separately the relationship between (1) spirituality with patient-rated worry as the outcome, (2) spirituality with follow-up health-care utilization as the outcome, and (3) patient-rated worry with follow-up health-care utilization as the outcome. In the above models, the following covariates were forced into each model: age, sex, cancer type, time from last cancer-related treatment to study start time, income, and type of medical insurance. These models were also fitted using outcomes ascertained at both 6 and 12 months. Interaction models between patient-rated worry and level of spirituality were also evaluated for an association with follow-up health-care utilization at 12 months to explore the role of spirituality in the relationship between patient-rated worry and health-care utilization. A P value of at least 0.05 was considered statistically significant.

Results

Study Participation

Of the 2,000 participants invited, 1,881 were deemed eligible (minus those who died or had wrong addresses). Baseline questionnaires were returned by 939 participants (baseline response rate of 50%). Seventeen wanted to participate only in the baseline survey. Of the 922 baseline participants, 691 returned the 6-month survey at the time of the analysis for this study, for a response rate of 76% when adjusted for deaths (182 no response, 18 deaths, 25 declined, 12 returned with wrong address). At 1 year, 691 surveys were mailed, with 588 surveys returned (58 no response, 17 deaths, 14 declined, 13 returned with wrong address, and one in hospice); a response rate of 87% was achieved after adjusting for deaths. Thirty-seven participants had missing information on spirituality, leaving a total of 551 included in this study. No differences in age, sex, and type of cancer were noted between patients included and excluded in the current analysis.

Characteristics of Study Participants

Demographic characteristics of the 551 study participants included in this study are shown in Table 1. We found that cancer survivors with low or high spirituality were more similar than different in all but two characteristics: highly spiritual survivors were more likely to be Protestant and male.

Prevalence of Spirituality and Patient Worry

Within our population, 271 (49%) survivors reported low spirituality and 280 (51%) reported high spirituality (Table 1). Also, at baseline, 277 (51%) survivors reported high levels of recurrence/progression-related worry, 190 survivors (35%) reported high levels of new malignancy–related worry, and 178 survivors (33%) reported high levels of treatment-related complication worry. As some participants may have reported one or more types of worry, this translates to 322 (59%) reporting any type of worry. Highly spiritual survivors reported significantly lower levels of high worry concerning recurrence/progression (6-month 27% vs. 38%, P < 0.01; 12-month 21% vs. 38%, P < 0.01), development of a different type of cancer (6-month 22% vs. 31%, P = 0.03; 12-month 15% vs. 26%, P < 0.01), and complications from treatment (6-month 17% vs. 30%, P < 0.01; 12-month 16% vs. 26%, P < 0.01). Highly spiritual survivors reported significantly lower levels for any type of worry at both 6 and 12 months (6 months 37% vs. 54%, P <0.01; 12 months 28% vs. 47%, P < 0.01) (Table 2).

Relationship Between Spirituality and Patient Worry

At the 6- and 12-month time points, after adjusting for covariates, highly spiritual survivors were significantly less likely to have worries than survivors who reported lower spirituality regarding disease recurrence/progression at 6 months (odds ratio [OR] = 0.61, 95% confidence interval [CI] 0.42–0.89, P < 0.01) and at 12 months (OR = 0.43, 95% CI 0.29–0.63, P < 0.01), complications from treatment at 6 months (OR = 0.50, 95% CI 0.33–0.76, P < 0.01) and at 12 months (OR = 0.54, 95% CI 0.35–0.83, P < 0.01), and development of a different type of cancer at 6 months (OR = 0.65, 95% CI 0.44–0.97, P = 0.04) and at 12 months (OR = 0.50, 95% CI 0.33–0.77, P < 0.01) (Table 3A).

Relationship Between Patient Worry and Follow-Up Health-Care Utilization

Survivors who were highly worried about disease recurrence/progression, development of another type of cancer, and/or complications from treatment were more likely to visit their providers for follow-up care when compared with survivors who were less worried at 6 months (OR = 1.81, 95% CI 1.04–3.12, P = 0.03) and at 12 months (OR = 1.49, 95% CI 1.00–2.22, P = 0.05). Similarly, survivors who were highly worried were also more likely to place phone calls to their follow-up providers for medical inquiries than survivors who were less worried at 6 months (OR = 2.21, 95% CI 1.48–3.31, P < 0.01) and at 12 months (OR = 1.74, 95% CI 1.20–2.53, P = 0.01). We did not observe differences in emergency room visits between survivors with low and those with high rates of worrying at both 6 and 12 months (Table 3B).

Relationship Between Spirituality and Health-Care Utilization

No significant differences were noted for the frequency of follow-up visits, changes in follow-up providers, and emergency room visits between the levels of spirituality at both 6 and 12 months. However, at 12 months, highly spiritual survivors were less likely to call their follow-up providers for medical inquiries compared to survivors with low spirituality scores (OR = 0.70, 95% CI 0.49–1.00, P = 0.04) (Table 3C).

Interaction Between Spirituality and Patient Worry With Health-Care Utilization

Interaction between patient-rated worry and level of spirituality as it relates to health-care utilization was not statistically significant (data not shown). This suggests that spirituality does not modify the effect of patient worry in producing change in follow-up health-care utilization.

Discussion

Our study examined the relationships between spirituality, patient-rated worry, and follow-up health-care utilization among cancer survivors and found that individuals who possess higher levels of spirituality tend to have less worry of disease recurrence/progression, development of treatment-related complications, and development of new cancers. These findings are consistent with previous research among patients with advanced or terminal cancers that consistently showed such correlations between spirituality and general measures of anxiety.^{[10], [15], [17], [19], [30] and [31]} Additionally, our study showed that a higher degree of worry about common concerns of cancer survivors is associated with more follow-up visits and calls to health-care providers. However, our data also showed that spirituality by itself is for the most part not associated with follow-up health-care utilization.

It has been documented that psychosocial factors like anxiety and spirituality can influence behaviors.^{[32], [33], [34], [35] and [36]} Our analysis showed that both discretionary and nondiscretionary indices of health-care utilization increased significantly among highly worried cancer survivors. However, these increases are independent of one's level of spirituality. These results suggest that cancer survivors with a high degree of worry about disease recurrence/progression, development of treatment-related complications, or development of a new cancer produce a change in care-seeking behavior and may concomitantly alter the health provider's need to see the patient. Our results also suggest that while spirituality has an impact on one's level of worry, being less spiritual does not necessarily alter a cancer survivor's care-seeking behavior.

Worried patients present a potential problem for clinicians in that they may need more attention during clinic visits,³⁷ may result in requests for more ancillary/diagnostic tests including imaging modalities,^{[38] and [39]} or may use more medications^{[40] and [41]} or resort to other alternative therapies^{[42], [43], [44] and [45]} available to reduce their worries. Given that cancer patients already receive many chemotherapeutic agents for their treatment, many of them are more inclined to undergo alternative therapies.^{[16], [43], [46], [47] and [48]} Spirituality-based interventions shown to be effective at reducing anxiety and increasing QOL may therefore have a role among cancer survivors. And because spirituality and religiosity are closely linked,²⁹ faith-based interventions may also benefit the patient.

Our study has several implications in the assessment of cancer survivors in multidisciplinary survivorship clinics. While much attention about assessing depression, anxiety, and QOL has been given to cancer survivors, our study shows that the evaluation of one's spirituality may have some merit as well. Participants with low spirituality and a high degree of worry may benefit from activities that enhance spirituality (e.g., yoga, meditation). Because of the increasing number of cancer survivors,^{[32] and [49]} development of clinic-based spiritual interventions to address common worries of cancer survivors may be appropriate. In addition to the implications for clinical practice, our study has implications for future research. While the literature has shown a correlation between spirituality and religiosity,²⁹ these two concepts are not the same.^{[1], [2], [50], [51] and [52]} It would have been interesting to compare outcomes by level of spirituality and religiosity, but our data revealed a high degree of correlation between these two concepts. Over 90% of individuals who are spiritual are also religious.^{[28], [53] and [54]} This may be the reason that some spirituality-based interventions have enhancement of religious activities as main approaches to improve spirituality.^{[28] and [53]}

While our study has the strengths associated with a prospective study in a relatively large number of cancer survivors treated in a single medical center, it has several limitations. Our participation rate at baseline was only 50%, although our retention rates at 6 and 12 months were on average 80%. Another limitation of our study is that the baseline surveys were conducted at different time intervals from last treatment, although this limitation also allowed us to include all kinds of cancer survivors in terms of disease and time interval from last cancer treatment. Analysis confined to patients who received treatment within the last 5 years (n = 371) showed essentially the same results. We also compared the baseline spirituality scores of the study participants according to time from last treatment to study participation (0–2, 2–5, >5) and showed no statistically significant differences. Additionally, we adjusted for time from last treatment to study participation in the multivariate analyses. Combining all the participants into one analysis allowed for our exploratory analyses to have stronger statistical power. Another limitation of our study is the crude measurement of patient worry. However, in the absence of validated instruments to measure these worries, we felt the measures reflected subjective ratings of common worries by cancer survivors. Health-care utilization would have been ideally measured continuously to better quantify the medical services utilized. However, because we included a heterogeneous group of cancer patients, this measure would be highly variable and depend on the type of disease and treatment received by the patient. Thus, type of disease and time period from last treatment were adjusted for in the multivariate analyses.

In summary, cancer survivors who possess higher levels of spirituality tend to have a lesser degree of worry over disease recurrence/progression, development of treatment complications, and development of new cancers. A higher degree of worry about the common concerns of cancer survivors is associated with more follow-up visits and calls to health providers. However, our data showed that, for the most part, spirituality is not associated with follow-up health-care utilization.

Acknowledgments

The authors thank Linda Bauer, Garrett Frost, and Gregory McFadden for their help in coordinating the study and processing the data. This work was supported by the University of Nebraska Medical Center–Eppley Cancer Center (Support Grant P30 CA 036727) and the Medical Student Research Program. The funding source had no role in the design, collection, analysis, and interpretation of the data or in the writing of the article.